Discussion:
OpenRISC Architecture: Patch set version 2
Jonas Bonn
2011-07-02 21:15:33 UTC
Permalink
This is version 2 of the OpenRISC architecture patch set. Changes
since version 1 include:

- general cleanups, removing unused code and variables, taking care of
compiler warnings, etc.
- cleanups for all the errors spotted by checkpatch.pl
- changes in response to the feedback received for version 1 of the patch set
- simplification of the fixmap code
- simplification of the DMA code
- calculate loops_per_jiffy from cpu frequency passed in via devicetree
- rework io.h to use more generic code
- rework the delay code, making it more generic

- also, the module code has been reworked, making it more generic; this is
not included in this patch as the generic parts are still pending review

Since version 1:
54 files changed, 793 insertions(+), 1900 deletions(-)

Once again, all this code can be found in the 'for-upstream' branch at:

git://openrisc.net/jonas/linux

Note, that there are three patches at the base of the branch that make
some changes to generic kernel code that are necessary for the OpenRISC
patches to work. Those patches aren't included in this post as they have
already been posted for review earlier.

In this series:

[PATCH v2 01/19] OpenRISC: Boot code
[PATCH v2 02/19] OpenRISC: Device tree
[PATCH v2 03/19] OpenRISC: Memory management
[PATCH v2 04/19] OpenRISC: Signal handling
[PATCH v2 05/19] OpenRISC: Build infrastructure
[PATCH v2 06/19] OpenRISC: PTrace
[PATCH v2 07/19] OpenRISC: DMA
[PATCH v2 08/19] OpenRISC: Timekeeping
[PATCH v2 09/19] OpenRISC: IRQ
[PATCH v2 10/19] OpenRISC: System calls
[PATCH v2 11/19] OpenRISC: Idle/Power management
[PATCH v2 12/19] OpenRISC: Scheduling/Process management
[PATCH v2 13/19] OpenRISC: GPIO
[PATCH v2 14/19] OpenRISC: Module support
[PATCH v2 15/19] OpenRISC: Traps
[PATCH v2 16/19] OpenRISC: Headers
[PATCH v2 17/19] OpenRISC: Library routines
[PATCH v2 18/19] OpenRISC: Miscellaneous
[PATCH v2 19/19] OpenRISC: Add MAINTAINERS entry

Looking forward to further feedback!

/Jonas

Diffstat for entire series:

MAINTAINERS | 8 +
arch/openrisc/Kconfig | 207 ++++
arch/openrisc/Makefile | 66 ++
arch/openrisc/README.openrisc | 102 ++
arch/openrisc/TODO.openrisc | 18 +
arch/openrisc/boot/Makefile | 15 +
arch/openrisc/boot/dts/atlys.dts | 78 ++
arch/openrisc/boot/dts/or1ksim.dts | 73 ++
arch/openrisc/configs/or1ksim_defconfig | 67 ++
arch/openrisc/include/asm/Kbuild | 64 ++
arch/openrisc/include/asm/asm-offsets.h | 1 +
arch/openrisc/include/asm/bitops.h | 59 ++
arch/openrisc/include/asm/bitops/__ffs.h | 33 +
arch/openrisc/include/asm/bitops/__fls.h | 33 +
arch/openrisc/include/asm/bitops/ffs.h | 32 +
arch/openrisc/include/asm/bitops/fls.h | 33 +
arch/openrisc/include/asm/byteorder.h | 1 +
arch/openrisc/include/asm/cache.h | 29 +
arch/openrisc/include/asm/cpuinfo.h | 34 +
arch/openrisc/include/asm/delay.h | 24 +
arch/openrisc/include/asm/dma-mapping.h | 62 ++
arch/openrisc/include/asm/elf.h | 109 ++
arch/openrisc/include/asm/fixmap.h | 88 ++
arch/openrisc/include/asm/gpio.h | 65 ++
arch/openrisc/include/asm/io.h | 51 +
arch/openrisc/include/asm/irq.h | 27 +
arch/openrisc/include/asm/irqflags.h | 29 +
arch/openrisc/include/asm/linkage.h | 25 +
arch/openrisc/include/asm/memblock.h | 24 +
arch/openrisc/include/asm/mmu.h | 26 +
arch/openrisc/include/asm/mmu_context.h | 43 +
arch/openrisc/include/asm/mutex.h | 27 +
arch/openrisc/include/asm/page.h | 112 +++
arch/openrisc/include/asm/param.h | 26 +
arch/openrisc/include/asm/pgalloc.h | 142 +++
arch/openrisc/include/asm/pgtable.h | 449 +++++++++
arch/openrisc/include/asm/processor.h | 112 +++
arch/openrisc/include/asm/prom.h | 92 ++
arch/openrisc/include/asm/ptrace.h | 124 +++
arch/openrisc/include/asm/serial.h | 36 +
arch/openrisc/include/asm/sigcontext.h | 38 +
arch/openrisc/include/asm/spinlock.h | 24 +
arch/openrisc/include/asm/spr.h | 42 +
arch/openrisc/include/asm/spr_defs.h | 604 +++++++++++
arch/openrisc/include/asm/syscall.h | 77 ++
arch/openrisc/include/asm/syscalls.h | 27 +
arch/openrisc/include/asm/system.h | 35 +
arch/openrisc/include/asm/thread_info.h | 149 +++
arch/openrisc/include/asm/timex.h | 36 +
arch/openrisc/include/asm/tlb.h | 34 +
arch/openrisc/include/asm/tlbflush.h | 72 ++
arch/openrisc/include/asm/uaccess.h | 385 +++++++
arch/openrisc/include/asm/unaligned.h | 51 +
arch/openrisc/include/asm/unistd.h | 29 +
arch/openrisc/kernel/Makefile | 14 +
arch/openrisc/kernel/asm-offsets.c | 70 ++
arch/openrisc/kernel/dma.c | 86 ++
arch/openrisc/kernel/entry.S | 1134 +++++++++++++++++++++
arch/openrisc/kernel/head.S | 1607 ++++++++++++++++++++++++++++++
arch/openrisc/kernel/idle.c | 77 ++
arch/openrisc/kernel/init_task.c | 41 +
arch/openrisc/kernel/irq.c | 172 ++++
arch/openrisc/kernel/module.c | 126 +++
arch/openrisc/kernel/or32_ksyms.c | 46 +
arch/openrisc/kernel/process.c | 305 ++++++
arch/openrisc/kernel/prom.c | 109 ++
arch/openrisc/kernel/ptrace.c | 574 +++++++++++
arch/openrisc/kernel/ptrace.h | 37 +
arch/openrisc/kernel/setup.c | 386 +++++++
arch/openrisc/kernel/signal.c | 404 ++++++++
arch/openrisc/kernel/sys_call_table.c | 28 +
arch/openrisc/kernel/sys_or32.c | 83 ++
arch/openrisc/kernel/time.c | 181 ++++
arch/openrisc/kernel/traps.c | 367 +++++++
arch/openrisc/kernel/vmlinux.h | 12 +
arch/openrisc/kernel/vmlinux.lds.S | 115 +++
arch/openrisc/lib/Makefile | 5 +
arch/openrisc/lib/delay.c | 60 ++
arch/openrisc/lib/string.S | 204 ++++
arch/openrisc/mm/Makefile | 5 +
arch/openrisc/mm/fault.c | 338 +++++++
arch/openrisc/mm/init.c | 283 ++++++
arch/openrisc/mm/ioremap.c | 113 +++
arch/openrisc/mm/tlb.c | 191 ++++
arch/openrisc/or1ksim.cfg | 888 +++++++++++++++++
85 files changed, 12310 insertions(+), 0 deletions(-)
Jonas Bonn
2011-07-02 21:15:39 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/ptrace.h | 124 ++++++++
arch/openrisc/kernel/ptrace.c | 574 ++++++++++++++++++++++++++++++++++++
arch/openrisc/kernel/ptrace.h | 37 +++
3 files changed, 735 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/ptrace.h
create mode 100644 arch/openrisc/kernel/ptrace.c
create mode 100644 arch/openrisc/kernel/ptrace.h

diff --git a/arch/openrisc/include/asm/ptrace.h b/arch/openrisc/include/asm/ptrace.h
new file mode 100644
index 0000000..d3f620e
--- /dev/null
+++ b/arch/openrisc/include/asm/ptrace.h
@@ -0,0 +1,124 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_PTRACE_H
+#define __ASM_OPENRISC_PTRACE_H
+
+#include <asm/spr_defs.h>
+/*
+ * This struct defines the way the registers are stored on the
+ * kernel stack during a system call or other kernel entry.
+ *
+ * this should only contain volatile regs
+ * since we can keep non-volatile in the thread_struct
+ * should set this up when only volatiles are saved
+ * by intr code.
+ *
+ * Since this is going on the stack, *CARE MUST BE TAKEN* to insure
+ * that the overall structure is a multiple of 16 bytes in length.
+ *
+ * Note that the offsets of the fields in this struct correspond with
+ * the values below.
+ */
+
+/*
+ * These are 'magic' values for PTRACE_PEEKUSR that return info about where a
+ * process is located in memory.
+ */
+#define PT_TEXT_ADDR 0x10000
+#define PT_DATA_ADDR 0x10004
+#define PT_TEXT_END_ADDR 0x10008
+
+#ifndef __ASSEMBLY__
+
+struct pt_regs {
+ union {
+ struct {
+ /* Named registers */
+ long sr; /* Stored in place of r0 */
+ long sp; /* r1 */
+ };
+ struct {
+ /* Old style */
+ long offset[2];
+ long gprs[30];
+ };
+ struct {
+ /* New style */
+ long gpr[32];
+ };
+ };
+ long pc;
+ long orig_gpr11; /* Used for restarting system calls */
+ long syscallno; /* Syscall no. (used by strace) */
+};
+#endif /* __ASSEMBLY__ */
+
+#ifdef __KERNEL__
+#define STACK_FRAME_OVERHEAD 128 /* size of minimum stack frame */
+//#define STACK_FRAME_OVERHEAD 0 /* size of minimum stack frame */
+
+#define instruction_pointer(regs) ((regs)->pc)
+#define user_mode(regs) (((regs)->sr & SPR_SR_SM) == 0)
+#define user_stack_pointer(regs) ((unsigned long)(regs)->sp)
+#define profile_pc(regs) instruction_pointer(regs)
+
+#define arch_has_single_step() (1)
+
+#endif /* __KERNEL__ */
+
+/*
+ * Offsets used by 'ptrace' system call interface.
+ */
+#define PT_SR 0
+#define PT_SP 4
+#define PT_GPR2 8
+#define PT_GPR3 12
+#define PT_GPR4 16
+#define PT_GPR5 20
+#define PT_GPR6 24
+#define PT_GPR7 28
+#define PT_GPR8 32
+#define PT_GPR9 36
+#define PT_GPR10 40
+#define PT_GPR11 44
+#define PT_GPR12 48
+#define PT_GPR13 52
+#define PT_GPR14 56
+#define PT_GPR15 60
+#define PT_GPR16 64
+#define PT_GPR17 68
+#define PT_GPR18 72
+#define PT_GPR19 76
+#define PT_GPR20 80
+#define PT_GPR21 84
+#define PT_GPR22 88
+#define PT_GPR23 92
+#define PT_GPR24 96
+#define PT_GPR25 100
+#define PT_GPR26 104
+#define PT_GPR27 108
+#define PT_GPR28 112
+#define PT_GPR29 116
+#define PT_GPR30 120
+#define PT_GPR31 124
+#define PT_PC 128
+#define PT_ORIG_GPR11 132
+#define PT_SYSCALLNO 136
+
+#endif /* __ASM_OPENRISC_PTRACE_H */
diff --git a/arch/openrisc/kernel/ptrace.c b/arch/openrisc/kernel/ptrace.c
new file mode 100644
index 0000000..a167a1e
--- /dev/null
+++ b/arch/openrisc/kernel/ptrace.c
@@ -0,0 +1,574 @@
+/*
+ * OpenRISC ptrace.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2005 Gyorgy Jeney <***@bsemi.com>
+ * Copyright (C) 2010 Julius Baxter <***@orsoc.se>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <stddef.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/audit.h>
+#include <linux/regset.h>
+#include <linux/tracehook.h>
+#include <linux/elf.h>
+
+#include <asm/thread_info.h>
+#include <asm/segment.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/system.h>
+
+#include "ptrace.h"
+
+/*
+ * retrieve the contents of OpenRISC userspace general registers
+ */
+static int genregs_get(struct task_struct *target,
+ const struct user_regset *regset,
+ unsigned int pos, unsigned int count,
+ void *kbuf, void __user * ubuf)
+{
+ const struct pt_regs *regs = task_pt_regs(target);
+ int ret;
+
+#if 0
+ /* r0 */
+ ret = user_regset_copyout_zero(&pos, &count, &kbuf, &ubuf,
+ 0, offsetof(struct pt_regs, regs));
+#endif
+
+ ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+ regs, 0, sizeof(*regs));
+
+/* put PPC here */
+
+#if 0
+ /* fill out rest of elf_gregset_t structure with zeroes */
+ if (!ret)
+ ret = user_regset_copyout_zero(&pos, &count, &kbuf, &ubuf,
+ sizeof(struct pt_regs), -1);
+#endif
+
+ return ret;
+}
+
+/*
+ * update the contents of the OpenRISC userspace general registers
+ */
+static int genregs_set(struct task_struct *target,
+ const struct user_regset *regset,
+ unsigned int pos, unsigned int count,
+ const void *kbuf, const void __user * ubuf)
+{
+ struct pt_regs *regs = task_pt_regs(target);
+ int ret;
+
+ /* PC */
+ ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+ regs, 0, offsetof(struct pt_regs, sr));
+
+ /* skip SR */
+ ret = user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf,
+ offsetof(struct pt_regs, sr),
+ offsetof(struct pt_regs, sp));
+
+ /* SP, r2 - r31 */
+ ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+ regs,
+ offsetof(struct pt_regs, sp),
+ sizeof(struct pt_regs));
+
+#if 0
+ /* read out the rest of the elf_gregset_t structure */
+ if (!ret)
+ ret = user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf,
+ sizeof(struct pt_regs), -1);
+#endif
+
+ return ret;
+}
+
+/*
+ * Define the register sets available on OpenRISC under Linux
+ */
+enum openrisc_regset {
+ REGSET_GENERAL,
+};
+
+static const struct user_regset openrisc_regsets[] = {
+ [REGSET_GENERAL] = {
+ .core_note_type = NT_PRSTATUS,
+ .n = ELF_NGREG,
+ .size = sizeof(long),
+ .align = sizeof(long),
+ .get = genregs_get,
+ .set = genregs_set,
+ },
+};
+
+static const struct user_regset_view user_openrisc_native_view = {
+ .name = "OpenRISC",
+ .e_machine = EM_OPENRISC,
+ .regsets = openrisc_regsets,
+ .n = ARRAY_SIZE(openrisc_regsets),
+};
+
+const struct user_regset_view *task_user_regset_view(struct task_struct *task)
+{
+ return &user_openrisc_native_view;
+}
+
+void user_enable_single_step(struct task_struct *child)
+{
+ set_tsk_thread_flag(child, TIF_SINGLESTEP);
+}
+
+void user_disable_single_step(struct task_struct *child)
+{
+ clear_tsk_thread_flag(child, TIF_SINGLESTEP);
+}
+
+/*
+ * does not yet catch signals sent when the child dies.
+ * in exit.c or in signal.c.
+ */
+
+#define OPC_MASK 0x3f
+#define OPC_SHIFT 26
+#define OPC_J 0x00
+#define OPC_JAL 0x01
+#define OPC_BNF 0x03
+#define OPC_BF 0x04
+#define OPC_SYSC 0x08
+#define OPC_JR 0x11
+#define OPC_JALR 0x12
+
+#define JUMP_IMM_MASK 0x03ffffff
+#define JUMP_IMM_SIGN 0x02000000
+
+/* Trap instruction on bit 15 of SR, fixed one, unconditional */
+#define OR1K_TRAP 0x2100000f
+
+/* Macro to extract branch offset with sign extension of immediate. */
+#define BRANCH_OFFSET(insn) ((insn & JUMP_IMM_SIGN) ? \
+ ((insn & JUMP_IMM_MASK) << 2) | 0xf0000000 : \
+ ((insn & JUMP_IMM_MASK) << 2))
+
+/* Get address of next instruction we can insert a breakpoint on */
+/* pc passed to function is address of instruction we've now just trapped on.
+ It is yet/next to be executed. We determine next instruction after PC to
+ place a l.trap on. If we should take a branch, we must remember this.
+*/
+static unsigned long
+get_next_address(struct task_struct *tsk, unsigned long pc, unsigned long insn)
+{
+ struct pt_regs *regs;
+ struct debug_info *dbg;
+ char opc;
+ unsigned long npc;
+ unsigned long rB;
+
+ dbg = &current_thread_info()->debug;
+
+ regs = task_pt_regs(tsk);
+ /* Extract opcode from instruction */
+ opc = (insn >> OPC_SHIFT) & OPC_MASK;
+
+ pr_debug(KERN_INFO
+ "ptrace get_next_address: pc 0x%.8lx insn. 0x%.8lx opc. 0x%.2x\n",
+ pc, insn, opc);
+
+ /*
+ Will always proceed to next instruction if this function was called
+ ie. we know we're not in a delay slot - the function calling this
+ checks dbg->branch_taken.
+ */
+ npc = pc + 4;
+
+ /* Check if we're to record a branch */
+ switch (opc) {
+ case OPC_BNF:
+ /* Check flag - we're branching if !flag */
+ if (!(((regs)->sr) & SPR_SR_F)) {
+ dbg->bp.branch = 1;
+ dbg->bp.branch_target = pc + BRANCH_OFFSET(insn);
+ }
+ break;
+ case OPC_BF:
+ /* Check flag - we're branching if flag */
+ if (((regs)->sr) & SPR_SR_F) {
+ dbg->bp.branch = 1;
+ dbg->bp.branch_target = pc + BRANCH_OFFSET(insn);
+ }
+ break;
+ case OPC_J:
+ case OPC_JAL:
+ /* PC-relative branch target encoded in instruction. Extract
+ and add it. */
+ dbg->bp.branch = 1;
+ dbg->bp.branch_target = pc + BRANCH_OFFSET(insn);
+ break;
+ case OPC_JR:
+ case OPC_JALR:
+ /* Register number holding branch target is encoded in rB slot
+ of instruction. Extract it. */
+ rB = (insn >> 11) & 0x1f;
+ pr_debug(KERN_INFO
+ "ptrace get_next_address: jump reg from r%ld=0x%08lx\n",
+ rB,
+ /* *((unsigned long*)((char*)regs + (rB<<2) + 4)) */
+ regs->gprs[rB - 2]);
+ if (rB < 2)
+ pr_debug(KERN_WARNING
+ "ptrace get_next_address(): Warning, JR with GPR < 2");
+ dbg->bp.branch = 1;
+ dbg->bp.branch_target =
+ /* *((unsigned long*)((char*)regs + (rB<<2) + 4)) */
+ regs->gprs[rB - 2];
+ break;
+#if 0
+ case OPC_SYSC:
+ /* Not sure we want to do this */
+ /* Remember - l.sys has no delay slot. */
+ npc = 0x900;
+ break;
+#endif
+ default:
+ break;
+ }
+
+ /* If setting a branch target, remember where branch was */
+ if (dbg->bp.branch)
+ dbg->bp.branch_insn_address = pc;
+
+ if (dbg->bp.branch)
+ pr_debug(KERN_INFO
+ "ptrace get_next_address: branch detected to 0x%.8lx\n",
+ dbg->bp.branch_target);
+ pr_debug(KERN_INFO "ptrace get_next_address: returning 0x%.8lx\n", npc);
+
+ return npc;
+}
+
+static inline int
+read_instr(struct task_struct *tsk, unsigned long addr, u32 * res)
+{
+ int ret;
+ u32 val;
+ ret = access_process_vm(tsk, addr & ~3, &val, sizeof(val), 0);
+ ret = ret == sizeof(val) ? 0 : -EIO;
+ *res = val;
+ return ret;
+}
+
+static int
+swap_insn(struct task_struct *tsk, unsigned long addr,
+ void *old_insn, void *new_insn, int size)
+{
+ int ret;
+
+ ret = access_process_vm(tsk, addr, old_insn, size, 0);
+ if (ret == size)
+ ret = access_process_vm(tsk, addr, new_insn, size, 1);
+ return ret;
+}
+
+static void
+add_breakpoint(struct task_struct *tsk, struct debug_info *dbg,
+ unsigned long addr)
+{
+ u32 new_insn = OR1K_TRAP;
+ int res;
+
+ res = swap_insn(tsk, addr, &dbg->bp.insn, &new_insn, 4);
+
+ pr_debug(KERN_INFO
+ "ptrace add_breakpoint: addr 0x%.8lx insn 0x%.8lx %d\n\n",
+ addr, dbg->bp.insn, res);
+
+ if (res == 4) {
+ dbg->bp.address = addr;
+ dbg->bp.set = 1;
+ }
+}
+
+/*
+ * Clear the breakpoint in the user program.
+ */
+static void clear_breakpoint(struct task_struct *tsk, struct debug_entry *bp)
+{
+ u32 old_insn;
+ int ret;
+ struct pt_regs *regs;
+ unsigned long pc;
+
+ unsigned long addr = bp->address;
+
+ regs = task_pt_regs(tsk);
+ pc = instruction_pointer(regs); /* This is NPC */
+
+ /* Either at the breakpoint, or in delay slot (pc will be at address of
+ * branch instruction - OR1K exception handlers do that, for now.
+ * Perhaps FIXME
+ */
+ if ((pc == addr) || ((bp->branch_insn_address == pc) && bp->branch))
+ ret = swap_insn(tsk, addr & ~3, &old_insn, &bp->insn, 4);
+ else {
+ /* Trapped for some other reason. */
+ pr_debug(KERN_INFO
+ "ptrace clear_breakpoint: not correct, pc 0x%.8lx bp 0x%.8lx\n",
+ pc, addr);
+ pr_debug(KERN_INFO
+ "ptrace clear_breakpoint: 0x%.8lx 0x%.8lx %ld\n",
+ bp->branch_insn_address, bp->branch_target,
+ bp->branch);
+ return;
+ }
+
+ pr_debug(KERN_INFO
+ "ptrace clear_breakpoint: addr 0x%.8lx insn 0x%.8lx %d\n",
+ addr, bp->insn, ret);
+
+ if (ret != 4 || old_insn != OR1K_TRAP)
+ pr_debug(KERN_ERR
+ "%s:%d: corrupted breakpoint at 0x%08lx (0x%08x)\n",
+ tsk->comm, task_pid_nr(tsk), addr, old_insn);
+
+ bp->set = 0;
+}
+
+void ptrace_set_bpt(struct task_struct *tsk)
+{
+ struct pt_regs *regs;
+ unsigned long pc;
+ u32 insn;
+ int res;
+
+ regs = task_pt_regs(tsk);
+ pc = instruction_pointer(regs); /* This is NPC */
+
+ res = read_instr(tsk, pc, &insn);
+ if (!res) {
+ struct debug_info *dbg = &current_thread_info()->debug;
+ unsigned long npc;
+
+ /* First check if breakpoint is still set, if so, probably
+ just let us run until we hit it.
+ This can occur because perhaps we caused an exception before
+ a jump or in a delay slot, and the OR1K exception handlers
+ will rewind us to the branch before the delay slot or jump
+ target.
+ This way we skip everything that goes on during an exception
+ and can hopefully just track the program's execution.
+ */
+ if (dbg->bp.set)
+ return;
+
+ /* Are we in delay slot of branch we should take? */
+ if (dbg->bp.branch == 1) {
+ /* Branch target is already determined. Set l.trap. */
+ npc = dbg->bp.branch_target;
+
+ /* Reset NPC to branch instruction - should be one
+ * before this one.
+ */
+ instruction_pointer(regs) = dbg->bp.branch_insn_address;
+
+ /* Now mark this as no longer needing to be taken */
+ dbg->bp.branch = 0;
+ } else {
+ npc = get_next_address(tsk, pc, insn);
+ }
+
+ add_breakpoint(tsk, dbg, npc);
+ }
+}
+
+/*
+ * Ensure no single-step breakpoint is pending. Returns non-zero
+ * value if child was being single-stepped.
+ */
+void ptrace_cancel_bpt(struct task_struct *tsk)
+{
+ clear_breakpoint(tsk,
+ (struct debug_entry *)&current_thread_info()->debug);
+}
+
+/*
+ * Called by kernel/ptrace.c when detaching..
+ *
+ * Make sure the single step bit is not set.
+ */
+void ptrace_disable(struct task_struct *child)
+{
+ pr_debug(KERN_WARNING "ptrace_disable(): TODO\n");
+
+ user_disable_single_step(child);
+ clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+}
+
+/*
+ * Read the word at offset "off" into the "struct user". We
+ * actually access the pt_regs stored on the kernel stack.
+ */
+static int ptrace_read_user(struct task_struct *tsk, unsigned long off,
+ unsigned long __user *ret)
+{
+ struct pt_regs *regs;
+ unsigned long tmp;
+
+ /* if (off & 3 || off >= sizeof(struct user))
+ return -EIO; */
+
+ regs = task_pt_regs(tsk);
+
+ tmp = 0;
+ if (off == PT_TEXT_ADDR)
+ tmp = tsk->mm->start_code;
+ else if (off == PT_DATA_ADDR)
+ tmp = tsk->mm->start_data;
+ else if (off == PT_TEXT_END_ADDR)
+ tmp = tsk->mm->end_code;
+ else if (off < sizeof(struct pt_regs))
+ tmp = *((unsigned long *)((char *)regs + off));
+
+ return put_user(tmp, ret);
+}
+
+/*
+ * Write the word at offset "off" into "struct user". We
+ * actually access the pt_regs stored on the kernel stack.
+ */
+static int ptrace_write_user(struct task_struct *tsk, unsigned long off,
+ unsigned long val)
+{
+ struct pt_regs *regs;
+
+ /*if (off & 3 || off >= sizeof(struct user))
+ return -EIO; */
+
+ if (off >= sizeof(struct pt_regs))
+ return 0;
+
+ regs = task_pt_regs(tsk);
+
+ if (off != offsetof(struct pt_regs, sr)) {
+ *((unsigned long *)((char *)regs + off)) = val;
+ } else {
+ /* Prevent any process from setting the SR flags and
+ * thus elevating privileges
+ */
+ }
+
+ return 0;
+}
+
+long arch_ptrace(struct task_struct *child, long request, unsigned long addr,
+ unsigned long data)
+{
+ int ret;
+ unsigned long __user *datap = (unsigned long __user *)data;
+
+ switch (request) {
+ /* read the word at location addr in the USER area. */
+ case PTRACE_PEEKUSR:
+ ret = ptrace_read_user(child, addr, datap);
+ break;
+ case PTRACE_POKEUSR:
+ ret = ptrace_write_user(child, addr, data);
+ break;
+ default:
+ ret = ptrace_request(child, request, addr, data);
+ break;
+ }
+
+ single_step_set(current);
+
+ return ret;
+}
+
+/* notification of system call entry/exit
+ * - triggered by current->work.syscall_trace
+ */
+asmlinkage long do_syscall_trace_enter(struct pt_regs *regs)
+{
+ long ret = 0;
+
+ if (test_thread_flag(TIF_SYSCALL_TRACE) &&
+ tracehook_report_syscall_entry(regs))
+ /*
+ * Tracing decided this syscall should not happen.
+ * We'll return a bogus call number to get an ENOSYS
+ * error, but leave the original number in <something>.
+ */
+ ret = -1L;
+
+/* if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
+ trace_sys_enter(regs, regs->syscallno);
+*/
+
+ /* Are these regs right??? */
+ if (unlikely(current->audit_context))
+ audit_syscall_entry(audit_arch(), regs->syscallno,
+ regs->gpr[3], regs->gpr[4],
+ regs->gpr[5], regs->gpr[6]);
+
+ return ret ? : regs->syscallno;
+
+#if 0
+/*FIXME : audit the rest of this */
+
+ /*
+ * this isn't the same as continuing with a signal, but it will do
+ * for normal use. strace only continues with a signal if the
+ * stopping signal is not SIGTRAP. -brl
+ */
+ if (current->exit_code) {
+ send_sig(current->exit_code, current, 1);
+ current->exit_code = 0;
+ }
+out:
+ /*FIXME: audit_arch isn't even defined for openrisc */
+ /*FIXME: What's with the register numbers here... makes no sense */
+ if (unlikely(current->audit_context) && !entryexit)
+ audit_syscall_entry(audit_arch(), regs->regs[2],
+ regs->regs[4], regs->regs[5],
+ regs->regs[6], regs->regs[7]);
+ /*RGD*/
+#endif
+}
+
+asmlinkage void do_syscall_trace_leave(struct pt_regs *regs)
+{
+ int step;
+
+ if (unlikely(current->audit_context))
+ audit_syscall_exit(AUDITSC_RESULT(regs->gpr[11]),
+ regs->gpr[11]);
+
+/* if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
+ trace_sys_exit(regs, regs->gprs[9]);
+*/
+
+ step = test_thread_flag(TIF_SINGLESTEP);
+ if (step || test_thread_flag(TIF_SYSCALL_TRACE))
+ tracehook_report_syscall_exit(regs, step);
+}
diff --git a/arch/openrisc/kernel/ptrace.h b/arch/openrisc/kernel/ptrace.h
new file mode 100644
index 0000000..157d6ea
--- /dev/null
+++ b/arch/openrisc/kernel/ptrace.h
@@ -0,0 +1,37 @@
+/*
+ * linux/arch/openrisc/kernel/ptrace.h
+ *
+ * Copyright (C) 2000-2003 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/ptrace.h>
+
+extern void ptrace_cancel_bpt(struct task_struct *);
+extern void ptrace_set_bpt(struct task_struct *);
+extern void ptrace_break(struct task_struct *, struct pt_regs *);
+
+/*
+ * Send SIGTRAP if we're single-stepping
+ */
+static inline void single_step_trap(struct task_struct *task)
+{
+ if (test_tsk_thread_flag (task, TIF_SINGLESTEP)){
+ ptrace_cancel_bpt(task);
+ send_sig(SIGTRAP, task, 1);
+ }
+}
+
+static inline void single_step_clear(struct task_struct *task)
+{
+ if (test_tsk_thread_flag(task, TIF_SINGLESTEP))
+ ptrace_cancel_bpt(task);
+}
+
+static inline void single_step_set(struct task_struct *task)
+{
+ if (test_tsk_thread_flag(task, TIF_SINGLESTEP))
+ ptrace_set_bpt(task);
+}
--
1.7.4.1
Marcin Slusarz
2011-07-03 19:40:09 UTC
Permalink
Post by Jonas Bonn
+ pr_debug(KERN_INFO
+ "ptrace get_next_address: pc 0x%.8lx insn. 0x%.8lx opc. 0x%.2x\n",
+ pc, insn, opc);
+
This evaluates to printk(KERN_DEBUG KERN_INFO "ptrace...", ...), which does not seem
to be ok ;). This mistake is repeated several times in this patchset...

Marcin
Arnd Bergmann
2011-07-05 15:42:29 UTC
Permalink
Post by Jonas Bonn
---
arch/openrisc/include/asm/ptrace.h | 124 ++++++++
arch/openrisc/kernel/ptrace.c | 574 ++++++++++++++++++++++++++++++++++++
arch/openrisc/kernel/ptrace.h | 37 +++
3 files changed, 735 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/ptrace.h
create mode 100644 arch/openrisc/kernel/ptrace.c
create mode 100644 arch/openrisc/kernel/ptrace.h
This patch still contains a bunch of commented out code that wants to be
cleaned up. I think you have done that in most other patches that needed
the same cleanup.

Arnd
Jonas Bonn
2011-07-05 16:05:07 UTC
Permalink
Post by Arnd Bergmann
Post by Jonas Bonn
---
arch/openrisc/include/asm/ptrace.h | 124 ++++++++
arch/openrisc/kernel/ptrace.c | 574 ++++++++++++++++++++++++++++++++++++
arch/openrisc/kernel/ptrace.h | 37 +++
3 files changed, 735 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/ptrace.h
create mode 100644 arch/openrisc/kernel/ptrace.c
create mode 100644 arch/openrisc/kernel/ptrace.h
This patch still contains a bunch of commented out code that wants to be
cleaned up. I think you have done that in most other patches that needed
the same cleanup.
Arnd
I've been looking a bit at the ptrace stuff the last couple of days.
One question that occurred to me was: is it mandatory to export the
"struct pt_regs" info in ptrace.h to userspace?

I'm not sure I like the layout of our struct pt_regs. It would be
better to lay out the registers in a way that mirrors the order that
they are saved during exception/syscall entry in order to make better
use of the cache.

Since the registers can be exported via a regset to userspace anyway
(which doesn't necessarily need to look like pt_regs), I don't see that
userspace really needs how we lay out the registers on the stack. By
not exporting pt_regs, I am free to change the layout... if it's
exported, it becomes ABI.

Comments?

/Jonas
Arnd Bergmann
2011-07-05 16:42:11 UTC
Permalink
Post by Jonas Bonn
I've been looking a bit at the ptrace stuff the last couple of days.
One question that occurred to me was: is it mandatory to export the
"struct pt_regs" info in ptrace.h to userspace?
I don't know, it certainly has been the case traditionally, but as
far as I can tell, this was mainly used by ancient gdb ports,
while newer versions just hardcode the offsets for PTRACE_PEEKUSR.
Post by Jonas Bonn
I'm not sure I like the layout of our struct pt_regs. It would be
better to lay out the registers in a way that mirrors the order that
they are saved during exception/syscall entry in order to make better
use of the cache.
Since the registers can be exported via a regset to userspace anyway
(which doesn't necessarily need to look like pt_regs), I don't see that
userspace really needs how we lay out the registers on the stack. By
not exporting pt_regs, I am free to change the layout... if it's
exported, it becomes ABI.
Yes. I think you can just change your ptrace code to not export
pt_regs through PEEKUSR but only through elf_gregset_t, which
you can define independently.

Arnd
Jonas Bonn
2011-07-02 21:15:36 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/cache.h | 29 ++
arch/openrisc/include/asm/fixmap.h | 88 ++++++
arch/openrisc/include/asm/memblock.h | 24 ++
arch/openrisc/include/asm/mmu.h | 26 ++
arch/openrisc/include/asm/mmu_context.h | 43 +++
arch/openrisc/include/asm/page.h | 112 ++++++++
arch/openrisc/include/asm/pgalloc.h | 142 ++++++++++
arch/openrisc/include/asm/pgtable.h | 449 +++++++++++++++++++++++++++++++
arch/openrisc/include/asm/tlb.h | 34 +++
arch/openrisc/include/asm/tlbflush.h | 72 +++++
arch/openrisc/include/asm/uaccess.h | 385 ++++++++++++++++++++++++++
arch/openrisc/mm/fault.c | 338 +++++++++++++++++++++++
arch/openrisc/mm/init.c | 283 +++++++++++++++++++
arch/openrisc/mm/ioremap.c | 113 ++++++++
arch/openrisc/mm/tlb.c | 191 +++++++++++++
15 files changed, 2329 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/cache.h
create mode 100644 arch/openrisc/include/asm/fixmap.h
create mode 100644 arch/openrisc/include/asm/memblock.h
create mode 100644 arch/openrisc/include/asm/mmu.h
create mode 100644 arch/openrisc/include/asm/mmu_context.h
create mode 100644 arch/openrisc/include/asm/page.h
create mode 100644 arch/openrisc/include/asm/pgalloc.h
create mode 100644 arch/openrisc/include/asm/pgtable.h
create mode 100644 arch/openrisc/include/asm/tlb.h
create mode 100644 arch/openrisc/include/asm/tlbflush.h
create mode 100644 arch/openrisc/include/asm/uaccess.h
create mode 100644 arch/openrisc/mm/fault.c
create mode 100644 arch/openrisc/mm/init.c
create mode 100644 arch/openrisc/mm/ioremap.c
create mode 100644 arch/openrisc/mm/tlb.c

diff --git a/arch/openrisc/include/asm/cache.h b/arch/openrisc/include/asm/cache.h
new file mode 100644
index 0000000..4ce7a01
--- /dev/null
+++ b/arch/openrisc/include/asm/cache.h
@@ -0,0 +1,29 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_CACHE_H
+#define __ASM_OPENRISC_CACHE_H
+
+/* FIXME: How can we replace these with values from the CPU...
+ * they shouldn't be hard-coded!
+ */
+
+#define L1_CACHE_BYTES 16
+#define L1_CACHE_SHIFT 4
+
+#endif /* __ASM_OPENRISC_CACHE_H */
diff --git a/arch/openrisc/include/asm/fixmap.h b/arch/openrisc/include/asm/fixmap.h
new file mode 100644
index 0000000..a6d27ad23
--- /dev/null
+++ b/arch/openrisc/include/asm/fixmap.h
@@ -0,0 +1,88 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_FIXMAP_H
+#define __ASM_OPENRISC_FIXMAP_H
+
+/* Why exactly do we need 2 empty pages between the top of the fixed
+ * addresses and the top of virtual memory? Something is using that
+ * memory space but not sure what right now... If you find it, leave
+ * a comment here.
+ */
+#define FIXADDR_TOP ((unsigned long) (-2*PAGE_SIZE))
+
+#include <linux/kernel.h>
+//#include <linux/threads.h>
+#include <asm/page.h>
+
+/*
+ * On OpenRISC we use these special fixed_addresses for doing ioremap
+ * early in the boot process before memory initialization is complete.
+ * This is used, in particular, by the early serial console code.
+ *
+ * It's not really 'fixmap', per se, but fits loosely into the same
+ * paradigm.
+ */
+enum fixed_addresses {
+ /*
+ * FIX_IOREMAP entries are useful for mapping physical address
+ * space before ioremap() is useable, e.g. really early in boot
+ * before kmalloc() is working.
+ */
+#define FIX_N_IOREMAPS 32
+ FIX_IOREMAP_BEGIN,
+ FIX_IOREMAP_END = FIX_IOREMAP_BEGIN + FIX_N_IOREMAPS - 1,
+ __end_of_fixed_addresses
+};
+
+#define FIXADDR_SIZE (__end_of_fixed_addresses << PAGE_SHIFT)
+/* FIXADDR_BOTTOM might be a better name here... */
+#define FIXADDR_START (FIXADDR_TOP - FIXADDR_SIZE)
+
+#define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT))
+#define __virt_to_fix(x) ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT)
+
+/*
+ * 'index to address' translation. If anyone tries to use the idx
+ * directly without tranlation, we catch the bug with a NULL-deference
+ * kernel oops. Illegal ranges of incoming indices are caught too.
+ */
+static __always_inline unsigned long fix_to_virt(const unsigned int idx)
+{
+ /*
+ * this branch gets completely eliminated after inlining,
+ * except when someone tries to use fixaddr indices in an
+ * illegal way. (such as mixing up address types or using
+ * out-of-range indices).
+ *
+ * If it doesn't get removed, the linker will complain
+ * loudly with a reasonably clear error message..
+ */
+ if (idx >= __end_of_fixed_addresses)
+ BUG();
+
+ return __fix_to_virt(idx);
+}
+
+static inline unsigned long virt_to_fix(const unsigned long vaddr)
+{
+ BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
+ return __virt_to_fix(vaddr);
+}
+
+#endif
diff --git a/arch/openrisc/include/asm/memblock.h b/arch/openrisc/include/asm/memblock.h
new file mode 100644
index 0000000..bbe5a1c
--- /dev/null
+++ b/arch/openrisc/include/asm/memblock.h
@@ -0,0 +1,24 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_MEMBLOCK_H
+#define __ASM_OPENRISC_MEMBLOCK_H
+
+/* empty */
+
+#endif /* __ASM_OPENRISC_MEMBLOCK_H */
diff --git a/arch/openrisc/include/asm/mmu.h b/arch/openrisc/include/asm/mmu.h
new file mode 100644
index 0000000..d069bc2
--- /dev/null
+++ b/arch/openrisc/include/asm/mmu.h
@@ -0,0 +1,26 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_MMU_H
+#define __ASM_OPENRISC_MMU_H
+
+#ifndef __ASSEMBLY__
+typedef unsigned long mm_context_t;
+#endif
+
+#endif
diff --git a/arch/openrisc/include/asm/mmu_context.h b/arch/openrisc/include/asm/mmu_context.h
new file mode 100644
index 0000000..a823716
--- /dev/null
+++ b/arch/openrisc/include/asm/mmu_context.h
@@ -0,0 +1,43 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_MMU_CONTEXT_H
+#define __ASM_OPENRISC_MMU_CONTEXT_H
+
+#include <asm-generic/mm_hooks.h>
+
+extern int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
+extern void destroy_context(struct mm_struct *mm);
+extern void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ struct task_struct *tsk);
+
+#define deactivate_mm(tsk,mm) do { } while (0)
+
+#define activate_mm(prev,next) switch_mm((prev),(next),NULL)
+
+/* current active pgd - this is similar to other processors pgd
+ * registers like cr3 on the i386
+ */
+
+extern volatile pgd_t *current_pgd; /* defined in arch/openrisc/mm/fault.c */
+
+static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+}
+
+#endif
diff --git a/arch/openrisc/include/asm/page.h b/arch/openrisc/include/asm/page.h
new file mode 100644
index 0000000..037b844
--- /dev/null
+++ b/arch/openrisc/include/asm/page.h
@@ -0,0 +1,112 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_PAGE_H
+#define __ASM_OPENRISC_PAGE_H
+
+
+/* PAGE_SHIFT determines the page size */
+
+#define PAGE_SHIFT 13
+#ifdef __ASSEMBLY__
+#define PAGE_SIZE (1 << PAGE_SHIFT)
+#else
+#define PAGE_SIZE (1UL << PAGE_SHIFT)
+#endif
+#define PAGE_MASK (~(PAGE_SIZE-1))
+
+#define PAGE_OFFSET 0xc0000000
+#define KERNELBASE PAGE_OFFSET
+
+/* #define ARCH_PFN_OFFSET (PAGE_OFFSET >> PAGE_SHIFT) */
+
+/* This is not necessarily the right place for this, but it's needed by
+ * drivers/of/fdt.c
+ */
+#include <asm/setup.h>
+
+#ifndef __ASSEMBLY__
+
+#define get_user_page(vaddr) __get_free_page(GFP_KERNEL)
+#define free_user_page(page, addr) free_page(addr)
+
+#define clear_page(page) memset((page), 0, PAGE_SIZE)
+#define copy_page(to,from) memcpy((to), (from), PAGE_SIZE)
+
+#define clear_user_page(page, vaddr, pg) clear_page(page)
+#define copy_user_page(to, from, vaddr, pg) copy_page(to, from)
+
+/*
+ * These are used to make use of C type-checking..
+ */
+typedef struct {
+ unsigned long pte;
+} pte_t;
+typedef struct {
+ unsigned long pgd;
+} pgd_t;
+typedef struct {
+ unsigned long pgprot;
+} pgprot_t;
+typedef struct page *pgtable_t;
+
+#define pte_val(x) ((x).pte)
+#define pgd_val(x) ((x).pgd)
+#define pgprot_val(x) ((x).pgprot)
+
+#define __pte(x) ((pte_t) { (x) } )
+#define __pgd(x) ((pgd_t) { (x) } )
+#define __pgprot(x) ((pgprot_t) { (x) } )
+
+extern unsigned long memory_start;
+extern unsigned long memory_end;
+
+#endif /* !__ASSEMBLY__ */
+
+
+#ifndef __ASSEMBLY__
+
+/* from linker script */
+
+#define __va(x) ((void *)((unsigned long)(x) + PAGE_OFFSET))
+#define __pa(x) ((unsigned long) (x) - PAGE_OFFSET)
+
+#define virt_to_pfn(kaddr) (__pa(kaddr) >> PAGE_SHIFT)
+#define pfn_to_virt(pfn) __va((pfn) << PAGE_SHIFT)
+
+#define virt_to_page(addr) (mem_map + (((unsigned long)(addr)-PAGE_OFFSET) >> PAGE_SHIFT))
+#define page_to_virt(page) ((((page) - mem_map) << PAGE_SHIFT) + PAGE_OFFSET)
+
+#define page_to_phys(page) ((dma_addr_t)page_to_pfn(page) << PAGE_SHIFT)
+
+#define pfn_valid(pfn) ((pfn) < max_mapnr)
+
+#define virt_addr_valid(kaddr) (((void *)(kaddr) >= (void *)PAGE_OFFSET) && \
+ ((void *)(kaddr) < (void *)memory_end))
+
+#endif /* __ASSEMBLY__ */
+
+
+#define VM_DATA_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \
+ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+
+#include <asm-generic/memory_model.h>
+#include <asm-generic/getorder.h>
+
+#endif /* __ASM_OPENRISC_PAGE_H */
diff --git a/arch/openrisc/include/asm/pgalloc.h b/arch/openrisc/include/asm/pgalloc.h
new file mode 100644
index 0000000..339bb52
--- /dev/null
+++ b/arch/openrisc/include/asm/pgalloc.h
@@ -0,0 +1,142 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_PGALLOC_H
+#define __ASM_OPENRISC_PGALLOC_H
+
+#include <asm/page.h>
+#include <linux/threads.h>
+#include <linux/mm.h>
+#include <linux/memblock.h>
+#include <linux/bootmem.h>
+
+extern int mem_init_done;
+
+#if 1
+#define pmd_populate_kernel(mm, pmd, pte) \
+ set_pmd(pmd, __pmd(_KERNPG_TABLE + __pa(pte)))
+
+static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd, struct page *pte)
+{
+ set_pmd(pmd, __pmd(_KERNPG_TABLE +
+ ((unsigned long)page_to_pfn(pte) <<
+ (unsigned long) PAGE_SHIFT)));
+}
+#endif
+
+#if 0
+#define pmd_populate_kernel(mm, pmd, pte) \
+ set_pmd(pmd, __pmd(_KERNPG_TABLE + __pa(pte)))
+#define pmd_populate(mm, pmd, pte) \
+ set_pmd(pmd, __pmd(__pa(pte)))
+
+#endif
+
+#if 0
+/* __PHX__ check */
+#define pmd_populate_kernel(mm, pmd, pte) pmd_set(pmd, pte)
+#define pmd_populate(mm, pmd, pte) pmd_set(pmd, page_address(pte))
+#endif
+
+
+#if 1
+/*
+ * Allocate and free page tables.
+ */
+
+static inline pgd_t *pgd_alloc(struct mm_struct *mm)
+{
+ pgd_t *ret = (pgd_t *)__get_free_page(GFP_KERNEL);
+
+ if (ret) {
+ memset(ret, 0, USER_PTRS_PER_PGD * sizeof(pgd_t));
+ memcpy(ret + USER_PTRS_PER_PGD, swapper_pg_dir + USER_PTRS_PER_PGD,
+ (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t));
+
+ }
+ return ret;
+}
+#endif
+
+#if 0
+/* __PHX__ check, this is supposed to be 2.6 style, but
+ * we use current_pgd (from mm->pgd) to load kernel pages
+ * so we need it initialized.
+ */
+extern inline pgd_t *pgd_alloc (struct mm_struct *mm)
+{
+ return (pgd_t *)get_zeroed_page(GFP_KERNEL);
+}
+#endif
+
+static inline void pgd_free (struct mm_struct *mm, pgd_t *pgd)
+{
+ free_page((unsigned long)pgd);
+}
+
+/**
+ * OK, this one's a bit tricky... ioremap can get called before memory is
+ * initialized (early serial console does this) and will want to alloc a page
+ * for its mapping. No userspace pages will ever get allocated before memory
+ * is initialized so this applies only to kernel pages. In the event that
+ * this is called before memory is initialized we allocate the page using
+ * the memblock infrastructure.
+ */
+
+static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+{
+ pte_t* pte;
+
+ if (likely(mem_init_done)) {
+ pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT);
+ } else {
+ pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+// pte = (pte_t *) __va(memblock_alloc(PAGE_SIZE, PAGE_SIZE));
+ }
+
+ if (pte)
+ clear_page(pte);
+ return pte;
+}
+
+static inline struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
+{
+ struct page *pte;
+ pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT, 0);
+ if (pte)
+ clear_page(page_address(pte));
+ return pte;
+}
+
+static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte)
+{
+ free_page((unsigned long)pte);
+}
+
+static inline void pte_free(struct mm_struct *mm, struct page *pte)
+{
+ __free_page(pte);
+}
+
+
+#define __pte_free_tlb(tlb,pte,addr) tlb_remove_page((tlb),(pte))
+#define pmd_pgtable(pmd) pmd_page(pmd)
+
+#define check_pgt_cache() do { } while (0)
+
+#endif
diff --git a/arch/openrisc/include/asm/pgtable.h b/arch/openrisc/include/asm/pgtable.h
new file mode 100644
index 0000000..3d5792a
--- /dev/null
+++ b/arch/openrisc/include/asm/pgtable.h
@@ -0,0 +1,449 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+/* or32 pgtable.h - macros and functions to manipulate page tables
+ *
+ * Based on:
+ * include/asm-cris/pgtable.h
+ */
+
+#ifndef __ASM_OPENRISC_PGTABLE_H
+#define __ASM_OPENRISC_PGTABLE_H
+
+//#include <asm-generic/4level-fixup.h>
+#include <asm-generic/pgtable-nopmd.h>
+
+#ifndef __ASSEMBLY__
+#include <asm/mmu.h>
+#include <asm/fixmap.h>
+
+/*
+ * The Linux memory management assumes a three-level page table setup. On
+ * or32, we use that, but "fold" the mid level into the top-level page
+ * table. Since the MMU TLB is software loaded through an interrupt, it
+ * supports any page table structure, so we could have used a three-level
+ * setup, but for the amounts of memory we normally use, a two-level is
+ * probably more efficient.
+ *
+ * This file contains the functions and defines necessary to modify and use
+ * the or32 page table tree.
+ */
+
+extern void paging_init(void);
+
+/* Certain architectures need to do special things when pte's
+ * within a page table are directly modified. Thus, the following
+ * hook is made available.
+ */
+#define set_pte(pteptr, pteval) ((*(pteptr)) = (pteval))
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
+/*
+ * (pmds are folded into pgds so this doesn't get actually called,
+ * but the define is needed for a generic inline function.)
+ */
+#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
+
+#define PGDIR_SHIFT (PAGE_SHIFT + (PAGE_SHIFT-2))
+#define PGDIR_SIZE (1UL << PGDIR_SHIFT)
+#define PGDIR_MASK (~(PGDIR_SIZE-1))
+
+/*
+ * entries per page directory level: we use a two-level, so
+ * we don't really have any PMD directory physically.
+ * pointers are 4 bytes so we can use the page size and
+ * divide it by 4 (shift by 2).
+ */
+#define PTRS_PER_PTE (1UL << (PAGE_SHIFT-2))
+
+#define PTRS_PER_PGD (1UL << (PAGE_SHIFT-2))
+
+/* calculate how many PGD entries a user-level program can use
+ * the first mappable virtual address is 0
+ * (TASK_SIZE is the maximum virtual address space)
+ */
+
+#define USER_PTRS_PER_PGD (TASK_SIZE/PGDIR_SIZE)
+#define FIRST_USER_ADDRESS 0
+
+/*
+ * Kernels own virtual memory area.
+ */
+
+/*
+ * The size and location of the vmalloc area are chosen so that modules
+ * placed in this area aren't more than a 28-bit signed offset from any
+ * kernel functions that they may need. This greatly simplifies handling
+ * of the relocations for l.j and l.jal instructions as we don't need to
+ * introduce any trampolines for reaching "distant" code.
+ *
+ * 64 MB of vmalloc area is comparable to what's available on other arches.
+ */
+
+#define VMALLOC_START (PAGE_OFFSET-0x04000000)
+#define VMALLOC_END (PAGE_OFFSET)
+#define VMALLOC_VMADDR(x) ((unsigned long)(x))
+
+/* Define some higher level generic page attributes.
+ *
+ * If you change _PAGE_CI definition be sure to change it in
+ * io.h for ioremap_nocache() too.
+ */
+
+/*
+ * An OR32 PTE looks like this:
+ *
+ * | 31 ... 10 | 9 | 8 ... 6 | 5 | 4 | 3 | 2 | 1 | 0 |
+ * Phys pg.num L PP Index D A WOM WBC CI CC
+ *
+ * L : link
+ * PPI: Page protection index
+ * D : Dirty
+ * A : Accessed
+ * WOM: Weakly ordered memory
+ * WBC: Write-back cache
+ * CI : Cache inhibit
+ * CC : Cache coherent
+ *
+ * The protection bits below should correspond to the layout of the actual
+ * PTE as per above
+ */
+
+#define _PAGE_CC 0x001 /* software: pte contains a translation */
+#define _PAGE_CI 0x002 /* cache inhibit */
+#define _PAGE_WBC 0x004 /* write back cache */
+#define _PAGE_FILE 0x004 /* set: pagecache, unset: swap (when !PRESENT) */
+#define _PAGE_WOM 0x008 /* weakly ordered memory */
+
+#define _PAGE_A 0x010 /* accessed */
+#define _PAGE_D 0x020 /* dirty */
+#define _PAGE_URE 0x040 /* user read enable */
+#define _PAGE_UWE 0x080 /* user write enable */
+
+#define _PAGE_SRE 0x100 /* superuser read enable */
+#define _PAGE_SWE 0x200 /* superuser write enable */
+#define _PAGE_EXEC 0x400 /* software: page is executable */
+#define _PAGE_U_SHARED 0x800 /* software: page is shared in user space */
+
+/* 0x001 is cache coherency bit, which should always be set to
+ * 1 - for SMP (when we support it)
+ * 0 - otherwise
+ *
+ * we just reuse this bit in software for _PAGE_PRESENT and
+ * force it to 0 when loading it into TLB.
+ */
+#define _PAGE_PRESENT _PAGE_CC
+#define _PAGE_USER _PAGE_URE
+#define _PAGE_WRITE (_PAGE_UWE | _PAGE_SWE)
+#define _PAGE_DIRTY _PAGE_D
+#define _PAGE_ACCESSED _PAGE_A
+#define _PAGE_NO_CACHE _PAGE_CI
+#define _PAGE_SHARED _PAGE_U_SHARED
+#define _PAGE_READ (_PAGE_URE | _PAGE_SRE)
+
+#define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _PAGE_BASE (_PAGE_PRESENT | _PAGE_ACCESSED)
+#define _PAGE_ALL (_PAGE_PRESENT | _PAGE_ACCESSED)
+#define _KERNPG_TABLE (_PAGE_BASE | _PAGE_SRE | _PAGE_SWE | _PAGE_ACCESSED | _PAGE_DIRTY)
+
+#define PAGE_NONE __pgprot(_PAGE_ALL)
+#define PAGE_READONLY __pgprot(_PAGE_ALL | _PAGE_URE | _PAGE_SRE )
+#define PAGE_READONLY_X __pgprot(_PAGE_ALL | _PAGE_URE | _PAGE_SRE | _PAGE_EXEC)
+#define PAGE_SHARED __pgprot(_PAGE_ALL | _PAGE_URE | _PAGE_SRE | _PAGE_UWE | _PAGE_SWE | _PAGE_SHARED)
+#define PAGE_SHARED_X __pgprot(_PAGE_ALL | _PAGE_URE | _PAGE_SRE | _PAGE_UWE | _PAGE_SWE | _PAGE_SHARED | _PAGE_EXEC)
+#define PAGE_COPY __pgprot(_PAGE_ALL | _PAGE_URE | _PAGE_SRE )
+#define PAGE_COPY_X __pgprot(_PAGE_ALL | _PAGE_URE | _PAGE_SRE | _PAGE_EXEC)
+
+#define PAGE_KERNEL __pgprot(_PAGE_ALL | _PAGE_SRE | _PAGE_SWE | _PAGE_SHARED | _PAGE_DIRTY | _PAGE_EXEC)
+#define PAGE_KERNEL_RO __pgprot(_PAGE_ALL | _PAGE_SRE | _PAGE_SHARED | _PAGE_DIRTY | _PAGE_EXEC)
+#define PAGE_KERNEL_NOCACHE __pgprot(_PAGE_ALL | _PAGE_SRE | _PAGE_SWE | _PAGE_SHARED | _PAGE_DIRTY | _PAGE_EXEC | _PAGE_CI)
+
+#define __P000 PAGE_NONE
+#define __P001 PAGE_READONLY_X
+#define __P010 PAGE_COPY
+#define __P011 PAGE_COPY_X
+#define __P100 PAGE_READONLY
+#define __P101 PAGE_READONLY_X
+#define __P110 PAGE_COPY
+#define __P111 PAGE_COPY_X
+
+#define __S000 PAGE_NONE
+#define __S001 PAGE_READONLY_X
+#define __S010 PAGE_SHARED
+#define __S011 PAGE_SHARED_X
+#define __S100 PAGE_READONLY
+#define __S101 PAGE_READONLY_X
+#define __S110 PAGE_SHARED
+#define __S111 PAGE_SHARED_X
+
+/* zero page used for uninitialized stuff */
+extern unsigned long empty_zero_page[2048];
+#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
+
+/* number of bits that fit into a memory pointer */
+#define BITS_PER_PTR (8*sizeof(unsigned long))
+
+/* to align the pointer to a pointer address */
+#define PTR_MASK (~(sizeof(void*)-1))
+
+/* sizeof(void*)==1<<SIZEOF_PTR_LOG2 */
+/* 64-bit machines, beware! SRB. */
+#define SIZEOF_PTR_LOG2 2
+
+/* to find an entry in a page-table */
+#define PAGE_PTR(address) \
+((unsigned long)(address)>>(PAGE_SHIFT-SIZEOF_PTR_LOG2)&PTR_MASK&~PAGE_MASK)
+
+/* to set the page-dir */
+#define SET_PAGE_DIR(tsk,pgdir)
+
+#define pte_none(x) (!pte_val(x))
+#define pte_present(x) (pte_val(x) & _PAGE_PRESENT)
+#define pte_clear(mm,addr,xp) do { pte_val(*(xp)) = 0; } while (0)
+
+#define pmd_none(x) (!pmd_val(x))
+#define pmd_bad(x) ((pmd_val(x) & (~PAGE_MASK)) != _KERNPG_TABLE)
+#define pmd_present(x) (pmd_val(x) & _PAGE_PRESENT)
+#define pmd_clear(xp) do { pmd_val(*(xp)) = 0; } while (0)
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+
+static inline int pte_read(pte_t pte) { return pte_val(pte) & _PAGE_READ; }
+static inline int pte_write(pte_t pte) { return pte_val(pte) & _PAGE_WRITE; }
+static inline int pte_exec(pte_t pte) { return pte_val(pte) & _PAGE_EXEC; }
+static inline int pte_dirty(pte_t pte) { return pte_val(pte) & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte) { return pte_val(pte) & _PAGE_ACCESSED; }
+static inline int pte_file(pte_t pte) { return pte_val(pte) & _PAGE_FILE; }
+static inline int pte_special(pte_t pte) { return 0; }
+static inline pte_t pte_mkspecial(pte_t pte) { return pte; }
+
+static inline pte_t pte_wrprotect(pte_t pte)
+{
+ pte_val(pte) &= ~(_PAGE_WRITE);
+ return pte;
+}
+
+static inline pte_t pte_rdprotect(pte_t pte)
+{
+ pte_val(pte) &= ~(_PAGE_READ);
+ return pte;
+}
+
+static inline pte_t pte_exprotect(pte_t pte)
+{
+ pte_val(pte) &= ~(_PAGE_EXEC);
+ return pte;
+}
+
+static inline pte_t pte_mkclean(pte_t pte)
+{
+ pte_val(pte) &= ~(_PAGE_DIRTY);
+ return pte;
+}
+
+static inline pte_t pte_mkold(pte_t pte)
+{
+ pte_val(pte) &= ~(_PAGE_ACCESSED);
+ return pte;
+}
+
+static inline pte_t pte_mkwrite(pte_t pte)
+{
+ pte_val(pte) |= _PAGE_WRITE;
+ return pte;
+}
+
+static inline pte_t pte_mkread(pte_t pte)
+{
+ pte_val(pte) |= _PAGE_READ;
+ return pte;
+}
+
+static inline pte_t pte_mkexec(pte_t pte)
+{
+ pte_val(pte) |= _PAGE_EXEC;
+ return pte;
+}
+
+static inline pte_t pte_mkdirty(pte_t pte)
+{
+ pte_val(pte) |= _PAGE_DIRTY;
+ return pte;
+}
+
+static inline pte_t pte_mkyoung(pte_t pte)
+{
+ pte_val(pte) |= _PAGE_ACCESSED;
+ return pte;
+}
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ */
+
+/* What actually goes as arguments to the various functions is less than
+ * obvious, but a rule of thumb is that struct page's goes as struct page *,
+ * really physical DRAM addresses are unsigned long's, and DRAM "virtual"
+ * addresses (the 0xc0xxxxxx's) goes as void *'s.
+ */
+
+static inline pte_t __mk_pte(void * page, pgprot_t pgprot)
+{
+ pte_t pte;
+ /* the PTE needs a physical address */
+ pte_val(pte) = __pa(page) | pgprot_val(pgprot);
+ return pte;
+}
+
+#define mk_pte(page, pgprot) __mk_pte(page_address(page), (pgprot))
+
+#define mk_pte_phys(physpage, pgprot) \
+({ \
+ pte_t __pte; \
+ \
+ pte_val(__pte) = (physpage) + pgprot_val(pgprot); \
+ __pte; \
+})
+
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{ pte_val(pte) = (pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot); return pte; }
+
+
+/* pte_val refers to a page in the 0x0xxxxxxx physical DRAM interval
+ * __pte_page(pte_val) refers to the "virtual" DRAM interval
+ * pte_pagenr refers to the page-number counted starting from the virtual DRAM start
+ */
+
+static inline unsigned long __pte_page(pte_t pte)
+{
+ /* the PTE contains a physical address */
+ return (unsigned long)__va(pte_val(pte) & PAGE_MASK);
+}
+
+#define pte_pagenr(pte) ((__pte_page(pte) - PAGE_OFFSET) >> PAGE_SHIFT)
+
+/* permanent address of a page */
+
+#define __page_address(page) (PAGE_OFFSET + (((page) - mem_map) << PAGE_SHIFT))
+#define pte_page(pte) (mem_map+pte_pagenr(pte))
+
+/* only the pte's themselves need to point to physical DRAM (see above)
+ * the pagetable links are purely handled within the kernel SW and thus
+ * don't need the __pa and __va transformations.
+ */
+
+extern inline void pmd_set(pmd_t * pmdp, pte_t * ptep)
+{ pmd_val(*pmdp) = _KERNPG_TABLE | (unsigned long) ptep; }
+
+#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
+#define pmd_page_kernel(pmd) ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+
+/* to find an entry in a page-table-directory. */
+#define pgd_index(address) ((address >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
+
+#define __pgd_offset(address) pgd_index(address)
+
+#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
+
+/* to find an entry in a kernel page-table-directory */
+#define pgd_offset_k(address) pgd_offset(&init_mm, address)
+
+#define __pmd_offset(address) \
+ (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
+
+/*
+ * the pte page can be thought of an array like this: pte_t[PTRS_PER_PTE]
+ *
+ * this macro returns the index of the entry in the pte page which would
+ * control the given virtual address
+ */
+#define __pte_offset(address) \
+ (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset_kernel(dir, address) \
+ ((pte_t *) pmd_page_kernel(*(dir)) + __pte_offset(address))
+#define pte_offset_map(dir, address) \
+ ((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))
+#define pte_offset_map_nested(dir, address) \
+ pte_offset_map(dir, address)
+
+#define pte_unmap(pte) do { } while (0)
+#define pte_unmap_nested(pte) do { } while (0)
+#define pte_pfn(x) ((unsigned long)(((x).pte)) >> PAGE_SHIFT)
+#define pfn_pte(pfn, prot) __pte((((pfn) << PAGE_SHIFT)) | pgprot_val(prot))
+
+#define pte_ERROR(e) \
+ printk("%s:%d: bad pte %p(%08lx).\n", __FILE__, __LINE__, &(e), pte_val(e))
+#define pgd_ERROR(e) \
+ printk("%s:%d: bad pgd %p(%08lx).\n", __FILE__, __LINE__, &(e), pgd_val(e))
+
+extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; /* defined in head.S */
+
+/*
+ * or32 doesn't have any external MMU info: the kernel page
+ * tables contain all the necessary information.
+ *
+ * Actually I am not sure on what this could be used for.
+ */
+static inline void update_mmu_cache(struct vm_area_struct * vma,
+ unsigned long address, pte_t * pte)
+{
+}
+
+/* __PHX__ FIXME, SWAP, this probably doesn't work */
+
+/* Encode and de-code a swap entry (must be !pte_none(e) && !pte_present(e)) */
+/* Since the PAGE_PRESENT bit is bit 4, we can use the bits above */
+
+#define __swp_type(x) (((x).val >> 5) & 0x7f)
+#define __swp_offset(x) ((x).val >> 12)
+#define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 5) | ((offset) << 12) })
+#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) })
+#define __swp_entry_to_pte(x) ((pte_t) { (x).val })
+
+/* Encode and decode a nonlinear file mapping entry */
+
+#define PTE_FILE_MAX_BITS 26
+#define pte_to_pgoff(x) (pte_val(x) >> 6)
+#define pgoff_to_pte(x) __pte(((x) << 6) | _PAGE_FILE)
+
+#define kern_addr_valid(addr) (1)
+
+#define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \
+ remap_pfn_range(vma, vaddr, pfn, size, prot)
+
+#include <asm-generic/pgtable.h>
+
+void *consistent_alloc(int gfp, size_t size, dma_addr_t *dma_handle);
+void consistent_free(void *vaddr);
+void consistent_sync(void *vaddr, size_t size, int direction);
+void consistent_sync_page(struct page *page, unsigned long offset,
+ size_t size, int direction);
+
+/*
+ * No page table caches to initialise
+ */
+#define pgtable_cache_init() do { } while (0)
+#define io_remap_page_range remap_page_range
+
+typedef pte_t *pte_addr_t;
+
+#endif /* __ASSEMBLY__ */
+#endif /* __ASM_OPENRISC_PGTABLE_H */
diff --git a/arch/openrisc/include/asm/tlb.h b/arch/openrisc/include/asm/tlb.h
new file mode 100644
index 0000000..fa4376a
--- /dev/null
+++ b/arch/openrisc/include/asm/tlb.h
@@ -0,0 +1,34 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_TLB_H__
+#define __ASM_OPENRISC_TLB_H__
+
+/*
+ * or32 doesn't need any special per-pte or
+ * per-vma handling..
+ */
+#define tlb_start_vma(tlb, vma) do { } while (0)
+#define tlb_end_vma(tlb, vma) do { } while (0)
+#define __tlb_remove_tlb_entry(tlb, ptep, address) do { } while (0)
+
+#define tlb_flush(tlb) flush_tlb_mm((tlb)->mm)
+#include <linux/pagemap.h>
+#include <asm-generic/tlb.h>
+
+#endif /* __ASM_OPENRISC_TLB_H__ */
diff --git a/arch/openrisc/include/asm/tlbflush.h b/arch/openrisc/include/asm/tlbflush.h
new file mode 100644
index 0000000..91adaf9
--- /dev/null
+++ b/arch/openrisc/include/asm/tlbflush.h
@@ -0,0 +1,72 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_TLBFLUSH_H
+#define __ASM_OPENRISC_TLBFLUSH_H
+
+#include <linux/mm.h>
+#include <asm/processor.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/current.h>
+/*#include <asm/spr_defs.h>*/
+#include <linux/sched.h>
+
+/*
+ * TLB flushing (implemented in arch/or32/mm/tlb.c):
+ *
+ * - flush_tlb() flushes the current mm struct TLBs
+ * - flush_tlb_all() flushes all processes TLBs
+ * - flush_tlb_mm(mm) flushes the specified mm context TLB's
+ * - flush_tlb_page(vma, vmaddr) flushes one page
+ * - flush_tlb_range(mm, start, end) flushes a range of pages
+ *
+ */
+
+void flush_tlb_all(void);
+void flush_tlb_mm(struct mm_struct* mm);
+void flush_tlb_page(struct vm_area_struct* vma, unsigned long addr);
+void flush_tlb_range(struct vm_area_struct* vma,
+ unsigned long start,
+ unsigned long end);
+
+
+#if 0
+static inline void flush_tlb_pgtables(struct mm_struct *mm,
+ unsigned long start, unsigned long end)
+{
+ /* OR32 does not keep any page table caches in TLB */
+}
+
+#endif
+
+static inline void flush_tlb(void)
+{
+ flush_tlb_mm(current->mm);
+}
+
+static inline void flush_tlb_kernel_range(unsigned long start,
+ unsigned long end)
+{
+ flush_tlb_range(NULL, start, end);
+}
+
+
+
+
+#endif /* __ASM_OPENRISC_TLBFLUSH_H */
diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h
new file mode 100644
index 0000000..a163341
--- /dev/null
+++ b/arch/openrisc/include/asm/uaccess.h
@@ -0,0 +1,385 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_UACCESS_H
+#define __ASM_OPENRISC_UACCESS_H
+
+/*
+ * User space memory access functions
+ */
+#include <linux/errno.h>
+#include <linux/thread_info.h>
+#include <linux/prefetch.h>
+#include <linux/string.h>
+#include <asm/page.h>
+#include <asm/thread_info.h>
+
+#define VERIFY_READ 0
+#define VERIFY_WRITE 1
+
+/*
+ * The fs value determines whether argument validity checking should be
+ * performed or not. If get_fs() == USER_DS, checking is performed, with
+ * get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons, these macros are grossly misnamed.
+ */
+
+//#define MAKE_MM_SEG(s) ((mm_segment_t) s)
+
+/* addr_limit is the maximum accessible address for the task. we misuse
+ * the KERNEL_DS and USER_DS values to both assign and compare the
+ * addr_limit values through the equally misnamed get/set_fs macros.
+ * (see above)
+ */
+
+#define KERNEL_DS (~0UL)
+#define get_ds() (KERNEL_DS)
+
+#define USER_DS (TASK_SIZE)
+#define get_fs() (current_thread_info()->addr_limit)
+#define set_fs(x) (current_thread_info()->addr_limit = (x))
+
+#define segment_eq(a,b) ((a) == (b))
+
+#if 0
+
+#define __kernel_ok (segment_eq(get_fs(), KERNEL_DS))
+#define __user_ok(addr,size) (((size) <= TASK_SIZE)&&((addr) <= TASK_SIZE-(size)))
+#endif
+
+/* Ensure that the range from addr to addr+size is all within the process'
+ * address space
+ */
+//#define __range_ok(addr,size) (size <= TASK_SIZE && addr <= (TASK_SIZE-size))
+#define __range_ok(addr,size) (size <= get_fs() && addr <= (get_fs()-size))
+
+/* Ensure that addr is below task's addr_limit */
+#define __addr_ok(addr) ((unsigned long) addr < get_fs())
+
+//#define __access_ok(addr,size) (__kernel_ok || __user_ok((addr),(size)))
+//#define __access_ok(addr,size) (__range_ok(addr,size))
+//#define access_ok(type,addr,size) __access_ok((unsigned long)(addr),(size))
+#define access_ok(type,addr,size) __range_ok((unsigned long)addr,(unsigned long)size)
+
+
+/*
+ * The exception table consists of pairs of addresses: the first is the
+ * address of an instruction that is allowed to fault, and the second is
+ * the address at which the program should continue. No registers are
+ * modified, so it is entirely up to the continuation code to figure out
+ * what to do.
+ *
+ * All the routines below use bits of fixup code that are out of line
+ * with the main instruction path. This means when everything is well,
+ * we don't even have to jump over them. Further, they do not intrude
+ * on our cache or tlb entries.
+ */
+
+struct exception_table_entry
+{
+ unsigned long insn, fixup;
+};
+
+/* Returns 0 if exception not found and fixup otherwise. */
+extern unsigned long search_exception_table(unsigned long);
+extern void sort_exception_table(void);
+
+/*
+ * These are the main single-value transfer routines. They automatically
+ * use the right size if we just have the right pointer type.
+ *
+ * This gets kind of ugly. We want to return _two_ values in "get_user()"
+ * and yet we don't want to do any pointers, because that is too much
+ * of a performance impact. Thus we have a few rather ugly macros here,
+ * and hide all the uglyness from the user.
+ *
+ * The "__xxx" versions of the user access functions are versions that
+ * do not verify the address space, that must have been done previously
+ * with a separate "access_ok()" call (this is used when we do multiple
+ * accesses to the same area of user memory).
+ *
+ * As we use the same address space for kernel and user data on the
+ * PowerPC, we can just do these as direct assignments. (Of course, the
+ * exception handling means that it's no longer "just"...)
+ */
+#define get_user(x,ptr) \
+ __get_user_check((x),(ptr),sizeof(*(ptr)))
+#define put_user(x,ptr) \
+ __put_user_check((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
+
+#define __get_user(x,ptr) \
+ __get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+#define __put_user(x,ptr) \
+ __put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
+
+extern long __put_user_bad(void);
+
+#define __put_user_nocheck(x,ptr,size) \
+({ \
+ long __pu_err; \
+ __put_user_size((x),(ptr),(size),__pu_err); \
+ __pu_err; \
+})
+
+#define __put_user_check(x,ptr,size) \
+({ \
+ long __pu_err = -EFAULT; \
+ __typeof__(*(ptr)) *__pu_addr = (ptr); \
+ if (access_ok(VERIFY_WRITE,__pu_addr,size)) \
+ __put_user_size((x),__pu_addr,(size),__pu_err); \
+ __pu_err; \
+})
+
+#define __put_user_size(x,ptr,size,retval) \
+do { \
+ retval = 0; \
+ switch (size) { \
+ case 1: __put_user_asm(x,ptr,retval,"l.sb"); break; \
+ case 2: __put_user_asm(x,ptr,retval,"l.sh"); break; \
+ case 4: __put_user_asm(x,ptr,retval,"l.sw"); break; \
+ case 8: __put_user_asm2(x,ptr,retval); break; \
+ default: __put_user_bad(); \
+ } \
+} while (0)
+
+struct __large_struct { unsigned long buf[100]; };
+#define __m(x) (*(struct __large_struct *)(x))
+
+/*
+ * We don't tell gcc that we are accessing memory, but this is OK
+ * because we do not write to any memory gcc knows about, so there
+ * are no aliasing issues.
+ */
+#define __put_user_asm(x, addr, err, op) \
+ __asm__ __volatile__( \
+ "1: "op" 0(%2),%1\n" \
+ "2:\n" \
+ ".section .fixup,\"ax\"\n" \
+ "3: l.addi %0,r0,%3\n" \
+ " l.j 2b\n" \
+ " l.nop \n" \
+ ".previous\n" \
+ ".section __ex_table,\"a\"\n" \
+ " .align 2\n" \
+ " .long 1b,3b\n" \
+ ".previous" \
+ : "=r"(err) \
+ : "r"(x), "r"(addr), "i"(-EFAULT), "0"(err))
+
+#define __put_user_asm2(x, addr, err) \
+ __asm__ __volatile__( \
+ "1: l.sw 0(%2),%1\n" \
+ "2: l.sw 4(%2),%H1\n" \
+ "3:\n" \
+ ".section .fixup,\"ax\"\n" \
+ "4: l.addi %0,r0,%3\n" \
+ " l.j 3b\n" \
+ " l.nop \n" \
+ ".previous\n" \
+ ".section __ex_table,\"a\"\n" \
+ " .align 2\n" \
+ " .long 1b,4b\n" \
+ " .long 2b,4b\n" \
+ ".previous" \
+ : "=r"(err) \
+ : "r"(x), "r"(addr), "i"(-EFAULT), "0"(err))
+
+#define __get_user_nocheck(x,ptr,size) \
+({ \
+ long __gu_err, __gu_val; \
+ __get_user_size(__gu_val,(ptr),(size),__gu_err); \
+ (x) = (__typeof__(*(ptr)))__gu_val; \
+ __gu_err; \
+})
+
+#define __get_user_check(x,ptr,size) \
+({ \
+ long __gu_err = -EFAULT, __gu_val = 0; \
+ const __typeof__(*(ptr)) *__gu_addr = (ptr); \
+ if (access_ok(VERIFY_READ,__gu_addr,size)) \
+ __get_user_size(__gu_val,__gu_addr,(size),__gu_err); \
+ (x) = (__typeof__(*(ptr)))__gu_val; \
+ __gu_err; \
+})
+
+extern long __get_user_bad(void);
+
+#define __get_user_size(x,ptr,size,retval) \
+do { \
+ retval = 0; \
+ switch (size) { \
+ case 1: __get_user_asm(x,ptr,retval,"l.lbz"); break; \
+ case 2: __get_user_asm(x,ptr,retval,"l.lhz"); break; \
+ case 4: __get_user_asm(x,ptr,retval,"l.lwz"); break; \
+ case 8: __get_user_asm2(x, ptr, retval); \
+ default: (x) = __get_user_bad(); \
+ } \
+} while (0)
+
+#define __get_user_asm(x, addr, err, op) \
+ __asm__ __volatile__( \
+ "1: "op" %1,0(%2)\n" \
+ "2:\n" \
+ ".section .fixup,\"ax\"\n" \
+ "3: l.addi %0,r0,%3\n" \
+ " l.addi %1,r0,0\n" \
+ " l.j 2b\n" \
+ " l.nop \n" \
+ ".previous\n" \
+ ".section __ex_table,\"a\"\n" \
+ " .align 2\n" \
+ " .long 1b,3b\n" \
+ ".previous" \
+ : "=r"(err), "=r"(x) \
+ : "r"(addr), "i"(-EFAULT), "0"(err))
+
+#define __get_user_asm2(x, addr, err) \
+ __asm__ __volatile__( \
+ "1: l.lwz %1,0(%2)\n" \
+ "2: l.lwz %H1,4(%2)\n" \
+ "3:\n" \
+ ".section .fixup,\"ax\"\n" \
+ "4: l.addi %0,r0,%3\n" \
+ " l.addi %1,r0,0\n" \
+ " l.addi %H1,r0,0\n" \
+ " l.j 3b\n" \
+ " l.nop \n" \
+ ".previous\n" \
+ ".section __ex_table,\"a\"\n" \
+ " .align 2\n" \
+ " .long 1b,4b\n" \
+ " .long 2b,4b\n" \
+ ".previous" \
+ : "=r"(err), "=&r"(x) \
+ : "r"(addr), "i"(-EFAULT), "0"(err))
+
+/* more complex routines */
+
+extern unsigned long __must_check
+__copy_tofrom_user(void *to, const void *from, unsigned long size);
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+#define __copy_from_user(to, from, size) \
+ __copy_tofrom_user(to, from, size)
+#define __copy_to_user(to, from, size) \
+ __copy_tofrom_user(to, from, size)
+
+#define __copy_to_user_inatomic __copy_to_user
+#define __copy_from_user_inatomic __copy_from_user
+
+static inline unsigned long
+copy_from_user(void *to, const void *from, unsigned long n)
+{
+ unsigned long over;
+
+ if (access_ok(VERIFY_READ, from, n))
+ return __copy_tofrom_user(to, from, n);
+ if ((unsigned long)from < TASK_SIZE) {
+ over = (unsigned long)from + n - TASK_SIZE;
+ return __copy_tofrom_user(to, from, n - over) + over;
+ }
+ return n;
+}
+
+static inline unsigned long
+copy_to_user(void *to, const void *from, unsigned long n)
+{
+ unsigned long over;
+
+ if (access_ok(VERIFY_WRITE, to, n))
+ return __copy_tofrom_user(to, from, n);
+ if ((unsigned long)to < TASK_SIZE) {
+ over = (unsigned long)to + n - TASK_SIZE;
+ return __copy_tofrom_user(to, from, n - over) + over;
+ }
+ return n;
+}
+
+
+extern unsigned long __clear_user(void *addr, unsigned long size);
+
+static inline __must_check unsigned long
+clear_user(void *addr, unsigned long size)
+{
+
+ if (access_ok(VERIFY_WRITE, addr, size))
+ return __clear_user(addr, size);
+ if ((unsigned long)addr < TASK_SIZE) {
+ unsigned long over = (unsigned long)addr + size - TASK_SIZE;
+ return __clear_user(addr, size - over) + over;
+ }
+ return size;
+}
+
+extern int __strncpy_from_user(char *dst, const char *src, long count);
+
+static inline long
+strncpy_from_user(char *dst, const char *src, long count)
+{
+ if (access_ok(VERIFY_READ, src, 1))
+ return __strncpy_from_user(dst, src, count);
+ return -EFAULT;
+}
+
+/*
+ * Return the size of a string (including the ending 0)
+ *
+ * Return 0 for error
+ */
+
+extern int __strnlen_user(const char *str, long len, unsigned long top);
+
+/*
+ * Returns the length of the string at str (including the null byte),
+ * or 0 if we hit a page we can't access,
+ * or something > len if we didn't find a null byte.
+ *
+ * The `top' parameter to __strnlen_user is to make sure that
+ * we can never overflow from the user area into kernel space.
+ */
+static inline long strnlen_user(const char __user *str, long len)
+{
+// unsigned long top = __kernel_ok? ~0UL: TASK_SIZE - 1;
+
+ unsigned long top = (unsigned long) get_fs();
+ unsigned long res = 0;
+
+ if (__addr_ok(str))
+ res = __strnlen_user(str, len, top);
+
+ return res;
+}
+
+//#define strlen_user(str) strnlen_user(str, 0x7ffffffe)
+#define strlen_user(str) strnlen_user(str, TASK_SIZE-1)
+
+#endif /* __ASM_OPENRISC_UACCESS_H */
diff --git a/arch/openrisc/mm/fault.c b/arch/openrisc/mm/fault.c
new file mode 100644
index 0000000..a5dce82
--- /dev/null
+++ b/arch/openrisc/mm/fault.c
@@ -0,0 +1,338 @@
+/*
+ * OpenRISC fault.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+
+#include <asm/uaccess.h>
+#include <asm/siginfo.h>
+#include <asm/signal.h>
+
+#define NUM_TLB_ENTRIES 64
+#define TLB_OFFSET(add) (((add) >> PAGE_SHIFT) & (NUM_TLB_ENTRIES-1))
+
+unsigned long pte_misses; /* updated by do_page_fault() */
+unsigned long pte_errors; /* updated by do_page_fault() */
+
+/* __PHX__ :: - check the vmalloc_fault in do_page_fault()
+ * - also look into include/asm-or32/mmu_context.h
+ */
+volatile pgd_t *current_pgd;
+
+extern void die(char *, struct pt_regs *, long);
+
+/*
+ * This routine handles page faults. It determines the address,
+ * and the problem, and then passes it off to one of the appropriate
+ * routines.
+ *
+ * If this routine detects a bad access, it returns 1, otherwise it
+ * returns 0.
+ */
+
+asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long address,
+ unsigned long vector, int write_acc)
+{
+ struct task_struct *tsk;
+ struct mm_struct *mm;
+ struct vm_area_struct *vma;
+ siginfo_t info;
+ int fault;
+
+ tsk = current;
+
+ /*
+ * We fault-in kernel-space virtual memory on-demand. The
+ * 'reference' page table is init_mm.pgd.
+ *
+ * NOTE! We MUST NOT take any locks for this case. We may
+ * be in an interrupt or a critical region, and should
+ * only copy the information from the master page table,
+ * nothing more.
+ *
+ * NOTE2: This is done so that, when updating the vmalloc
+ * mappings we don't have to walk all processes pgdirs and
+ * add the high mappings all at once. Instead we do it as they
+ * are used. However vmalloc'ed page entries have the PAGE_GLOBAL
+ * bit set so sometimes the TLB can use a lingering entry.
+ *
+ * This verifies that the fault happens in kernel space
+ * and that the fault was not a protection error.
+ */
+
+ if (address >= VMALLOC_START &&
+ (vector != 0x300 && vector != 0x400) &&
+ !user_mode(regs))
+ goto vmalloc_fault;
+
+ /* If exceptions were enabled, we can reenable them here */
+ if (user_mode(regs)) {
+ /* Exception was in userspace: reenable interrupts */
+ local_irq_enable();
+ } else {
+ /* If exception was in a syscall, then IRQ's may have
+ * been enabled or disabled. If they were enabled,
+ * reenable them.
+ */
+ if (regs->sr && (SPR_SR_IEE | SPR_SR_TEE))
+ local_irq_enable();
+ }
+
+ mm = tsk->mm;
+ info.si_code = SEGV_MAPERR;
+
+ /*
+ * If we're in an interrupt or have no user
+ * context, we must not take the fault..
+ */
+
+ if (in_interrupt() || !mm)
+ goto no_context;
+
+ down_read(&mm->mmap_sem);
+ vma = find_vma(mm, address);
+
+ if (!vma)
+ goto bad_area;
+
+ if (vma->vm_start <= address)
+ goto good_area;
+
+ if (!(vma->vm_flags & VM_GROWSDOWN))
+ goto bad_area;
+
+ if (user_mode(regs)) {
+ /*
+ * accessing the stack below usp is always a bug.
+ * we get page-aligned addresses so we can only check
+ * if we're within a page from usp, but that might be
+ * enough to catch brutal errors at least.
+ */
+ if (address + PAGE_SIZE < regs->sp)
+ goto bad_area;
+ }
+ if (expand_stack(vma, address))
+ goto bad_area;
+
+ /*
+ * Ok, we have a good vm_area for this memory access, so
+ * we can handle it..
+ */
+
+good_area:
+ info.si_code = SEGV_ACCERR;
+
+ /* first do some preliminary protection checks */
+
+ if (write_acc) {
+ if (!(vma->vm_flags & VM_WRITE))
+ goto bad_area;
+ } else {
+ /* not present */
+ if (!(vma->vm_flags & (VM_READ | VM_EXEC)))
+ goto bad_area;
+ }
+
+ /* are we trying to execute nonexecutable area */
+ if ((vector == 0x400) && !(vma->vm_page_prot.pgprot & _PAGE_EXEC))
+ goto bad_area;
+
+ /*
+ * If for any reason at all we couldn't handle the fault,
+ * make sure we exit gracefully rather than endlessly redo
+ * the fault.
+ */
+
+ fault = handle_mm_fault(mm, vma, address, write_acc);
+ if (unlikely(fault & VM_FAULT_ERROR)) {
+ if (fault & VM_FAULT_OOM)
+ goto out_of_memory;
+ else if (fault & VM_FAULT_SIGBUS)
+ goto do_sigbus;
+ BUG();
+ }
+ /*RGD modeled on Cris */
+ if (fault & VM_FAULT_MAJOR)
+ tsk->maj_flt++;
+ else
+ tsk->min_flt++;
+
+ up_read(&mm->mmap_sem);
+ return;
+
+ /*
+ * Something tried to access memory that isn't in our memory map..
+ * Fix it, but check if it's kernel or user first..
+ */
+
+bad_area:
+ up_read(&mm->mmap_sem);
+
+bad_area_nosemaphore:
+
+ /* User mode accesses just cause a SIGSEGV */
+
+ if (user_mode(regs)) {
+ info.si_signo = SIGSEGV;
+ info.si_errno = 0;
+ /* info.si_code has been set above */
+ info.si_addr = (void *)address;
+ force_sig_info(SIGSEGV, &info, tsk);
+ return;
+ }
+
+no_context:
+
+ /* Are we prepared to handle this kernel fault?
+ *
+ * (The kernel has valid exception-points in the source
+ * when it acesses user-memory. When it fails in one
+ * of those points, we find it in a table and do a jump
+ * to some fixup code that loads an appropriate error
+ * code)
+ */
+
+ {
+ const struct exception_table_entry *entry;
+
+ __asm__ __volatile__("l.nop 42");
+
+ if ((entry = search_exception_tables(regs->pc)) != NULL) {
+ /* Adjust the instruction pointer in the stackframe */
+ regs->pc = entry->fixup;
+ return;
+ }
+ }
+
+ /*
+ * Oops. The kernel tried to access some bad page. We'll have to
+ * terminate things with extreme prejudice.
+ */
+
+ if ((unsigned long)(address) < PAGE_SIZE)
+ printk(KERN_ALERT
+ "Unable to handle kernel NULL pointer dereference");
+ else
+ printk(KERN_ALERT "Unable to handle kernel access");
+ printk(" at virtual address 0x%08lx\n", address);
+
+ die("Oops", regs, write_acc);
+
+ do_exit(SIGKILL);
+
+ /*
+ * We ran out of memory, or some other thing happened to us that made
+ * us unable to handle the page fault gracefully.
+ */
+
+out_of_memory:
+ __asm__ __volatile__("l.nop 42");
+ __asm__ __volatile__("l.nop 1");
+
+ up_read(&mm->mmap_sem);
+ printk("VM: killing process %s\n", tsk->comm);
+ if (user_mode(regs))
+ do_exit(SIGKILL);
+ goto no_context;
+
+do_sigbus:
+ up_read(&mm->mmap_sem);
+
+ /*
+ * Send a sigbus, regardless of whether we were in kernel
+ * or user mode.
+ */
+ info.si_signo = SIGBUS;
+ info.si_errno = 0;
+ info.si_code = BUS_ADRERR;
+ info.si_addr = (void *)address;
+ force_sig_info(SIGBUS, &info, tsk);
+
+ /* Kernel mode? Handle exceptions or die */
+ if (!user_mode(regs))
+ goto no_context;
+ return;
+
+vmalloc_fault:
+ {
+ /*
+ * Synchronize this task's top level page-table
+ * with the 'reference' page table.
+ *
+ * Use current_pgd instead of tsk->active_mm->pgd
+ * since the latter might be unavailable if this
+ * code is executed in a misfortunately run irq
+ * (like inside schedule() between switch_mm and
+ * switch_to...).
+ */
+
+ int offset = pgd_index(address);
+ pgd_t *pgd, *pgd_k;
+ pud_t *pud, *pud_k;
+ pmd_t *pmd, *pmd_k;
+ pte_t *pte_k;
+
+/*
+ phx_warn("do_page_fault(): vmalloc_fault will not work, "
+ "since current_pgd assign a proper value somewhere\n"
+ "anyhow we don't need this at the moment\n");
+
+ phx_mmu("vmalloc_fault");
+*/
+ pgd = (pgd_t *)current_pgd + offset;
+ pgd_k = init_mm.pgd + offset;
+
+ /* Since we're two-level, we don't need to do both
+ * set_pgd and set_pmd (they do the same thing). If
+ * we go three-level at some point, do the right thing
+ * with pgd_present and set_pgd here.
+ *
+ * Also, since the vmalloc area is global, we don't
+ * need to copy individual PTE's, it is enough to
+ * copy the pgd pointer into the pte page of the
+ * root task. If that is there, we'll find our pte if
+ * it exists.
+ */
+
+ pud = pud_offset(pgd, address);
+ pud_k = pud_offset(pgd_k, address);
+ if (!pud_present(*pud_k))
+ goto no_context;
+
+ pmd = pmd_offset(pud, address);
+ pmd_k = pmd_offset(pud_k, address);
+
+ if (!pmd_present(*pmd_k))
+ goto bad_area_nosemaphore;
+
+ set_pmd(pmd, *pmd_k);
+
+ /* Make sure the actual PTE exists as well to
+ * catch kernel vmalloc-area accesses to non-mapped
+ * addresses. If we don't do this, this will just
+ * silently loop forever.
+ */
+
+ pte_k = pte_offset_kernel(pmd_k, address);
+ if (!pte_present(*pte_k))
+ goto no_context;
+
+ return;
+ }
+}
diff --git a/arch/openrisc/mm/init.c b/arch/openrisc/mm/init.c
new file mode 100644
index 0000000..069191f
--- /dev/null
+++ b/arch/openrisc/mm/init.c
@@ -0,0 +1,283 @@
+/*
+ * OpenRISC idle.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/smp.h>
+#include <linux/bootmem.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/blkdev.h> /* for initrd_* */
+#include <linux/pagemap.h>
+#include <linux/memblock.h>
+
+#include <asm/system.h>
+#include <asm/segment.h>
+#include <asm/pgalloc.h>
+#include <asm/pgtable.h>
+#include <asm/dma.h>
+#include <asm/io.h>
+#include <asm/tlb.h>
+#include <asm/mmu_context.h>
+#include <asm/kmap_types.h>
+#include <asm/fixmap.h>
+#include <asm/tlbflush.h>
+
+int mem_init_done = 0;
+
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+
+static void __init zone_sizes_init(void)
+{
+ unsigned long zones_size[MAX_NR_ZONES];
+
+ /* Clear the zone sizes */
+ memset(zones_size, 0, sizeof(zones_size));
+
+ /*
+ * We use only ZONE_NORMAL
+ */
+ zones_size[ZONE_NORMAL] = max_low_pfn;
+
+ free_area_init(zones_size);
+}
+
+extern const char _s_kernel_ro[], _e_kernel_ro[];
+
+/*
+ * Map all physical memory into kernel's address space.
+ *
+ * This is explicitly coded for two-level page tables, so if you need
+ * something else then this needs to change.
+ */
+static void __init map_ram(void)
+{
+ unsigned long v, p, e;
+ pgprot_t prot;
+ pgd_t *pge;
+ pud_t *pue;
+ pmd_t *pme;
+ pte_t *pte;
+ /* These mark extents of read-only kernel pages...
+ * ...from vmlinux.lds.S
+ */
+ struct memblock_region *region;
+
+ v = PAGE_OFFSET;
+
+ for_each_memblock(memory, region) {
+ p = (u32) region->base & PAGE_MASK;
+ e = p + (u32) region->size;
+
+ v = (u32) __va(p);
+ pge = pgd_offset_k(v);
+
+ while (p < e) {
+ int j;
+ pue = pud_offset(pge, v);
+ pme = pmd_offset(pue, v);
+
+ if ((u32) pue != (u32) pge || (u32) pme != (u32) pge) {
+ panic
+ ("%s: OR1K kernel hardcoded for two-level page tables",
+ __func__);
+ }
+
+ /* Alloc one page for holding PTE's... */
+ pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+ set_pmd(pme, __pmd(_KERNPG_TABLE + __pa(pte)));
+
+ /* Fill the newly allocated page with PTE'S */
+ for (j = 0; p < e && j < PTRS_PER_PGD;
+ v += PAGE_SIZE, p += PAGE_SIZE, j++, pte++) {
+ if (v >= (u32) _e_kernel_ro ||
+ v < (u32) _s_kernel_ro)
+ prot = PAGE_KERNEL;
+ else
+ prot = PAGE_KERNEL_RO;
+
+ set_pte(pte, mk_pte_phys(p, prot));
+ }
+
+ pge++;
+ }
+
+ printk(KERN_INFO "%s: Memory: 0x%x-0x%x\n", __func__,
+ region->base, region->base + region->size);
+ }
+}
+
+void __init paging_init(void)
+{
+ extern void tlb_init(void);
+
+ unsigned long end;
+ int i;
+
+ printk("Setting up paging and PTEs.\n");
+
+ /* clear out the init_mm.pgd that will contain the kernel's mappings */
+
+ for (i = 0; i < PTRS_PER_PGD; i++)
+ swapper_pg_dir[i] = __pgd(0);
+
+ /* make sure the current pgd table points to something sane
+ * (even if it is most probably not used until the next
+ * switch_mm)
+ */
+ current_pgd = init_mm.pgd;
+
+ end = (unsigned long)__va(max_low_pfn * PAGE_SIZE);
+
+ map_ram();
+
+ zone_sizes_init();
+
+ /* self modifying code ;) */
+ /* Since the old TLB miss handler has been running up until now,
+ * the kernel pages are still all RW, so we can still modify the
+ * text directly... after this change and a TLB flush, the kernel
+ * pages will become RO.
+ */
+ {
+ extern unsigned long dtlb_miss_handler;
+ extern unsigned long itlb_miss_handler;
+
+ unsigned long *dtlb_vector = __va(0x900);
+ unsigned long *itlb_vector = __va(0xa00);
+
+ printk("dtlb_miss_handler %p\n", &dtlb_miss_handler);
+ *dtlb_vector = ((unsigned long)&dtlb_miss_handler -
+ (unsigned long)dtlb_vector) >> 2;
+
+ printk("itlb_miss_handler %p\n", &itlb_miss_handler);
+ *itlb_vector = ((unsigned long)&itlb_miss_handler -
+ (unsigned long)itlb_vector) >> 2;
+ }
+
+ /* Invalidate instruction caches after code modification */
+ mtspr(SPR_ICBIR, 0x900);
+ mtspr(SPR_ICBIR, 0xa00);
+
+ /* New TLB miss handlers and kernel page tables are in now place.
+ * Make sure that page flags get updated for all pages in TLB by
+ * flushing the TLB and forcing all TLB entries to be recreated
+ * from their page table flags.
+ */
+ flush_tlb_all();
+}
+
+/* References to section boundaries */
+
+extern char _stext, _etext, _edata, __bss_start, _end;
+extern char __init_begin, __init_end;
+
+static int __init free_pages_init(void)
+{
+ int reservedpages, pfn;
+
+ /* this will put all low memory onto the freelists */
+ totalram_pages = free_all_bootmem();
+
+ reservedpages = 0;
+ for (pfn = 0; pfn < max_low_pfn; pfn++) {
+ /*
+ * Only count reserved RAM pages
+ */
+ if (PageReserved(mem_map + pfn))
+ reservedpages++;
+ }
+
+ return reservedpages;
+}
+
+static void __init set_max_mapnr_init(void)
+{
+ max_mapnr = num_physpages = max_low_pfn;
+}
+
+void __init mem_init(void)
+{
+ int codesize, reservedpages, datasize, initsize;
+
+ if (!mem_map)
+ BUG();
+
+ set_max_mapnr_init();
+
+ high_memory = (void *)__va(max_low_pfn * PAGE_SIZE);
+
+ /* clear the zero-page */
+ memset((void *)empty_zero_page, 0, PAGE_SIZE);
+
+ reservedpages = free_pages_init();
+
+ codesize = (unsigned long)&_etext - (unsigned long)&_stext;
+ datasize = (unsigned long)&_edata - (unsigned long)&_etext;
+ initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin;
+
+ printk(KERN_INFO
+ "Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init, %ldk highmem)\n",
+ (unsigned long)nr_free_pages() << (PAGE_SHIFT - 10),
+ max_mapnr << (PAGE_SHIFT - 10), codesize >> 10,
+ reservedpages << (PAGE_SHIFT - 10), datasize >> 10,
+ initsize >> 10, (unsigned long)(0 << (PAGE_SHIFT - 10))
+ );
+
+ printk("mem_init_done ...........................................\n");
+ mem_init_done = 1;
+ return;
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+ printk(KERN_INFO "Freeing initrd memory: %ldk freed\n",
+ (end - start) >> 10);
+
+ for (; start < end; start += PAGE_SIZE) {
+ ClearPageReserved(virt_to_page(start));
+ init_page_count(virt_to_page(start));
+ free_page(start);
+ totalram_pages++;
+ }
+}
+#endif
+
+void free_initmem(void)
+{
+ unsigned long addr;
+
+ addr = (unsigned long)(&__init_begin);
+ for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) {
+ ClearPageReserved(virt_to_page(addr));
+ init_page_count(virt_to_page(addr));
+ free_page(addr);
+ totalram_pages++;
+ }
+ printk(KERN_INFO "Freeing unused kernel memory: %luk freed\n",
+ ((unsigned long)&__init_end -
+ (unsigned long)&__init_begin) >> 10);
+}
diff --git a/arch/openrisc/mm/ioremap.c b/arch/openrisc/mm/ioremap.c
new file mode 100644
index 0000000..5e12976
--- /dev/null
+++ b/arch/openrisc/mm/ioremap.c
@@ -0,0 +1,113 @@
+/*
+ * OpenRISC ioremap.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/vmalloc.h>
+#include <linux/io.h>
+#include <asm/pgalloc.h>
+#include <asm/kmap_types.h>
+#include <asm/fixmap.h>
+#include <asm/bug.h>
+#include <asm/pgtable.h>
+#include <linux/sched.h>
+#include <asm/tlbflush.h>
+
+extern int mem_init_done;
+
+static unsigned int fixmaps_used __initdata;
+
+/*
+ * Remap an arbitrary physical address space into the kernel virtual
+ * address space. Needed when the kernel wants to access high addresses
+ * directly.
+ *
+ * NOTE! We need to allow non-page-aligned mappings too: we will obviously
+ * have to convert them into an offset in a page-aligned mapping, but the
+ * caller shouldn't need to know that small detail.
+ */
+void __iomem* __init_refok
+__ioremap(phys_addr_t addr, unsigned long size, unsigned long flags)
+{
+ phys_addr_t p;
+ unsigned long v;
+ unsigned long offset, last_addr;
+ struct vm_struct *area = NULL;
+ pgprot_t prot;
+
+ /* Don't allow wraparound or zero size */
+ last_addr = addr + size - 1;
+ if (!size || last_addr < addr)
+ return NULL;
+
+ /*
+ * Mappings have to be page-aligned
+ */
+ offset = addr & ~PAGE_MASK;
+ p = addr & PAGE_MASK;
+ size = PAGE_ALIGN(last_addr+1) - p;
+
+ if (likely (mem_init_done)) {
+ area = get_vm_area(size, VM_IOREMAP);
+ if (!area)
+ return NULL;
+ v = (unsigned long) area->addr;
+ } else {
+ if ((fixmaps_used + (size >> PAGE_SHIFT)) > FIX_N_IOREMAPS)
+ return NULL;
+ v = fix_to_virt(FIX_IOREMAP_BEGIN+fixmaps_used);
+ fixmaps_used += (size >> PAGE_SHIFT);
+ }
+
+ prot = __pgprot(_PAGE_ALL | _PAGE_SRE | _PAGE_SWE |
+ _PAGE_SHARED | _PAGE_DIRTY | _PAGE_EXEC | flags);
+
+ if (ioremap_page_range(v, v + size, p, prot)) {
+ if (likely(mem_init_done))
+ vfree(area->addr);
+ else
+ fixmaps_used -= (size >> PAGE_SHIFT);
+ return NULL;
+ }
+
+
+ return (void __iomem *) (offset + (char *)v);
+}
+
+void iounmap(void *addr)
+{
+ /* If the page is from the fixmap pool then we just clear out
+ * the fixmap mapping.
+ */
+ if (unlikely((unsigned long)addr > FIXADDR_START)) {
+ /* This is a bit broken... we don't really know
+ * how big the area is so it's difficult to know
+ * how many fixed pages to invalidate...
+ * just flush tlb and hope for the best...
+ * consider this a FIXME
+ *
+ * Really we should be clearing out one or more page
+ * table entries for these virtual addresses so that
+ * future references cause a page fault... for now, we
+ * rely on two things:
+ * i) this code never gets called on known boards
+ * ii) invalid accesses to the freed areas aren't made
+ */
+ flush_tlb_all();
+ return;
+ }
+
+ return vfree((void *) (PAGE_MASK & (unsigned long) addr));
+}
diff --git a/arch/openrisc/mm/tlb.c b/arch/openrisc/mm/tlb.c
new file mode 100644
index 0000000..498f29b
--- /dev/null
+++ b/arch/openrisc/mm/tlb.c
@@ -0,0 +1,191 @@
+/*
+ * OpenRISC tlb.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Julius Baxter <***@orsoc.se>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+
+#include <asm/system.h>
+#include <asm/segment.h>
+#include <asm/tlbflush.h>
+#include <asm/pgtable.h>
+#include <asm/mmu_context.h>
+#include <asm/spr_defs.h>
+
+#define NO_CONTEXT -1
+
+#define NUM_DTLB_SETS (1 << ((mfspr(SPR_IMMUCFGR) & SPR_IMMUCFGR_NTS) >> \
+ SPR_DMMUCFGR_NTS_OFF))
+#define NUM_ITLB_SETS (1 << ((mfspr(SPR_IMMUCFGR) & SPR_IMMUCFGR_NTS) >> \
+ SPR_IMMUCFGR_NTS_OFF))
+#define DTLB_OFFSET(addr) (((addr) >> PAGE_SHIFT) & (NUM_DTLB_SETS-1))
+#define ITLB_OFFSET(addr) (((addr) >> PAGE_SHIFT) & (NUM_ITLB_SETS-1))
+/*
+ * Invalidate all TLB entries.
+ *
+ * This comes down to setting the 'valid' bit for all xTLBMR registers to 0.
+ * Easiest way to accomplish this is to just zero out the xTLBMR register
+ * completely.
+ *
+ */
+
+void flush_tlb_all(void)
+{
+ int i;
+ unsigned long num_tlb_sets;
+
+ /* Determine number of sets for IMMU. */
+ /* FIXME: Assumption is I & D nsets equal. */
+ num_tlb_sets = NUM_ITLB_SETS;
+
+ for (i = 0; i < num_tlb_sets; i++) {
+ mtspr_off(SPR_DTLBMR_BASE(0), i, 0);
+ mtspr_off(SPR_ITLBMR_BASE(0), i, 0);
+ }
+}
+
+#define have_dtlbeir (mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_TEIRI)
+#define have_itlbeir (mfspr(SPR_IMMUCFGR) & SPR_IMMUCFGR_TEIRI)
+
+/*
+ * Invalidate a single page. This is what the xTLBEIR register is for.
+ *
+ * There's no point in checking the vma for PAGE_EXEC to determine whether it's
+ * the data or instruction TLB that should be flushed... that would take more
+ * than the few instructions that the following compiles down to!
+ *
+ * The case where we don't have the xTLBEIR register really only works for
+ * MMU's with a single way and is hard-coded that way.
+ */
+
+#define flush_dtlb_page_eir(addr) mtspr(SPR_DTLBEIR, addr)
+#define flush_dtlb_page_no_eir(addr) mtspr_off(SPR_DTLBMR_BASE(0), DTLB_OFFSET(addr), 0);
+
+#define flush_itlb_page_eir(addr) mtspr(SPR_ITLBEIR, addr)
+#define flush_itlb_page_no_eir(addr) mtspr_off(SPR_ITLBMR_BASE(0), ITLB_OFFSET(addr), 0);
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long addr)
+{
+ if (have_dtlbeir)
+ flush_dtlb_page_eir(addr);
+ else
+ flush_dtlb_page_no_eir(addr);
+
+ if (have_itlbeir)
+ flush_itlb_page_eir(addr);
+ else
+ flush_itlb_page_no_eir(addr);
+}
+
+void flush_tlb_range(struct vm_area_struct *vma,
+ unsigned long start, unsigned long end)
+{
+ int addr;
+ bool dtlbeir;
+ bool itlbeir;
+
+ dtlbeir = have_dtlbeir;
+ itlbeir = have_itlbeir;
+
+ for (addr = start; addr < end; addr += PAGE_SIZE) {
+ if (dtlbeir)
+ flush_dtlb_page_eir(addr);
+ else
+ flush_dtlb_page_no_eir(addr);
+
+ if (itlbeir)
+ flush_itlb_page_eir(addr);
+ else
+ flush_itlb_page_no_eir(addr);
+ }
+}
+
+/*
+ * Invalidate the selected mm context only.
+ *
+ * FIXME: Due to some bug here, we're flushing everything for now.
+ * This should be changed to loop over over mm and call flush_tlb_range.
+ */
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+
+ /* Was seeing bugs with the mm struct passed to us. Scrapped most of
+ this function. */
+ /* Several architctures do this */
+ flush_tlb_all();
+}
+
+/* called in schedule() just before actually doing the switch_to */
+
+void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ struct task_struct *next_tsk)
+{
+ /* remember the pgd for the fault handlers
+ * this is similar to the pgd register in some other CPU's.
+ * we need our own copy of it because current and active_mm
+ * might be invalid at points where we still need to derefer
+ * the pgd.
+ */
+ current_pgd = next->pgd;
+
+ /* We don't have context support implemented, so flush all
+ * entries belonging to previous map
+ */
+
+ if (prev != next)
+ flush_tlb_mm(prev);
+
+}
+
+/*
+ * Initialize the context related info for a new mm_struct
+ * instance.
+ */
+
+int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
+{
+ mm->context = NO_CONTEXT;
+ return 0;
+}
+
+/* called by __exit_mm to destroy the used MMU context if any before
+ * destroying the mm itself. this is only called when the last user of the mm
+ * drops it.
+ */
+
+void destroy_context(struct mm_struct *mm)
+{
+ flush_tlb_mm(mm);
+
+}
+
+/* called once during VM initialization, from init.c */
+
+void __init tlb_init(void)
+{
+ /* Do nothing... */
+ /* invalidate the entire TLB */
+ /* flush_tlb_all(); */
+}
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:37 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/sigcontext.h | 38 +++
arch/openrisc/kernel/signal.c | 404 ++++++++++++++++++++++++++++++++
2 files changed, 442 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/sigcontext.h
create mode 100644 arch/openrisc/kernel/signal.c

diff --git a/arch/openrisc/include/asm/sigcontext.h b/arch/openrisc/include/asm/sigcontext.h
new file mode 100644
index 0000000..54a5c50
--- /dev/null
+++ b/arch/openrisc/include/asm/sigcontext.h
@@ -0,0 +1,38 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_SIGCONTEXT_H
+#define __ASM_OPENRISC_SIGCONTEXT_H
+
+#include <asm/ptrace.h>
+
+/* This struct is saved by setup_frame in signal.c, to keep the current
+ context while a signal handler is executed. It's restored by sys_sigreturn.
+
+ To keep things simple, we use pt_regs here even though normally you just
+ specify the list of regs to save. Then we can use copy_from_user on the
+ entire regs instead of a bunch of get_user's as well...
+*/
+
+struct sigcontext {
+ struct pt_regs regs; /* needs to be first */
+ unsigned long oldmask;
+ unsigned long usp; /* usp before stacking this gunk on it */
+};
+
+#endif /* __ASM_OPENRISC_SIGCONTEXT_H */
diff --git a/arch/openrisc/kernel/signal.c b/arch/openrisc/kernel/signal.c
new file mode 100644
index 0000000..9d1b839
--- /dev/null
+++ b/arch/openrisc/kernel/signal.c
@@ -0,0 +1,404 @@
+/*
+ * OpenRISC signal.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/errno.h>
+#include <linux/wait.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/stddef.h>
+#include <linux/tracehook.h>
+
+#include <asm/processor.h>
+#include <asm/ucontext.h>
+#include <asm/uaccess.h>
+
+#include "ptrace.h"
+
+#define DEBUG_SIG 0
+
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
+asmlinkage long
+_sys_sigaltstack(const stack_t *uss, stack_t *uoss, struct pt_regs *regs)
+{
+ return do_sigaltstack(uss, uoss, regs->sp);
+}
+
+struct rt_sigframe {
+ struct siginfo *pinfo;
+ void *puc;
+ struct siginfo info;
+ struct ucontext uc;
+ unsigned char retcode[16]; /* trampoline code */
+};
+
+static int restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc)
+{
+ unsigned int err = 0;
+ unsigned long old_usp;
+
+ /* Alwys make any pending restarted system call return -EINTR */
+ current_thread_info()->restart_block.fn = do_no_restart_syscall;
+
+ /* restore the regs from &sc->regs (same as sc, since regs is first)
+ * (sc is already checked for VERIFY_READ since the sigframe was
+ * checked in sys_sigreturn previously)
+ */
+
+ if (__copy_from_user(regs, sc, sizeof(struct pt_regs)))
+ goto badframe;
+
+ /* make sure the SM-bit is cleared so user-mode cannot fool us */
+ regs->sr &= ~SPR_SR_SM;
+
+ /* restore the old USP as it was before we stacked the sc etc.
+ * (we cannot just pop the sigcontext since we aligned the sp and
+ * stuff after pushing it)
+ */
+
+ err |= __get_user(old_usp, &sc->usp);
+
+ regs->sp = old_usp;
+
+ /* TODO: the other ports use regs->orig_XX to disable syscall checks
+ * after this completes, but we don't use that mechanism. maybe we can
+ * use it now ?
+ */
+
+ return err;
+
+badframe:
+ return 1;
+}
+
+asmlinkage long _sys_rt_sigreturn(struct pt_regs *regs)
+{
+ struct rt_sigframe *frame = (struct rt_sigframe __user *)regs->sp;
+ sigset_t set;
+ stack_t st;
+
+ /*
+ * Since we stacked the signal on a dword boundary,
+ * then frame should be dword aligned here. If it's
+ * not, then the user is trying to mess with us.
+ */
+ if (((long)frame) & 3)
+ goto badframe;
+
+ if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+ goto badframe;
+ if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+ goto badframe;
+
+ sigdelsetmask(&set, ~_BLOCKABLE);
+ spin_lock_irq(&current->sighand->siglock);
+ current->blocked = set;
+ recalc_sigpending();
+ spin_unlock_irq(&current->sighand->siglock);
+
+ if (restore_sigcontext(regs, &frame->uc.uc_mcontext))
+ goto badframe;
+
+ if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
+ goto badframe;
+ /* It is more difficult to avoid calling this function than to
+ call it and ignore errors. */
+ do_sigaltstack(&st, NULL, regs->sp);
+
+ single_step_trap(current);
+
+ return regs->gpr[11];
+
+badframe:
+ force_sig(SIGSEGV, current);
+ return 0;
+}
+
+/*
+ * Set up a signal frame.
+ */
+
+static int setup_sigcontext(struct sigcontext *sc, struct pt_regs *regs,
+ unsigned long mask)
+{
+ int err = 0;
+ unsigned long usp = regs->sp;
+
+ /* copy the regs. they are first in sc so we can use sc directly */
+
+ err |= __copy_to_user(sc, regs, sizeof(struct pt_regs));
+
+ /* then some other stuff */
+
+ err |= __put_user(mask, &sc->oldmask);
+
+ err |= __put_user(usp, &sc->usp);
+
+ return err;
+}
+
+static inline unsigned long align_sigframe(unsigned long sp)
+{
+ return sp & ~3UL;
+}
+
+/*
+ * Work out where the signal frame should go. It's either on the user stack
+ * or the alternate stack.
+ */
+
+static inline void __user *get_sigframe(struct k_sigaction *ka,
+ struct pt_regs *regs, size_t frame_size)
+{
+ unsigned long sp = regs->sp;
+ int onsigstack = on_sig_stack(sp);
+
+ /* redzone */
+ sp -= STACK_FRAME_OVERHEAD;
+
+ /* This is the X/Open sanctioned signal stack switching. */
+ if ((ka->sa.sa_flags & SA_ONSTACK) && !onsigstack) {
+ if (current->sas_ss_size)
+ sp = current->sas_ss_sp + current->sas_ss_size;
+ }
+
+ sp = align_sigframe(sp - frame_size);
+
+ /*
+ * If we are on the alternate signal stack and would overflow it, don't.
+ * Return an always-bogus address instead so we will die with SIGSEGV.
+ */
+ if (onsigstack && !likely(on_sig_stack(sp)))
+ return (void __user *)-1L;
+
+ return (void __user *)sp;
+}
+
+/* grab and setup a signal frame.
+ *
+ * basically we stack a lot of state info, and arrange for the
+ * user-mode program to return to the kernel using either a
+ * trampoline which performs the syscall sigreturn, or a provided
+ * user-mode trampoline.
+ */
+static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+ sigset_t *set, struct pt_regs *regs)
+{
+ struct rt_sigframe *frame;
+ unsigned long return_ip;
+ int err = 0;
+
+ frame = get_sigframe(ka, regs, sizeof(*frame));
+
+ if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+ goto give_sigsegv;
+
+ err |= __put_user(&frame->info, &frame->pinfo);
+ err |= __put_user(&frame->uc, &frame->puc);
+
+ if (ka->sa.sa_flags & SA_SIGINFO)
+ err |= copy_siginfo_to_user(&frame->info, info);
+ if (err)
+ goto give_sigsegv;
+
+ /* Clear all the bits of the ucontext we don't use. */
+ err |= __clear_user(&frame->uc, offsetof(struct ucontext, uc_mcontext));
+ err |= __put_user(0, &frame->uc.uc_flags);
+ err |= __put_user(NULL, &frame->uc.uc_link);
+ err |= __put_user((void *)current->sas_ss_sp,
+ &frame->uc.uc_stack.ss_sp);
+ err |= __put_user(sas_ss_flags(regs->sp), &frame->uc.uc_stack.ss_flags);
+ err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
+ err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, set->sig[0]);
+
+ err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+ if (err)
+ goto give_sigsegv;
+
+ /* trampoline - the desired return ip is the retcode itself */
+ return_ip = (unsigned long)&frame->retcode;
+ /* This is l.ori r11,r0,__NR_sigreturn, l.sys 1 */
+ err |= __put_user(0xa960, (short *)(frame->retcode + 0));
+ err |= __put_user(__NR_rt_sigreturn, (short *)(frame->retcode + 2));
+ err |= __put_user(0x20000001, (unsigned long *)(frame->retcode + 4));
+ err |= __put_user(0x15000000, (unsigned long *)(frame->retcode + 8));
+
+ if (err)
+ goto give_sigsegv;
+
+ /* TODO what is the current->exec_domain stuff and invmap ? */
+
+ /* Set up registers for signal handler */
+ regs->pc = (unsigned long)ka->sa.sa_handler; /* what we enter NOW */
+ regs->gpr[9] = (unsigned long)return_ip; /* what we enter LATER */
+ regs->gpr[3] = (unsigned long)sig; /* arg 1: signo */
+ regs->gpr[4] = (unsigned long)&frame->info; /* arg 2: (siginfo_t*) */
+ regs->gpr[5] = (unsigned long)&frame->uc; /* arg 3: ucontext */
+
+ /* actually move the usp to reflect the stacked frame */
+ regs->sp = (unsigned long)frame;
+
+ return;
+
+give_sigsegv:
+ if (sig == SIGSEGV)
+ ka->sa.sa_handler = SIG_DFL;
+ force_sig(SIGSEGV, current);
+}
+
+static inline void
+handle_signal(unsigned long sig,
+ siginfo_t *info, struct k_sigaction *ka,
+ sigset_t *oldset, struct pt_regs *regs)
+{
+ setup_rt_frame(sig, ka, info, oldset, regs);
+
+ if (ka->sa.sa_flags & SA_ONESHOT)
+ ka->sa.sa_handler = SIG_DFL;
+
+ spin_lock_irq(&current->sighand->siglock);
+ sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
+ if (!(ka->sa.sa_flags & SA_NODEFER))
+ sigaddset(&current->blocked, sig);
+ recalc_sigpending();
+
+ spin_unlock_irq(&current->sighand->siglock);
+}
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it doesn't
+ * want to handle. Thus you cannot kill init even with a SIGKILL even by
+ * mistake.
+ *
+ * Also note that the regs structure given here as an argument, is the latest
+ * pushed pt_regs. It may or may not be the same as the first pushed registers
+ * when the initial usermode->kernelmode transition took place. Therefore
+ * we can use user_mode(regs) to see if we came directly from kernel or user
+ * mode below.
+ */
+
+void do_signal(struct pt_regs *regs)
+{
+ siginfo_t info;
+ int signr;
+ struct k_sigaction ka;
+
+ /*
+ * We want the common case to go fast, which
+ * is why we may in certain cases get here from
+ * kernel mode. Just return without doing anything
+ * if so.
+ */
+ if (!user_mode(regs))
+ return;
+
+ single_step_clear(current);
+
+ signr = get_signal_to_deliver(&info, &ka, regs, NULL);
+
+ /* If we are coming out of a syscall then we need
+ * to check if the syscall was interrupted and wants to be
+ * restarted after handling the signal. If so, the original
+ * syscall number is put back into r11 and the PC rewound to
+ * point at the l.sys instruction that resulted in the
+ * original syscall. Syscall results other than the four
+ * below mean that the syscall executed to completion and no
+ * restart is necessary.
+ */
+ if (regs->syscallno) {
+ int restart = 0;
+
+ switch (regs->gpr[11]) {
+ case -ERESTART_RESTARTBLOCK:
+ case -ERESTARTNOHAND:
+ /* Restart if there is no signal handler */
+ restart = (signr <= 0);
+ break;
+ case -ERESTARTSYS:
+ /* Restart if there no signal handler or
+ * SA_RESTART flag is set */
+ restart = (signr <= 0 || (ka.sa.sa_flags & SA_RESTART));
+ break;
+ case -ERESTARTNOINTR:
+ /* Always restart */
+ restart = 1;
+ break;
+ }
+
+ if (restart) {
+ if (regs->gpr[11] == -ERESTART_RESTARTBLOCK)
+ regs->gpr[11] = __NR_restart_syscall;
+ else
+ regs->gpr[11] = regs->orig_gpr11;
+ regs->pc -= 4;
+ } else {
+ regs->gpr[11] = -EINTR;
+ }
+ }
+
+ if (signr <= 0) {
+ /* no signal to deliver so we just put the saved sigmask
+ * back */
+ if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+ clear_thread_flag(TIF_RESTORE_SIGMASK);
+ sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+ }
+
+ } else { /* signr > 0 */
+ sigset_t *oldset;
+
+ if (current_thread_info()->flags & _TIF_RESTORE_SIGMASK)
+ oldset = &current->saved_sigmask;
+ else
+ oldset = &current->blocked;
+
+ /* Whee! Actually deliver the signal. */
+ handle_signal(signr, &info, &ka, oldset, regs);
+ /* a signal was successfully delivered; the saved
+ * sigmask will have been stored in the signal frame,
+ * and will be restored by sigreturn, so we can simply
+ * clear the TIF_RESTORE_SIGMASK flag */
+ if (test_thread_flag(TIF_RESTORE_SIGMASK))
+ clear_thread_flag(TIF_RESTORE_SIGMASK);
+
+ tracehook_signal_handler(signr, &info, &ka, regs,
+ test_thread_flag(TIF_SINGLESTEP));
+ }
+
+ single_step_set(current);
+
+ return;
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs)
+{
+ if (current_thread_info()->flags & _TIF_SIGPENDING)
+ do_signal(regs);
+
+ if (current_thread_info()->flags & _TIF_NOTIFY_RESUME) {
+ clear_thread_flag(TIF_NOTIFY_RESUME);
+ tracehook_notify_resume(regs);
+ if (current->replacement_session_keyring)
+ key_replace_session_keyring();
+ }
+}
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:35 UTC
Permalink
The OpenRISC architecture uses the device tree infrastructure for the
platform description. This is currently limited to having a device tree
built into the kernel, but work is underway within the OpenRISC project
to define how this device tree blob should be passed into the kernel from
an external resource.

Signed-off-by: Jonas Bonn <***@southpole.se>
Cc: devicetree-***@lists.ozlabs.org
---
arch/openrisc/boot/dts/atlys.dts | 78 +++++++++++++++++++++++++
arch/openrisc/boot/dts/or1ksim.dts | 73 ++++++++++++++++++++++++
arch/openrisc/include/asm/prom.h | 92 ++++++++++++++++++++++++++++++
arch/openrisc/kernel/prom.c | 109 ++++++++++++++++++++++++++++++++++++
arch/openrisc/kernel/setup.c | 3 +-
5 files changed, 353 insertions(+), 2 deletions(-)
create mode 100644 arch/openrisc/boot/dts/atlys.dts
create mode 100644 arch/openrisc/boot/dts/or1ksim.dts
create mode 100644 arch/openrisc/include/asm/prom.h
create mode 100644 arch/openrisc/kernel/prom.c

diff --git a/arch/openrisc/boot/dts/atlys.dts b/arch/openrisc/boot/dts/atlys.dts
new file mode 100644
index 0000000..cd028b9
--- /dev/null
+++ b/arch/openrisc/boot/dts/atlys.dts
@@ -0,0 +1,78 @@
+/dts-v1/;
+/ {
+ compatible = "digilent,atlys";
+ #size-cells = <1>;
+ #address-cells = <1>;
+
+ chosen {
+ bootargs = "console=tty0 console=uart,mmio,0x90000000,115200 debug video=ocfb:640x480-***@60";
+ };
+
+ memory {
+ device_type = "memory";
+ reg = <0x00000000 0x8000000>;
+ };
+
+ cpus {
+ ***@0 {
+ compatible = "opencores,openrisc-1200";
+ clock-frequency = <50000000>;
+ };
+ };
+
+ /* The "soc" node */
+
+ soc {
+ /* Every node with children must specify address-cells and
+ size-cells; they are not inherited */
+ #address-cells = <1>;
+ #size-cells = <1>;
+ device_type = "soc";
+
+ /* This is also a 1:1 mapping but limits size of mappable
+ region to 0x40000000 */
+ ranges = <0x80000000 0x80000000 0x40000000>;
+
+ /* An alternative is to specify an empty ranges here to
+ indicates 1:1 mapping of SOC addresses (device addresses)
+ to CPU addresses */
+ /*ranges;*/
+
+ /* This SOC has only a single, implicit PIC, so define it
+ early and put the interrupt-parent property in the 'soc'
+ node so that it's inherited by all the children using
+ interrupts.
+ */
+ interrupt-parent = <&pic>;
+
+ /* Devices with interrupts need a 'parent', so we need to
+ define a controller */
+ pic: ***@0 {
+ compatible = "opencores,or1k-pic";
+ #interrupt-cells = <1>;
+ interrupt-controller;
+ };
+
+ serial0: ***@90000000 {
+ /* FIXME: device_type is still needed here...
+ should remove requirement from of_serial driver */
+ device_type = "serial";
+ compatible = "opencores,uart", "ns16550a";
+ reg = <0x90000000 0x100>;
+ interrupts = <2>;
+ clock-frequency = <50000000>;
+ };
+
+ enet0: ***@92000000 {
+ compatible = "opencores,ethoc", "ethoc";
+ reg = <0x92000000 0x53>;
+ interrupts = <4>;
+ local-mac-address = [02 de ad be ef 02];
+ };
+
+ fb0: ***@97000000 {
+ compatible = "opencores,ocfb", "ocfb";
+ reg = <0x97000000 0x1000>;
+ };
+ };
+};
diff --git a/arch/openrisc/boot/dts/or1ksim.dts b/arch/openrisc/boot/dts/or1ksim.dts
new file mode 100644
index 0000000..6a025ea
--- /dev/null
+++ b/arch/openrisc/boot/dts/or1ksim.dts
@@ -0,0 +1,73 @@
+/dts-v1/;
+/ {
+ compatible = "opencores,or1ksim";
+ #size-cells = <1>;
+ #address-cells = <1>;
+
+ chosen {
+ bootargs = "console=uart,mmio,0x90000000,115200 debug";
+ };
+
+ memory {
+ device_type = "memory";
+ reg = <0x00000000 0x2000000>;
+ };
+
+ cpus {
+ ***@0 {
+ compatible = "opencores,openrisc-1200";
+ clock-frequency = <20000000>;
+ };
+ };
+
+ /* The "soc" node */
+
+ soc {
+ /* Every node with children must specify address-cells and
+ size-cells; they are not inherited */
+ #address-cells = <1>;
+ #size-cells = <1>;
+ device_type = "soc";
+
+ /* This is also a 1:1 mapping but limits size of mappable
+ region to 0x40000000 */
+ ranges = <0x80000000 0x80000000 0x40000000>;
+
+ /* An alternative is to specify an empty ranges here to
+ indicates 1:1 mapping of SOC addresses (device addresses)
+ to CPU addresses */
+ /*ranges;*/
+
+ /* This SOC has only a single, implicit PIC, so define it
+ early and put the interrupt-parent property in the 'soc'
+ node so that it's inherited by all the children using
+ interrupts.
+ */
+ interrupt-parent = <&pic>;
+
+ /* Devices with interrupts need a 'parent', so we need to
+ define a controller */
+ pic: ***@0 {
+ compatible = "opencores,or1k-pic";
+ #interrupt-cells = <1>;
+ interrupt-controller;
+ };
+
+ serial0: ***@90000000 {
+ /* FIXME: device_type is still needed here...
+ should remove requirement from of_serial driver */
+ device_type = "serial";
+ compatible = "opencores,uart", "ns16550a";
+ reg = <0x90000000 0x100>;
+ interrupts = <2>;
+ clock-frequency = <20000000>;
+ };
+
+ enet0: ***@92000000 {
+ compatible = "opencores,ethoc", "ethoc";
+ reg = <0x92000000 0x53>;
+ interrupts = <4>;
+ local-mac-address = [02 de ad be ef 02];
+ };
+ };
+};
diff --git a/arch/openrisc/include/asm/prom.h b/arch/openrisc/include/asm/prom.h
new file mode 100644
index 0000000..3818902
--- /dev/null
+++ b/arch/openrisc/include/asm/prom.h
@@ -0,0 +1,92 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/of.h> /* linux/of.h gets to determine #include ordering */
+
+#ifndef _ASM_OPENRISC_PROM_H
+#define _ASM_OPENRISC_PROM_H
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
+
+#include <linux/types.h>
+#include <asm/irq.h>
+#include <asm/atomic.h>
+#include <linux/of_irq.h>
+#include <linux/of_fdt.h>
+#include <linux/of_address.h>
+#include <linux/proc_fs.h>
+#include <linux/platform_device.h>
+#define HAVE_ARCH_DEVTREE_FIXUPS
+
+/* Other Prototypes */
+extern int early_uartlite_console(void);
+
+#ifdef CONFIG_PCI
+/*
+ * PCI <-> OF matching functions
+ * (XXX should these be here?)
+ */
+struct pci_bus;
+struct pci_dev;
+extern int pci_device_from_OF_node(struct device_node *node,
+ u8 *bus, u8 *devfn);
+extern struct device_node *pci_busdev_to_OF_node(struct pci_bus *bus,
+ int devfn);
+extern struct device_node *pci_device_to_OF_node(struct pci_dev *dev);
+extern void pci_create_OF_bus_map(void);
+#endif
+
+/* Parse the ibm,dma-window property of an OF node into the busno, phys and
+ * size parameters.
+ */
+void of_parse_dma_window(struct device_node *dn, const void *dma_window_prop,
+ unsigned long *busno, unsigned long *phys, unsigned long *size);
+
+extern void kdump_move_device_tree(void);
+
+/* CPU OF node matching */
+struct device_node *of_get_cpu_node(int cpu, unsigned int *thread);
+
+/* Get the MAC address */
+extern const void *of_get_mac_address(struct device_node *np);
+
+/**
+ * of_irq_map_pci - Resolve the interrupt for a PCI device
+ * @pdev: the device whose interrupt is to be resolved
+ * @out_irq: structure of_irq filled by this function
+ *
+ * This function resolves the PCI interrupt for a given PCI device. If a
+ * device-node exists for a given pci_dev, it will use normal OF tree
+ * walking. If not, it will implement standard swizzling and walk up the
+ * PCI tree until an device-node is found, at which point it will finish
+ * resolving using the OF tree walking.
+ */
+struct pci_dev;
+extern int of_irq_map_pci(struct pci_dev *pdev, struct of_irq *out_irq);
+
+/* This routine is here to provide compatibility with how powerpc
+ * handles IRQ mapping for OF device nodes. We precompute and permanently
+ * register them in the platform_device objects, whereas powerpc computes them
+ * on request.
+ */
+static inline void irq_dispose_mapping(unsigned int virq)
+{
+}
+
+#endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+#endif /* _ASM_OPENRISC_PROM_H */
diff --git a/arch/openrisc/kernel/prom.c b/arch/openrisc/kernel/prom.c
new file mode 100644
index 0000000..2bb5a8c
--- /dev/null
+++ b/arch/openrisc/kernel/prom.c
@@ -0,0 +1,109 @@
+/*
+ * OpenRISC prom.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Architecture specific procedures for creating, accessing and
+ * interpreting the device tree.
+ *
+ */
+
+#include <stdarg.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/threads.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/stringify.h>
+#include <linux/delay.h>
+#include <linux/initrd.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/kexec.h>
+#include <linux/debugfs.h>
+#include <linux/irq.h>
+#include <linux/memblock.h>
+#include <linux/of_fdt.h>
+
+#include <asm/prom.h>
+#include <asm/page.h>
+#include <asm/processor.h>
+#include <asm/irq.h>
+#include <linux/io.h>
+#include <asm/system.h>
+#include <asm/mmu.h>
+#include <asm/pgtable.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+
+extern char cmd_line[COMMAND_LINE_SIZE];
+
+void __init early_init_dt_add_memory_arch(u64 base, u64 size)
+{
+ size &= PAGE_MASK;
+ memblock_add(base, size);
+}
+
+void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align)
+{
+ return __va(memblock_alloc(size, align));
+}
+
+void __init early_init_devtree(void *params)
+{
+ u8 *alloc;
+
+ /* Setup flat device-tree pointer */
+ initial_boot_params = params;
+
+
+ /* Retrieve various informations from the /chosen node of the
+ * device-tree, including the platform type, initrd location and
+ * size, TCE reserve, and more ...
+ */
+ of_scan_flat_dt(early_init_dt_scan_chosen, cmd_line);
+
+ /* Scan memory nodes and rebuild MEMBLOCKs */
+ memblock_init();
+ of_scan_flat_dt(early_init_dt_scan_root, NULL);
+ of_scan_flat_dt(early_init_dt_scan_memory, NULL);
+
+ /* Save command line for /proc/cmdline and then parse parameters */
+ strlcpy(boot_command_line, cmd_line, COMMAND_LINE_SIZE);
+// parse_early_param();
+
+ memblock_analyze();
+
+ /* We must copy the flattend device tree from init memory to regular
+ * memory because the device tree references the strings in it
+ * directly.
+ */
+
+ alloc = __va(memblock_alloc(initial_boot_params->totalsize, PAGE_SIZE));
+
+ memcpy(alloc, initial_boot_params, initial_boot_params->totalsize);
+
+ initial_boot_params = (void *) alloc;
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+void __init early_init_dt_setup_initrd_arch(unsigned long start,
+ unsigned long end)
+{
+ initrd_start = (unsigned long)__va(start);
+ initrd_end = (unsigned long)__va(end);
+ initrd_below_start_ok = 1;
+}
+#endif
diff --git a/arch/openrisc/kernel/setup.c b/arch/openrisc/kernel/setup.c
index 49342e9..6ce6583 100644
--- a/arch/openrisc/kernel/setup.c
+++ b/arch/openrisc/kernel/setup.c
@@ -173,8 +173,7 @@ void __init setup_cpuinfo(void)
unsigned long iccfgr,dccfgr;
unsigned long cache_set_size, cache_ways;;

- cpu = (struct device_node *) of_find_compatible_node(NULL,
- NULL, "opencores,openrisc-1200");
+ cpu = of_find_compatible_node(NULL, NULL, "opencores,openrisc-1200");
if (!cpu) {
panic("No compatible CPU found in device tree...\n");
}
--
1.7.4.1
Segher Boessenkool
2011-07-03 18:07:45 UTC
Permalink
Post by Jonas Bonn
The OpenRISC architecture uses the device tree infrastructure for the
platform description.
Hurray, world domination is one step closer ;-)
Post by Jonas Bonn
+ memory {
+ device_type = "memory";
+ reg = <0x00000000 0x8000000>;
+ };
Might want to write this as 0x08000000 (easier to read).
Post by Jonas Bonn
+ cpus {
Any node with a textual unit address (@0) needs to have a "reg" property
with the same address (and then the parent node, "cpus", should have
#address-cells and #size-cells defined, probably to 1 and 0 resp. in
this
case).

You can just name the node "cpu" here.
Post by Jonas Bonn
+ compatible = "opencores,or1k-pic";
+ #interrupt-cells = <1>;
+ interrupt-controller;
+ };
Same thing here. Although, does your interrupt controller have any
registers?
You should describe them then.
Post by Jonas Bonn
+
+ /* FIXME: device_type is still needed here...
+ should remove requirement from of_serial driver */
+ device_type = "serial";
+ compatible = "opencores,uart", "ns16550a";
Opencores has and always will have only one UART implementation? You
should
make the "compatible" name more specific ("opencores,uart-16550a"
perhaps).
Post by Jonas Bonn
+ compatible = "opencores,ethoc", "ethoc";
Probably shouldn't have the "ethoc" there, way too generic a name.
Post by Jonas Bonn
+ reg = <0x92000000 0x53>;
0x53? Usually you write the size of the register block that is decoded,
not just the size of the defined registers; so it would be 0x100 or
something
like that.


Segher
Grant Likely
2011-07-03 20:51:10 UTC
Permalink
Post by Jonas Bonn
The OpenRISC architecture uses the device tree infrastructure for the
platform description. =A0This is currently limited to having a device=
tree
Post by Jonas Bonn
built into the kernel, but work is underway within the OpenRISC proje=
ct
Post by Jonas Bonn
to define how this device tree blob should be passed into the kernel =
from
Post by Jonas Bonn
an external resource.
---
=A0arch/openrisc/boot/dts/atlys.dts =A0 | =A0 78 ++++++++++++++++++++=
+++++
Post by Jonas Bonn
=A0arch/openrisc/boot/dts/or1ksim.dts | =A0 73 ++++++++++++++++++++++=
++
Post by Jonas Bonn
=A0arch/openrisc/include/asm/prom.h =A0 | =A0 92 ++++++++++++++++++++=
++++++++++
Post by Jonas Bonn
=A0arch/openrisc/kernel/prom.c =A0 =A0 =A0 =A0| =A0109 ++++++++++++++=
++++++++++++++++++++++
Post by Jonas Bonn
=A0arch/openrisc/kernel/setup.c =A0 =A0 =A0 | =A0 =A03 +-
=A05 files changed, 353 insertions(+), 2 deletions(-)
=A0create mode 100644 arch/openrisc/boot/dts/atlys.dts
=A0create mode 100644 arch/openrisc/boot/dts/or1ksim.dts
=A0create mode 100644 arch/openrisc/include/asm/prom.h
=A0create mode 100644 arch/openrisc/kernel/prom.c
diff --git a/arch/openrisc/boot/dts/atlys.dts b/arch/openrisc/boot/dt=
s/atlys.dts
Post by Jonas Bonn
new file mode 100644
index 0000000..cd028b9
--- /dev/null
+++ b/arch/openrisc/boot/dts/atlys.dts
@@ -0,0 +1,78 @@
+/dts-v1/;
+/ {
+ =A0 =A0 =A0 compatible =3D "digilent,atlys";
+ =A0 =A0 =A0 #size-cells =3D <1>;
+ =A0 =A0 =A0 #address-cells =3D <1>;
+
+ =A0 =A0 =A0 chosen {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 bootargs =3D "console=3Dtty0 console=3D=
+ =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 memory {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 device_type =3D "memory";
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 reg =3D <0x00000000 0x8000000>;
+ =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 cpus {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,openrisc-1200";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 clock-frequency =3D <50=
000000>;

As Segher metioned, the cpus node should have #address/size-cells and
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+ =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 /* The "soc" node */
+
+ =A0 =A0 =A0 soc {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* Every node with children must specif=
y address-cells and
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0size-cells; they are not inherit=
ed */
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 #address-cells =3D <1>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 #size-cells =3D <1>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 device_type =3D "soc";
Drop device type. device_type only makes sense for systems with real
OpenFirmware. Always use a 'compatible' property to identify a
device. In this case you probably want the compatible list to include
the name of the bus bridge plus "simple-bus".

Does it really make sense to have an SoC node for this board? I
assume this is for a soft CPU system on a Diligent FPGA board. It is
best if the device tree structure matches the actual bus layout of the
chip and/or FPGA design. If the devices are directly on the CPU bus,
then it is better to do without an soc node entirely and just put
everything at the root.
Post by Jonas Bonn
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* This is also a 1:1 mapping but limit=
s size of mappable
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0region to 0x40000000 */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 ranges =3D <0x80000000 0x80000000 0x400=
00000>;
Post by Jonas Bonn
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* An alternative is to specify an empt=
y ranges here to
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0indicates 1:1 mapping of SOC add=
resses (device addresses)
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0to CPU addresses */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*ranges;*/
ranges is a well documented property. You can drop the commented out
empty ranges, and the comments attached to it.
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* This SOC has only a single, implicit=
PIC, so define it
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0early and put the interrupt-pare=
nt property in the 'soc'
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0node so that it's inherited by a=
ll the children using
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0interrupts.
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupt-parent =3D <&pic>;
Put this at the root of the tree so that every node inherits it
(unless explicitly overridden). You may very well end up with devices
at the root of the tree which are not children of the SoC node.
Post by Jonas Bonn
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* Devices with interrupts need a 'pare=
nt', so we need to
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0define a controller */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,or1k-pic";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 #interrupt-cells =3D <1=
;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupt-controller;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* FIXME: device_type i=
s still needed here...
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0should remove re=
quirement from of_serial driver */
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 device_type =3D "serial=
";

device_type should not be necessary at all any more in the current
kernel. You can remove this now.
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,uart", "ns16550a";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 reg =3D <0x90000000 0x1=
00>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupts =3D <2>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 clock-frequency =3D <50=
000000>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,ethoc", "ethoc";

"ethoc" is too generic. Drop it. Also, since this represents a soft
IP core, "opencores,ethoc" should probably include some indication of
which version of the core is instantiated (same goes for the other
"opencores,..." values in this file.

=46inally, every new "compatible" value that is used in this device tre=
e
needs to be documented. Look in Documentation/devicetree/bindings.
The documentation is used to guard against "binding drift" where the
meaning of a compatible property is interpreted differently by
different people.
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 reg =3D <0x92000000 0x5=
3>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupts =3D <4>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 local-mac-address =3D [=
02 de ad be ef 02];

Don't do this. If you don't have a valid Ethernet address, it is
probably better to leave the property out entirely, and let the kernel
randomly generate one.
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,ocfb", "ocfb";

"ocfb" too generic.
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 reg =3D <0x97000000 0x1=
000>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+ =A0 =A0 =A0 };
+};
diff --git a/arch/openrisc/boot/dts/or1ksim.dts b/arch/openrisc/boot/=
dts/or1ksim.dts
Post by Jonas Bonn
new file mode 100644
index 0000000..6a025ea
--- /dev/null
+++ b/arch/openrisc/boot/dts/or1ksim.dts
@@ -0,0 +1,73 @@
+/dts-v1/;
+/ {
+ =A0 =A0 =A0 compatible =3D "opencores,or1ksim";
+ =A0 =A0 =A0 #size-cells =3D <1>;
+ =A0 =A0 =A0 #address-cells =3D <1>;
+
+ =A0 =A0 =A0 chosen {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 bootargs =3D "console=3Duart,mmio,0x900=
00000,115200 debug";
Post by Jonas Bonn
+ =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 memory {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 device_type =3D "memory";
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 reg =3D <0x00000000 0x2000000>;
+ =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 cpus {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,openrisc-1200";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 clock-frequency =3D <20=
000000>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+ =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 /* The "soc" node */
+
+ =A0 =A0 =A0 soc {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* Every node with children must specif=
y address-cells and
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0size-cells; they are not inherit=
ed */
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 #address-cells =3D <1>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 #size-cells =3D <1>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 device_type =3D "soc";
Drop device_type, blah blah blah... Most of my comments on the
previous file apply here too.
Post by Jonas Bonn
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* This is also a 1:1 mapping but limit=
s size of mappable
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0region to 0x40000000 */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 ranges =3D <0x80000000 0x80000000 0x400=
00000>;
Post by Jonas Bonn
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* An alternative is to specify an empt=
y ranges here to
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0indicates 1:1 mapping of SOC add=
resses (device addresses)
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0to CPU addresses */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*ranges;*/
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* This SOC has only a single, implicit=
PIC, so define it
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0early and put the interrupt-pare=
nt property in the 'soc'
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0node so that it's inherited by a=
ll the children using
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0interrupts.
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupt-parent =3D <&pic>;
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* Devices with interrupts need a 'pare=
nt', so we need to
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0define a controller */
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,or1k-pic";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 #interrupt-cells =3D <1=
;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupt-controller;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* FIXME: device_type i=
s still needed here...
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0should remove re=
quirement from of_serial driver */
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 device_type =3D "serial=
";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,uart", "ns16550a";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 reg =3D <0x90000000 0x1=
00>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupts =3D <2>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 clock-frequency =3D <20=
000000>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 compatible =3D "opencor=
es,ethoc", "ethoc";
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 reg =3D <0x92000000 0x5=
3>;
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 interrupts =3D <4>;
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 local-mac-address =3D [=
02 de ad be ef 02];
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 };
+ =A0 =A0 =A0 };
+};
diff --git a/arch/openrisc/include/asm/prom.h b/arch/openrisc/include=
/asm/prom.h
Post by Jonas Bonn
new file mode 100644
index 0000000..3818902
--- /dev/null
+++ b/arch/openrisc/include/asm/prom.h
@@ -0,0 +1,92 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works o=
f
Post by Jonas Bonn
+ * others. =A0All original copyrights apply as per the original sour=
ce
Post by Jonas Bonn
+ * declaration.
+ *
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or mod=
ify
Post by Jonas Bonn
+ * it under the terms of the GNU General Public License as published=
by
Post by Jonas Bonn
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/of.h> =A0/* linux/of.h gets to determine #include or=
dering */
Post by Jonas Bonn
+
+#ifndef _ASM_OPENRISC_PROM_H
+#define _ASM_OPENRISC_PROM_H
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
+
+#include <linux/types.h>
+#include <asm/irq.h>
+#include <asm/atomic.h>
+#include <linux/of_irq.h>
+#include <linux/of_fdt.h>
+#include <linux/of_address.h>
+#include <linux/proc_fs.h>
+#include <linux/platform_device.h>
+#define HAVE_ARCH_DEVTREE_FIXUPS
+
+/* Other Prototypes */
+extern int early_uartlite_console(void);
+
+#ifdef CONFIG_PCI
+/*
+ * PCI <-> OF matching functions
+ * (XXX should these be here?)
+ */
+struct pci_bus;
+struct pci_dev;
+extern int pci_device_from_OF_node(struct device_node *node,
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
=A0 =A0 u8 *bus, u8 *devfn);
Post by Jonas Bonn
+extern struct device_node *pci_busdev_to_OF_node(struct pci_bus *bus=
,
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 int devfn);
Post by Jonas Bonn
+extern struct device_node *pci_device_to_OF_node(struct pci_dev *dev=
);
Post by Jonas Bonn
+extern void pci_create_OF_bus_map(void);
+#endif
+
+/* Parse the ibm,dma-window property of an OF node into the busno, p=
hys and
Post by Jonas Bonn
+ * size parameters.
+ */
+void of_parse_dma_window(struct device_node *dn, const void *dma_win=
dow_prop,
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 unsigned long *busno, unsigned long *ph=
ys, unsigned long *size);
Post by Jonas Bonn
+
+extern void kdump_move_device_tree(void);
+
+/* CPU OF node matching */
+struct device_node *of_get_cpu_node(int cpu, unsigned int *thread);
+
+/* Get the MAC address */
+extern const void *of_get_mac_address(struct device_node *np);
+
+/**
+ * of_irq_map_pci - Resolve the interrupt for a PCI device
+ *
+ * This function resolves the PCI interrupt for a given PCI device. =
If a
Post by Jonas Bonn
+ * device-node exists for a given pci_dev, it will use normal OF tre=
e
Post by Jonas Bonn
+ * walking. If not, it will implement standard swizzling and walk up=
the
Post by Jonas Bonn
+ * PCI tree until an device-node is found, at which point it will fi=
nish
Post by Jonas Bonn
+ * resolving using the OF tree walking.
+ */
+struct pci_dev;
+extern int of_irq_map_pci(struct pci_dev *pdev, struct of_irq *out_i=
rq);
Post by Jonas Bonn
+
+/* This routine is here to provide compatibility with how powerpc
+ * handles IRQ mapping for OF device nodes. =A0We precompute and per=
manently
Post by Jonas Bonn
+ * register them in the platform_device objects, whereas powerpc com=
putes them
Post by Jonas Bonn
+ * on request.
+ */
+static inline void irq_dispose_mapping(unsigned int virq)
+{
+}
The need for this should hopefully go away in Linux v3.1 when I get
the irq_domain stuff merged.
Post by Jonas Bonn
+
+#endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+#endif /* _ASM_OPENRISC_PROM_H */
diff --git a/arch/openrisc/kernel/prom.c b/arch/openrisc/kernel/prom.=
c
Post by Jonas Bonn
new file mode 100644
index 0000000..2bb5a8c
--- /dev/null
+++ b/arch/openrisc/kernel/prom.c
@@ -0,0 +1,109 @@
+/*
+ * OpenRISC prom.c
+ *
+ * Linux architectural port borrowing liberally from similar works o=
f
Post by Jonas Bonn
+ * others. =A0All original copyrights apply as per the original sour=
ce
Post by Jonas Bonn
+ * declaration.
+ *
+ *
+ * =A0 =A0 =A0This program is free software; you can redistribute it=
and/or
Post by Jonas Bonn
+ * =A0 =A0 =A0modify it under the terms of the GNU General Public Li=
cense
Post by Jonas Bonn
+ * =A0 =A0 =A0as published by the Free Software Foundation; either v=
ersion
Post by Jonas Bonn
+ * =A0 =A0 =A02 of the License, or (at your option) any later versio=
n.
Post by Jonas Bonn
+ *
+ * Architecture specific procedures for creating, accessing and
+ * interpreting the device tree.
+ *
+ */
+
+#include <stdarg.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/threads.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/stringify.h>
+#include <linux/delay.h>
+#include <linux/initrd.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/kexec.h>
+#include <linux/debugfs.h>
+#include <linux/irq.h>
+#include <linux/memblock.h>
+#include <linux/of_fdt.h>
+
+#include <asm/prom.h>
+#include <asm/page.h>
+#include <asm/processor.h>
+#include <asm/irq.h>
+#include <linux/io.h>
+#include <asm/system.h>
+#include <asm/mmu.h>
+#include <asm/pgtable.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+
+extern char cmd_line[COMMAND_LINE_SIZE];
+
+void __init early_init_dt_add_memory_arch(u64 base, u64 size)
+{
+ =A0 =A0 =A0 size &=3D PAGE_MASK;
+ =A0 =A0 =A0 memblock_add(base, size);
+}
+
+void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align)
+{
+ =A0 =A0 =A0 return __va(memblock_alloc(size, align));
+}
+
+void __init early_init_devtree(void *params)
+{
+ =A0 =A0 =A0 u8 *alloc;
+
+ =A0 =A0 =A0 /* Setup flat device-tree pointer */
+ =A0 =A0 =A0 initial_boot_params =3D params;
+
+
+ =A0 =A0 =A0 /* Retrieve various informations from the /chosen node =
of the
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0* device-tree, including the platform type, initrd l=
ocation and
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0* size, TCE reserve, and more ...
+ =A0 =A0 =A0 =A0*/
+ =A0 =A0 =A0 of_scan_flat_dt(early_init_dt_scan_chosen, cmd_line);
+
+ =A0 =A0 =A0 /* Scan memory nodes and rebuild MEMBLOCKs */
+ =A0 =A0 =A0 memblock_init();
+ =A0 =A0 =A0 of_scan_flat_dt(early_init_dt_scan_root, NULL);
+ =A0 =A0 =A0 of_scan_flat_dt(early_init_dt_scan_memory, NULL);
+
+ =A0 =A0 =A0 /* Save command line for /proc/cmdline and then parse p=
arameters */
Post by Jonas Bonn
+ =A0 =A0 =A0 strlcpy(boot_command_line, cmd_line, COMMAND_LINE_SIZE)=
;
Post by Jonas Bonn
+// =A0 =A0 parse_early_param();
This should just be removed?
Post by Jonas Bonn
+
+ =A0 =A0 =A0 memblock_analyze();
+
+ =A0 =A0 =A0 /* We must copy the flattend device tree from init memo=
ry to regular
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0* memory because the device tree references the stri=
ngs in it
Post by Jonas Bonn
+ =A0 =A0 =A0 =A0* directly.
+ =A0 =A0 =A0 =A0*/
+
+ =A0 =A0 =A0 alloc =3D __va(memblock_alloc(initial_boot_params->tota=
lsize, PAGE_SIZE));
Post by Jonas Bonn
+
+ =A0 =A0 =A0 memcpy(alloc, initial_boot_params, initial_boot_params-=
totalsize);
+
+ =A0 =A0 =A0 initial_boot_params =3D (void *) alloc;
If alloc gets declared as a 'void*', then this cast can be removed.
Post by Jonas Bonn
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+void __init early_init_dt_setup_initrd_arch(unsigned long start,
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 unsigned long end)
+{
+ =A0 =A0 =A0 initrd_start =3D (unsigned long)__va(start);
+ =A0 =A0 =A0 initrd_end =3D (unsigned long)__va(end);
+ =A0 =A0 =A0 initrd_below_start_ok =3D 1;
+}
+#endif
diff --git a/arch/openrisc/kernel/setup.c b/arch/openrisc/kernel/setu=
p.c
Post by Jonas Bonn
index 49342e9..6ce6583 100644
--- a/arch/openrisc/kernel/setup.c
+++ b/arch/openrisc/kernel/setup.c
@@ -173,8 +173,7 @@ void __init setup_cpuinfo(void)
=A0 =A0 =A0 =A0unsigned long iccfgr,dccfgr;
=A0 =A0 =A0 =A0unsigned long cache_set_size, cache_ways;;
- =A0 =A0 =A0 cpu =3D (struct device_node *) of_find_compatible_node(=
NULL,
Post by Jonas Bonn
- =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
=A0 =A0 NULL, "opencores,openrisc-1200");
Post by Jonas Bonn
+ =A0 =A0 =A0 cpu =3D of_find_compatible_node(NULL, NULL, "opencores,=
openrisc-1200");

This looks odd. Is it a stale hunk from a previous patch version?
Post by Jonas Bonn
=A0 =A0 =A0 =A0if (!cpu) {
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0panic("No compatible CPU found in devi=
ce tree...\n");
Post by Jonas Bonn
=A0 =A0 =A0 =A0}
--
1.7.4.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kerne=
l" in
Post by Jonas Bonn
More majordomo info at =A0http://vger.kernel.org/majordomo-info.html
Please read the FAQ at =A0http://www.tux.org/lkml/
--=20
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.
Benjamin Herrenschmidt
2011-07-03 21:39:38 UTC
Permalink
Post by Jonas Bonn
Post by Jonas Bonn
+
+#ifdef CONFIG_PCI
+/*
+ * PCI <-> OF matching functions
+ * (XXX should these be here?)
+ */
+struct pci_bus;
+struct pci_dev;
+extern int pci_device_from_OF_node(struct device_node *node,
+ u8 *bus, u8 *devfn);
+extern struct device_node *pci_busdev_to_OF_node(struct pci_bus
*bus,
Post by Jonas Bonn
+ int devfn);
+extern struct device_node *pci_device_to_OF_node(struct pci_dev
*dev);
Post by Jonas Bonn
+extern void pci_create_OF_bus_map(void);
+#endif
Don't copy these from powermac or microblaze. the OF_bus_map is
something that should not spread :-)

You get all you need of the above from generic code with my patches to
PCI <-> OF matching, which should be in linux-next and are going to be
merged in the next merge window.

Cheers,
Ben.
Jonas Bonn
2011-07-04 04:58:28 UTC
Permalink
Post by Grant Likely
Does it really make sense to have an SoC node for this board? I
assume this is for a soft CPU system on a Diligent FPGA board. It is
best if the device tree structure matches the actual bus layout of the
chip and/or FPGA design. If the devices are directly on the CPU bus,
then it is better to do without an soc node entirely and just put
everything at the root.
Not sure I quite understand the distinction here. Yes, it's a soft CPU,
but conceptually it has an internal bus (Wishbone) comparable to the
Avalon bus that the peripherals sit on. I'd say it's an SoC; but I need
to ask: what constitutes a valid use of the "soc" node?
Post by Grant Likely
Post by Jonas Bonn
diff --git a/arch/openrisc/kernel/setup.c b/arch/openrisc/kernel/setup.c
index 49342e9..6ce6583 100644
--- a/arch/openrisc/kernel/setup.c
+++ b/arch/openrisc/kernel/setup.c
@@ -173,8 +173,7 @@ void __init setup_cpuinfo(void)
unsigned long iccfgr,dccfgr;
unsigned long cache_set_size, cache_ways;;
- cpu = (struct device_node *) of_find_compatible_node(NULL,
- NULL, "opencores,openrisc-1200");
+ cpu = of_find_compatible_node(NULL, NULL, "opencores,openrisc-1200");
This looks odd. Is it a stale hunk from a previous patch version?
Hmmm... no, it's a hunk that should have been in patch 01/19 of the
series. That said, there's a bit of devicetree code in that patch too.
Can I just say, "please have a look there, too", or do I need to pull
those bits out into the devicetree patch?

I've pasted in just the relevant parts from patch 01/19 below for ease
of review.

/Jonas
+static inline unsigned int fcpu(struct device_node *cpu, char *n)
+{
+ const int *val;
+ return (val = of_get_property(cpu, n, NULL)) ? *val : 0;
+}
+
+void __init setup_cpuinfo(void)
+{
+ struct device_node *cpu = NULL;
+ unsigned long iccfgr,dccfgr;
+ unsigned long cache_set_size, cache_ways;;
+
+ cpu = (struct device_node *) of_find_compatible_node(NULL,
+ NULL,
"opencores,openrisc-1200");
+ if (!cpu) {
+ panic("No compatible CPU found in device tree...\n");
+ }
+
+ iccfgr = mfspr(SPR_ICCFGR);
+ cache_ways = 1 << (iccfgr & SPR_ICCFGR_NCW);
+ cache_set_size = 1 << ((iccfgr & SPR_ICCFGR_NCS) >> 3);
+ cpuinfo.icache_block_size = 16 << ((iccfgr & SPR_ICCFGR_CBS) >>
7);
+ cpuinfo.icache_size = cache_set_size * cache_ways *
cpuinfo.icache_block_size;
+
+ dccfgr = mfspr(SPR_DCCFGR);
+ cache_ways = 1 << (dccfgr & SPR_DCCFGR_NCW);
+ cache_set_size = 1 << ((dccfgr & SPR_DCCFGR_NCS) >> 3);
+ cpuinfo.dcache_block_size = 16 << ((dccfgr & SPR_DCCFGR_CBS) >>
7);
+ cpuinfo.dcache_size = cache_set_size * cache_ways *
cpuinfo.dcache_block_size;
+
+ cpuinfo.clock_frequency = fcpu(cpu, "clock-frequency");
+
+ of_node_put(cpu);
+
+ print_cpuinfo();
+}
+
+/**
+ * or32_early_setup
+ *
+ * Handles the pointer to the device tree that this kernel is to use
+ * for establishing the available platform devices.
+ *
+ * For now, this is limited to using the built-in device tree. In the
future,
+ * it is intended that this function will take a pointer to the device
tree
+ * that is potentially built-in, but potentially also passed in by the
+ * bootloader, or discovered by some equally clever means...
+ */
+
+void __init or32_early_setup(void) {
+
+ early_init_devtree((void *) __dtb_start);
+
+ printk(KERN_INFO "Compiled-in FDT at 0x%08x\n",
+ (unsigned int) __dtb_start);
+}
+
+const struct of_device_id openrisc_bus_ids[] = {
+ { .type = "soc", },
+ { .compatible = "soc", },
+ {},
+};
+
+static int __init openrisc_device_probe(void)
+{
+ of_platform_bus_probe(NULL, openrisc_bus_ids, NULL);
+ return 0;
+}
+device_initcall(openrisc_device_probe);
Grant Likely
2011-07-04 05:35:52 UTC
Permalink
Post by Jonas Bonn
Post by Grant Likely
Does it really make sense to have an SoC node for this board? I
assume this is for a soft CPU system on a Diligent FPGA board. It is
best if the device tree structure matches the actual bus layout of the
chip and/or FPGA design. If the devices are directly on the CPU bus,
then it is better to do without an soc node entirely and just put
everything at the root.
Not sure I quite understand the distinction here. Yes, it's a soft CPU,
but conceptually it has an internal bus (Wishbone) comparable to the
Avalon bus that the peripherals sit on. I'd say it's an SoC; but I need
to ask: what constitutes a valid use of the "soc" node?
An soc node is really just a convenient collection node for devices on
an SoC. It can be used, or not, depending on the SoC and whoever is
writing the bindings for it.

Regardless, the device tree hierarchy should reflect the actual layout
of the buses, and devices on the same bus segment should be siblings
in the tree. Often when I work on an FPGA system, I don't even bother
with an soc node because it is kind of meaningless, and I just start
with knowing that the root of the tree represents the CPU local bus,
and then creating nodes for each bridge to another bus. If all the
devices are on the same bus segment, then it looks pretty much like
what you wrote except that the node name for the bridge would probably
be something like "wishbone-bus".

Basically, I'm suggesting to make the DT model as representative of
the FPGA design as possible.
Post by Jonas Bonn
Post by Grant Likely
Post by Jonas Bonn
diff --git a/arch/openrisc/kernel/setup.c b/arch/openrisc/kernel/setup.c
index 49342e9..6ce6583 100644
--- a/arch/openrisc/kernel/setup.c
+++ b/arch/openrisc/kernel/setup.c
@@ -173,8 +173,7 @@ void __init setup_cpuinfo(void)
unsigned long iccfgr,dccfgr;
unsigned long cache_set_size, cache_ways;;
- cpu = (struct device_node *) of_find_compatible_node(NULL,
- NULL, "opencores,openrisc-1200");
+ cpu = of_find_compatible_node(NULL, NULL, "opencores,openrisc-1200");
This looks odd. Is it a stale hunk from a previous patch version?
Hmmm... no, it's a hunk that should have been in patch 01/19 of the
series. That said, there's a bit of devicetree code in that patch too.
Can I just say, "please have a look there, too", or do I need to pull
those bits out into the devicetree patch?
No, you don't need to pull it out into this patch, but when you respin
you should fix the series to not include this hunk in patch 2.
Post by Jonas Bonn
I've pasted in just the relevant parts from patch 01/19 below for ease
of review.
/Jonas
+static inline unsigned int fcpu(struct device_node *cpu, char *n)
+{
+ const int *val;
+ return (val = of_get_property(cpu, n, NULL)) ? *val : 0;
be32_to_cpup(val) should be used here.

However, of_property_read_u32() is now available to be used with the
correct data size checking.
Post by Jonas Bonn
+}
+
+void __init setup_cpuinfo(void)
+{
+ struct device_node *cpu = NULL;
No need to initialize this to NULL, it is immediately set at the start
of the function.
Post by Jonas Bonn
+ unsigned long iccfgr,dccfgr;
+ unsigned long cache_set_size, cache_ways;;
+
+ cpu = (struct device_node *) of_find_compatible_node(NULL,
+ NULL,
"opencores,openrisc-1200");
Why the cast? I looks like it can be dropped.
Post by Jonas Bonn
+ if (!cpu) {
+ panic("No compatible CPU found in device tree...\n");
+ }
+
+ iccfgr = mfspr(SPR_ICCFGR);
+ cache_ways = 1 << (iccfgr & SPR_ICCFGR_NCW);
+ cache_set_size = 1 << ((iccfgr & SPR_ICCFGR_NCS) >> 3);
+ cpuinfo.icache_block_size = 16 << ((iccfgr & SPR_ICCFGR_CBS) >>
7);
+ cpuinfo.icache_size = cache_set_size * cache_ways *
cpuinfo.icache_block_size;
+
+ dccfgr = mfspr(SPR_DCCFGR);
+ cache_ways = 1 << (dccfgr & SPR_DCCFGR_NCW);
+ cache_set_size = 1 << ((dccfgr & SPR_DCCFGR_NCS) >> 3);
+ cpuinfo.dcache_block_size = 16 << ((dccfgr & SPR_DCCFGR_CBS) >>
7);
+ cpuinfo.dcache_size = cache_set_size * cache_ways *
cpuinfo.dcache_block_size;
+
+ cpuinfo.clock_frequency = fcpu(cpu, "clock-frequency");
+
+ of_node_put(cpu);
+
+ print_cpuinfo();
+}
+
+/**
+ * or32_early_setup
+ *
+ * Handles the pointer to the device tree that this kernel is to use
+ * for establishing the available platform devices.
+ *
+ * For now, this is limited to using the built-in device tree. In the
future,
+ * it is intended that this function will take a pointer to the device
tree
+ * that is potentially built-in, but potentially also passed in by the
+ * bootloader, or discovered by some equally clever means...
+ */
+
+void __init or32_early_setup(void) {
+
+ early_init_devtree((void *) __dtb_start);
Isn't __dtb_start already a void*?
Post by Jonas Bonn
+
+ printk(KERN_INFO "Compiled-in FDT at 0x%08x\n",
pr_info()

use %p for pointers.
Post by Jonas Bonn
+ (unsigned int) __dtb_start);
+}
+
+const struct of_device_id openrisc_bus_ids[] = {
+ { .type = "soc", },
Never use .type for new code. Only .compatible should be checked.
Post by Jonas Bonn
+ { .compatible = "soc", },
Use "simple-bus". "soc" isn't defined to mean anything (see the ePAPR
spec).
Post by Jonas Bonn
+ {},
+};
+
+static int __init openrisc_device_probe(void)
+{
+ of_platform_bus_probe(NULL, openrisc_bus_ids, NULL);
Use of_platform_populate() instead (in linux-next, or the
devicetree/next branch at git://git.secretlab.ca/git/linux-2.6).
of_platform_populate() is better at handling devices at the root of
the device tree, and it properly ignores nodes that don't have a
compatible property.

g.
Jonas Bonn
2011-07-02 21:15:38 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/Kconfig | 207 +++++++++++++++++++++++++++++++
arch/openrisc/Makefile | 66 ++++++++++
arch/openrisc/boot/Makefile | 15 +++
arch/openrisc/configs/or1ksim_defconfig | 67 ++++++++++
arch/openrisc/include/asm/Kbuild | 64 ++++++++++
arch/openrisc/kernel/Makefile | 14 ++
arch/openrisc/kernel/vmlinux.lds.S | 115 +++++++++++++++++
arch/openrisc/lib/Makefile | 5 +
arch/openrisc/mm/Makefile | 5 +
9 files changed, 558 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/Kconfig
create mode 100644 arch/openrisc/Makefile
create mode 100644 arch/openrisc/boot/Makefile
create mode 100644 arch/openrisc/configs/or1ksim_defconfig
create mode 100644 arch/openrisc/include/asm/Kbuild
create mode 100644 arch/openrisc/kernel/Makefile
create mode 100644 arch/openrisc/kernel/vmlinux.lds.S
create mode 100644 arch/openrisc/lib/Makefile
create mode 100644 arch/openrisc/mm/Makefile

diff --git a/arch/openrisc/Kconfig b/arch/openrisc/Kconfig
new file mode 100644
index 0000000..4558baf
--- /dev/null
+++ b/arch/openrisc/Kconfig
@@ -0,0 +1,207 @@
+#
+# For a description of the syntax of this configuration file,
+# see Documentation/kbuild/config-language.txt.
+#
+
+config OPENRISC
+ def_bool y
+ select OF
+ select OF_EARLY_FLATTREE
+ select HAVE_MEMBLOCK
+ select ARCH_WANT_OPTIONAL_GPIOLIB
+ select HAVE_ARCH_TRACEHOOK
+ select HAVE_GENERIC_HARDIRQS
+ select GENERIC_IRQ_CHIP
+ select GENERIC_IRQ_PROBE
+ select GENERIC_IRQ_SHOW
+ select GENERIC_IOMAP
+
+config MMU
+ def_bool y
+
+config WISHBONE_BUS_BIG_ENDIAN
+ def_bool y
+
+config SYMBOL_PREFIX
+ string
+ default ""
+
+config HAVE_DMA_ATTRS
+ def_bool y
+
+config UID16
+ def_bool y
+
+config RWSEM_GENERIC_SPINLOCK
+ def_bool y
+
+config RWSEM_XCHGADD_ALGORITHM
+ def_bool n
+
+config GENERIC_HWEIGHT
+ def_bool y
+
+config GENERIC_IOMAP
+ def_bool y
+
+config NO_IOPORT
+ def_bool y
+
+config GENERIC_GPIO
+ def_bool y
+
+config GENERIC_CLOCKEVENTS
+ def_bool y
+
+config TRACE_IRQFLAGS_SUPPORT
+ def_bool y
+
+# For now, use generic checksum functions
+#These can be reimplemented in assembly later if so inclined
+config GENERIC_CSUM
+ def_bool y
+
+config GENERIC_FIND_NEXT_BIT
+ def_bool y
+
+source "init/Kconfig"
+
+
+menu "Processor type and features"
+
+choice
+ prompt "Subarchitecture"
+ default OR1K_1200
+
+config OR1K_1200
+ bool "OR1200"
+ help
+ Generic OpenRISC 1200 architecture
+
+endchoice
+
+config OPENRISC_BUILTIN_DTB
+ string "Builtin DTB"
+ default ""
+
+menu "Class II Instructions"
+
+config OPENRISC_HAVE_INST_FF1
+ bool "Have instruction l.ff1"
+ default y
+ help
+ Select this if your implementation has the Class II instruction l.ff1
+
+config OPENRISC_HAVE_INST_FL1
+ bool "Have instruction l.fl1"
+ default y
+ help
+ Select this if your implementation has the Class II instruction l.fl1
+
+config OPENRISC_HAVE_INST_MUL
+ bool "Have instruction l.mul for hardware multiply"
+ default y
+ help
+ Select this if your implementation has a hardware multiply instruction
+
+config OPENRISC_HAVE_INST_DIV
+ bool "Have instruction l.div for hardware divide"
+ default y
+ help
+ Select this if your implementation has a hardware divide instruction
+endmenu
+
+
+source "kernel/time/Kconfig"
+source kernel/Kconfig.hz
+source kernel/Kconfig.preempt
+source "mm/Kconfig"
+
+config OPENRISC_NO_SPR_SR_DSX
+ bool "use SPR_SR_DSX software emulation" if OR1K_1200
+ default y
+ help
+ SPR_SR_DSX bit is status register bit indicating whether
+ the last exception has happened in delay slot.
+
+ OpenRISC architecture makes it optional to have it implemented
+ in hardware and the OR1200 does not have it.
+
+ Say N here if you know that your OpenRISC processor has
+ SPR_SR_DSX bit implemented. Say Y if you are unsure.
+
+config CMDLINE
+ string "Default kernel command string"
+ default ""
+ help
+ On some architectures there is currently no way for the boot loader
+ to pass arguments to the kernel. For these architectures, you should
+ supply some command-line options at build time by entering them
+ here.
+
+menu "Debugging options"
+
+config DEBUG_STACKOVERFLOW
+ bool "Check for kernel stack overflow"
+ default y
+ help
+ Make extra checks for space avaliable on stack in some
+ critical functions. This will cause kernel to run a bit slower,
+ but will catch most of kernel stack overruns and exit gracefuly.
+
+ Say Y if you are unsure.
+
+config JUMP_UPON_UNHANDLED_EXCEPTION
+ bool "Try to die gracefully"
+ default y
+ help
+ Now this puts kernel into infinite loop after first oops. Till
+ your kernel crashes this doesn't have any influence.
+
+ Say Y if you are unsure.
+
+config OPENRISC_EXCEPTION_DEBUG
+ bool "Print processor state at each exception"
+ default n
+ help
+ This option will make your kernel unusable for all but kernel
+ debugging.
+
+ Say N if you are unsure.
+
+config OPENRISC_ESR_EXCEPTION_BUG_CHECK
+ bool "Check for possible ESR exception bug"
+ default n
+ help
+ This option enables some checks that might expose some problems
+ in kernel.
+
+ Say N if you are unsure.
+
+endmenu
+
+endmenu
+
+menu "Executable file formats"
+
+source "fs/Kconfig.binfmt"
+
+endmenu
+
+source "net/Kconfig"
+
+source "drivers/Kconfig"
+
+source "fs/Kconfig"
+
+source "security/Kconfig"
+
+source "crypto/Kconfig"
+
+source "lib/Kconfig"
+
+menu "Kernel hacking"
+
+source "lib/Kconfig.debug"
+
+endmenu
diff --git a/arch/openrisc/Makefile b/arch/openrisc/Makefile
new file mode 100644
index 0000000..9881c45
--- /dev/null
+++ b/arch/openrisc/Makefile
@@ -0,0 +1,66 @@
+# BK Id: %F% %I% %G% %U% %#%
+#
+# This file is included by the global makefile so that you can add your own
+# architecture-specific flags and dependencies. Remember to do have actions
+# for "archclean" and "archdep" for cleaning up and making dependencies for
+# this architecture
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License. See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+# Copyright (C) 1994 by Linus Torvalds
+
+# Changes for or32 by Matjaz Breskvar
+# Based on:
+# arch/i386/Makefile
+
+KBUILD_DEFCONFIG := or1ksim_defconfig
+
+LDFLAGS :=
+OBJCOPYFLAGS := -O binary -R .note -R .comment -S
+LDFLAGS_vmlinux :=
+LIBGCC = $(shell $(CC) $(CFLAGS) -print-file-name=libgcc.a)
+
+#KBUILD_CFLAGS += -pipe -g -mredzone=0
+KBUILD_CFLAGS += -pipe -ffixed-r10
+#KBUILD_CFLAGS += -fsigned-char -pipe -ffixed-r10 \
+# -ffixed-r16 -ffixed-r17 -ffixed-r18 -ffixed-r19 \
+# -ffixed-r20 -ffixed-r21 -ffixed-r22 -ffixed-r23 \
+# -ffixed-r24 -ffixed-r25 -ffixed-r26 -ffixed-r27 \
+# -ffixed-r28 -ffixed-r29 -ffixed-r30 -ffixed-r31 \
+# -Wno-uninitialized
+
+#KBUILD_CFLAGS += $(call cc-option,-mredzone=128)
+#KBUILD_CFLAGS += -mredzone=0
+
+ifeq ($(CONFIG_OPENRISC_HAVE_INST_MUL),y)
+ KBUILD_CFLAGS += $(call cc-option,-mhard-mul)
+else
+ KBUILD_CFLAGS += $(call cc-option,-msoft-mul)
+endif
+
+ifeq ($(CONFIG_OPENRISC_HAVE_INST_DIV),y)
+ KBUILD_CFLAGS += $(call cc-option,-mhard-div)
+else
+ KBUILD_CFLAGS += $(call cc-option,-msoft-div)
+endif
+
+
+
+head-y := arch/openrisc/kernel/head.o arch/openrisc/kernel/init_task.o
+
+
+core-y += arch/openrisc/lib/ \
+ arch/openrisc/kernel/ \
+ arch/openrisc/mm/
+libs-y += $(LIBGCC)
+
+ifneq '$(CONFIG_OPENRISC_BUILTIN_DTB)' '""'
+BUILTIN_DTB := y
+else
+BUILTIN_DTB := n
+endif
+core-$(BUILTIN_DTB) += arch/openrisc/boot/
+
+all: vmlinux
diff --git a/arch/openrisc/boot/Makefile b/arch/openrisc/boot/Makefile
new file mode 100644
index 0000000..98ca185
--- /dev/null
+++ b/arch/openrisc/boot/Makefile
@@ -0,0 +1,15 @@
+
+
+ifneq '$(CONFIG_OPENRISC_BUILTIN_DTB)' '""'
+BUILTIN_DTB := $(patsubst "%",%,$(CONFIG_OPENRISC_BUILTIN_DTB)).dtb.o
+else
+BUILTIN_DTB :=
+endif
+obj-y += $(BUILTIN_DTB)
+
+clean-files := *.dtb.S
+
+#DTC_FLAGS ?= -p 1024
+
+$(obj)/%.dtb: $(src)/dts/%.dts
+ $(call cmd,dtc)
diff --git a/arch/openrisc/configs/or1ksim_defconfig b/arch/openrisc/configs/or1ksim_defconfig
new file mode 100644
index 0000000..af52508
--- /dev/null
+++ b/arch/openrisc/configs/or1ksim_defconfig
@@ -0,0 +1,67 @@
+CONFIG_CROSS_COMPILE="or32-linux-"
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_BLK_DEV_INITRD=y
+# CONFIG_RD_GZIP is not set
+CONFIG_EXPERT=y
+# CONFIG_SYSCTL_SYSCALL is not set
+# CONFIG_KALLSYMS is not set
+# CONFIG_ELF_CORE is not set
+# CONFIG_EPOLL is not set
+# CONFIG_TIMERFD is not set
+# CONFIG_EVENTFD is not set
+# CONFIG_AIO is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+# CONFIG_COMPAT_BRK is not set
+CONFIG_SLOB=y
+CONFIG_MODULES=y
+# CONFIG_BLOCK is not set
+CONFIG_OPENRISC_BUILTIN_DTB="or1ksim"
+CONFIG_NO_HZ=y
+CONFIG_HZ_100=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+# CONFIG_INET_DIAG is not set
+CONFIG_TCP_CONG_ADVANCED=y
+# CONFIG_TCP_CONG_BIC is not set
+# CONFIG_TCP_CONG_CUBIC is not set
+# CONFIG_TCP_CONG_WESTWOOD is not set
+# CONFIG_TCP_CONG_HTCP is not set
+# CONFIG_IPV6 is not set
+# CONFIG_WIRELESS is not set
+CONFIG_DEVTMPFS=y
+CONFIG_DEVTMPFS_MOUNT=y
+# CONFIG_PREVENT_FIRMWARE_BUILD is not set
+# CONFIG_FW_LOADER is not set
+CONFIG_PROC_DEVICETREE=y
+CONFIG_NETDEVICES=y
+CONFIG_MICREL_PHY=y
+CONFIG_NET_ETHERNET=y
+CONFIG_ETHOC=y
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+# CONFIG_WLAN is not set
+# CONFIG_INPUT is not set
+# CONFIG_SERIO is not set
+# CONFIG_VT is not set
+# CONFIG_LEGACY_PTYS is not set
+# CONFIG_DEVKMEM is not set
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_SERIAL_OF_PLATFORM=y
+# CONFIG_HW_RANDOM is not set
+# CONFIG_HWMON is not set
+# CONFIG_MFD_SUPPORT is not set
+# CONFIG_FB_OC_SHMEM is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_DNOTIFY is not set
+CONFIG_TMPFS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_ENABLE_WARN_DEPRECATED is not set
+# CONFIG_ENABLE_MUST_CHECK is not set
diff --git a/arch/openrisc/include/asm/Kbuild b/arch/openrisc/include/asm/Kbuild
new file mode 100644
index 0000000..11162e6
--- /dev/null
+++ b/arch/openrisc/include/asm/Kbuild
@@ -0,0 +1,64 @@
+include include/asm-generic/Kbuild.asm
+
+header-y += spr_defs.h
+
+generic-y += atomic.h
+generic-y += auxvec.h
+generic-y += bitsperlong.h
+generic-y += bug.h
+generic-y += bugs.h
+generic-y += cacheflush.h
+generic-y += checksum.h
+generic-y += cmpxchg.h
+generic-y += cmpxchg-local.h
+generic-y += cpumask.h
+generic-y += cputime.h
+generic-y += current.h
+generic-y += device.h
+generic-y += div64.h
+generic-y += dma.h
+generic-y += emergency-restart.h
+generic-y += errno.h
+generic-y += fb.h
+generic-y += fcntl.h
+generic-y += ftrace.h
+generic-y += futex.h
+generic-y += hardirq.h
+generic-y += hw_irq.h
+generic-y += ioctl.h
+generic-y += ioctls.h
+generic-y += ipcbuf.h
+generic-y += irq_regs.h
+generic-y += kdebug.h
+generic-y += kmap_types.h
+generic-y += local.h
+generic-y += mman.h
+generic-y += module.h
+generic-y += msgbuf.h
+generic-y += pci.h
+generic-y += percpu.h
+generic-y += poll.h
+generic-y += posix_types.h
+generic-y += resource.h
+generic-y += rmap.h
+generic-y += scatterlist.h
+generic-y += sections.h
+generic-y += segment.h
+generic-y += sembuf.h
+generic-y += setup.h
+generic-y += shmbuf.h
+generic-y += shmparam.h
+generic-y += siginfo.h
+generic-y += signal.h
+generic-y += socket.h
+generic-y += sockios.h
+generic-y += statfs.h
+generic-y += stat.h
+generic-y += string.h
+generic-y += swab.h
+generic-y += termbits.h
+generic-y += termios.h
+generic-y += topology.h
+generic-y += types.h
+generic-y += ucontext.h
+generic-y += user.h
diff --git a/arch/openrisc/kernel/Makefile b/arch/openrisc/kernel/Makefile
new file mode 100644
index 0000000..9a4c270
--- /dev/null
+++ b/arch/openrisc/kernel/Makefile
@@ -0,0 +1,14 @@
+#
+# Makefile for the linux kernel.
+#
+
+extra-y := head.o vmlinux.lds init_task.o
+
+obj-y := setup.o idle.o or32_ksyms.o process.o dma.o \
+ traps.o time.o irq.o entry.o ptrace.o signal.o sys_or32.o \
+ sys_call_table.o
+
+obj-$(CONFIG_MODULES) += module.o
+obj-$(CONFIG_OF) += prom.o
+
+clean:
diff --git a/arch/openrisc/kernel/vmlinux.lds.S b/arch/openrisc/kernel/vmlinux.lds.S
new file mode 100644
index 0000000..2d69a85
--- /dev/null
+++ b/arch/openrisc/kernel/vmlinux.lds.S
@@ -0,0 +1,115 @@
+/*
+ * OpenRISC vmlinux.lds.S
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * ld script for OpenRISC architecture
+ */
+
+/* TODO
+ * - clean up __offset & stuff
+ * - change all 8192 aligment to PAGE !!!
+ * - recheck if all aligments are really needed
+ */
+
+# define LOAD_OFFSET PAGE_OFFSET
+# define LOAD_BASE PAGE_OFFSET
+
+#include <asm/page.h>
+#include <asm/cache.h>
+#include <asm-generic/vmlinux.lds.h>
+
+OUTPUT_FORMAT("elf32-or32", "elf32-or32", "elf32-or32")
+jiffies = jiffies_64 + 4;
+
+SECTIONS
+{
+ /* Read-only sections, merged into text segment: */
+ . = LOAD_BASE ;
+
+ /* _s_kernel_ro must be page aligned */
+ . = ALIGN(PAGE_SIZE);
+ _s_kernel_ro = .;
+
+ .text : AT(ADDR(.text) - LOAD_OFFSET)
+ {
+ _stext = .;
+ TEXT_TEXT
+ SCHED_TEXT
+ LOCK_TEXT
+ KPROBES_TEXT
+ IRQENTRY_TEXT
+ *(.fixup)
+ *(.text.__*)
+ _etext = .;
+ }
+ /* TODO: Check if fixup and text.__* are really necessary
+ * fixup is definitely necessary
+ */
+
+ _sdata = .;
+
+ /* Page alignment required for RO_DATA_SECTION */
+ RO_DATA_SECTION(PAGE_SIZE)
+ _e_kernel_ro = .;
+
+ /* Whatever comes after _e_kernel_ro had better be page-aligend, too */
+
+ /* 32 here is cacheline size... recheck this */
+ RW_DATA_SECTION(32, PAGE_SIZE, PAGE_SIZE)
+
+ _edata = .;
+
+ EXCEPTION_TABLE(4)
+ NOTES
+
+ /* Init code and data */
+ . = ALIGN(PAGE_SIZE);
+ __init_begin = .;
+
+ HEAD_TEXT_SECTION
+
+ /* Page aligned */
+ INIT_TEXT_SECTION(PAGE_SIZE)
+
+ /* Align __setup_start on 16 byte boundary */
+ INIT_DATA_SECTION(16)
+
+ PERCPU_SECTION(L1_CACHE_BYTES)
+
+ __init_end = .;
+
+ . = ALIGN(PAGE_SIZE);
+ .initrd : AT(ADDR(.initrd) - LOAD_OFFSET)
+ {
+ __initrd_start = .;
+ *(.initrd)
+ __initrd_end = .;
+ FILL (0);
+ . = ALIGN (PAGE_SIZE);
+ }
+
+ __vmlinux_end = .; /* last address of the physical file */
+
+ BSS_SECTION(0, 0, 0x20)
+
+ _end = .;
+
+ /* Throw in the debugging sections */
+ STABS_DEBUG
+ DWARF_DEBUG
+
+ /* Sections to be discarded -- must be last */
+ DISCARDS
+}
diff --git a/arch/openrisc/lib/Makefile b/arch/openrisc/lib/Makefile
new file mode 100644
index 0000000..966f65d
--- /dev/null
+++ b/arch/openrisc/lib/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for or32 specific library files..
+#
+
+obj-y = string.o delay.o
diff --git a/arch/openrisc/mm/Makefile b/arch/openrisc/mm/Makefile
new file mode 100644
index 0000000..324ba26
--- /dev/null
+++ b/arch/openrisc/mm/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the linux openrisc-specific parts of the memory manager.
+#
+
+obj-y := fault.o tlb.o init.o ioremap.o
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:52 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
MAINTAINERS | 8 ++++++++
1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 29801f7..12e2ef4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4668,6 +4668,14 @@ F: drivers/of
F: include/linux/of*.h
K: of_get_property

+OPENRISC ARCHITECTURE
+M: Jonas Bonn <***@southpole.se>
+W: http://openrisc.net
+L: ***@lists.openrisc.net
+S: Maintained
+T: git git://openrisc.net/~jonas/linux
+F: arch/openrisc
+
OPL4 DRIVER
M: Clemens Ladisch <***@ladisch.de>
L: alsa-***@alsa-project.org (moderated for non-subscribers)
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:51 UTC
Permalink
Adds README file, TODO list, config file for or1ksim simulator, and a
couple of other pieces that didn't seem to fit into any other patch.

Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/README.openrisc | 102 ++++
arch/openrisc/TODO.openrisc | 18 +
arch/openrisc/kernel/asm-offsets.c | 70 +++
arch/openrisc/kernel/or32_ksyms.c | 46 ++
arch/openrisc/or1ksim.cfg | 888 ++++++++++++++++++++++++++++++++++++
5 files changed, 1124 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/README.openrisc
create mode 100644 arch/openrisc/TODO.openrisc
create mode 100644 arch/openrisc/kernel/asm-offsets.c
create mode 100644 arch/openrisc/kernel/or32_ksyms.c
create mode 100644 arch/openrisc/or1ksim.cfg

diff --git a/arch/openrisc/README.openrisc b/arch/openrisc/README.openrisc
new file mode 100644
index 0000000..fbf656f
--- /dev/null
+++ b/arch/openrisc/README.openrisc
@@ -0,0 +1,102 @@
+OpenRISC Linux
+==============
+
+This is a port of Linux to the OpenRISC class of microprocessors; the initial
+target architecture, specifically, is the 32-bit OpenRISC 1000 family (or1k).
+
+For more information about OpenRISC processors and ongoing development:
+
+ website http://openrisc.net
+
+Additional information, though not kept up to date by the site maintainers,
+can be found at http://opencores.org/openrisc
+
+For information about Linux on OpenRISC, please contact South Pole AB.
+
+ email: ***@southpole.se
+
+ website: http://southpole.se
+ http://southpoleconsulting.com
+
+---------------------------------------------------------------------
+
+Build instructions for OpenRISC toolchain and Linux
+===================================================
+
+In order to build and run Linux for OpenRISC, you'll need at least a basic
+toolchain and, perhaps, the architectural simulator. Steps to get these bits
+in place are outlined here.
+
+1) The toolchain can be obtained from openrisc.net. Instructions for building
+a toolchain can be found at:
+
+http://openrisc.net/toolchain-build.html
+
+2) or1ksim (optional)
+
+or1ksim is the architectural simulator which will allow you to actually run
+your OpenRISC Linux kernel if you don't have an OpenRISC processor at hand.
+
+ git clone git://openrisc.net/jonas/or1ksim-svn
+
+ cd or1ksim
+ ./configure --prefix=$OPENRISC_PREFIX
+ make
+ make install
+
+3) Linux kernel
+
+Build the kernel as usual
+
+ make ARCH=openrisc defconfig
+ make ARCH=openrisc
+
+4) Run in architectural simulator
+
+There's a configuration file for the architectural simulator (or1ksim) in the
+arch/openrisc directory, allowing you to run your kernel with the following
+incantation:
+
+ sim -f arch/openrisc/or1ksim.cfg vmlinux
+
+---------------------------------------------------------------------
+
+Terminology
+===========
+
+In the code, the following particles are used on symbols to limit the scope
+to more or less specific processor implementations:
+
+openrisc: the OpenRISC class of processors
+or1k: the OpenRISC 1000 family of processors
+or1200: the OpenRISC 1200 processor
+
+---------------------------------------------------------------------
+
+History
+========
+
+18. 11. 2003 Matjaz Breskvar (***@bsemi.com)
+ initial port of linux to OpenRISC/or32 architecture.
+ all the core stuff is implemented and seams usable.
+
+08. 12. 2003 Matjaz Breskvar (***@bsemi.com)
+ complete change of TLB miss handling.
+ rewrite of exceptions handling.
+ fully functional sash-3.6 in default initrd.
+ a much improved version with changes all around.
+
+10. 04. 2004 Matjaz Breskvar (***@bsemi.com)
+ alot of bugfixes all over.
+ ethernet support, functional http and telnet servers.
+ running many standard linux apps.
+
+26. 06. 2004 Matjaz Breskvar (***@bsemi.com)
+ port to 2.6.x
+
+30. 11. 2004 Matjaz Breskvar (***@bsemi.com)
+ lots of bugfixes and enhancments.
+ added opencores framebuffer driver.
+
+09. 10. 2010 Jonas Bonn (***@southpole.se)
+ major rewrite to bring up to par with upstream Linux 2.6.36
diff --git a/arch/openrisc/TODO.openrisc b/arch/openrisc/TODO.openrisc
new file mode 100644
index 0000000..814ecc1
--- /dev/null
+++ b/arch/openrisc/TODO.openrisc
@@ -0,0 +1,18 @@
+The OpenRISC Linux port is fully functional and has been tracking upstream
+since 2.6.35. There are, however, remaining items to be completed within
+the coming months. Here's a list of known-to-be-less-than-stellar items
+that are due for investigation shortly, i.e. our TODO list:
+
+Implement the rest of the DMA API... dma_map_sg, etc.
+
+Consolidate usage of memblock and bootmem... move everything over to memblock.
+
+Clean up defconfigs:
+- remove initramfs, cross-compile, local version, etc.
+
+Finish the renaming cleanup... there are references to or32 in the code
+which was an older name for the architecture. The name we've settled on is
+or1k and this change is slowly trickling through the stack. For the time
+being, or32 is equivalent to or1k.
+
+Implement optimized version of memcpy and memset
diff --git a/arch/openrisc/kernel/asm-offsets.c b/arch/openrisc/kernel/asm-offsets.c
new file mode 100644
index 0000000..08d6c1c
--- /dev/null
+++ b/arch/openrisc/kernel/asm-offsets.c
@@ -0,0 +1,70 @@
+/*
+ * OpenRISC asm-offsets.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This program is used to generate definitions needed by
+ * assembly language modules.
+ *
+ * We use the technique used in the OSF Mach kernel code:
+ * generate asm statements containing #defines,
+ * compile this file to assembler, and then extract the
+ * #defines from the assembly-language output.
+ */
+
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <asm/io.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/thread_info.h>
+
+#define DEFINE(sym, val) \
+ asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+
+#define BLANK() asm volatile("\n->" : : )
+
+int main(void)
+{
+ /* offsets into the task_struct */
+ DEFINE(TASK_STATE, offsetof(struct task_struct, state));
+ DEFINE(TASK_FLAGS, offsetof(struct task_struct, flags));
+ DEFINE(TASK_PTRACE, offsetof(struct task_struct, ptrace));
+ DEFINE(TASK_THREAD, offsetof(struct task_struct, thread));
+ DEFINE(TASK_MM, offsetof(struct task_struct, mm));
+ DEFINE(TASK_ACTIVE_MM, offsetof(struct task_struct, active_mm));
+
+ /* offsets into thread_info */
+ DEFINE(TI_TASK, offsetof(struct thread_info, task));
+ DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
+ DEFINE(TI_PREEMPT, offsetof(struct thread_info, preempt_count));
+ DEFINE(TI_KSP, offsetof(struct thread_info, ksp));
+
+ DEFINE(PT_SIZE, sizeof(struct pt_regs));
+
+ /* Interrupt register frame */
+ DEFINE(STACK_FRAME_OVERHEAD, STACK_FRAME_OVERHEAD);
+ DEFINE(INT_FRAME_SIZE, STACK_FRAME_OVERHEAD + sizeof(struct pt_regs));
+
+ DEFINE(NUM_USER_SEGMENTS, TASK_SIZE >> 28);
+ return 0;
+}
diff --git a/arch/openrisc/kernel/or32_ksyms.c b/arch/openrisc/kernel/or32_ksyms.c
new file mode 100644
index 0000000..83ccf7c
--- /dev/null
+++ b/arch/openrisc/kernel/or32_ksyms.c
@@ -0,0 +1,46 @@
+/*
+ * OpenRISC or32_ksyms.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/elfcore.h>
+#include <linux/sched.h>
+#include <linux/in6.h>
+#include <linux/interrupt.h>
+#include <linux/vmalloc.h>
+#include <linux/semaphore.h>
+
+#include <asm/processor.h>
+#include <asm/uaccess.h>
+#include <asm/checksum.h>
+#include <asm/io.h>
+#include <asm/hardirq.h>
+#include <asm/delay.h>
+#include <asm/pgalloc.h>
+
+#define DECLARE_EXPORT(name) extern void name(void); EXPORT_SYMBOL(name)
+
+/* compiler generated symbols */
+DECLARE_EXPORT(__udivsi3);
+DECLARE_EXPORT(__divsi3);
+DECLARE_EXPORT(__umodsi3);
+DECLARE_EXPORT(__modsi3);
+DECLARE_EXPORT(__muldi3);
+DECLARE_EXPORT(__ashrdi3);
+DECLARE_EXPORT(__ashldi3);
+DECLARE_EXPORT(__lshrdi3);
+
+EXPORT_SYMBOL(__copy_tofrom_user);
diff --git a/arch/openrisc/or1ksim.cfg b/arch/openrisc/or1ksim.cfg
new file mode 100644
index 0000000..65f5e87
--- /dev/null
+++ b/arch/openrisc/or1ksim.cfg
@@ -0,0 +1,888 @@
+/* sim.cfg -- Simulator configuration script file
+ Copyright (C) 2001-2002, Marko Mlinar, ***@opencores.org
+
+This file is part of OpenRISC 1000 Architectural Simulator.
+It contains the default configuration and help about configuring
+the simulator.
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; if not, write to the Free Software
+Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */
+
+
+/* INTRODUCTION
+
+ The ork1sim has various parameters, that are set in configuration files
+ like this one. The user can switch between configurations at startup by
+ specifying the required configuration file with the -f <filename.cfg> option.
+ If no configuration file is specified or1ksim searches for the default
+ configuration file sim.cfg. First it searches for './sim.cfg'. If this
+ file is not found, it searches for '~/or1k/sim.cfg'. If this file is
+ not found too, it reverts to the built-in default configuration.
+
+ NOTE: Users should not rely on the built-in configuration, since the
+ default configuration may differ between version.
+ Rather create a configuration file that sets all critical values.
+
+ This file may contain (standard C) comments only - no // support.
+
+ Configure files may be be included, using:
+ include "file_name_to_include"
+
+ Like normal configuration files, the included file is divided into
+ sections. Each section is described in detail also.
+
+ Some section have subsections. One example of such a subsection is:
+
+ device <index>
+ instance specific parameters...
+ enddevice
+
+ which creates a device instance.
+*/
+
+
+/* MEMORY SECTION
+
+ This section specifies how the memory is generated and the blocks
+ it consists of.
+
+ type = random/unknown/pattern
+ Specifies the initial memory values.
+ 'random' generates random memory using seed 'random_seed'.
+ 'pattern' fills memory with 'pattern'.
+ 'unknown' does not specify how memory should be generated,
+ leaving the memory in a undefined state. This is the fastest
+ option.
+
+ random_seed = <value>
+ random seed for randomizer, used if type = 'random'.
+
+ pattern = <value>
+ pattern to fill memory, used if type = 'pattern'.
+
+ nmemories = <value>
+ number of memory instances connected
+
+ baseaddr = <hex_value>
+ memory start address
+
+ size = <hex_value>
+ memory size
+
+ name = "<string>"
+ memory block name
+
+ ce = <value>
+ chip enable index of the memory instance
+
+ mc = <value>
+ memory controller this memory is connected to
+
+ delayr = <value>
+ cycles, required for read access, -1 if instance does not support reading
+
+ delayw = <value>
+ cycles, required for write access, -1 if instance does not support writing
+
+ log = "<filename>"
+ filename, where to log memory accesses to, no log, if log command is not specified
+*/
+
+
+section memory
+ /*random_seed = 12345
+ type = random*/
+ pattern = 0x00
+ type = unknown /* Fastest */
+
+ name = "FLASH"
+ ce = 0
+ mc = 0
+ baseaddr = 0xf0000000
+ size = 0x01000000
+ delayr = 1
+ delayw = -1
+end
+
+section memory
+ /*random_seed = 12345
+ type = random*/
+ pattern = 0x00
+ type = unknown /* Fastest */
+
+ name = "RAM"
+ ce = 1
+ mc = 0
+ baseaddr = 0x00000000
+ size = 0x02000000
+ delayr = 1
+ delayw = 1
+end
+
+section memory
+ /*random_seed = 12345
+ type = random*/
+ pattern = 0x00
+ type = unknown /* Fastest */
+
+ name = "SRAM"
+ mc = 0
+ ce = 2
+ baseaddr = 0xa4000000
+ size = 0x00100000
+ delayr = 1
+ delayw = 2
+end
+
+
+/* IMMU SECTION
+
+ This section configures the Instruction Memory Manangement Unit
+
+ enabled = 0/1
+ '0': disabled
+ '1': enabled
+ (NOTE: UPR bit is set)
+
+ nsets = <value>
+ number of ITLB sets; must be power of two
+
+ nways = <value>
+ number of ITLB ways
+
+ pagesize = <value>
+ instruction page size; must be power of two
+
+ entrysize = <value>
+ instruction entry size in bytes
+
+ ustates = <value>
+ number of ITLB usage states (2, 3, 4 etc., max is 4)
+
+ hitdelay = <value>
+ number of cycles immu hit costs
+
+ missdelay = <value>
+ number of cycles immu miss costs
+*/
+
+section immu
+ enabled = 1
+ nsets = 64
+ nways = 1
+ pagesize = 8192
+ hitdelay = 0
+ missdelay = 0
+end
+
+
+/* DMMU SECTION
+
+ This section configures the Data Memory Manangement Unit
+
+ enabled = 0/1
+ '0': disabled
+ '1': enabled
+ (NOTE: UPR bit is set)
+
+ nsets = <value>
+ number of DTLB sets; must be power of two
+
+ nways = <value>
+ number of DTLB ways
+
+ pagesize = <value>
+ data page size; must be power of two
+
+ entrysize = <value>
+ data entry size in bytes
+
+ ustates = <value>
+ number of DTLB usage states (2, 3, 4 etc., max is 4)
+
+ hitdelay = <value>
+ number of cycles dmmu hit costs
+
+ missdelay = <value>
+ number of cycles dmmu miss costs
+*/
+
+section dmmu
+ enabled = 1
+ nsets = 64
+ nways = 1
+ pagesize = 8192
+ hitdelay = 0
+ missdelay = 0
+end
+
+
+/* IC SECTION
+
+ This section configures the Instruction Cache
+
+ enabled = 0/1
+ '0': disabled
+ '1': enabled
+ (NOTE: UPR bit is set)
+
+ nsets = <value>
+ number of IC sets; must be power of two
+
+ nways = <value>
+ number of IC ways
+
+ blocksize = <value>
+ IC block size in bytes; must be power of two
+
+ ustates = <value>
+ number of IC usage states (2, 3, 4 etc., max is 4)
+
+ hitdelay = <value>
+ number of cycles ic hit costs
+
+ missdelay = <value>
+ number of cycles ic miss costs
+*/
+
+section ic
+ enabled = 0
+ nsets = 512
+ nways = 1
+ blocksize = 16
+ hitdelay = 1
+ missdelay = 1
+end
+
+
+/* DC SECTION
+
+ This section configures the Data Cache
+
+ enabled = 0/1
+ '0': disabled
+ '1': enabled
+ (NOTE: UPR bit is set)
+
+ nsets = <value>
+ number of DC sets; must be power of two
+
+ nways = <value>
+ number of DC ways
+
+ blocksize = <value>
+ DC block size in bytes; must be power of two
+
+ ustates = <value>
+ number of DC usage states (2, 3, 4 etc., max is 4)
+
+ load_hitdelay = <value>
+ number of cycles dc load hit costs
+
+ load_missdelay = <value>
+ number of cycles dc load miss costs
+
+ store_hitdelay = <value>
+ number of cycles dc load hit costs
+
+ store_missdelay = <value>
+ number of cycles dc load miss costs
+*/
+
+section dc
+ enabled = 0
+ nsets = 512
+ nways = 1
+ blocksize = 16
+ load_hitdelay = 1
+ load_missdelay = 1
+ store_hitdelay = 1
+ store_missdelay = 1
+end
+
+
+/* SIM SECTION
+
+ This section specifies how or1ksim should behave.
+
+ verbose = 0/1
+ '0': don't print extra messages
+ '1': print extra messages
+
+ debug = 0-9
+ 0 : no debug messages
+ 1-9: debug message level.
+ higher numbers produce more messages
+
+ profile = 0/1
+ '0': don't generate profiling file 'sim.profile'
+ '1': don't generate profiling file 'sim.profile'
+
+ prof_fn = "<filename>"
+ optional filename for the profiling file.
+ valid only if 'profile' is set
+
+ mprofile = 0/1
+ '0': don't generate memory profiling file 'sim.mprofile'
+ '1': generate memory profiling file 'sim.mprofile'
+
+ mprof_fn = "<filename>"
+ optional filename for the memory profiling file.
+ valid only if 'mprofile' is set
+
+ history = 0/1
+ '0': don't track execution flow
+ '1': track execution flow
+ Execution flow can be tracked for the simulator's
+ 'hist' command. Useful for back-trace debugging.
+
+ iprompt = 0/1
+ '0': start in <not interactive prompt> (so what do we start in ???)
+ '1': start in interactive prompt.
+
+ exe_log = 0/1
+ '0': don't generate execution log.
+ '1': generate execution log.
+
+ exe_log = default/hardware/simple/software
+ type of execution log, default is used when not specified
+
+ exe_log_start = <value>
+ index of first instruction to start logging, default = 0
+
+ exe_log_end = <value>
+ index of last instruction to end logging; not limited, if omitted
+
+ exe_log_marker = <value>
+ <value> specifies number of instructions before horizontal marker is
+ printed; if zero, markers are disabled (default)
+
+ exe_log_fn = "<filename>"
+ filename for the exection log file.
+ valid only if 'exe_log' is set
+
+ clkcycle = <value>[ps|ns|us|ms]
+ specifies time measurement for one cycle
+*/
+
+section sim
+ verbose = 1
+ debug = 0
+ profile = 0
+ history = 0
+
+ clkcycle = 10ns
+end
+
+
+/* SECTION VAPI
+
+ This section configures the Verification API, used for Advanced
+ Core Verification.
+
+ enabled = 0/1
+ '0': disbable VAPI server
+ '1': enable/start VAPI server
+
+ server_port = <value>
+ TCP/IP port to start VAPI server on
+
+ log_enabled = 0/1
+ '0': disable VAPI requests logging
+ '1': enable VAPI requests logging
+
+ hide_device_id = 0/1
+ '0': don't log device id (for compatability with old version)
+ '1': log device id
+
+
+ vapi_fn = <filename>
+ filename for the log file.
+ valid only if log_enabled is set
+*/
+
+section VAPI
+ enabled = 0
+ server_port = 9998
+ log_enabled = 0
+ vapi_log_fn = "vapi.log"
+end
+
+
+/* CPU SECTION
+
+ This section specifies various CPU parameters.
+
+ ver = <value>
+ rev = <value>
+ specifies version and revision of the CPU used
+
+ upr = <value>
+ changes the upr register
+
+ sr = <value>
+ sets the initial Supervision Register value
+ supervisor mode (SM) and fixed one (FO) set = 0x8001
+ exception prefix high (EPH, ***@0xf0000000) = 0x4000
+ together, (SM | FO | EPH) = 0xc001
+ superscalar = 0/1
+ '0': CPU is scalar
+ '1': CPU is superscalar
+ (modify cpu/or32/execute.c to tune superscalar model)
+
+ hazards = 0/1
+ '0': don't track data hazards in superscalar CPU
+ '1': track data hazards in superscalar CPU
+ If tracked, data hazards can be displayed using the
+ simulator's 'r' command.
+
+ dependstats = 0/1
+ '0': don't calculate inter-instruction dependencies.
+ '1': calculate inter-instruction dependencies.
+ If calculated, inter-instruction dependencies can be
+ displayed using the simulator's 'stat' command.
+
+ sbuf_len = <value>
+ length of store buffer (<= 256), 0 = disabled
+*/
+
+section cpu
+ ver = 0x12
+ cfg = 0x00
+ rev = 0x01
+ sr = 0x8001 /*SPR_SR_FO | SPR_SR_SM | SPR_SR_EPH */
+ /* upr = */
+ superscalar = 0
+ hazards = 0
+ dependstats = 0
+ sbuf_len = 0
+end
+
+
+/* PM SECTION
+
+ This section specifies Power Management parameters
+
+ enabled = 0/1
+ '0': disable power management
+ '1': enable power management
+*/
+
+section pm
+ enabled = 0
+end
+
+
+/* BPB SECTION
+
+ This section specifies how branch prediction should behave.
+
+ enabled = 0/1
+ '0': disable branch prediction
+ '1': enable branch prediction
+
+ btic = 0/1
+ '0': disable branch target instruction cache model
+ '1': enable branch target instruction cache model
+
+ sbp_bf_fwd = 0/1
+ Static branch prediction for 'l.bf'
+ '0': don't use forward prediction
+ '1': use forward prediction
+
+ sbp_bnf_fwd = 0/1
+ Static branch prediction for 'l.bnf'
+ '0': don't use forward prediction
+ '1': use forward prediction
+
+ hitdelay = <value>
+ number of cycles bpb hit costs
+
+ missdelay = <value>
+ number of cycles bpb miss costs
+*/
+
+section bpb
+ enabled = 0
+ btic = 0
+ sbp_bf_fwd = 0
+ sbp_bnf_fwd = 0
+ hitdelay = 0
+ missdelay = 0
+end
+
+
+/* DEBUG SECTION
+
+ This sections specifies how the debug unit should behave.
+
+ enabled = 0/1
+ '0': disable debug unit
+ '1': enable debug unit
+
+ gdb_enabled = 0/1
+ '0': don't start gdb server
+ '1': start gdb server at port 'server_port'
+
+ server_port = <value>
+ TCP/IP port to start gdb server on
+ valid only if gdb_enabled is set
+
+ vapi_id = <hex_value>
+ Used to create "fake" vapi log file containing the JTAG proxy messages.
+*/
+section debug
+ enabled = 0
+/* gdb_enabled = 0 */
+/* server_port = 9999*/
+ rsp_enabled = 1
+ rsp_port = 50001
+end
+
+
+/* MC SECTION
+
+ This section configures the memory controller
+
+ enabled = 0/1
+ '0': disable memory controller
+ '1': enable memory controller
+
+ baseaddr = <hex_value>
+ address of first MC register
+
+ POC = <hex_value>
+ Power On Configuration register
+
+ index = <value>
+ Index of this memory controller amongst all the memory controllers
+*/
+
+section mc
+ enabled = 0
+ baseaddr = 0x93000000
+ POC = 0x00000008 /* Power on configuration register */
+ index = 0
+end
+
+
+/* UART SECTION
+
+ This section configures the UARTs
+
+ enabled = <0|1>
+ Enable/disable the peripheral. By default if it is enabled.
+
+ baseaddr = <hex_value>
+ address of first UART register for this device
+
+
+ channel = <channeltype>:<args>
+
+ The channel parameter indicates the source of received UART characters
+ and the sink for transmitted UART characters.
+
+ The <channeltype> can be either "file", "xterm", "tcp", "fd", or "tty"
+ (without quotes).
+
+ A) To send/receive characters from a pair of files, use a file
+ channel:
+
+ channel=file:<rxfile>,<txfile>
+
+ B) To create an interactive terminal window, use an xterm channel:
+
+ channel=xterm:[<xterm_arg>]*
+
+ C) To create a bidirectional tcp socket which one could, for example,
+ access via telnet, use a tcp channel:
+
+ channel=tcp:<port number>
+
+ D) To cause the UART to read/write from existing numeric file
+ descriptors, use an fd channel:
+
+ channel=fd:<rx file descriptor num>,<tx file descriptor num>
+
+ E) To connect the UART to a physical serial port, create a tty
+ channel:
+
+ channel=tty:device=/dev/ttyS0,baud=9600
+
+ irq = <value>
+ irq number for this device
+
+ 16550 = 0/1
+ '0': this device is a UART16450
+ '1': this device is a UART16550
+
+ jitter = <value>
+ in msecs... time to block, -1 to disable it
+
+ vapi_id = <hex_value>
+ VAPI id of this instance
+*/
+
+section uart
+ enabled = 1
+ baseaddr = 0x90000000
+ irq = 2
+ /* channel = "file:uart0.rx,uart0.tx" */
+ channel = "tcp:10084"
+/* channel = "xterm:" */
+ jitter = -1 /* async behaviour */
+ 16550 = 1
+end
+
+
+/* DMA SECTION
+
+ This section configures the DMAs
+
+ enabled = <0|1>
+ Enable/disable the peripheral. By default if it is enabled.
+
+ baseaddr = <hex_value>
+ address of first DMA register for this device
+
+ irq = <value>
+ irq number for this device
+
+ vapi_id = <hex_value>
+ VAPI id of this instance
+*/
+
+section dma
+ enabled = 1
+ baseaddr = 0x9a000000
+ irq = 11
+end
+
+
+/* ETHERNET SECTION
+
+ This section configures the ETHERNETs
+
+ enabled = <0|1>
+ Enable/disable the peripheral. By default if it is enabled.
+
+ baseaddr = <hex_value>
+ address of first ethernet register for this device
+
+ dma = <value>
+ which controller is this ethernet "connected" to
+
+ irq = <value>
+ ethernet mac IRQ level
+
+ rtx_type = <value>
+ use 0 - file interface, 1 - socket interface
+
+ rx_channel = <value>
+ DMA channel used for RX
+
+ tx_channel = <value>
+ DMA channel used for TX
+
+ rxfile = "<filename>"
+ filename, where to read data from
+
+ txfile = "<filename>"
+ filename, where to write data to
+
+ sockif = "<ifacename>"
+ interface name of ethernet socket
+
+ vapi_id = <hex_value>
+ VAPI id of this instance
+*/
+
+section ethernet
+ enabled = 1
+ baseaddr = 0x92000000
+ /* dma = 0 */
+ irq = 4
+ rtx_type = "tap"
+ tap_dev = "tap0"
+ /* tx_channel = 0 */
+ /* rx_channel = 1 */
+ rxfile = "eth0.rx"
+ txfile = "eth0.tx"
+ sockif = "eth0"
+end
+
+
+/* GPIO SECTION
+
+ This section configures the GPIOs
+
+ enabled = <0|1>
+ Enable/disable the peripheral. By default if it is enabled.
+
+ baseaddr = <hex_value>
+ address of first GPIO register for this device
+
+ irq = <value>
+ irq number for this device
+
+ base_vapi_id = <hex_value>
+ first VAPI id of this instance
+ GPIO uses 8 consecutive VAPI IDs
+*/
+
+section gpio
+ enabled = 0
+ baseaddr = 0x91000000
+ irq = 3
+ base_vapi_id = 0x0200
+end
+
+/* VGA SECTION
+
+ This section configures the VGA/LCD controller
+
+ enabled = <0|1>
+ Enable/disable the peripheral. By default if it is enabled.
+
+ baseaddr = <hex_value>
+ address of first VGA register
+
+ irq = <value>
+ irq number for this device
+
+ refresh_rate = <value>
+ number of cycles between screen dumps
+
+ filename = "<filename>"
+ template name for generated names (e.g. "primary" produces "primary0023.bmp")
+*/
+
+section vga
+ enabled = 0
+ baseaddr = 0x97100000
+ irq = 8
+ refresh_rate = 100000
+ filename = "primary"
+end
+
+
+/* TICK TIMER SECTION
+
+ This section configures tick timer
+
+ enabled = 0/1
+ whether tick timer is enabled
+*/
+
+section pic
+ enabled = 1
+ edge_trigger = 1
+end
+
+/* FB SECTION
+
+ This section configures the frame buffer
+
+ enabled = <0|1>
+ Enable/disable the peripheral. By default if it is enabled.
+
+ baseaddr = <hex_value>
+ base address of frame buffer
+
+ paladdr = <hex_value>
+ base address of first palette entry
+
+ refresh_rate = <value>
+ number of cycles between screen dumps
+
+ filename = "<filename>"
+ template name for generated names (e.g. "primary" produces "primary0023.bmp")
+*/
+
+section fb
+ enabled = 0
+ baseaddr = 0x97000000
+ refresh_rate = 1000000
+ filename = "primary"
+end
+
+
+/* KBD SECTION
+
+ This section configures the PS/2 compatible keyboard
+
+ baseaddr = <hex_value>
+ base address of the keyboard device
+
+ rxfile = "<filename>"
+ filename, where to read data from
+*/
+
+section kbd
+ enabled = 0
+ irq = 5
+ baseaddr = 0x94000000
+ rxfile = "kbd.rx"
+end
+
+
+/* ATA SECTION
+
+ This section configures the ATA/ATAPI host controller
+
+ baseaddr = <hex_value>
+ address of first ATA register
+
+ enabled = <0|1>
+ Enable/disable the peripheral. By default if it is enabled.
+
+ irq = <value>
+ irq number for this device
+
+ debug = <value>
+ debug level for ata models.
+ 0: no debug messages
+ 1: verbose messages
+ 3: normal messages (more messages than verbose)
+ 5: debug messages (normal debug messages)
+ 7: flow control messages (debug statemachine flows)
+ 9: low priority message (display everything the code does)
+
+ dev_type0/1 = <value>
+ ata device 0 type
+ 0: NO_CONNeCT: none (not connected)
+ 1: FILE : simulated harddisk
+ 2: LOCAL : local system harddisk
+
+ dev_file0/1 = "<filename>"
+ filename for simulated ATA device
+ valid only if dev_type0 == 1
+
+ dev_size0/1 = <value>
+ size of simulated hard-disk (in MBytes)
+ valid only if dev_type0 == 1
+
+ dev_packet0/1 = <value>
+ 0: simulated ATA device does NOT implement PACKET command feature set
+ 1: simulated ATA device does implement PACKET command feature set
+
+ FIXME: irq number
+*/
+
+section ata
+ enabled = 0
+ baseaddr = 0x9e000000
+ irq = 15
+
+end
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:50 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/lib/delay.c | 60 +++++++++++++
arch/openrisc/lib/string.S | 204 ++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 264 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/lib/delay.c
create mode 100644 arch/openrisc/lib/string.S

diff --git a/arch/openrisc/lib/delay.c b/arch/openrisc/lib/delay.c
new file mode 100644
index 0000000..01d9740
--- /dev/null
+++ b/arch/openrisc/lib/delay.c
@@ -0,0 +1,60 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation
+ *
+ * Precise Delay Loops
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <asm/delay.h>
+#include <asm/timex.h>
+#include <asm/processor.h>
+
+int __devinit read_current_timer(unsigned long *timer_value)
+{
+ *timer_value = mfspr(SPR_TTCR);
+ return 0;
+}
+
+void __delay(unsigned long cycles)
+{
+ cycles_t target = get_cycles() + cycles;
+
+ while (get_cycles() < target)
+ cpu_relax();
+}
+EXPORT_SYMBOL(__delay);
+
+inline void __const_udelay(unsigned long xloops)
+{
+ unsigned long long loops;
+
+ loops = xloops * loops_per_jiffy * HZ;
+
+ __delay(loops >> 32);
+}
+EXPORT_SYMBOL(__const_udelay);
+
+void __udelay(unsigned long usecs)
+{
+ __const_udelay(usecs * 0x10C7UL); /* 2**32 / 1000000 (rounded up) */
+}
+EXPORT_SYMBOL(__udelay);
+
+void __ndelay(unsigned long nsecs)
+{
+ __const_udelay(nsecs * 0x5UL); /* 2**32 / 1000000000 (rounded up) */
+}
+EXPORT_SYMBOL(__ndelay);
diff --git a/arch/openrisc/lib/string.S b/arch/openrisc/lib/string.S
new file mode 100644
index 0000000..465f04b
--- /dev/null
+++ b/arch/openrisc/lib/string.S
@@ -0,0 +1,204 @@
+/*
+ * OpenRISC string.S
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/linkage.h>
+#include <asm/errno.h>
+
+ /*
+ * this can be optimized by doing gcc inline assemlby with
+ * proper constraints (no need to save args registers...)
+ *
+ */
+
+
+/*
+ *
+ * int __copy_tofrom_user(void *to, const void *from, unsigned long size);
+ *
+ * NOTE: it returns number of bytes NOT copied !!!
+ *
+ */
+ .global __copy_tofrom_user
+__copy_tofrom_user:
+ l.addi r1,r1,-12
+ l.sw 0(r1),r6
+ l.sw 4(r1),r4
+ l.sw 8(r1),r3
+
+ l.addi r11,r5,0
+2: l.sfeq r11,r0
+ l.bf 1f
+ l.addi r11,r11,-1
+8: l.lbz r6,0(r4)
+9: l.sb 0(r3),r6
+ l.addi r3,r3,1
+ l.j 2b
+ l.addi r4,r4,1
+1:
+ l.addi r11,r11,1 // r11 holds the return value
+
+ l.lwz r6,0(r1)
+ l.lwz r4,4(r1)
+ l.lwz r3,8(r1)
+ l.jr r9
+ l.addi r1,r1,12
+
+ .section .fixup, "ax"
+99:
+ l.j 1b
+ l.nop
+ .previous
+
+ .section __ex_table, "a"
+ .long 8b, 99b // read fault
+ .long 9b, 99b // write fault
+ .previous
+
+/*
+ * unsigned long clear_user(void *addr, unsigned long size) ;
+ *
+ * NOTE: it returns number of bytes NOT cleared !!!
+ */
+ .global __clear_user
+__clear_user:
+ l.addi r1,r1,-8
+ l.sw 0(r1),r4
+ l.sw 4(r1),r3
+
+2: l.sfeq r4,r0
+ l.bf 1f
+ l.addi r4,r4,-1
+9: l.sb 0(r3),r0
+ l.j 2b
+ l.addi r3,r3,1
+
+1:
+ l.addi r11,r4,1
+
+ l.lwz r4,0(r1)
+ l.lwz r3,4(r1)
+ l.jr r9
+ l.addi r1,r1,8
+
+ .section .fixup, "ax"
+99:
+ l.j 1b
+ l.nop
+ .previous
+
+ .section __ex_table, "a"
+ .long 9b, 99b // write fault
+ .previous
+
+/*
+ * long strncpy_from_user(char *dst, const char *src, long count)
+ *
+ *
+ */
+ .global __strncpy_from_user
+__strncpy_from_user:
+ l.addi r1,r1,-16
+ l.sw 0(r1),r6
+ l.sw 4(r1),r5
+ l.sw 8(r1),r4
+ l.sw 12(r1),r3
+
+ l.addi r11,r5,0
+2: l.sfeq r5,r0
+ l.bf 1f
+ l.addi r5,r5,-1
+8: l.lbz r6,0(r4)
+ l.sfeq r6,r0
+ l.bf 1f
+9: l.sb 0(r3),r6
+ l.addi r3,r3,1
+ l.j 2b
+ l.addi r4,r4,1
+1:
+ l.lwz r6,0(r1)
+ l.addi r5,r5,1
+ l.sub r11,r11,r5 // r11 holds the return value
+
+ l.lwz r6,0(r1)
+ l.lwz r5,4(r1)
+ l.lwz r4,8(r1)
+ l.lwz r3,12(r1)
+ l.jr r9
+ l.addi r1,r1,16
+
+ .section .fixup, "ax"
+99:
+ l.movhi r11,hi(-EFAULT)
+ l.ori r11,r11,lo(-EFAULT)
+
+ l.lwz r6,0(r1)
+ l.lwz r5,4(r1)
+ l.lwz r4,8(r1)
+ l.lwz r3,12(r1)
+ l.jr r9
+ l.addi r1,r1,16
+ .previous
+
+ .section __ex_table, "a"
+ .long 8b, 99b // read fault
+ .previous
+
+/*
+ * extern int __strnlen_user(const char *str, long len, unsigned long top);
+ *
+ *
+ * RTRN: - length of a string including NUL termination character
+ * - on page fault 0
+ */
+
+ .global __strnlen_user
+__strnlen_user:
+ l.addi r1,r1,-8
+ l.sw 0(r1),r6
+ l.sw 4(r1),r3
+
+ l.addi r11,r0,0
+2: l.sfeq r11,r4
+ l.bf 1f
+ l.addi r11,r11,1
+8: l.lbz r6,0(r3)
+ l.sfeq r6,r0
+ l.bf 1f
+ l.sfgeu r3,r5 // are we over the top ?
+ l.bf 99f
+ l.j 2b
+ l.addi r3,r3,1
+
+1:
+ l.lwz r6,0(r1)
+ l.lwz r3,4(r1)
+ l.jr r9
+ l.addi r1,r1,8
+
+ .section .fixup, "ax"
+99:
+ l.addi r11,r0,0
+
+ l.lwz r6,0(r1)
+ l.lwz r3,4(r1)
+ l.jr r9
+ l.addi r1,r1,8
+ .previous
+
+ .section __ex_table, "a"
+ .long 8b, 99b // read fault
+ .previous
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:48 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/kernel/traps.c | 367 ++++++++++++++++++++++++++++++++++++++++++
1 files changed, 367 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/kernel/traps.c

diff --git a/arch/openrisc/kernel/traps.c b/arch/openrisc/kernel/traps.c
new file mode 100644
index 0000000..e75ab98
--- /dev/null
+++ b/arch/openrisc/kernel/traps.c
@@ -0,0 +1,367 @@
+/*
+ * OpenRISC traps.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Here we handle the break vectors not used by the system call
+ * mechanism, as well as some general stack/register dumping
+ * things.
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/kmod.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/timer.h>
+#include <linux/mm.h>
+#include <linux/kallsyms.h>
+#include <asm/uaccess.h>
+
+#include <asm/system.h>
+#include <asm/segment.h>
+#include <asm/io.h>
+#include <asm/pgtable.h>
+
+extern char _etext, _stext;
+
+int kstack_depth_to_print = 0x180;
+
+static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
+{
+ return p > (void *)tinfo && p < (void *)tinfo + THREAD_SIZE - 3;
+}
+
+void show_trace(struct task_struct *task, unsigned long *stack)
+{
+ struct thread_info *context;
+ unsigned long addr;
+
+ context = (struct thread_info *)
+ ((unsigned long)stack & (~(THREAD_SIZE - 1)));
+
+ while (valid_stack_ptr(context, stack)) {
+ addr = *stack++;
+ if (__kernel_text_address(addr)) {
+ printk(" [<%08lx>]", addr);
+ print_symbol(" %s", addr);
+ printk("\n");
+ }
+ }
+ printk(" =======================\n");
+}
+
+/* displays a short stack trace */
+void show_stack(struct task_struct *task, unsigned long *esp)
+{
+ unsigned long addr, *stack;
+ int i;
+
+ if (esp == NULL)
+ esp = (unsigned long *)&esp;
+
+ stack = esp;
+
+ printk("Stack dump [0x%08lx]:\n", (unsigned long)esp);
+ for (i = 0; i < kstack_depth_to_print; i++) {
+ if (kstack_end(stack))
+ break;
+ if (__get_user(addr, stack)) {
+ /* This message matches "failing address" marked
+ s390 in ksymoops, so lines containing it will
+ not be filtered out by ksymoops. */
+ printk("Failing address 0x%lx\n", (unsigned long)stack);
+ break;
+ }
+ stack++;
+
+ printk("sp + %02d: 0x%08lx\n", i * 4, addr);
+ }
+ printk("\n");
+
+ show_trace(task, esp);
+
+ return;
+}
+
+void show_trace_task(struct task_struct *tsk)
+{
+ /*
+ * TODO: SysRq-T trace dump...
+ */
+}
+
+/*
+ * The architecture-independent backtrace generator
+ */
+void dump_stack(void)
+{
+ unsigned long stack;
+
+ show_stack(current, &stack);
+}
+
+void show_registers(struct pt_regs *regs)
+{
+ int i;
+ int in_kernel = 1;
+ unsigned long esp;
+
+ esp = (unsigned long)(&regs->sp);
+ if (user_mode(regs))
+ in_kernel = 0;
+
+ printk("CPU #: %d\n"
+ " PC: %08lx SR: %08lx SP: %08lx\n",
+ smp_processor_id(), regs->pc, regs->sr, regs->sp);
+ printk("GPR00: %08lx GPR01: %08lx GPR02: %08lx GPR03: %08lx\n",
+ 0L, regs->gpr[1], regs->gpr[2], regs->gpr[3]);
+ printk("GPR04: %08lx GPR05: %08lx GPR06: %08lx GPR07: %08lx\n",
+ regs->gpr[4], regs->gpr[5], regs->gpr[6], regs->gpr[7]);
+ printk("GPR08: %08lx GPR09: %08lx GPR10: %08lx GPR11: %08lx\n",
+ regs->gpr[8], regs->gpr[9], regs->gpr[10], regs->gpr[11]);
+ printk("GPR12: %08lx GPR13: %08lx GPR14: %08lx GPR15: %08lx\n",
+ regs->gpr[12], regs->gpr[13], regs->gpr[14], regs->gpr[15]);
+ printk("GPR16: %08lx GPR17: %08lx GPR18: %08lx GPR19: %08lx\n",
+ regs->gpr[16], regs->gpr[17], regs->gpr[18], regs->gpr[19]);
+ printk("GPR20: %08lx GPR21: %08lx GPR22: %08lx GPR23: %08lx\n",
+ regs->gpr[20], regs->gpr[21], regs->gpr[22], regs->gpr[23]);
+ printk("GPR24: %08lx GPR25: %08lx GPR26: %08lx GPR27: %08lx\n",
+ regs->gpr[24], regs->gpr[25], regs->gpr[26], regs->gpr[27]);
+ printk("GPR28: %08lx GPR29: %08lx GPR30: %08lx GPR31: %08lx\n",
+ regs->gpr[28], regs->gpr[29], regs->gpr[30], regs->gpr[31]);
+ printk(" RES: %08lx oGPR11: %08lx syscallno: %08lx\n",
+ regs->gpr[11], regs->orig_gpr11, regs->syscallno);
+
+ printk("Process %s (pid: %d, stackpage=%08lx)\n",
+ current->comm, current->pid, (unsigned long)current);
+ /*
+ * When in-kernel, we also print out the stack and code at the
+ * time of the fault..
+ */
+ if (in_kernel) {
+
+ printk("\nStack: ");
+ show_stack(NULL, (unsigned long *)esp);
+
+ printk("\nCode: ");
+ if (regs->pc < PAGE_OFFSET)
+ goto bad;
+
+ for (i = -24; i < 24; i++) {
+ unsigned char c;
+ if (__get_user(c, &((unsigned char *)regs->pc)[i])) {
+bad:
+ printk(" Bad PC value.");
+ break;
+ }
+
+ if (i == 0)
+ printk("(%02x) ", c);
+ else
+ printk("%02x ", c);
+ }
+ }
+ printk("\n");
+}
+
+void nommu_dump_state(struct pt_regs *regs,
+ unsigned long ea, unsigned long vector)
+{
+ int i;
+ unsigned long addr, stack = regs->sp;
+
+ printk("\n\r[nommu_dump_state] :: ea %lx, vector %lx\n\r", ea, vector);
+
+ printk("CPU #: %d\n"
+ " PC: %08lx SR: %08lx SP: %08lx\n",
+ 0, regs->pc, regs->sr, regs->sp);
+ printk("GPR00: %08lx GPR01: %08lx GPR02: %08lx GPR03: %08lx\n",
+ 0L, regs->gpr[1], regs->gpr[2], regs->gpr[3]);
+ printk("GPR04: %08lx GPR05: %08lx GPR06: %08lx GPR07: %08lx\n",
+ regs->gpr[4], regs->gpr[5], regs->gpr[6], regs->gpr[7]);
+ printk("GPR08: %08lx GPR09: %08lx GPR10: %08lx GPR11: %08lx\n",
+ regs->gpr[8], regs->gpr[9], regs->gpr[10], regs->gpr[11]);
+ printk("GPR12: %08lx GPR13: %08lx GPR14: %08lx GPR15: %08lx\n",
+ regs->gpr[12], regs->gpr[13], regs->gpr[14], regs->gpr[15]);
+ printk("GPR16: %08lx GPR17: %08lx GPR18: %08lx GPR19: %08lx\n",
+ regs->gpr[16], regs->gpr[17], regs->gpr[18], regs->gpr[19]);
+ printk("GPR20: %08lx GPR21: %08lx GPR22: %08lx GPR23: %08lx\n",
+ regs->gpr[20], regs->gpr[21], regs->gpr[22], regs->gpr[23]);
+ printk("GPR24: %08lx GPR25: %08lx GPR26: %08lx GPR27: %08lx\n",
+ regs->gpr[24], regs->gpr[25], regs->gpr[26], regs->gpr[27]);
+ printk("GPR28: %08lx GPR29: %08lx GPR30: %08lx GPR31: %08lx\n",
+ regs->gpr[28], regs->gpr[29], regs->gpr[30], regs->gpr[31]);
+ printk(" RES: %08lx oGPR11: %08lx syscallno: %08lx\n",
+ regs->gpr[11], regs->orig_gpr11, regs->syscallno);
+
+ printk("Process %s (pid: %d, stackpage=%08lx)\n",
+ ((struct task_struct *)(__pa(current)))->comm,
+ ((struct task_struct *)(__pa(current)))->pid,
+ (unsigned long)current);
+
+ printk("\nStack: ");
+ printk("Stack dump [0x%08lx]:\n", (unsigned long)stack);
+ for (i = 0; i < kstack_depth_to_print; i++) {
+ if (((long)stack & (THREAD_SIZE - 1)) == 0)
+ break;
+ stack++;
+
+ printk("%lx :: sp + %02d: 0x%08lx\n", stack, i * 4,
+ *((unsigned long *)(__pa(stack))));
+ }
+ printk("\n");
+
+ printk("Call Trace: ");
+ i = 1;
+ while (((long)stack & (THREAD_SIZE - 1)) != 0) {
+ addr = *((unsigned long *)__pa(stack));
+ stack++;
+
+ if (kernel_text_address(addr)) {
+ if (i && ((i % 6) == 0))
+ printk("\n ");
+ printk(" [<%08lx>]", addr);
+ i++;
+ }
+ }
+ printk("\n");
+
+ printk("\nCode: ");
+
+ for (i = -24; i < 24; i++) {
+ unsigned char c;
+ c = ((unsigned char *)(__pa(regs->pc)))[i];
+
+ if (i == 0)
+ printk("(%02x) ", c);
+ else
+ printk("%02x ", c);
+ }
+ printk("\n");
+}
+
+/* This is normally the 'Oops' routine */
+void die(const char *str, struct pt_regs *regs, long err)
+{
+
+ console_verbose();
+ printk("\n%s#: %04lx\n", str, err & 0xffff);
+ show_registers(regs);
+#ifdef CONFIG_JUMP_UPON_UNHANDLED_EXCEPTION
+ printk("\n\nUNHANDLED_EXCEPTION: entering infinite loop\n");
+
+ /* shut down interrupts */
+ local_irq_disable();
+
+ __asm__ __volatile__("l.nop 1");
+ do {} while(1);
+#endif
+ do_exit(SIGSEGV);
+}
+
+/* This is normally the 'Oops' routine */
+void die_if_kernel(const char *str, struct pt_regs *regs, long err)
+{
+ if (user_mode(regs))
+ return;
+
+ die(str, regs, err);
+}
+
+void unhandled_exception(struct pt_regs *regs, int ea, int vector)
+{
+ printk("Unable to handle exception at EA =0x%x, vector 0x%x",
+ ea, vector);
+ die("Oops", regs, 9);
+}
+
+void __init trap_init(void)
+{
+ /* Nothing needs to be done */
+}
+
+asmlinkage void do_trap(struct pt_regs *regs, unsigned long address)
+{
+ siginfo_t info;
+ memset(&info, 0, sizeof(info));
+ info.si_signo = SIGTRAP;
+ info.si_code = TRAP_TRACE;
+ info.si_addr = (void *)address;
+ force_sig_info(SIGTRAP, &info, current);
+
+ if (!(test_tsk_thread_flag(current, TIF_SINGLESTEP)))
+ regs->pc += 4;
+}
+
+asmlinkage void do_unaligned_access(struct pt_regs *regs, unsigned long address)
+{
+ siginfo_t info;
+
+ if (user_mode(regs)) {
+ /* Send a SIGSEGV */
+ info.si_signo = SIGSEGV;
+ info.si_errno = 0;
+ /* info.si_code has been set above */
+ info.si_addr = (void *)address;
+ force_sig_info(SIGSEGV, &info, current);
+ } else {
+ printk("KERNEL: Unaligned Access 0x%.8lx\n", address);
+ show_registers(regs);
+ die("Die:", regs, address);
+ }
+
+}
+
+asmlinkage void do_bus_fault(struct pt_regs *regs, unsigned long address)
+{
+ siginfo_t info;
+
+ if (user_mode(regs)) {
+ /* Send a SIGBUS */
+ info.si_signo = SIGBUS;
+ info.si_errno = 0;
+ info.si_code = BUS_ADRERR;
+ info.si_addr = (void *)address;
+ force_sig_info(SIGBUS, &info, current);
+ } else { /* Kernel mode */
+ printk("KERNEL: Bus error (SIGBUS) 0x%.8lx\n", address);
+ show_registers(regs);
+ die("Die:", regs, address);
+ }
+}
+
+asmlinkage void do_illegal_instruction(struct pt_regs *regs,
+ unsigned long address)
+{
+ siginfo_t info;
+
+ if (user_mode(regs)) {
+ /* Send a SIGILL */
+ info.si_signo = SIGILL;
+ info.si_errno = 0;
+ info.si_code = ILL_ILLOPC;
+ info.si_addr = (void *)address;
+ force_sig_info(SIGBUS, &info, current);
+ } else { /* Kernel mode */
+ printk("KERNEL: Illegal instruction (SIGILL) 0x%.8lx\n",
+ address);
+ show_registers(regs);
+ die("Die:", regs, address);
+ }
+}
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:49 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/asm-offsets.h | 1 +
arch/openrisc/include/asm/bitops.h | 59 +++
arch/openrisc/include/asm/bitops/__ffs.h | 33 ++
arch/openrisc/include/asm/bitops/__fls.h | 33 ++
arch/openrisc/include/asm/bitops/ffs.h | 32 ++
arch/openrisc/include/asm/bitops/fls.h | 33 ++
arch/openrisc/include/asm/byteorder.h | 1 +
arch/openrisc/include/asm/cpuinfo.h | 34 ++
arch/openrisc/include/asm/delay.h | 24 ++
arch/openrisc/include/asm/elf.h | 109 ++++++
arch/openrisc/include/asm/io.h | 51 +++
arch/openrisc/include/asm/linkage.h | 25 ++
arch/openrisc/include/asm/mutex.h | 27 ++
arch/openrisc/include/asm/param.h | 26 ++
arch/openrisc/include/asm/processor.h | 112 ++++++
arch/openrisc/include/asm/serial.h | 36 ++
arch/openrisc/include/asm/spinlock.h | 24 ++
arch/openrisc/include/asm/spr.h | 42 ++
arch/openrisc/include/asm/spr_defs.h | 604 ++++++++++++++++++++++++++++++
arch/openrisc/include/asm/system.h | 35 ++
arch/openrisc/include/asm/unaligned.h | 51 +++
arch/openrisc/kernel/vmlinux.h | 12 +
22 files changed, 1404 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/asm-offsets.h
create mode 100644 arch/openrisc/include/asm/bitops.h
create mode 100644 arch/openrisc/include/asm/bitops/__ffs.h
create mode 100644 arch/openrisc/include/asm/bitops/__fls.h
create mode 100644 arch/openrisc/include/asm/bitops/ffs.h
create mode 100644 arch/openrisc/include/asm/bitops/fls.h
create mode 100644 arch/openrisc/include/asm/byteorder.h
create mode 100644 arch/openrisc/include/asm/cpuinfo.h
create mode 100644 arch/openrisc/include/asm/delay.h
create mode 100644 arch/openrisc/include/asm/elf.h
create mode 100644 arch/openrisc/include/asm/io.h
create mode 100644 arch/openrisc/include/asm/linkage.h
create mode 100644 arch/openrisc/include/asm/mutex.h
create mode 100644 arch/openrisc/include/asm/param.h
create mode 100644 arch/openrisc/include/asm/processor.h
create mode 100644 arch/openrisc/include/asm/serial.h
create mode 100644 arch/openrisc/include/asm/spinlock.h
create mode 100644 arch/openrisc/include/asm/spr.h
create mode 100644 arch/openrisc/include/asm/spr_defs.h
create mode 100644 arch/openrisc/include/asm/system.h
create mode 100644 arch/openrisc/include/asm/unaligned.h
create mode 100644 arch/openrisc/kernel/vmlinux.h

diff --git a/arch/openrisc/include/asm/asm-offsets.h b/arch/openrisc/include/asm/asm-offsets.h
new file mode 100644
index 0000000..d370ee3
--- /dev/null
+++ b/arch/openrisc/include/asm/asm-offsets.h
@@ -0,0 +1 @@
+#include <generated/asm-offsets.h>
diff --git a/arch/openrisc/include/asm/bitops.h b/arch/openrisc/include/asm/bitops.h
new file mode 100644
index 0000000..a9e11ef
--- /dev/null
+++ b/arch/openrisc/include/asm/bitops.h
@@ -0,0 +1,59 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_BITOPS_H
+#define __ASM_OPENRISC_BITOPS_H
+
+/*
+ * Where we haven't written assembly versions yet, we fall back to the
+ * generic implementations. Otherwise, we pull in our (hopefully)
+ * optimized versions.
+ */
+
+#include <linux/irqflags.h>
+#include <linux/compiler.h>
+
+/*
+ * clear_bit may not imply a memory barrier
+ */
+#ifndef smp_mb__before_clear_bit
+#define smp_mb__before_clear_bit() smp_mb()
+#define smp_mb__after_clear_bit() smp_mb()
+#endif
+
+#include <asm/bitops/__ffs.h>
+#include <asm-generic/bitops/ffz.h>
+#include <asm/bitops/fls.h>
+#include <asm/bitops/__fls.h>
+#include <asm-generic/bitops/fls64.h>
+#include <asm-generic/bitops/find.h>
+
+#ifndef _LINUX_BITOPS_H
+#error only <linux/bitops.h> can be included directly
+#endif
+
+#include <asm-generic/bitops/sched.h>
+#include <asm/bitops/ffs.h>
+#include <asm-generic/bitops/hweight.h>
+#include <asm-generic/bitops/lock.h>
+
+#include <asm-generic/bitops/atomic.h>
+#include <asm-generic/bitops/non-atomic.h>
+#include <asm-generic/bitops/ext2-atomic.h>
+
+#endif /* __ASM_GENERIC_BITOPS_H */
diff --git a/arch/openrisc/include/asm/bitops/__ffs.h b/arch/openrisc/include/asm/bitops/__ffs.h
new file mode 100644
index 0000000..6c8368a
--- /dev/null
+++ b/arch/openrisc/include/asm/bitops/__ffs.h
@@ -0,0 +1,33 @@
+/*
+ * OpenRISC Linux
+ *
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC___FFS_H
+#define __ASM_OPENRISC___FFS_H
+
+
+#ifdef CONFIG_OPENRISC_HAVE_INST_FF1
+
+static inline unsigned long __ffs(unsigned long x)
+{
+ int ret;
+
+ __asm__ ("l.ff1 %0,%1"
+ : "=r" (ret)
+ : "r" (x));
+
+ return ret-1;
+}
+
+#else
+#include <asm-generic/bitops/__ffs.h>
+#endif
+
+#endif /* __ASM_OPENRISC___FFS_H */
diff --git a/arch/openrisc/include/asm/bitops/__fls.h b/arch/openrisc/include/asm/bitops/__fls.h
new file mode 100644
index 0000000..c4ecdb4
--- /dev/null
+++ b/arch/openrisc/include/asm/bitops/__fls.h
@@ -0,0 +1,33 @@
+/*
+ * OpenRISC Linux
+ *
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC___FLS_H
+#define __ASM_OPENRISC___FLS_H
+
+
+#ifdef CONFIG_OPENRISC_HAVE_INST_FL1
+
+static inline unsigned long __fls(unsigned long x)
+{
+ int ret;
+
+ __asm__ ("l.fl1 %0,%1"
+ : "=r" (ret)
+ : "r" (x));
+
+ return ret-1;
+}
+
+#else
+#include <asm-generic/bitops/__fls.h>
+#endif
+
+#endif /* __ASM_OPENRISC___FLS_H */
diff --git a/arch/openrisc/include/asm/bitops/ffs.h b/arch/openrisc/include/asm/bitops/ffs.h
new file mode 100644
index 0000000..9de4624
--- /dev/null
+++ b/arch/openrisc/include/asm/bitops/ffs.h
@@ -0,0 +1,32 @@
+/*
+ * OpenRISC Linux
+ *
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_FFS_H
+#define __ASM_OPENRISC_FFS_H
+
+#ifdef CONFIG_OPENRISC_HAVE_INST_FF1
+
+static inline int ffs(int x)
+{
+ int ret;
+
+ __asm__ ("l.ff1 %0,%1"
+ : "=r" (ret)
+ : "r" (x));
+
+ return ret;
+}
+
+#else
+#include <asm-generic/bitops/ffs.h>
+#endif
+
+#endif /* __ASM_OPENRISC_FFS_H */
diff --git a/arch/openrisc/include/asm/bitops/fls.h b/arch/openrisc/include/asm/bitops/fls.h
new file mode 100644
index 0000000..9efbf9a
--- /dev/null
+++ b/arch/openrisc/include/asm/bitops/fls.h
@@ -0,0 +1,33 @@
+/*
+ * OpenRISC Linux
+ *
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_FLS_H
+#define __ASM_OPENRISC_FLS_H
+
+
+#ifdef CONFIG_OPENRISC_HAVE_INST_FL1
+
+static inline int fls(int x)
+{
+ int ret;
+
+ __asm__ ("l.fl1 %0,%1"
+ : "=r" (ret)
+ : "r" (x));
+
+ return ret;
+}
+
+#else
+#include <asm-generic/bitops/fls.h>
+#endif
+
+#endif /* __ASM_OPENRISC_FLS_H */
diff --git a/arch/openrisc/include/asm/byteorder.h b/arch/openrisc/include/asm/byteorder.h
new file mode 100644
index 0000000..60d14f7
--- /dev/null
+++ b/arch/openrisc/include/asm/byteorder.h
@@ -0,0 +1 @@
+#include <linux/byteorder/big_endian.h>
diff --git a/arch/openrisc/include/asm/cpuinfo.h b/arch/openrisc/include/asm/cpuinfo.h
new file mode 100644
index 0000000..917318b
--- /dev/null
+++ b/arch/openrisc/include/asm/cpuinfo.h
@@ -0,0 +1,34 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_CPUINFO_H
+#define __ASM_OPENRISC_CPUINFO_H
+
+struct cpuinfo {
+ u32 clock_frequency;
+
+ u32 icache_size;
+ u32 icache_block_size;
+
+ u32 dcache_size;
+ u32 dcache_block_size;
+};
+
+extern struct cpuinfo cpuinfo;
+
+#endif /* __ASM_OPENRISC_CPUINFO_H */
diff --git a/arch/openrisc/include/asm/delay.h b/arch/openrisc/include/asm/delay.h
new file mode 100644
index 0000000..17f8bf5
--- /dev/null
+++ b/arch/openrisc/include/asm/delay.h
@@ -0,0 +1,24 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_DELAY_H
+#define __ASM_OPENRISC_DELAY_H
+
+#include <asm-generic/delay.h>
+
+extern unsigned long loops_per_jiffy;
+
+#endif
diff --git a/arch/openrisc/include/asm/elf.h b/arch/openrisc/include/asm/elf.h
new file mode 100644
index 0000000..a52afb4
--- /dev/null
+++ b/arch/openrisc/include/asm/elf.h
@@ -0,0 +1,109 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_ELF_H
+#define __ASM_OPENRISC_ELF_H
+
+/*
+ * ELF register definitions..
+ */
+#include <asm/types.h>
+#include <asm/ptrace.h>
+
+
+/* The OR1K relocation types... not all relevant for module loader */
+#define R_OR32_NONE 0
+#define R_OR32_32 1
+#define R_OR32_16 2
+#define R_OR32_8 3
+#define R_OR32_CONST 4
+#define R_OR32_CONSTH 5
+#define R_OR32_JUMPTARG 6
+#define R_OR32_VTINHERIT 7
+#define R_OR32_VTENTRY 8
+
+typedef unsigned long elf_greg_t;
+
+/* Note that NGREG is defined to ELF_NGREG in include/linux/elfcore.h, and is
+ thus exposed to user-space. */
+#define ELF_NGREG (sizeof (struct pt_regs) / sizeof(elf_greg_t))
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+
+/* A placeholder; OR32 does not have fp support yes, so no fp regs for now. */
+typedef unsigned long elf_fpregset_t;
+
+/* This should be moved to include/linux/elf.h */
+#define EM_OR32 0x8472
+#define EM_OPENRISC 92 /* OpenRISC 32-bit embedded processor */
+
+/*
+ * These are used to set parameters in the core dumps.
+ */
+#define ELF_ARCH EM_OR32
+#define ELF_CLASS ELFCLASS32
+#define ELF_DATA ELFDATA2MSB
+
+#ifdef __KERNEL__
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+
+#define elf_check_arch(x) \
+ (((x)->e_machine == EM_OR32) || ((x)->e_machine == EM_OPENRISC))
+
+/* This is the location that an ET_DYN program is loaded if exec'ed. Typical
+ use of this is to invoke "./ld.so someprog" to test out a new version of
+ the loader. We need to make sure that it is out of the way of the program
+ that it will "exec", and that there is sufficient room for the brk. */
+
+#define ELF_ET_DYN_BASE (0x08000000)
+
+/*
+ * Enable dump using regset.
+ * This covers all of general/DSP/FPU regs.
+ */
+#define CORE_DUMP_USE_REGSET
+
+#define ELF_EXEC_PAGESIZE 8192
+
+#define ELF_CORE_COPY_REGS(gregs, regs) \
+ memcpy(gregs, regs, \
+ sizeof(struct pt_regs) < sizeof(elf_gregset_t) ? \
+ sizeof(struct pt_regs) : sizeof(elf_gregset_t));
+
+
+/* This yields a mask that user programs can use to figure out what
+ instruction set this cpu supports. This could be done in userspace,
+ but it's not easy, and we've already done it here. */
+
+#define ELF_HWCAP (0)
+
+/* This yields a string that ld.so will use to load implementation
+ specific libraries for optimization. This is more specific in
+ intent than poking at uname or /proc/cpuinfo.
+
+ For the moment, we have only optimizations for the Intel generations,
+ but that could change... */
+
+#define ELF_PLATFORM (NULL)
+
+#define SET_PERSONALITY(ex) set_personality(PER_LINUX)
+
+#endif /* __KERNEL__ */
+#endif
diff --git a/arch/openrisc/include/asm/io.h b/arch/openrisc/include/asm/io.h
new file mode 100644
index 0000000..c877679
--- /dev/null
+++ b/arch/openrisc/include/asm/io.h
@@ -0,0 +1,51 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_IO_H
+#define __ASM_OPENRISC_IO_H
+
+/*
+ * PCI: can we really do 0 here if we have no port IO?
+ */
+#define IO_SPACE_LIMIT 0
+
+/* OpenRISC has no port IO */
+#define HAVE_ARCH_PIO_SIZE 1
+#define PIO_RESERVED 0X0UL
+#define PIO_OFFSET 0
+#define PIO_MASK 0
+
+#include <asm-generic/io.h>
+
+extern void __iomem * __ioremap(phys_addr_t offset, unsigned long size,
+ unsigned long flags);
+
+static inline void __iomem * ioremap(phys_addr_t offset, unsigned long size)
+{
+ return __ioremap(offset, size, 0);
+}
+
+/* #define _PAGE_CI 0x002 */
+static inline void __iomem * ioremap_nocache(phys_addr_t offset,
+ unsigned long size)
+{
+ return __ioremap(offset, size, 0x002);
+}
+
+extern void iounmap(void *addr);
+
+#endif
diff --git a/arch/openrisc/include/asm/linkage.h b/arch/openrisc/include/asm/linkage.h
new file mode 100644
index 0000000..e263875
--- /dev/null
+++ b/arch/openrisc/include/asm/linkage.h
@@ -0,0 +1,25 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_LINKAGE_H
+#define __ASM_OPENRISC_LINKAGE_H
+
+#define __ALIGN .align 0
+#define __ALIGN_STR ".align 0"
+
+#endif /* __ASM_OPENRISC_LINKAGE_H */
diff --git a/arch/openrisc/include/asm/mutex.h b/arch/openrisc/include/asm/mutex.h
new file mode 100644
index 0000000..b85a0cf
--- /dev/null
+++ b/arch/openrisc/include/asm/mutex.h
@@ -0,0 +1,27 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+/*
+ * Pull in the generic implementation for the mutex fastpath.
+ *
+ * TODO: implement optimized primitives instead, or leave the generic
+ * implementation in place, or pick the atomic_xchg() based generic
+ * implementation. (see asm-generic/mutex-xchg.h for details)
+ */
+
+#include <asm-generic/mutex-dec.h>
diff --git a/arch/openrisc/include/asm/param.h b/arch/openrisc/include/asm/param.h
new file mode 100644
index 0000000..c39a336
--- /dev/null
+++ b/arch/openrisc/include/asm/param.h
@@ -0,0 +1,26 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_PARAM_H
+#define __ASM_OPENRISC_PARAM_H
+
+#define EXEC_PAGESIZE 8192
+
+#include <asm-generic/param.h>
+
+#endif /* __ASM_OPENRISC_PARAM_H */
diff --git a/arch/openrisc/include/asm/processor.h b/arch/openrisc/include/asm/processor.h
new file mode 100644
index 0000000..d9a85c9
--- /dev/null
+++ b/arch/openrisc/include/asm/processor.h
@@ -0,0 +1,112 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_PROCESSOR_H
+#define __ASM_OPENRISC_PROCESSOR_H
+
+#include <asm/spr_defs.h>
+#include <asm/page.h>
+#include <asm/ptrace.h>
+#define STACK_TOP TASK_SIZE
+#define STACK_TOP_MAX STACK_TOP
+/* Kernel and user SR register setting */
+#define KERNEL_SR (SPR_SR_DME | SPR_SR_IME | SPR_SR_ICE \
+ | SPR_SR_DCE | SPR_SR_SM)
+#define USER_SR (SPR_SR_DME | SPR_SR_IME | SPR_SR_ICE \
+ | SPR_SR_DCE | SPR_SR_IEE | SPR_SR_TEE)
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+#define current_text_addr() ({ __label__ _l; _l: &&_l; })
+
+/*
+ * User space process size. This is hardcoded into a few places,
+ * so don't change it unless you know what you are doing.
+ */
+
+#define TASK_SIZE (0x80000000UL)
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define TASK_UNMAPPED_BASE (TASK_SIZE / 8 * 3)
+
+#ifndef __ASSEMBLY__
+
+struct task_struct;
+
+struct thread_struct {
+};
+
+/*
+ * At user->kernel entry, the pt_regs struct is stacked on the top of the
+ * kernel-stack. This macro allows us to find those regs for a task.
+ * Notice that subsequent pt_regs stackings, like recursive interrupts
+ * occurring while we're in the kernel, won't affect this - only the first
+ * user->kernel transition registers are reached by this (i.e. not regs
+ * for running signal handler)
+ */
+#define user_regs(thread_info) (((struct pt_regs *)((unsigned long)(thread_info) + THREAD_SIZE - STACK_FRAME_OVERHEAD)) - 1)
+
+/*
+ * Dito but for the currently running task
+ */
+
+#define task_pt_regs(task) user_regs(task_thread_info(task))
+#define current_regs() user_regs(current_thread_info())
+
+extern inline void prepare_to_copy(struct task_struct *tsk)
+{
+}
+
+#define INIT_SP (sizeof(init_stack) + (unsigned long) &init_stack)
+
+#define INIT_THREAD { }
+
+
+#define KSTK_EIP(tsk) (task_pt_regs(tsk)->pc);
+#define KSTK_ESP(tsk) (task_pt_regs(tsk)->sp);
+
+
+extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+
+void start_thread(struct pt_regs *regs, unsigned long nip, unsigned long sp);
+void release_thread(struct task_struct *);
+unsigned long get_wchan(struct task_struct *p);
+
+/*
+ * Free current thread data structures etc..
+ */
+
+extern inline void exit_thread(void)
+{
+ /* Nothing needs to be done. */
+}
+
+/*
+ * Return saved PC of a blocked thread. For now, this is the "user" PC
+ */
+extern unsigned long thread_saved_pc(struct task_struct *t);
+
+#define init_stack (init_thread_union.stack)
+
+#define cpu_relax() do { } while (0)
+
+#endif /* __ASSEMBLY__ */
+#endif /* __ASM_OPENRISC_PROCESSOR_H */
diff --git a/arch/openrisc/include/asm/serial.h b/arch/openrisc/include/asm/serial.h
new file mode 100644
index 0000000..270a452
--- /dev/null
+++ b/arch/openrisc/include/asm/serial.h
@@ -0,0 +1,36 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_SERIAL_H
+#define __ASM_OPENRISC_SERIAL_H
+
+#ifdef __KERNEL__
+
+#include <asm/cpuinfo.h>
+
+/* There's a generic version of this file, but it assumes a 1.8MHz UART clk...
+ * this, on the other hand, assumes the UART clock is tied to the system
+ * clock... 8250_early.c (early 8250 serial console) actually uses this, so
+ * it needs to be correct to get the early console working.
+ */
+
+#define BASE_BAUD (cpuinfo.clock_frequency/16)
+
+#endif /* __KERNEL__ */
+
+#endif /* __ASM_OPENRISC_SERIAL_H */
diff --git a/arch/openrisc/include/asm/spinlock.h b/arch/openrisc/include/asm/spinlock.h
new file mode 100644
index 0000000..fd00a3a
--- /dev/null
+++ b/arch/openrisc/include/asm/spinlock.h
@@ -0,0 +1,24 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_SPINLOCK_H
+#define __ASM_OPENRISC_SPINLOCK_H
+
+#error "or32 doesn't do SMP yet"
+
+#endif
diff --git a/arch/openrisc/include/asm/spr.h b/arch/openrisc/include/asm/spr.h
new file mode 100644
index 0000000..1cccb42
--- /dev/null
+++ b/arch/openrisc/include/asm/spr.h
@@ -0,0 +1,42 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_SPR_H
+#define __ASM_OPENRISC_SPR_H
+
+#define mtspr(_spr, _val) __asm__ __volatile__ ( \
+ "l.mtspr r0,%1,%0" \
+ : : "K" (_spr), "r" (_val))
+#define mtspr_off(_spr, _off, _val) __asm__ __volatile__ ( \
+ "l.mtspr %0,%1,%2" \
+ : : "r" (_off), "r" (_val), "K" (_spr))
+
+static inline unsigned long mfspr(unsigned long add)
+{
+ unsigned long ret;
+ __asm__ __volatile__ ("l.mfspr %0,r0,%1" : "=r" (ret) : "K" (add));
+ return ret;
+}
+
+static inline unsigned long mfspr_off(unsigned long add, unsigned long offset)
+{
+ unsigned long ret;
+ __asm__ __volatile__ ("l.mfspr %0,%1,%2" : "=r" (ret)
+ : "r" (offset), "K" (add));
+ return ret;
+}
+
+#endif
diff --git a/arch/openrisc/include/asm/spr_defs.h b/arch/openrisc/include/asm/spr_defs.h
new file mode 100644
index 0000000..5dbc668
--- /dev/null
+++ b/arch/openrisc/include/asm/spr_defs.h
@@ -0,0 +1,604 @@
+/*
+ * OpenRISC Linux
+ *
+ * SPR Definitions
+ *
+ * Copyright (C) 2000 Damjan Lampret
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2008, 2010 Embecosm Limited
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This file is part of OpenRISC 1000 Architectural Simulator.
+ */
+
+#ifndef SPR_DEFS__H
+#define SPR_DEFS__H
+
+/* Definition of special-purpose registers (SPRs). */
+
+#define MAX_GRPS (32)
+#define MAX_SPRS_PER_GRP_BITS (11)
+#define MAX_SPRS_PER_GRP (1 << MAX_SPRS_PER_GRP_BITS)
+#define MAX_SPRS (0x10000)
+
+/* Base addresses for the groups */
+#define SPRGROUP_SYS (0 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_DMMU (1 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_IMMU (2 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_DC (3 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_IC (4 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_MAC (5 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_D (6 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_PC (7 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_PM (8 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_PIC (9 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_TT (10 << MAX_SPRS_PER_GRP_BITS)
+#define SPRGROUP_FP (11 << MAX_SPRS_PER_GRP_BITS)
+
+/* System control and status group */
+#define SPR_VR (SPRGROUP_SYS + 0)
+#define SPR_UPR (SPRGROUP_SYS + 1)
+#define SPR_CPUCFGR (SPRGROUP_SYS + 2)
+#define SPR_DMMUCFGR (SPRGROUP_SYS + 3)
+#define SPR_IMMUCFGR (SPRGROUP_SYS + 4)
+#define SPR_DCCFGR (SPRGROUP_SYS + 5)
+#define SPR_ICCFGR (SPRGROUP_SYS + 6)
+#define SPR_DCFGR (SPRGROUP_SYS + 7)
+#define SPR_PCCFGR (SPRGROUP_SYS + 8)
+#define SPR_NPC (SPRGROUP_SYS + 16) /* CZ 21/06/01 */
+#define SPR_SR (SPRGROUP_SYS + 17) /* CZ 21/06/01 */
+#define SPR_PPC (SPRGROUP_SYS + 18) /* CZ 21/06/01 */
+#define SPR_FPCSR (SPRGROUP_SYS + 20) /* CZ 21/06/01 */
+#define SPR_EPCR_BASE (SPRGROUP_SYS + 32) /* CZ 21/06/01 */
+#define SPR_EPCR_LAST (SPRGROUP_SYS + 47) /* CZ 21/06/01 */
+#define SPR_EEAR_BASE (SPRGROUP_SYS + 48)
+#define SPR_EEAR_LAST (SPRGROUP_SYS + 63)
+#define SPR_ESR_BASE (SPRGROUP_SYS + 64)
+#define SPR_ESR_LAST (SPRGROUP_SYS + 79)
+#define SPR_GPR_BASE (SPRGROUP_SYS + 1024)
+
+/* Data MMU group */
+#define SPR_DMMUCR (SPRGROUP_DMMU + 0)
+#define SPR_DTLBEIR (SPRGROUP_DMMU + 2)
+#define SPR_DTLBMR_BASE(WAY) (SPRGROUP_DMMU + 0x200 + (WAY) * 0x100)
+#define SPR_DTLBMR_LAST(WAY) (SPRGROUP_DMMU + 0x27f + (WAY) * 0x100)
+#define SPR_DTLBTR_BASE(WAY) (SPRGROUP_DMMU + 0x280 + (WAY) * 0x100)
+#define SPR_DTLBTR_LAST(WAY) (SPRGROUP_DMMU + 0x2ff + (WAY) * 0x100)
+
+/* Instruction MMU group */
+#define SPR_IMMUCR (SPRGROUP_IMMU + 0)
+#define SPR_ITLBEIR (SPRGROUP_IMMU + 2)
+#define SPR_ITLBMR_BASE(WAY) (SPRGROUP_IMMU + 0x200 + (WAY) * 0x100)
+#define SPR_ITLBMR_LAST(WAY) (SPRGROUP_IMMU + 0x27f + (WAY) * 0x100)
+#define SPR_ITLBTR_BASE(WAY) (SPRGROUP_IMMU + 0x280 + (WAY) * 0x100)
+#define SPR_ITLBTR_LAST(WAY) (SPRGROUP_IMMU + 0x2ff + (WAY) * 0x100)
+
+/* Data cache group */
+#define SPR_DCCR (SPRGROUP_DC + 0)
+#define SPR_DCBPR (SPRGROUP_DC + 1)
+#define SPR_DCBFR (SPRGROUP_DC + 2)
+#define SPR_DCBIR (SPRGROUP_DC + 3)
+#define SPR_DCBWR (SPRGROUP_DC + 4)
+#define SPR_DCBLR (SPRGROUP_DC + 5)
+#define SPR_DCR_BASE(WAY) (SPRGROUP_DC + 0x200 + (WAY) * 0x200)
+#define SPR_DCR_LAST(WAY) (SPRGROUP_DC + 0x3ff + (WAY) * 0x200)
+
+/* Instruction cache group */
+#define SPR_ICCR (SPRGROUP_IC + 0)
+#define SPR_ICBPR (SPRGROUP_IC + 1)
+#define SPR_ICBIR (SPRGROUP_IC + 2)
+#define SPR_ICBLR (SPRGROUP_IC + 3)
+#define SPR_ICR_BASE(WAY) (SPRGROUP_IC + 0x200 + (WAY) * 0x200)
+#define SPR_ICR_LAST(WAY) (SPRGROUP_IC + 0x3ff + (WAY) * 0x200)
+
+/* MAC group */
+#define SPR_MACLO (SPRGROUP_MAC + 1)
+#define SPR_MACHI (SPRGROUP_MAC + 2)
+
+/* Debug group */
+#define SPR_DVR(N) (SPRGROUP_D + (N))
+#define SPR_DCR(N) (SPRGROUP_D + 8 + (N))
+#define SPR_DMR1 (SPRGROUP_D + 16)
+#define SPR_DMR2 (SPRGROUP_D + 17)
+#define SPR_DWCR0 (SPRGROUP_D + 18)
+#define SPR_DWCR1 (SPRGROUP_D + 19)
+#define SPR_DSR (SPRGROUP_D + 20)
+#define SPR_DRR (SPRGROUP_D + 21)
+
+/* Performance counters group */
+#define SPR_PCCR(N) (SPRGROUP_PC + (N))
+#define SPR_PCMR(N) (SPRGROUP_PC + 8 + (N))
+
+/* Power management group */
+#define SPR_PMR (SPRGROUP_PM + 0)
+
+/* PIC group */
+#define SPR_PICMR (SPRGROUP_PIC + 0)
+#define SPR_PICPR (SPRGROUP_PIC + 1)
+#define SPR_PICSR (SPRGROUP_PIC + 2)
+
+/* Tick Timer group */
+#define SPR_TTMR (SPRGROUP_TT + 0)
+#define SPR_TTCR (SPRGROUP_TT + 1)
+
+/*
+ * Bit definitions for the Version Register
+ *
+ */
+#define SPR_VR_VER 0xff000000 /* Processor version */
+#define SPR_VR_CFG 0x00ff0000 /* Processor configuration */
+#define SPR_VR_RES 0x0000ffc0 /* Reserved */
+#define SPR_VR_REV 0x0000003f /* Processor revision */
+
+#define SPR_VR_VER_OFF 24
+#define SPR_VR_CFG_OFF 16
+#define SPR_VR_REV_OFF 0
+
+/*
+ * Bit definitions for the Unit Present Register
+ *
+ */
+#define SPR_UPR_UP 0x00000001 /* UPR present */
+#define SPR_UPR_DCP 0x00000002 /* Data cache present */
+#define SPR_UPR_ICP 0x00000004 /* Instruction cache present */
+#define SPR_UPR_DMP 0x00000008 /* Data MMU present */
+#define SPR_UPR_IMP 0x00000010 /* Instruction MMU present */
+#define SPR_UPR_MP 0x00000020 /* MAC present */
+#define SPR_UPR_DUP 0x00000040 /* Debug unit present */
+#define SPR_UPR_PCUP 0x00000080 /* Performance counters unit present */
+#define SPR_UPR_PMP 0x00000100 /* Power management present */
+#define SPR_UPR_PICP 0x00000200 /* PIC present */
+#define SPR_UPR_TTP 0x00000400 /* Tick timer present */
+#define SPR_UPR_RES 0x00fe0000 /* Reserved */
+#define SPR_UPR_CUP 0xff000000 /* Context units present */
+
+/*
+ * JPB: Bit definitions for the CPU configuration register
+ *
+ */
+#define SPR_CPUCFGR_NSGF 0x0000000f /* Number of shadow GPR files */
+#define SPR_CPUCFGR_CGF 0x00000010 /* Custom GPR file */
+#define SPR_CPUCFGR_OB32S 0x00000020 /* ORBIS32 supported */
+#define SPR_CPUCFGR_OB64S 0x00000040 /* ORBIS64 supported */
+#define SPR_CPUCFGR_OF32S 0x00000080 /* ORFPX32 supported */
+#define SPR_CPUCFGR_OF64S 0x00000100 /* ORFPX64 supported */
+#define SPR_CPUCFGR_OV64S 0x00000200 /* ORVDX64 supported */
+#define SPR_CPUCFGR_RES 0xfffffc00 /* Reserved */
+
+/*
+ * JPB: Bit definitions for the Debug configuration register and other
+ * constants.
+ *
+ */
+
+#define SPR_DCFGR_NDP 0x00000007 /* Number of matchpoints mask */
+#define SPR_DCFGR_NDP1 0x00000000 /* One matchpoint supported */
+#define SPR_DCFGR_NDP2 0x00000001 /* Two matchpoints supported */
+#define SPR_DCFGR_NDP3 0x00000002 /* Three matchpoints supported */
+#define SPR_DCFGR_NDP4 0x00000003 /* Four matchpoints supported */
+#define SPR_DCFGR_NDP5 0x00000004 /* Five matchpoints supported */
+#define SPR_DCFGR_NDP6 0x00000005 /* Six matchpoints supported */
+#define SPR_DCFGR_NDP7 0x00000006 /* Seven matchpoints supported */
+#define SPR_DCFGR_NDP8 0x00000007 /* Eight matchpoints supported */
+#define SPR_DCFGR_WPCI 0x00000008 /* Watchpoint counters implemented */
+
+#define MATCHPOINTS_TO_NDP(n) (1 == n ? SPR_DCFGR_NDP1 : \
+ 2 == n ? SPR_DCFGR_NDP2 : \
+ 3 == n ? SPR_DCFGR_NDP3 : \
+ 4 == n ? SPR_DCFGR_NDP4 : \
+ 5 == n ? SPR_DCFGR_NDP5 : \
+ 6 == n ? SPR_DCFGR_NDP6 : \
+ 7 == n ? SPR_DCFGR_NDP7 : SPR_DCFGR_NDP8)
+#define MAX_MATCHPOINTS 8
+#define MAX_WATCHPOINTS (MAX_MATCHPOINTS + 2)
+
+/*
+ * Bit definitions for the Supervision Register
+ *
+ */
+#define SPR_SR_SM 0x00000001 /* Supervisor Mode */
+#define SPR_SR_TEE 0x00000002 /* Tick timer Exception Enable */
+#define SPR_SR_IEE 0x00000004 /* Interrupt Exception Enable */
+#define SPR_SR_DCE 0x00000008 /* Data Cache Enable */
+#define SPR_SR_ICE 0x00000010 /* Instruction Cache Enable */
+#define SPR_SR_DME 0x00000020 /* Data MMU Enable */
+#define SPR_SR_IME 0x00000040 /* Instruction MMU Enable */
+#define SPR_SR_LEE 0x00000080 /* Little Endian Enable */
+#define SPR_SR_CE 0x00000100 /* CID Enable */
+#define SPR_SR_F 0x00000200 /* Condition Flag */
+#define SPR_SR_CY 0x00000400 /* Carry flag */
+#define SPR_SR_OV 0x00000800 /* Overflow flag */
+#define SPR_SR_OVE 0x00001000 /* Overflow flag Exception */
+#define SPR_SR_DSX 0x00002000 /* Delay Slot Exception */
+#define SPR_SR_EPH 0x00004000 /* Exception Prefix High */
+#define SPR_SR_FO 0x00008000 /* Fixed one */
+#define SPR_SR_SUMRA 0x00010000 /* Supervisor SPR read access */
+#define SPR_SR_RES 0x0ffe0000 /* Reserved */
+#define SPR_SR_CID 0xf0000000 /* Context ID */
+
+/*
+ * Bit definitions for the Data MMU Control Register
+ *
+ */
+#define SPR_DMMUCR_P2S 0x0000003e /* Level 2 Page Size */
+#define SPR_DMMUCR_P1S 0x000007c0 /* Level 1 Page Size */
+#define SPR_DMMUCR_VADDR_WIDTH 0x0000f800 /* Virtual ADDR Width */
+#define SPR_DMMUCR_PADDR_WIDTH 0x000f0000 /* Physical ADDR Width */
+
+/*
+ * Bit definitions for the Instruction MMU Control Register
+ *
+ */
+#define SPR_IMMUCR_P2S 0x0000003e /* Level 2 Page Size */
+#define SPR_IMMUCR_P1S 0x000007c0 /* Level 1 Page Size */
+#define SPR_IMMUCR_VADDR_WIDTH 0x0000f800 /* Virtual ADDR Width */
+#define SPR_IMMUCR_PADDR_WIDTH 0x000f0000 /* Physical ADDR Width */
+
+/*
+ * Bit definitions for the Data TLB Match Register
+ *
+ */
+#define SPR_DTLBMR_V 0x00000001 /* Valid */
+#define SPR_DTLBMR_PL1 0x00000002 /* Page Level 1 (if 0 then PL2) */
+#define SPR_DTLBMR_CID 0x0000003c /* Context ID */
+#define SPR_DTLBMR_LRU 0x000000c0 /* Least Recently Used */
+#define SPR_DTLBMR_VPN 0xfffff000 /* Virtual Page Number */
+
+/*
+ * Bit definitions for the Data TLB Translate Register
+ *
+ */
+#define SPR_DTLBTR_CC 0x00000001 /* Cache Coherency */
+#define SPR_DTLBTR_CI 0x00000002 /* Cache Inhibit */
+#define SPR_DTLBTR_WBC 0x00000004 /* Write-Back Cache */
+#define SPR_DTLBTR_WOM 0x00000008 /* Weakly-Ordered Memory */
+#define SPR_DTLBTR_A 0x00000010 /* Accessed */
+#define SPR_DTLBTR_D 0x00000020 /* Dirty */
+#define SPR_DTLBTR_URE 0x00000040 /* User Read Enable */
+#define SPR_DTLBTR_UWE 0x00000080 /* User Write Enable */
+#define SPR_DTLBTR_SRE 0x00000100 /* Supervisor Read Enable */
+#define SPR_DTLBTR_SWE 0x00000200 /* Supervisor Write Enable */
+#define SPR_DTLBTR_PPN 0xfffff000 /* Physical Page Number */
+
+/*
+ * Bit definitions for the Instruction TLB Match Register
+ *
+ */
+#define SPR_ITLBMR_V 0x00000001 /* Valid */
+#define SPR_ITLBMR_PL1 0x00000002 /* Page Level 1 (if 0 then PL2) */
+#define SPR_ITLBMR_CID 0x0000003c /* Context ID */
+#define SPR_ITLBMR_LRU 0x000000c0 /* Least Recently Used */
+#define SPR_ITLBMR_VPN 0xfffff000 /* Virtual Page Number */
+
+/*
+ * Bit definitions for the Instruction TLB Translate Register
+ *
+ */
+#define SPR_ITLBTR_CC 0x00000001 /* Cache Coherency */
+#define SPR_ITLBTR_CI 0x00000002 /* Cache Inhibit */
+#define SPR_ITLBTR_WBC 0x00000004 /* Write-Back Cache */
+#define SPR_ITLBTR_WOM 0x00000008 /* Weakly-Ordered Memory */
+#define SPR_ITLBTR_A 0x00000010 /* Accessed */
+#define SPR_ITLBTR_D 0x00000020 /* Dirty */
+#define SPR_ITLBTR_SXE 0x00000040 /* User Read Enable */
+#define SPR_ITLBTR_UXE 0x00000080 /* User Write Enable */
+#define SPR_ITLBTR_PPN 0xfffff000 /* Physical Page Number */
+
+/*
+ * Bit definitions for Data Cache Control register
+ *
+ */
+#define SPR_DCCR_EW 0x000000ff /* Enable ways */
+
+/*
+ * Bit definitions for Insn Cache Control register
+ *
+ */
+#define SPR_ICCR_EW 0x000000ff /* Enable ways */
+
+/*
+ * Bit definitions for Data Cache Configuration Register
+ *
+ */
+
+#define SPR_DCCFGR_NCW 0x00000007
+#define SPR_DCCFGR_NCS 0x00000078
+#define SPR_DCCFGR_CBS 0x00000080
+#define SPR_DCCFGR_CWS 0x00000100
+#define SPR_DCCFGR_CCRI 0x00000200
+#define SPR_DCCFGR_CBIRI 0x00000400
+#define SPR_DCCFGR_CBPRI 0x00000800
+#define SPR_DCCFGR_CBLRI 0x00001000
+#define SPR_DCCFGR_CBFRI 0x00002000
+#define SPR_DCCFGR_CBWBRI 0x00004000
+
+#define SPR_DCCFGR_NCW_OFF 0
+#define SPR_DCCFGR_NCS_OFF 3
+#define SPR_DCCFGR_CBS_OFF 7
+
+/*
+ * Bit definitions for Instruction Cache Configuration Register
+ *
+ */
+#define SPR_ICCFGR_NCW 0x00000007
+#define SPR_ICCFGR_NCS 0x00000078
+#define SPR_ICCFGR_CBS 0x00000080
+#define SPR_ICCFGR_CCRI 0x00000200
+#define SPR_ICCFGR_CBIRI 0x00000400
+#define SPR_ICCFGR_CBPRI 0x00000800
+#define SPR_ICCFGR_CBLRI 0x00001000
+
+#define SPR_ICCFGR_NCW_OFF 0
+#define SPR_ICCFGR_NCS_OFF 3
+#define SPR_ICCFGR_CBS_OFF 7
+
+/*
+ * Bit definitions for Data MMU Configuration Register
+ *
+ */
+
+#define SPR_DMMUCFGR_NTW 0x00000003
+#define SPR_DMMUCFGR_NTS 0x0000001C
+#define SPR_DMMUCFGR_NAE 0x000000E0
+#define SPR_DMMUCFGR_CRI 0x00000100
+#define SPR_DMMUCFGR_PRI 0x00000200
+#define SPR_DMMUCFGR_TEIRI 0x00000400
+#define SPR_DMMUCFGR_HTR 0x00000800
+
+#define SPR_DMMUCFGR_NTW_OFF 0
+#define SPR_DMMUCFGR_NTS_OFF 2
+
+/*
+ * Bit definitions for Instruction MMU Configuration Register
+ *
+ */
+
+#define SPR_IMMUCFGR_NTW 0x00000003
+#define SPR_IMMUCFGR_NTS 0x0000001C
+#define SPR_IMMUCFGR_NAE 0x000000E0
+#define SPR_IMMUCFGR_CRI 0x00000100
+#define SPR_IMMUCFGR_PRI 0x00000200
+#define SPR_IMMUCFGR_TEIRI 0x00000400
+#define SPR_IMMUCFGR_HTR 0x00000800
+
+#define SPR_IMMUCFGR_NTW_OFF 0
+#define SPR_IMMUCFGR_NTS_OFF 2
+
+/*
+ * Bit definitions for Debug Control registers
+ *
+ */
+#define SPR_DCR_DP 0x00000001 /* DVR/DCR present */
+#define SPR_DCR_CC 0x0000000e /* Compare condition */
+#define SPR_DCR_SC 0x00000010 /* Signed compare */
+#define SPR_DCR_CT 0x000000e0 /* Compare to */
+
+/* Bit results with SPR_DCR_CC mask */
+#define SPR_DCR_CC_MASKED 0x00000000
+#define SPR_DCR_CC_EQUAL 0x00000002
+#define SPR_DCR_CC_LESS 0x00000004
+#define SPR_DCR_CC_LESSE 0x00000006
+#define SPR_DCR_CC_GREAT 0x00000008
+#define SPR_DCR_CC_GREATE 0x0000000a
+#define SPR_DCR_CC_NEQUAL 0x0000000c
+
+/* Bit results with SPR_DCR_CT mask */
+#define SPR_DCR_CT_DISABLED 0x00000000
+#define SPR_DCR_CT_IFEA 0x00000020
+#define SPR_DCR_CT_LEA 0x00000040
+#define SPR_DCR_CT_SEA 0x00000060
+#define SPR_DCR_CT_LD 0x00000080
+#define SPR_DCR_CT_SD 0x000000a0
+#define SPR_DCR_CT_LSEA 0x000000c0
+#define SPR_DCR_CT_LSD 0x000000e0
+/* SPR_DCR_CT_LSD doesn't seem to be implemented anywhere in or1ksim. 2004-1-30 HP */
+
+/*
+ * Bit definitions for Debug Mode 1 register
+ *
+ */
+#define SPR_DMR1_CW 0x000fffff /* Chain register pair data */
+#define SPR_DMR1_CW0_AND 0x00000001
+#define SPR_DMR1_CW0_OR 0x00000002
+#define SPR_DMR1_CW0 (SPR_DMR1_CW0_AND | SPR_DMR1_CW0_OR)
+#define SPR_DMR1_CW1_AND 0x00000004
+#define SPR_DMR1_CW1_OR 0x00000008
+#define SPR_DMR1_CW1 (SPR_DMR1_CW1_AND | SPR_DMR1_CW1_OR)
+#define SPR_DMR1_CW2_AND 0x00000010
+#define SPR_DMR1_CW2_OR 0x00000020
+#define SPR_DMR1_CW2 (SPR_DMR1_CW2_AND | SPR_DMR1_CW2_OR)
+#define SPR_DMR1_CW3_AND 0x00000040
+#define SPR_DMR1_CW3_OR 0x00000080
+#define SPR_DMR1_CW3 (SPR_DMR1_CW3_AND | SPR_DMR1_CW3_OR)
+#define SPR_DMR1_CW4_AND 0x00000100
+#define SPR_DMR1_CW4_OR 0x00000200
+#define SPR_DMR1_CW4 (SPR_DMR1_CW4_AND | SPR_DMR1_CW4_OR)
+#define SPR_DMR1_CW5_AND 0x00000400
+#define SPR_DMR1_CW5_OR 0x00000800
+#define SPR_DMR1_CW5 (SPR_DMR1_CW5_AND | SPR_DMR1_CW5_OR)
+#define SPR_DMR1_CW6_AND 0x00001000
+#define SPR_DMR1_CW6_OR 0x00002000
+#define SPR_DMR1_CW6 (SPR_DMR1_CW6_AND | SPR_DMR1_CW6_OR)
+#define SPR_DMR1_CW7_AND 0x00004000
+#define SPR_DMR1_CW7_OR 0x00008000
+#define SPR_DMR1_CW7 (SPR_DMR1_CW7_AND | SPR_DMR1_CW7_OR)
+#define SPR_DMR1_CW8_AND 0x00010000
+#define SPR_DMR1_CW8_OR 0x00020000
+#define SPR_DMR1_CW8 (SPR_DMR1_CW8_AND | SPR_DMR1_CW8_OR)
+#define SPR_DMR1_CW9_AND 0x00040000
+#define SPR_DMR1_CW9_OR 0x00080000
+#define SPR_DMR1_CW9 (SPR_DMR1_CW9_AND | SPR_DMR1_CW9_OR)
+#define SPR_DMR1_RES1 0x00300000 /* Reserved */
+#define SPR_DMR1_ST 0x00400000 /* Single-step trace*/
+#define SPR_DMR1_BT 0x00800000 /* Branch trace */
+#define SPR_DMR1_RES2 0xff000000 /* Reserved */
+
+/*
+ * Bit definitions for Debug Mode 2 register. AWTC and WGB corrected by JPB
+ *
+ */
+#define SPR_DMR2_WCE0 0x00000001 /* Watchpoint counter 0 enable */
+#define SPR_DMR2_WCE1 0x00000002 /* Watchpoint counter 0 enable */
+#define SPR_DMR2_AWTC 0x00000ffc /* Assign watchpoints to counters */
+#define SPR_DMR2_AWTC_OFF 2 /* Bit offset to AWTC field */
+#define SPR_DMR2_WGB 0x003ff000 /* Watchpoints generating breakpoint */
+#define SPR_DMR2_WGB_OFF 12 /* Bit offset to WGB field */
+#define SPR_DMR2_WBS 0xffc00000 /* JPB: Watchpoint status */
+#define SPR_DMR2_WBS_OFF 22 /* Bit offset to WBS field */
+
+/*
+ * Bit definitions for Debug watchpoint counter registers
+ *
+ */
+#define SPR_DWCR_COUNT 0x0000ffff /* Count */
+#define SPR_DWCR_MATCH 0xffff0000 /* Match */
+#define SPR_DWCR_MATCH_OFF 16 /* Match bit offset */
+
+/*
+ * Bit definitions for Debug stop register
+ *
+ */
+#define SPR_DSR_RSTE 0x00000001 /* Reset exception */
+#define SPR_DSR_BUSEE 0x00000002 /* Bus error exception */
+#define SPR_DSR_DPFE 0x00000004 /* Data Page Fault exception */
+#define SPR_DSR_IPFE 0x00000008 /* Insn Page Fault exception */
+#define SPR_DSR_TTE 0x00000010 /* Tick Timer exception */
+#define SPR_DSR_AE 0x00000020 /* Alignment exception */
+#define SPR_DSR_IIE 0x00000040 /* Illegal Instruction exception */
+#define SPR_DSR_IE 0x00000080 /* Interrupt exception */
+#define SPR_DSR_DME 0x00000100 /* DTLB miss exception */
+#define SPR_DSR_IME 0x00000200 /* ITLB miss exception */
+#define SPR_DSR_RE 0x00000400 /* Range exception */
+#define SPR_DSR_SCE 0x00000800 /* System call exception */
+#define SPR_DSR_FPE 0x00001000 /* Floating Point Exception */
+#define SPR_DSR_TE 0x00002000 /* Trap exception */
+
+/*
+ * Bit definitions for Debug reason register
+ *
+ */
+#define SPR_DRR_RSTE 0x00000001 /* Reset exception */
+#define SPR_DRR_BUSEE 0x00000002 /* Bus error exception */
+#define SPR_DRR_DPFE 0x00000004 /* Data Page Fault exception */
+#define SPR_DRR_IPFE 0x00000008 /* Insn Page Fault exception */
+#define SPR_DRR_TTE 0x00000010 /* Tick Timer exception */
+#define SPR_DRR_AE 0x00000020 /* Alignment exception */
+#define SPR_DRR_IIE 0x00000040 /* Illegal Instruction exception */
+#define SPR_DRR_IE 0x00000080 /* Interrupt exception */
+#define SPR_DRR_DME 0x00000100 /* DTLB miss exception */
+#define SPR_DRR_IME 0x00000200 /* ITLB miss exception */
+#define SPR_DRR_RE 0x00000400 /* Range exception */
+#define SPR_DRR_SCE 0x00000800 /* System call exception */
+#define SPR_DRR_FPE 0x00001000 /* Floating Point Exception */
+#define SPR_DRR_TE 0x00002000 /* Trap exception */
+
+/*
+ * Bit definitions for Performance counters mode registers
+ *
+ */
+#define SPR_PCMR_CP 0x00000001 /* Counter present */
+#define SPR_PCMR_UMRA 0x00000002 /* User mode read access */
+#define SPR_PCMR_CISM 0x00000004 /* Count in supervisor mode */
+#define SPR_PCMR_CIUM 0x00000008 /* Count in user mode */
+#define SPR_PCMR_LA 0x00000010 /* Load access event */
+#define SPR_PCMR_SA 0x00000020 /* Store access event */
+#define SPR_PCMR_IF 0x00000040 /* Instruction fetch event*/
+#define SPR_PCMR_DCM 0x00000080 /* Data cache miss event */
+#define SPR_PCMR_ICM 0x00000100 /* Insn cache miss event */
+#define SPR_PCMR_IFS 0x00000200 /* Insn fetch stall event */
+#define SPR_PCMR_LSUS 0x00000400 /* LSU stall event */
+#define SPR_PCMR_BS 0x00000800 /* Branch stall event */
+#define SPR_PCMR_DTLBM 0x00001000 /* DTLB miss event */
+#define SPR_PCMR_ITLBM 0x00002000 /* ITLB miss event */
+#define SPR_PCMR_DDS 0x00004000 /* Data dependency stall event */
+#define SPR_PCMR_WPE 0x03ff8000 /* Watchpoint events */
+
+/*
+ * Bit definitions for the Power management register
+ *
+ */
+#define SPR_PMR_SDF 0x0000000f /* Slow down factor */
+#define SPR_PMR_DME 0x00000010 /* Doze mode enable */
+#define SPR_PMR_SME 0x00000020 /* Sleep mode enable */
+#define SPR_PMR_DCGE 0x00000040 /* Dynamic clock gating enable */
+#define SPR_PMR_SUME 0x00000080 /* Suspend mode enable */
+
+/*
+ * Bit definitions for PICMR
+ *
+ */
+#define SPR_PICMR_IUM 0xfffffffc /* Interrupt unmask */
+
+/*
+ * Bit definitions for PICPR
+ *
+ */
+#define SPR_PICPR_IPRIO 0xfffffffc /* Interrupt priority */
+
+/*
+ * Bit definitions for PICSR
+ *
+ */
+#define SPR_PICSR_IS 0xffffffff /* Interrupt status */
+
+/*
+ * Bit definitions for Tick Timer Control Register
+ *
+ */
+
+#define SPR_TTCR_CNT 0xffffffff /* Count, time period */
+#define SPR_TTMR_TP 0x0fffffff /* Time period */
+#define SPR_TTMR_IP 0x10000000 /* Interrupt Pending */
+#define SPR_TTMR_IE 0x20000000 /* Interrupt Enable */
+#define SPR_TTMR_DI 0x00000000 /* Disabled */
+#define SPR_TTMR_RT 0x40000000 /* Restart tick */
+#define SPR_TTMR_SR 0x80000000 /* Single run */
+#define SPR_TTMR_CR 0xc0000000 /* Continuous run */
+#define SPR_TTMR_M 0xc0000000 /* Tick mode */
+
+/*
+ * Bit definitions for the FP Control Status Register
+ *
+ */
+#define SPR_FPCSR_FPEE 0x00000001 /* Floating Point Exception Enable */
+#define SPR_FPCSR_RM 0x00000006 /* Rounding Mode */
+#define SPR_FPCSR_OVF 0x00000008 /* Overflow Flag */
+#define SPR_FPCSR_UNF 0x00000010 /* Underflow Flag */
+#define SPR_FPCSR_SNF 0x00000020 /* SNAN Flag */
+#define SPR_FPCSR_QNF 0x00000040 /* QNAN Flag */
+#define SPR_FPCSR_ZF 0x00000080 /* Zero Flag */
+#define SPR_FPCSR_IXF 0x00000100 /* Inexact Flag */
+#define SPR_FPCSR_IVF 0x00000200 /* Invalid Flag */
+#define SPR_FPCSR_INF 0x00000400 /* Infinity Flag */
+#define SPR_FPCSR_DZF 0x00000800 /* Divide By Zero Flag */
+#define SPR_FPCSR_ALLF (SPR_FPCSR_OVF | SPR_FPCSR_UNF | SPR_FPCSR_SNF | \
+ SPR_FPCSR_QNF | SPR_FPCSR_ZF | SPR_FPCSR_IXF | \
+ SPR_FPCSR_IVF | SPR_FPCSR_INF | SPR_FPCSR_DZF)
+
+#define FPCSR_RM_RN (0<<1)
+#define FPCSR_RM_RZ (1<<1)
+#define FPCSR_RM_RIP (2<<1)
+#define FPCSR_RM_RIN (3<<1)
+
+/*
+ * l.nop constants
+ *
+ */
+#define NOP_NOP 0x0000 /* Normal nop instruction */
+#define NOP_EXIT 0x0001 /* End of simulation */
+#define NOP_REPORT 0x0002 /* Simple report */
+/*#define NOP_PRINTF 0x0003 Simprintf instruction (obsolete)*/
+#define NOP_PUTC 0x0004 /* JPB: Simputc instruction */
+#define NOP_CNT_RESET 0x0005 /* Reset statistics counters */
+#define NOP_GET_TICKS 0x0006 /* JPB: Get # ticks running */
+#define NOP_GET_PS 0x0007 /* JPB: Get picosecs/cycle */
+#define NOP_REPORT_FIRST 0x0400 /* Report with number */
+#define NOP_REPORT_LAST 0x03ff /* Report with number */
+
+#endif /* SPR_DEFS__H */
diff --git a/arch/openrisc/include/asm/system.h b/arch/openrisc/include/asm/system.h
new file mode 100644
index 0000000..cf65888
--- /dev/null
+++ b/arch/openrisc/include/asm/system.h
@@ -0,0 +1,35 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_SYSTEM_H
+#define __ASM_OPENRISC_SYSTEM_H
+
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
+
+#include <asm/spr.h>
+#include <asm-generic/system.h>
+
+/* We probably need this definition, but the generic system.h provides it
+ * and it's not used on our arch anyway...
+ */
+/*#define nop() __asm__ __volatile__ ("l.nop"::)*/
+
+#endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+#endif /* __ASM_OPENRISC_SYSTEM_H */
diff --git a/arch/openrisc/include/asm/unaligned.h b/arch/openrisc/include/asm/unaligned.h
new file mode 100644
index 0000000..1141cbd
--- /dev/null
+++ b/arch/openrisc/include/asm/unaligned.h
@@ -0,0 +1,51 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_UNALIGNED_H
+#define __ASM_OPENRISC_UNALIGNED_H
+
+/*
+ * This is copied from the generic implementation and the C-struct
+ * variant replaced with the memmove variant. The GCC compiler
+ * for the OR32 arch optimizes too aggressively for the C-struct
+ * variant to work, so use the memmove variant instead.
+ *
+ * It may be worth considering implementing the unaligned access
+ * exception handler and allowing unaligned accesses (access_ok.h)...
+ * not sure if it would be much of a performance win without further
+ * investigation.
+ */
+#include <asm/byteorder.h>
+
+#if defined(__LITTLE_ENDIAN)
+# include <linux/unaligned/le_memmove.h>
+# include <linux/unaligned/be_byteshift.h>
+# include <linux/unaligned/generic.h>
+# define get_unaligned __get_unaligned_le
+# define put_unaligned __put_unaligned_le
+#elif defined(__BIG_ENDIAN)
+# include <linux/unaligned/be_memmove.h>
+# include <linux/unaligned/le_byteshift.h>
+# include <linux/unaligned/generic.h>
+# define get_unaligned __get_unaligned_be
+# define put_unaligned __put_unaligned_be
+#else
+# error need to define endianess
+#endif
+
+#endif /* __ASM_OPENRISC_UNALIGNED_H */
diff --git a/arch/openrisc/kernel/vmlinux.h b/arch/openrisc/kernel/vmlinux.h
new file mode 100644
index 0000000..ee842a2
--- /dev/null
+++ b/arch/openrisc/kernel/vmlinux.h
@@ -0,0 +1,12 @@
+#ifndef __OPENRISC_VMLINUX_H_
+#define __OPENRISC_VMLINUX_H_
+
+extern char _stext, _etext, _edata, _end;
+#ifdef CONFIG_BLK_DEV_INITRD
+extern char __initrd_start, __initrd_end;
+extern char __initramfs_start;
+#endif
+
+extern u32 __dtb_start[];
+
+#endif
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:47 UTC
Permalink
This is still our old module support implementation. There's a pending
patch to move a lot of this functionality into default hooks that are
available in core code to all arches; if that patch is accepted, this
patch becomes significantly smaller.

Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/kernel/module.c | 126 +++++++++++++++++++++++++++++++++++++++++
1 files changed, 126 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/kernel/module.c

diff --git a/arch/openrisc/kernel/module.c b/arch/openrisc/kernel/module.c
new file mode 100644
index 0000000..952b129
--- /dev/null
+++ b/arch/openrisc/kernel/module.c
@@ -0,0 +1,126 @@
+/*
+ * OpenRISC module.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/moduleloader.h>
+#include <linux/elf.h>
+#include <linux/vmalloc.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+
+void *module_alloc(unsigned long size)
+{
+ pr_debug("module_alloc size: %lu\n", size);
+
+ if (size == 0)
+ return NULL;
+
+ return vmalloc(size);
+}
+
+
+/* Free memory returned from module_alloc */
+void module_free(struct module *mod, void *module_region)
+{
+ vfree(module_region);
+ /* FIXME: If module_region == mod->init_region, trim exception
+ table entries. */
+}
+
+/* We don't need anything special. */
+int module_frob_arch_sections(Elf_Ehdr *hdr,
+ Elf_Shdr *sechdrs,
+ char *secstrings,
+ struct module *mod)
+{
+ return 0;
+}
+
+int apply_relocate_add(Elf32_Shdr *sechdrs,
+ const char *strtab,
+ unsigned int symindex,
+ unsigned int relsec,
+ struct module *me)
+{
+ unsigned int i;
+ Elf32_Rela *rel = (void *)sechdrs[relsec].sh_addr;
+ Elf32_Sym *sym;
+// Elf32_Addr relocation;
+ uint32_t *location;
+ uint32_t value;
+
+ pr_debug("Applying relocate section %u to %u\n", relsec,
+ sechdrs[relsec].sh_info);
+ for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) {
+ /* This is where to make the change */
+ location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr
+ + rel[i].r_offset;
+
+ /* This is the symbol it is referring to. Note that all
+ undefined symbols have been resolved. */
+ sym = (Elf32_Sym *)sechdrs[symindex].sh_addr
+ + ELF32_R_SYM(rel[i].r_info);
+ value = sym->st_value + rel[i].r_addend;
+
+ switch(ELF32_R_TYPE(rel[i].r_info)) {
+ case R_OR32_32:
+ *location = value;
+ break;
+ case R_OR32_CONST:
+ location = (uint16_t*)location + 1;
+ *((uint16_t*)location) = (uint16_t) (value);
+ break;
+ case R_OR32_CONSTH:
+ location = (uint16_t*)location + 1;
+ *((uint16_t*)location) = (uint16_t) (value >> 16);
+ break;
+ case R_OR32_JUMPTARG:
+ value -= (uint32_t)location;
+ value >>= 2;
+ value &= 0x03ffffff;
+ value |= *location & 0xfc000000;
+ *location = value;
+ break;
+ default:
+ pr_err("module %s: Unknown relocation: %u\n",
+ me->name, ELF32_R_TYPE(rel[i].r_info));
+ break;
+ }
+ }
+
+ return 0;
+}
+
+int apply_relocate(Elf32_Shdr *sechdrs,
+ const char *strtab,
+ unsigned int symindex,
+ unsigned int relsec,
+ struct module *me)
+{
+ pr_err("module %s: REL relocation unsupported\n", me->name);
+ return -ENOEXEC;
+}
+
+int module_finalize(const Elf_Ehdr *hdr,
+ const Elf_Shdr *sechdrs,
+ struct module *me)
+{
+ return 0;
+}
+
+void module_arch_cleanup(struct module *mod)
+{
+}
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:45 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/thread_info.h | 149 +++++++++++++++
arch/openrisc/kernel/idle.c | 2 +-
arch/openrisc/kernel/process.c | 305 +++++++++++++++++++++++++++++++
3 files changed, 455 insertions(+), 1 deletions(-)
create mode 100644 arch/openrisc/include/asm/thread_info.h
create mode 100644 arch/openrisc/kernel/process.c

diff --git a/arch/openrisc/include/asm/thread_info.h b/arch/openrisc/include/asm/thread_info.h
new file mode 100644
index 0000000..8a8fb50
--- /dev/null
+++ b/arch/openrisc/include/asm/thread_info.h
@@ -0,0 +1,149 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef _ASM_THREAD_INFO_H
+#define _ASM_THREAD_INFO_H
+
+#ifdef __KERNEL__
+
+#ifndef __ASSEMBLY__
+#include <asm/types.h>
+#include <asm/processor.h>
+#endif
+
+
+/* THREAD_SIZE is the size of the task_struct/kernel_stack combo.
+ * normally, the stack is found by doing something like p + THREAD_SIZE
+ * in or32, a page is 8192 bytes, which seems like a sane size
+ */
+
+#define THREAD_SIZE_ORDER 0
+#define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER)
+
+/*
+ * low level task data that entry.S needs immediate access to
+ * - this struct should fit entirely inside of one cache line
+ * - this struct shares the supervisor stack pages
+ * - if the contents of this structure are changed, the assembly constants
+ * must also be changed
+ */
+#ifndef __ASSEMBLY__
+
+typedef unsigned long mm_segment_t;
+
+struct debug_entry {
+ unsigned long address; /* Address of breakpoint */
+ unsigned long insn; /* Instruction replaced */
+ unsigned long set; /* Is BP active? */
+ unsigned long branch; /* BP is after a branch */
+ unsigned long branch_target; /* Branch address */
+ unsigned long branch_insn_address; /* PC of l.b/l.j */
+};
+
+struct debug_info {
+ struct debug_entry bp;
+};
+
+struct thread_info {
+ struct task_struct *task; /* main task structure */
+ struct exec_domain *exec_domain; /* execution domain */
+ unsigned long flags; /* low level flags */
+ __u32 cpu; /* current CPU */
+ __s32 preempt_count; /* 0 => preemptable, <0 => BUG */
+
+ mm_segment_t addr_limit; /* thread address space:
+ 0-0x7FFFFFFF for user-thead
+ 0-0xFFFFFFFF for kernel-thread
+ */
+ struct restart_block restart_block;
+ __u8 supervisor_stack[0];
+
+ /* saved context data */
+ unsigned long ksp;
+
+ struct debug_info debug;
+};
+#endif
+
+/*
+ * macros/functions for gaining access to the thread information structure
+ *
+ * preempt_count needs to be 1 initially, until the scheduler is functional.
+ */
+#ifndef __ASSEMBLY__
+#define INIT_THREAD_INFO(tsk) \
+{ \
+ .task = &tsk, \
+ .exec_domain = &default_exec_domain, \
+ .flags = 0, \
+ .cpu = 0, \
+ .preempt_count = 1, \
+ .addr_limit = KERNEL_DS, \
+ .restart_block = { \
+ .fn = do_no_restart_syscall, \
+ }, \
+ .ksp = 0, \
+}
+
+#define init_thread_info (init_thread_union.thread_info)
+
+/* how to get the thread information struct from C */
+register struct thread_info *current_thread_info_reg asm("r10");
+#define current_thread_info() (current_thread_info_reg)
+
+#define get_thread_info(ti) get_task_struct((ti)->task)
+#define put_thread_info(ti) put_task_struct((ti)->task)
+
+#endif /* !__ASSEMBLY__ */
+
+/*
+ * thread information flags
+ * these are process state flags that various assembly files may need to
+ * access
+ * - pending work-to-be-done flags are in LSW
+ * - other flags in MSW
+ */
+#define TIF_SYSCALL_TRACE 0 /* syscall trace active */
+#define TIF_NOTIFY_RESUME 1 /* resumption notification requested */
+#define TIF_SIGPENDING 2 /* signal pending */
+#define TIF_NEED_RESCHED 3 /* rescheduling necessary */
+#define TIF_SINGLESTEP 4 /* restore singlestep on return to user
+ * mode
+ */
+#define TIF_SYSCALL_TRACEPOINT 8 /* for ftrace syscall instrumentation */
+#define TIF_RESTORE_SIGMASK 9
+#define TIF_POLLING_NRFLAG 16 /* true if poll_idle() is polling * TIF_NEED_RESCHED
+ */
+#define TIF_MEMDIE 17
+
+#define _TIF_SYSCALL_TRACE (1<<TIF_SYSCALL_TRACE)
+#define _TIF_NOTIFY_RESUME (1<<TIF_NOTIFY_RESUME)
+#define _TIF_SIGPENDING (1<<TIF_SIGPENDING)
+#define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED)
+#define _TIF_SINGLESTEP (1<<TIF_SINGLESTEP)
+#define _TIF_RESTORE_SIGMASK (1<<TIF_RESTORE_SIGMASK)
+#define _TIF_POLLING_NRFLAG (1<<TIF_POLLING_NRFLAG)
+
+
+/* Work to do when returning from interrupt/exception */
+/* For OpenRISC, this is anything in the LSW other than syscall trace */
+#define _TIF_WORK_MASK (0xff & ~(_TIF_SYSCALL_TRACE|_TIF_SINGLESTEP))
+
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_THREAD_INFO_H */
diff --git a/arch/openrisc/kernel/idle.c b/arch/openrisc/kernel/idle.c
index 495a772..d5bc5f8 100644
--- a/arch/openrisc/kernel/idle.c
+++ b/arch/openrisc/kernel/idle.c
@@ -62,7 +62,7 @@ void cpu_idle(void)
local_irq_disable();
/* Don't trace irqs off for idle */
stop_critical_timings();
- if (powersave != NULL)
+ if (!need_resched() && powersave != NULL)
powersave();
start_critical_timings();
local_irq_enable();
diff --git a/arch/openrisc/kernel/process.c b/arch/openrisc/kernel/process.c
new file mode 100644
index 0000000..852f618
--- /dev/null
+++ b/arch/openrisc/kernel/process.c
@@ -0,0 +1,305 @@
+/*
+ * OpenRISC process.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This file handles the architecture-dependent parts of process handling...
+ */
+
+#define __KERNEL_SYSCALLS__
+#include <stdarg.h>
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/elfcore.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/init_task.h>
+#include <linux/mqueue.h>
+#include <linux/fs.h>
+
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/spr_defs.h>
+
+#include <linux/smp.h>
+
+/*
+ * Pointer to Current thread info structure.
+ *
+ * Used at user space -> kernel transitions.
+ */
+struct thread_info *current_thread_info_set[NR_CPUS] = { &init_thread_info, };
+
+void machine_restart(void)
+{
+ printk(KERN_INFO "*** MACHINE RESTART ***\n");
+ __asm__("l.nop 1");
+}
+
+/*
+ * Similar to machine_power_off, but don't shut off power. Add code
+ * here to freeze the system for e.g. post-mortem debug purpose when
+ * possible. This halt has nothing to do with the idle halt.
+ */
+void machine_halt(void)
+{
+ printk(KERN_INFO "*** MACHINE HALT ***\n");
+ __asm__("l.nop 1");
+}
+
+/* If or when software power-off is implemented, add code here. */
+void machine_power_off(void)
+{
+ printk(KERN_INFO "*** MACHINE POWER OFF ***\n");
+ __asm__("l.nop 1");
+}
+
+void (*pm_power_off) (void) = machine_power_off;
+
+/*
+ * When a process does an "exec", machine state like FPU and debug
+ * registers need to be reset. This is a hook function for that.
+ * Currently we don't have any such state to reset, so this is empty.
+ */
+void flush_thread(void)
+{
+}
+
+void show_regs(struct pt_regs *regs)
+{
+ extern void show_registers(struct pt_regs *regs);
+
+ /* __PHX__ cleanup this mess */
+ show_registers(regs);
+}
+
+unsigned long thread_saved_pc(struct task_struct *t)
+{
+ return (unsigned long)user_regs(t->stack)->pc;
+}
+
+void release_thread(struct task_struct *dead_task)
+{
+}
+
+/*
+ * Copy the thread-specific (arch specific) info from the current
+ * process to the new one p
+ */
+extern asmlinkage void ret_from_fork(void);
+
+int
+copy_thread(unsigned long clone_flags, unsigned long usp,
+ unsigned long unused, struct task_struct *p, struct pt_regs *regs)
+{
+ struct pt_regs *childregs;
+ struct pt_regs *kregs;
+ unsigned long sp = (unsigned long)task_stack_page(p) + THREAD_SIZE;
+ struct thread_info *ti;
+ unsigned long top_of_kernel_stack;
+
+ top_of_kernel_stack = sp;
+
+ p->set_child_tid = p->clear_child_tid = NULL;
+
+ /* Copy registers */
+ /* redzone */
+ sp -= STACK_FRAME_OVERHEAD;
+ sp -= sizeof(struct pt_regs);
+ childregs = (struct pt_regs *)sp;
+
+ /* Copy parent registers */
+ *childregs = *regs;
+
+ if ((childregs->sr & SPR_SR_SM) == 1) {
+ /* for kernel thread, set `current_thread_info'
+ * and stackptr in new task
+ */
+ childregs->sp = (unsigned long)task_stack_page(p) + THREAD_SIZE;
+ childregs->gpr[10] = (unsigned long)task_thread_info(p);
+ } else {
+ childregs->sp = usp;
+ }
+
+ childregs->gpr[11] = 0; /* Result from fork() */
+
+ /*
+ * The way this works is that at some point in the future
+ * some task will call _switch to switch to the new task.
+ * That will pop off the stack frame created below and start
+ * the new task running at ret_from_fork. The new task will
+ * do some house keeping and then return from the fork or clone
+ * system call, using the stack frame created above.
+ */
+ /* redzone */
+ sp -= STACK_FRAME_OVERHEAD;
+ sp -= sizeof(struct pt_regs);
+ kregs = (struct pt_regs *)sp;
+
+ ti = task_thread_info(p);
+ ti->ksp = sp;
+
+ /* kregs->sp must store the location of the 'pre-switch' kernel stack
+ * pointer... for a newly forked process, this is simply the top of
+ * the kernel stack.
+ */
+ kregs->sp = top_of_kernel_stack;
+ kregs->gpr[3] = (unsigned long)current; /* arg to schedule_tail */
+ kregs->gpr[10] = (unsigned long)task_thread_info(p);
+ kregs->gpr[9] = (unsigned long)ret_from_fork;
+
+ return 0;
+}
+
+/*
+ * Set up a thread for executing a new program
+ */
+void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long sp)
+{
+ unsigned long sr = regs->sr & ~SPR_SR_SM;
+
+ set_fs(USER_DS);
+ memset(regs->gpr, 0, sizeof(regs->gpr));
+
+ regs->pc = pc;
+ regs->sr = sr;
+ regs->sp = sp;
+
+/* printk("start thread, ksp = %lx\n", current_thread_info()->ksp);*/
+}
+
+/* Fill in the fpu structure for a core dump. */
+int dump_fpu(struct pt_regs *regs, elf_fpregset_t * fpu)
+{
+ /* TODO */
+ return 0;
+}
+
+extern struct thread_info *_switch(struct thread_info *old_ti,
+ struct thread_info *new_ti);
+
+struct task_struct *__switch_to(struct task_struct *old,
+ struct task_struct *new)
+{
+ struct task_struct *last;
+ struct thread_info *new_ti, *old_ti;
+ unsigned long flags;
+
+ local_irq_save(flags);
+
+ /* current_set is an array of saved current pointers
+ * (one for each cpu). we need them at user->kernel transition,
+ * while we save them at kernel->user transition
+ */
+ new_ti = new->stack;
+ old_ti = old->stack;
+
+ current_thread_info_set[smp_processor_id()] = new_ti;
+ last = (_switch(old_ti, new_ti))->task;
+
+ local_irq_restore(flags);
+
+ return last;
+}
+
+/*
+ * fill in the user structure for a core dump..
+ */
+void dump_thread(struct pt_regs *regs, struct user *dump)
+{
+ /* TODO */
+}
+
+extern void _kernel_thread_helper(void);
+
+void __noreturn kernel_thread_helper(int (*fn) (void *), void *arg)
+{
+ do_exit(fn(arg));
+}
+
+/*
+ * Create a kernel thread.
+ */
+int kernel_thread(int (*fn) (void *), void *arg, unsigned long flags)
+{
+ struct pt_regs regs;
+
+ memset(&regs, 0, sizeof(regs));
+
+ regs.gpr[20] = (unsigned long)fn;
+ regs.gpr[22] = (unsigned long)arg;
+ regs.sr = mfspr(SPR_SR);
+ regs.pc = (unsigned long)_kernel_thread_helper;
+
+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED,
+ 0, &regs, 0, NULL, NULL);
+}
+
+/*
+ * sys_execve() executes a new program.
+ */
+asmlinkage long _sys_execve(const char __user *name,
+ const char __user * const __user *argv,
+ const char __user * const __user *envp,
+ struct pt_regs *regs)
+{
+ int error;
+ char *filename;
+
+ filename = getname(name);
+ error = PTR_ERR(filename);
+
+ if (IS_ERR(filename))
+ goto out;
+
+ error = do_execve(filename, argv, envp, regs);
+ putname(filename);
+
+out:
+ return error;
+}
+
+unsigned long get_wchan(struct task_struct *p)
+{
+ /* TODO */
+
+ return 0;
+}
+
+int kernel_execve(const char *filename, char *const argv[], char *const envp[])
+{
+ register long __res asm("r11") = __NR_execve;
+ register long __a asm("r3") = (long)(filename);
+ register long __b asm("r4") = (long)(argv);
+ register long __c asm("r5") = (long)(envp);
+ __asm__ volatile ("l.sys 1"
+ : "=r" (__res), "=r"(__a), "=r"(__b), "=r"(__c)
+ : "0"(__res), "1"(__a), "2"(__b), "3"(__c)
+ : "r6", "r7", "r8", "r12", "r13", "r15",
+ "r17", "r19", "r21", "r23", "r25", "r27",
+ "r29", "r31");
+ __asm__ volatile ("l.nop");
+ return __res;
+}
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:44 UTC
Permalink
Minimal functionality...

Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/kernel/idle.c | 77 +++++++++++++++++++++++++++++++++++++++++++
1 files changed, 77 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/kernel/idle.c

diff --git a/arch/openrisc/kernel/idle.c b/arch/openrisc/kernel/idle.c
new file mode 100644
index 0000000..495a772
--- /dev/null
+++ b/arch/openrisc/kernel/idle.c
@@ -0,0 +1,77 @@
+/*
+ * OpenRISC idle.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Idle daemon for or32. Idle daemon will handle any action
+ * that needs to be taken when the system becomes idle.
+ */
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/tick.h>
+
+#include <asm/pgtable.h>
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/mmu.h>
+#include <asm/cache.h>
+#include <asm/pgalloc.h>
+
+void (*powersave) (void) = NULL;
+
+static inline void pm_idle(void)
+{
+ barrier();
+}
+
+void cpu_idle(void)
+{
+ set_thread_flag(TIF_POLLING_NRFLAG);
+
+ /* endless idle loop with no priority at all */
+ while (1) {
+ tick_nohz_stop_sched_tick(1);
+
+ while (!need_resched()) {
+ check_pgt_cache();
+ rmb();
+
+ clear_thread_flag(TIF_POLLING_NRFLAG);
+
+ local_irq_disable();
+ /* Don't trace irqs off for idle */
+ stop_critical_timings();
+ if (powersave != NULL)
+ powersave();
+ start_critical_timings();
+ local_irq_enable();
+ set_thread_flag(TIF_POLLING_NRFLAG);
+ }
+
+ tick_nohz_restart_sched_tick();
+ preempt_enable_no_resched();
+ schedule();
+ preempt_disable();
+ }
+}
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:46 UTC
Permalink
Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/gpio.h | 65 ++++++++++++++++++++++++++++++++++++++
1 files changed, 65 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/gpio.h

diff --git a/arch/openrisc/include/asm/gpio.h b/arch/openrisc/include/asm/gpio.h
new file mode 100644
index 0000000..0b0d174
--- /dev/null
+++ b/arch/openrisc/include/asm/gpio.h
@@ -0,0 +1,65 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_GPIO_H
+#define __ASM_OPENRISC_GPIO_H
+
+#include <linux/errno.h>
+#include <asm-generic/gpio.h>
+
+#ifdef CONFIG_GPIOLIB
+
+/*
+ * OpenRISC (or1k) does not have on-chip GPIO's so there is not really
+ * any standardized implementation that makes sense here. If passing
+ * through gpiolib becomes a bottleneck then it may make sense, on a
+ * case-by-case basis, to implement these inlined/rapid versions.
+ *
+ * Just call gpiolib.
+ */
+static inline int gpio_get_value(unsigned int gpio)
+{
+ return __gpio_get_value(gpio);
+}
+
+static inline void gpio_set_value(unsigned int gpio, int value)
+{
+ __gpio_set_value(gpio, value);
+}
+
+static inline int gpio_cansleep(unsigned int gpio)
+{
+ return __gpio_cansleep(gpio);
+}
+
+/*
+ * Not implemented, yet.
+ */
+static inline int gpio_to_irq(unsigned int gpio)
+{
+ return -ENOSYS;
+}
+
+static inline int irq_to_gpio(unsigned int irq)
+{
+ return -EINVAL;
+}
+
+#endif /* CONFIG_GPIOLIB */
+
+#endif /* __ASM_OPENRISC_GPIO_H */
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:43 UTC
Permalink
The OpenRISC Linux kernel conforms to the "generic" syscall interface which
contains only the reduced set of syscalls deemed necessary for new
architectures. Unfortunately, the uClibc port for OpenRISC does not fully
support this reduced set; as such, an additional patch available out-of-tree
needs to be applied to the kernel in order to use the current uClibc. This
is just a temporary measure until the libc port can be straightened out; it
is likely that OpenRISC will make the transition to glibc shortly where the
generic syscall interface is better supported.

Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/syscall.h | 77 ++++++++++++++++++++++++++++++
arch/openrisc/include/asm/syscalls.h | 27 +++++++++++
arch/openrisc/include/asm/unistd.h | 29 +++++++++++
arch/openrisc/kernel/sys_call_table.c | 28 +++++++++++
arch/openrisc/kernel/sys_or32.c | 83 +++++++++++++++++++++++++++++++++
5 files changed, 244 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/syscall.h
create mode 100644 arch/openrisc/include/asm/syscalls.h
create mode 100644 arch/openrisc/include/asm/unistd.h
create mode 100644 arch/openrisc/kernel/sys_call_table.c
create mode 100644 arch/openrisc/kernel/sys_or32.c

diff --git a/arch/openrisc/include/asm/syscall.h b/arch/openrisc/include/asm/syscall.h
new file mode 100644
index 0000000..9f03370
--- /dev/null
+++ b/arch/openrisc/include/asm/syscall.h
@@ -0,0 +1,77 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_SYSCALL_H__
+#define __ASM_OPENRISC_SYSCALL_H__
+
+#include <linux/err.h>
+#include <linux/sched.h>
+
+static inline int
+syscall_get_nr(struct task_struct *task, struct pt_regs *regs)
+{
+ return regs->syscallno ? regs->syscallno : -1;
+}
+
+static inline void
+syscall_rollback(struct task_struct *task, struct pt_regs *regs)
+{
+ regs->gpr[11] = regs->orig_gpr11;
+}
+
+static inline long
+syscall_get_error(struct task_struct *task, struct pt_regs *regs)
+{
+ return IS_ERR_VALUE(regs->gpr[11]) ? regs->gpr[11] : 0;
+}
+
+static inline long
+syscall_get_return_value(struct task_struct *task, struct pt_regs *regs)
+{
+ return regs->gpr[11];
+}
+
+static inline void
+syscall_set_return_value(struct task_struct *task, struct pt_regs *regs,
+ int error, long val)
+{
+ if (error)
+ regs->gpr[11] = -error;
+ else
+ regs->gpr[11] = val;
+}
+
+static inline void
+syscall_get_arguments(struct task_struct *task, struct pt_regs *regs,
+ unsigned int i, unsigned int n, unsigned long *args)
+{
+ BUG_ON(i + n > 6);
+
+ memcpy(args, &regs->gpr[3 + i], n * sizeof(args[0]));
+}
+
+static inline void
+syscall_set_arguments(struct task_struct *task, struct pt_regs *regs,
+ unsigned int i, unsigned int n, const unsigned long *args)
+{
+ BUG_ON(i + n > 6);
+
+ memcpy(&regs->gpr[3 + i], args, n * sizeof(args[0]));
+}
+
+#endif
diff --git a/arch/openrisc/include/asm/syscalls.h b/arch/openrisc/include/asm/syscalls.h
new file mode 100644
index 0000000..84a978a
--- /dev/null
+++ b/arch/openrisc/include/asm/syscalls.h
@@ -0,0 +1,27 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_SYSCALLS_H
+#define __ASM_OPENRISC_SYSCALLS_H
+
+asmlinkage long sys_or1k_atomic(unsigned long type, unsigned long *v1,
+ unsigned long *v2);
+
+#include <asm-generic/syscalls.h>
+
+#endif /* __ASM_OPENRISC_SYSCALLS_H */
diff --git a/arch/openrisc/include/asm/unistd.h b/arch/openrisc/include/asm/unistd.h
new file mode 100644
index 0000000..4a3d6ad
--- /dev/null
+++ b/arch/openrisc/include/asm/unistd.h
@@ -0,0 +1,29 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#if !defined(__ASM_OPENRISC_UNISTD_H) || defined(__SYSCALL)
+#define __ASM_OPENRISC_UNISTD_H
+
+#define __ARCH_HAVE_MMU
+
+#include <asm-generic/unistd.h>
+
+#define __NR_or1k_atomic __NR_arch_specific_syscall
+__SYSCALL(__NR_or1k_atomic, sys_or1k_atomic)
+
+#endif /* __ASM_OPENRISC_UNISTD_H */
diff --git a/arch/openrisc/kernel/sys_call_table.c b/arch/openrisc/kernel/sys_call_table.c
new file mode 100644
index 0000000..e1f8ce8
--- /dev/null
+++ b/arch/openrisc/kernel/sys_call_table.c
@@ -0,0 +1,28 @@
+/*
+ * OpenRISC sys_call_table.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/syscalls.h>
+#include <linux/signal.h>
+#include <linux/unistd.h>
+
+#include <asm/syscalls.h>
+
+#undef __SYSCALL
+#define __SYSCALL(nr, call) [nr] = (call),
+
+void *sys_call_table[__NR_syscalls] = {
+#include <asm/unistd.h>
+};
diff --git a/arch/openrisc/kernel/sys_or32.c b/arch/openrisc/kernel/sys_or32.c
new file mode 100644
index 0000000..a327239
--- /dev/null
+++ b/arch/openrisc/kernel/sys_or32.c
@@ -0,0 +1,83 @@
+/*
+ * OpenRISC sys_or32.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This file contains various random system calls that
+ * have a non-standard calling sequence on some platforms.
+ * Since we don't have to do any backwards compatibility, our
+ * versions are done in the most "normal" way possible.
+ */
+
+#include <linux/errno.h>
+#include <linux/syscalls.h>
+#include <linux/mm.h>
+
+#include <asm/syscalls.h>
+
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+ unsigned long prot, unsigned long flags,
+ unsigned long fd, unsigned long pgoff)
+{
+ return sys_mmap_pgoff(addr, len, prot, flags, fd, pgoff);
+}
+
+asmlinkage long sys_mmap(unsigned long addr, unsigned long len,
+ unsigned long prot, unsigned long flags,
+ unsigned long fd, off_t pgoff)
+{
+ if (pgoff & ~PAGE_MASK)
+ return -EINVAL;
+
+ return sys_mmap_pgoff(addr, len, prot, flags, fd, pgoff >> PAGE_SHIFT);
+}
+
+/* These are secondary entry points as the primary entry points are defined in
+ * entry.S where we add the 'regs' parameter value
+ */
+
+asmlinkage long _sys_clone(unsigned long clone_flags, unsigned long newsp,
+ int __user *parent_tid, int __user *child_tid,
+ struct pt_regs *regs)
+{
+ long ret;
+
+ /* FIXME: Is alignment necessary? */
+ /* newsp = ALIGN(newsp, 4); */
+
+ if (!newsp)
+ newsp = regs->sp;
+
+ ret = do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
+
+ return ret;
+}
+
+asmlinkage int _sys_fork(struct pt_regs *regs)
+{
+#ifdef CONFIG_MMU
+ return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
+#else
+ return -EINVAL;
+#endif
+}
+
+asmlinkage int _sys_vfork(struct pt_regs *regs)
+{
+ /* This doesn't seem to work */
+// return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs, 0, NULL, NULL);
+
+ /* This works */
+ return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
+}
--
1.7.4.1
Arnd Bergmann
2011-07-05 15:39:34 UTC
Permalink
Post by Jonas Bonn
+
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+ unsigned long prot, unsigned long flags,
+ unsigned long fd, unsigned long pgoff)
+{
+ return sys_mmap_pgoff(addr, len, prot, flags, fd, pgoff);
+}
+
+asmlinkage long sys_mmap(unsigned long addr, unsigned long len,
+ unsigned long prot, unsigned long flags,
+ unsigned long fd, off_t pgoff)
+{
+ if (pgoff & ~PAGE_MASK)
+ return -EINVAL;
+
+ return sys_mmap_pgoff(addr, len, prot, flags, fd, pgoff >> PAGE_SHIFT);
+}
You shouldn't really need these two. Either provide just sys_mmap2,
or "#define sys_mmap2 sys_mmap_pgoff" and change your user implementation.

Arnd
Jonas Bonn
2011-07-02 21:15:42 UTC
Permalink
This patch adds support for the OpenRISC PIC.

Signed-off-by: Jonas Bonn <***@southpole.se>
Cc: ***@linutronix.de
---
arch/openrisc/include/asm/irq.h | 27 ++++++
arch/openrisc/include/asm/irqflags.h | 29 ++++++
arch/openrisc/kernel/irq.c | 172 ++++++++++++++++++++++++++++++++++
3 files changed, 228 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/irq.h
create mode 100644 arch/openrisc/include/asm/irqflags.h
create mode 100644 arch/openrisc/kernel/irq.c

diff --git a/arch/openrisc/include/asm/irq.h b/arch/openrisc/include/asm/irq.h
new file mode 100644
index 0000000..eb612b1
--- /dev/null
+++ b/arch/openrisc/include/asm/irq.h
@@ -0,0 +1,27 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_IRQ_H__
+#define __ASM_OPENRISC_IRQ_H__
+
+#define NR_IRQS 32
+#include <asm-generic/irq.h>
+
+#define NO_IRQ (-1)
+
+#endif /* __ASM_OPENRISC_IRQ_H__ */
diff --git a/arch/openrisc/include/asm/irqflags.h b/arch/openrisc/include/asm/irqflags.h
new file mode 100644
index 0000000..dc86c65
--- /dev/null
+++ b/arch/openrisc/include/asm/irqflags.h
@@ -0,0 +1,29 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ * et al.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef ___ASM_OPENRISC_IRQFLAGS_H
+#define ___ASM_OPENRISC_IRQFLAGS_H
+
+#include <asm/spr_defs.h>
+
+#define ARCH_IRQ_DISABLED 0x00
+#define ARCH_IRQ_ENABLED (SPR_SR_IEE|SPR_SR_TEE)
+
+#include <asm-generic/irqflags.h>
+
+#endif /* ___ASM_OPENRISC_IRQFLAGS_H */
diff --git a/arch/openrisc/kernel/irq.c b/arch/openrisc/kernel/irq.c
new file mode 100644
index 0000000..80146f1
--- /dev/null
+++ b/arch/openrisc/kernel/irq.c
@@ -0,0 +1,172 @@
+/*
+ * OpenRISC irq.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/ptrace.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/of.h>
+#include <linux/ftrace.h>
+#include <linux/irq.h>
+#include <linux/seq_file.h>
+#include <linux/kernel_stat.h>
+
+#include <linux/irqflags.h>
+
+/* read interrupt enabled status */
+unsigned long arch_local_save_flags(void)
+{
+ return (mfspr(SPR_SR) & (SPR_SR_IEE|SPR_SR_TEE));
+}
+EXPORT_SYMBOL(arch_local_save_flags);
+
+/* set interrupt enabled status */
+void arch_local_irq_restore(unsigned long flags)
+{
+ mtspr(SPR_SR, ((mfspr(SPR_SR) & ~(SPR_SR_IEE|SPR_SR_TEE)) | flags));
+}
+EXPORT_SYMBOL(arch_local_irq_restore);
+
+
+/* OR1K PIC implementation */
+
+/* We're a couple of cycles faster than the generic implementations with
+ * these 'fast' versions.
+ */
+
+static void or1k_pic_mask(struct irq_data *data)
+{
+ mtspr(SPR_PICMR, mfspr(SPR_PICMR) & ~(1UL << data->irq));
+}
+
+static void or1k_pic_unmask(struct irq_data *data)
+{
+ mtspr(SPR_PICMR, mfspr(SPR_PICMR) | (1UL << data->irq));
+}
+
+static void or1k_pic_ack(struct irq_data *data)
+{
+ /* EDGE-triggered interrupts need to be ack'ed in order to clear
+ * the latch.
+ * LEVER-triggered interrupts do not need to be ack'ed; however,
+ * ack'ing the interrupt has no ill-effect and is quicker than
+ * trying to figure out what type it is...
+ */
+
+ /* The OpenRISC 1000 spec says to write a 1 to the bit to ack the
+ * interrupt, but the OR1200 does this backwards and requires a 0
+ * to be written...
+ */
+
+#ifdef CONFIG_OR1K_1200
+ /* There are two oddities with the OR1200 PIC implementation:
+ * i) LEVEL-triggered interrupts are latched and need to be cleared
+ * ii) the interrupt latch is cleared by writing a 0 to the bit,
+ * as opposed to a 1 as mandated by the spec
+ */
+
+ mtspr(SPR_PICSR, mfspr(SPR_PICSR) & ~(1UL << data->irq));
+#else
+ WARN(1, "Interrupt handling possibily broken\n");
+ mtspr(SPR_PICSR, (1UL << irq));
+#endif
+}
+
+static void or1k_pic_mask_ack(struct irq_data *data)
+{
+ /* Comments for pic_ack apply here, too */
+
+#ifdef CONFIG_OR1K_1200
+ mtspr(SPR_PICSR, mfspr(SPR_PICSR) & ~(1UL << data->irq));
+#else
+ WARN(1, "Interrupt handling possibily broken\n");
+ mtspr(SPR_PICSR, (1UL << irq));
+#endif
+}
+
+static int or1k_pic_set_type(struct irq_data *data, unsigned int flow_type)
+{
+ /* There's nothing to do in the PIC configuration when changing
+ * flow type. Level and edge-triggered interrupts are both
+ * supported, but it's PIC-implementation specific which type
+ * is handled. */
+
+ return irq_setup_alt_chip(data, flow_type);
+}
+
+static inline int pic_get_irq(int first)
+{
+ int irq;
+
+ irq = ffs(mfspr(SPR_PICSR) >> first);
+
+ return irq ? irq + first - 1 : NO_IRQ;
+}
+
+static void __init or1k_irq_init(void)
+{
+ struct irq_chip_generic *gc;
+ struct irq_chip_type *ct;
+
+ /* Disable all interrupts until explicitly requested */
+ mtspr(SPR_PICMR, (0UL));
+
+ gc = irq_alloc_generic_chip("or1k-PIC", 1, 0, 0, handle_level_irq);
+ ct = gc->chip_types;
+
+ ct->chip.irq_unmask = or1k_pic_unmask;
+ ct->chip.irq_mask = or1k_pic_mask;
+ ct->chip.irq_ack = or1k_pic_ack;
+ ct->chip.irq_mask_ack = or1k_pic_mask_ack;
+ ct->chip.irq_set_type = or1k_pic_set_type;
+
+ /* The OR1K PIC can handle both level and edge trigged
+ * interrupts in roughly the same manner
+ */
+#if 0
+ /* FIXME: chip.type??? */
+ ct->chip.type = IRQ_TYPE_EDGE_BOTH | IRQ_TYPE_LEVEL_MASK;
+#endif
+
+ irq_setup_generic_chip(gc, IRQ_MSK(NR_IRQS), 0,
+ IRQ_NOREQUEST, IRQ_LEVEL | IRQ_NOPROBE);
+}
+
+void __init init_IRQ(void)
+{
+ or1k_irq_init();
+}
+
+void __irq_entry do_IRQ(struct pt_regs *regs)
+{
+ int irq = -1;
+ struct pt_regs *old_regs = set_irq_regs(regs);
+
+ irq_enter();
+
+ while ((irq = pic_get_irq(irq + 1)) != NO_IRQ)
+ generic_handle_irq(irq);
+
+ irq_exit();
+ set_irq_regs(old_regs);
+}
+
+unsigned int irq_create_of_mapping(struct device_node *controller,
+ const u32 *intspec, unsigned int intsize)
+{
+ return intspec[0];
+}
+EXPORT_SYMBOL_GPL(irq_create_of_mapping);
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:41 UTC
Permalink
Implements support for the OpenRISC timer which is a 28 bit cycle counter
that can be read out of a special purpose register. This counter is
used as a both a clock event and clocksource device.

Signed-off-by: Jonas Bonn <***@southpole.se>
Cc: ***@linutronix.de
---
arch/openrisc/include/asm/timex.h | 36 ++++++++
arch/openrisc/kernel/time.c | 181 +++++++++++++++++++++++++++++++++++++
2 files changed, 217 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/timex.h
create mode 100644 arch/openrisc/kernel/time.c

diff --git a/arch/openrisc/include/asm/timex.h b/arch/openrisc/include/asm/timex.h
new file mode 100644
index 0000000..9935cad
--- /dev/null
+++ b/arch/openrisc/include/asm/timex.h
@@ -0,0 +1,36 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_TIMEX_H
+#define __ASM_OPENRISC_TIMEX_H
+
+#define get_cycles get_cycles
+
+#include <asm-generic/timex.h>
+#include <asm/spr.h>
+#include <asm/spr_defs.h>
+
+static inline cycles_t get_cycles(void)
+{
+ return mfspr(SPR_TTCR);
+}
+
+/* This isn't really used any more */
+#define CLOCK_TICK_RATE 1000
+
+#define ARCH_HAS_READ_CURRENT_TIMER
+
+#endif
diff --git a/arch/openrisc/kernel/time.c b/arch/openrisc/kernel/time.c
new file mode 100644
index 0000000..bd946ef
--- /dev/null
+++ b/arch/openrisc/kernel/time.c
@@ -0,0 +1,181 @@
+/*
+ * OpenRISC time.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/time.h>
+#include <linux/timex.h>
+#include <linux/interrupt.h>
+#include <linux/ftrace.h>
+
+#include <linux/clocksource.h>
+#include <linux/clockchips.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+
+#include <asm/cpuinfo.h>
+
+static int openrisc_timer_set_next_event(unsigned long delta,
+ struct clock_event_device *dev)
+{
+ u32 c;
+
+ /* Read 32-bit counter value, add delta, mask off the low 28 bits.
+ * We're guaranteed delta won't be bigger than 28 bits because the
+ * generic timekeeping code ensures that for us.
+ */
+ c = mfspr(SPR_TTCR);
+ c += delta;
+ c &= SPR_TTMR_TP;
+
+ /* Set counter and enable interrupt.
+ * Keep timer in continuous mode always.
+ */
+ mtspr(SPR_TTMR, SPR_TTMR_CR | SPR_TTMR_IE | c);
+
+ return 0;
+}
+
+static void openrisc_timer_set_mode(enum clock_event_mode mode,
+ struct clock_event_device *evt)
+{
+ switch (mode) {
+ case CLOCK_EVT_MODE_PERIODIC:
+ pr_debug(KERN_INFO "%s: periodic\n", __func__);
+ BUG();
+ break;
+ case CLOCK_EVT_MODE_ONESHOT:
+ pr_debug(KERN_INFO "%s: oneshot\n", __func__);
+ break;
+ case CLOCK_EVT_MODE_UNUSED:
+ pr_debug(KERN_INFO "%s: unused\n", __func__);
+ break;
+ case CLOCK_EVT_MODE_SHUTDOWN:
+ pr_debug(KERN_INFO "%s: shutdown\n", __func__);
+ break;
+ case CLOCK_EVT_MODE_RESUME:
+ pr_debug(KERN_INFO "%s: resume\n", __func__);
+ break;
+ }
+}
+
+/* This is the clock event device based on the OR1K tick timer.
+ * As the timer is being used as a continuous clock-source (required for HR
+ * timers) we cannot enable the PERIODIC feature. The tick timer can run using
+ * one-shot events, so no problem.
+ */
+
+static struct clock_event_device clockevent_openrisc_timer = {
+ .name = "openrisc_timer_clockevent",
+ .features = CLOCK_EVT_FEAT_ONESHOT,
+ .rating = 300,
+ .set_next_event = openrisc_timer_set_next_event,
+ .set_mode = openrisc_timer_set_mode,
+};
+
+static inline void timer_ack(void)
+{
+ /* Clear the IP bit and disable further interrupts */
+ /* This can be done very simply... we just need to keep the timer
+ running, so just maintain the CR bits while clearing the rest
+ of the register
+ */
+ mtspr(SPR_TTMR, SPR_TTMR_CR);
+}
+
+/*
+ * The timer interrupt is mostly handled in generic code nowadays... this
+ * function just acknowledges the interrupt and fires the event handler that
+ * has been set on the clockevent device by the generic time management code.
+ *
+ * This function needs to be called by the timer exception handler and that's
+ * all the exception handler needs to do.
+ */
+
+irqreturn_t __irq_entry timer_interrupt(struct pt_regs *regs)
+{
+ struct pt_regs *old_regs = set_irq_regs(regs);
+ struct clock_event_device *evt = &clockevent_openrisc_timer;
+
+ timer_ack();
+
+ /*
+ * update_process_times() expects us to have called irq_enter().
+ */
+ irq_enter();
+ evt->event_handler(evt);
+ irq_exit();
+
+ set_irq_regs(old_regs);
+
+ return IRQ_HANDLED;
+}
+
+static __init void openrisc_clockevent_init(void)
+{
+ clockevents_calc_mult_shift(&clockevent_openrisc_timer,
+ cpuinfo.clock_frequency, 4);
+
+ /* We only have 28 bits */
+ clockevent_openrisc_timer.max_delta_ns =
+ clockevent_delta2ns((u32) 0x0fffffff, &clockevent_openrisc_timer);
+ clockevent_openrisc_timer.min_delta_ns =
+ clockevent_delta2ns(1, &clockevent_openrisc_timer);
+ clockevent_openrisc_timer.cpumask = cpumask_of(0);
+ clockevents_register_device(&clockevent_openrisc_timer);
+}
+
+/**
+ * Clocksource: Based on OpenRISC timer/counter
+ *
+ * This sets up the OpenRISC Tick Timer as a clock source. The tick timer
+ * is 32 bits wide and runs at the CPU clock frequency.
+ */
+
+static cycle_t openrisc_timer_read(struct clocksource *cs)
+{
+ return (cycle_t) mfspr(SPR_TTCR);
+}
+
+static struct clocksource openrisc_timer = {
+ .name = "openrisc_timer",
+ .rating = 200,
+ .read = openrisc_timer_read,
+ .mask = CLOCKSOURCE_MASK(32),
+ .flags = CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static int __init openrisc_timer_init(void)
+{
+ if (clocksource_register_hz(&openrisc_timer, cpuinfo.clock_frequency))
+ panic("failed to register clocksource");
+
+ /* Enable the incrementer: 'continuous' mode with interrupt disabled */
+ mtspr(SPR_TTMR, SPR_TTMR_CR);
+
+ return 0;
+}
+
+void __init time_init(void)
+{
+ u32 upr;
+
+ upr = mfspr(SPR_UPR);
+ if (!(upr & SPR_UPR_TTP))
+ panic("Linux not supported on devices without tick timer");
+
+ openrisc_timer_init();
+ openrisc_clockevent_init();
+}
--
1.7.4.1
Jonas Bonn
2011-07-02 21:15:40 UTC
Permalink
Simple DMA implementation. Allows for allocation of coherent memory
(simply uncached) for DMA operations.

Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/include/asm/dma-mapping.h | 62 ++++++++++++++++++++++
arch/openrisc/kernel/dma.c | 86 +++++++++++++++++++++++++++++++
2 files changed, 148 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/dma-mapping.h
create mode 100644 arch/openrisc/kernel/dma.c

diff --git a/arch/openrisc/include/asm/dma-mapping.h b/arch/openrisc/include/asm/dma-mapping.h
new file mode 100644
index 0000000..3d68ed8
--- /dev/null
+++ b/arch/openrisc/include/asm/dma-mapping.h
@@ -0,0 +1,62 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_DMA_MAPPING_H
+#define __ASM_OPENRISC_DMA_MAPPING_H
+
+/*
+ * See Documentation/PCI/PCI-DMA-mapping.txt and
+ * Documentation/DMA-API.txt for documentation.
+ */
+
+#include <linux/dma-debug.h>
+#include <asm-generic/dma-coherent.h>
+
+#define DMA_ERROR_CODE (~(dma_addr_t)0x0)
+
+int dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
+
+int dma_supported(struct device *dev, u64 mask);
+
+int dma_set_mask(struct device *dev, u64 mask);
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+void *or1k_dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flag);
+void or1k_dma_free_coherent(struct device *dev, size_t size, void *vaddr,
+ dma_addr_t dma_handle);
+
+static inline void *dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flag)
+{
+ void *memory;
+
+ memory = or1k_dma_alloc_coherent(dev, size, dma_handle, flag);
+
+ debug_dma_alloc_coherent(dev, size, *dma_handle, memory);
+ return memory;
+}
+
+static inline void dma_free_coherent(struct device *dev, size_t size,
+ void *cpu_addr, dma_addr_t dma_handle)
+{
+ debug_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+ or1k_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+}
+
+#endif /* __ASM_OPENRISC_DMA_MAPPING_H */
diff --git a/arch/openrisc/kernel/dma.c b/arch/openrisc/kernel/dma.c
new file mode 100644
index 0000000..e038732a
--- /dev/null
+++ b/arch/openrisc/kernel/dma.c
@@ -0,0 +1,86 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * DMA mapping callbacks...
+ * As alloc_coherent is the only DMA callback being used currently, that's
+ * the only thing implemented properly. The rest need looking into...
+ */
+
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/gfp.h>
+#include <linux/dma-debug.h>
+#include <linux/io.h>
+#include <linux/vmalloc.h>
+
+/*
+ * Alloc "coherent" memory, which for OpenRISC means simply uncached.
+ */
+void *or1k_dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flag)
+{
+ int order;
+ unsigned long page, va;
+ pgprot_t prot;
+ struct vm_struct *area;
+
+ /* Only allocate page size areas. */
+ size = PAGE_ALIGN(size);
+ order = get_order(size);
+
+ page = __get_free_pages(flag, order);
+ if (!page)
+ return NULL;
+
+ /* Allocate some common virtual space to map the new pages. */
+ area = get_vm_area(size, VM_ALLOC);
+ if (area == NULL) {
+ free_pages(page, order);
+ return NULL;
+ }
+ va = (unsigned long)area->addr;
+
+ /* This gives us the real physical address of the first page. */
+ *dma_handle = __pa(page);
+
+ prot = PAGE_KERNEL_NOCACHE;
+
+ /* This isn't so much ioremap as just simply 'remap' */
+ if (ioremap_page_range(va, va + size, *dma_handle, prot)) {
+ vfree(area->addr);
+ return NULL;
+ }
+
+ return (void *)va;
+}
+
+void or1k_dma_free_coherent(struct device *dev, size_t size, void *vaddr,
+ dma_addr_t dma_handle)
+{
+ vfree(vaddr);
+}
+
+/* Number of entries preallocated for DMA-API debugging */
+#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)
+
+static int __init dma_init(void)
+{
+ dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
+
+ return 0;
+}
+
+fs_initcall(dma_init);
--
1.7.4.1
Arnd Bergmann
2011-07-05 15:37:29 UTC
Permalink
Post by Jonas Bonn
+void *or1k_dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flag)
+{
+ int order;
+ unsigned long page, va;
+ pgprot_t prot;
+ struct vm_struct *area;
+
+ /* Only allocate page size areas. */
+ size = PAGE_ALIGN(size);
+ order = get_order(size);
+
+ page = __get_free_pages(flag, order);
+ if (!page)
+ return NULL;
+
+ /* Allocate some common virtual space to map the new pages. */
+ area = get_vm_area(size, VM_ALLOC);
+ if (area == NULL) {
+ free_pages(page, order);
+ return NULL;
+ }
+ va = (unsigned long)area->addr;
+
+ /* This gives us the real physical address of the first page. */
+ *dma_handle = __pa(page);
+
+ prot = PAGE_KERNEL_NOCACHE;
+
+ /* This isn't so much ioremap as just simply 'remap' */
+ if (ioremap_page_range(va, va + size, *dma_handle, prot)) {
+ vfree(area->addr);
+ return NULL;
+ }
+
+ return (void *)va;
+}
This will result in having conflicting mappings, one with and another
without caching, which a lot of CPU architectures don't like. Are you
sure that you can handle this with or1k?

I think at the very least you will need to flush the cache for
the linear mapping, to avoid writing back dirty cache lines over
the DMA buffer.

You can save a little memory by using alloc_pages_exact instead of
get_free_pages, which always gives you a power-of-two size.

Also, isn't get_vm_area+ioremap_page_range the same as ioremap
on or1k?

In the case that ioremap_page_ranges fails, I think you have a
memory leak, or worse, because areas is not backed by the pages at that
moment.

Arnd
Jonas Bonn
2011-07-08 07:36:19 UTC
Permalink
Post by Arnd Bergmann
This will result in having conflicting mappings, one with and another
without caching, which a lot of CPU architectures don't like. Are you
sure that you can handle this with or1k?
Arnd
Hi Arnd,
I put a bit more thought into what we actually need for DMA and came up
with the implementation in this patch. This is a lot simpler than what
we had before, but it seems to fit nicely with the model that we have.

The jist of it is that allocation of coherent memory just needs to
return pages with the cache-inhibit bit set so we now:

i) call __get_free_pages (via alloc_pages_exact)
ii) set the CI bit on those pages
iii) flush the pages from cache

This should mean that we don't have any conflicting page table entries
for these pages anymore.

I'd appreciate if you could look over this implementation to make sure
I haven't missed anything fundamental.

Thanks,
Jonas

PS: Having a bit of an issue with git send-email and in a bit of a hurry
so I just pasted the patch into this mail... should be ok since it's
just for review anyway; just hoping the mail client doesn't mess up the
formatting.

-----------------------------------------------------------------
Post by Arnd Bergmann
From b2dc6eb85a19b4601180c9dcec440364daf7d9db Mon Sep 17 00:00:00 2001
From: Jonas Bonn <***@southpole.se>
Date: Sat, 4 Jun 2011 21:56:48 +0300
Subject: [PATCH] OpenRISC: DMA

Simple DMA implementation. Allows for allocation of coherent memory
(simply uncached) for DMA operations.

Signed-off-by: Jonas Bonn <***@southpole.se>
---


arch/openrisc/include/asm/dma-mapping.h | 62 +++++++++++++++
arch/openrisc/kernel/dma.c | 131
+++++++++++++++++++++++++++++++
2 files changed, 193 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/include/asm/dma-mapping.h
create mode 100644 arch/openrisc/kernel/dma.c

diff --git a/arch/openrisc/include/asm/dma-mapping.h
b/arch/openrisc/include/asm/dma-mapping.h
new file mode 100644
index 0000000..3d68ed8
--- /dev/null
+++ b/arch/openrisc/include/asm/dma-mapping.h
@@ -0,0 +1,62 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * OpenRISC implementation:
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef __ASM_OPENRISC_DMA_MAPPING_H
+#define __ASM_OPENRISC_DMA_MAPPING_H
+
+/*
+ * See Documentation/PCI/PCI-DMA-mapping.txt and
+ * Documentation/DMA-API.txt for documentation.
+ */
+
+#include <linux/dma-debug.h>
+#include <asm-generic/dma-coherent.h>
+
+#define DMA_ERROR_CODE (~(dma_addr_t)0x0)
+
+int dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
+
+int dma_supported(struct device *dev, u64 mask);
+
+int dma_set_mask(struct device *dev, u64 mask);
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h,
f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+void *or1k_dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flag);
+void or1k_dma_free_coherent(struct device *dev, size_t size, void
*vaddr,
+ dma_addr_t dma_handle);
+
+static inline void *dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flag)
+{
+ void *memory;
+
+ memory = or1k_dma_alloc_coherent(dev, size, dma_handle, flag);
+
+ debug_dma_alloc_coherent(dev, size, *dma_handle, memory);
+ return memory;
+}
+
+static inline void dma_free_coherent(struct device *dev, size_t size,
+ void *cpu_addr, dma_addr_t dma_handle)
+{
+ debug_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+ or1k_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+}
+
+#endif /* __ASM_OPENRISC_DMA_MAPPING_H */
diff --git a/arch/openrisc/kernel/dma.c b/arch/openrisc/kernel/dma.c
new file mode 100644
index 0000000..54d4773
--- /dev/null
+++ b/arch/openrisc/kernel/dma.c
@@ -0,0 +1,131 @@
+/*
+ * OpenRISC Linux
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * DMA mapping callbacks...
+ * As alloc_coherent is the only DMA callback being used currently,
that's
+ * the only thing implemented properly. The rest need looking into...
+ */
+
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/gfp.h>
+#include <linux/dma-debug.h>
+#include <linux/io.h>
+#include <linux/vmalloc.h>
+#include <linux/sched.h>
+
+#include <asm/cpuinfo.h>
+#include <asm/spr_defs.h>
+#include <asm/tlbflush.h>
+
+static int page_set_nocache(pte_t* pte, unsigned long addr,
+ unsigned long next, struct mm_walk* walk)
+{
+ unsigned long cl;
+
+ pte_val(*pte) |= _PAGE_CI;
+
+ /*
+ * Flush the page out of the TLB so that the new page flags get
+ * picked up next time there's an access
+ */
+ flush_tlb_page(NULL, addr);
+
+ /* Flush page out of dcache */
+ for (cl = __pa(addr); cl < __pa(next); cl +=
cpuinfo.dcache_block_size)
+ mtspr(SPR_DCBFR, cl);
+
+ return 0;
+}
+
+static int page_clear_nocache(pte_t* pte, unsigned long addr,
+ unsigned long next, struct mm_walk* walk)
+{
+ pte_val(*pte) &= ~_PAGE_CI;
+
+ /*
+ * Flush the page out of the TLB so that the new page flags get
+ * picked up next time there's an access
+ */
+ flush_tlb_page(NULL, addr);
+
+ return 0;
+}
+
+/*
+ * Alloc "coherent" memory, which for OpenRISC means simply uncached.
+ *
+ * This function effectively just calls __get_free_pages, sets the
+ * cache-inhibit bit on those pages, and makes sure that the pages are
+ * flushed out of the cache before they are used.
+ *
+ */
+void *or1k_dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t gfp)
+{
+ unsigned long va;
+ void* page;
+ struct mm_walk walk = {
+ .pte_entry = page_set_nocache,
+ .mm = &init_mm
+ };
+
+ page = alloc_pages_exact(size, gfp);
+ if (!page)
+ return NULL;
+
+ /* This gives us the real physical address of the first page. */
+ *dma_handle = __pa(page);
+
+ va = (unsigned long) page;
+
+ /*
+ * We need to iterate through the pages, clearing the dcache for
+ * them and setting the cache-inhibit bit.
+ */
+ if (walk_page_range(va, va+size, &walk)) {
+ free_pages_exact(page, size);
+ return NULL;
+ }
+
+ return (void*) va;
+}
+
+void or1k_dma_free_coherent(struct device *dev, size_t size, void
*vaddr,
+ dma_addr_t dma_handle)
+{
+ unsigned long va = (unsigned long) vaddr;
+ struct mm_walk walk = {
+ .pte_entry = page_clear_nocache,
+ .mm = &init_mm
+ };
+
+ /* walk_page_range shouldn't be able to fail here */
+ WARN_ON(walk_page_range(va, va+size, &walk));
+
+ free_pages_exact(vaddr, size);
+}
+
+/* Number of entries preallocated for DMA-API debugging */
+#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)
+
+static int __init dma_init(void)
+{
+ dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
+
+ return 0;
+}
+fs_initcall(dma_init);
--
1.7.1
Arnd Bergmann
2011-07-08 07:44:30 UTC
Permalink
Post by Jonas Bonn
I put a bit more thought into what we actually need for DMA and came up
with the implementation in this patch. This is a lot simpler than what
we had before, but it seems to fit nicely with the model that we have.
The jist of it is that allocation of coherent memory just needs to
i) call __get_free_pages (via alloc_pages_exact)
ii) set the CI bit on those pages
iii) flush the pages from cache
This should mean that we don't have any conflicting page table entries
for these pages anymore.
I'd appreciate if you could look over this implementation to make sure
I haven't missed anything fundamental.
Looks ok to me. A few more points for future optimization though:

* If your CPU can do different page sizes, you should make the kernel
mapping use the largest possible page size. This will however make
the dma mapping part much harder.

* If you ever want to map a frame buffer with this, you should
add a write-combining mapping type, uncached accesses are really
slow.

* Cache coherency is good. If you can make the hardware do cache-coherent
DMA, then just use cacheable pages for alloc_coherent.

Arnd
Jonas Bonn
2011-07-02 21:15:34 UTC
Permalink
Architecture code and early setup routines for booting Linux.

Signed-off-by: Jonas Bonn <***@southpole.se>
---
arch/openrisc/kernel/entry.S | 1134 +++++++++++++++++++++++++++
arch/openrisc/kernel/head.S | 1607 ++++++++++++++++++++++++++++++++++++++
arch/openrisc/kernel/init_task.c | 41 +
arch/openrisc/kernel/setup.c | 387 +++++++++
4 files changed, 3169 insertions(+), 0 deletions(-)
create mode 100644 arch/openrisc/kernel/entry.S
create mode 100644 arch/openrisc/kernel/head.S
create mode 100644 arch/openrisc/kernel/init_task.c
create mode 100644 arch/openrisc/kernel/setup.c

diff --git a/arch/openrisc/kernel/entry.S b/arch/openrisc/kernel/entry.S
new file mode 100644
index 0000000..7a967cb
--- /dev/null
+++ b/arch/openrisc/kernel/entry.S
@@ -0,0 +1,1134 @@
+/*
+ * OpenRISC entry.S
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2005 Gyorgy Jeney <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/linkage.h>
+
+#include <asm/processor.h>
+#include <asm/unistd.h>
+#include <asm/thread_info.h>
+#include <asm/errno.h>
+#include <asm/spr_defs.h>
+#include <asm/page.h>
+#include <asm/mmu.h>
+#include <asm/pgtable.h>
+#include <asm/asm-offsets.h>
+
+#define DISABLE_INTERRUPTS(t1,t2) \
+ l.mfspr t2,r0,SPR_SR ;\
+ l.movhi t1,hi(~(SPR_SR_IEE|SPR_SR_TEE)) ;\
+ l.ori t1,t1,lo(~(SPR_SR_IEE|SPR_SR_TEE)) ;\
+ l.and t2,t2,t1 ;\
+ l.mtspr r0,t2,SPR_SR
+
+#define ENABLE_INTERRUPTS(t1) \
+ l.mfspr t1,r0,SPR_SR ;\
+ l.ori t1,t1,lo(SPR_SR_IEE|SPR_SR_TEE) ;\
+ l.mtspr r0,t1,SPR_SR
+
+/* =========================================================[ macros ]=== */
+
+/*
+ * We need to disable interrupts at beginning of RESTORE_ALL
+ * since interrupt might come in after we've loaded EPC return address
+ * and overwrite EPC with address somewhere in RESTORE_ALL
+ * which is of course wrong!
+ */
+
+#define RESTORE_ALL \
+ DISABLE_INTERRUPTS(r3,r4) ;\
+ l.lwz r3,PT_PC(r1) ;\
+ l.mtspr r0,r3,SPR_EPCR_BASE ;\
+ l.lwz r3,PT_SR(r1) ;\
+ l.mtspr r0,r3,SPR_ESR_BASE ;\
+ l.lwz r2,PT_GPR2(r1) ;\
+ l.lwz r3,PT_GPR3(r1) ;\
+ l.lwz r4,PT_GPR4(r1) ;\
+ l.lwz r5,PT_GPR5(r1) ;\
+ l.lwz r6,PT_GPR6(r1) ;\
+ l.lwz r7,PT_GPR7(r1) ;\
+ l.lwz r8,PT_GPR8(r1) ;\
+ l.lwz r9,PT_GPR9(r1) ;\
+ l.lwz r10,PT_GPR10(r1) ;\
+ l.lwz r11,PT_GPR11(r1) ;\
+ l.lwz r12,PT_GPR12(r1) ;\
+ l.lwz r13,PT_GPR13(r1) ;\
+ l.lwz r14,PT_GPR14(r1) ;\
+ l.lwz r15,PT_GPR15(r1) ;\
+ l.lwz r16,PT_GPR16(r1) ;\
+ l.lwz r17,PT_GPR17(r1) ;\
+ l.lwz r18,PT_GPR18(r1) ;\
+ l.lwz r19,PT_GPR19(r1) ;\
+ l.lwz r20,PT_GPR20(r1) ;\
+ l.lwz r21,PT_GPR21(r1) ;\
+ l.lwz r22,PT_GPR22(r1) ;\
+ l.lwz r23,PT_GPR23(r1) ;\
+ l.lwz r24,PT_GPR24(r1) ;\
+ l.lwz r25,PT_GPR25(r1) ;\
+ l.lwz r26,PT_GPR26(r1) ;\
+ l.lwz r27,PT_GPR27(r1) ;\
+ l.lwz r28,PT_GPR28(r1) ;\
+ l.lwz r29,PT_GPR29(r1) ;\
+ l.lwz r30,PT_GPR30(r1) ;\
+ l.lwz r31,PT_GPR31(r1) ;\
+ l.lwz r1,PT_SP(r1) ;\
+ l.rfe
+
+
+#define EXCEPTION_ENTRY(handler) \
+ .global handler ;\
+handler: ;\
+ /* r1, EPCR, ESR a already saved */ ;\
+ l.sw PT_GPR2(r1),r2 ;\
+ l.sw PT_GPR3(r1),r3 ;\
+ l.sw PT_ORIG_GPR11(r1),r11 ;\
+ /* r4 already save */ ;\
+ l.sw PT_GPR5(r1),r5 ;\
+ l.sw PT_GPR6(r1),r6 ;\
+ l.sw PT_GPR7(r1),r7 ;\
+ l.sw PT_GPR8(r1),r8 ;\
+ l.sw PT_GPR9(r1),r9 ;\
+ /* r10 already saved */ ;\
+ l.sw PT_GPR11(r1),r11 ;\
+ /* r12 already saved */ ;\
+ l.sw PT_GPR13(r1),r13 ;\
+ l.sw PT_GPR14(r1),r14 ;\
+ l.sw PT_GPR15(r1),r15 ;\
+ l.sw PT_GPR16(r1),r16 ;\
+ l.sw PT_GPR17(r1),r17 ;\
+ l.sw PT_GPR18(r1),r18 ;\
+ l.sw PT_GPR19(r1),r19 ;\
+ l.sw PT_GPR20(r1),r20 ;\
+ l.sw PT_GPR21(r1),r21 ;\
+ l.sw PT_GPR22(r1),r22 ;\
+ l.sw PT_GPR23(r1),r23 ;\
+ l.sw PT_GPR24(r1),r24 ;\
+ l.sw PT_GPR25(r1),r25 ;\
+ l.sw PT_GPR26(r1),r26 ;\
+ l.sw PT_GPR27(r1),r27 ;\
+ l.sw PT_GPR28(r1),r28 ;\
+ l.sw PT_GPR29(r1),r29 ;\
+ /* r30 already save */ ;\
+/* l.sw PT_GPR30(r1),r30*/ ;\
+ l.sw PT_GPR31(r1),r31 ;\
+ l.sw PT_SYSCALLNO(r1),r0
+
+#define UNHANDLED_EXCEPTION(handler,vector) \
+ .global handler ;\
+handler: ;\
+ /* r1, EPCR, ESR already saved */ ;\
+ l.sw PT_GPR2(r1),r2 ;\
+ l.sw PT_GPR3(r1),r3 ;\
+ l.sw PT_ORIG_GPR11(r1),r11 ;\
+ l.sw PT_GPR5(r1),r5 ;\
+ l.sw PT_GPR6(r1),r6 ;\
+ l.sw PT_GPR7(r1),r7 ;\
+ l.sw PT_GPR8(r1),r8 ;\
+ l.sw PT_GPR9(r1),r9 ;\
+ /* r10 already saved */ ;\
+ l.sw PT_GPR11(r1),r11 ;\
+ /* r12 already saved */ ;\
+ l.sw PT_GPR13(r1),r13 ;\
+ l.sw PT_GPR14(r1),r14 ;\
+ l.sw PT_GPR15(r1),r15 ;\
+ l.sw PT_GPR16(r1),r16 ;\
+ l.sw PT_GPR17(r1),r17 ;\
+ l.sw PT_GPR18(r1),r18 ;\
+ l.sw PT_GPR19(r1),r19 ;\
+ l.sw PT_GPR20(r1),r20 ;\
+ l.sw PT_GPR21(r1),r21 ;\
+ l.sw PT_GPR22(r1),r22 ;\
+ l.sw PT_GPR23(r1),r23 ;\
+ l.sw PT_GPR24(r1),r24 ;\
+ l.sw PT_GPR25(r1),r25 ;\
+ l.sw PT_GPR26(r1),r26 ;\
+ l.sw PT_GPR27(r1),r27 ;\
+ l.sw PT_GPR28(r1),r28 ;\
+ l.sw PT_GPR29(r1),r29 ;\
+ /* r31 already saved */ ;\
+ l.sw PT_GPR30(r1),r30 ;\
+/* l.sw PT_GPR31(r1),r31 */ ;\
+ l.sw PT_SYSCALLNO(r1),r0 ;\
+ l.addi r3,r1,0 ;\
+ /* r4 is exception EA */ ;\
+ l.addi r5,r0,vector ;\
+ l.jal unhandled_exception ;\
+ l.nop ;\
+ l.j _ret_from_exception ;\
+ l.nop
+
+/*
+ * NOTE: one should never assume that SPR_EPC, SPR_ESR, SPR_EEAR
+ * contain the same values as when exception we're handling
+ * occured. in fact they never do. if you need them use
+ * values saved on stack (for SPR_EPC, SPR_ESR) or content
+ * of r4 (for SPR_EEAR). for details look at EXCEPTION_HANDLE()
+ * in 'arch/or32/kernel/head.S'
+ */
+
+/* =====================================================[ exceptions] === */
+
+/* ---[ 0x100: RESET exception ]----------------------------------------- */
+
+EXCEPTION_ENTRY(_tng_kernel_start)
+ l.jal _start
+ l.andi r0,r0,0
+
+/* ---[ 0x200: BUS exception ]------------------------------------------- */
+
+EXCEPTION_ENTRY(_bus_fault_handler)
+ /* r4: EA of fault (set by EXCEPTION_HANDLE) */
+ l.jal do_bus_fault
+ l.addi r3,r1,0 /* pt_regs */
+
+ l.j _ret_from_exception
+ l.nop
+
+/* ---[ 0x300: Data Page Fault exception ]------------------------------- */
+
+EXCEPTION_ENTRY(_data_page_fault_handler)
+ /* set up parameters for do_page_fault */
+ l.addi r3,r1,0 // pt_regs
+ /* r4 set be EXCEPTION_HANDLE */ // effective address of fault
+ l.ori r5,r0,0x300 // exception vector
+
+ /*
+ * __PHX__: TODO
+ *
+ * all this can be written much simpler. look at
+ * DTLB miss handler in the CONFIG_GUARD_PROTECTED_CORE part
+ */
+#ifdef CONFIG_OPENRISC_NO_SPR_SR_DSX
+ l.lwz r6,PT_PC(r3) // address of an offending insn
+ l.lwz r6,0(r6) // instruction that caused pf
+
+ l.srli r6,r6,26 // check opcode for jump insn
+ l.sfeqi r6,0 // l.j
+ l.bf 8f
+ l.sfeqi r6,1 // l.jal
+ l.bf 8f
+ l.sfeqi r6,3 // l.bnf
+ l.bf 8f
+ l.sfeqi r6,4 // l.bf
+ l.bf 8f
+ l.sfeqi r6,0x11 // l.jr
+ l.bf 8f
+ l.sfeqi r6,0x12 // l.jalr
+ l.bf 8f
+
+ l.nop
+
+ l.j 9f
+ l.nop
+8:
+
+ l.lwz r6,PT_PC(r3) // address of an offending insn
+ l.addi r6,r6,4
+ l.lwz r6,0(r6) // instruction that caused pf
+ l.srli r6,r6,26 // get opcode
+9:
+
+#else
+
+ l.mfspr r6,r0,SPR_SR // SR
+// l.lwz r6,PT_SR(r3) // ESR
+ l.andi r6,r6,SPR_SR_DSX // check for delay slot exception
+ l.sfeqi r6,0x1 // exception happened in delay slot
+ l.bnf 7f
+ l.lwz r6,PT_PC(r3) // address of an offending insn
+
+ l.addi r6,r6,4 // offending insn is in delay slot
+7:
+ l.lwz r6,0(r6) // instruction that caused pf
+ l.srli r6,r6,26 // check opcode for write access
+#endif
+
+ l.sfgeui r6,0x34 // check opcode for write access
+ l.bnf 1f
+ l.sfleui r6,0x37
+ l.bnf 1f
+ l.ori r6,r0,0x1 // write access
+ l.j 2f
+ l.nop
+1: l.ori r6,r0,0x0 // !write access
+2:
+
+ /* call fault.c handler in or32/mm/fault.c */
+ l.jal do_page_fault
+ l.nop
+ l.j _ret_from_exception
+ l.nop
+
+/* ---[ 0x400: Insn Page Fault exception ]------------------------------- */
+
+EXCEPTION_ENTRY(_insn_page_fault_handler)
+ /* set up parameters for do_page_fault */
+ l.addi r3,r1,0 // pt_regs
+ /* r4 set be EXCEPTION_HANDLE */ // effective address of fault
+ l.ori r5,r0,0x400 // exception vector
+ l.ori r6,r0,0x0 // !write access
+
+ /* call fault.c handler in or32/mm/fault.c */
+ l.jal do_page_fault
+ l.nop
+ l.j _ret_from_exception
+ l.nop
+
+
+/* ---[ 0x500: Timer exception ]----------------------------------------- */
+
+EXCEPTION_ENTRY(_timer_handler)
+ l.jal timer_interrupt
+ l.addi r3,r1,0 /* pt_regs */
+
+ l.j _ret_from_intr
+ l.nop
+
+/* ---[ 0x600: Aligment exception ]-------------------------------------- */
+
+EXCEPTION_ENTRY(_alignment_handler)
+ /* r4: EA of fault (set by EXCEPTION_HANDLE) */
+ l.jal do_unaligned_access
+ l.addi r3,r1,0 /* pt_regs */
+
+ l.j _ret_from_exception
+ l.nop
+
+#if 0
+EXCEPTION_ENTRY(_aligment_handler)
+// l.mfspr r2,r0,SPR_EEAR_BASE /* Load the efective addres */
+ l.addi r2,r4,0
+// l.mfspr r5,r0,SPR_EPCR_BASE /* Load the insn address */
+ l.lwz r5,PT_PC(r1)
+
+ l.lwz r3,0(r5) /* Load insn */
+ l.srli r4,r3,26 /* Shift left to get the insn opcode */
+
+ l.sfeqi r4,0x00 /* Check if the load/store insn is in delay slot */
+ l.bf jmp
+ l.sfeqi r4,0x01
+ l.bf jmp
+ l.sfeqi r4,0x03
+ l.bf jmp
+ l.sfeqi r4,0x04
+ l.bf jmp
+ l.sfeqi r4,0x11
+ l.bf jr
+ l.sfeqi r4,0x12
+ l.bf jr
+ l.nop
+ l.j 1f
+ l.addi r5,r5,4 /* Increment PC to get return insn address */
+
+jmp:
+ l.slli r4,r3,6 /* Get the signed extended jump length */
+ l.srai r4,r4,4
+
+ l.lwz r3,4(r5) /* Load the real load/store insn */
+
+ l.add r5,r5,r4 /* Calculate jump target address */
+
+ l.j 1f
+ l.srli r4,r3,26 /* Shift left to get the insn opcode */
+
+jr:
+ l.slli r4,r3,9 /* Shift to get the reg nb */
+ l.andi r4,r4,0x7c
+
+ l.lwz r3,4(r5) /* Load the real load/store insn */
+
+ l.add r4,r4,r1 /* Load the jump register value from the stack */
+ l.lwz r5,0(r4)
+
+ l.srli r4,r3,26 /* Shift left to get the insn opcode */
+
+
+1:
+// l.mtspr r0,r5,SPR_EPCR_BASE
+ l.sw PT_PC(r1),r5
+
+ l.sfeqi r4,0x26
+ l.bf lhs
+ l.sfeqi r4,0x25
+ l.bf lhz
+ l.sfeqi r4,0x22
+ l.bf lws
+ l.sfeqi r4,0x21
+ l.bf lwz
+ l.sfeqi r4,0x37
+ l.bf sh
+ l.sfeqi r4,0x35
+ l.bf sw
+ l.nop
+
+1: l.j 1b /* I don't know what to do */
+ l.nop
+
+lhs: l.lbs r5,0(r2)
+ l.slli r5,r5,8
+ l.lbz r6,1(r2)
+ l.or r5,r5,r6
+ l.srli r4,r3,19
+ l.andi r4,r4,0x7c
+ l.add r4,r4,r1
+ l.j align_end
+ l.sw 0(r4),r5
+
+lhz: l.lbz r5,0(r2)
+ l.slli r5,r5,8
+ l.lbz r6,1(r2)
+ l.or r5,r5,r6
+ l.srli r4,r3,19
+ l.andi r4,r4,0x7c
+ l.add r4,r4,r1
+ l.j align_end
+ l.sw 0(r4),r5
+
+lws: l.lbs r5,0(r2)
+ l.slli r5,r5,24
+ l.lbz r6,1(r2)
+ l.slli r6,r6,16
+ l.or r5,r5,r6
+ l.lbz r6,2(r2)
+ l.slli r6,r6,8
+ l.or r5,r5,r6
+ l.lbz r6,3(r2)
+ l.or r5,r5,r6
+ l.srli r4,r3,19
+ l.andi r4,r4,0x7c
+ l.add r4,r4,r1
+ l.j align_end
+ l.sw 0(r4),r5
+
+lwz: l.lbz r5,0(r2)
+ l.slli r5,r5,24
+ l.lbz r6,1(r2)
+ l.slli r6,r6,16
+ l.or r5,r5,r6
+ l.lbz r6,2(r2)
+ l.slli r6,r6,8
+ l.or r5,r5,r6
+ l.lbz r6,3(r2)
+ l.or r5,r5,r6
+ l.srli r4,r3,19
+ l.andi r4,r4,0x7c
+ l.add r4,r4,r1
+ l.j align_end
+ l.sw 0(r4),r5
+
+sh:
+ l.srli r4,r3,9
+ l.andi r4,r4,0x7c
+ l.add r4,r4,r1
+ l.lwz r5,0(r4)
+ l.sb 1(r2),r5
+ l.srli r5,r5,8
+ l.j align_end
+ l.sb 0(r2),r5
+
+sw:
+ l.srli r4,r3,9
+ l.andi r4,r4,0x7c
+ l.add r4,r4,r1
+ l.lwz r5,0(r4)
+ l.sb 3(r2),r5
+ l.srli r5,r5,8
+ l.sb 2(r2),r5
+ l.srli r5,r5,8
+ l.sb 1(r2),r5
+ l.srli r5,r5,8
+ l.j align_end
+ l.sb 0(r2),r5
+
+align_end:
+ l.j _ret_from_intr
+ l.nop
+#endif
+
+/* ---[ 0x700: Illegal insn exception ]---------------------------------- */
+
+EXCEPTION_ENTRY(_illegal_instruction_handler)
+ /* r4: EA of fault (set by EXCEPTION_HANDLE) */
+ l.jal do_illegal_instruction
+ l.addi r3,r1,0 /* pt_regs */
+
+ l.j _ret_from_exception
+ l.nop
+
+/* ---[ 0x800: External interrupt exception ]---------------------------- */
+
+EXCEPTION_ENTRY(_external_irq_handler)
+#ifdef CONFIG_OPENRISC_ESR_EXCEPTION_BUG_CHECK
+ l.lwz r4,PT_SR(r1) // were interrupts enabled ?
+ l.andi r4,r4,SPR_SR_IEE
+ l.sfeqi r4,0
+ l.bnf 1f // ext irq enabled, all ok.
+ l.nop
+
+ l.addi r1,r1,-0x8
+ l.movhi r3,hi(42f)
+ l.ori r3,r3,lo(42f)
+ l.sw 0x0(r1),r3
+ l.jal printk
+ l.sw 0x4(r1),r4
+ l.addi r1,r1,0x8
+
+ .section .rodata, "a"
+42:
+ .string "\n\rESR interrupt bug: in _external_irq_handler (ESR %x)\n\r"
+ .align 4
+ .previous
+
+ l.ori r4,r4,SPR_SR_IEE // fix the bug
+// l.sw PT_SR(r1),r4
+1:
+#endif
+ l.addi r3,r1,0
+ l.movhi r8,hi(do_IRQ)
+ l.ori r8,r8,lo(do_IRQ)
+ l.jalr r8
+ l.nop
+ l.j _ret_from_intr
+ l.nop
+
+/* ---[ 0x900: DTLB miss exception ]------------------------------------- */
+
+
+/* ---[ 0xa00: ITLB miss exception ]------------------------------------- */
+
+
+/* ---[ 0xb00: Range exception ]----------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0xb00,0xb00)
+
+/* ---[ 0xc00: Syscall exception ]--------------------------------------- */
+
+/*
+ * Syscalls are a special type of exception in that they are
+ * _explicitly_ invoked by userspace and can therefore be
+ * held to conform to the same ABI as normal functions with
+ * respect to whether registers are preserved across the call
+ * or not.
+ */
+
+/* Upon syscall entry we just save the callee-saved registers
+ * and not the call-clobbered ones.
+ */
+
+_string_syscall_return:
+ .string "syscall return %ld \n\r\0"
+ .align 4
+
+ENTRY(_sys_call_handler)
+ /* syscalls run with interrupts enabled */
+ ENABLE_INTERRUPTS(r29) // enable interrupts, r29 is temp
+
+ /* r1, EPCR, ESR a already saved */
+ l.sw PT_GPR2(r1),r2
+ /* r3-r8 must be saved because syscall restart relies
+ * on us being able to restart the syscall args... technically
+ * they should be clobbered, otherwise
+ */
+ l.sw PT_GPR3(r1),r3
+ /* r4 already saved */
+ /* r4 holds the EEAR address of the fault, load the original r4 */
+ l.lwz r4,PT_GPR4(r1)
+ l.sw PT_GPR5(r1),r5
+ l.sw PT_GPR6(r1),r6
+ l.sw PT_GPR7(r1),r7
+ l.sw PT_GPR8(r1),r8
+ l.sw PT_GPR9(r1),r9
+ /* r10 already saved */
+ l.sw PT_GPR11(r1),r11
+ l.sw PT_ORIG_GPR11(r1),r11
+ /* r12,r13 already saved */
+
+ /* r14-r28 (even) aren't touched by the syscall fast path below
+ * so we don't need to save them. However, the functions that return
+ * to userspace via a call to switch() DO need to save these because
+ * switch() effectively clobbers them... saving these registers for
+ * such functions is handled in their syscall wrappers (see fork, vfork,
+ * and clone, below).
+
+ /* r30 is the only register we clobber in the fast path */
+ /* r30 already saved */
+/* l.sw PT_GPR30(r1),r30 */
+ /* This is used by do_signal to determine whether to check for
+ * syscall restart or not */
+ l.sw PT_SYSCALLNO(r1),r11
+
+_syscall_check_trace_enter:
+ /* If TIF_SYSCALL_TRACE is set, then we want to do syscall tracing */
+ l.lwz r30,TI_FLAGS(r10)
+ l.andi r30,r30,_TIF_SYSCALL_TRACE
+ l.sfne r30,r0
+ l.bf _syscall_trace_enter
+ l.nop
+
+_syscall_check:
+ /* Ensure that the syscall number is reasonable */
+ l.sfgeui r11,__NR_syscalls
+ l.bf _syscall_badsys
+ l.nop
+
+_syscall_call:
+ l.movhi r29,hi(sys_call_table)
+ l.ori r29,r29,lo(sys_call_table)
+ l.slli r11,r11,2
+ l.add r29,r29,r11
+ l.lwz r29,0(r29)
+
+ l.jalr r29
+ l.nop
+
+_syscall_return:
+ /* All syscalls return here... just pay attention to ret_from_fork
+ * which does it in a round-about way.
+ */
+ l.sw PT_GPR11(r1),r11 // save return value
+
+#if 0
+_syscall_debug:
+ l.movhi r3,hi(_string_syscall_return)
+ l.ori r3,r3,lo(_string_syscall_return)
+ l.ori r27,r0,1
+ l.sw -4(r1),r27
+ l.sw -8(r1),r11
+ l.addi r1,r1,-8
+ l.movhi r27,hi(printk)
+ l.ori r27,r27,lo(printk)
+ l.jalr r27
+ l.nop
+ l.addi r1,r1,8
+#endif
+
+_syscall_check_trace_leave:
+ /* r30 is a callee-saved register so this should still hold the
+ * _TIF_SYSCALL_TRACE flag from _syscall_check_trace_enter above...
+ * _syscall_trace_leave expects syscall result to be in pt_regs->r11.
+ */
+ l.sfne r30,r0
+ l.bf _syscall_trace_leave
+ l.nop
+
+/* This is where the exception-return code begins... interrupts need to be
+ * disabled the rest of the way here because we can't afford to miss any
+ * interrupts that set NEED_RESCHED or SIGNALPENDING... really true? */
+
+_syscall_check_work:
+ /* Here we need to disable interrupts */
+ DISABLE_INTERRUPTS(r27,r29)
+ l.lwz r30,TI_FLAGS(r10)
+ l.andi r30,r30,_TIF_WORK_MASK
+ l.sfne r30,r0
+
+ l.bnf _syscall_resume_userspace
+ l.nop
+
+ /* Work pending follows a different return path, so we need to
+ * make sure that all the call-saved registers get into pt_regs
+ * before branching...
+ */
+ l.sw PT_GPR14(r1),r14
+ l.sw PT_GPR16(r1),r16
+ l.sw PT_GPR18(r1),r18
+ l.sw PT_GPR20(r1),r20
+ l.sw PT_GPR22(r1),r22
+ l.sw PT_GPR24(r1),r24
+ l.sw PT_GPR26(r1),r26
+ l.sw PT_GPR28(r1),r28
+
+ /* _work_pending needs to be called with interrupts disabled */
+ l.j _work_pending
+ l.nop
+
+_syscall_resume_userspace:
+// ENABLE_INTERRUPTS(r29)
+
+
+/* This is the hot path for returning to userspace from a syscall. If there's
+ * work to be done and the branch to _work_pending was taken above, then the
+ * return to userspace will be done via the normal exception return path...
+ * that path restores _all_ registers and will overwrite the "clobbered"
+ * registers with whatever garbage is in pt_regs -- that's OK because those
+ * registers are clobbered anyway and because the extra work is insignificant
+ * in the context of the extra work that _work_pending is doing.
+
+/* Once again, syscalls are special and only guarantee to preserve the
+ * same registers as a normal function call */
+
+/* The assumption here is that the registers r14-r28 (even) are untouched and
+ * don't need to be restored... be sure that that's really the case!
+ */
+
+/* This is still too much... we should only be restoring what we actually
+ * clobbered... we should even be using 'scratch' (odd) regs above so that
+ * we don't need to restore anything, hardly...
+ */
+
+ l.lwz r2,PT_GPR2(r1)
+
+ /* Restore args */
+ /* r3-r8 are technically clobbered, but syscall restart needs these
+ * to be restored...
+ */
+ l.lwz r3,PT_GPR3(r1)
+ l.lwz r4,PT_GPR4(r1)
+ l.lwz r5,PT_GPR5(r1)
+ l.lwz r6,PT_GPR6(r1)
+ l.lwz r7,PT_GPR7(r1)
+ l.lwz r8,PT_GPR8(r1)
+
+ l.lwz r9,PT_GPR9(r1)
+ l.lwz r10,PT_GPR10(r1)
+ l.lwz r11,PT_GPR11(r1)
+
+ /* r30 is the only register we clobber in the fast path */
+ l.lwz r30,PT_GPR30(r1)
+
+ /* Here we use r13-r19 (odd) as scratch regs */
+ l.lwz r13,PT_PC(r1)
+ l.lwz r15,PT_SR(r1)
+ l.lwz r1,PT_SP(r1)
+ /* Interrupts need to be disabled for setting EPCR and ESR
+ * so that another interrupt doesn't come in here and clobber
+ * them before we can use them for our l.rfe */
+ DISABLE_INTERRUPTS(r17,r19)
+ l.mtspr r0,r13,SPR_EPCR_BASE
+ l.mtspr r0,r15,SPR_ESR_BASE
+ l.rfe
+
+/* End of hot path!
+ * Keep the below tracing and error handling out of the hot path...
+*/
+
+_syscall_trace_enter:
+ /* Here we pass pt_regs to do_syscall_trace_enter. Make sure
+ * that function is really getting all the info it needs as
+ * pt_regs isn't a complete set of userspace regs, just the
+ * ones relevant to the syscall...
+ *
+ * Note use of delay slot for setting argument.
+ */
+ l.jal do_syscall_trace_enter
+ l.addi r3,r1,0
+
+ /* Restore arguments (not preserved across do_syscall_trace_enter)
+ * so that we can do the syscall for real and return to the syscall
+ * hot path.
+ */
+ l.lwz r11,PT_SYSCALLNO(r1)
+ l.lwz r3,PT_GPR3(r1)
+ l.lwz r4,PT_GPR4(r1)
+ l.lwz r5,PT_GPR5(r1)
+ l.lwz r6,PT_GPR6(r1)
+ l.lwz r7,PT_GPR7(r1)
+
+ l.j _syscall_check
+ l.lwz r8,PT_GPR8(r1)
+
+_syscall_trace_leave:
+ l.jal do_syscall_trace_leave
+ l.addi r3,r1,0
+
+ l.j _syscall_check_work
+ l.nop
+
+_syscall_badsys:
+ /* Here we effectively pretend to have executed an imaginary
+ * syscall that returns -ENOSYS and then return to the regular
+ * syscall hot path.
+ * Note that "return value" is set in the delay slot...
+ */
+ l.j _syscall_return
+ l.addi r11,r0,-ENOSYS
+
+/******* END SYSCALL HANDLING *******/
+
+/* ---[ 0xd00: Trap exception ]------------------------------------------ */
+
+UNHANDLED_EXCEPTION(_vector_0xd00,0xd00)
+
+/* ---[ 0xe00: Trap exception ]------------------------------------------ */
+
+EXCEPTION_ENTRY(_trap_handler)
+ /* r4: EA of fault (set by EXCEPTION_HANDLE) */
+ l.jal do_trap
+ l.addi r3,r1,0 /* pt_regs */
+
+ l.j _ret_from_exception
+ l.nop
+
+/* ---[ 0xf00: Reserved exception ]-------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0xf00,0xf00)
+
+/* ---[ 0x1000: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1000,0x1000)
+
+/* ---[ 0x1100: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1100,0x1100)
+
+/* ---[ 0x1200: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1200,0x1200)
+
+/* ---[ 0x1300: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1300,0x1300)
+
+/* ---[ 0x1400: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1400,0x1400)
+
+/* ---[ 0x1500: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1500,0x1500)
+
+/* ---[ 0x1600: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1600,0x1600)
+
+/* ---[ 0x1700: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1700,0x1700)
+
+/* ---[ 0x1800: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1800,0x1800)
+
+/* ---[ 0x1900: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1900,0x1900)
+
+/* ---[ 0x1a00: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1a00,0x1a00)
+
+/* ---[ 0x1b00: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1b00,0x1b00)
+
+/* ---[ 0x1c00: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1c00,0x1c00)
+
+/* ---[ 0x1d00: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1d00,0x1d00)
+
+/* ---[ 0x1e00: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1e00,0x1e00)
+
+/* ---[ 0x1f00: Reserved exception ]------------------------------------- */
+
+UNHANDLED_EXCEPTION(_vector_0x1f00,0x1f00)
+
+/* ========================================================[ return ] === */
+
+_work_pending:
+ /*
+ * if (current_thread_info->flags & _TIF_NEED_RESCHED)
+ * schedule();
+ */
+ l.lwz r5,TI_FLAGS(r10)
+ l.andi r3,r5,_TIF_NEED_RESCHED
+ l.sfnei r3,0
+ l.bnf _work_notifysig
+ l.nop
+ l.jal schedule
+ l.nop
+ l.j _resume_userspace
+ l.nop
+
+/* Handle pending signals and notify-resume requests.
+ * do_notify_resume must be passed the latest pushed pt_regs, not
+ * necessarily the "userspace" ones. Also, pt_regs->syscallno
+ * must be set so that the syscall restart functionality works.
+ */
+_work_notifysig:
+ l.jal do_notify_resume
+ l.ori r3,r1,0 /* pt_regs */
+
+_resume_userspace:
+ DISABLE_INTERRUPTS(r3,r4)
+ l.lwz r3,TI_FLAGS(r10)
+ l.andi r3,r3,_TIF_WORK_MASK
+ l.sfnei r3,0
+ l.bf _work_pending
+ l.nop
+
+_restore_all:
+ RESTORE_ALL
+ /* This returns to userspace code */
+
+
+ENTRY(_ret_from_intr)
+ENTRY(_ret_from_exception)
+ l.lwz r4,PT_SR(r1)
+ l.andi r3,r4,SPR_SR_SM
+ l.sfeqi r3,0
+ l.bnf _restore_all
+ l.nop
+ l.j _resume_userspace
+ l.nop
+
+ENTRY(ret_from_fork)
+ l.jal schedule_tail
+ l.nop
+
+ /* _syscall_returns expect r11 to contain return value */
+ l.lwz r11,PT_GPR11(r1)
+
+ /* The syscall fast path return expects call-saved registers
+ * r12-r28 to be untouched, so we restore them here as they
+ * will have been effectively clobbered when arriving here
+ * via the call to switch()
+ */
+ l.lwz r12,PT_GPR12(r1)
+ l.lwz r14,PT_GPR14(r1)
+ l.lwz r16,PT_GPR16(r1)
+ l.lwz r18,PT_GPR18(r1)
+ l.lwz r20,PT_GPR20(r1)
+ l.lwz r22,PT_GPR22(r1)
+ l.lwz r24,PT_GPR24(r1)
+ l.lwz r26,PT_GPR26(r1)
+ l.lwz r28,PT_GPR28(r1)
+
+ l.j _syscall_return
+ l.nop
+
+/* Since syscalls don't save call-clobbered registers, the args to
+ * kernel_thread_helper will need to be passed through callee-saved
+ * registers and copied to the parameter registers when the thread
+ * begins running.
+ *
+ * See arch/openrisc/kernel/process.c:
+ * The args are passed as follows:
+ * arg1 (r3) : passed in r20
+ * arg2 (r4) : passed in r22
+ */
+
+ENTRY(_kernel_thread_helper)
+ l.or r3,r20,r0
+ l.or r4,r22,r0
+ l.movhi r31,hi(kernel_thread_helper)
+ l.ori r31,r31,lo(kernel_thread_helper)
+ l.jr r31
+ l.nop
+
+
+/* ========================================================[ switch ] === */
+
+/*
+ * This routine switches between two different tasks. The process
+ * state of one is saved on its kernel stack. Then the state
+ * of the other is restored from its kernel stack. The memory
+ * management hardware is updated to the second process's state.
+ * Finally, we can return to the second process, via the 'return'.
+ *
+ * Note: there are two ways to get to the "going out" portion
+ * of this code; either by coming in via the entry (_switch)
+ * or via "fork" which must set up an environment equivalent
+ * to the "_switch" path. If you change this (or in particular, the
+ * SAVE_REGS macro), you'll have to change the fork code also.
+ */
+
+
+/* _switch MUST never lay on page boundry, cause it runs from
+ * effective addresses and beeing interrupted by iTLB miss would kill it.
+ * dTLB miss seams to never accour in the bad place since data accesses
+ * are from task structures which are always page aligned.
+ *
+ * The problem happens in RESTORE_ALL_NO_R11 where we first set the EPCR
+ * register, then load the previous register values and only at the end call
+ * the l.rfe instruction. If get TLB miss in beetwen the EPCR register gets
+ * garbled and we end up calling l.rfe with the wrong EPCR. (same probably
+ * holds for ESR)
+ *
+ * To avoid this problems it is sufficient to align _switch to
+ * some nice round number smaller than it's size...
+ */
+
+/* ABI rules apply here... we either enter _switch via schedule() or via
+ * an imaginary call to which we shall return at return_from_fork. Either
+ * way, we are a function call and only need to preserve the callee-saved
+ * registers when we return. As such, we don't need to save the registers
+ * on the stack that we won't be returning as they were...
+ */
+
+ .align 0x400
+ENTRY(_switch)
+ /* We don't store SR as _switch only gets called in a context where
+ * the SR will be the same going in and coming out... */
+
+ /* Set up new pt_regs struct for saving task state */
+ l.addi r1,r1,-(INT_FRAME_SIZE)
+
+ /* No need to store r1/PT_SP as it goes into KSP below */
+ l.sw PT_GPR2(r1),r2
+ l.sw PT_GPR9(r1),r9
+ /* This is wrong, r12 shouldn't be here... but GCC is broken for the time being
+ * and expects r12 to be callee-saved... */
+ l.sw PT_GPR12(r1),r12
+ l.sw PT_GPR14(r1),r14
+ l.sw PT_GPR16(r1),r16
+ l.sw PT_GPR18(r1),r18
+ l.sw PT_GPR20(r1),r20
+ l.sw PT_GPR22(r1),r22
+ l.sw PT_GPR24(r1),r24
+ l.sw PT_GPR26(r1),r26
+ l.sw PT_GPR28(r1),r28
+ l.sw PT_GPR30(r1),r30
+
+ l.addi r11,r10,0 /* Save old 'current' to 'last' return value*/
+
+ /* We use thread_info->ksp for storing the address of the above
+ * structure so that we can get back to it later... we don't want
+ * to lose the value of thread_info->ksp, though, so store it as
+ * pt_regs->sp so that we can easily restore it when we are made
+ * live again...
+ */
+
+ /* Save the old value of thread_info->ksp as pt_regs->sp */
+ l.lwz r29,TI_KSP(r10)
+ l.sw PT_SP(r1),r29
+
+ /* Swap kernel stack pointers */
+ l.sw TI_KSP(r10),r1 /* Save old stack pointer */
+ l.or r10,r4,r0 /* Set up new current_thread_info */
+ l.lwz r1,TI_KSP(r10) /* Load new stack pointer */
+
+ /* Restore the old value of thread_info->ksp */
+ l.lwz r29,PT_SP(r1)
+ l.sw TI_KSP(r10),r29
+
+ /* ...and restore the registers, except r11 because the return value
+ * has already been set above.
+ */
+ l.lwz r2,PT_GPR2(r1)
+ l.lwz r9,PT_GPR9(r1)
+ /* No need to restore r10 */
+ /* ...and do not restore r11 */
+
+ /* This is wrong, r12 shouldn't be here... but GCC is broken for the time being
+ * and expects r12 to be callee-saved... */
+ l.lwz r12,PT_GPR12(r1)
+ l.lwz r14,PT_GPR14(r1)
+ l.lwz r16,PT_GPR16(r1)
+ l.lwz r18,PT_GPR18(r1)
+ l.lwz r20,PT_GPR20(r1)
+ l.lwz r22,PT_GPR22(r1)
+ l.lwz r24,PT_GPR24(r1)
+ l.lwz r26,PT_GPR26(r1)
+ l.lwz r28,PT_GPR28(r1)
+ l.lwz r30,PT_GPR30(r1)
+
+ /* Unwind stack to pre-switch state */
+ l.addi r1,r1,(INT_FRAME_SIZE)
+
+ /* Return via the link-register back to where we 'came from', where that can be
+ * either schedule() or return_from_fork()... */
+ l.jr r9
+ l.nop
+
+/* ==================================================================== */
+
+/* These all use the delay slot for setting the argument register, so the
+ * jump is always happening after the l.addi instruction.
+ *
+ * These are all just wrappers that don't touch the link-register r9, so the
+ * return from the "real" syscall function will return back to the syscall
+ * code that did the l.jal that brought us here.
+ */
+
+/* fork requires that we save all the callee-saved registers because they
+ * are all effectively clobbered by the call to _switch. Here we store
+ * all the registers that aren't touched by the syscall fast path and thus
+ * weren't saved there.
+ */
+
+_fork_save_extra_regs_and_call:
+ l.sw PT_GPR14(r1),r14
+ l.sw PT_GPR16(r1),r16
+ l.sw PT_GPR18(r1),r18
+ l.sw PT_GPR20(r1),r20
+ l.sw PT_GPR22(r1),r22
+ l.sw PT_GPR24(r1),r24
+ l.sw PT_GPR26(r1),r26
+ l.jr r29
+ l.sw PT_GPR28(r1),r28
+
+ENTRY(sys_clone)
+ l.movhi r29,hi(_sys_clone)
+ l.ori r29,r29,lo(_sys_clone)
+ l.j _fork_save_extra_regs_and_call
+ l.addi r7,r1,0
+
+ENTRY(sys_fork)
+ l.movhi r29,hi(_sys_fork)
+ l.ori r29,r29,lo(_sys_fork)
+ l.j _fork_save_extra_regs_and_call
+ l.addi r3,r1,0
+
+ENTRY(sys_vfork)
+ l.movhi r29,hi(_sys_vfork)
+ l.ori r29,r29,lo(_sys_vfork)
+ l.j _fork_save_extra_regs_and_call
+ l.addi r3,r1,0
+
+ENTRY(sys_execve)
+ l.j _sys_execve
+ l.addi r6,r1,0
+
+ENTRY(sys_sigaltstack)
+ l.j _sys_sigaltstack
+ l.addi r5,r1,0
+
+ENTRY(sys_rt_sigreturn)
+ l.j _sys_rt_sigreturn
+ l.addi r3,r1,0
+
+/* This is a catch-all syscall for atomic instructions for the OpenRISC 1000.
+ * The functions takes a variable number of parameters depending on which
+ * particular flavour of atomic you want... parameter 1 is a flag identifying
+ * the atomic in question. Currently, this function implements the
+ * following variants:
+ *
+ * XCHG:
+ * @flag: 1
+ * @ptr1:
+ * @ptr2:
+ * Atomically exchange the values in pointers 1 and 2.
+ *
+ */
+
+ENTRY(sys_or1k_atomic)
+ /* FIXME: This ignores r3 and always does an XCHG */
+ DISABLE_INTERRUPTS(r17,r19)
+ l.lwz r30,0(r4)
+ l.lwz r28,0(r5)
+ l.sw 0(r4),r28
+ l.sw 0(r5),r30
+ ENABLE_INTERRUPTS(r17)
+ l.jr r9
+ l.or r11,r0,r0
+
+/* ============================================================[ EOF ]=== */
diff --git a/arch/openrisc/kernel/head.S b/arch/openrisc/kernel/head.S
new file mode 100644
index 0000000..c75018d
--- /dev/null
+++ b/arch/openrisc/kernel/head.S
@@ -0,0 +1,1607 @@
+/*
+ * OpenRISC head.S
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/linkage.h>
+#include <linux/threads.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <asm/processor.h>
+#include <asm/page.h>
+#include <asm/mmu.h>
+#include <asm/pgtable.h>
+#include <asm/cache.h>
+#include <asm/spr_defs.h>
+#include <asm/asm-offsets.h>
+
+#define tophys(rd,rs) \
+ l.movhi rd,hi(-KERNELBASE) ;\
+ l.add rd,rd,rs
+
+#define CLEAR_GPR(gpr) \
+ l.or gpr,r0,r0
+
+#define LOAD_SYMBOL_2_GPR(gpr,symbol) \
+ l.movhi gpr,hi(symbol) ;\
+ l.ori gpr,gpr,lo(symbol)
+
+
+#define UART_BASE_ADD 0x90000000
+
+#define EXCEPTION_SR (SPR_SR_DME | SPR_SR_IME | SPR_SR_DCE | SPR_SR_ICE | SPR_SR_SM)
+#define SYSCALL_SR (SPR_SR_DME | SPR_SR_IME | SPR_SR_DCE | SPR_SR_ICE | SPR_SR_IEE | SPR_SR_TEE | SPR_SR_SM)
+
+/* ============================================[ tmp store locations ]=== */
+
+/*
+ * emergency_print temporary stores
+ */
+#define EMERGENCY_PRINT_STORE_GPR4 l.sw 0x20(r0),r4
+#define EMERGENCY_PRINT_LOAD_GPR4 l.lwz r4,0x20(r0)
+
+#define EMERGENCY_PRINT_STORE_GPR5 l.sw 0x24(r0),r5
+#define EMERGENCY_PRINT_LOAD_GPR5 l.lwz r5,0x24(r0)
+
+#define EMERGENCY_PRINT_STORE_GPR6 l.sw 0x28(r0),r6
+#define EMERGENCY_PRINT_LOAD_GPR6 l.lwz r6,0x28(r0)
+
+#define EMERGENCY_PRINT_STORE_GPR7 l.sw 0x2c(r0),r7
+#define EMERGENCY_PRINT_LOAD_GPR7 l.lwz r7,0x2c(r0)
+
+#define EMERGENCY_PRINT_STORE_GPR8 l.sw 0x30(r0),r8
+#define EMERGENCY_PRINT_LOAD_GPR8 l.lwz r8,0x30(r0)
+
+#define EMERGENCY_PRINT_STORE_GPR9 l.sw 0x34(r0),r9
+#define EMERGENCY_PRINT_LOAD_GPR9 l.lwz r9,0x34(r0)
+
+
+/*
+ * TLB miss handlers temorary stores
+ */
+#define EXCEPTION_STORE_GPR9 l.sw 0x10(r0),r9
+#define EXCEPTION_LOAD_GPR9 l.lwz r9,0x10(r0)
+
+#define EXCEPTION_STORE_GPR2 l.sw 0x64(r0),r2
+#define EXCEPTION_LOAD_GPR2 l.lwz r2,0x64(r0)
+
+#define EXCEPTION_STORE_GPR3 l.sw 0x68(r0),r3
+#define EXCEPTION_LOAD_GPR3 l.lwz r3,0x68(r0)
+
+#define EXCEPTION_STORE_GPR4 l.sw 0x6c(r0),r4
+#define EXCEPTION_LOAD_GPR4 l.lwz r4,0x6c(r0)
+
+#define EXCEPTION_STORE_GPR5 l.sw 0x70(r0),r5
+#define EXCEPTION_LOAD_GPR5 l.lwz r5,0x70(r0)
+
+#define EXCEPTION_STORE_GPR6 l.sw 0x74(r0),r6
+#define EXCEPTION_LOAD_GPR6 l.lwz r6,0x74(r0)
+
+
+/*
+ * EXCEPTION_HANDLE temporary stores
+ */
+
+#define EXCEPTION_T_STORE_GPR30 l.sw 0x78(r0),r30
+#define EXCEPTION_T_LOAD_GPR30(reg) l.lwz reg,0x78(r0)
+
+#define EXCEPTION_T_STORE_GPR10 l.sw 0x7c(r0),r10
+#define EXCEPTION_T_LOAD_GPR10(reg) l.lwz reg,0x7c(r0)
+
+#define EXCEPTION_T_STORE_SP l.sw 0x80(r0),r1
+#define EXCEPTION_T_LOAD_SP(reg) l.lwz reg,0x80(r0)
+
+/*
+ * For UNHANLDED_EXCEPTION
+ */
+
+#define EXCEPTION_T_STORE_GPR31 l.sw 0x84(r0),r31
+#define EXCEPTION_T_LOAD_GPR31(reg) l.lwz reg,0x84(r0)
+
+/* =========================================================[ macros ]=== */
+
+
+#define GET_CURRENT_PGD(reg,t1) \
+ LOAD_SYMBOL_2_GPR(reg,current_pgd) ;\
+ tophys (t1,reg) ;\
+ l.lwz reg,0(t1)
+
+
+/*
+ * DSCR: this is a common hook for handling exceptions. it will save
+ * the needed registers, set up stack and pointer to current
+ * then jump to the handler while enabling MMU
+ *
+ * PRMS: handler - a function to jump to. it has to save the
+ * remaining registers to kernel stack, call
+ * appropriate arch-independant exception handler
+ * and finaly jump to ret_from_except
+ *
+ * PREQ: unchanged state from the time exception happened
+ *
+ * POST: SAVED the following registers original value
+ * to the new created exception frame pointed to by r1
+ *
+ * r1 - ksp pointing to the new (exception) frame
+ * r4 - EEAR exception EA
+ * r10 - current pointing to current_thread_info struct
+ * r12 - syscall 0, since we didn't come from syscall
+ * r13 - temp it actually contains new SR, not needed anymore
+ * r31 - handler address of the handler we'll jump to
+ *
+ * handler has to save remaining registers to the exception
+ * ksp frame *before* tainting them!
+ *
+ * NOTE: this function is not reentrant per se. reentrancy is guaranteed
+ * by processor disabling all exceptions/interrupts when exception
+ * accours.
+ *
+ * OPTM: no need to make it so wasteful to extract ksp when in user mode
+ */
+
+#define EXCEPTION_HANDLE(handler) \
+ EXCEPTION_T_STORE_GPR30 ;\
+ l.mfspr r30,r0,SPR_ESR_BASE ;\
+ l.andi r30,r30,SPR_SR_SM ;\
+ l.sfeqi r30,0 ;\
+ EXCEPTION_T_STORE_GPR10 ;\
+ l.bnf 2f /* kernel_mode */ ;\
+ EXCEPTION_T_STORE_SP /* delay slot */ ;\
+1: /* user_mode: */ ;\
+ LOAD_SYMBOL_2_GPR(r1,current_thread_info_set) ;\
+ tophys (r30,r1) ;\
+ /* r10: current_thread_info */ ;\
+ l.lwz r10,0(r30) ;\
+ tophys (r30,r10) ;\
+ l.lwz r1,(TI_KSP)(r30) ;\
+ /* fall through */ ;\
+2: /* kernel_mode: */ ;\
+ /* create new stack frame, save only needed gprs */ ;\
+ /* r1: KSP, r10: current, r4: EEAR, r31: __pa(KSP) */ ;\
+ /* r12: temp, syscall indicator */ ;\
+ l.addi r1,r1,-(INT_FRAME_SIZE) ;\
+ /* r1 is KSP, r30 is __pa(KSP) */ ;\
+ tophys (r30,r1) ;\
+ l.sw PT_GPR12(r30),r12 ;\
+ l.mfspr r12,r0,SPR_EPCR_BASE ;\
+ l.sw PT_PC(r30),r12 ;\
+ l.mfspr r12,r0,SPR_ESR_BASE ;\
+ l.sw PT_SR(r30),r12 ;\
+ /* save r30 */ ;\
+ EXCEPTION_T_LOAD_GPR30(r12) ;\
+ l.sw PT_GPR30(r30),r12 ;\
+ /* save r10 as was prior to exception */ ;\
+ EXCEPTION_T_LOAD_GPR10(r12) ;\
+ l.sw PT_GPR10(r30),r12 ;\
+ /* save PT_SP as was prior to exception */ ;\
+ EXCEPTION_T_LOAD_SP(r12) ;\
+ l.sw PT_SP(r30),r12 ;\
+ /* save exception r4, set r4 = EA */ ;\
+ l.sw PT_GPR4(r30),r4 ;\
+ l.mfspr r4,r0,SPR_EEAR_BASE ;\
+ /* r12 == 1 if we come from syscall */ ;\
+ CLEAR_GPR(r12) ;\
+ /* ----- turn on MMU ----- */ ;\
+ l.ori r30,r0,(EXCEPTION_SR) ;\
+ l.mtspr r0,r30,SPR_ESR_BASE ;\
+ /* r30: EA address of handler */ ;\
+ LOAD_SYMBOL_2_GPR(r30,handler) ;\
+ l.mtspr r0,r30,SPR_EPCR_BASE ;\
+ l.rfe
+
+/*
+ * this doesn't work
+ *
+ *
+ * #ifdef CONFIG_JUMP_UPON_UNHANDLED_EXCEPTION
+ * #define UNHANDLED_EXCEPTION(handler) \
+ * l.ori r3,r0,0x1 ;\
+ * l.mtspr r0,r3,SPR_SR ;\
+ * l.movhi r3,hi(0xf0000100) ;\
+ * l.ori r3,r3,lo(0xf0000100) ;\
+ * l.jr r3 ;\
+ * l.nop 1
+ *
+ * #endif
+ */
+
+/* DSCR: this is the same as EXCEPTION_HANDLE(), we are just
+ * a bit more carefull (if we have a PT_SP or current pointer
+ * corruption) and set them up from 'current_set'
+ *
+ */
+#define UNHANDLED_EXCEPTION(handler) \
+ EXCEPTION_T_STORE_GPR31 ;\
+ EXCEPTION_T_STORE_GPR10 ;\
+ EXCEPTION_T_STORE_SP ;\
+ /* temporary store r3, r9 into r1, r10 */ ;\
+ l.addi r1,r3,0x0 ;\
+ l.addi r10,r9,0x0 ;\
+ /* the string referenced by r3 must be low enough */ ;\
+ l.jal _emergency_print ;\
+ l.ori r3,r0,lo(_string_unhandled_exception) ;\
+ l.mfspr r3,r0,SPR_NPC ;\
+ l.jal _emergency_print_nr ;\
+ l.andi r3,r3,0x1f00 ;\
+ /* the string referenced by r3 must be low enough */ ;\
+ l.jal _emergency_print ;\
+ l.ori r3,r0,lo(_string_epc_prefix) ;\
+ l.jal _emergency_print_nr ;\
+ l.mfspr r3,r0,SPR_EPCR_BASE ;\
+ l.jal _emergency_print ;\
+ l.ori r3,r0,lo(_string_nl) ;\
+ /* end of printing */ ;\
+ l.addi r3,r1,0x0 ;\
+ l.addi r9,r10,0x0 ;\
+ /* extract current, ksp from current_set */ ;\
+ LOAD_SYMBOL_2_GPR(r1,_unhandled_stack_top) ;\
+ LOAD_SYMBOL_2_GPR(r10,init_thread_union) ;\
+ /* create new stack frame, save only needed gprs */ ;\
+ /* r1: KSP, r10: current, r31: __pa(KSP) */ ;\
+ /* r12: temp, syscall indicator, r13 temp */ ;\
+ l.addi r1,r1,-(INT_FRAME_SIZE) ;\
+ /* r1 is KSP, r31 is __pa(KSP) */ ;\
+ tophys (r31,r1) ;\
+ l.sw PT_GPR12(r31),r12 ;\
+ l.mfspr r12,r0,SPR_EPCR_BASE ;\
+ l.sw PT_PC(r31),r12 ;\
+ l.mfspr r12,r0,SPR_ESR_BASE ;\
+ l.sw PT_SR(r31),r12 ;\
+ /* save r31 */ ;\
+ EXCEPTION_T_LOAD_GPR31(r12) ;\
+ l.sw PT_GPR31(r31),r12 ;\
+ /* save r10 as was prior to exception */ ;\
+ EXCEPTION_T_LOAD_GPR10(r12) ;\
+ l.sw PT_GPR10(r31),r12 ;\
+ /* save PT_SP as was prior to exception */ ;\
+ EXCEPTION_T_LOAD_SP(r12) ;\
+ l.sw PT_SP(r31),r12 ;\
+ l.sw PT_GPR13(r31),r13 ;\
+ /* --> */ ;\
+ /* save exception r4, set r4 = EA */ ;\
+ l.sw PT_GPR4(r31),r4 ;\
+ l.mfspr r4,r0,SPR_EEAR_BASE ;\
+ /* r12 == 1 if we come from syscall */ ;\
+ CLEAR_GPR(r12) ;\
+ /* ----- play a MMU trick ----- */ ;\
+ l.ori r31,r0,(EXCEPTION_SR) ;\
+ l.mtspr r0,r31,SPR_ESR_BASE ;\
+ /* r31: EA address of handler */ ;\
+ LOAD_SYMBOL_2_GPR(r31,handler) ;\
+ l.mtspr r0,r31,SPR_EPCR_BASE ;\
+ l.rfe
+
+/* =====================================================[ exceptions] === */
+
+/* ---[ 0x100: RESET exception ]----------------------------------------- */
+ .org 0x100
+ /* Jump to .init code at _start which lives in the .head section
+ * and will be discarded after boot.
+ */
+ LOAD_SYMBOL_2_GPR(r4, _start)
+ tophys (r3,r4) /* MMU disabled */
+ l.jr r3
+ l.nop
+
+/* ---[ 0x200: BUS exception ]------------------------------------------- */
+ .org 0x200
+_dispatch_bus_fault:
+ EXCEPTION_HANDLE(_bus_fault_handler)
+
+/* ---[ 0x300: Data Page Fault exception ]------------------------------- */
+ .org 0x300
+_dispatch_do_dpage_fault:
+// totaly disable timer interrupt
+// l.mtspr r0,r0,SPR_TTMR
+// DEBUG_TLB_PROBE(0x300)
+// EXCEPTION_DEBUG_VALUE_ER_ENABLED(0x300)
+ EXCEPTION_HANDLE(_data_page_fault_handler)
+
+/* ---[ 0x400: Insn Page Fault exception ]------------------------------- */
+ .org 0x400
+_dispatch_do_ipage_fault:
+// totaly disable timer interrupt
+// l.mtspr r0,r0,SPR_TTMR
+// DEBUG_TLB_PROBE(0x400)
+// EXCEPTION_DEBUG_VALUE_ER_ENABLED(0x400)
+ EXCEPTION_HANDLE(_insn_page_fault_handler)
+
+/* ---[ 0x500: Timer exception ]----------------------------------------- */
+ .org 0x500
+ EXCEPTION_HANDLE(_timer_handler)
+
+/* ---[ 0x600: Aligment exception ]-------------------------------------- */
+ .org 0x600
+ EXCEPTION_HANDLE(_alignment_handler)
+
+/* ---[ 0x700: Illegal insn exception ]---------------------------------- */
+ .org 0x700
+ EXCEPTION_HANDLE(_illegal_instruction_handler)
+
+/* ---[ 0x800: External interrupt exception ]---------------------------- */
+ .org 0x800
+ EXCEPTION_HANDLE(_external_irq_handler)
+
+/* ---[ 0x900: DTLB miss exception ]------------------------------------- */
+ .org 0x900
+ l.j boot_dtlb_miss_handler
+ l.nop
+
+/* ---[ 0xa00: ITLB miss exception ]------------------------------------- */
+ .org 0xa00
+ l.j boot_itlb_miss_handler
+ l.nop
+
+/* ---[ 0xb00: Range exception ]----------------------------------------- */
+ .org 0xb00
+ UNHANDLED_EXCEPTION(_vector_0xb00)
+
+/* ---[ 0xc00: Syscall exception ]--------------------------------------- */
+ .org 0xc00
+ EXCEPTION_HANDLE(_sys_call_handler)
+
+/* ---[ 0xd00: Trap exception ]------------------------------------------ */
+ .org 0xd00
+ UNHANDLED_EXCEPTION(_vector_0xd00)
+
+/* ---[ 0xe00: Trap exception ]------------------------------------------ */
+ .org 0xe00
+// UNHANDLED_EXCEPTION(_vector_0xe00)
+ EXCEPTION_HANDLE(_trap_handler)
+
+/* ---[ 0xf00: Reserved exception ]-------------------------------------- */
+ .org 0xf00
+ UNHANDLED_EXCEPTION(_vector_0xf00)
+
+/* ---[ 0x1000: Reserved exception ]------------------------------------- */
+ .org 0x1000
+ UNHANDLED_EXCEPTION(_vector_0x1000)
+
+/* ---[ 0x1100: Reserved exception ]------------------------------------- */
+ .org 0x1100
+ UNHANDLED_EXCEPTION(_vector_0x1100)
+
+/* ---[ 0x1200: Reserved exception ]------------------------------------- */
+ .org 0x1200
+ UNHANDLED_EXCEPTION(_vector_0x1200)
+
+/* ---[ 0x1300: Reserved exception ]------------------------------------- */
+ .org 0x1300
+ UNHANDLED_EXCEPTION(_vector_0x1300)
+
+/* ---[ 0x1400: Reserved exception ]------------------------------------- */
+ .org 0x1400
+ UNHANDLED_EXCEPTION(_vector_0x1400)
+
+/* ---[ 0x1500: Reserved exception ]------------------------------------- */
+ .org 0x1500
+ UNHANDLED_EXCEPTION(_vector_0x1500)
+
+/* ---[ 0x1600: Reserved exception ]------------------------------------- */
+ .org 0x1600
+ UNHANDLED_EXCEPTION(_vector_0x1600)
+
+/* ---[ 0x1700: Reserved exception ]------------------------------------- */
+ .org 0x1700
+ UNHANDLED_EXCEPTION(_vector_0x1700)
+
+/* ---[ 0x1800: Reserved exception ]------------------------------------- */
+ .org 0x1800
+ UNHANDLED_EXCEPTION(_vector_0x1800)
+
+/* ---[ 0x1900: Reserved exception ]------------------------------------- */
+ .org 0x1900
+ UNHANDLED_EXCEPTION(_vector_0x1900)
+
+/* ---[ 0x1a00: Reserved exception ]------------------------------------- */
+ .org 0x1a00
+ UNHANDLED_EXCEPTION(_vector_0x1a00)
+
+/* ---[ 0x1b00: Reserved exception ]------------------------------------- */
+ .org 0x1b00
+ UNHANDLED_EXCEPTION(_vector_0x1b00)
+
+/* ---[ 0x1c00: Reserved exception ]------------------------------------- */
+ .org 0x1c00
+ UNHANDLED_EXCEPTION(_vector_0x1c00)
+
+/* ---[ 0x1d00: Reserved exception ]------------------------------------- */
+ .org 0x1d00
+ UNHANDLED_EXCEPTION(_vector_0x1d00)
+
+/* ---[ 0x1e00: Reserved exception ]------------------------------------- */
+ .org 0x1e00
+ UNHANDLED_EXCEPTION(_vector_0x1e00)
+
+/* ---[ 0x1f00: Reserved exception ]------------------------------------- */
+ .org 0x1f00
+ UNHANDLED_EXCEPTION(_vector_0x1f00)
+
+ .org 0x2000
+/* ===================================================[ kernel start ]=== */
+
+/* .text*/
+
+/* This early stuff belongs in HEAD, but some of the functions below definitely
+ * don't... */
+
+ __HEAD
+ .global _start
+_start:
+ /*
+ * ensure a deterministic start
+ */
+
+ l.ori r3,r0,0x1
+ l.mtspr r0,r3,SPR_SR
+
+ CLEAR_GPR(r1)
+ CLEAR_GPR(r2)
+ CLEAR_GPR(r3)
+ CLEAR_GPR(r4)
+ CLEAR_GPR(r5)
+ CLEAR_GPR(r6)
+ CLEAR_GPR(r7)
+ CLEAR_GPR(r8)
+ CLEAR_GPR(r9)
+ CLEAR_GPR(r10)
+ CLEAR_GPR(r11)
+ CLEAR_GPR(r12)
+ CLEAR_GPR(r13)
+ CLEAR_GPR(r14)
+ CLEAR_GPR(r15)
+ CLEAR_GPR(r16)
+ CLEAR_GPR(r17)
+ CLEAR_GPR(r18)
+ CLEAR_GPR(r19)
+ CLEAR_GPR(r20)
+ CLEAR_GPR(r21)
+ CLEAR_GPR(r22)
+ CLEAR_GPR(r23)
+ CLEAR_GPR(r24)
+ CLEAR_GPR(r25)
+ CLEAR_GPR(r26)
+ CLEAR_GPR(r27)
+ CLEAR_GPR(r28)
+ CLEAR_GPR(r29)
+ CLEAR_GPR(r30)
+ CLEAR_GPR(r31)
+
+ /*
+ * set up initial ksp and current
+ */
+ LOAD_SYMBOL_2_GPR(r1,init_thread_union+0x2000) // setup kernel stack
+ LOAD_SYMBOL_2_GPR(r10,init_thread_union) // setup current
+ tophys (r31,r10)
+ l.sw TI_KSP(r31), r1
+
+ l.ori r4,r0,0x0
+
+
+ /*
+ * .data contains initialized data,
+ * .bss contains uninitialized data - clear it up
+ */
+clear_bss:
+ LOAD_SYMBOL_2_GPR(r24, __bss_start)
+ LOAD_SYMBOL_2_GPR(r26, _end)
+ tophys(r28,r24)
+ tophys(r30,r26)
+ CLEAR_GPR(r24)
+ CLEAR_GPR(r26)
+1:
+ l.sw (0)(r28),r0
+ l.sfltu r28,r30
+ l.bf 1b
+ l.addi r28,r28,4
+
+enable_ic:
+ l.jal _ic_enable
+ l.nop
+
+enable_dc:
+ l.jal _dc_enable
+ l.nop
+
+flush_tlb:
+ /*
+ * I N V A L I D A T E T L B e n t r i e s
+ */
+ LOAD_SYMBOL_2_GPR(r5,SPR_DTLBMR_BASE(0))
+ LOAD_SYMBOL_2_GPR(r6,SPR_ITLBMR_BASE(0))
+ l.addi r7,r0,128 /* Maximum number of sets */
+1:
+ l.mtspr r5,r0,0x0
+ l.mtspr r6,r0,0x0
+
+ l.addi r5,r5,1
+ l.addi r6,r6,1
+ l.sfeq r7,r0
+ l.bnf 1b
+ l.addi r7,r7,-1
+
+
+/* The MMU needs to be enabled before or32_early_setup is called */
+
+enable_mmu:
+ /*
+ * enable dmmu & immu
+ * SR[5] = 0, SR[6] = 0, 6th and 7th bit of SR set to 0
+ */
+ l.mfspr r30,r0,SPR_SR
+ l.movhi r28,hi(SPR_SR_DME | SPR_SR_IME)
+ l.ori r28,r28,lo(SPR_SR_DME | SPR_SR_IME)
+ l.or r30,r30,r28
+ l.mtspr r0,r30,SPR_SR
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+
+ // reset the simulation counters
+ l.nop 5
+
+ LOAD_SYMBOL_2_GPR(r24, or32_early_setup)
+ l.jalr r24
+ l.nop
+
+clear_regs:
+ /*
+ * clear all GPRS to increase determinism
+ */
+ CLEAR_GPR(r2)
+ CLEAR_GPR(r3)
+ CLEAR_GPR(r4)
+ CLEAR_GPR(r5)
+ CLEAR_GPR(r6)
+ CLEAR_GPR(r7)
+ CLEAR_GPR(r8)
+ CLEAR_GPR(r9)
+ CLEAR_GPR(r11)
+ CLEAR_GPR(r12)
+ CLEAR_GPR(r13)
+ CLEAR_GPR(r14)
+ CLEAR_GPR(r15)
+ CLEAR_GPR(r16)
+ CLEAR_GPR(r17)
+ CLEAR_GPR(r18)
+ CLEAR_GPR(r19)
+ CLEAR_GPR(r20)
+ CLEAR_GPR(r21)
+ CLEAR_GPR(r22)
+ CLEAR_GPR(r23)
+ CLEAR_GPR(r24)
+ CLEAR_GPR(r25)
+ CLEAR_GPR(r26)
+ CLEAR_GPR(r27)
+ CLEAR_GPR(r28)
+ CLEAR_GPR(r29)
+ CLEAR_GPR(r30)
+ CLEAR_GPR(r31)
+
+jump_start_kernel:
+ /*
+ * jump to kernel entry (start_kernel)
+ */
+ LOAD_SYMBOL_2_GPR(r30, start_kernel)
+ l.jr r30
+ l.nop
+
+/* ========================================[ cache ]=== */
+
+ /* aligment here so we don't change memory offsets with
+ * memory controler defined
+ */
+ .align 0x2000
+
+_ic_enable:
+ /* Check if IC present and skip enabling otherwise */
+ l.mfspr r24,r0,SPR_UPR
+ l.andi r26,r24,SPR_UPR_ICP
+ l.sfeq r26,r0
+ l.bf 9f
+ l.nop
+
+ /* Disable IC */
+ l.mfspr r6,r0,SPR_SR
+ l.addi r5,r0,-1
+ l.xori r5,r5,SPR_SR_ICE
+ l.and r5,r6,r5
+ l.mtspr r0,r5,SPR_SR
+
+ /* Establish cache block size
+ If BS=0, 16;
+ If BS=1, 32;
+ r14 contain block size
+ */
+ l.mfspr r24,r0,SPR_ICCFGR
+ l.andi r26,r24,SPR_ICCFGR_CBS
+ l.srli r28,r26,7
+ l.ori r30,r0,16
+ l.sll r14,r30,r28
+
+ /* Establish number of cache sets
+ r16 contains number of cache sets
+ r28 contains log(# of cache sets)
+ */
+ l.andi r26,r24,SPR_ICCFGR_NCS
+ l.srli r28,r26,3
+ l.ori r30,r0,1
+ l.sll r16,r30,r28
+
+ /* Invalidate IC */
+ l.addi r6,r0,0
+ l.sll r5,r14,r28
+// l.mul r5,r14,r16
+// l.trap 1
+// l.addi r5,r0,IC_SIZE
+1:
+ l.mtspr r0,r6,SPR_ICBIR
+ l.sfne r6,r5
+ l.bf 1b
+ l.add r6,r6,r14
+ // l.addi r6,r6,IC_LINE
+
+ /* Enable IC */
+ l.mfspr r6,r0,SPR_SR
+ l.ori r6,r6,SPR_SR_ICE
+ l.mtspr r0,r6,SPR_SR
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+ l.nop
+9:
+ l.jr r9
+ l.nop
+
+_dc_enable:
+ /* Check if DC present and skip enabling otherwise */
+ l.mfspr r24,r0,SPR_UPR
+ l.andi r26,r24,SPR_UPR_DCP
+ l.sfeq r26,r0
+ l.bf 9f
+ l.nop
+
+ /* Disable DC */
+ l.mfspr r6,r0,SPR_SR
+ l.addi r5,r0,-1
+ l.xori r5,r5,SPR_SR_DCE
+ l.and r5,r6,r5
+ l.mtspr r0,r5,SPR_SR
+
+ /* Establish cache block size
+ If BS=0, 16;
+ If BS=1, 32;
+ r14 contain block size
+ */
+ l.mfspr r24,r0,SPR_DCCFGR
+ l.andi r26,r24,SPR_DCCFGR_CBS
+ l.srli r28,r26,7
+ l.ori r30,r0,16
+ l.sll r14,r30,r28
+
+ /* Establish number of cache sets
+ r16 contains number of cache sets
+ r28 contains log(# of cache sets)
+ */
+ l.andi r26,r24,SPR_DCCFGR_NCS
+ l.srli r28,r26,3
+ l.ori r30,r0,1
+ l.sll r16,r30,r28
+
+ /* Invalidate DC */
+ l.addi r6,r0,0
+ l.sll r5,r14,r28
+1:
+ l.mtspr r0,r6,SPR_DCBIR
+ l.sfne r6,r5
+ l.bf 1b
+ l.add r6,r6,r14
+
+ /* Enable DC */
+ l.mfspr r6,r0,SPR_SR
+ l.ori r6,r6,SPR_SR_DCE
+ l.mtspr r0,r6,SPR_SR
+9:
+ l.jr r9
+ l.nop
+
+/* ===============================================[ page table masks ]=== */
+
+/* bit 4 is used in hardware as write back cache bit. we never use this bit
+ * explicitly, so we can reuse it as _PAGE_FILE bit and mask it out when
+ * writing into hardware pte's
+ */
+
+#define DTLB_UP_CONVERT_MASK 0x3fa
+#define ITLB_UP_CONVERT_MASK 0x3a
+
+/* for SMP we'd have (this is a bit subtle, CC must be always set
+ * for SMP, but since we have _PAGE_PRESENT bit always defined
+ * we can just modify the mask)
+ */
+#define DTLB_SMP_CONVERT_MASK 0x3fb
+#define ITLB_SMP_CONVERT_MASK 0x3b
+
+/* ---[ boot dtlb miss handler ]----------------------------------------- */
+
+boot_dtlb_miss_handler:
+
+/* mask for DTLB_MR register: - (0) sets V (valid) bit,
+ * - (31-12) sets bits belonging to VPN (31-12)
+ */
+#define DTLB_MR_MASK 0xfffff001
+
+/* mask for DTLB_TR register: - (2) sets CI (cache inhibit) bit,
+ * - (4) sets A (access) bit,
+ * - (5) sets D (dirty) bit,
+ * - (8) sets SRE (superuser read) bit
+ * - (9) sets SWE (superuser write) bit
+ * - (31-12) sets bits belonging to VPN (31-12)
+ */
+#define DTLB_TR_MASK 0xfffff332
+
+/* These are for masking out the VPN/PPN value from the MR/TR registers...
+ * it's not the same as the PFN */
+#define VPN_MASK 0xfffff000
+#define PPN_MASK 0xfffff000
+
+
+ EXCEPTION_STORE_GPR6
+
+#if 0
+ l.mfspr r6,r0,SPR_ESR_BASE //
+ l.andi r6,r6,SPR_SR_SM // are we in kernel mode ?
+ l.sfeqi r6,0 // r6 == 0x1 --> SM
+ l.bf exit_with_no_dtranslation //
+ l.nop
+#endif
+
+ /* this could be optimized by moving storing of
+ * non r6 registers here, and jumping r6 restore
+ * if not in supervisor mode
+ */
+
+ EXCEPTION_STORE_GPR2
+ EXCEPTION_STORE_GPR3
+ EXCEPTION_STORE_GPR4
+ EXCEPTION_STORE_GPR5
+
+ l.mfspr r4,r0,SPR_EEAR_BASE // get the offending EA
+
+immediate_translation:
+ CLEAR_GPR(r6)
+
+ l.srli r3,r4,0xd // r3 <- r4 / 8192 (sets are relative to page size (8Kb) NOT VPN size (4Kb)
+
+ l.mfspr r6, r0, SPR_DMMUCFGR
+ l.andi r6, r6, SPR_DMMUCFGR_NTS
+ l.srli r6, r6, SPR_DMMUCFGR_NTS_OFF
+ l.ori r5, r0, 0x1
+ l.sll r5, r5, r6 // r5 = number DMMU sets
+ l.addi r6, r5, -1 // r6 = nsets mask
+ l.and r2, r3, r6 // r2 <- r3 % NSETS_MASK
+
+ l.or r6,r6,r4 // r6 <- r4
+ l.ori r6,r6,~(VPN_MASK) // r6 <- VPN :VPN .xfff - clear up lo(r6) to 0x**** *fff
+ l.movhi r5,hi(DTLB_MR_MASK) // r5 <- ffff:0000.x000
+ l.ori r5,r5,lo(DTLB_MR_MASK) // r5 <- ffff:1111.x001 - apply DTLB_MR_MASK
+ l.and r5,r5,r6 // r5 <- VPN :VPN .x001 - we have DTLBMR entry
+ l.mtspr r2,r5,SPR_DTLBMR_BASE(0) // set DTLBMR
+
+ /* set up DTLB with no translation for EA <= 0xbfffffff */
+ LOAD_SYMBOL_2_GPR(r6,0xbfffffff)
+ l.sfgeu r6,r4 // flag if r6 >= r4 (if 0xbfffffff >= EA)
+ l.bf 1f // goto out
+ l.and r3,r4,r4 // delay slot :: 24 <- r4 (if flag==1)
+
+ tophys(r3,r4) // r3 <- PA
+1:
+ l.ori r3,r3,~(PPN_MASK) // r3 <- PPN :PPN .xfff - clear up lo(r6) to 0x**** *fff
+ l.movhi r5,hi(DTLB_TR_MASK) // r5 <- ffff:0000.x000
+ l.ori r5,r5,lo(DTLB_TR_MASK) // r5 <- ffff:1111.x330 - apply DTLB_MR_MASK
+ l.and r5,r5,r3 // r5 <- PPN :PPN .x330 - we have DTLBTR entry
+ l.mtspr r2,r5,SPR_DTLBTR_BASE(0) // set DTLBTR
+
+ EXCEPTION_LOAD_GPR6
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR2
+
+ l.rfe // SR <- ESR, PC <- EPC
+
+exit_with_no_dtranslation:
+ /* EA out of memory or not in supervisor mode */
+ EXCEPTION_LOAD_GPR6
+ EXCEPTION_LOAD_GPR4
+ l.j _dispatch_bus_fault
+
+/* ---[ boot itlb miss handler ]----------------------------------------- */
+
+boot_itlb_miss_handler:
+
+/* mask for ITLB_MR register: - sets V (valid) bit,
+ * - sets bits belonging to VPN (15-12)
+ */
+#define ITLB_MR_MASK 0xfffff001
+
+/* mask for ITLB_TR register: - sets A (access) bit,
+ * - sets SXE (superuser execute) bit
+ * - sets bits belonging to VPN (15-12)
+ */
+#define ITLB_TR_MASK 0xfffff050
+
+/*
+#define VPN_MASK 0xffffe000
+#define PPN_MASK 0xffffe000
+*/
+
+
+
+ EXCEPTION_STORE_GPR2
+ EXCEPTION_STORE_GPR3
+ EXCEPTION_STORE_GPR4
+ EXCEPTION_STORE_GPR5
+ EXCEPTION_STORE_GPR6
+
+#if 0
+ l.mfspr r6,r0,SPR_ESR_BASE //
+ l.andi r6,r6,SPR_SR_SM // are we in kernel mode ?
+ l.sfeqi r6,0 // r6 == 0x1 --> SM
+ l.bf exit_with_no_itranslation
+ l.nop
+#endif
+
+
+ l.mfspr r4,r0,SPR_EEAR_BASE // get the offending EA
+
+earlyearly:
+ CLEAR_GPR(r6)
+
+ l.srli r3,r4,0xd // r3 <- r4 / 8192 (sets are relative to page size (8Kb) NOT VPN size (4Kb)
+
+ l.mfspr r6, r0, SPR_IMMUCFGR
+ l.andi r6, r6, SPR_IMMUCFGR_NTS
+ l.srli r6, r6, SPR_IMMUCFGR_NTS_OFF
+ l.ori r5, r0, 0x1
+ l.sll r5, r5, r6 // r5 = number IMMU sets from IMMUCFGR
+ l.addi r6, r5, -1 // r6 = nsets mask
+ l.and r2, r3, r6 // r2 <- r3 % NSETS_MASK
+
+ l.or r6,r6,r4 // r6 <- r4
+ l.ori r6,r6,~(VPN_MASK) // r6 <- VPN :VPN .xfff - clear up lo(r6) to 0x**** *fff
+ l.movhi r5,hi(ITLB_MR_MASK) // r5 <- ffff:0000.x000
+ l.ori r5,r5,lo(ITLB_MR_MASK) // r5 <- ffff:1111.x001 - apply ITLB_MR_MASK
+ l.and r5,r5,r6 // r5 <- VPN :VPN .x001 - we have ITLBMR entry
+ l.mtspr r2,r5,SPR_ITLBMR_BASE(0) // set ITLBMR
+
+ /*
+ * set up ITLB with no translation for EA <= 0x0fffffff
+ *
+ * we need this for head.S mapping (EA = PA). if we move all functions
+ * which run with mmu enabled into entry.S, we might be able to eliminate this.
+ *
+ */
+ LOAD_SYMBOL_2_GPR(r6,0x0fffffff)
+ l.sfgeu r6,r4 // flag if r6 >= r4 (if 0xb0ffffff >= EA)
+ l.bf 1f // goto out
+ l.and r3,r4,r4 // delay slot :: 24 <- r4 (if flag==1)
+
+ tophys(r3,r4) // r3 <- PA
+1:
+ l.ori r3,r3,~(PPN_MASK) // r3 <- PPN :PPN .xfff - clear up lo(r6) to 0x**** *fff
+ l.movhi r5,hi(ITLB_TR_MASK) // r5 <- ffff:0000.x000
+ l.ori r5,r5,lo(ITLB_TR_MASK) // r5 <- ffff:1111.x050 - apply ITLB_MR_MASK
+ l.and r5,r5,r3 // r5 <- PPN :PPN .x050 - we have ITLBTR entry
+ l.mtspr r2,r5,SPR_ITLBTR_BASE(0) // set ITLBTR
+
+ EXCEPTION_LOAD_GPR6
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR2
+
+ l.rfe // SR <- ESR, PC <- EPC
+
+exit_with_no_itranslation:
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR6
+ l.j _dispatch_bus_fault
+ l.nop
+
+/* ====================================================================== */
+/*
+ * Stuff below here shouldn't go into .head section... maybe this stuff
+ * can be moved to entry.S ???
+ */
+
+/* ==============================================[ DTLB miss handler ]=== */
+
+/*
+ * Comments:
+ * Exception handlers are entered with MMU off so the following handler
+ * needs to use physical addressing
+ *
+ */
+
+ .text
+ENTRY(dtlb_miss_handler)
+ EXCEPTION_STORE_GPR2
+ EXCEPTION_STORE_GPR3
+ EXCEPTION_STORE_GPR4
+ EXCEPTION_STORE_GPR5
+ EXCEPTION_STORE_GPR6
+ /*
+ * get EA of the miss
+ */
+ l.mfspr r2,r0,SPR_EEAR_BASE
+ /*
+ * pmd = (pmd_t *)(current_pgd + pgd_index(daddr));
+ */
+ GET_CURRENT_PGD(r3,r5) // r3 is current_pgd, r5 is temp
+ l.srli r4,r2,0x18 // >> PAGE_SHIFT + (PAGE_SHIFT - 2)
+ l.slli r4,r4,0x2 // to get address << 2
+ l.add r5,r4,r3 // r4 is pgd_index(daddr)
+ /*
+ * if (pmd_none(*pmd))
+ * goto pmd_none:
+ */
+ tophys (r4,r5)
+ l.lwz r3,0x0(r4) // get *pmd value
+ l.sfne r3,r0
+ l.bnf d_pmd_none
+ l.andi r3,r3,~PAGE_MASK //0x1fff // ~PAGE_MASK
+ /*
+ * if (pmd_bad(*pmd))
+ * pmd_clear(pmd)
+ * goto pmd_bad:
+ */
+// l.sfeq r3,r0 // check *pmd value
+// l.bf d_pmd_good
+ l.addi r3,r0,0xffffe000 // PAGE_MASK
+// l.j d_pmd_bad
+// l.sw 0x0(r4),r0 // clear pmd
+d_pmd_good:
+ /*
+ * pte = *pte_offset(pmd, daddr);
+ */
+ l.lwz r4,0x0(r4) // get **pmd value
+ l.and r4,r4,r3 // & PAGE_MASK
+ l.srli r5,r2,0xd // >> PAGE_SHIFT, r2 == EEAR
+ l.andi r3,r5,0x7ff // (1UL << PAGE_SHIFT - 2) - 1
+ l.slli r3,r3,0x2 // to get address << 2
+ l.add r3,r3,r4
+ l.lwz r2,0x0(r3) // this is pte at last
+ /*
+ * if (!pte_present(pte))
+ */
+ l.andi r4,r2,0x1
+ l.sfne r4,r0 // is pte present
+ l.bnf d_pte_not_present
+ l.addi r3,r0,0xffffe3fa // PAGE_MASK | DTLB_UP_CONVERT_MASK
+ /*
+ * fill DTLB TR register
+ */
+ l.and r4,r2,r3 // apply the mask
+ // Determine number of DMMU sets
+ l.mfspr r6, r0, SPR_DMMUCFGR
+ l.andi r6, r6, SPR_DMMUCFGR_NTS
+ l.srli r6, r6, SPR_DMMUCFGR_NTS_OFF
+ l.ori r3, r0, 0x1
+ l.sll r3, r3, r6 // r3 = number DMMU sets DMMUCFGR
+ l.addi r6, r3, -1 // r6 = nsets mask
+ l.and r5, r5, r6 // calc offset: & (NUM_TLB_ENTRIES-1)
+ //NUM_TLB_ENTRIES
+ l.mtspr r5,r4,SPR_DTLBTR_BASE(0)
+ /*
+ * fill DTLB MR register
+ */
+ l.mfspr r2,r0,SPR_EEAR_BASE
+ l.addi r3,r0,0xffffe000 // PAGE_MASK
+ l.and r4,r2,r3 // apply PAGE_MASK to EA (__PHX__ do we really need this?)
+ l.ori r4,r4,0x1 // set hardware valid bit: DTBL_MR entry
+ l.mtspr r5,r4,SPR_DTLBMR_BASE(0)
+
+ EXCEPTION_LOAD_GPR2
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR6
+ l.rfe
+d_pmd_bad:
+ l.nop 1
+ EXCEPTION_LOAD_GPR2
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR6
+ l.rfe
+d_pmd_none:
+d_pte_not_present:
+ EXCEPTION_LOAD_GPR2
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR6
+ l.j _dispatch_do_dpage_fault
+ l.nop
+
+/* ==============================================[ ITLB miss handler ]=== */
+ENTRY(itlb_miss_handler)
+ EXCEPTION_STORE_GPR2
+ EXCEPTION_STORE_GPR3
+ EXCEPTION_STORE_GPR4
+ EXCEPTION_STORE_GPR5
+ EXCEPTION_STORE_GPR6
+ /*
+ * get EA of the miss
+ */
+ l.mfspr r2,r0,SPR_EEAR_BASE
+
+ /*
+ * pmd = (pmd_t *)(current_pgd + pgd_index(daddr));
+ *
+ */
+ GET_CURRENT_PGD(r3,r5) // r3 is current_pgd, r5 is temp
+ l.srli r4,r2,0x18 // >> PAGE_SHIFT + (PAGE_SHIFT - 2)
+ l.slli r4,r4,0x2 // to get address << 2
+ l.add r5,r4,r3 // r4 is pgd_index(daddr)
+ /*
+ * if (pmd_none(*pmd))
+ * goto pmd_none:
+ */
+ tophys (r4,r5)
+ l.lwz r3,0x0(r4) // get *pmd value
+ l.sfne r3,r0
+ l.bnf i_pmd_none
+ l.andi r3,r3,0x1fff // ~PAGE_MASK
+ /*
+ * if (pmd_bad(*pmd))
+ * pmd_clear(pmd)
+ * goto pmd_bad:
+ */
+
+// l.sfeq r3,r0 // check *pmd value
+// l.bf i_pmd_good
+ l.addi r3,r0,0xffffe000 // PAGE_MASK
+// l.j i_pmd_bad
+// l.sw 0x0(r4),r0 // clear pmd
+
+i_pmd_good:
+ /*
+ * pte = *pte_offset(pmd, iaddr);
+ *
+ */
+ l.lwz r4,0x0(r4) // get **pmd value
+ l.and r4,r4,r3 // & PAGE_MASK
+ l.srli r5,r2,0xd // >> PAGE_SHIFT, r2 == EEAR
+ l.andi r3,r5,0x7ff // (1UL << PAGE_SHIFT - 2) - 1
+ l.slli r3,r3,0x2 // to get address << 2
+ l.add r3,r3,r4
+ l.lwz r2,0x0(r3) // this is pte at last
+ /*
+ * if (!pte_present(pte))
+ *
+ */
+ l.andi r4,r2,0x1
+ l.sfne r4,r0 // is pte present
+ l.bnf i_pte_not_present
+ l.addi r3,r0,0xffffe03a // PAGE_MASK | ITLB_UP_CONVERT_MASK
+ /*
+ * fill ITLB TR register
+ */
+ l.and r4,r2,r3 // apply the mask
+ l.andi r3,r2,0x7c0 // _PAGE_EXEC | _PAGE_SRE | _PAGE_SWE | _PAGE_URE | _PAGE_UWE
+// l.andi r3,r2,0x400 // _PAGE_EXEC
+ l.sfeq r3,r0
+ l.bf itlb_tr_fill //_workaround
+ // Determine number of IMMU sets
+ l.mfspr r6, r0, SPR_IMMUCFGR
+ l.andi r6, r6, SPR_IMMUCFGR_NTS
+ l.srli r6, r6, SPR_IMMUCFGR_NTS_OFF
+ l.ori r3, r0, 0x1
+ l.sll r3, r3, r6 // r3 = number IMMU sets IMMUCFGR
+ l.addi r6, r3, -1 // r6 = nsets mask
+ l.and r5, r5, r6 // calc offset: & (NUM_TLB_ENTRIES-1)
+
+/*
+ * __PHX__ :: fixme
+ * we should not just blindly set executable flags,
+ * but it does help with ping. the clean way would be to find out
+ * (and fix it) why stack doesn't have execution permissions
+ */
+
+itlb_tr_fill_workaround:
+ l.ori r4,r4,0xc0 // | (SPR_ITLBTR_UXE | ITLBTR_SXE)
+itlb_tr_fill:
+ l.mtspr r5,r4,SPR_ITLBTR_BASE(0)
+ /*
+ * fill DTLB MR register
+ */
+ l.mfspr r2,r0,SPR_EEAR_BASE
+ l.addi r3,r0,0xffffe000 // PAGE_MASK
+ l.and r4,r2,r3 // apply PAGE_MASK to EA (__PHX__ do we really need this?)
+ l.ori r4,r4,0x1 // set hardware valid bit: DTBL_MR entry
+ l.mtspr r5,r4,SPR_ITLBMR_BASE(0)
+
+ EXCEPTION_LOAD_GPR2
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR6
+ l.rfe
+
+i_pmd_bad:
+ l.nop 1
+ EXCEPTION_LOAD_GPR2
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR6
+ l.rfe
+i_pmd_none:
+i_pte_not_present:
+ EXCEPTION_LOAD_GPR2
+ EXCEPTION_LOAD_GPR3
+ EXCEPTION_LOAD_GPR4
+ EXCEPTION_LOAD_GPR5
+ EXCEPTION_LOAD_GPR6
+ l.j _dispatch_do_ipage_fault
+ l.nop
+
+/* ==============================================[ boot tlb handlers ]=== */
+
+
+/* =================================================[ debugging aids ]=== */
+
+ .align 64
+_immu_trampoline:
+ .space 64
+_immu_trampoline_top:
+
+#define TRAMP_SLOT_0 (0x0)
+#define TRAMP_SLOT_1 (0x4)
+#define TRAMP_SLOT_2 (0x8)
+#define TRAMP_SLOT_3 (0xc)
+#define TRAMP_SLOT_4 (0x10)
+#define TRAMP_SLOT_5 (0x14)
+#define TRAMP_FRAME_SIZE (0x18)
+
+ENTRY(_immu_trampoline_workaround)
+ // r2 EEA
+ // r6 is physical EEA
+ tophys(r6,r2)
+
+ LOAD_SYMBOL_2_GPR(r5,_immu_trampoline)
+ tophys (r3,r5) // r3 is trampoline (physical)
+
+ LOAD_SYMBOL_2_GPR(r4,0x15000000)
+ l.sw TRAMP_SLOT_0(r3),r4
+ l.sw TRAMP_SLOT_1(r3),r4
+ l.sw TRAMP_SLOT_4(r3),r4
+ l.sw TRAMP_SLOT_5(r3),r4
+
+ // EPC = EEA - 0x4
+ l.lwz r4,0x0(r6) // load op @ EEA + 0x0 (fc address)
+ l.sw TRAMP_SLOT_3(r3),r4 // store it to _immu_trampoline_data
+ l.lwz r4,-0x4(r6) // load op @ EEA - 0x4 (f8 address)
+ l.sw TRAMP_SLOT_2(r3),r4 // store it to _immu_trampoline_data
+
+ l.srli r5,r4,26 // check opcode for write access
+ l.sfeqi r5,0 // l.j
+ l.bf 0f
+ l.sfeqi r5,0x11 // l.jr
+ l.bf 1f
+ l.sfeqi r5,1 // l.jal
+ l.bf 2f
+ l.sfeqi r5,0x12 // l.jalr
+ l.bf 3f
+ l.sfeqi r5,3 // l.bnf
+ l.bf 4f
+ l.sfeqi r5,4 // l.bf
+ l.bf 5f
+99:
+ l.nop
+ l.j 99b // should never happen
+ l.nop 1
+
+ // r2 is EEA
+ // r3 is trampoline address (physical)
+ // r4 is instruction
+ // r6 is physical(EEA)
+ //
+ // r5
+
+2: // l.jal
+
+ /* 19 20 aa aa l.movhi r9,0xaaaa
+ * a9 29 bb bb l.ori r9,0xbbbb
+ *
+ * where 0xaaaabbbb is EEA + 0x4 shifted right 2
+ */
+
+ l.addi r6,r2,0x4 // this is 0xaaaabbbb
+
+ // l.movhi r9,0xaaaa
+ l.ori r5,r0,0x1920 // 0x1920 == l.movhi r9
+ l.sh (TRAMP_SLOT_0+0x0)(r3),r5
+ l.srli r5,r6,16
+ l.sh (TRAMP_SLOT_0+0x2)(r3),r5
+
+ // l.ori r9,0xbbbb
+ l.ori r5,r0,0xa929 // 0xa929 == l.ori r9
+ l.sh (TRAMP_SLOT_1+0x0)(r3),r5
+ l.andi r5,r6,0xffff
+ l.sh (TRAMP_SLOT_1+0x2)(r3),r5
+
+ /* falthrough, need to set up new jump offset */
+
+
+0: // l.j
+ l.slli r6,r4,6 // original offset shifted left 6 - 2
+// l.srli r6,r6,6 // original offset shifted right 2
+
+ l.slli r4,r2,4 // old jump position: EEA shifted left 4
+// l.srli r4,r4,6 // old jump position: shifted right 2
+
+ l.addi r5,r3,0xc // new jump position (physical)
+ l.slli r5,r5,4 // new jump position: shifted left 4
+
+ // calculate new jump offset
+ // new_off = old_off + (old_jump - new_jump)
+
+ l.sub r5,r4,r5 // old_jump - new_jump
+ l.add r5,r6,r5 // orig_off + (old_jump - new_jump)
+ l.srli r5,r5,6 // new offset shifted right 2
+
+ // r5 is new jump offset
+ // l.j has opcode 0x0...
+ l.sw TRAMP_SLOT_2(r3),r5 // write it back
+
+ l.j trampoline_out
+ l.nop
+
+/* ----------------------------- */
+
+3: // l.jalr
+
+ /* 19 20 aa aa l.movhi r9,0xaaaa
+ * a9 29 bb bb l.ori r9,0xbbbb
+ *
+ * where 0xaaaabbbb is EEA + 0x4 shifted right 2
+ */
+
+ l.addi r6,r2,0x4 // this is 0xaaaabbbb
+
+ // l.movhi r9,0xaaaa
+ l.ori r5,r0,0x1920 // 0x1920 == l.movhi r9
+ l.sh (TRAMP_SLOT_0+0x0)(r3),r5
+ l.srli r5,r6,16
+ l.sh (TRAMP_SLOT_0+0x2)(r3),r5
+
+ // l.ori r9,0xbbbb
+ l.ori r5,r0,0xa929 // 0xa929 == l.ori r9
+ l.sh (TRAMP_SLOT_1+0x0)(r3),r5
+ l.andi r5,r6,0xffff
+ l.sh (TRAMP_SLOT_1+0x2)(r3),r5
+
+ l.lhz r5,(TRAMP_SLOT_2+0x0)(r3) // load hi part of jump instruction
+ l.andi r5,r5,0x3ff // clear out opcode part
+ l.ori r5,r5,0x4400 // opcode changed from l.jalr -> l.jr
+ l.sh (TRAMP_SLOT_2+0x0)(r3),r5 // write it back
+
+ /* falthrough */
+
+1: // l.jr
+ l.j trampoline_out
+ l.nop
+
+/* ----------------------------- */
+
+4: // l.bnf
+5: // l.bf
+ l.slli r6,r4,6 // original offset shifted left 6 - 2
+// l.srli r6,r6,6 // original offset shifted right 2
+
+ l.slli r4,r2,4 // old jump position: EEA shifted left 4
+// l.srli r4,r4,6 // old jump position: shifted right 2
+
+ l.addi r5,r3,0xc // new jump position (physical)
+ l.slli r5,r5,4 // new jump position: shifted left 4
+
+ // calculate new jump offset
+ // new_off = old_off + (old_jump - new_jump)
+
+ l.add r6,r6,r4 // (orig_off + old_jump)
+ l.sub r6,r6,r5 // (orig_off + old_jump) - new_jump
+ l.srli r6,r6,6 // new offset shifted right 2
+
+ // r6 is new jump offset
+ l.lwz r4,(TRAMP_SLOT_2+0x0)(r3) // load jump instruction
+ l.srli r4,r4,16
+ l.andi r4,r4,0xfc00 // get opcode part
+ l.slli r4,r4,16
+ l.or r6,r4,r6 // l.b(n)f new offset
+ l.sw TRAMP_SLOT_2(r3),r6 // write it back
+
+ /* we need to add l.j to EEA + 0x8 */
+ tophys (r4,r2) // may not be needed (due to shifts down_
+ l.addi r4,r4,(0x8 - 0x8) // jump target = r2 + 0x8 (compensate for 0x8)
+ // jump position = r5 + 0x8 (0x8 compensated)
+ l.sub r4,r4,r5 // jump offset = target - new_position + 0x8
+
+ l.slli r4,r4,4 // the amount of info in imediate of jump
+ l.srli r4,r4,6 // jump instruction with offset
+ l.sw TRAMP_SLOT_4(r3),r4 // write it to 4th slot
+
+ /* fallthrough */
+
+trampoline_out:
+ // set up new EPC to point to our trampoline code
+ LOAD_SYMBOL_2_GPR(r5,_immu_trampoline)
+ l.mtspr r0,r5,SPR_EPCR_BASE
+
+ // immu_trampoline is (4x) CACHE_LINE aligned
+ // and only 6 instructions long,
+ // so we need to invalidate only 2 lines
+
+ /* Establish cache block size
+ If BS=0, 16;
+ If BS=1, 32;
+ r14 contain block size
+ */
+ l.mfspr r21,r0,SPR_ICCFGR
+ l.andi r21,r21,SPR_ICCFGR_CBS
+ l.srli r21,r21,7
+ l.ori r23,r0,16
+ l.sll r14,r23,r21
+
+ l.mtspr r0,r5,SPR_ICBIR
+ l.add r5,r5,r14
+ l.mtspr r0,r5,SPR_ICBIR
+
+ l.jr r9
+ l.nop
+
+
+/*
+ * DSCR: prints a string referenced by r3.
+ *
+ * PRMS: r3 - address of the first character of null
+ * terminated string to be printed
+ *
+ * PREQ: UART at UART_BASE_ADD has to be initialized
+ *
+ * POST: caller should be aware that r3, r9 are changed
+ */
+ENTRY(_emergency_print)
+ EMERGENCY_PRINT_STORE_GPR4
+ EMERGENCY_PRINT_STORE_GPR5
+ EMERGENCY_PRINT_STORE_GPR6
+ EMERGENCY_PRINT_STORE_GPR7
+2:
+ l.lbz r7,0(r3)
+ l.sfeq r7,r0
+ l.bf 9f
+ l.nop
+
+// putc:
+ l.movhi r4,hi(UART_BASE_ADD)
+
+ l.addi r6,r0,0x20
+1: l.lbz r5,5(r4)
+ l.andi r5,r5,0x20
+ l.sfeq r5,r6
+ l.bnf 1b
+ l.nop
+
+ l.sb 0(r4),r7
+
+ l.addi r6,r0,0x60
+1: l.lbz r5,5(r4)
+ l.andi r5,r5,0x60
+ l.sfeq r5,r6
+ l.bnf 1b
+ l.nop
+
+ /* next character */
+ l.j 2b
+ l.addi r3,r3,0x1
+
+9:
+ EMERGENCY_PRINT_LOAD_GPR7
+ EMERGENCY_PRINT_LOAD_GPR6
+ EMERGENCY_PRINT_LOAD_GPR5
+ EMERGENCY_PRINT_LOAD_GPR4
+ l.jr r9
+ l.nop
+
+ENTRY(_emergency_print_nr)
+ EMERGENCY_PRINT_STORE_GPR4
+ EMERGENCY_PRINT_STORE_GPR5
+ EMERGENCY_PRINT_STORE_GPR6
+ EMERGENCY_PRINT_STORE_GPR7
+ EMERGENCY_PRINT_STORE_GPR8
+
+ l.addi r8,r0,32 // shift register
+
+1: /* remove leading zeros */
+ l.addi r8,r8,-0x4
+ l.srl r7,r3,r8
+ l.andi r7,r7,0xf
+
+ /* don't skip the last zero if number == 0x0 */
+ l.sfeqi r8,0x4
+ l.bf 2f
+ l.nop
+
+ l.sfeq r7,r0
+ l.bf 1b
+ l.nop
+
+2:
+ l.srl r7,r3,r8
+
+ l.andi r7,r7,0xf
+ l.sflts r8,r0
+ l.bf 9f
+
+ l.sfgtui r7,0x9
+ l.bnf 8f
+ l.nop
+ l.addi r7,r7,0x27
+
+8:
+ l.addi r7,r7,0x30
+// putc:
+ l.movhi r4,hi(UART_BASE_ADD)
+
+ l.addi r6,r0,0x20
+1: l.lbz r5,5(r4)
+ l.andi r5,r5,0x20
+ l.sfeq r5,r6
+ l.bnf 1b
+ l.nop
+
+ l.sb 0(r4),r7
+
+ l.addi r6,r0,0x60
+1: l.lbz r5,5(r4)
+ l.andi r5,r5,0x60
+ l.sfeq r5,r6
+ l.bnf 1b
+ l.nop
+
+ /* next character */
+ l.j 2b
+ l.addi r8,r8,-0x4
+
+9:
+ EMERGENCY_PRINT_LOAD_GPR8
+ EMERGENCY_PRINT_LOAD_GPR7
+ EMERGENCY_PRINT_LOAD_GPR6
+ EMERGENCY_PRINT_LOAD_GPR5
+ EMERGENCY_PRINT_LOAD_GPR4
+ l.jr r9
+ l.nop
+
+
+/*
+ * This should be used for debugging only.
+ * It messes up the Linux early serial output
+ * somehow, so use it sparingly and essentially
+ * only if you need to debug something that goes wrong
+ * before Linux gets the early serial going.
+ *
+ * Furthermore, you'll have to make sure you set the
+ * UART_DEVISOR correctly according to the system
+ * clock rate.
+ *
+ *
+ */
+
+
+
+#define SYS_CLK 20000000
+//#define SYS_CLK 1843200
+#define OR32_CONSOLE_BAUD 115200
+#define UART_DIVISOR SYS_CLK/(16*OR32_CONSOLE_BAUD)
+
+ENTRY(_early_uart_init)
+ l.movhi r3,hi(UART_BASE_ADD)
+
+ l.addi r4,r0,0x7
+ l.sb 0x2(r3),r4
+
+ l.addi r4,r0,0x0
+ l.sb 0x1(r3),r4
+
+ l.addi r4,r0,0x3
+ l.sb 0x3(r3),r4
+
+ l.lbz r5,3(r3)
+ l.ori r4,r5,0x80
+ l.sb 0x3(r3),r4
+ l.addi r4,r0,((UART_DIVISOR>>8) & 0x000000ff)
+ l.sb UART_DLM(r3),r4
+ l.addi r4,r0,((UART_DIVISOR) & 0x000000ff)
+ l.sb UART_DLL(r3),r4
+ l.sb 0x3(r3),r5
+
+ l.jr r9
+ l.nop
+
+_string_copying_linux:
+ .string "\n\n\n\n\n\rCopying Linux... \0"
+
+_string_ok_booting:
+ .string "Ok, booting the kernel.\n\r\0"
+
+_string_unhandled_exception:
+ .string "\n\rRunarunaround: Unhandled exception 0x\0"
+
+_string_epc_prefix:
+ .string ": EPC=0x\0"
+
+_string_nl:
+ .string "\n\r\0"
+
+ .global _string_esr_irq_bug
+_string_esr_irq_bug:
+ .string "\n\rESR external interrupt bug, for details look into entry.S\n\r\0"
+
+
+
+/* ========================================[ page aligned structures ]=== */
+
+/*
+ * .data section should be page aligned
+ * (look into arch/or32/kernel/vmlinux.lds)
+ */
+ .section .data,"aw"
+ .align 8192
+ .global empty_zero_page
+empty_zero_page:
+ .space 8192
+
+ .global swapper_pg_dir
+swapper_pg_dir:
+ .space 8192
+
+ .global _unhandled_stack
+_unhandled_stack:
+ .space 8192
+_unhandled_stack_top:
+
+/* ============================================================[ EOF ]=== */
diff --git a/arch/openrisc/kernel/init_task.c b/arch/openrisc/kernel/init_task.c
new file mode 100644
index 0000000..d05c338
--- /dev/null
+++ b/arch/openrisc/kernel/init_task.c
@@ -0,0 +1,41 @@
+/*
+ * OpenRISC init_task.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/init_task.h>
+#include <linux/mqueue.h>
+
+static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
+static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+
+/*
+ * Initial thread structure.
+ *
+ * We need to make sure that this is THREAD_SIZE aligned due to the
+ * way process stacks are handled. This is done by having a special
+ * "init_task" linker map entry..
+ */
+union thread_union init_thread_union __init_task_data =
+ { INIT_THREAD_INFO(init_task) };
+
+/*
+ * Initial task structure.
+ *
+ * All other task structs will be allocated on slabs in fork.c
+ */
+struct task_struct init_task = INIT_TASK(init_task);
+
+EXPORT_SYMBOL(init_task);
diff --git a/arch/openrisc/kernel/setup.c b/arch/openrisc/kernel/setup.c
new file mode 100644
index 0000000..49342e9
--- /dev/null
+++ b/arch/openrisc/kernel/setup.c
@@ -0,0 +1,387 @@
+/*
+ * OpenRISC setup.c
+ *
+ * Linux architectural port borrowing liberally from similar works of
+ * others. All original copyrights apply as per the original source
+ * declaration.
+ *
+ * Modifications for the OpenRISC architecture:
+ * Copyright (C) 2003 Matjaz Breskvar <***@bsemi.com>
+ * Copyright (C) 2010-2011 Jonas Bonn <***@southpole.se>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This file handles the architecture-dependent parts of initialization
+ */
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/tty.h>
+#include <linux/ioport.h>
+#include <linux/delay.h>
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/seq_file.h>
+#include <linux/serial.h>
+#include <linux/initrd.h>
+#include <linux/of_fdt.h>
+#include <linux/of.h>
+#include <linux/memblock.h>
+#include <linux/device.h>
+#include <linux/of_platform.h>
+
+#include <asm/segment.h>
+#include <asm/system.h>
+#include <asm/pgtable.h>
+#include <asm/types.h>
+#include <asm/setup.h>
+#include <asm/io.h>
+#include <asm/cpuinfo.h>
+#include <asm/delay.h>
+
+#include "vmlinux.h"
+
+char __initdata cmd_line[COMMAND_LINE_SIZE] = CONFIG_CMDLINE;
+
+static unsigned long __init setup_memory(void)
+{
+ unsigned long bootmap_size;
+ unsigned long ram_start_pfn;
+ unsigned long free_ram_start_pfn;
+ unsigned long ram_end_pfn;
+ phys_addr_t memory_start, memory_end;
+ struct memblock_region* region;
+
+ memory_end = memory_start = 0;
+
+ /* Find main memory where is the kernel */
+ for_each_memblock(memory, region) {
+ memory_start = region->base;
+ memory_end = region->base + region->size;
+ printk(KERN_INFO "%s: Memory: 0x%x-0x%x\n", __func__,
+ memory_start, memory_end);
+ }
+
+ if (! memory_end) {
+ panic("No memory!");
+ }
+
+ ram_start_pfn = PFN_UP(memory_start);
+ /* free_ram_start_pfn is first page after kernel */
+ free_ram_start_pfn = PFN_UP(__pa(&_end));
+ ram_end_pfn = PFN_DOWN(memblock_end_of_DRAM());
+
+ max_pfn = ram_end_pfn;
+
+ /*
+ * initialize the boot-time allocator (with low memory only).
+ *
+ * This makes the memory from the end of the kernel to the end of
+ * RAM usable.
+ * init_bootmem sets the global values min_low_pfn, max_low_pfn.
+ */
+ bootmap_size = init_bootmem(free_ram_start_pfn,
+ ram_end_pfn-ram_start_pfn);
+ free_bootmem(PFN_PHYS(free_ram_start_pfn),
+ (ram_end_pfn-free_ram_start_pfn)<< PAGE_SHIFT);
+ reserve_bootmem(PFN_PHYS(free_ram_start_pfn), bootmap_size,
+ BOOTMEM_DEFAULT);
+
+ for_each_memblock(reserved, region) {
+ printk(KERN_INFO "Reserved - 0x%08x-0x%08x\n",
+ (u32) region->base, (u32) region->size);
+ reserve_bootmem(region->base, region->size, BOOTMEM_DEFAULT);
+ }
+
+ return(ram_end_pfn);
+}
+
+
+struct cpuinfo cpuinfo;
+
+static void print_cpuinfo(void) {
+ unsigned long upr = mfspr(SPR_UPR);
+ unsigned long vr = mfspr(SPR_VR);
+ unsigned int version;
+ unsigned int revision;
+
+ version = (vr & SPR_VR_VER) >> 24;
+ revision = (vr & SPR_VR_REV);
+
+ printk(KERN_INFO "CPU: OpenRISC-%x (revision %d) @%d MHz\n",
+ version, revision, cpuinfo.clock_frequency / 1000000);
+
+ if (!(upr & SPR_UPR_UP)) {
+ printk(KERN_INFO "-- no UPR register... unable to detect configuration\n");
+ return;
+ }
+
+ if (upr & SPR_UPR_DCP)
+ printk(KERN_INFO "-- dcache: %4d bytes total, %2d bytes/line, %d way(s)\n",
+ cpuinfo.dcache_size, cpuinfo.dcache_block_size, 1);
+ else
+ printk(KERN_INFO "-- dcache disabled\n");
+ if (upr & SPR_UPR_ICP)
+ printk(KERN_INFO "-- icache: %4d bytes total, %2d bytes/line, %d way(s)\n",
+ cpuinfo.icache_size, cpuinfo.icache_block_size, 1);
+ else
+ printk(KERN_INFO "-- icache disabled\n");
+
+ if (upr & SPR_UPR_DMP)
+ printk(KERN_INFO "-- dmmu: %4d entries, %lu way(s)\n",
+ 1 << ((mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_NTS) >> 2),
+ 1 + (mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_NTW));
+ if (upr & SPR_UPR_IMP)
+ printk(KERN_INFO "-- immu: %4d entries, %lu way(s)\n",
+ 1 << ((mfspr(SPR_IMMUCFGR) & SPR_IMMUCFGR_NTS) >> 2),
+ 1 + (mfspr(SPR_IMMUCFGR) & SPR_IMMUCFGR_NTW));
+
+ printk(KERN_INFO "-- additional features:\n");
+ if (upr & SPR_UPR_DUP)
+ printk(KERN_INFO "-- debug unit\n");
+ if (upr & SPR_UPR_PCUP)
+ printk(KERN_INFO "-- performance counters\n");
+ if (upr & SPR_UPR_PMP)
+ printk(KERN_INFO "-- power management\n");
+ if (upr & SPR_UPR_PICP)
+ printk(KERN_INFO "-- PIC\n");
+ if (upr & SPR_UPR_TTP)
+ printk(KERN_INFO "-- timer\n");
+ if (upr & SPR_UPR_CUP)
+ printk(KERN_INFO "-- custom unit(s)\n");
+}
+
+static inline unsigned int fcpu(struct device_node *cpu, char *n)
+{
+ const int *val;
+ return (val = of_get_property(cpu, n, NULL)) ? *val : 0;
+}
+
+void __init setup_cpuinfo(void)
+{
+ struct device_node *cpu = NULL;
+ unsigned long iccfgr,dccfgr;
+ unsigned long cache_set_size, cache_ways;;
+
+ cpu = (struct device_node *) of_find_compatible_node(NULL,
+ NULL, "opencores,openrisc-1200");
+ if (!cpu) {
+ panic("No compatible CPU found in device tree...\n");
+ }
+
+ iccfgr = mfspr(SPR_ICCFGR);
+ cache_ways = 1 << (iccfgr & SPR_ICCFGR_NCW);
+ cache_set_size = 1 << ((iccfgr & SPR_ICCFGR_NCS) >> 3);
+ cpuinfo.icache_block_size = 16 << ((iccfgr & SPR_ICCFGR_CBS) >> 7);
+ cpuinfo.icache_size = cache_set_size * cache_ways * cpuinfo.icache_block_size;
+
+ dccfgr = mfspr(SPR_DCCFGR);
+ cache_ways = 1 << (dccfgr & SPR_DCCFGR_NCW);
+ cache_set_size = 1 << ((dccfgr & SPR_DCCFGR_NCS) >> 3);
+ cpuinfo.dcache_block_size = 16 << ((dccfgr & SPR_DCCFGR_CBS) >> 7);
+ cpuinfo.dcache_size = cache_set_size * cache_ways * cpuinfo.dcache_block_size;
+
+ cpuinfo.clock_frequency = fcpu(cpu, "clock-frequency");
+
+ of_node_put(cpu);
+
+ print_cpuinfo();
+}
+
+/**
+ * or32_early_setup
+ *
+ * Handles the pointer to the device tree that this kernel is to use
+ * for establishing the available platform devices.
+ *
+ * For now, this is limited to using the built-in device tree. In the future,
+ * it is intended that this function will take a pointer to the device tree
+ * that is potentially built-in, but potentially also passed in by the
+ * bootloader, or discovered by some equally clever means...
+ */
+
+void __init or32_early_setup(void) {
+
+ early_init_devtree((void *) __dtb_start);
+
+ printk(KERN_INFO "Compiled-in FDT at 0x%08x\n",
+ (unsigned int) __dtb_start);
+}
+
+const struct of_device_id openrisc_bus_ids[] = {
+ { .type = "soc", },
+ { .compatible = "soc", },
+ {},
+};
+
+static int __init openrisc_device_probe(void)
+{
+ of_platform_bus_probe(NULL, openrisc_bus_ids, NULL);
+ return 0;
+}
+device_initcall(openrisc_device_probe);
+
+
+static inline unsigned long extract_value_bits(unsigned long reg,
+ short bit_nr, short width)
+{
+ return((reg >> bit_nr) & (0 << width));
+}
+
+static inline unsigned long extract_value(unsigned long reg,
+ unsigned long mask)
+{
+ while (!(mask & 0x1)) {
+ reg = reg >> 1;
+ mask = mask >> 1;
+ }
+ return(mask & reg);
+}
+
+void __init detect_unit_config(unsigned long upr, unsigned long mask,
+ char *text, void (*func)(void))
+{
+ if (text != NULL)
+ printk("%s", text);
+
+ if ( upr & mask ) {
+ if (func != NULL)
+ func();
+ else
+ printk("present\n");
+ }
+ else
+ printk("not present\n");
+}
+
+/*
+ * calibrate_delay
+ *
+ * Lightweight calibrate_delay implementation that calculates loops_per_jiffy
+ * from the clock frequency passed in via the device tree
+ *
+ */
+
+void __cpuinit calibrate_delay(void)
+{
+ const int *val;
+ struct device_node *cpu = NULL;
+ cpu = of_find_compatible_node(NULL, NULL, "opencores,openrisc-1200");
+ val = of_get_property(cpu, "clock-frequency", NULL);
+ if (!val) {
+ panic("no cpu 'clock-frequency' parameter in device tree");
+ }
+ loops_per_jiffy = *val / HZ;
+ pr_cont("%lu.%02lu BogoMIPS (lpj=%lu)\n",
+ loops_per_jiffy/(500000/HZ),
+ (loops_per_jiffy/(5000/HZ)) % 100, loops_per_jiffy);
+}
+
+void __init setup_arch(char **cmdline_p)
+{
+ unsigned long max_low_pfn;
+
+ unflatten_device_tree();
+
+ setup_cpuinfo();
+
+ /* process 1's initial memory region is the kernel code/data */
+ init_mm.start_code = (unsigned long) &_stext;
+ init_mm.end_code = (unsigned long) &_etext;
+ init_mm.end_data = (unsigned long) &_edata;
+ init_mm.brk = (unsigned long) &_end;
+
+#ifdef CONFIG_BLK_DEV_INITRD
+ initrd_start = (unsigned long)&__initrd_start;
+ initrd_end = (unsigned long)&__initrd_end;
+ if (initrd_start == initrd_end) {
+ initrd_start = 0;
+ initrd_end = 0;
+ }
+ initrd_below_start_ok = 1;
+#endif
+
+ /* setup bootmem allocator */
+ max_low_pfn = setup_memory();
+
+ /* paging_init() sets up the MMU and marks all pages as reserved */
+ paging_init();
+
+#if defined(CONFIG_VT) && defined(CONFIG_DUMMY_CONSOLE)
+ if(!conswitchp)
+ conswitchp = &dummy_con;
+#endif
+
+ *cmdline_p = cmd_line;
+
+ printk(KERN_INFO "OpenRISC Linux -- http://openrisc.net\n");
+}
+
+static int show_cpuinfo(struct seq_file *m, void *v)
+{
+ unsigned long vr;
+ int version, revision;
+
+ vr = mfspr(SPR_VR);
+ version = (vr & SPR_VR_VER) >> 24;
+ revision = vr & SPR_VR_REV;
+
+ return seq_printf(m,
+ "cpu\t\t: OpenRISC-%x\n"
+ "revision\t: %d\n"
+ "frequency\t: %ld\n"
+ "dcache size\t: %d bytes\n"
+ "dcache block size\t: %d bytes\n"
+ "icache size\t: %d bytes\n"
+ "icache block size\t: %d bytes\n"
+ "immu\t\t: %d entries, %lu ways\n"
+ "dmmu\t\t: %d entries, %lu ways\n"
+ "bogomips\t: %lu.%02lu\n",
+
+ version,
+ revision,
+ loops_per_jiffy * HZ,
+ cpuinfo.dcache_size,
+ cpuinfo.dcache_block_size,
+ cpuinfo.icache_size,
+ cpuinfo.icache_block_size,
+ 1 << ((mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_NTS) >> 2),
+ 1 + (mfspr(SPR_DMMUCFGR) & SPR_DMMUCFGR_NTW),
+ 1 << ((mfspr(SPR_IMMUCFGR) & SPR_IMMUCFGR_NTS) >> 2),
+ 1 + (mfspr(SPR_IMMUCFGR) & SPR_IMMUCFGR_NTW),
+ (loops_per_jiffy * HZ) / 500000,
+ ((loops_per_jiffy * HZ) / 5000) % 100);
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+ /* We only have one CPU... */
+ return *pos < 1 ? (void *)1 : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+ ++*pos;
+ return NULL;
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+
+struct seq_operations cpuinfo_op = {
+ .start = c_start,
+ .next = c_next,
+ .stop = c_stop,
+ .show = show_cpuinfo,
+};
--
1.7.4.1
Arnd Bergmann
2011-07-05 15:56:33 UTC
Permalink
Post by Jonas Bonn
This is version 2 of the OpenRISC architecture patch set. Changes
- general cleanups, removing unused code and variables, taking care of
compiler warnings, etc.
- cleanups for all the errors spotted by checkpatch.pl
- changes in response to the feedback received for version 1 of the patch set
- simplification of the fixmap code
- simplification of the DMA code
- calculate loops_per_jiffy from cpu frequency passed in via devicetree
- rework io.h to use more generic code
- rework the delay code, making it more generic
- also, the module code has been reworked, making it more generic; this is
not included in this patch as the generic parts are still pending review
I think this is basically ready for inclusion in 3.1. There are a few details
that you can still improve, but I see no reason to hold up the merge into
mainline. After you have addressed the comments that are still outstanding
at this time, please add "Reviewed-by: Arnd Bergmann <***@arndb.de>"
to all the patches.

The most important part to work out are the device tree bindings, as discussed
for patch 2. Once you have a version of that patch that everyone is happy
with, please post the entire series including the patches that address common
code once more.
Post by Jonas Bonn
git://openrisc.net/jonas/linux
IMO it would be best if Stephen added that branch to linux-next right
away, so we get some test exposure on your common code changes before
the merge window.

Arnd
Jonas Bonn
2011-07-06 15:51:31 UTC
Permalink
Post by Arnd Bergmann
IMO it would be best if Stephen added that branch to linux-next right
away, so we get some test exposure on your common code changes before
the merge window.
Hi Stephen,

I'd like to request the inclusion of the OpenRISC architecture kernel
tree into the linux-next repository. The relevant branch is the
'for-upstream' branch of the repository at:

git://openrisc.net/jonas/linux

Thanks,
Jonas
Stephen Rothwell
2011-07-16 03:55:37 UTC
Permalink
Hi Jonas,
Post by Jonas Bonn
Post by Arnd Bergmann
IMO it would be best if Stephen added that branch to linux-next right
away, so we get some test exposure on your common code changes before
the merge window.
I'd like to request the inclusion of the OpenRISC architecture kernel
tree into the linux-next repository. The relevant branch is the
git://openrisc.net/jonas/linux
I will include that tree in the next linux-next release.

Thanks for adding your subsystem tree as a participant of linux-next. As
you may know, this is not a judgment of your code. The purpose of
linux-next is for integration testing and to lower the impact of
conflicts between subsystems in the next merge window.

You will need to ensure that the patches/commits in your tree/series have
been:
* submitted under GPL v2 (or later) and include the Contributor's
Signed-off-by,
* posted to the relevant mailing list,
* reviewed by you (or another maintainer of your subsystem tree),
* successfully unit tested, and
* destined for the current or next Linux merge window.

Basically, this should be just what you would send to Linus (or ask him
to fetch). It is allowed to be rebased if you deem it necessary.
--
Cheers,
Stephen Rothwell
***@canb.auug.org.au

Legal Stuff:
By participating in linux-next, your subsystem tree contributions are
public and will be included in the linux-next trees. You may be sent
e-mail messages indicating errors or other issues when the
patches/commits from your subsystem tree are merged and tested in
linux-next. These messages may also be cross-posted to the linux-next
mailing list, the linux-kernel mailing list, etc. The linux-next tree
project and IBM (my employer) make no warranties regarding the linux-next
project, the testing procedures, the results, the e-mails, etc. If you
don't agree to these ground rules, let me know and I'll remove your tree
from participation in linux-next.
Stephen Rothwell
2011-07-16 04:01:17 UTC
Permalink
Hi Arnd,
Ok, I guess that means you also won't be able to merge the arm-soc tree
before the merge window, right?
I meant to ask you to include the master branch of
git://git.kernel.org/pub/scm/linux/kernel/git/arm/linux-arm-soc.git
but we first had to move it into the right location. Can you add
that when you get back?
I will add this tree in the next linux-next release.

Thanks for adding your subsystem tree as a participant of linux-next. As
you may know, this is not a judgment of your code. The purpose of
linux-next is for integration testing and to lower the impact of
conflicts between subsystems in the next merge window.

You will need to ensure that the patches/commits in your tree/series have
been:
* submitted under GPL v2 (or later) and include the Contributor's
Signed-off-by,
* posted to the relevant mailing list,
* reviewed by you (or another maintainer of your subsystem tree),
* successfully unit tested, and
* destined for the current or next Linux merge window.

Basically, this should be just what you would send to Linus (or ask him
to fetch). It is allowed to be rebased if you deem it necessary.
--
Cheers,
Stephen Rothwell
***@canb.auug.org.au

Legal Stuff:
By participating in linux-next, your subsystem tree contributions are
public and will be included in the linux-next trees. You may be sent
e-mail messages indicating errors or other issues when the
patches/commits from your subsystem tree are merged and tested in
linux-next. These messages may also be cross-posted to the linux-next
mailing list, the linux-kernel mailing list, etc. The linux-next tree
project and IBM (my employer) make no warranties regarding the linux-next
project, the testing procedures, the results, the e-mails, etc. If you
don't agree to these ground rules, let me know and I'll remove your tree
from participation in linux-next.
Loading...