Discussion:
[PATCH v5, part4 00/41] Simplify mem_init() implementations and kill num_physpages
Jiang Liu
2013-05-08 15:50:57 UTC
Permalink
The original goal of this patchset is to fix the bug reported by
https://bugzilla.kernel.org/show_bug.cgi?id=53501
Now it has also been expanded to reduce common code used by memory
initializion.

This is the last part, previous three patch sets could be accessed at:
http://marc.info/?l=linux-mm&m=136289696323825&w=2
http://marc.info/?l=linux-mm&m=136290291524901&w=2
http://marc.info/?l=linux-mm&m=136525653817198&w=2

This patchset applies to
git://git.cmpxchg.org/linux-mmotm.git v3.9-rc8-mmotm-2013-04-25-16-24

You may access the patch seriea at:
git://github.com/jiangliu/linux.git mem_init_v5

V3->V4:
1) rebase to git://git.cmpxchg.org/linux-mmotm.git
2) add support for ARC and metag architectures
3) minor code refinement

V4->V5:
1) rebase to v3.9-rc8-mmotm-2013-04-25-16-24
2) address comments from previous round of review

Patch 1-7:
1) add comments for global variables exported by vmlinux.lds
2) normalize global variables exported by vmlinux.lds
Patch 8:
Introduce helper functions mem_init_print_info() and
get_num_physpages()
Patch 9:
Avoid using global variable num_physpages at runtime
Patch 10:
Don't update num_physpages in memory_hotplug.c
Patch 11-40:
Modify arch mm initialization code to:
1) Simplify mem_init() by using mem_init_print_info()
2) Prepare for killing global variable num_physpages
Patch 41:
Kill the global variable num_physpages

With all patches applied, mem_init(), free_initmem(), free_initrd_mem()
could be as simple as below. This patch series has reduced about 1.2K
lines of code in total.

#ifndef CONFIG_DISCONTIGMEM
void __init
mem_init(void)
{
max_mapnr = max_low_pfn;
free_all_bootmem();
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);

mem_init_print_info(NULL);
}
#endif /* CONFIG_DISCONTIGMEM */

void
free_initmem(void)
{
free_initmem_default(-1);
}

#ifdef CONFIG_BLK_DEV_INITRD
void
free_initrd_mem(unsigned long start, unsigned long end)
{
free_reserved_area(start, end, -1, "initrd");
}
#endif

Due to hardware resource limitations, I have only tested this on x86_64.
And the messages reported on an x86_64 system are:
Log message before applying patches:
Memory: 7745676k/8910848k available (6934k kernel code, 836024k absent, 329148k reserved, 6343k data, 1012k init)
Log message after applying patches:
Memory: 7744624K/8074824K available (6969K kernel code, 1011K data, 2828K rodata, 1016K init, 9640K bss, 330200K reserved)

Any help to review or test these patch series are welcomed!

Regards!
Gerry

Jiang Liu (41):
vmlinux.lds: add comments for global variables and clean up useless
declarations
avr32: normalize global variables exported by vmlinux.lds
c6x: normalize global variables exported by vmlinux.lds
h8300: normalize global variables exported by vmlinux.lds
score: normalize global variables exported by vmlinux.lds
tile: normalize global variables exported by vmlinux.lds
UML: normalize global variables exported by vmlinux.lds
mm: introduce helper function mem_init_print_info() to simplify
mem_init()
mm: use totalram_pages instead of num_physpages at runtime
mm/hotplug: prepare for removing num_physpages
mm/alpha: prepare for removing num_physpages and simplify mem_init()
mm/ARC: prepare for removing num_physpages and simplify mem_init()
mm/ARM: prepare for removing num_physpages and simplify mem_init()
mm/ARM64: prepare for removing num_physpages and simplify mem_init()
mm/AVR32: prepare for removing num_physpages and simplify mem_init()
mm/blackfin: prepare for removing num_physpages and simplify
mem_init()
mm/c6x: prepare for removing num_physpages and simplify mem_init()
mm/cris: prepare for removing num_physpages and simplify mem_init()
mm/frv: prepare for removing num_physpages and simplify mem_init()
mm/h8300: prepare for removing num_physpages and simplify mem_init()
mm/hexagon: prepare for removing num_physpages and simplify
mem_init()
mm/IA64: prepare for removing num_physpages and simplify mem_init()
mm/m32r: prepare for removing num_physpages and simplify mem_init()
mm/m68k: prepare for removing num_physpages and simplify mem_init()
mm/metag: prepare for removing num_physpages and simplify mem_init()
mm/microblaze: prepare for removing num_physpages and simplify
mem_init()
mm/MIPS: prepare for removing num_physpages and simplify mem_init()
mm/mn10300: prepare for removing num_physpages and simplify
mem_init()
mm/openrisc: prepare for removing num_physpages and simplify
mem_init()
mm/PARISC: prepare for removing num_physpages and simplify mem_init()
mm/ppc: prepare for removing num_physpages and simplify mem_init()
mm/s390: prepare for removing num_physpages and simplify mem_init()
mm/score: prepare for removing num_physpages and simplify mem_init()
mm/SH: prepare for removing num_physpages and simplify mem_init()
mm/SPARC: prepare for removing num_physpages and simplify mem_init()
mm/tile: prepare for removing num_physpages and simplify mem_init()
mm/um: prepare for removing num_physpages and simplify mem_init()
mm/unicore32: prepare for removing num_physpages and simplify
mem_init()
mm/x86: prepare for removing num_physpages and simplify mem_init()
mm/xtensa: prepare for removing num_physpages and simplify mem_init()
mm: kill global variable num_physpages

arch/alpha/mm/init.c | 32 ++-------------------
arch/alpha/mm/numa.c | 34 ++--------------------
arch/arc/mm/init.c | 36 ++---------------------
arch/arm/mm/init.c | 47 ++----------------------------
arch/arm64/mm/init.c | 48 ++-----------------------------
arch/avr32/kernel/setup.c | 2 +-
arch/avr32/kernel/vmlinux.lds.S | 4 +--
arch/avr32/mm/init.c | 29 +++----------------
arch/blackfin/mm/init.c | 38 ++++---------------------
arch/c6x/kernel/vmlinux.lds.S | 4 +--
arch/c6x/mm/init.c | 11 +------
arch/cris/mm/init.c | 33 ++-------------------
arch/frv/kernel/setup.c | 13 +++++----
arch/frv/mm/init.c | 49 +++++++++----------------------
arch/h8300/boot/compressed/misc.c | 1 -
arch/h8300/kernel/vmlinux.lds.S | 2 ++
arch/h8300/mm/init.c | 34 ++++++----------------
arch/hexagon/mm/init.c | 4 +--
arch/ia64/mm/contig.c | 11 -------
arch/ia64/mm/discontig.c | 3 --
arch/ia64/mm/init.c | 27 +-----------------
arch/m32r/mm/discontig.c | 6 +---
arch/m32r/mm/init.c | 49 ++++---------------------------
arch/m68k/mm/init.c | 31 ++------------------
arch/metag/mm/init.c | 8 +-----
arch/microblaze/mm/init.c | 51 ++++-----------------------------
arch/mips/mm/init.c | 57 ++++++++++++-------------------------
arch/mips/pci/pci-lantiq.c | 2 +-
arch/mips/sgi-ip27/ip27-memory.c | 21 ++------------
arch/mn10300/mm/init.c | 26 ++---------------
arch/openrisc/mm/init.c | 44 +++-------------------------
arch/parisc/mm/init.c | 46 ++----------------------------
arch/powerpc/mm/mem.c | 56 ++++++++----------------------------
arch/s390/mm/init.c | 17 ++---------
arch/score/kernel/vmlinux.lds.S | 1 +
arch/score/mm/init.c | 26 ++---------------
arch/sh/mm/init.c | 25 +++-------------
arch/sparc/kernel/leon_smp.c | 3 --
arch/sparc/mm/init_32.c | 33 ++-------------------
arch/sparc/mm/init_64.c | 23 ++-------------
arch/tile/include/asm/sections.h | 2 +-
arch/tile/kernel/setup.c | 20 ++++++-------
arch/tile/kernel/vmlinux.lds.S | 4 ++-
arch/tile/mm/init.c | 17 ++---------
arch/um/include/asm/common.lds.S | 1 -
arch/um/kernel/dyn.lds.S | 6 ++--
arch/um/kernel/mem.c | 4 +--
arch/um/kernel/uml.lds.S | 7 +++--
arch/unicore32/mm/init.c | 49 ++-----------------------------
arch/x86/kernel/cpu/amd.c | 2 +-
arch/x86/kernel/setup.c | 2 --
arch/x86/mm/init_32.c | 30 ++-----------------
arch/x86/mm/init_64.c | 20 +------------
arch/x86/mm/numa_32.c | 2 --
arch/xtensa/mm/init.c | 27 ++----------------
fs/fuse/inode.c | 2 +-
include/asm-generic/sections.h | 21 +++++++++++++-
include/linux/mm.h | 13 ++++++++-
kernel/power/snapshot.c | 4 +--
mm/memory.c | 2 --
mm/memory_hotplug.c | 4 ---
mm/nommu.c | 2 --
mm/page_alloc.c | 52 +++++++++++++++++++++++++++++++++
net/ipv4/inet_fragment.c | 2 +-
64 files changed, 256 insertions(+), 1026 deletions(-)
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:50:58 UTC
Permalink
This patch documents global variables exported from vmlinux.lds.
1) Add comments about usage guidelines for global variables exported
from vmlinux.lds.S.
2) Remove unused __initdata_begin[] and __initdata_end[].

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Arnd Bergmann <***@arndb.de>
Cc: Arnd Bergmann <***@arndb.de>
Cc: linux-***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
include/asm-generic/sections.h | 21 ++++++++++++++++++++-
1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/include/asm-generic/sections.h b/include/asm-generic/sections.h
index c1a1216..f1a24b5 100644
--- a/include/asm-generic/sections.h
+++ b/include/asm-generic/sections.h
@@ -3,6 +3,26 @@

/* References to section boundaries */

+/*
+ * Usage guidelines:
+ * _text, _data: architecture specific, don't use them in arch-independent code
+ * [_stext, _etext]: contains .text.* sections, may also contain .rodata.*
+ * and/or .init.* sections
+ * [_sdata, _edata]: contains .data.* sections, may also contain .rodata.*
+ * and/or .init.* sections.
+ * [__start_rodata, __end_rodata]: contains .rodata.* sections
+ * [__init_begin, __init_end]: contains .init.* sections, but .init.text.*
+ * may be out of this range on some architectures.
+ * [_sinittext, _einittext]: contains .init.text.* sections
+ * [__bss_start, __bss_stop]: contains BSS sections
+ *
+ * Following global variables are optional and may be unavailable on some
+ * architectures and/or kernel configurations.
+ * _text, _data
+ * __kprobes_text_start, __kprobes_text_end
+ * __entry_text_start, __entry_text_end
+ * __ctors_start, __ctors_end
+ */
extern char _text[], _stext[], _etext[];
extern char _data[], _sdata[], _edata[];
extern char __bss_start[], __bss_stop[];
@@ -12,7 +32,6 @@ extern char _end[];
extern char __per_cpu_load[], __per_cpu_start[], __per_cpu_end[];
extern char __kprobes_text_start[], __kprobes_text_end[];
extern char __entry_text_start[], __entry_text_end[];
-extern char __initdata_begin[], __initdata_end[];
extern char __start_rodata[], __end_rodata[];

/* Start and end of .ctors section - used for constructor calls. */
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:50:59 UTC
Permalink
Normalize global variables exported by vmlinux.lds to conform usage
guidelines from include/asm-generic/sections.h.

Use _text to mark the start of the kernel image including the head text,
and _stext to mark the start of the .text section.

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Hans-Christian Egtvedt <***@samfundet.no>
Cc: Haavard Skinnemoen <***@gmail.com>
Cc: linux-***@vger.kernel.org
---
arch/avr32/kernel/setup.c | 2 +-
arch/avr32/kernel/vmlinux.lds.S | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/avr32/kernel/setup.c b/arch/avr32/kernel/setup.c
index b4247f4..209ae5a 100644
--- a/arch/avr32/kernel/setup.c
+++ b/arch/avr32/kernel/setup.c
@@ -555,7 +555,7 @@ void __init setup_arch (char **cmdline_p)
{
struct clk *cpu_clk;

- init_mm.start_code = (unsigned long)_text;
+ 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;
diff --git a/arch/avr32/kernel/vmlinux.lds.S b/arch/avr32/kernel/vmlinux.lds.S
index 9cd2bd9..a458917 100644
--- a/arch/avr32/kernel/vmlinux.lds.S
+++ b/arch/avr32/kernel/vmlinux.lds.S
@@ -23,7 +23,7 @@ SECTIONS
{
. = CONFIG_ENTRY_ADDRESS;
.init : AT(ADDR(.init) - LOAD_OFFSET) {
- _stext = .;
+ _text = .;
__init_begin = .;
_sinittext = .;
*(.text.reset)
@@ -46,7 +46,7 @@ SECTIONS

.text : AT(ADDR(.text) - LOAD_OFFSET) {
_evba = .;
- _text = .;
+ _stext = .;
*(.ex.text)
*(.irq.text)
KPROBES_TEXT
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:00 UTC
Permalink
Normalize global variables exported by vmlinux.lds to conform usage
guidelines from include/asm-generic/sections.h.

Use _text to mark the start of the kernel image including the head text,
and _stext to mark the start of the .text section.

This patch also fixes possible bugs due to current address layout that
[__init_begin, __init_end] is a sub-range of [_stext, _etext] and pages
within range [__init_begin, __init_end] will be freed by free_initmem().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Mark Salter <***@redhat.com>
Cc: Aurelien Jacquiot <a-***@ti.com>
Cc: linux-c6x-***@linux-c6x.org
Cc: linux-***@vger.kernel.org
---
arch/c6x/kernel/vmlinux.lds.S | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/c6x/kernel/vmlinux.lds.S b/arch/c6x/kernel/vmlinux.lds.S
index 1d81c4c..279d807 100644
--- a/arch/c6x/kernel/vmlinux.lds.S
+++ b/arch/c6x/kernel/vmlinux.lds.S
@@ -54,16 +54,15 @@ SECTIONS
}

. = ALIGN(PAGE_SIZE);
+ __init_begin = .;
.init :
{
- _stext = .;
_sinittext = .;
HEAD_TEXT
INIT_TEXT
_einittext = .;
}

- __init_begin = _stext;
INIT_DATA_SECTION(16)

PERCPU_SECTION(128)
@@ -74,6 +73,7 @@ SECTIONS
.text :
{
_text = .;
+ _stext = .;
TEXT_TEXT
SCHED_TEXT
LOCK_TEXT
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:01 UTC
Permalink
Generate mandatory global variables __bss_start/__bss_stop in
file vmlinux.lds.

Also remove one unused declaration of _text.

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Yoshinori Sato <***@users.sourceforge.jp>
Cc: Jiang Liu <***@huawei.com>
Cc: linux-***@vger.kernel.org
---
arch/h8300/boot/compressed/misc.c | 1 -
arch/h8300/kernel/vmlinux.lds.S | 2 ++
2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/h8300/boot/compressed/misc.c b/arch/h8300/boot/compressed/misc.c
index 51ab6cb..4a1e3dd 100644
--- a/arch/h8300/boot/compressed/misc.c
+++ b/arch/h8300/boot/compressed/misc.c
@@ -79,7 +79,6 @@ static void error(char *m);

int puts(const char *);

-extern int _text; /* Defined in vmlinux.lds.S */
extern int _end;
static unsigned long free_mem_ptr;
static unsigned long free_mem_end_ptr;
diff --git a/arch/h8300/kernel/vmlinux.lds.S b/arch/h8300/kernel/vmlinux.lds.S
index 03d356d..3253fed 100644
--- a/arch/h8300/kernel/vmlinux.lds.S
+++ b/arch/h8300/kernel/vmlinux.lds.S
@@ -132,10 +132,12 @@ SECTIONS
{
. = ALIGN(0x4) ;
__sbss = . ;
+ ___bss_start = . ;
*(.bss*)
. = ALIGN(0x4) ;
*(COMMON)
. = ALIGN(0x4) ;
+ ___bss_stop = . ;
__ebss = . ;
__end = . ;
__ramstart = .;
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:02 UTC
Permalink
Generate mandatory global variables _sdata in file vmlinux.lds.

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Chen Liqin <***@sunplusct.com>
Cc: Lennox Wu <***@gmail.com>
Cc: linux-***@vger.kernel.org
---
arch/score/kernel/vmlinux.lds.S | 1 +
1 file changed, 1 insertion(+)

diff --git a/arch/score/kernel/vmlinux.lds.S b/arch/score/kernel/vmlinux.lds.S
index eebcbaa..7274b5c 100644
--- a/arch/score/kernel/vmlinux.lds.S
+++ b/arch/score/kernel/vmlinux.lds.S
@@ -49,6 +49,7 @@ SECTIONS
}

. = ALIGN(16);
+ _sdata = .; /* Start of data section */
RODATA

EXCEPTION_TABLE(16)
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:03 UTC
Permalink
Normalize global variables exported by vmlinux.lds to conform usage
guidelines from include/asm-generic/sections.h.

1) Use _text to mark the start of the kernel image including the head
text, and _stext to mark the start of the .text section.
2) Export mandatory global variables __init_begin and __init_end.

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Chris Metcalf <***@tilera.com>
Cc: Rusty Russell <***@rustcorp.com.au>
Cc: Bjorn Helgaas <***@google.com>
Cc: "David S. Miller" <***@davemloft.net>
Cc: Wen Congyang <***@cn.fujitsu.com>
Cc: David Howells <***@redhat.com>
Cc: linux-***@vger.kernel.org
---
arch/tile/include/asm/sections.h | 2 +-
arch/tile/kernel/setup.c | 4 ++--
arch/tile/kernel/vmlinux.lds.S | 4 +++-
arch/tile/mm/init.c | 2 +-
4 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/arch/tile/include/asm/sections.h b/arch/tile/include/asm/sections.h
index d062d46..7d8a935 100644
--- a/arch/tile/include/asm/sections.h
+++ b/arch/tile/include/asm/sections.h
@@ -34,7 +34,7 @@ extern char __sys_cmpxchg_grab_lock[];
extern char __start_atomic_asm_code[], __end_atomic_asm_code[];
#endif

-/* Handle the discontiguity between _sdata and _stext. */
+/* Handle the discontiguity between _sdata and _text. */
static inline int arch_is_kernel_data(unsigned long addr)
{
return addr >= (unsigned long)_sdata &&
diff --git a/arch/tile/kernel/setup.c b/arch/tile/kernel/setup.c
index 7a5aa1a..41818e3 100644
--- a/arch/tile/kernel/setup.c
+++ b/arch/tile/kernel/setup.c
@@ -307,8 +307,8 @@ static void __cpuinit store_permanent_mappings(void)
hv_store_mapping(addr, pages << PAGE_SHIFT, pa);
}

- hv_store_mapping((HV_VirtAddr)_stext,
- (uint32_t)(_einittext - _stext), 0);
+ hv_store_mapping((HV_VirtAddr)_text,
+ (uint32_t)(_einittext - _text), 0);
}

/*
diff --git a/arch/tile/kernel/vmlinux.lds.S b/arch/tile/kernel/vmlinux.lds.S
index 631f10d..a13ed90 100644
--- a/arch/tile/kernel/vmlinux.lds.S
+++ b/arch/tile/kernel/vmlinux.lds.S
@@ -27,7 +27,6 @@ SECTIONS
.intrpt1 (LOAD_OFFSET) : AT ( 0 ) /* put at the start of physical memory */
{
_text = .;
- _stext = .;
*(.intrpt1)
} :intrpt1 =0

@@ -36,6 +35,7 @@ SECTIONS

/* Now the real code */
. = ALIGN(0x20000);
+ _stext = .;
.text : AT (ADDR(.text) - LOAD_OFFSET) {
HEAD_TEXT
SCHED_TEXT
@@ -58,11 +58,13 @@ SECTIONS
#define LOAD_OFFSET PAGE_OFFSET

. = ALIGN(PAGE_SIZE);
+ __init_begin = .;
VMLINUX_SYMBOL(_sinitdata) = .;
INIT_DATA_SECTION(16) :data =0
PERCPU_SECTION(L2_CACHE_BYTES)
. = ALIGN(PAGE_SIZE);
VMLINUX_SYMBOL(_einitdata) = .;
+ __init_end = .;

_sdata = .; /* Start of data section */

diff --git a/arch/tile/mm/init.c b/arch/tile/mm/init.c
index 45ce26d..f2ac2f4 100644
--- a/arch/tile/mm/init.c
+++ b/arch/tile/mm/init.c
@@ -562,7 +562,7 @@ static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
prot = ktext_set_nocache(prot);
}

- BUG_ON(address != (unsigned long)_stext);
+ BUG_ON(address != (unsigned long)_text);
pte = NULL;
for (; address < (unsigned long)_einittext;
pfn++, address += PAGE_SIZE) {
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:04 UTC
Permalink
Normalize global variables exported by vmlinux.lds to conform usage
guidelines from include/asm-generic/sections.h.

1) Use _text to mark the start of the kernel image including the head
text, and _stext to mark the start of the .text section.
2) Export mandatory global variables __bss_stop.
3) Adjust __init_begin and __init_end to avoid acrossing .text and
.data sections.

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Jeff Dike <***@addtoit.com>
Cc: Richard Weinberger <***@nod.at>
Cc: Al Viro <***@zeniv.linux.org.uk>
Cc: user-mode-linux-***@lists.sourceforge.net
Cc: linux-***@vger.kernel.org
---
arch/um/include/asm/common.lds.S | 1 -
arch/um/kernel/dyn.lds.S | 6 ++++--
arch/um/kernel/uml.lds.S | 7 +++++--
3 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/arch/um/include/asm/common.lds.S b/arch/um/include/asm/common.lds.S
index 4938de5..1dd5bd8 100644
--- a/arch/um/include/asm/common.lds.S
+++ b/arch/um/include/asm/common.lds.S
@@ -57,7 +57,6 @@
*(.uml.initcall.init)
__uml_initcall_end = .;
}
- __init_end = .;

SECURITY_INIT

diff --git a/arch/um/kernel/dyn.lds.S b/arch/um/kernel/dyn.lds.S
index fb8fd6f..adde088 100644
--- a/arch/um/kernel/dyn.lds.S
+++ b/arch/um/kernel/dyn.lds.S
@@ -14,8 +14,6 @@ SECTIONS
__binary_start = .;
. = ALIGN(4096); /* Init code and data */
_text = .;
- _stext = .;
- __init_begin = .;
INIT_TEXT_SECTION(PAGE_SIZE)

. = ALIGN(PAGE_SIZE);
@@ -67,6 +65,7 @@ SECTIONS
} =0x90909090
.plt : { *(.plt) }
.text : {
+ _stext = .;
TEXT_TEXT
SCHED_TEXT
LOCK_TEXT
@@ -91,7 +90,9 @@ SECTIONS

#include <asm/common.lds.S>

+ __init_begin = .;
init.data : { INIT_DATA }
+ __init_end = .;

/* Ensure the __preinit_array_start label is properly aligned. We
could instead move the label definition inside the section, but
@@ -155,6 +156,7 @@ SECTIONS
. = ALIGN(32 / 8);
. = ALIGN(32 / 8);
}
+ __bss_stop = .;
_end = .;
PROVIDE (end = .);

diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S
index ff65fb4..6899195 100644
--- a/arch/um/kernel/uml.lds.S
+++ b/arch/um/kernel/uml.lds.S
@@ -20,13 +20,12 @@ SECTIONS
. = START + SIZEOF_HEADERS;

_text = .;
- _stext = .;
- __init_begin = .;
INIT_TEXT_SECTION(0)
. = ALIGN(PAGE_SIZE);

.text :
{
+ _stext = .;
TEXT_TEXT
SCHED_TEXT
LOCK_TEXT
@@ -62,7 +61,10 @@ SECTIONS

#include <asm/common.lds.S>

+ __init_begin = .;
init.data : { INIT_DATA }
+ __init_end = .;
+
.data :
{
INIT_TASK_DATA(KERNEL_STACK_SIZE)
@@ -97,6 +99,7 @@ SECTIONS
PROVIDE(_bss_start = .);
SBSS(0)
BSS(0)
+ __bss_stop = .;
_end = .;
PROVIDE (end = .);
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:05 UTC
Permalink
Introduce helper function mem_init_print_info() to simplify mem_init()
across different architectures, which also unifies the format and
information printed.

Function mem_init_print_info() calculates memory statistics information
without walking each page, so it should be a little faster on some
architectures.

Also introduce another helper get_num_physpages() to kill the global
variable num_physpages.

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Andrew Morton <***@linux-foundation.org>
Cc: Mel Gorman <***@suse.de>
Cc: Michel Lespinasse <***@google.com>
Cc: Rik van Riel <***@redhat.com>
Cc: Minchan Kim <***@kernel.org>
Cc: Marek Szyprowski <***@samsung.com>
Cc: linux-***@kvack.org
Cc: linux-***@vger.kernel.org
---
include/linux/mm.h | 12 ++++++++++++
mm/page_alloc.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 64 insertions(+)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 86014c9..66e5fb8 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1313,6 +1313,7 @@ extern void free_highmem_page(struct page *page);
#endif

extern void adjust_managed_page_count(struct page *page, long count);
+extern void mem_init_print_info(const char *str);

/* Free the reserved page into the buddy system, so it gets managed. */
static inline void __free_reserved_page(struct page *page)
@@ -1349,6 +1350,17 @@ static inline unsigned long free_initmem_default(int poison)
poison, "unused kernel");
}

+static inline unsigned long get_num_physpages(void)
+{
+ int nid;
+ unsigned long phys_pages = 0;
+
+ for_each_online_node(nid)
+ phys_pages += node_present_pages(nid);
+
+ return phys_pages;
+}
+
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP
/*
* With CONFIG_HAVE_MEMBLOCK_NODE_MAP set, an architecture may initialise its
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index d49cb72..4f0ec15 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -61,6 +61,7 @@
#include <linux/hugetlb.h>
#include <linux/sched/rt.h>

+#include <asm/sections.h>
#include <asm/tlbflush.h>
#include <asm/div64.h>
#include "internal.h"
@@ -5225,6 +5226,57 @@ void free_highmem_page(struct page *page)
}
#endif

+
+void __init mem_init_print_info(const char *str)
+{
+ unsigned long physpages, codesize, datasize, rosize;
+ unsigned long init_code_size, init_data_size;
+
+ physpages = get_num_physpages();
+ codesize = _etext - _stext;
+ datasize = _edata - _sdata;
+ rosize = __end_rodata - __start_rodata;
+ init_data_size = __init_end - __init_begin;
+ init_code_size = _einittext - _sinittext;
+
+ /*
+ * Detect special cases and adjust section sizes accordingly:
+ * 1) .init.* may be embedded into .data sections
+ * 2) .init.text.* may be out of [__init_begin, __init_end],
+ * please refer to arch/tile/kernel/vmlinux.lds.S.
+ * 3) .rodata.* may be embedded into .text or .data sections.
+ */
+#define adj_init_size(start, end, size, pos, adj) \
+ if (start <= pos && pos < end && size > adj) \
+ size -= adj;
+
+ adj_init_size(__init_begin, __init_end, init_data_size,
+ _sinittext, init_code_size);
+ adj_init_size(_stext, _etext, codesize, _sinittext, init_code_size);
+ adj_init_size(_sdata, _edata, datasize, __init_begin, init_data_size);
+ adj_init_size(_stext, _etext, codesize, __start_rodata, rosize);
+ adj_init_size(_sdata, _edata, datasize, __start_rodata, rosize);
+
+#undef adj_init_size
+
+ printk("Memory: %luK/%luK available "
+ "(%luK kernel code, %luK rwdata, %luK rodata, "
+ "%luK init, %luK bss, %luK reserved"
+#ifdef CONFIG_HIGHMEM
+ ", %luK highmem"
+#endif
+ "%s%s)\n",
+ nr_free_pages() << (PAGE_SHIFT-10), physpages << (PAGE_SHIFT-10),
+ codesize >> 10, datasize >> 10, rosize >> 10,
+ (init_data_size + init_code_size) >> 10,
+ (__bss_stop - __bss_start) >> 10,
+ (physpages - totalram_pages) << (PAGE_SHIFT-10),
+#ifdef CONFIG_HIGHMEM
+ totalhigh_pages << (PAGE_SHIFT-10),
+#endif
+ str ? ", " : "", str ? str : "");
+}
+
/**
* set_dma_reserve - set the specified number of pages reserved in the first zone
* @new_dma_reserve: The number of pages to mark reserved
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:06 UTC
Permalink
The global variable num_physpages is scheduled to be removed, so use
totalram_pages instead of num_physpages at runtime.

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Miklos Szeredi <***@szeredi.hu>
Cc: "David S. Miller" <***@davemloft.net>
Cc: Alexey Kuznetsov <***@ms2.inr.ac.ru>
Cc: James Morris <***@namei.org>
Cc: Hideaki YOSHIFUJI <***@linux-ipv6.org>
Cc: Patrick McHardy <***@trash.net>
Cc: fuse-***@lists.sourceforge.net
Cc: linux-***@vger.kernel.org
Cc: ***@vger.kernel.org
---
fs/fuse/inode.c | 2 +-
kernel/power/snapshot.c | 4 ++--
net/ipv4/inet_fragment.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 17ec1f7..cd6dc27 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -785,7 +785,7 @@ static const struct super_operations fuse_super_operations = {
static void sanitize_global_limit(unsigned *limit)
{
if (*limit == 0)
- *limit = ((num_physpages << PAGE_SHIFT) >> 13) /
+ *limit = ((totalram_pages << PAGE_SHIFT) >> 13) /
sizeof(struct fuse_req);

if (*limit >= 1 << 16)
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 0de2857..8b5d1cd 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -1651,7 +1651,7 @@ unsigned long snapshot_get_image_size(void)
static int init_header(struct swsusp_info *info)
{
memset(info, 0, sizeof(struct swsusp_info));
- info->num_physpages = num_physpages;
+ info->num_physpages = get_num_physpages();
info->image_pages = nr_copy_pages;
info->pages = snapshot_get_image_size();
info->size = info->pages;
@@ -1795,7 +1795,7 @@ static int check_header(struct swsusp_info *info)
char *reason;

reason = check_image_kernel(info);
- if (!reason && info->num_physpages != num_physpages)
+ if (!reason && info->num_physpages != get_num_physpages())
reason = "memory size";
if (reason) {
printk(KERN_ERR "PM: Image mismatch: %s\n", reason);
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index e97d66a..f90079d 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -93,7 +93,7 @@ void inet_frags_init(struct inet_frags *f)
}
rwlock_init(&f->lock);

- f->rnd = (u32) ((num_physpages ^ (num_physpages>>7)) ^
+ f->rnd = (u32) ((totalram_pages ^ (totalram_pages >> 7)) ^
(jiffies ^ (jiffies >> 6)));

setup_timer(&f->secret_timer, inet_frag_secret_rebuild,
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:07 UTC
Permalink
Prepare for removing num_physpages.

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Wen Congyang <***@cn.fujitsu.com>
Cc: Tang Chen <***@cn.fujitsu.com>
Cc: Yasuaki Ishimatsu <***@jp.fujitsu.com>
Cc: linux-***@kvack.org
Cc: linux-***@vger.kernel.org
---
mm/memory_hotplug.c | 4 ----
1 file changed, 4 deletions(-)

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 304ada7..283755a 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -760,10 +760,6 @@ EXPORT_SYMBOL_GPL(restore_online_page_callback);

void __online_page_set_limits(struct page *page)
{
- unsigned long pfn = page_to_pfn(page);
-
- if (pfn >= num_physpages)
- num_physpages = pfn + 1;
}
EXPORT_SYMBOL_GPL(__online_page_set_limits);
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:08 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Richard Henderson <***@twiddle.net>
Cc: Ivan Kokshaysky <***@jurassic.park.msu.ru>
Cc: Matt Turner <***@gmail.com>
Cc: David Howells <***@redhat.com>
Cc: linux-***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
arch/alpha/mm/init.c | 32 ++------------------------------
arch/alpha/mm/numa.c | 34 ++--------------------------------
2 files changed, 4 insertions(+), 62 deletions(-)

diff --git a/arch/alpha/mm/init.c b/arch/alpha/mm/init.c
index ca07a97..04c933c 100644
--- a/arch/alpha/mm/init.c
+++ b/arch/alpha/mm/init.c
@@ -277,42 +277,14 @@ srm_paging_stop (void)
#endif

#ifndef CONFIG_DISCONTIGMEM
-static void __init
-printk_memory_info(void)
-{
- unsigned long codesize, reservedpages, datasize, initsize, tmp;
- extern int page_is_ram(unsigned long) __init;
-
- /* printk all informations */
- reservedpages = 0;
- for (tmp = 0; tmp < max_low_pfn; tmp++)
- /*
- * Only count reserved RAM pages
- */
- if (page_is_ram(tmp) && PageReserved(mem_map+tmp))
- reservedpages++;
-
- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- datasize = (unsigned long) &_edata - (unsigned long) &_data;
- initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
-
- printk("Memory: %luk/%luk available (%luk kernel code, %luk reserved, %luk data, %luk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- max_mapnr << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10);
-}
-
void __init
mem_init(void)
{
- max_mapnr = num_physpages = max_low_pfn;
+ max_mapnr = max_low_pfn;
free_all_bootmem();
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);

- printk_memory_info();
+ mem_init_print_info(NULL);
}
#endif /* CONFIG_DISCONTIGMEM */

diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c
index 857452c..0894b3a8 100644
--- a/arch/alpha/mm/numa.c
+++ b/arch/alpha/mm/numa.c
@@ -129,8 +129,6 @@ setup_memory_node(int nid, void *kernel_end)
if (node_max_pfn > max_low_pfn)
max_pfn = max_low_pfn = node_max_pfn;

- num_physpages += node_max_pfn - node_min_pfn;
-
#if 0 /* we'll try this one again in a little while */
/* Cute trick to make sure our local node data is on local memory */
node_data[nid] = (pg_data_t *)(__va(node_min_pfn << PAGE_SHIFT));
@@ -324,37 +322,9 @@ void __init paging_init(void)

void __init mem_init(void)
{
- unsigned long codesize, reservedpages, datasize, initsize, pfn;
- extern int page_is_ram(unsigned long) __init;
- unsigned long nid, i;
high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT);
-
- reservedpages = 0;
- for_each_online_node(nid) {
- /*
- * This will free up the bootmem, ie, slot 0 memory
- */
- free_all_bootmem_node(NODE_DATA(nid));
-
- pfn = NODE_DATA(nid)->node_start_pfn;
- for (i = 0; i < node_spanned_pages(nid); i++, pfn++)
- if (page_is_ram(pfn) &&
- PageReserved(nid_page_nr(nid, i)))
- reservedpages++;
- }
-
- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- datasize = (unsigned long) &_edata - (unsigned long) &_data;
- initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
-
- printk("Memory: %luk/%luk available (%luk kernel code, %luk reserved, "
- "%luk data, %luk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10);
+ free_all_bootmem();
+ mem_init_print_info(NULL);
#if 0
mem_stress();
#endif
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:09 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Vineet Gupta <***@synopsys.com>
Cc: James Hogan <***@imgtec.com>
Cc: Rob Herring <***@calxeda.com>
Cc: linux-***@vger.kernel.org
---
arch/arc/mm/init.c | 36 +++---------------------------------
1 file changed, 3 insertions(+), 33 deletions(-)

diff --git a/arch/arc/mm/init.c b/arch/arc/mm/init.c
index 78d8c31..8ba6562 100644
--- a/arch/arc/mm/init.c
+++ b/arch/arc/mm/init.c
@@ -74,7 +74,7 @@ void __init setup_arch_memory(void)
/* Last usable page of low mem (no HIGHMEM yet for ARC port) */
max_low_pfn = max_pfn = PFN_DOWN(end_mem);

- max_mapnr = num_physpages = max_low_pfn - min_low_pfn;
+ max_mapnr = max_low_pfn - min_low_pfn;

/*------------- reserve kernel image -----------------------*/
memblock_reserve(CONFIG_LINUX_LINK_BASE,
@@ -84,7 +84,7 @@ void __init setup_arch_memory(void)

/*-------------- node setup --------------------------------*/
memset(zones_size, 0, sizeof(zones_size));
- zones_size[ZONE_NORMAL] = num_physpages;
+ zones_size[ZONE_NORMAL] = max_low_pfn - min_low_pfn;

/*
* We can't use the helper free_area_init(zones[]) because it uses
@@ -106,39 +106,9 @@ void __init setup_arch_memory(void)
*/
void __init mem_init(void)
{
- int codesize, datasize, initsize, reserved_pages, free_pages;
- int tmp;
-
high_memory = (void *)(CONFIG_LINUX_LINK_BASE + arc_mem_sz);
-
free_all_bootmem();
-
- /* count all reserved pages [kernel code/data/mem_map..] */
- reserved_pages = 0;
- for (tmp = 0; tmp < max_mapnr; tmp++)
- if (PageReserved(mem_map + tmp))
- reserved_pages++;
-
- /* XXX: nr_free_pages() is equivalent */
- free_pages = max_mapnr - reserved_pages;
-
- /*
- * For the purpose of display below, split the "reserve mem"
- * kernel code/data is already shown explicitly,
- * Show any other reservations (mem_map[ ] et al)
- */
- reserved_pages -= (((unsigned int)_end - CONFIG_LINUX_LINK_BASE) >>
- PAGE_SHIFT);
-
- codesize = _etext - _text;
- datasize = _end - _etext;
- initsize = __init_end - __init_begin;
-
- pr_info("Memory Available: %dM / %ldM (%dK code, %dK data, %dK init, %dK reserv)\n",
- PAGES_TO_MB(free_pages),
- TO_MB(arc_mem_sz),
- TO_KB(codesize), TO_KB(datasize), TO_KB(initsize),
- PAGES_TO_KB(reserved_pages));
+ mem_init_print_info(NULL);
}

/*
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Vineet Gupta
2013-05-29 08:41:30 UTC
Permalink
Hi Jiang,
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
---
arch/arc/mm/init.c | 36 +++---------------------------------
1 file changed, 3 insertions(+), 33 deletions(-)
diff --git a/arch/arc/mm/init.c b/arch/arc/mm/init.c
index 78d8c31..8ba6562 100644
--- a/arch/arc/mm/init.c
+++ b/arch/arc/mm/init.c
@@ -74,7 +74,7 @@ void __init setup_arch_memory(void)
/* Last usable page of low mem (no HIGHMEM yet for ARC port) */
max_low_pfn = max_pfn = PFN_DOWN(end_mem);
- max_mapnr = num_physpages = max_low_pfn - min_low_pfn;
+ max_mapnr = max_low_pfn - min_low_pfn;
/*------------- reserve kernel image -----------------------*/
memblock_reserve(CONFIG_LINUX_LINK_BASE,
@@ -84,7 +84,7 @@ void __init setup_arch_memory(void)
/*-------------- node setup --------------------------------*/
memset(zones_size, 0, sizeof(zones_size));
- zones_size[ZONE_NORMAL] = num_physpages;
+ zones_size[ZONE_NORMAL] = max_low_pfn - min_low_pfn;
/*
* We can't use the helper free_area_init(zones[]) because it uses
@@ -106,39 +106,9 @@ void __init setup_arch_memory(void)
*/
void __init mem_init(void)
{
- int codesize, datasize, initsize, reserved_pages, free_pages;
- int tmp;
-
high_memory = (void *)(CONFIG_LINUX_LINK_BASE + arc_mem_sz);
-
free_all_bootmem();
-
What baseline is this code against, since mainline looks like following:

high_memory = (void *)(CONFIG_LINUX_LINK_BASE + arc_mem_sz);

totalram_pages = free_all_bootmem();

So I would have expected the following

- totalram_pages = free_all_bootmem();
+ free_all_bootmem();

Aha, you missed out CCing all maintainers on "[PATCH v7, part3 14/16]" or rather
"[PATCH v8, part3 13/14]" - and it is difficult to dig this out from all the
patches that fly by on linux-arch.

Same goes for "[PATCH v8, part3 01/14]" and "[PATCH v8, part3 02/14]"
Post by Jiang Liu
- /* count all reserved pages [kernel code/data/mem_map..] */
- reserved_pages = 0;
- for (tmp = 0; tmp < max_mapnr; tmp++)
- if (PageReserved(mem_map + tmp))
- reserved_pages++;
-
- /* XXX: nr_free_pages() is equivalent */
- free_pages = max_mapnr - reserved_pages;
-
- /*
- * For the purpose of display below, split the "reserve mem"
- * kernel code/data is already shown explicitly,
- * Show any other reservations (mem_map[ ] et al)
- */
- reserved_pages -= (((unsigned int)_end - CONFIG_LINUX_LINK_BASE) >>
- PAGE_SHIFT);
-
- codesize = _etext - _text;
- datasize = _end - _etext;
- initsize = __init_end - __init_begin;
-
- pr_info("Memory Available: %dM / %ldM (%dK code, %dK data, %dK init, %dK reserv)\n",
- PAGES_TO_MB(free_pages),
- TO_MB(arc_mem_sz),
- TO_KB(codesize), TO_KB(datasize), TO_KB(initsize),
- PAGES_TO_KB(reserved_pages));
+ mem_init_print_info(NULL);
}
/*
The Changes look OK though. I managed to build your github tree (mem_init_v5
branch). And it seems to work ok.

Acked-by: Vineet Gupta <***@synopsys.com> # for arch/arc

Thx
-Vineet

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Liu Jiang
2013-05-29 13:05:43 UTC
Permalink
Post by Vineet Gupta
Hi Jiang,
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
---
arch/arc/mm/init.c | 36 +++---------------------------------
1 file changed, 3 insertions(+), 33 deletions(-)
diff --git a/arch/arc/mm/init.c b/arch/arc/mm/init.c
index 78d8c31..8ba6562 100644
--- a/arch/arc/mm/init.c
+++ b/arch/arc/mm/init.c
@@ -74,7 +74,7 @@ void __init setup_arch_memory(void)
/* Last usable page of low mem (no HIGHMEM yet for ARC port) */
max_low_pfn = max_pfn = PFN_DOWN(end_mem);
- max_mapnr = num_physpages = max_low_pfn - min_low_pfn;
+ max_mapnr = max_low_pfn - min_low_pfn;
/*------------- reserve kernel image -----------------------*/
memblock_reserve(CONFIG_LINUX_LINK_BASE,
@@ -84,7 +84,7 @@ void __init setup_arch_memory(void)
/*-------------- node setup --------------------------------*/
memset(zones_size, 0, sizeof(zones_size));
- zones_size[ZONE_NORMAL] = num_physpages;
+ zones_size[ZONE_NORMAL] = max_low_pfn - min_low_pfn;
/*
* We can't use the helper free_area_init(zones[]) because it uses
@@ -106,39 +106,9 @@ void __init setup_arch_memory(void)
*/
void __init mem_init(void)
{
- int codesize, datasize, initsize, reserved_pages, free_pages;
- int tmp;
-
high_memory = (void *)(CONFIG_LINUX_LINK_BASE + arc_mem_sz);
-
free_all_bootmem();
-
high_memory = (void *)(CONFIG_LINUX_LINK_BASE + arc_mem_sz);
totalram_pages = free_all_bootmem();
So I would handve expected the following
- totalram_pages = free_all_bootmem();
+ free_all_bootmem();
Aha, you missed out CCing all maintainers on "[PATCH v7, part3 14/16]" or rather
"[PATCH v8, part3 13/14]" - and it is difficult to dig this out from all the
patches that fly by on linux-arch.
Same goes for "[PATCH v8, part3 01/14]" and "[PATCH v8, part3 02/14]"
Hi Vineet,
This patchset is based on Andrew's mmotm tree, there are some
differences
between mmotm and the upstream kernel, sorry for the inconvenience.
Once I sent those common patches to all related maintainers, but
the CC list
is too long and also caused other troubles. Next time I will send all
patches to
linux-arch maillist.
Post by Vineet Gupta
Post by Jiang Liu
- /* count all reserved pages [kernel code/data/mem_map..] */
- reserved_pages = 0;
- for (tmp = 0; tmp < max_mapnr; tmp++)
- if (PageReserved(mem_map + tmp))
- reserved_pages++;
-
- /* XXX: nr_free_pages() is equivalent */
- free_pages = max_mapnr - reserved_pages;
-
- /*
- * For the purpose of display below, split the "reserve mem"
- * kernel code/data is already shown explicitly,
- * Show any other reservations (mem_map[ ] et al)
- */
- reserved_pages -= (((unsigned int)_end - CONFIG_LINUX_LINK_BASE) >>
- PAGE_SHIFT);
-
- codesize = _etext - _text;
- datasize = _end - _etext;
- initsize = __init_end - __init_begin;
-
- pr_info("Memory Available: %dM / %ldM (%dK code, %dK data, %dK init, %dK reserv)\n",
- PAGES_TO_MB(free_pages),
- TO_MB(arc_mem_sz),
- TO_KB(codesize), TO_KB(datasize), TO_KB(initsize),
- PAGES_TO_KB(reserved_pages));
+ mem_init_print_info(NULL);
}
/*
The Changes look OK though. I managed to build your github tree (mem_init_v5
branch). And it seems to work ok.
Thanks for review and tests!
Regards!
Gerry
Post by Vineet Gupta
Thx
-Vineet
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:10 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Russell King <***@arm.linux.org.uk>
Cc: Catalin Marinas <***@arm.com>
Cc: Will Deacon <***@arm.com>
Cc: linux-arm-***@lists.infradead.org
Cc: linux-***@vger.kernel.org
---
arch/arm/mm/init.c | 47 ++---------------------------------------------
1 file changed, 2 insertions(+), 45 deletions(-)

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index add4fcb..7a911d1 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -582,9 +582,6 @@ static void __init free_highpages(void)
*/
void __init mem_init(void)
{
- unsigned long reserved_pages, free_pages;
- struct memblock_region *reg;
- int i;
#ifdef CONFIG_HAVE_TCM
/* These pointers are filled in on TCM detection */
extern u32 dtcm_end;
@@ -605,47 +602,7 @@ void __init mem_init(void)

free_highpages();

- reserved_pages = free_pages = 0;
-
- for_each_bank(i, &meminfo) {
- struct membank *bank = &meminfo.bank[i];
- unsigned int pfn1, pfn2;
- struct page *page, *end;
-
- pfn1 = bank_pfn_start(bank);
- pfn2 = bank_pfn_end(bank);
-
- page = pfn_to_page(pfn1);
- end = pfn_to_page(pfn2 - 1) + 1;
-
- do {
- if (PageReserved(page))
- reserved_pages++;
- else if (!page_count(page))
- free_pages++;
- page++;
- } while (page < end);
- }
-
- /*
- * Since our memory may not be contiguous, calculate the
- * real number of pages we have in this system
- */
- printk(KERN_INFO "Memory:");
- num_physpages = 0;
- for_each_memblock(memory, reg) {
- unsigned long pages = memblock_region_memory_end_pfn(reg) -
- memblock_region_memory_base_pfn(reg);
- num_physpages += pages;
- printk(" %ldMB", pages >> (20 - PAGE_SHIFT));
- }
- printk(" = %luMB total\n", num_physpages >> (20 - PAGE_SHIFT));
-
- printk(KERN_NOTICE "Memory: %luk/%luk available, %luk reserved, %luK highmem\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- free_pages << (PAGE_SHIFT-10),
- reserved_pages << (PAGE_SHIFT-10),
- totalhigh_pages << (PAGE_SHIFT-10));
+ mem_init_print_info(NULL);

#define MLK(b, t) b, t, ((t) - (b)) >> 10
#define MLM(b, t) b, t, ((t) - (b)) >> 20
@@ -711,7 +668,7 @@ void __init mem_init(void)
BUG_ON(PKMAP_BASE + LAST_PKMAP * PAGE_SIZE > PAGE_OFFSET);
#endif

- if (PAGE_SIZE >= 16384 && num_physpages <= 128) {
+ if (PAGE_SIZE >= 16384 && get_num_physpages() <= 128) {
extern int sysctl_overcommit_memory;
/*
* On a machine this small we won't get
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:11 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Russell King <***@arm.linux.org.uk>
Cc: Catalin Marinas <***@arm.com>
Cc: Will Deacon <***@arm.com>
Cc: linux-arm-***@lists.infradead.org
Cc: linux-***@vger.kernel.org
---
arch/arm64/mm/init.c | 48 +++---------------------------------------------
1 file changed, 3 insertions(+), 45 deletions(-)

diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 0f2cf5d..821e788 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -272,59 +272,17 @@ static void __init free_unused_memmap(void)
*/
void __init mem_init(void)
{
- unsigned long reserved_pages, free_pages;
- struct memblock_region *reg;
-
arm64_swiotlb_init();

max_mapnr = pfn_to_page(max_pfn + PHYS_PFN_OFFSET) - mem_map;

#ifndef CONFIG_SPARSEMEM_VMEMMAP
- /* this will put all unused low memory onto the freelists */
free_unused_memmap();
#endif
-
+ /* this will put all unused low memory onto the freelists */
free_all_bootmem();

- reserved_pages = free_pages = 0;
-
- for_each_memblock(memory, reg) {
- unsigned int pfn1, pfn2;
- struct page *page, *end;
-
- pfn1 = __phys_to_pfn(reg->base);
- pfn2 = pfn1 + __phys_to_pfn(reg->size);
-
- page = pfn_to_page(pfn1);
- end = pfn_to_page(pfn2 - 1) + 1;
-
- do {
- if (PageReserved(page))
- reserved_pages++;
- else if (!page_count(page))
- free_pages++;
- page++;
- } while (page < end);
- }
-
- /*
- * Since our memory may not be contiguous, calculate the real number
- * of pages we have in this system.
- */
- pr_info("Memory:");
- num_physpages = 0;
- for_each_memblock(memory, reg) {
- unsigned long pages = memblock_region_memory_end_pfn(reg) -
- memblock_region_memory_base_pfn(reg);
- num_physpages += pages;
- printk(" %ldMB", pages >> (20 - PAGE_SHIFT));
- }
- printk(" = %luMB total\n", num_physpages >> (20 - PAGE_SHIFT));
-
- pr_notice("Memory: %luk/%luk available, %luk reserved\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- free_pages << (PAGE_SHIFT-10),
- reserved_pages << (PAGE_SHIFT-10));
+ mem_init_print_info();

#define MLK(b, t) b, t, ((t) - (b)) >> 10
#define MLM(b, t) b, t, ((t) - (b)) >> 20
@@ -366,7 +324,7 @@ void __init mem_init(void)
BUILD_BUG_ON(TASK_SIZE_64 > MODULES_VADDR);
BUG_ON(TASK_SIZE_64 > MODULES_VADDR);

- if (PAGE_SIZE >= 16384 && num_physpages <= 128) {
+ if (PAGE_SIZE >= 16384 && get_num_physpages() <= 128) {
extern int sysctl_overcommit_memory;
/*
* On a machine this small we won't get anywhere without
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:12 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Hans-Christian Egtvedt <***@samfundet.no>
Cc: Haavard Skinnemoen <***@gmail.com>
Cc: linux-***@vger.kernel.org
---
arch/avr32/mm/init.c | 29 ++++-------------------------
1 file changed, 4 insertions(+), 25 deletions(-)

diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c
index 7e8d55a..c1706a0 100644
--- a/arch/avr32/mm/init.c
+++ b/arch/avr32/mm/init.c
@@ -100,26 +100,16 @@ void __init paging_init(void)

void __init mem_init(void)
{
- int codesize, reservedpages, datasize, initsize;
- int nid, i;
+ pg_data_t *pgdat;

- reservedpages = 0;
high_memory = NULL;

/* this will put all low memory onto the freelists */
- for_each_online_node(nid) {
- pg_data_t *pgdat = NODE_DATA(nid);
- unsigned long node_pages = 0;
+ for_each_online_pgdat(pgdat) {
void *node_high_memory;

- num_physpages += pgdat->node_present_pages;
-
if (pgdat->node_spanned_pages != 0)
- node_pages = free_all_bootmem_node(pgdat);
-
- for (i = 0; i < node_pages; i++)
- if (PageReserved(pgdat->node_mem_map + i))
- reservedpages++;
+ free_all_bootmem_node(pgdat);

node_high_memory = (void *)((pgdat->node_start_pfn
+ pgdat->node_spanned_pages)
@@ -130,18 +120,7 @@ void __init mem_init(void)

max_mapnr = MAP_NR(high_memory);

- codesize = (unsigned long)_etext - (unsigned long)_text;
- datasize = (unsigned long)_edata - (unsigned long)_data;
- initsize = (unsigned long)__init_end - (unsigned long)__init_begin;
-
- printk ("Memory: %luk/%luk available (%dk kernel code, "
- "%dk reserved, %dk data, %dk init)\n",
- nr_free_pages() << (PAGE_SHIFT - 10),
- totalram_pages << (PAGE_SHIFT - 10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT - 10),
- datasize >> 10,
- initsize >> 10);
+ mem_init_print_info(NULL);
}

void free_initmem(void)
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:13 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Mike Frysinger <***@gentoo.org>
Cc: Bob Liu <***@gmail.com>
Cc: uclinux-dist-***@blackfin.uclinux.org
Cc: linux-***@vger.kernel.org
---
arch/blackfin/mm/init.c | 38 ++++++--------------------------------
1 file changed, 6 insertions(+), 32 deletions(-)

diff --git a/arch/blackfin/mm/init.c b/arch/blackfin/mm/init.c
index 1cc8607..e4b6e11 100644
--- a/arch/blackfin/mm/init.c
+++ b/arch/blackfin/mm/init.c
@@ -90,43 +90,17 @@ asmlinkage void __init init_pda(void)

void __init mem_init(void)
{
- unsigned int codek = 0, datak = 0, initk = 0;
- unsigned int reservedpages = 0, freepages = 0;
- unsigned long tmp;
- unsigned long start_mem = memory_start;
- unsigned long end_mem = memory_end;
+ char buf[64];

- end_mem &= PAGE_MASK;
- high_memory = (void *)end_mem;
-
- start_mem = PAGE_ALIGN(start_mem);
- max_mapnr = num_physpages = MAP_NR(high_memory);
- printk(KERN_DEBUG "Kernel managed physical pages: %lu\n", num_physpages);
+ high_memory = (void *)(memory_end & PAGE_MASK);
+ max_mapnr = MAP_NR(high_memory);
+ printk(KERN_DEBUG "Kernel managed physical pages: %lu\n", max_mapnr);

/* This will put all low memory onto the freelists. */
free_all_bootmem();

- reservedpages = 0;
- for (tmp = ARCH_PFN_OFFSET; tmp < max_mapnr; tmp++)
- if (PageReserved(pfn_to_page(tmp)))
- reservedpages++;
- freepages = max_mapnr - ARCH_PFN_OFFSET - reservedpages;
-
- /* do not count in kernel image between _rambase and _ramstart */
- reservedpages -= (_ramstart - _rambase) >> PAGE_SHIFT;
-#if (defined(CONFIG_BFIN_EXTMEM_ICACHEABLE) && ANOMALY_05000263)
- reservedpages += (_ramend - memory_end - DMA_UNCACHED_REGION) >> PAGE_SHIFT;
-#endif
-
- codek = (_etext - _stext) >> 10;
- initk = (__init_end - __init_begin) >> 10;
- datak = ((_ramstart - _rambase) >> 10) - codek - initk;
-
- printk(KERN_INFO
- "Memory available: %luk/%luk RAM, "
- "(%uk init code, %uk kernel code, %uk data, %uk dma, %uk reserved)\n",
- (unsigned long) freepages << (PAGE_SHIFT-10), (_ramend - CONFIG_PHY_RAM_BASE_ADDRESS) >> 10,
- initk, codek, datak, DMA_UNCACHED_REGION >> 10, (reservedpages << (PAGE_SHIFT-10)));
+ snprintf(buf, sizeof(buf) - 1, "%uK DMA", DMA_UNCACHED_REGION >> 10);
+ mem_init_print_info(buf);
}

#ifdef CONFIG_BLK_DEV_INITRD
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Sonic Zhang
2013-05-25 13:25:47 UTC
Permalink
Hi Jiang
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
---
arch/blackfin/mm/init.c | 38 ++++++--------------------------------
1 file changed, 6 insertions(+), 32 deletions(-)
diff --git a/arch/blackfin/mm/init.c b/arch/blackfin/mm/init.c
index 1cc8607..e4b6e11 100644
--- a/arch/blackfin/mm/init.c
+++ b/arch/blackfin/mm/init.c
@@ -90,43 +90,17 @@ asmlinkage void __init init_pda(void)
void __init mem_init(void)
{
- unsigned int codek = 0, datak = 0, initk = 0;
- unsigned int reservedpages = 0, freepages = 0;
- unsigned long tmp;
- unsigned long start_mem = memory_start;
- unsigned long end_mem = memory_end;
+ char buf[64];
- end_mem &= PAGE_MASK;
- high_memory = (void *)end_mem;
-
- start_mem = PAGE_ALIGN(start_mem);
- max_mapnr = num_physpages = MAP_NR(high_memory);
- printk(KERN_DEBUG "Kernel managed physical pages: %lu\n", num_physpages);
+ high_memory = (void *)(memory_end & PAGE_MASK);
+ max_mapnr = MAP_NR(high_memory);
+ printk(KERN_DEBUG "Kernel managed physical pages: %lu\n", max_mapnr);
/* This will put all low memory onto the freelists. */
free_all_bootmem();
- reservedpages = 0;
- for (tmp = ARCH_PFN_OFFSET; tmp < max_mapnr; tmp++)
- if (PageReserved(pfn_to_page(tmp)))
- reservedpages++;
- freepages = max_mapnr - ARCH_PFN_OFFSET - reservedpages;
-
- /* do not count in kernel image between _rambase and _ramstart */
- reservedpages -= (_ramstart - _rambase) >> PAGE_SHIFT;
-#if (defined(CONFIG_BFIN_EXTMEM_ICACHEABLE) && ANOMALY_05000263)
- reservedpages += (_ramend - memory_end - DMA_UNCACHED_REGION) >> PAGE_SHIFT;
-#endif
-
- codek = (_etext - _stext) >> 10;
- initk = (__init_end - __init_begin) >> 10;
- datak = ((_ramstart - _rambase) >> 10) - codek - initk;
-
- printk(KERN_INFO
- "Memory available: %luk/%luk RAM, "
- "(%uk init code, %uk kernel code, %uk data, %uk dma, %uk reserved)\n",
- (unsigned long) freepages << (PAGE_SHIFT-10), (_ramend - CONFIG_PHY_RAM_BASE_ADDRESS) >> 10,
- initk, codek, datak, DMA_UNCACHED_REGION >> 10, (reservedpages << (PAGE_SHIFT-10)));
You can't remove all these memory information for blackfin. They are
useful on blackfin platform.

Regards,

Sonic

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Liu Jiang
2013-05-26 13:55:34 UTC
Permalink
Post by Sonic Zhang
Hi Jiang
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
---
arch/blackfin/mm/init.c | 38 ++++++--------------------------------
1 file changed, 6 insertions(+), 32 deletions(-)
diff --git a/arch/blackfin/mm/init.c b/arch/blackfin/mm/init.c
index 1cc8607..e4b6e11 100644
--- a/arch/blackfin/mm/init.c
+++ b/arch/blackfin/mm/init.c
@@ -90,43 +90,17 @@ asmlinkage void __init init_pda(void)
void __init mem_init(void)
{
- unsigned int codek = 0, datak = 0, initk = 0;
- unsigned int reservedpages = 0, freepages = 0;
- unsigned long tmp;
- unsigned long start_mem = memory_start;
- unsigned long end_mem = memory_end;
+ char buf[64];
- end_mem &= PAGE_MASK;
- high_memory = (void *)end_mem;
-
- start_mem = PAGE_ALIGN(start_mem);
- max_mapnr = num_physpages = MAP_NR(high_memory);
- printk(KERN_DEBUG "Kernel managed physical pages: %lu\n", num_physpages);
+ high_memory = (void *)(memory_end & PAGE_MASK);
+ max_mapnr = MAP_NR(high_memory);
+ printk(KERN_DEBUG "Kernel managed physical pages: %lu\n", max_mapnr);
/* This will put all low memory onto the freelists. */
free_all_bootmem();
- reservedpages = 0;
- for (tmp = ARCH_PFN_OFFSET; tmp < max_mapnr; tmp++)
- if (PageReserved(pfn_to_page(tmp)))
- reservedpages++;
- freepages = max_mapnr - ARCH_PFN_OFFSET - reservedpages;
-
- /* do not count in kernel image between _rambase and _ramstart */
- reservedpages -= (_ramstart - _rambase) >> PAGE_SHIFT;
-#if (defined(CONFIG_BFIN_EXTMEM_ICACHEABLE) && ANOMALY_05000263)
- reservedpages += (_ramend - memory_end - DMA_UNCACHED_REGION) >> PAGE_SHIFT;
-#endif
-
- codek = (_etext - _stext) >> 10;
- initk = (__init_end - __init_begin) >> 10;
- datak = ((_ramstart - _rambase) >> 10) - codek - initk;
-
- printk(KERN_INFO
- "Memory available: %luk/%luk RAM, "
- "(%uk init code, %uk kernel code, %uk data, %uk dma, %uk reserved)\n",
- (unsigned long) freepages << (PAGE_SHIFT-10), (_ramend - CONFIG_PHY_RAM_BASE_ADDRESS) >> 10,
- initk, codek, datak, DMA_UNCACHED_REGION >> 10, (reservedpages << (PAGE_SHIFT-10)));
You can't remove all these memory information for blackfin. They are
useful on blackfin platform.
Regards,
Sonic
Hi Sonic,
Thanks for review!
We are not trying to remove these code, but replacing it
with a generic
helper function mem_init_print_info(), which will print similar (actual
more info)
boot message. I have no blackfin platforms to generate a real boot
message,
so could only share an example boot messages on x86:
Memory: 7744624K/8074824K available (6969K kernel code, 1011K data,
2828K rodata, 1016K init, 9640K bss, 330200K reserved)

Regards!
Gerry

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:14 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Mark Salter <***@redhat.com>
Cc: Aurelien Jacquiot <a-***@ti.com>
Cc: linux-c6x-***@linux-c6x.org
Cc: linux-***@vger.kernel.org
---
arch/c6x/mm/init.c | 11 +----------
1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/arch/c6x/mm/init.c b/arch/c6x/mm/init.c
index 2c51474..066f75c 100644
--- a/arch/c6x/mm/init.c
+++ b/arch/c6x/mm/init.c
@@ -57,21 +57,12 @@ void __init paging_init(void)

void __init mem_init(void)
{
- int codek, datak;
- unsigned long tmp;
- unsigned long len = memory_end - memory_start;
-
high_memory = (void *)(memory_end & PAGE_MASK);

/* this will put all memory onto the freelists */
free_all_bootmem();

- codek = (_etext - _stext) >> 10;
- datak = (_end - _sdata) >> 10;
-
- tmp = nr_free_pages() << PAGE_SHIFT;
- printk(KERN_INFO "Memory: %luk/%luk RAM (%dk kernel code, %dk data)\n",
- tmp >> 10, len >> 10, codek, datak);
+ mem_init_print_info(NULL);
}

#ifdef CONFIG_BLK_DEV_INITRD
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:15 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Jesper Nilsson <***@axis.com>
Cc: Mikael Starvik <***@axis.com>
Cc: linux-cris-***@axis.com
Cc: linux-***@vger.kernel.org
---
arch/cris/mm/init.c | 33 ++-------------------------------
1 file changed, 2 insertions(+), 31 deletions(-)

diff --git a/arch/cris/mm/init.c b/arch/cris/mm/init.c
index 52b8b56..c81af5b 100644
--- a/arch/cris/mm/init.c
+++ b/arch/cris/mm/init.c
@@ -19,9 +19,6 @@ unsigned long empty_zero_page;
void __init
mem_init(void)
{
- int codesize, reservedpages, datasize, initsize;
- unsigned long tmp;
-
BUG_ON(!mem_map);

/* max/min_low_pfn was set by setup.c
@@ -29,35 +26,9 @@ mem_init(void)
*
* high_memory was also set in setup.c
*/
-
- max_mapnr = num_physpages = max_low_pfn - min_low_pfn;
-
- /* this will put all memory onto the freelists */
+ max_mapnr = max_low_pfn - min_low_pfn;
free_all_bootmem();
-
- reservedpages = 0;
- for (tmp = 0; tmp < max_mapnr; tmp++) {
- /*
- * Only count reserved RAM pages
- */
- if (PageReserved(mem_map + tmp))
- reservedpages++;
- }
-
- 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)\n" ,
- nr_free_pages() << (PAGE_SHIFT-10),
- max_mapnr << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10
- );
+ mem_init_print_info(NULL);
}

/* free the pages occupied by initialization code */
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:16 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: David Howells <***@redhat.com>
Cc: Andi Kleen <***@linux.intel.com>
Cc: Geert Uytterhoeven <***@linux-m68k.org>
Cc: Greg Kroah-Hartman <***@linuxfoundation.org>
Cc: linux-***@vger.kernel.org
---
arch/frv/kernel/setup.c | 13 +++++++------
arch/frv/mm/init.c | 49 ++++++++++++++---------------------------------
2 files changed, 21 insertions(+), 41 deletions(-)

diff --git a/arch/frv/kernel/setup.c b/arch/frv/kernel/setup.c
index a513647..f78f8cb 100644
--- a/arch/frv/kernel/setup.c
+++ b/arch/frv/kernel/setup.c
@@ -876,6 +876,7 @@ late_initcall(setup_arch_serial);
static void __init setup_linux_memory(void)
{
unsigned long bootmap_size, low_top_pfn, kstart, kend, high_mem;
+ unsigned long physpages;

kstart = (unsigned long) &__kernel_image_start - PAGE_OFFSET;
kend = (unsigned long) &__kernel_image_end - PAGE_OFFSET;
@@ -893,19 +894,19 @@ static void __init setup_linux_memory(void)
);

/* pass the memory that the kernel can immediately use over to the bootmem allocator */
- max_mapnr = num_physpages = (memory_end - memory_start) >> PAGE_SHIFT;
+ max_mapnr = physpages = (memory_end - memory_start) >> PAGE_SHIFT;
low_top_pfn = (KERNEL_LOWMEM_END - KERNEL_LOWMEM_START) >> PAGE_SHIFT;
high_mem = 0;

- if (num_physpages > low_top_pfn) {
+ if (physpages > low_top_pfn) {
#ifdef CONFIG_HIGHMEM
- high_mem = num_physpages - low_top_pfn;
+ high_mem = physpages - low_top_pfn;
#else
- max_mapnr = num_physpages = low_top_pfn;
+ max_mapnr = physpages = low_top_pfn;
#endif
}
else {
- low_top_pfn = num_physpages;
+ low_top_pfn = physpages;
}

min_low_pfn = memory_start >> PAGE_SHIFT;
@@ -979,7 +980,7 @@ static void __init setup_uclinux_memory(void)
free_bootmem(memory_start, memory_end - memory_start);

high_memory = (void *) (memory_end & PAGE_MASK);
- max_mapnr = num_physpages = ((unsigned long) high_memory - PAGE_OFFSET) >> PAGE_SHIFT;
+ max_mapnr = ((unsigned long) high_memory - PAGE_OFFSET) >> PAGE_SHIFT;

min_low_pfn = memory_start >> PAGE_SHIFT;
max_low_pfn = memory_end >> PAGE_SHIFT;
diff --git a/arch/frv/mm/init.c b/arch/frv/mm/init.c
index 4215822..e9e2b6f 100644
--- a/arch/frv/mm/init.c
+++ b/arch/frv/mm/init.c
@@ -78,7 +78,7 @@ void __init paging_init(void)
memset((void *) empty_zero_page, 0, PAGE_SIZE);

#ifdef CONFIG_HIGHMEM
- if (num_physpages - num_mappedpages) {
+ if (get_num_physpages() - num_mappedpages) {
pgd_t *pge;
pud_t *pue;
pmd_t *pme;
@@ -96,7 +96,7 @@ void __init paging_init(void)
*/
zones_size[ZONE_NORMAL] = max_low_pfn - min_low_pfn;
#ifdef CONFIG_HIGHMEM
- zones_size[ZONE_HIGHMEM] = num_physpages - num_mappedpages;
+ zones_size[ZONE_HIGHMEM] = get_num_physpages() - num_mappedpages;
#endif

free_area_init(zones_size);
@@ -114,45 +114,24 @@ void __init paging_init(void)
*/
void __init mem_init(void)
{
- unsigned long npages = (memory_end - memory_start) >> PAGE_SHIFT;
- unsigned long tmp;
-#ifdef CONFIG_MMU
- unsigned long loop, pfn;
- int datapages = 0;
-#endif
- int codek = 0, datak = 0;
+ unsigned long code_size = _etext - _stext;

/* this will put all low memory onto the freelists */
free_all_bootmem();
+#if defined(CONFIG_MMU) && defined(CONFIG_HIGHMEM)
+ {
+ unsigned long pfn;

-#ifdef CONFIG_MMU
- for (loop = 0 ; loop < npages ; loop++)
- if (PageReserved(&mem_map[loop]))
- datapages++;
-
-#ifdef CONFIG_HIGHMEM
- for (pfn = num_physpages - 1; pfn >= num_mappedpages; pfn--)
- free_highmem_page(&mem_map[pfn]);
-#endif
-
- codek = ((unsigned long) &_etext - (unsigned long) &_stext) >> 10;
- datak = datapages << (PAGE_SHIFT - 10);
-
-#else
- codek = (_etext - _stext) >> 10;
- datak = 0; //(__bss_stop - _sdata) >> 10;
+ for (pfn = get_num_physpages() - 1;
+ pfn >= num_mappedpages; pfn--)
+ free_highmem_page(&mem_map[pfn]);
+ }
#endif

- tmp = nr_free_pages() << PAGE_SHIFT;
- printk("Memory available: %luKiB/%luKiB RAM, %luKiB/%luKiB ROM (%dKiB kernel code, %dKiB data)\n",
- tmp >> 10,
- npages << (PAGE_SHIFT - 10),
- (rom_length > 0) ? ((rom_length >> 10) - codek) : 0,
- rom_length >> 10,
- codek,
- datak
- );
-
+ mem_init_print_info(NULL);
+ if (rom_length > 0 && rom_length >= code_size)
+ printk("Memory available: %luKiB/%luKiB ROM\n",
+ (rom_length - code_size) >> 10, rom_length >> 10);
} /* end mem_init() */

/*****************************************************************************/
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:17 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Yoshinori Sato <***@users.sourceforge.jp>
Cc: Geert Uytterhoeven <***@linux-m68k.org>
Cc: linux-***@vger.kernel.org
---
arch/h8300/mm/init.c | 34 ++++++++--------------------------
1 file changed, 8 insertions(+), 26 deletions(-)

diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c
index 22fd869..0088f3a 100644
--- a/arch/h8300/mm/init.c
+++ b/arch/h8300/mm/init.c
@@ -121,40 +121,22 @@ void __init paging_init(void)

void __init mem_init(void)
{
- int codek = 0, datak = 0, initk = 0;
- /* DAVIDM look at setup memory map generically with reserved area */
- unsigned long tmp;
- extern unsigned long _ramend, _ramstart;
- unsigned long len = &_ramend - &_ramstart;
- unsigned long start_mem = memory_start; /* DAVIDM - these must start at end of kernel */
- unsigned long end_mem = memory_end; /* DAVIDM - this must not include kernel stack at top */
+ unsigned long codesize = _etext - _stext;

#ifdef DEBUG
- printk(KERN_DEBUG "Mem_init: start=%lx, end=%lx\n", start_mem, end_mem);
+ pr_debug("Mem_init: start=%lx, end=%lx\n", memory_start, memory_end);
#endif

- end_mem &= PAGE_MASK;
- high_memory = (void *) end_mem;
-
- start_mem = PAGE_ALIGN(start_mem);
- max_mapnr = num_physpages = MAP_NR(high_memory);
+ high_memory = (void *) (memory_end & PAGE_MASK);
+ max_mapnr = MAP_NR(high_memory);

/* this will put all low memory onto the freelists */
free_all_bootmem();

- codek = (_etext - _stext) >> 10;
- datak = (__bss_stop - _sdata) >> 10;
- initk = (__init_begin - __init_end) >> 10;
-
- tmp = nr_free_pages() << PAGE_SHIFT;
- printk(KERN_INFO "Memory available: %luk/%luk RAM, %luk/%luk ROM (%dk kernel code, %dk data)\n",
- tmp >> 10,
- len >> 10,
- (rom_length > 0) ? ((rom_length >> 10) - codek) : 0,
- rom_length >> 10,
- codek,
- datak
- );
+ mem_init_print_info(NULL);
+ if (rom_length > 0 && rom_length > codesize)
+ pr_info("Memory available: %luK/%luK ROM\n",
+ (rom_length - codesize) >> 10, rom_length >> 10);
}
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Sergei Shtylyov
2013-05-08 16:26:40 UTC
Permalink
Hello.
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
---
arch/h8300/mm/init.c | 34 ++++++++--------------------------
1 file changed, 8 insertions(+), 26 deletions(-)
diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c
index 22fd869..0088f3a 100644
--- a/arch/h8300/mm/init.c
+++ b/arch/h8300/mm/init.c
@@ -121,40 +121,22 @@ void __init paging_init(void)
void __init mem_init(void)
{
- int codek = 0, datak = 0, initk = 0;
- /* DAVIDM look at setup memory map generically with reserved area */
- unsigned long tmp;
- extern unsigned long _ramend, _ramstart;
- unsigned long len = &_ramend - &_ramstart;
- unsigned long start_mem = memory_start; /* DAVIDM - these must start at end of kernel */
- unsigned long end_mem = memory_end; /* DAVIDM - this must not include kernel stack at top */
+ unsigned long codesize = _etext - _stext;
#ifdef DEBUG
- printk(KERN_DEBUG "Mem_init: start=%lx, end=%lx\n", start_mem, end_mem);
+ pr_debug("Mem_init: start=%lx, end=%lx\n", memory_start, memory_end);
#endif
pr_debug() only prints something if DEBUG is #define'd, so you can
drop the #ifdef here.

WBR, Sergei
Sergei Shtylyov
2013-05-08 19:29:36 UTC
Permalink
Hello.
Post by Sergei Shtylyov
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
---
arch/h8300/mm/init.c | 34 ++++++++--------------------------
1 file changed, 8 insertions(+), 26 deletions(-)
diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c
index 22fd869..0088f3a 100644
--- a/arch/h8300/mm/init.c
+++ b/arch/h8300/mm/init.c
@@ -121,40 +121,22 @@ void __init paging_init(void)
void __init mem_init(void)
{
- int codek = 0, datak = 0, initk = 0;
- /* DAVIDM look at setup memory map generically with reserved
area */
- unsigned long tmp;
- extern unsigned long _ramend, _ramstart;
- unsigned long len = &_ramend - &_ramstart;
- unsigned long start_mem = memory_start; /* DAVIDM - these must
start at end of kernel */
- unsigned long end_mem = memory_end; /* DAVIDM - this must not
include kernel stack at top */
+ unsigned long codesize = _etext - _stext;
#ifdef DEBUG
- printk(KERN_DEBUG "Mem_init: start=%lx, end=%lx\n", start_mem,
end_mem);
+ pr_debug("Mem_init: start=%lx, end=%lx\n", memory_start,
memory_end);
#endif
pr_debug() only prints something if DEBUG is #define'd, so you can
drop the #ifdef here.
Although, not necessarily: it also supports CONFIG_DYNAMIC_DEBUG --
look at how pr_debug() is defined.
So this doesn't seem to be an equivalent change, and I suggest not doing
it at all.

WBR, Sergei

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Cody P Schafer
2013-05-08 23:35:22 UTC
Permalink
Post by Sergei Shtylyov
Although, not necessarily: it also supports CONFIG_DYNAMIC_DEBUG --
look at how pr_debug() is defined.
So this doesn't seem to be an equivalent change, and I suggest not doing
it at all.
WBR, Sergei
pr_devel() should get the same behavior: no code emitted unless debug is
defined, if it is, output at KERN_DEBUG level.
Liu Jiang
2013-05-12 15:17:14 UTC
Permalink
Post by Cody P Schafer
Post by Sergei Shtylyov
Although, not necessarily: it also supports CONFIG_DYNAMIC_DEBUG --
look at how pr_debug() is defined.
So this doesn't seem to be an equivalent change, and I suggest not doing
it at all.
WBR, Sergei
pr_devel() should get the same behavior: no code emitted unless debug
is defined, if it is, output at KERN_DEBUG level.
Thanks Cody and Sergei, will use pr_devel() instead in next version.
Gerry

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:18 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Richard Kuo <***@codeaurora.org>
Cc: linux-***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
arch/hexagon/mm/init.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/hexagon/mm/init.c b/arch/hexagon/mm/init.c
index 0ab5b43..88977e4 100644
--- a/arch/hexagon/mm/init.c
+++ b/arch/hexagon/mm/init.c
@@ -71,9 +71,7 @@ void __init mem_init(void)
{
/* No idea where this is actually declared. Seems to evade LXR. */
free_all_bootmem();
- num_physpages = bootmem_lastpg-ARCH_PFN_OFFSET;
-
- printk(KERN_INFO "totalram_pages = %ld\n", totalram_pages);
+ mem_init_print_info(NULL);

/*
* To-Do: someone somewhere should wipe out the bootmem map
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:19 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Tony Luck <***@intel.com>
Cc: Fenghua Yu <***@intel.com>
Cc: Andrew Morton <***@linux-foundation.org>
Cc: Zhang Yanfei <***@cn.fujitsu.com>
Cc: linux-***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
arch/ia64/mm/contig.c | 11 -----------
arch/ia64/mm/discontig.c | 3 ---
arch/ia64/mm/init.c | 27 +--------------------------
3 files changed, 1 insertion(+), 40 deletions(-)

diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index 955fd25..b975c8b 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -295,14 +295,6 @@ find_memory (void)
alloc_per_cpu_data();
}

-static int count_pages(u64 start, u64 end, void *arg)
-{
- unsigned long *count = arg;
-
- *count += (end - start) >> PAGE_SHIFT;
- return 0;
-}
-
/*
* Set up the page tables.
*/
@@ -313,9 +305,6 @@ paging_init (void)
unsigned long max_dma;
unsigned long max_zone_pfns[MAX_NR_ZONES];

- num_physpages = 0;
- efi_memmap_walk(count_pages, &num_physpages);
-
memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
#ifdef CONFIG_ZONE_DMA
max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 254ae8e..3aef17c 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -37,7 +37,6 @@ struct early_node_data {
struct ia64_node_data *node_data;
unsigned long pernode_addr;
unsigned long pernode_size;
- unsigned long num_physpages;
#ifdef CONFIG_ZONE_DMA
unsigned long num_dma_physpages;
#endif
@@ -732,7 +731,6 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n
{
unsigned long end = start + len;

- mem_data[node].num_physpages += len >> PAGE_SHIFT;
#ifdef CONFIG_ZONE_DMA
if (start <= __pa(MAX_DMA_ADDRESS))
mem_data[node].num_dma_physpages +=
@@ -778,7 +776,6 @@ void __init paging_init(void)
#endif

for_each_online_node(node) {
- num_physpages += mem_data[node].num_physpages;
pfn_offset = mem_data[node].min_pfn;

#ifdef CONFIG_VIRTUAL_MEM_MAP
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index b5b71e8..d4382dc 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -546,19 +546,6 @@ int __init register_active_ranges(u64 start, u64 len, int nid)
return 0;
}

-static int __init
-count_reserved_pages(u64 start, u64 end, void *arg)
-{
- unsigned long num_reserved = 0;
- unsigned long *count = arg;
-
- for (; start < end; start += PAGE_SIZE)
- if (PageReserved(virt_to_page(start)))
- ++num_reserved;
- *count += num_reserved;
- return 0;
-}
-
int
find_max_min_low_pfn (u64 start, u64 end, void *arg)
{
@@ -597,7 +584,6 @@ __setup("nolwsys", nolwsys_setup);
void __init
mem_init (void)
{
- long reserved_pages, codesize, datasize, initsize;
pg_data_t *pgdat;
int i;

@@ -625,18 +611,7 @@ mem_init (void)
if (pgdat->bdata->node_bootmem_map)
free_all_bootmem_node(pgdat);

- reserved_pages = 0;
- efi_memmap_walk(count_reserved_pages, &reserved_pages);
-
- 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 (%luk code, %luk reserved, "
- "%luk data, %luk init)\n", nr_free_pages() << (PAGE_SHIFT - 10),
- num_physpages << (PAGE_SHIFT - 10), codesize >> 10,
- reserved_pages << (PAGE_SHIFT - 10), datasize >> 10, initsize >> 10);
-
+ mem_init_print_info(NULL);

/*
* For fsyscall entrpoints with no light-weight handler, use the ordinary
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:20 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Hirokazu Takata <***@linux-m32r.org>
Cc: linux-***@ml.linux-m32r.org
Cc: linux-m32r-***@ml.linux-m32r.org
Cc: linux-***@vger.kernel.org
---
arch/m32r/mm/discontig.c | 6 +-----
arch/m32r/mm/init.c | 49 +++++-----------------------------------------
2 files changed, 6 insertions(+), 49 deletions(-)

diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c
index 2c468e8..2719630 100644
--- a/arch/m32r/mm/discontig.c
+++ b/arch/m32r/mm/discontig.c
@@ -129,11 +129,10 @@ unsigned long __init setup_memory(void)
#define START_PFN(nid) (NODE_DATA(nid)->bdata->node_min_pfn)
#define MAX_LOW_PFN(nid) (NODE_DATA(nid)->bdata->node_low_pfn)

-unsigned long __init zone_sizes_init(void)
+void __init zone_sizes_init(void)
{
unsigned long zones_size[MAX_NR_ZONES], zholes_size[MAX_NR_ZONES];
unsigned long low, start_pfn;
- unsigned long holes = 0;
int nid, i;
mem_prof_t *mp;

@@ -147,7 +146,6 @@ unsigned long __init zone_sizes_init(void)
low = MAX_LOW_PFN(nid);
zones_size[ZONE_DMA] = low - start_pfn;
zholes_size[ZONE_DMA] = mp->holes;
- holes += zholes_size[ZONE_DMA];

node_set_state(nid, N_NORMAL_MEMORY);
free_area_init_node(nid, zones_size, start_pfn, zholes_size);
@@ -161,6 +159,4 @@ unsigned long __init zone_sizes_init(void)
NODE_DATA(1)->node_zones->watermark[WMARK_MIN] = 0;
NODE_DATA(1)->node_zones->watermark[WMARK_LOW] = 0;
NODE_DATA(1)->node_zones->watermark[WMARK_HIGH] = 0;
-
- return holes;
}
diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c
index c421c31..9c94839 100644
--- a/arch/m32r/mm/init.c
+++ b/arch/m32r/mm/init.c
@@ -40,7 +40,6 @@ unsigned long mmu_context_cache_dat;
#else
unsigned long mmu_context_cache_dat[NR_CPUS];
#endif
-static unsigned long hole_pages;

/*
* function prototype
@@ -57,7 +56,7 @@ void free_initrd_mem(unsigned long, unsigned long);
#define MAX_LOW_PFN(nid) (NODE_DATA(nid)->bdata->node_low_pfn)

#ifndef CONFIG_DISCONTIGMEM
-unsigned long __init zone_sizes_init(void)
+void __init zone_sizes_init(void)
{
unsigned long zones_size[MAX_NR_ZONES] = {0, };
unsigned long max_dma;
@@ -83,11 +82,9 @@ unsigned long __init zone_sizes_init(void)
#endif /* CONFIG_MMU */

free_area_init_node(0, zones_size, start_pfn, 0);
-
- return 0;
}
#else /* CONFIG_DISCONTIGMEM */
-extern unsigned long zone_sizes_init(void);
+extern void zone_sizes_init(void);
#endif /* CONFIG_DISCONTIGMEM */

/*======================================================================*
@@ -105,24 +102,7 @@ void __init paging_init(void)
for (i = 0 ; i < USER_PTRS_PER_PGD * 2 ; i++)
pgd_val(pg_dir[i]) = 0;
#endif /* CONFIG_MMU */
- hole_pages = zone_sizes_init();
-}
-
-int __init reservedpages_count(void)
-{
- int reservedpages, nid, i;
-
- reservedpages = 0;
- for_each_online_node(nid) {
- unsigned long flags;
- pgdat_resize_lock(NODE_DATA(nid), &flags);
- for (i = 0 ; i < MAX_LOW_PFN(nid) - START_PFN(nid) ; i++)
- if (PageReserved(nid_page_nr(nid, i)))
- reservedpages++;
- pgdat_resize_unlock(NODE_DATA(nid), &flags);
- }
-
- return reservedpages;
+ zone_sizes_init();
}

/*======================================================================*
@@ -131,20 +111,13 @@ int __init reservedpages_count(void)
*======================================================================*/
void __init mem_init(void)
{
- int codesize, reservedpages, datasize, initsize;
int nid;
#ifndef CONFIG_MMU
extern unsigned long memory_end;
#endif

- num_physpages = 0;
- for_each_online_node(nid)
- num_physpages += MAX_LOW_PFN(nid) - START_PFN(nid) + 1;
-
- num_physpages -= hole_pages;
-
#ifndef CONFIG_DISCONTIGMEM
- max_mapnr = num_physpages;
+ max_mapnr = get_num_physpages();
#endif /* CONFIG_DISCONTIGMEM */

#ifdef CONFIG_MMU
@@ -160,19 +133,7 @@ void __init mem_init(void)
for_each_online_node(nid)
free_all_bootmem_node(NODE_DATA(nid));

- reservedpages = reservedpages_count() - hole_pages;
- codesize = (unsigned long) &_etext - (unsigned long)&_text;
- 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)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10);
+ mem_init_print_info(NULL);
}

/*======================================================================*
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:21 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Greg Ungerer <***@uclinux.org>
Cc: Geert Uytterhoeven <***@linux-m68k.org>
Cc: linux-***@lists.linux-m68k.org
Cc: linux-***@vger.kernel.org
---
arch/m68k/mm/init.c | 31 ++-----------------------------
1 file changed, 2 insertions(+), 29 deletions(-)

diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c
index 2485a8c..0723141 100644
--- a/arch/m68k/mm/init.c
+++ b/arch/m68k/mm/init.c
@@ -149,33 +149,11 @@ void __init print_memmap(void)
void __init mem_init(void)
{
pg_data_t *pgdat;
- int codepages = 0;
- int datapages = 0;
- int initpages = 0;
int i;

/* this will put all memory onto the freelists */
- num_physpages = 0;
- for_each_online_pgdat(pgdat) {
- num_physpages += pgdat->node_present_pages;
-
+ for_each_online_pgdat(pgdat)
free_all_bootmem_node(pgdat);
- for (i = 0; i < pgdat->node_spanned_pages; i++) {
- struct page *page = pgdat->node_mem_map + i;
- char *addr = page_to_virt(page);
-
- if (!PageReserved(page))
- continue;
- if (addr >= _text &&
- addr < _etext)
- codepages++;
- else if (addr >= __init_begin &&
- addr < __init_end)
- initpages++;
- else
- datapages++;
- }
- }

#if defined(CONFIG_MMU) && !defined(CONFIG_SUN3) && !defined(CONFIG_COLDFIRE)
/* insert pointer tables allocated so far into the tablelist */
@@ -190,12 +168,7 @@ void __init mem_init(void)
init_pointer_table((unsigned long)zero_pgtable);
#endif

- pr_info("Memory: %luk/%luk available (%dk kernel code, %dk data, %dk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- totalram_pages << (PAGE_SHIFT-10),
- codepages << (PAGE_SHIFT-10),
- datapages << (PAGE_SHIFT-10),
- initpages << (PAGE_SHIFT-10));
+ mem_init_print_info(NULL);
print_memmap();
}
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:22 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
---
arch/metag/mm/init.c | 8 +-------
1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/arch/metag/mm/init.c b/arch/metag/mm/init.c
index 279d701..e00586f 100644
--- a/arch/metag/mm/init.c
+++ b/arch/metag/mm/init.c
@@ -388,22 +388,16 @@ void __init mem_init(void)
reset_all_zones_managed_pages();
for (tmp = highstart_pfn; tmp < highend_pfn; tmp++)
free_highmem_page(pfn_to_page(tmp));
- num_physpages += totalhigh_pages;
#endif /* CONFIG_HIGHMEM */

for_each_online_node(nid) {
pg_data_t *pgdat = NODE_DATA(nid);

- num_physpages += pgdat->node_present_pages;
-
if (pgdat->node_spanned_pages)
free_all_bootmem_node(pgdat);
}

- pr_info("Memory: %luk/%luk available\n",
- (unsigned long)nr_free_pages() << (PAGE_SHIFT - 10),
- num_physpages << (PAGE_SHIFT - 10));
-
+ mem_init_print_info(NULL);
show_mem(0);

return;
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:23 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Michal Simek <***@monstr.eu>
Cc: microblaze-***@itee.uq.edu.au
Cc: linux-***@vger.kernel.org
---
arch/microblaze/mm/init.c | 51 ++++++---------------------------------------
1 file changed, 6 insertions(+), 45 deletions(-)

diff --git a/arch/microblaze/mm/init.c b/arch/microblaze/mm/init.c
index fb7f248..b0a4a36 100644
--- a/arch/microblaze/mm/init.c
+++ b/arch/microblaze/mm/init.c
@@ -71,24 +71,17 @@ static void __init highmem_init(void)
kmap_prot = PAGE_KERNEL;
}

-static unsigned long highmem_setup(void)
+static void highmem_setup(void)
{
unsigned long pfn;
- unsigned long reservedpages = 0;

for (pfn = max_low_pfn; pfn < max_pfn; ++pfn) {
struct page *page = pfn_to_page(pfn);

/* FIXME not sure about */
- if (memblock_is_reserved(pfn << PAGE_SHIFT))
- continue;
- free_highmem_page(page);
- reservedpages++;
+ if (!memblock_is_reserved(pfn << PAGE_SHIFT))
+ free_highmem_page(page);
}
- pr_info("High memory: %luk\n",
- totalhigh_pages << (PAGE_SHIFT-10));
-
- return reservedpages;
}
#endif /* CONFIG_HIGHMEM */

@@ -167,13 +160,12 @@ void __init setup_memory(void)
* min_low_pfn - the first page (mm/bootmem.c - node_boot_start)
* max_low_pfn
* max_mapnr - the first unused page (mm/bootmem.c - node_low_pfn)
- * num_physpages - number of all pages
*/

/* memory start is from the kernel end (aligned) to higher addr */
min_low_pfn = memory_start >> PAGE_SHIFT; /* minimum for allocation */
/* RAM is assumed contiguous */
- num_physpages = max_mapnr = memory_size >> PAGE_SHIFT;
+ max_mapnr = memory_size >> PAGE_SHIFT;
max_low_pfn = ((u64)memory_start + (u64)lowmem_size) >> PAGE_SHIFT;
max_pfn = ((u64)memory_start + (u64)memory_size) >> PAGE_SHIFT;

@@ -246,46 +238,15 @@ void free_initmem(void)

void __init mem_init(void)
{
- pg_data_t *pgdat;
- unsigned long reservedpages = 0, codesize, initsize, datasize, bsssize;
-
high_memory = (void *)__va(memory_start + lowmem_size - 1);

/* this will put all memory onto the freelists */
free_all_bootmem();
-
- for_each_online_pgdat(pgdat) {
- unsigned long i;
- struct page *page;
-
- for (i = 0; i < pgdat->node_spanned_pages; i++) {
- if (!pfn_valid(pgdat->node_start_pfn + i))
- continue;
- page = pgdat_page_nr(pgdat, i);
- if (PageReserved(page))
- reservedpages++;
- }
- }
-
#ifdef CONFIG_HIGHMEM
- reservedpages -= highmem_setup();
+ highmem_setup();
#endif

- codesize = (unsigned long)&_sdata - (unsigned long)&_stext;
- datasize = (unsigned long)&_edata - (unsigned long)&_sdata;
- initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin;
- bsssize = (unsigned long)&__bss_stop - (unsigned long)&__bss_start;
-
- pr_info("Memory: %luk/%luk available (%luk kernel code, ",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10);
- pr_cont("%luk reserved, %luk data, %luk bss, %luk init)\n",
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- bsssize >> 10,
- initsize >> 10);
-
+ mem_init_print_info(str);
#ifdef CONFIG_MMU
pr_info("Kernel virtual memory layout:\n");
pr_info(" * 0x%08lx..0x%08lx : fixmap\n", FIXADDR_START, FIXADDR_TOP);
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:24 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Ralf Baechle <***@linux-mips.org>
Cc: David Daney <***@cavium.com>
Cc: Arnd Bergmann <***@arndb.de>
Cc: Jiri Kosina <***@suse.cz>
Cc: John Crispin <***@openwrt.org>
Cc: Greg Kroah-Hartman <***@linuxfoundation.org>
Cc: Minchan Kim <***@kernel.org>
Cc: linux-***@linux-mips.org
Cc: linux-***@vger.kernel.org
---
arch/mips/mm/init.c | 57 ++++++++++++--------------------------
arch/mips/pci/pci-lantiq.c | 2 +-
arch/mips/sgi-ip27/ip27-memory.c | 21 ++------------
3 files changed, 21 insertions(+), 59 deletions(-)

diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c
index de4ff2f..0392a1a 100644
--- a/arch/mips/mm/init.c
+++ b/arch/mips/mm/init.c
@@ -358,11 +358,24 @@ void __init paging_init(void)
static struct kcore_list kcore_kseg0;
#endif

-void __init mem_init(void)
+static inline void mem_init_free_highmem(void)
{
- unsigned long codesize, reservedpages, datasize, initsize;
- unsigned long tmp, ram;
+#ifdef CONFIG_HIGHMEM
+ unsigned long tmp;

+ for (tmp = highstart_pfn; tmp < highend_pfn; tmp++) {
+ struct page *page = pfn_to_page(tmp);
+
+ if (!page_is_ram(tmp))
+ SetPageReserved(page);
+ else
+ free_highmem_page(page);
+ }
+#endif
+}
+
+void __init mem_init(void)
+{
#ifdef CONFIG_HIGHMEM
#ifdef CONFIG_DISCONTIGMEM
#error "CONFIG_HIGHMEM and CONFIG_DISCONTIGMEM dont work together yet"
@@ -375,32 +388,8 @@ void __init mem_init(void)

free_all_bootmem();
setup_zero_pages(); /* Setup zeroed pages. */
-
- reservedpages = ram = 0;
- for (tmp = 0; tmp < max_low_pfn; tmp++)
- if (page_is_ram(tmp) && pfn_valid(tmp)) {
- ram++;
- if (PageReserved(pfn_to_page(tmp)))
- reservedpages++;
- }
- num_physpages = ram;
-
-#ifdef CONFIG_HIGHMEM
- for (tmp = highstart_pfn; tmp < highend_pfn; tmp++) {
- struct page *page = pfn_to_page(tmp);
-
- if (!page_is_ram(tmp)) {
- SetPageReserved(page);
- continue;
- }
- free_highmem_page(page);
- }
- num_physpages += totalhigh_pages;
-#endif
-
- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- datasize = (unsigned long) &_edata - (unsigned long) &_etext;
- initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
+ mem_init_free_highmem();
+ mem_init_print_info(NULL);

#ifdef CONFIG_64BIT
if ((unsigned long) &_text > (unsigned long) CKSEG0)
@@ -409,16 +398,6 @@ void __init mem_init(void)
kclist_add(&kcore_kseg0, (void *) CKSEG0,
0x80000000 - 4, KCORE_TEXT);
#endif
-
- printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
- "%ldk reserved, %ldk data, %ldk init, %ldk highmem)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- ram << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10,
- totalhigh_pages << (PAGE_SHIFT-10));
}
#endif /* !CONFIG_NEED_MULTIPLE_NODES */

diff --git a/arch/mips/pci/pci-lantiq.c b/arch/mips/pci/pci-lantiq.c
index 879077b..cb1ef99 100644
--- a/arch/mips/pci/pci-lantiq.c
+++ b/arch/mips/pci/pci-lantiq.c
@@ -89,7 +89,7 @@ static inline u32 ltq_calc_bar11mask(void)
u32 mem, bar11mask;

/* BAR11MASK value depends on available memory on system. */
- mem = num_physpages * PAGE_SIZE;
+ mem = get_num_physpages() * PAGE_SIZE;
bar11mask = (0x0ffffff0 & ~((1 << (fls(mem) - 1)) - 1)) | 8;

return bar11mask;
diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index 936e617..d074680 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -357,8 +357,6 @@ static void __init szmem(void)
int slot;
cnodeid_t node;

- num_physpages = 0;
-
for_each_online_node(node) {
nodebytes = 0;
for (slot = 0; slot < MAX_MEM_SLOTS; slot++) {
@@ -381,7 +379,6 @@ static void __init szmem(void)
slot = MAX_MEM_SLOTS;
continue;
}
- num_physpages += slot_psize;
memblock_add_node(PFN_PHYS(slot_getbasepfn(node, slot)),
PFN_PHYS(slot_psize), node);
}
@@ -480,10 +477,9 @@ void __init paging_init(void)

void __init mem_init(void)
{
- unsigned long codesize, datasize, initsize, tmp;
unsigned node;

- high_memory = (void *) __va(num_physpages << PAGE_SHIFT);
+ high_memory = (void *) __va(get_num_physpages() << PAGE_SHIFT);

for_each_online_node(node) {
/*
@@ -494,18 +490,5 @@ void __init mem_init(void)

setup_zero_pages(); /* This comes from node 0 */

- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- datasize = (unsigned long) &_edata - (unsigned long) &_etext;
- initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
-
- tmp = nr_free_pages();
- printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
- "%ldk reserved, %ldk data, %ldk init, %ldk highmem)\n",
- tmp << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- (num_physpages - tmp) << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10,
- totalhigh_pages << (PAGE_SHIFT-10));
+ mem_init_print_info(NULL);
}
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:25 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: David Howells <***@redhat.com>
Cc: Koichi Yasutake <***@jp.panasonic.com>
Cc: linux-am33-***@redhat.com
Cc: linux-***@vger.kernel.org
---
arch/mn10300/mm/init.c | 26 ++------------------------
1 file changed, 2 insertions(+), 24 deletions(-)

diff --git a/arch/mn10300/mm/init.c b/arch/mn10300/mm/init.c
index d7312aa..f43d993 100644
--- a/arch/mn10300/mm/init.c
+++ b/arch/mn10300/mm/init.c
@@ -99,15 +99,12 @@ void __init paging_init(void)
*/
void __init mem_init(void)
{
- int codesize, reservedpages, datasize, initsize;
- int tmp;
-
BUG_ON(!mem_map);

#define START_PFN (contig_page_data.bdata->node_min_pfn)
#define MAX_LOW_PFN (contig_page_data.bdata->node_low_pfn)

- max_mapnr = num_physpages = MAX_LOW_PFN - START_PFN;
+ max_mapnr = MAX_LOW_PFN - START_PFN;
high_memory = (void *) __va(MAX_LOW_PFN * PAGE_SIZE);

/* clear the zero-page */
@@ -116,26 +113,7 @@ void __init mem_init(void)
/* this will put all low memory onto the freelists */
free_all_bootmem();

- reservedpages = 0;
- for (tmp = 0; tmp < num_physpages; tmp++)
- if (PageReserved(&mem_map[tmp]))
- reservedpages++;
-
- 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",
- nr_free_pages() << (PAGE_SHIFT - 10),
- max_mapnr << (PAGE_SHIFT - 10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT - 10),
- datasize >> 10,
- initsize >> 10,
- totalhigh_pages << (PAGE_SHIFT - 10));
+ mem_init_print_info(NULL);
}

/*
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:26 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Jonas Bonn <***@southpole.se>
Cc: David Howells <***@redhat.com>
Cc: Arnd Bergmann <***@arndb.de>
Cc: ***@lists.openrisc.net
Cc: linux-***@vger.kernel.org
---
arch/openrisc/mm/init.c | 44 ++++----------------------------------------
1 file changed, 4 insertions(+), 40 deletions(-)

diff --git a/arch/openrisc/mm/init.c b/arch/openrisc/mm/init.c
index da26482..ac438ab 100644
--- a/arch/openrisc/mm/init.c
+++ b/arch/openrisc/mm/init.c
@@ -202,56 +202,20 @@ void __init paging_init(void)

/* References to section boundaries */

-static int __init free_pages_init(void)
-{
- int reservedpages, pfn;
-
- /* this will put all low memory onto the freelists */
- 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;
-
BUG_ON(!mem_map);

- set_max_mapnr_init();
-
+ max_mapnr = max_low_pfn;
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;
+ /* this will put all low memory onto the freelists */
+ free_all_bootmem();

- 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))
- );
+ mem_init_print_info(NULL);

printk("mem_init_done ...........................................\n");
mem_init_done = 1;
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:27 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: "James E.J. Bottomley" <***@parisc-linux.org>
Cc: Helge Deller <***@gmx.de>
Cc: Thomas Gleixner <***@linutronix.de>
Cc: Michal Hocko <***@suse.cz>
Cc: David Rientjes <***@google.com>
Cc: linux-***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
arch/parisc/mm/init.c | 46 +++-------------------------------------------
1 file changed, 3 insertions(+), 43 deletions(-)

diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index 1fe9d841..f80c175 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -214,7 +214,6 @@ static void __init setup_bootmem(void)
mem_limit_func(); /* check for "mem=" argument */

mem_max = 0;
- num_physpages = 0;
for (i = 0; i < npmem_ranges; i++) {
unsigned long rsize;

@@ -229,10 +228,8 @@ static void __init setup_bootmem(void)
npmem_ranges = i + 1;
mem_max = mem_limit;
}
- num_physpages += pmem_ranges[i].pages;
break;
}
- num_physpages += pmem_ranges[i].pages;
mem_max += rsize;
}

@@ -532,7 +529,7 @@ void free_initmem(void)
* pages are no-longer executable */
flush_icache_range(init_begin, init_end);

- num_physpages += free_initmem_default(-1);
+ free_initmem_default(-1);

/* set up a new led state on systems shipped LED State panel */
pdc_chassis_send_status(PDC_CHASSIS_DIRECT_BCOMPLETE);
@@ -580,8 +577,6 @@ unsigned long pcxl_dma_start __read_mostly;

void __init mem_init(void)
{
- int codesize, reservedpages, datasize, initsize;
-
/* Do sanity checks on page table constants */
BUILD_BUG_ON(PTE_ENTRY_SIZE != sizeof(pte_t));
BUILD_BUG_ON(PMD_ENTRY_SIZE != sizeof(pmd_t));
@@ -603,33 +598,6 @@ void __init mem_init(void)
}
#endif

- codesize = (unsigned long)_etext - (unsigned long)_text;
- datasize = (unsigned long)_edata - (unsigned long)_etext;
- initsize = (unsigned long)__init_end - (unsigned long)__init_begin;
-
- reservedpages = 0;
-{
- unsigned long pfn;
-#ifdef CONFIG_DISCONTIGMEM
- int i;
-
- for (i = 0; i < npmem_ranges; i++) {
- for (pfn = node_start_pfn(i); pfn < node_end_pfn(i); pfn++) {
- if (PageReserved(pfn_to_page(pfn)))
- reservedpages++;
- }
- }
-#else /* !CONFIG_DISCONTIGMEM */
- for (pfn = 0; pfn < max_pfn; pfn++) {
- /*
- * Only count reserved RAM pages
- */
- if (PageReserved(pfn_to_page(pfn)))
- reservedpages++;
- }
-#endif
-}
-
#ifdef CONFIG_PA11
if (hppa_dma_ops == &pcxl_dma_ops) {
pcxl_dma_start = (unsigned long)SET_MAP_OFFSET(MAP_START);
@@ -643,15 +611,7 @@ void __init mem_init(void)
parisc_vmalloc_start = SET_MAP_OFFSET(MAP_START);
#endif

- printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10
- );
-
+ mem_init_print_info(NULL);
#ifdef CONFIG_DEBUG_KERNEL /* double-sanity-check paranoia */
printk("virtual kernel memory layout:\n"
" vmalloc : 0x%p - 0x%p (%4ld MB)\n"
@@ -1099,6 +1059,6 @@ void flush_tlb_all(void)
#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
- num_physpages += free_reserved_area(start, end, -1, "initrd");
+ free_reserved_area(start, end, -1, "initrd");
}
#endif
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:28 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Benjamin Herrenschmidt <***@kernel.crashing.org>
Cc: Paul Mackerras <***@samba.org>
Cc: linuxppc-***@lists.ozlabs.org
Cc: linux-***@vger.kernel.org
---
arch/powerpc/mm/mem.c | 56 +++++++++++--------------------------------------
1 file changed, 12 insertions(+), 44 deletions(-)

diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index b890245..4e24f1c 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -299,46 +299,27 @@ void __init paging_init(void)

void __init mem_init(void)
{
-#ifdef CONFIG_NEED_MULTIPLE_NODES
- int nid;
-#endif
- pg_data_t *pgdat;
- unsigned long i;
- struct page *page;
- unsigned long reservedpages = 0, codesize, initsize, datasize, bsssize;
-
#ifdef CONFIG_SWIOTLB
swiotlb_init(0);
#endif

- num_physpages = memblock_phys_mem_size() >> PAGE_SHIFT;
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);

#ifdef CONFIG_NEED_MULTIPLE_NODES
- for_each_online_node(nid) {
- if (NODE_DATA(nid)->node_spanned_pages != 0) {
- printk("freeing bootmem node %d\n", nid);
- free_all_bootmem_node(NODE_DATA(nid));
- }
+ {
+ pg_data_t *pgdat;
+
+ for_each_online_pgdat(pgdat)
+ if (pgdat->node_spanned_pages != 0) {
+ printk("freeing bootmem node %d\n",
+ pgdat->node_id);
+ free_all_bootmem_node(pgdat);
+ }
}
#else
max_mapnr = max_pfn;
free_all_bootmem();
#endif
- for_each_online_pgdat(pgdat) {
- for (i = 0; i < pgdat->node_spanned_pages; i++) {
- if (!pfn_valid(pgdat->node_start_pfn + i))
- continue;
- page = pgdat_page_nr(pgdat, i);
- if (PageReserved(page))
- reservedpages++;
- }
- }
-
- codesize = (unsigned long)&_sdata - (unsigned long)&_stext;
- datasize = (unsigned long)&_edata - (unsigned long)&_sdata;
- initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin;
- bsssize = (unsigned long)&__bss_stop - (unsigned long)&__bss_start;

#ifdef CONFIG_HIGHMEM
{
@@ -348,13 +329,9 @@ void __init mem_init(void)
for (pfn = highmem_mapnr; pfn < max_mapnr; ++pfn) {
phys_addr_t paddr = (phys_addr_t)pfn << PAGE_SHIFT;
struct page *page = pfn_to_page(pfn);
- if (memblock_is_reserved(paddr))
- continue;
- free_highmem_page(page);
- reservedpages--;
+ if (!memblock_is_reserved(paddr))
+ free_highmem_page(page);
}
- printk(KERN_DEBUG "High memory: %luk\n",
- totalhigh_pages << (PAGE_SHIFT-10));
}
#endif /* CONFIG_HIGHMEM */

@@ -367,16 +344,7 @@ void __init mem_init(void)
(mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY) - 1;
#endif

- printk(KERN_INFO "Memory: %luk/%luk available (%luk kernel code, "
- "%luk reserved, %luk data, %luk bss, %luk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- bsssize >> 10,
- initsize >> 10);
-
+ mem_init_print_info(NULL);
#ifdef CONFIG_PPC32
pr_info("Kernel virtual memory layout:\n");
pr_info(" * 0x%08lx..0x%08lx : fixmap\n", FIXADDR_START, FIXADDR_TOP);
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Benjamin Herrenschmidt
2013-05-15 00:32:34 UTC
Permalink
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
No objection, I haven't had a chance to actually build/boot test though.

BTW. A recommended way of doing so which is pretty easy even if you
don't have access to powerpc hardware nowadays is to use
qemu-system-ppc64 with -M pseries.

You can find cross compilers for the kernel on kernel.org and you can
feed qemu with some distro installer ISO.

Cheers,
Ben.
Post by Jiang Liu
---
arch/powerpc/mm/mem.c | 56 +++++++++++--------------------------------------
1 file changed, 12 insertions(+), 44 deletions(-)
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index b890245..4e24f1c 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -299,46 +299,27 @@ void __init paging_init(void)
void __init mem_init(void)
{
-#ifdef CONFIG_NEED_MULTIPLE_NODES
- int nid;
-#endif
- pg_data_t *pgdat;
- unsigned long i;
- struct page *page;
- unsigned long reservedpages = 0, codesize, initsize, datasize, bsssize;
-
#ifdef CONFIG_SWIOTLB
swiotlb_init(0);
#endif
- num_physpages = memblock_phys_mem_size() >> PAGE_SHIFT;
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);
#ifdef CONFIG_NEED_MULTIPLE_NODES
- for_each_online_node(nid) {
- if (NODE_DATA(nid)->node_spanned_pages != 0) {
- printk("freeing bootmem node %d\n", nid);
- free_all_bootmem_node(NODE_DATA(nid));
- }
+ {
+ pg_data_t *pgdat;
+
+ for_each_online_pgdat(pgdat)
+ if (pgdat->node_spanned_pages != 0) {
+ printk("freeing bootmem node %d\n",
+ pgdat->node_id);
+ free_all_bootmem_node(pgdat);
+ }
}
#else
max_mapnr = max_pfn;
free_all_bootmem();
#endif
- for_each_online_pgdat(pgdat) {
- for (i = 0; i < pgdat->node_spanned_pages; i++) {
- if (!pfn_valid(pgdat->node_start_pfn + i))
- continue;
- page = pgdat_page_nr(pgdat, i);
- if (PageReserved(page))
- reservedpages++;
- }
- }
-
- codesize = (unsigned long)&_sdata - (unsigned long)&_stext;
- datasize = (unsigned long)&_edata - (unsigned long)&_sdata;
- initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin;
- bsssize = (unsigned long)&__bss_stop - (unsigned long)&__bss_start;
#ifdef CONFIG_HIGHMEM
{
@@ -348,13 +329,9 @@ void __init mem_init(void)
for (pfn = highmem_mapnr; pfn < max_mapnr; ++pfn) {
phys_addr_t paddr = (phys_addr_t)pfn << PAGE_SHIFT;
struct page *page = pfn_to_page(pfn);
- if (memblock_is_reserved(paddr))
- continue;
- free_highmem_page(page);
- reservedpages--;
+ if (!memblock_is_reserved(paddr))
+ free_highmem_page(page);
}
- printk(KERN_DEBUG "High memory: %luk\n",
- totalhigh_pages << (PAGE_SHIFT-10));
}
#endif /* CONFIG_HIGHMEM */
@@ -367,16 +344,7 @@ void __init mem_init(void)
(mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY) - 1;
#endif
- printk(KERN_INFO "Memory: %luk/%luk available (%luk kernel code, "
- "%luk reserved, %luk data, %luk bss, %luk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- bsssize >> 10,
- initsize >> 10);
-
+ mem_init_print_info(NULL);
#ifdef CONFIG_PPC32
pr_info("Kernel virtual memory layout:\n");
pr_info(" * 0x%08lx..0x%08lx : fixmap\n", FIXADDR_START, FIXADDR_TOP);
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Liu Jiang
2013-05-15 15:49:42 UTC
Permalink
Post by Benjamin Herrenschmidt
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
No objection, I haven't had a chance to actually build/boot test though.
BTW. A recommended way of doing so which is pretty easy even if you
don't have access to powerpc hardware nowadays is to use
qemu-system-ppc64 with -M pseries.
You can find cross compilers for the kernel on kernel.org and you can
feed qemu with some distro installer ISO.
Hi Benjamin,'
Thanks for review! As could I assume an "Acked-by"?
I have installed all cross-compiler from ftp.kernel.org and done
basic
building tests for these patches. But I haven't run new kernel with qemu
yet, that's a really good suggestion, will try it next time.
Regards!
Gerry
Post by Benjamin Herrenschmidt
Cheers,
Ben.
Post by Jiang Liu
---
arch/powerpc/mm/mem.c | 56 +++++++++++--------------------------------------
1 file changed, 12 insertions(+), 44 deletions(-)
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index b890245..4e24f1c 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -299,46 +299,27 @@ void __init paging_init(void)
void __init mem_init(void)
{
-#ifdef CONFIG_NEED_MULTIPLE_NODES
- int nid;
-#endif
- pg_data_t *pgdat;
- unsigned long i;
- struct page *page;
- unsigned long reservedpages = 0, codesize, initsize, datasize, bsssize;
-
#ifdef CONFIG_SWIOTLB
swiotlb_init(0);
#endif
- num_physpages = memblock_phys_mem_size() >> PAGE_SHIFT;
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);
#ifdef CONFIG_NEED_MULTIPLE_NODES
- for_each_online_node(nid) {
- if (NODE_DATA(nid)->node_spanned_pages != 0) {
- printk("freeing bootmem node %d\n", nid);
- free_all_bootmem_node(NODE_DATA(nid));
- }
+ {
+ pg_data_t *pgdat;
+
+ for_each_online_pgdat(pgdat)
+ if (pgdat->node_spanned_pages != 0) {
+ printk("freeing bootmem node %d\n",
+ pgdat->node_id);
+ free_all_bootmem_node(pgdat);
+ }
}
#else
max_mapnr = max_pfn;
free_all_bootmem();
#endif
- for_each_online_pgdat(pgdat) {
- for (i = 0; i < pgdat->node_spanned_pages; i++) {
- if (!pfn_valid(pgdat->node_start_pfn + i))
- continue;
- page = pgdat_page_nr(pgdat, i);
- if (PageReserved(page))
- reservedpages++;
- }
- }
-
- codesize = (unsigned long)&_sdata - (unsigned long)&_stext;
- datasize = (unsigned long)&_edata - (unsigned long)&_sdata;
- initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin;
- bsssize = (unsigned long)&__bss_stop - (unsigned long)&__bss_start;
#ifdef CONFIG_HIGHMEM
{
@@ -348,13 +329,9 @@ void __init mem_init(void)
for (pfn = highmem_mapnr; pfn < max_mapnr; ++pfn) {
phys_addr_t paddr = (phys_addr_t)pfn << PAGE_SHIFT;
struct page *page = pfn_to_page(pfn);
- if (memblock_is_reserved(paddr))
- continue;
- free_highmem_page(page);
- reservedpages--;
+ if (!memblock_is_reserved(paddr))
+ free_highmem_page(page);
}
- printk(KERN_DEBUG "High memory: %luk\n",
- totalhigh_pages << (PAGE_SHIFT-10));
}
#endif /* CONFIG_HIGHMEM */
@@ -367,16 +344,7 @@ void __init mem_init(void)
(mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY) - 1;
#endif
- printk(KERN_INFO "Memory: %luk/%luk available (%luk kernel code, "
- "%luk reserved, %luk data, %luk bss, %luk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- bsssize >> 10,
- initsize >> 10);
-
+ mem_init_print_info(NULL);
#ifdef CONFIG_PPC32
pr_info("Kernel virtual memory layout:\n");
pr_info(" * 0x%08lx..0x%08lx : fixmap\n", FIXADDR_START, FIXADDR_TOP);
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:29 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Martin Schwidefsky <***@de.ibm.com>
Cc: Heiko Carstens <***@de.ibm.com>
Cc: ***@de.ibm.com
Cc: linux-***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
arch/s390/mm/init.c | 17 ++---------------
1 file changed, 2 insertions(+), 15 deletions(-)

diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index 24d52aa..771c27a 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -133,9 +133,7 @@ void __init paging_init(void)

void __init mem_init(void)
{
- unsigned long codesize, reservedpages, datasize, initsize;
-
- max_mapnr = num_physpages = max_low_pfn;
+ max_mapnr = max_low_pfn;
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);

/* Setup guest page hinting */
@@ -145,18 +143,7 @@ void __init mem_init(void)
free_all_bootmem();
setup_zero_pages(); /* Setup zeroed pages. */

- reservedpages = 0;
-
- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- datasize = (unsigned long) &_edata - (unsigned long) &_etext;
- initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
- printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, %ldk data, %ldk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- max_mapnr << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >>10,
- initsize >> 10);
+ mem_init_print_info(NULL);
printk("Write protected kernel read-only data: %#lx - %#lx\n",
(unsigned long)&_stext,
PFN_ALIGN((unsigned long)&_eshared) - 1);
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:31 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Paul Mundt <***@linux-sh.org>
Cc: Wen Congyang <***@cn.fujitsu.com>
Cc: Tang Chen <***@cn.fujitsu.com>
Cc: linux-***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
arch/sh/mm/init.c | 25 ++++---------------------
1 file changed, 4 insertions(+), 21 deletions(-)

diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index aecd913..3826596 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -407,24 +407,18 @@ unsigned int mem_init_done = 0;

void __init mem_init(void)
{
- int codesize, datasize, initsize;
- int nid;
+ pg_data_t *pgdat;

iommu_init();

- num_physpages = 0;
high_memory = NULL;

- for_each_online_node(nid) {
- pg_data_t *pgdat = NODE_DATA(nid);
+ for_each_online_pgdat(pgdat) {
void *node_high_memory;

- num_physpages += pgdat->node_present_pages;
-
if (pgdat->node_spanned_pages)
free_all_bootmem_node(pgdat);

-
node_high_memory = (void *)__va((pgdat->node_start_pfn +
pgdat->node_spanned_pages) <<
PAGE_SHIFT);
@@ -441,19 +435,8 @@ void __init mem_init(void)

vsyscall_init();

- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- 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 data, %dk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- datasize >> 10,
- initsize >> 10);
-
- printk(KERN_INFO "virtual kernel memory layout:\n"
+ mem_init_print_info(NULL);
+ pr_info("virtual kernel memory layout:\n"
" fixmap : 0x%08lx - 0x%08lx (%4ld kB)\n"
#ifdef CONFIG_HIGHMEM
" pkmap : 0x%08lx - 0x%08lx (%4ld kB)\n"
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:30 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Chen Liqin <***@sunplusct.com>
Cc: Lennox Wu <***@gmail.com>
Cc: linux-***@vger.kernel.org
---
arch/score/mm/init.c | 26 ++------------------------
1 file changed, 2 insertions(+), 24 deletions(-)

diff --git a/arch/score/mm/init.c b/arch/score/mm/init.c
index 579fc4e..2a223d8 100644
--- a/arch/score/mm/init.c
+++ b/arch/score/mm/init.c
@@ -77,33 +77,11 @@ void __init paging_init(void)

void __init mem_init(void)
{
- unsigned long codesize, reservedpages, datasize, initsize;
- unsigned long tmp, ram = 0;
-
high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT);
free_all_bootmem();
setup_zero_page(); /* Setup zeroed pages. */
- reservedpages = 0;
-
- for (tmp = 0; tmp < max_low_pfn; tmp++)
- if (page_is_ram(tmp)) {
- ram++;
- if (PageReserved(pfn_to_page(tmp)))
- reservedpages++;
- }
-
- num_physpages = ram;
- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- datasize = (unsigned long) &_edata - (unsigned long) &_etext;
- initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
-
- printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
- "%ldk reserved, %ldk data, %ldk init, %ldk highmem)\n",
- (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
- ram << (PAGE_SHIFT-10), codesize >> 10,
- reservedpages << (PAGE_SHIFT-10), datasize >> 10,
- initsize >> 10,
- totalhigh_pages << (PAGE_SHIFT-10));
+
+ mem_init_print_info(NULL);
}
#endif /* !CONFIG_NEED_MULTIPLE_NODES */
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:32 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: "David S. Miller" <***@davemloft.net>
Cc: Sam Ravnborg <***@ravnborg.org>
Cc: Yasuaki Ishimatsu <***@jp.fujitsu.com>
Cc: Tang Chen <***@cn.fujitsu.com>
Cc: ***@vger.kernel.org
Cc: linux-***@vger.kernel.org
---
arch/sparc/kernel/leon_smp.c | 3 ---
arch/sparc/mm/init_32.c | 33 +++------------------------------
arch/sparc/mm/init_64.c | 23 ++---------------------
3 files changed, 5 insertions(+), 54 deletions(-)

diff --git a/arch/sparc/kernel/leon_smp.c b/arch/sparc/kernel/leon_smp.c
index 6cfc1b0..d7aa524 100644
--- a/arch/sparc/kernel/leon_smp.c
+++ b/arch/sparc/kernel/leon_smp.c
@@ -254,15 +254,12 @@ void __init leon_smp_done(void)
/* Free unneeded trap tables */
if (!cpu_present(1)) {
free_reserved_page(virt_to_page(&trapbase_cpu1));
- num_physpages++;
}
if (!cpu_present(2)) {
free_reserved_page(virt_to_page(&trapbase_cpu2));
- num_physpages++;
}
if (!cpu_present(3)) {
free_reserved_page(virt_to_page(&trapbase_cpu3));
- num_physpages++;
}
/* Ok, they are spinning and ready to go. */
smp_processors_ready = 1;
diff --git a/arch/sparc/mm/init_32.c b/arch/sparc/mm/init_32.c
index e96afed..25d10cf 100644
--- a/arch/sparc/mm/init_32.c
+++ b/arch/sparc/mm/init_32.c
@@ -288,10 +288,6 @@ static void map_high_region(unsigned long start_pfn, unsigned long end_pfn)

void __init mem_init(void)
{
- int codepages = 0;
- int datapages = 0;
- int initpages = 0;
- int reservedpages = 0;
int i;

if (PKMAP_BASE+LAST_PKMAP*PAGE_SIZE >= FIXADDR_START) {
@@ -329,8 +325,6 @@ void __init mem_init(void)
unsigned long start_pfn = sp_banks[i].base_addr >> PAGE_SHIFT;
unsigned long end_pfn = (sp_banks[i].base_addr + sp_banks[i].num_bytes) >> PAGE_SHIFT;

- num_physpages += sp_banks[i].num_bytes >> PAGE_SHIFT;
-
if (end_pfn <= highstart_pfn)
continue;

@@ -340,39 +334,18 @@ void __init mem_init(void)
map_high_region(start_pfn, end_pfn);
}

- codepages = (((unsigned long) &_etext) - ((unsigned long)&_start));
- codepages = PAGE_ALIGN(codepages) >> PAGE_SHIFT;
- datapages = (((unsigned long) &_edata) - ((unsigned long)&_etext));
- datapages = PAGE_ALIGN(datapages) >> PAGE_SHIFT;
- initpages = (((unsigned long) &__init_end) - ((unsigned long) &__init_begin));
- initpages = PAGE_ALIGN(initpages) >> PAGE_SHIFT;
-
- /* Ignore memory holes for the purpose of counting reserved pages */
- for (i=0; i < max_low_pfn; i++)
- if (test_bit(i >> (20 - PAGE_SHIFT), sparc_valid_addr_bitmap)
- && PageReserved(pfn_to_page(i)))
- reservedpages++;
-
- printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init, %ldk highmem)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT - 10),
- codepages << (PAGE_SHIFT-10),
- reservedpages << (PAGE_SHIFT - 10),
- datapages << (PAGE_SHIFT-10),
- initpages << (PAGE_SHIFT-10),
- totalhigh_pages << (PAGE_SHIFT-10));
+ mem_init_print_info(NULL);
}

void free_initmem (void)
{
- num_physpages += free_initmem_default(POISON_FREE_INITMEM);
+ free_initmem_default(POISON_FREE_INITMEM);
}

#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
- num_physpages += free_reserved_area(start, end, POISON_FREE_INITMEM,
- "initrd");
+ free_reserved_area(start, end, POISON_FREE_INITMEM, "initrd");
}
#endif

diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index c93c7c2..b0e6b8f 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -2038,7 +2038,6 @@ static void __init register_page_bootmem_info(void)
}
void __init mem_init(void)
{
- unsigned long codepages, datapages, initpages;
unsigned long addr, last;

addr = PAGE_OFFSET + kern_base;
@@ -2056,11 +2055,6 @@ void __init mem_init(void)
register_page_bootmem_info();
free_all_bootmem();

- /* We subtract one to account for the mem_map_zero page
- * allocated below.
- */
- num_physpages = totalram_pages - 1;
-
/*
* Set up the zero page, mark it reserved, so that page count
* is not manipulated when freeing the page from user ptes.
@@ -2072,19 +2066,7 @@ void __init mem_init(void)
}
mark_page_reserved(mem_map_zero);

- codepages = (((unsigned long) _etext) - ((unsigned long) _start));
- codepages = PAGE_ALIGN(codepages) >> PAGE_SHIFT;
- datapages = (((unsigned long) _edata) - ((unsigned long) _etext));
- datapages = PAGE_ALIGN(datapages) >> PAGE_SHIFT;
- initpages = (((unsigned long) __init_end) - ((unsigned long) __init_begin));
- initpages = PAGE_ALIGN(initpages) >> PAGE_SHIFT;
-
- printk("Memory: %luk available (%ldk kernel code, %ldk data, %ldk init) [%016lx,%016lx]\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- codepages << (PAGE_SHIFT-10),
- datapages << (PAGE_SHIFT-10),
- initpages << (PAGE_SHIFT-10),
- PAGE_OFFSET, (last_valid_pfn << PAGE_SHIFT));
+ mem_init_print_info(NULL);

if (tlb_type == cheetah || tlb_type == cheetah_plus)
cheetah_ecache_flush_init();
@@ -2124,8 +2106,7 @@ void free_initmem(void)
#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
- num_physpages += free_reserved_area(start, end, POISON_FREE_INITMEM,
- "initrd");
+ free_reserved_area(start, end, POISON_FREE_INITMEM, "initrd");
}
#endif
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Sam Ravnborg
2013-05-08 19:02:08 UTC
Permalink
Post by Jiang Liu
Prepare for removing num_physpages and simplify mem_init().
Acked-by: Sam Ravnborg <***@ravnborg.org>
Jiang Liu
2013-05-08 15:51:33 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Acked-by: Chris Metcalf <***@tilera.com>
Cc: Bjorn Helgaas <***@google.com>
Cc: "David S. Miller" <***@davemloft.net>
Cc: Wen Congyang <***@cn.fujitsu.com>
Cc: linux-***@vger.kernel.org
---
arch/tile/kernel/setup.c | 16 ++++++++--------
arch/tile/mm/init.c | 15 +--------------
2 files changed, 9 insertions(+), 22 deletions(-)

diff --git a/arch/tile/kernel/setup.c b/arch/tile/kernel/setup.c
index 41818e3..68b5426 100644
--- a/arch/tile/kernel/setup.c
+++ b/arch/tile/kernel/setup.c
@@ -329,6 +329,7 @@ static void __init setup_memory(void)
#if defined(CONFIG_HIGHMEM) || defined(__tilegx__)
long lowmem_pages;
#endif
+ unsigned long physpages = 0;

/* We are using a char to hold the cpu_2_node[] mapping */
BUILD_BUG_ON(MAX_NUMNODES > 127);
@@ -388,8 +389,8 @@ static void __init setup_memory(void)
continue;
}
}
- if (num_physpages + PFN_DOWN(range.size) > maxmem_pfn) {
- int max_size = maxmem_pfn - num_physpages;
+ if (physpages + PFN_DOWN(range.size) > maxmem_pfn) {
+ int max_size = maxmem_pfn - physpages;
if (max_size > 0) {
pr_err("Maxmem reduced node %d to %d pages\n",
i, max_size);
@@ -446,7 +447,7 @@ static void __init setup_memory(void)
node_start_pfn[i] = start;
node_end_pfn[i] = end;
node_controller[i] = range.controller;
- num_physpages += size;
+ physpages += size;
max_pfn = end;

/* Mark node as online */
@@ -465,7 +466,7 @@ static void __init setup_memory(void)
* we're willing to use at 8 million pages (32GB of 4KB pages).
*/
cap = 8 * 1024 * 1024; /* 8 million pages */
- if (num_physpages > cap) {
+ if (physpages > cap) {
int num_nodes = num_online_nodes();
int cap_each = cap / num_nodes;
unsigned long dropped_pages = 0;
@@ -476,10 +477,10 @@ static void __init setup_memory(void)
node_end_pfn[i] = node_start_pfn[i] + cap_each;
}
}
- num_physpages -= dropped_pages;
+ physpages -= dropped_pages;
pr_warning("Only using %ldMB memory;"
" ignoring %ldMB.\n",
- num_physpages >> (20 - PAGE_SHIFT),
+ physpages >> (20 - PAGE_SHIFT),
dropped_pages >> (20 - PAGE_SHIFT));
pr_warning("Consider using a larger page size.\n");
}
@@ -497,7 +498,7 @@ static void __init setup_memory(void)

lowmem_pages = (mappable_physpages > MAXMEM_PFN) ?
MAXMEM_PFN : mappable_physpages;
- highmem_pages = (long) (num_physpages - lowmem_pages);
+ highmem_pages = (long) (physpages - lowmem_pages);

pr_notice("%ldMB HIGHMEM available.\n",
pages_to_mb(highmem_pages > 0 ? highmem_pages : 0));
@@ -514,7 +515,6 @@ static void __init setup_memory(void)
pr_warning("Use a HIGHMEM enabled kernel.\n");
max_low_pfn = MAXMEM_PFN;
max_pfn = MAXMEM_PFN;
- num_physpages = MAXMEM_PFN;
node_end_pfn[0] = MAXMEM_PFN;
} else {
pr_notice("%ldMB memory available.\n",
diff --git a/arch/tile/mm/init.c b/arch/tile/mm/init.c
index f2ac2f4..e182958 100644
--- a/arch/tile/mm/init.c
+++ b/arch/tile/mm/init.c
@@ -821,7 +821,6 @@ static void __init set_max_mapnr_init(void)

void __init mem_init(void)
{
- int codesize, datasize, initsize;
int i;
#ifndef __tilegx__
void *last;
@@ -853,19 +852,7 @@ void __init mem_init(void)
set_non_bootmem_pages_init();
#endif

- codesize = (unsigned long)&_etext - (unsigned long)&_text;
- datasize = (unsigned long)&_end - (unsigned long)&_sdata;
- initsize = (unsigned long)&_einittext - (unsigned long)&_sinittext;
- initsize += (unsigned long)&_einitdata - (unsigned long)&_sinitdata;
-
- pr_info("Memory: %luk/%luk available (%dk kernel code, %dk data, %dk init, %ldk highmem)\n",
- (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- datasize >> 10,
- initsize >> 10,
- (unsigned long) (totalhigh_pages << (PAGE_SHIFT-10))
- );
+ mem_init_print_info(NULL);

/*
* In debug mode, dump some interesting memory mappings.
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:34 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Jeff Dike <***@addtoit.com>
Cc: Richard Weinberger <***@nod.at>
Cc: user-mode-linux-***@lists.sourceforge.net
Cc: linux-***@vger.kernel.org
---
arch/um/kernel/mem.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index a7dc6c1..819008f 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -70,10 +70,8 @@ void __init mem_init(void)
#ifdef CONFIG_HIGHMEM
setup_highmem(end_iomem, highmem);
#endif
- num_physpages = totalram_pages;
max_pfn = totalram_pages;
- printk(KERN_INFO "Memory: %luk available\n",
- nr_free_pages() << (PAGE_SHIFT-10));
+ mem_init_print_info(NULL);
kmalloc_ok = 1;
}
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:35 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Guan Xuetao <***@mprc.pku.edu.cn>
Cc: Michal Hocko <***@suse.cz>
Cc: David Rientjes <***@google.com>
Cc: linux-***@vger.kernel.org
---
arch/unicore32/mm/init.c | 49 ++--------------------------------------------
1 file changed, 2 insertions(+), 47 deletions(-)

diff --git a/arch/unicore32/mm/init.c b/arch/unicore32/mm/init.c
index 119b9e8..39a967a 100644
--- a/arch/unicore32/mm/init.c
+++ b/arch/unicore32/mm/init.c
@@ -383,10 +383,6 @@ static void __init free_unused_memmap(struct meminfo *mi)
*/
void __init mem_init(void)
{
- unsigned long reserved_pages, free_pages;
- struct memblock_region *reg;
- int i;
-
max_mapnr = pfn_to_page(max_pfn + PHYS_PFN_OFFSET) - mem_map;

free_unused_memmap(&meminfo);
@@ -394,48 +390,7 @@ void __init mem_init(void)
/* this will put all unused low memory onto the freelists */
free_all_bootmem();

- reserved_pages = free_pages = 0;
-
- for_each_bank(i, &meminfo) {
- struct membank *bank = &meminfo.bank[i];
- unsigned int pfn1, pfn2;
- struct page *page, *end;
-
- pfn1 = bank_pfn_start(bank);
- pfn2 = bank_pfn_end(bank);
-
- page = pfn_to_page(pfn1);
- end = pfn_to_page(pfn2 - 1) + 1;
-
- do {
- if (PageReserved(page))
- reserved_pages++;
- else if (!page_count(page))
- free_pages++;
- page++;
- } while (page < end);
- }
-
- /*
- * Since our memory may not be contiguous, calculate the
- * real number of pages we have in this system
- */
- printk(KERN_INFO "Memory:");
- num_physpages = 0;
- for_each_memblock(memory, reg) {
- unsigned long pages = memblock_region_memory_end_pfn(reg) -
- memblock_region_memory_base_pfn(reg);
- num_physpages += pages;
- printk(" %ldMB", pages >> (20 - PAGE_SHIFT));
- }
- printk(" = %luMB total\n", num_physpages >> (20 - PAGE_SHIFT));
-
- printk(KERN_NOTICE "Memory: %luk/%luk available, %luk reserved, %luK highmem\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- free_pages << (PAGE_SHIFT-10),
- reserved_pages << (PAGE_SHIFT-10),
- totalhigh_pages << (PAGE_SHIFT-10));
-
+ mem_init_print_info(NULL);
printk(KERN_NOTICE "Virtual kernel memory layout:\n"
" vector : 0x%08lx - 0x%08lx (%4ld kB)\n"
" vmalloc : 0x%08lx - 0x%08lx (%4ld MB)\n"
@@ -464,7 +419,7 @@ void __init mem_init(void)
BUILD_BUG_ON(TASK_SIZE > MODULES_VADDR);
BUG_ON(TASK_SIZE > MODULES_VADDR);

- if (PAGE_SIZE >= 16384 && num_physpages <= 128) {
+ if (PAGE_SIZE >= 16384 && get_num_physpages() <= 128) {
/*
* On a machine this small we won't get
* anywhere without overcommit, so turn
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:36 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Thomas Gleixner <***@linutronix.de>
Cc: Ingo Molnar <***@redhat.com>
Cc: "H. Peter Anvin" <***@zytor.com>
Cc: ***@kernel.org
Cc: Andreas Herrmann <***@amd.com>
Cc: Tang Chen <***@cn.fujitsu.com>
Cc: Wen Congyang <***@cn.fujitsu.com>
Cc: Jianguo Wu <***@huawei.com>
Cc: linux-***@vger.kernel.org
---
arch/x86/kernel/cpu/amd.c | 2 +-
arch/x86/kernel/setup.c | 2 --
arch/x86/mm/init_32.c | 30 ++----------------------------
arch/x86/mm/init_64.c | 20 +-------------------
arch/x86/mm/numa_32.c | 2 --
5 files changed, 4 insertions(+), 52 deletions(-)

diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 5013a48..c587a87 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -90,7 +90,7 @@ static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c)
static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c)
{
u32 l, h;
- int mbytes = num_physpages >> (20-PAGE_SHIFT);
+ int mbytes = get_num_physpages() >> (20-PAGE_SHIFT);

if (c->x86_model < 6) {
/* Based on AMD doc 20734R - June 2000 */
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index c8dbdfd..58f1a8ad 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -1041,8 +1041,6 @@ void __init setup_arch(char **cmdline_p)
/* max_low_pfn get updated here */
find_low_pfn_range();
#else
- num_physpages = max_pfn;
-
check_x2apic();

/* How many end-of-memory variables you have, grandma! */
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index 9fa46ba..4287f1f 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -660,10 +660,8 @@ void __init initmem_init(void)
highstart_pfn = max_low_pfn;
printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
pages_to_mb(highend_pfn - highstart_pfn));
- num_physpages = highend_pfn;
high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
#else
- num_physpages = max_low_pfn;
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
#endif

@@ -671,7 +669,7 @@ void __init initmem_init(void)
sparse_memory_present_with_active_regions(0);

#ifdef CONFIG_FLATMEM
- max_mapnr = num_physpages;
+ max_mapnr = IS_ENABLED(CONFIG_HIGHMEM) ? highend_pfn : max_low_pfn;
#endif
__vmalloc_start_set = true;

@@ -739,9 +737,6 @@ static void __init test_wp_bit(void)

void __init mem_init(void)
{
- int codesize, reservedpages, datasize, initsize;
- int tmp;
-
pci_iommu_alloc();

#ifdef CONFIG_FLATMEM
@@ -761,30 +756,9 @@ void __init mem_init(void)
/* this will put all low memory onto the freelists */
free_all_bootmem();

- reservedpages = 0;
- for (tmp = 0; tmp < max_low_pfn; tmp++)
- /*
- * Only count reserved RAM pages:
- */
- if (page_is_ram(tmp) && PageReserved(pfn_to_page(tmp)))
- reservedpages++;
-
after_bootmem = 1;

- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- 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",
- nr_free_pages() << (PAGE_SHIFT-10),
- num_physpages << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10,
- totalhigh_pages << (PAGE_SHIFT-10));
-
+ mem_init_print_info(NULL);
printk(KERN_INFO "virtual kernel memory layout:\n"
" fixmap : 0x%08lx - 0x%08lx (%4ld kB)\n"
#ifdef CONFIG_HIGHMEM
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 65116e5..650264b 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -1043,9 +1043,6 @@ static void __init register_page_bootmem_info(void)

void __init mem_init(void)
{
- long codesize, reservedpages, datasize, initsize;
- unsigned long absent_pages;
-
pci_iommu_alloc();

/* clear_bss() already clear the empty_zero_page */
@@ -1054,28 +1051,13 @@ void __init mem_init(void)

/* this will put all memory onto the freelists */
free_all_bootmem();
-
- absent_pages = absent_pages_in_range(0, max_pfn);
- reservedpages = max_pfn - totalram_pages - absent_pages;
after_bootmem = 1;

- codesize = (unsigned long) &_etext - (unsigned long) &_text;
- datasize = (unsigned long) &_edata - (unsigned long) &_etext;
- initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
-
/* Register memory areas for /proc/kcore */
kclist_add(&kcore_vsyscall, (void *)VSYSCALL_START,
VSYSCALL_END - VSYSCALL_START, KCORE_OTHER);

- printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
- "%ldk absent, %ldk reserved, %ldk data, %ldk init)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- max_pfn << (PAGE_SHIFT-10),
- codesize >> 10,
- absent_pages << (PAGE_SHIFT-10),
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10);
+ mem_init_print_info(NULL);
}

#ifdef CONFIG_DEBUG_RODATA
diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c
index 73a6d73..0342d27 100644
--- a/arch/x86/mm/numa_32.c
+++ b/arch/x86/mm/numa_32.c
@@ -83,10 +83,8 @@ void __init initmem_init(void)
highstart_pfn = max_low_pfn;
printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
pages_to_mb(highend_pfn - highstart_pfn));
- num_physpages = highend_pfn;
high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
#else
- num_physpages = max_low_pfn;
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
#endif
printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:37 UTC
Permalink
Prepare for removing num_physpages and simplify mem_init().

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Chris Zankel <***@zankel.net>
Cc: Max Filippov <***@gmail.com>
Cc: Geert Uytterhoeven <***@linux-m68k.org>
Cc: linux-***@linux-xtensa.org
Cc: linux-***@vger.kernel.org
---
arch/xtensa/mm/init.c | 27 ++-------------------------
1 file changed, 2 insertions(+), 25 deletions(-)

diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c
index dc6e009..267428e 100644
--- a/arch/xtensa/mm/init.c
+++ b/arch/xtensa/mm/init.c
@@ -173,12 +173,8 @@ void __init zones_init(void)

void __init mem_init(void)
{
- unsigned long codesize, reservedpages, datasize, initsize;
- unsigned long highmemsize, tmp, ram;
-
- max_mapnr = num_physpages = max_low_pfn - ARCH_PFN_OFFSET;
+ max_mapnr = max_low_pfn - ARCH_PFN_OFFSET;
high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT);
- highmemsize = 0;

#ifdef CONFIG_HIGHMEM
#error HIGHGMEM not implemented in init.c
@@ -186,26 +182,7 @@ void __init mem_init(void)

free_all_bootmem();

- reservedpages = ram = 0;
- for (tmp = 0; tmp < max_mapnr; tmp++) {
- ram++;
- if (PageReserved(mem_map+tmp))
- reservedpages++;
- }
-
- codesize = (unsigned long) _etext - (unsigned long) _stext;
- datasize = (unsigned long) _edata - (unsigned long) _sdata;
- initsize = (unsigned long) __init_end - (unsigned long) __init_begin;
-
- printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, "
- "%ldk data, %ldk init %ldk highmem)\n",
- nr_free_pages() << (PAGE_SHIFT-10),
- ram << (PAGE_SHIFT-10),
- codesize >> 10,
- reservedpages << (PAGE_SHIFT-10),
- datasize >> 10,
- initsize >> 10,
- highmemsize >> 10);
+ mem_init_print_info(NULL);
}

#ifdef CONFIG_BLK_DEV_INITRD
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Jiang Liu
2013-05-08 15:51:38 UTC
Permalink
Now all references to num_physpages have been removed, so kill it.

Signed-off-by: Jiang Liu <***@huawei.com>
Cc: Mel Gorman <***@suse.de>
Cc: Michel Lespinasse <***@google.com>
Cc: Rik van Riel <***@redhat.com>
Cc: Jiang Liu <***@huawei.com>
Cc: Hugh Dickins <***@google.com>
Cc: David Rientjes <***@google.com>
Cc: Al Viro <***@zeniv.linux.org.uk>
Cc: Konstantin Khlebnikov <***@openvz.org>
Cc: linux-***@kvack.org
Cc: linux-***@vger.kernel.org
---
include/linux/mm.h | 1 -
mm/memory.c | 2 --
mm/nommu.c | 2 --
3 files changed, 5 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 66e5fb8..f02b34f 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -29,7 +29,6 @@ struct writeback_control;
extern unsigned long max_mapnr;
#endif

-extern unsigned long num_physpages;
extern unsigned long totalram_pages;
extern void * high_memory;
extern int page_cluster;
diff --git a/mm/memory.c b/mm/memory.c
index f7a1fba..0022366 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -82,7 +82,6 @@ EXPORT_SYMBOL(max_mapnr);
EXPORT_SYMBOL(mem_map);
#endif

-unsigned long num_physpages;
/*
* A number of key systems in x86 including ioremap() rely on the assumption
* that high_memory defines the upper bound on direct map memory, then end
@@ -92,7 +91,6 @@ unsigned long num_physpages;
*/
void * high_memory;

-EXPORT_SYMBOL(num_physpages);
EXPORT_SYMBOL(high_memory);

/*
diff --git a/mm/nommu.c b/mm/nommu.c
index 247ef84..09cfde8 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -56,7 +56,6 @@
void *high_memory;
struct page *mem_map;
unsigned long max_mapnr;
-unsigned long num_physpages;
unsigned long highest_memmap_pfn;
struct percpu_counter vm_committed_as;
int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */
@@ -85,7 +84,6 @@ unsigned long vm_memory_committed(void)
EXPORT_SYMBOL_GPL(vm_memory_committed);

EXPORT_SYMBOL(mem_map);
-EXPORT_SYMBOL(num_physpages);

/* list of mapped, potentially shareable regions */
static struct kmem_cache *vm_region_jar;
--
1.7.9.5

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to ***@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"***@kvack.org"> ***@kvack.org </a>
Loading...