kernel-4.4 branch is ready

[Jaap Jan Meijer]

Hi Chih-Wei,

I tested your work on my HP Omni10 (Bay Trail tablet) and it works great,
thank you again for your great work!

I attached some Bluetooth patches to add missing ACPI ID's for bcm43241
chips and two fixes to get IOCTL working in mixed arch situations. Also
there is a fw package I extracted from the Win10 bcm driver that should go
into /lib/firmware/brcm. After that we could start using btattach for UART
BT devices. Do you agree with this? Should I push to Sourceforge?


The biggest problem I was having with vanilla 4.4 where MMC issues even
with a patch set from Adrian Hunter, I must have messed up somewhere. I'm
running your kernel now with Android from internal eMMC and SD-card and
haven't encountered any I/O errors so far.
One more strange problem, on my tablet with brcmfmac these is no WiFi
strength information from the connected network, other networks do have
information. On my laptop with iwlwifi there are no problems. It started on
kernel 4.2, not sure yet what thats about.

[spin877]

thanks great job.

I would like to report a problem

1) Mouse does not work vmware-Workstation 12 pro
2) with the firmware of the network adapter rtl8723bs on (Bay Trail
tablet)

[ 121.866155] init: Starting service 'wpa_supplicant'...
[ 121.967330] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 121.974317] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 121.974329] Request firmware failed with error 0xfffffffe
[ 121.978213] init: Service 'wpa_supplicant' (pid 3334) exited with status
255
[ 121.978237] init: Service 'wpa_supplicant' (pid 3334) killing any
children in process group
[ 126.980853] init: Starting service 'wpa_supplicant'...
[ 127.018092] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 127.018126] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 127.018131] Request firmware failed with error 0xfffffffe
[ 127.021275] init: Service 'wpa_supplicant' (pid 3336) exited with status
255
[ 127.021293] init: Service 'wpa_supplicant' (pid 3336) killing any
children in process group
[ 132.027654] init: Starting service 'wpa_supplicant'...
[ 132.054851] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 132.054885] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 132.054891] Request firmware failed with error 0xfffffffe
[ 132.058628] init: Service 'wpa_supplicant' (pid 3342) exited with status
255
[ 132.058691] init: Service 'wpa_supplicant' (pid 3342) killing any
children in process group
[ 142.003407] init: Starting service 'wpa_supplicant'...
[ 142.038020] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 142.038070] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 142.038075] Request firmware failed with error 0xfffffffe
[ 142.041793] init: Service 'wpa_supplicant' (pid 3346) exited with status
255
[ 142.041856] init: Service 'wpa_supplicant' (pid 3346) killing any
children in process group

[Povilas Staniulis]

Error -2 is ENOENT (aka. File not found).
Do you have rtlwifi/rtl8723bs_nic.bin in /lib/firmware ?

Post by spin877
thanks great job.
I would like to report a problem
1) Mouse does not work vmware-Workstation 12 pro
2) with the firmware of the network adapter rtl8723bs on (Bay Trail
tablet)
[ 121.866155] init: Starting service 'wpa_supplicant'...
[ 121.967330] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 121.974317] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 121.974329] Request firmware failed with error 0xfffffffe
[ 121.978213] init: Service 'wpa_supplicant' (pid 3334) exited with
status 255
[ 121.978237] init: Service 'wpa_supplicant' (pid 3334) killing any
children in process group
[ 126.980853] init: Starting service 'wpa_supplicant'...
[ 127.018092] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 127.018126] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 127.018131] Request firmware failed with error 0xfffffffe
[ 127.021275] init: Service 'wpa_supplicant' (pid 3336) exited with
status 255
[ 127.021293] init: Service 'wpa_supplicant' (pid 3336) killing any
children in process group
[ 132.027654] init: Starting service 'wpa_supplicant'...
[ 132.054851] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 132.054885] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 132.054891] Request firmware failed with error 0xfffffffe
[ 132.058628] init: Service 'wpa_supplicant' (pid 3342) exited with
status 255
[ 132.058691] init: Service 'wpa_supplicant' (pid 3342) killing any
children in process group
[ 142.003407] init: Starting service 'wpa_supplicant'...
[ 142.038020] rtl8723bs: accquire FW from file:rtlwifi/rtl8723bs_nic.bin
[ 142.038070] rtl8723bs mmc1:0001:1: Direct firmware load for
rtlwifi/rtl8723bs_nic.bin failed with error -2
[ 142.038075] Request firmware failed with error 0xfffffffe
[ 142.041793] init: Service 'wpa_supplicant' (pid 3346) exited with
status 255
[ 142.041856] init: Service 'wpa_supplicant' (pid 3346) killing any
children in process group
--
You received this message because you are subscribed to the Google
Groups "Android-x86" group.
To unsubscribe from this group and stop receiving emails from it, send
Visit this group at https://groups.google.com/group/android-x86.
For more options, visit https://groups.google.com/d/optout.

[spin877]

Il giorno giovedì 28 gennaio 2016 01:54:10 UTC+1, Povilas Staniulis ha

Post by Povilas Staniulis
Error -2 is ENOENT (aka. File not found).
Do you have rtlwifi/rtl8723bs_nic.bin in /lib/firmware ?

I've added and everything is back to work

[Jaap Jan Meijer]

Post by Jaap Jan Meijer
The biggest problem I was having with vanilla 4.4 where MMC issues even
with a patch set from Adrian Hunter, I must have messed up somewhere. I'm
running your kernel now with Android from internal eMMC and SD-card and
haven't encountered any I/O errors so far.

I'm sorry, I spoke to soon. The MMC situation is slightly improved with the
Hunter patches but still the complete MMC bus freezes up after a while.
First ADMA errors and than I/O errors and timeouts. Kernel 4.0.9 seems to
do work better with MMC.


I'm now testing some workarounds. With DMA (sdhci.debug_quirks=0x40) seems
to prevent freezing. Also I read about disabling 1.8V mode, will try that
as well.

[Mario Holzinger]

Tested marshmallow 32bit with kernel 4.4 (onńly include my kernel patch for
GSL1680 I2C touchscreen support)
sorry can not sync lollipop sources atm, drm cound not be pulled from git
repo

On my "Microsoft Surface Pro 2" Wifi still mentioned "Autentication failed"
so no wireless connection can be established (works with JJ Meijer's kernel
repo) ==> Marvell 88W8797 (mwifiex)
On my "TrekStor Surftab Wintron 7.0" Wifi with RTL8723BS wirks well (also
my GSL1680 Touchscreen I've patched)

I know marshmallow is not the best to report back but it is confusing why
Marvell 88W8797 wont work. it seems that ther is someting in android-x86
kernel that breaks my Wifi.
I I find time between getting my 7 inch Tablet working better (need to
implement nearly everry HW support on my own :-) ) I will share my sollution

thanks from my side
Mario

[Mario Holzinger]

Post by Mario Holzinger

Post by Mario Holzinger
Tested marshmallow 32bit with kernel 4.4 (onńly include my kernel patch

for

Post by Mario Holzinger
GSL1680 I2C touchscreen support)
sorry can not sync lollipop sources atm, drm cound not be pulled from

git

Post by Mario Holzinger
repo
On my "Microsoft Surface Pro 2" Wifi still mentioned "Autentication

failed"

Post by Mario Holzinger
so no wireless connection can be established (works with JJ Meijer's

kernel

Post by Mario Holzinger
repo) ==> Marvell 88W8797 (mwifiex)

Hmm, what patches does it have?

Post by Mario Holzinger
On my "TrekStor Surftab Wintron 7.0" Wifi with RTL8723BS wirks well

(also my

Post by Mario Holzinger
GSL1680 Touchscreen I've patched)

Please submit the patch.
My tests are with no modifications except the integration

of https://github.com/onitake/gslx680-acpi for my tablet touchcontroller

Post by Mario Holzinger

Post by Mario Holzinger
I know marshmallow is not the best to report back but it is confusing

why

Post by Mario Holzinger
Marvell 88W8797 wont work. it seems that ther is someting in android-x86
kernel that breaks my Wifi.
I I find time between getting my 7 inch Tablet working better (need to
implement nearly everry HW support on my own :-) ) I will share my

sollution

I do not want to provide a patch yet as long as it is such a dirty one like
it is at the moment
Integration of the complete Github repo into drivers/staging + adapt
driver/staging/Kconfig, driver/staging/Makefile and enable the module to
build (as module) during complete clean android build
I need to place a firmware extracted from the windows Touchscreen driver
(tools included in onitake's repo)
Autostart kernel module at boot time with init.sh

regards Mario

[Mario Holzinger]

This post might be inappropriate. Click to display it.

[youling 257]

the new remix os for pc4.4.2 kernelhas GSLx680 on my bay tabletit
isn't work welli will upload picture.

圚 2016幎1月29日星期五 UTC+8䞊午1:15:07Mario Holzinger写道

Post by Mario Holzinger
only for testing here is the Patch to implement GSLx680 touch controllers.
but be aware that it is only working if the correkt firmware is aso
included in device/generic/common/firmware and if tha module is loaded
manually or by startup script
The patch has included my firmware for Trekstor Surftab Wintron 7.0
extracted from the Windwos 10 driver firmware
it is only a alpha version because I don't know at the moment how I shell
calibrate the touchscreen.
Maybe I find a way to hack the firmware to ready to use. (that would be
the best sollution)
under ubuntu I used xinput calibrator but on android the built in
calibration do not work for me atm.

[youling 257]

the new remix os for pci use its kernel、firmware、modulescopy to Android
x86 6.0.1

圚 2016幎1月29日星期五 UTC+8䞊午1:15:07Mario Holzinger写道

Post by Mario Holzinger
only for testing here is the Patch to implement GSLx680 touch controllers.
but be aware that it is only working if the correkt firmware is aso
included in device/generic/common/firmware and if tha module is loaded
manually or by startup script
The patch has included my firmware for Trekstor Surftab Wintron 7.0
extracted from the Windwos 10 driver firmware
it is only a alpha version because I don't know at the moment how I shell
calibrate the touchscreen.
Maybe I find a way to hack the firmware to ready to use. (that would be
the best sollution)
under ubuntu I used xinput calibrator but on android the built in
calibration do not work for me atm.

[Mario Holzinger]

sorry for highjacking the thread again.

what do you think about the redefined patch for
hardware/libsensors/iio-sensors.cpp

hal.sensors.iio.accel.matrix=-000+000-
first sensor inverted for x
second sensor for y
third sensor inverted for z


hal.sensors.iio.gyro.matrix=+000+000+
first sensor for x
second sensor for y
third sensor for z

defines a matrix each for every aspect of the sensors in one property

+ = positive sensor value to be used
- = inverted sensor value to be used
0 = ignor sensor value for this axis

the first 3 characters stands for the first enumerated sensor were to be
used

e.g.
-00 = use first sensor inverted for x axis
0+0 = use first sensor for y axis

the second 3 characters are for the 2nd sensor
and the last 3 are for the 3rd sensor

best regards
Mario

PS: I have no Gyro so this could not be tested by my selfe, and also other
sensors can easily ported with this code blocks
+ //hal.sensors.iio.accel.matrix=-000+000- ==> default configuration
+ char cmatrix[PROPERTY_VALUE_MAX];
+ int imatrix[9];
+
+ property_get("hal.sensors.iio.accel.matrix", cmatrix, "-000+000-" );
+
+ for (int i = 0, j = 0; i < PROPERTY_VALUE_MAX; i++)
+ {
+ if (cmatrix[i] == '+') {imatrix[j] = 1; j++;}
+ if (cmatrix[i] == '-') {imatrix[j] = -1; j++;}
+ if (cmatrix[i] == '0') {imatrix[j] = 0; j++;}
+ }
+
+ for (int i = 0; i < ret; ++i)
+ {
+ if (imatrix[0] != 0) { data[i].acceleration.x = imatrix[0] * scale *
read_sysfs_int("in_accel_x_raw"); }
+ if (imatrix[1] != 0) { data[i].acceleration.x = imatrix[1] * scale *
read_sysfs_int("in_accel_y_raw"); }
+ if (imatrix[2] != 0) { data[i].acceleration.x = imatrix[2] * scale *
read_sysfs_int("in_accel_z_raw"); }
+ if (imatrix[3] != 0) { data[i].acceleration.y = imatrix[3] * scale *
read_sysfs_int("in_accel_x_raw"); }
+ if (imatrix[4] != 0) { data[i].acceleration.y = imatrix[4] * scale *
read_sysfs_int("in_accel_y_raw"); }
+ if (imatrix[5] != 0) { data[i].acceleration.y = imatrix[5] * scale *
read_sysfs_int("in_accel_z_raw"); }
+ if (imatrix[6] != 0) { data[i].acceleration.z = imatrix[6] * scale *
read_sysfs_int("in_accel_x_raw"); }
+ if (imatrix[7] != 0) { data[i].acceleration.z = imatrix[7] * scale *
read_sysfs_int("in_accel_y_raw"); }
+ if (imatrix[8] != 0) { data[i].acceleration.z = imatrix[8] * scale *
read_sysfs_int("in_accel_z_raw"); }
+ data[i].acceleration.status = SENSOR_STATUS_ACCURACY_HIGH;

[Jaap Jan Meijer]

Hi Chih Wei,

I just noticed 32-bit binder is not enabled is not enabled for x64 kernel.
Did you maybe forget it or do you plan to switch to x64 user space as well?
I haven't tried x64 user space myself, is it usable?

[Jaap Jan Meijer]

Post by Jaap Jan Meijer
Hi Chih Wei,
I just noticed 32-bit binder is not enabled is not enabled for x64 kernel.

We don't need it.
To be clearer, that patch is only used to support
32-bit userspace without binder version 8,
i.e., android version prior to kitkat (included).
Since I have no plan to use kernel-4.4 with
kitkat-x86, I'm not going to apply that patch.

Thanks for the explanation. I'm still on 32 bit user space for my cm builds
for which I do need to enable it. The only reason I switched to x64 kernel
is boot time on by tablet (for some reason there is a long delay with x86).

Post by Jaap Jan Meijer
Did you maybe forget it or do you plan to switch to x64 user space as well?

Already have -- the android_x86_64 target.

Post by Jaap Jan Meijer
I haven't tried x64 user space myself, is it usable?

We have fixed several critical 64-bit issues.
(thanks to the patches of Paulo and Jide)
IMO it's quite stable now.
Surely we will continue fixing 64-bit issues
if more are found.

Sounds good, I'll do some testing as soon as I get cm buildable from the
Android-x86 repo.

[Milan Parikh]

Any chance you can post your build for the Surface 3 (non-pro)? I don't
have a Linux machine to build it myself, or I wouldn't bother you.

[Milan Parikh]

Thank you! I'll try it later today.

Post by Milan Parikh
Any chance you can post your build for the Surface 3 (non-pro)? I don't

have

Post by Milan Parikh
a Linux machine to build it myself, or I wouldn't bother you.

Try android-x86-6.0_20160129.iso
https://sourceforge.net/projects/android-x86/files/Testing/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Chris Rogers]

There seems to be a regression on he Dell Venue 8 Pro. The power button is
no longer recognized at all. Could be an IRQ conflict? Also, is the sleep
behavior set to mem or freeze? Baytrail doesn't have mem support, so it
wouldn't properly sleep. I suppose the real solution to this is to build
for my own device.

Post by Milan Parikh
Any chance you can post your build for the Surface 3 (non-pro)? I don't

have

Post by Milan Parikh
a Linux machine to build it myself, or I wouldn't bother you.

Try android-x86-6.0_20160129.iso
https://sourceforge.net/projects/android-x86/files/Testing/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[JJ Meijer]

Hi,

For lollipop or later, what you need is
TARGET_USES_64_BIT_BINDER := true
for 64-bit kernel.

That worked, thanks!


There seems to be a regression on he Dell Venue 8 Pro. The power button is

Post by Chris Rogers
no longer recognized at all. Could be an IRQ conflict? Also, is the sleep
behavior set to mem or freeze? Baytrail doesn't have mem support, so it
wouldn't properly sleep. I suppose the real solution to this is to build
for my own device.

Can you please supply output of "cat /proc/interrupts"? Is pata_legacy in
there? If so you have the same IRQ conflict as on my HP Omni10. Chih-Wei,
can't we just remove this obsolete (I think) module? Otherwise we can just
blacklist it I guess.

[Jianlong Liu]

From my brief experiment with 4.5 RC2 (Mauro's branch), it's been fixed (I
think, pretty sure I didn't comment the module out). Some other things
weren't working though, so I went back to 4.4.

Post by JJ Meijer
Can you please supply output of "cat /proc/interrupts"? Is pata_legacy in
there? If so you have the same IRQ conflict as on my HP Omni10. Chih-Wei,
can't we just remove this obsolete (I think) module? Otherwise we can just
blacklist it I guess.

[Jianlong Liu]

And yes, it's the same IRQ conflict (I checked a while ago and removed the
module).

Post by JJ Meijer
Can you please supply output of "cat /proc/interrupts"? Is pata_legacy in
there? If so you have the same IRQ conflict as on my HP Omni10. Chih-Wei,
can't we just remove this obsolete (I think) module? Otherwise we can just
blacklist it I guess.

[Daniel N]

kernel4.4 work great for my t90chi.

Finally it is able to boot up, wifi, blacklight, touchscreen is working
bt, power button, sound is not working.

and sometimes it will be freeze.
Besides, the touchscreen will not work after will times reboot.


Chih-Wei Huangæ–Œ 2016幎2月2日星期二 UTC+8䞋午7時10分19秒寫道

Post by Jaap Jan Meijer

Post by JJ Meijer
Can you please supply output of "cat /proc/interrupts"? Is pata_legacy

in

Post by JJ Meijer
there? If so you have the same IRQ conflict as on my HP Omni10.

Chih-Wei,

Post by JJ Meijer
can't we just remove this obsolete (I think) module? Otherwise we can

just

Post by JJ Meijer
blacklist it I guess.

OK, let's remove it if nobody needs it.
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Povilas Staniulis]

From what I read on the T100 group, kernel 4.4 still freezes with MMC
errors.
Some devs there suggest disabling intel cstates on Baytrail to improve
stability: intel_idle.max_cstate=0 (or 1)
This does increase power drain significantly though.

It seems that cstates somehow mess up the MMC bus...

Post by Jaap Jan Meijer
The biggest problem I was having with vanilla 4.4 where MMC issues
even with a patch set from Adrian Hunter, I must have messed up
somewhere. I'm running your kernel now with Android from internal
eMMC and SD-card and haven't encountered any I/O errors so far.
I'm sorry, I spoke to soon. The MMC situation is slightly improved
with the Hunter patches but still the complete MMC bus freezes up
after a while. First ADMA errors and than I/O errors and timeouts.
Kernel 4.0.9 seems to do work better with MMC.
I'm now testing some workarounds. With DMA (sdhci.debug_quirks=0x40)
seems to prevent freezing. Also I read about disabling 1.8V mode, will
try that as well.
--
You received this message because you are subscribed to the Google
Groups "Android-x86" group.
To unsubscribe from this group and stop receiving emails from it, send
Visit this group at https://groups.google.com/group/android-x86.
For more options, visit https://groups.google.com/d/optout.

[Chieh-Min Wang]

I was trying to build the 4.4 kernel, however, I encountered a build error.
Any idea how can I resolve this?

Thanks in advance :).

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[zhongtian xie]

maybe lack of bcm wireless driver, Generally it download automatic  but
sometimes maybe something wrong , and you can download it to
kernel/drivers/net/wireless/wl/,

vim kernel/drivers/net/wireless/wl/build.mk
=========================================================================================================================
# parts of build/core/tasks/kernel.mk

WL_ENABLED := $(if $(wildcard $(WL_PATH)),$(shell grep ^CONFIG_WL=[my]
$(KERNEL_CONFIG_FILE)))
WL_ARCH_CHANGED := $(if $(shell file $(WL_PATH)/lib/wlc_hybrid.o_shipped |
grep -s $(if $(filter x86,$(TARGET_KERNEL_ARCH)),80386,x86-64)),,FORCE)
*WL_SRC := $(WL_PATH)/hybrid-v35$(if $(filter
x86,$(TARGET_KERNEL_ARCH)),,_64)-nodebug-pcoem-6_30_223_271.tar.gz*
$(WL_SRC):
@echo Downloading $(@F)...
$(hide) curl http://www.broadcom.com/docs/linux_sta/$(@F) > $@

$(WL_PATH)/Makefile : $(WL_SRC) $(wildcard $(WL_PATH)/*.patch)
$(WL_ARCH_CHANGED) $(KERNEL_ARCH_CHANGED)
$(hide) tar zxf $< -C $(@D) --overwrite && \
patch -p5 -d $(@D) -i wl.patch && \
patch -p1 -d $(@D) -i linux-recent.patch

$(INSTALLED_KERNEL_TARGET): $(if $(WL_ENABLED),$(WL_PATH)/Makefile)

==========================================================================================================================

Post by Chieh-Min Wang
I was trying to build the 4.4 kernel, however, I encountered a build
error. Any idea how can I resolve this?
Thanks in advance :).

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

--
You received this message because you are subscribed to the Google Groups
"Android-x86" group.
To unsubscribe from this group and stop receiving emails from it, send an
Visit this group at https://groups.google.com/group/android-x86.
For more options, visit https://groups.google.com/d/optout.

[Chieh-Min Wang]

Thanks, I was using the old build/core/tasks/kernel.mk for kernel 3.18, and
it always downloads the old version of bcm driver that's what cause the
problem.

Post by zhongtian xie
maybe lack of bcm wireless driver, Generally it download automatic  but
sometimes maybe something wrong , and you can download it to
kernel/drivers/net/wireless/wl/,
vim kernel/drivers/net/wireless/wl/build.mk
=========================================================================================================================
# parts of build/core/tasks/kernel.mk
WL_ENABLED := $(if $(wildcard $(WL_PATH)),$(shell grep ^CONFIG_WL=[my]
$(KERNEL_CONFIG_FILE)))
WL_ARCH_CHANGED := $(if $(shell file $(WL_PATH)/lib/wlc_hybrid.o_shipped |
grep -s $(if $(filter x86,$(TARGET_KERNEL_ARCH)),80386,x86-64)),,FORCE)
*WL_SRC := $(WL_PATH)/hybrid-v35$(if $(filter
x86,$(TARGET_KERNEL_ARCH)),,_64)-nodebug-pcoem-6_30_223_271.tar.gz*
@echo Downloading $(@F)...
$(WL_PATH)/Makefile : $(WL_SRC) $(wildcard $(WL_PATH)/*.patch)
$(WL_ARCH_CHANGED) $(KERNEL_ARCH_CHANGED)
$(INSTALLED_KERNEL_TARGET): $(if $(WL_ENABLED),$(WL_PATH)/Makefile)
==========================================================================================================================

Post by Chieh-Min Wang
I was trying to build the 4.4 kernel, however, I encountered a build
error. Any idea how can I resolve this?
Thanks in advance :).

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

--
You received this message because you are subscribed to the Google Groups
"Android-x86" group.
To unsubscribe from this group and stop receiving emails from it, send an
<javascript:>.
Visit this group at https://groups.google.com/group/android-x86.
For more options, visit https://groups.google.com/d/optout.

[Chieh-Min Wang]

I build kernel 4.4 for android 4.4, however, it failed to start. Can I use
kernel 4.4 for android 4.4?

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Jaap Jan Meijer]

Post by Jaap Jan Meijer
I attached some Bluetooth patches to add missing ACPI ID's for bcm43241
chips and two fixes to get IOCTL working in mixed arch situations. Also
there is a fw package I extracted from the Win10 bcm driver that should

go

Post by Jaap Jan Meijer
into /lib/firmware/brcm. After that we could start using btattach for

UART

Post by Jaap Jan Meijer
BT devices. Do you agree with this? Should I push to Sourceforge?

Looks good. Please push them.
About the 0001-Add-missing-ACPI-IDs-for-bcm43241.patch,
has it been submitted to the upstream?

The ioctl stuff is pushed upstream now. For the missing ACPI ID's I still
have to create a patch they will be able to apply. During my quest I
noticed there are a lot of missing ID's so I hope they will just push it at
once or find a better solution, some have been pushed already.

http://www.driveridentifier.com//scan/driver_file_detail.php/driver-detail/262C118A233E4B45BC8EB76D27B63F9A/1811458/md5/1583152406/ACPI%5CBCM2E54?scanid=262C118A233E4B45BC8EB76D27B63F9A&inf_file_id=1811458&md5=md5&item_id=1583152406&hardware_id=ACPI\BCM2E54&friendly_pc_model=broadcom-serial-bus-driver-over-uart-bus-enumerator

Post by Jaap Jan Meijer
The biggest problem I was having with vanilla 4.4 where MMC issues even

with

Post by Jaap Jan Meijer
a patch set from Adrian Hunter, I must have messed up somewhere. I'm
running your kernel now with Android from internal eMMC and SD-card and
haven't encountered any I/O errors so far.

I haven't made a lot of testing yet.

Alright, I hope you will find some time to test, I'll try get some time to
do some testing myself.

[spin877]

missing firmware rtlwifi/rtl8723bs_nic.bin
I added in the kernel module

try adding these two files in

/drivers/staging/rtl8723bs/hal/

[spin877]

Post by spin877
missing firmware rtlwifi/rtl8723bs_nic.bin
I added in the kernel module
try adding these two files in

Ok.

Ok . But I have just imitated the "master " you. joke :-)
Look at version 4.0.9

https://sourceforge.net/p/android-x86/kernel/ci/android-4.0/tree/drivers/staging/rtl8723bs/hal/HalHWImg8723B_FW.c

Post by spin877
/drivers/staging/rtl8723bs/hal/

The firmware should not be put into the kernel directly.
I've pushed the rtl8723bs firmware into
the firmware dir. Please sync to get it.
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Mario Holzinger]

I know I'm not an professional coder, but maybe some of you can optimize
the code a bit.

this patch + the 2 lines in /system/build.prop

hal.sensors.iio.accel.axis=1
hal.sensors.iio.accel.inv=7

makes my Trekstor Surftab Wintron 7.0 tablet act as expected :-)

Thanks again, for all the tips.
here is a patch for hardware/libsesnors/ to xchange all possibilities of
axis and invert all axis
with this I can mod /system/build.prop (needs root access)
to support my tablets accelerometers perfect
default is untoched as it was before :-)
best regards
Mario

[Hypo Turtle]

Would something like this work (and deal with inverse);

template<> int Sensor<ID_GYROSCOPE>::readEvents(sensors_event_t *data, int
cnt)
{
static float scale = read_sysfs_float((*nodes)[0]);
int ret = SensorBase::readEvents(data, cnt);
// TODO: read orientation from the properties
char rota[6] = property_get("hal.sensors.iio");
// or propery_get("hal.sensors.iio", rota, " ,x, ,y, ,z");
for (int i = 0; i < ret; ++i) {
data[i].gyro.x = rota[1]scale * read_sysfs_int(
"in_anglvel_rota[2]_raw");
data[i].gyro.y = rota[3]scale * read_sysfs_int(
"in_anglvel_rota[4]_raw");
data[i].gyro.z = rota[5]scale * read_sysfs_int(
"in_anglvel_rota[6]_raw");
data[i].gyro.status = SENSOR_STATUS_ACCURACY_HIGH;
}
return ret;
}




With
*hal.sensors.iio = -,x,-,y,-,z*or* hal.sensors.iio = ,x,,y,,z*

Thanks again, for all the tips.
here is a patch for hardware/libsesnors/ to xchange all possibilities of
axis and invert all axis

Thanks a lot for the patch.
I would suggest you implement it in a 3x3 conversion matrix
in_accel_{x,y,z}_raw => acceleration.{x,y,z}
The conversion matrix could be described by a property
containing 9 elements of array, say
hal.sensors.iio.accel.matrix=1,0,0,0,1,0,0,0,1

with this I can mod /system/build.prop (needs root access)
to support my tablets accelerometers perfect
default is untoched as it was before :-)

--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[youling 257]

need gslx680 patch
for https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.8/tree/

圚 2016幎2月20日星期六 UTC+8䞋午8:05:29Mario Holzinger写道

At first thx for all your great work.
I'v tested most of my devices with kernel-4.4 and whant to share what I
need to patch so far
Microsoft Surface Pro 2
- seems to wotk out of the box great (also my patch that was allready
integreated for wifi is working great)
- bluetooth Yes
- rotation Yes
- Type Cover Yes
...
Trekstore Surtab Wintron 7.0
- Wifi RTL8723BS updated sources from github (kernel) and add binaries in
(device/generic/common/firmware/rtlwifi) ==> see patches
- Touchscreen GSL1680 I2C is working with integrating the acpi kernel
module ==> see patches
- rotation is working out of the box but is misaligned 90° clockwise ==>
need to investigate further where to add on offset for my tablet

Post by spin877
missing firmware rtlwifi/rtl8723bs_nic.bin
I added in the kernel module
try adding these two files in
/drivers/staging/rtl8723bs/hal/

The firmware should not be put into the kernel directly.
I've pushed the rtl8723bs firmware into
the firmware dir. Please sync to get it.
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Mario Holzinger]

what did you mean?

there is already a working version of the gslx680 driver included
there is a updated version available at
https://github.com/onitake/gslx680-acpi

for android-x86 you could apply the attachment that replaces the current
version with the latest one from onitake

I sucessfully use the updated driver on ma android builds but the essential
thing is that you need a propper silead_ts.fw from
https://github.com/onitake/gsl-firmware
or selfmade one that fits as good as possible to your device.
with a good calibrated firmware you can also avoid recalibration under
android. mine fits perfectly on my Trekstor 7.0

regards
Mario

need gslx680 patch for
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.8/tree/
圚 2016幎2月20日星期六 UTC+8䞋午8:05:29Mario Holzinger写道

At first thx for all your great work.
I'v tested most of my devices with kernel-4.4 and whant to share what I
need to patch so far
Microsoft Surface Pro 2
- seems to wotk out of the box great (also my patch that was allready
integreated for wifi is working great)
- bluetooth Yes
- rotation Yes
- Type Cover Yes
...
Trekstore Surtab Wintron 7.0
- Wifi RTL8723BS updated sources from github (kernel) and add binaries in
(device/generic/common/firmware/rtlwifi) ==> see patches
- Touchscreen GSL1680 I2C is working with integrating the acpi kernel
module ==> see patches
- rotation is working out of the box but is misaligned 90° clockwise ==>
need to investigate further where to add on offset for my tablet

Post by spin877
missing firmware rtlwifi/rtl8723bs_nic.bin
I added in the kernel module
try adding these two files in
/drivers/staging/rtl8723bs/hal/

The firmware should not be put into the kernel directly.
I've pushed the rtl8723bs firmware into
the firmware dir. Please sync to get it.
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[youling 257]

0001-updated-gslx680-driver-from-https-github.com-onitake.patch , only
update
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/drivers/staging/gslx680/
now ,kernel-4.8 branch is ready, didn't
has https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.8/tree/drivers/staging/gslx680,
need patch
圚 2016幎9月4日星期日 UTC+8䞋午9:15:42Mario Holzinger写道

Post by Mario Holzinger
what did you mean?
there is already a working version of the gslx680 driver included
there is a updated version available at
https://github.com/onitake/gslx680-acpi
for android-x86 you could apply the attachment that replaces the current
version with the latest one from onitake
I sucessfully use the updated driver on ma android builds but the
essential thing is that you need a propper silead_ts.fw from
https://github.com/onitake/gsl-firmware
or selfmade one that fits as good as possible to your device.
with a good calibrated firmware you can also avoid recalibration under
android. mine fits perfectly on my Trekstor 7.0
regards
Mario

need gslx680 patch for
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.8/tree/
圚 2016幎2月20日星期六 UTC+8䞋午8:05:29Mario Holzinger写道

At first thx for all your great work.
I'v tested most of my devices with kernel-4.4 and whant to share what I
need to patch so far
Microsoft Surface Pro 2
- seems to wotk out of the box great (also my patch that was allready
integreated for wifi is working great)
- bluetooth Yes
- rotation Yes
- Type Cover Yes
...
Trekstore Surtab Wintron 7.0
- Wifi RTL8723BS updated sources from github (kernel) and add binaries
in (device/generic/common/firmware/rtlwifi) ==> see patches
- Touchscreen GSL1680 I2C is working with integrating the acpi kernel
module ==> see patches
- rotation is working out of the box but is misaligned 90° clockwise ==>
need to investigate further where to add on offset for my tablet

Post by spin877
missing firmware rtlwifi/rtl8723bs_nic.bin
I added in the kernel module
try adding these two files in
/drivers/staging/rtl8723bs/hal/

The firmware should not be put into the kernel directly.
I've pushed the rtl8723bs firmware into
the firmware dir. Please sync to get it.
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Mario Holzinger]

as soon as i get kernel-4-8 compiled properly I will support you with a
patch.
Than Chih-Wei Huang can decide if he likes to merge it.

I also play a bit to compile the 4.8 branch outside the android tree (and
kernel master branch from Linus Torvalds) but without success so far

best regards
Mario

btw. I also tried to mod my manifests/default.xml to get kernel-4.8
automatically when syncing but it fails due to the currently facing syncing
issues from sourceforge (on my side)

Post by youling 257
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/drivers/staging/gslx680/

now ,kernel-4.8 branch is ready, didn't has

Well, I didn't say kernel-4.8 is ready.
Nobody said that.
The branch was pushed to sf.net
just for GPL compliance.
It's the kernel source I used to
build the 6.0-rc2-for-s3 iso.
It's not officially created or maintained
for android-x86 porting.
You are free to try it but
don't expect it too much.
What's the next kernel for porting
is not decided.
I'm still waiting the next LTS kernel
but it's yet to be decided.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.8/tree/drivers/staging/gslx680,

need patch

[youling 257]

https://groups.google.com/d/msg/android-x86/nseLSalOtGI/Gh9UcPa1CQAJ , only
compile vmlinuz and modules success, rename vmlinuz to kernel
make android-x86_defconfig *time* *make* -j8*time make* -j8 targz-pkg -i

圚 2016幎9月5日星期䞀 UTC+8䞊午2:01:41Mario Holzinger写道

Post by Mario Holzinger
as soon as i get kernel-4-8 compiled properly I will support you with a
patch.
Than Chih-Wei Huang can decide if he likes to merge it.
I also play a bit to compile the 4.8 branch outside the android tree (and
kernel master branch from Linus Torvalds) but without success so far
best regards
Mario
btw. I also tried to mod my manifests/default.xml to get kernel-4.8
automatically when syncing but it fails due to the currently facing syncing
issues from sourceforge (on my side)

Post by youling 257
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/drivers/staging/gslx680/

now ,kernel-4.8 branch is ready, didn't has

Well, I didn't say kernel-4.8 is ready.
Nobody said that.
The branch was pushed to sf.net
just for GPL compliance.
It's the kernel source I used to
build the 6.0-rc2-for-s3 iso.
It's not officially created or maintained
for android-x86 porting.
You are free to try it but
don't expect it too much.
What's the next kernel for porting
is not decided.
I'm still waiting the next LTS kernel
but it's yet to be decided.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.8/tree/drivers/staging/gslx680,

need patch

[Mario Holzinger]

ok found out that my implemented gslx680_ts_acpi driver in
kernel/drivers/staging/gslx680 is now obsolete and replaced by silead_ts
wich comes directly with the kernel
the disadvantage is that the firmware is not able to be loaded anymore
because it is especially created for the gslx680_ts_acpi module

my luck is that my win10 firmware can be loaded and is working is I rename
it as mssl1680.fw however the calibration is a mess.

I will provide a patch as soon as I find out how to fix the calibration
directly in the win10.fw

regards
Mario

[youling 257]

4.9 is longterm4.9.13

圚 2016幎9月5日星期䞀 UTC+8䞊午12:35:36Chih-Wei Huang写道

Post by youling 257
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/drivers/staging/gslx680/

now ,kernel-4.8 branch is ready, didn't has

Well, I didn't say kernel-4.8 is ready.
Nobody said that.
The branch was pushed to sf.net
just for GPL compliance.
It's the kernel source I used to
build the 6.0-rc2-for-s3 iso.
It's not officially created or maintained
for android-x86 porting.
You are free to try it but
don't expect it too much.
What's the next kernel for porting
is not decided.
I'm still waiting the next LTS kernel
but it's yet to be decided.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.8/tree/drivers/staging/gslx680,

need patch

[Cemehtypnahov]

that if you make an application requires Root rights, which will replace
the patch for rotation, in the system folder? because ordinary users can
not be integrated this file in the system.


среЎа, 27 яМваря 2016 г., 17:02:00 UTC+8 пПльзПватель Chih-Wei Huang

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Jon West]

There are a couple patches still needed to solve issues with the multitouch
on the Surface Pro 3

https://github.com/shvr/fedora-surface-pro-3-kernel/commit/9033a2d97a2c40a4260928b609908f65d2edfbe0
https://github.com/shvr/fedora-surface-pro-3-kernel/blob/master/HID-multitouch-Add-MT_QUIRK_NOT_SEEN_MEANS_UP-to-MT_.patch

https://github.com/shvr/fedora-surface-pro-3-kernel/commit/3602b1729fe4e0a38f28828cb80d2c2765eb9a62
https://github.com/shvr/fedora-surface-pro-3-kernel/blob/master/HID-multitouch-Ignore-invalid-reports.patch

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Jon West]

I am sorry then. I am still getting the hang of things in this group.

Post by Jon West

Post by Jon West
There are a couple patches still needed to solve issues with the

multitouch

Post by Jon West
on the Surface Pro 3

https://github.com/shvr/fedora-surface-pro-3-kernel/commit/9033a2d97a2c40a4260928b609908f65d2edfbe0
https://github.com/shvr/fedora-surface-pro-3-kernel/blob/master/HID-multitouch-Add-MT_QUIRK_NOT_SEEN_MEANS_UP-to-MT_.patch
https://github.com/shvr/fedora-surface-pro-3-kernel/commit/3602b1729fe4e0a38f28828cb80d2c2765eb9a62
https://github.com/shvr/fedora-surface-pro-3-kernel/blob/master/HID-multitouch-Ignore-invalid-reports.patch
Thank you for the willing to help but
please check our kernel repository first
otherwise you are wasting our time to
check them again.
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[youling 257]

the new remix os for pci use its kernel、firmware、modulescopy to Android
x86 6.0.1 0129

圚 2016幎1月27日星期䞉 UTC+8䞋午5:02:00Chih-Wei Huang写道

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Andrew Watts]

If I am correct its only after kernel 4.5 that skylake's power saving
feature's are implemented properly right?

I ran 6.0 RC-2 on my Core M 6y30... When I looked at the cpu it was
(claimed) to be running fast - up to 2Ghz
and 4 cores (I assume this was the kernel misrecognising Hyperthreading on
the dual core)
Although it wasn't running hot.

Also power use didn't seem to be an issue.
I downloaded a pre-build 7.0 image... Now power consumption there is an
issue!

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[dgdn]

did location of wl-driver change sometime in the last week by any chance????

https://git.solus-project.com/packages/broadcom-sta/commit/?h=broadcom-sta-6.30.223.271-88


getting the following that was working fine on dec 2??

[ 99% 48020/48180] build kernel/drivers/net/wireless/wl/Makefile
FAILED: /bin/bash -c "tar zxf
kernel/drivers/net/wireless/wl/hybrid-v35_64-nodebug-pcoem-6_30_223_271.tar.gz
-C kernel/drivers/net/wireless/wl --overwrite && patch -p5 -d
kernel/drivers/net/wireless/wl -i wl.patch && patch -p1 -d
kernel/drivers/net/wireless/wl -i linux-recent.patch"

gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error is not recoverable: exiting now
[ 99% 48020/48180] host C: bison <= external/bison/src/InadequacyList.c
external/bison/src/InadequacyList.c:67:57: warning: unused parameter 'i'
[-Wunused-parameter]
ContributionIndex i)

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Jon West]

No, their zip is malformed or incomplete. Search around for it to grab a
good zip and place it in the wl dir. You can disable the curl too so it
doesn't try to grab another. This is only temporary until Broadcom
realizesthey posted a bad zip.

Post by dgdn
did location of wl-driver change sometime in the last week by any chance????
https://git.solus-project.com/packages/broadcom-sta/commit/?h=broadcom-sta-6.30.223.271-88
getting the following that was working fine on dec 2??
[ 99% 48020/48180] build kernel/drivers/net/wireless/wl/Makefile
FAILED: /bin/bash -c "tar zxf
kernel/drivers/net/wireless/wl/hybrid-v35_64-nodebug-pcoem-6_30_223_271.tar.gz
-C kernel/drivers/net/wireless/wl --overwrite && patch -p5 -d
kernel/drivers/net/wireless/wl -i wl.patch && patch -p1 -d
kernel/drivers/net/wireless/wl -i linux-recent.patch"
gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error is not recoverable: exiting now
[ 99% 48020/48180] host C: bison <= external/bison/src/InadequacyList.c
external/bison/src/InadequacyList.c:67:57: warning: unused parameter 'i'
[-Wunused-parameter]
ContributionIndex i)

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[dgdn]

that maybe BUT....I pulled the sources from the new link, added to my wl
folder and everything built fine.....

Post by Jon West
No, their zip is malformed or incomplete. Search around for it to grab a
good zip and place it in the wl dir. You can disable the curl too so it
doesn't try to grab another. This is only temporary until Broadcom
realizesthey posted a bad zip.

Post by dgdn
did location of wl-driver change sometime in the last week by any chance????
https://git.solus-project.com/packages/broadcom-sta/commit/?h=broadcom-sta-6.30.223.271-88
getting the following that was working fine on dec 2??
[ 99% 48020/48180] build kernel/drivers/net/wireless/wl/Makefile
FAILED: /bin/bash -c "tar zxf
kernel/drivers/net/wireless/wl/hybrid-v35_64-nodebug-pcoem-6_30_223_271.tar.gz
-C kernel/drivers/net/wireless/wl --overwrite && patch -p5 -d
kernel/drivers/net/wireless/wl -i wl.patch && patch -p1 -d
kernel/drivers/net/wireless/wl -i linux-recent.patch"
gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error is not recoverable: exiting now
[ 99% 48020/48180] host C: bison <= external/bison/src/InadequacyList.c
external/bison/src/InadequacyList.c:67:57: warning: unused parameter 'i'
[-Wunused-parameter]
ContributionIndex i)

Hi all,
I've pushed the kernel-4.4 branch into the SF git server.
* AOSP experimental/android-4.4 branch
for Android specific features
* Intel drm-intel-next-queued branch
to support newer chips
(test ok on Surface 3)
* Some patches from LKML to support newer
peripherals or fix bugs.
(thanks to Peter's mint-17.3-for-surface-pro-3)
* Rob Herring's virtio-gpu patches
(I didn't tested yet)
* Other older patches applied by android-x86
* uvesafb (v86d) is broken -- not sure it's due to
the kernel 4.4 regression or missing necessary configs
(on the top of android-x86 tree)
cd kernel
git fetch x86 kernel-4.4
git checkout x86/kernel-4.4
cd ..
Then build the android-x86 image as usual.
https://sourceforge.net/p/android-x86/kernel/ci/kernel-4.4/tree/
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[dgdn]

ok thank you.....will resync and test within the next few days.....in the
mean time

more recent test builds are here:

https://drive.google.com/drive/folders/0B3Vjym21YjOTTERmbk1MZjJ6Ylk?usp=sharing

Post by dgdn

Post by dgdn
did location of wl-driver change sometime in the last week by any

chance????
https://git.solus-project.com/packages/broadcom-sta/commit/?h=broadcom-sta-6.30.223.271-88
Thank you for pointing it out.
I've updated to the new URL.