8723bu and 8812au compile error

Question

8723bu and 8812au compile error

Closed this issue 7 years ago · 29 comments

Answer 1 · 2016-10-02T12:26:16.000Z

Hi youling257,

thanks for reporting the issue,

I'm not seeing the issue while building nougat-x86,
so, is it happening when building marshamallow-x86 (I have not yet tested) ?

Have you also trided to delete kernel and obj/kernel/vmlinux and vmlinux.o before rebuilding?
M.

Answer 2 · 2016-10-02T12:31:35.000Z

I've retried and the rtl8723bu objects are there and they get installed in the iso image

INSTALL drivers/staging/rtl8723bu/r8723bu.ko

Could you also post the build error you see, to check what is the problem?
Thx
M.

Answer 3 · 2016-10-02T12:34:43.000Z

/root/kernel/drivers/staging/rtl8723bu/core/rtw_mlme.c: In function ‘rtw_restructure_ht_ie’:
/root/kernel/drivers/staging/rtl8723bu/core/rtw_mlme.c:3663:3: warning: this ‘if’ clause does not guard... [-Wmisleading-indentation]
if (stbc_rx_enable)
^~
/root/kernel/drivers/staging/rtl8723bu/core/rtw_mlme.c:3666:18: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the ‘if’
set_mcs_rate_by_mask(ht_capie.supp_mcs_set, MCS_RATE_1R);
^~~~~~~~~~~~~~~~~~~~

/root/kernel/drivers/staging/rtl8812au/core/rtw_security.c: In function ‘aes_cipher’:
/root/kernel/drivers/staging/rtl8812au/core/rtw_security.c:1598:5: warning: this ‘for’ clause does not guard... [-Wmisleading-indentation]
for (j = 0; j < 8; j++)
^~~
/root/kernel/drivers/staging/rtl8812au/core/rtw_security.c:1601:2: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the ‘for’
payload_index = hdrlen + 8;
^~~~~~~~~~~~~
/root/kernel/drivers/staging/rtl8812au/core/rtw_security.c: In function ‘aes_decipher’:
/root/kernel/drivers/staging/rtl8812au/core/rtw_security.c:1984:5: warning: this ‘for’ clause does not guard... [-Wmisleading-indentation]
for (j = 0; j < 8; j++)
^~~
/root/kernel/drivers/staging/rtl8812au/core/rtw_security.c:1987:2: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the ‘for’
payload_index = hdrlen + 8;
^~~~~~~~~~~~~
/root/kernel/drivers/staging/rtl8812au/core/rtw_mlme_ext.c: In function ‘rtw_delba_check’:
/root/kernel/drivers/staging/rtl8812au/core/rtw_mlme_ext.c:12243:7: warning: this ‘else’ clause does not guard... [-Wmisleading-indentation]
else
^~~~
/root/kernel/drivers/staging/rtl8812au/core/rtw_mlme_ext.c:12245:8: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the ‘else’
psta->recvreorder_ctrl[i].enable = _FALSE;
^~~~
/root/kernel/drivers/staging/rtl8812au/core/rtw_mlme_ext.c:12246:7: warning: this ‘if’ clause does not guard... [-Wmisleading-indentation]
if (ret != _FAIL)
^~
/root/kernel/drivers/staging/rtl8812au/core/rtw_mlme_ext.c:12248:8: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the ‘if’
rtw_reset_continual_no_rx_packet(psta, i);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Answer 4 · 2016-10-02T12:42:17.000Z

3 hours ago , i first compile get more 8723bu error and more 8812au error , 8723bu.ko, 957.9 KB (957,900 字节) , 8812au.ko , 1.7 MB (1,702,432 字节)

Answer 5 · 2016-10-02T13:30:15.000Z

8723bs，closed wlan，open again，connected wlan，but Unable to access network，have to soft reboot，then can access network

Answer 6 · 2016-11-17T18:24:58.000Z

I'm also getting build errors with them (kernel 4.9rc4, kernel 4.9rc5):
Kernel: arch/x86/boot/bzImage is ready (#3)
ERROR: "memset" [drivers/staging/rtl8812au/r8812au.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8812au/r8812au.ko] undefined!
ERROR: "memset" [drivers/staging/rtl8723bu/r8723bu.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8723bu/r8723bu.ko] undefined!
ERROR: "memset" [drivers/staging/rtl8723bs/r8723bs.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8723bs/r8723bs.ko] undefined!
ERROR: "memset" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
/home/hkoivuneva/mbp750evo/home/hkoivuneva/nougat-x86/kernel/scripts/Makefile.modpost:91: recipe for target '__modpost' failed
make[2]: *** [__modpost] Error 1
/home/hkoivuneva/mbp750evo/home/hkoivuneva/nougat-x86/kernel/Makefile:1188: recipe for target 'modules' failed
make[1]: *** [modules] Error 2
make[1]: Leaving directory '/home/hkoivuneva/mbp750evo/home/hkoivuneva/nougat-x86/out/target/product/x86_64/obj/kernel'
Makefile:150: recipe for target 'sub-make' failed
make: *** [sub-make] Error 2
make: Leaving directory '/home/hkoivuneva/mbp750evo/home/hkoivuneva/nougat-x86/kernel'
ninja: build stopped: subcommand failed.
build/core/ninja.mk:148: recipe for target 'ninja_wrapper' failed
make: *** [ninja_wrapper] Error 1
make: Leaving directory '/home/hkoivuneva/mbp750evo/home/hkoivuneva/nougat-x86'

Not related to this issue, but in my case connected in a way.. With 4.9rc5 I had an error with kernel/drivers/media/dvb-frontends/gp8psk-fe.c:
WARNING: modpost: missing MODULE_LICENSE() in drivers/media/dvb-frontends/gp8psk-fe.o
see include/linux/module.h for more information
It was fixed by adding MODULE_LICENSE("GPL"); to the bottom of the file. It said it's GPL license in the top so I just put that to the bottom and the build continued -- but stopped at the rtl8812au, rtl8723bu & rtl8732bs & rtl8192u errors at the top of this comment. I stopped there because I had no idea how to define memset & memcpy :D

Thank you!

Answer 7 · 2016-11-17T19:32:02.000Z

Hi,
If you don't need those wifi drivers, you could Simply comment the CONFIG
Lines in the Android defconfig file.

The issue is due to removed asm code in kernel 4.9, I'm waiting for fixes
by mantainers

I'm having issues to boot with my own kernel 4.8, did you do something
special to be able to boot?
What defconfig have you used?

Mauro

Il 17/nov/2016 19:24, "Henri Koivuneva" notifications@github.com ha
scritto:

I'm also getting build errors with them (kernel 4.9rc4, kernel 4.9rc5):
Kernel: arch/x86/boot/bzImage is ready (#3)
ERROR: "memset" [drivers/staging/rtl8812au/r8812au.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8812au/r8812au.ko] undefined!
ERROR: "memset" [drivers/staging/rtl8723bu/r8723bu.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8723bu/r8723bu.ko] undefined!
ERROR: "memset" [drivers/staging/rtl8723bs/r8723bs.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8723bs/r8723bs.ko] undefined!
ERROR: "memset" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
ERROR: "memcpy" [drivers/staging/rtl8192u/r8192u_usb.ko] undefined!
/home/hkoivuneva/mbp750evo/home/hkoivuneva/nougat-x86/
kernel/scripts/Makefile.modpost:91: recipe for target '__modpost' failed
make[2]: *** [__modpost] Error 1
/home/hkoivuneva/mbp750evo/home/hkoivuneva/nougat-x86/kernel/Makefile:1188:
recipe for target 'modules' failed
make[1]: *** [modules] Error 2
make[1]: Leaving directory '/home/hkoivuneva/mbp750evo/
home/hkoivuneva/nougat-x86/out/target/product/x86_64/obj/kernel'
Makefile:150: recipe for target 'sub-make' failed
make: *** [sub-make] Error 2
make: Leaving directory '/home/hkoivuneva/mbp750evo/
home/hkoivuneva/nougat-x86/kernel'
ninja: build stopped: subcommand failed.
build/core/ninja.mk:148: recipe for target 'ninja_wrapper' failed
make: *** [ninja_wrapper] Error 1
make: Leaving directory '/home/hkoivuneva/mbp750evo/
home/hkoivuneva/nougat-x86'

Kernel 4.8 compiles without issues and seems to work well too on Surface
Pro 3! Love it!

Not related to this issue, but in my case connected in a way.. With 4.9rc5
I had an error with kernel/drivers/media/dvb-frontends/gp8psk-fe.c:
WARNING: modpost: missing MODULE_LICENSE() in drivers/media/dvb-frontends/
gp8psk-fe.o
see include/linux/module.h for more information

It was fixed by adding MODULE_LICENSE("GPL"); to the bottom of the file.
It said it's GPL license in the top so I just put that to the bottom and
the build continued -- but stopped at the rtl8812au, rtl8723bu & rtl8732bs
& rtl8192u errors at the top of this comment. I stopped there because I had
no idea how to define memset & memcpy :D

Thank you!

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#2 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AH9VlK0N6fKG2dvjYGfzdZqbwbNuwK5aks5q_Jv6gaJpZM4KL_FK
.

Answer 8 · 2016-11-17T20:59:38.000Z

Ooh right! Interesting. I used your defconfig from this repo, I considered disabling the drivers and adding =y to mwifi drivers but didn't change anything after all! Building Nougat (not cm14). I wonder if it's device-related?

With the successfully compiled 4.8 on nougat and your mesa 13.0.1 apps kept crashing though so I went back to default 4.4. I'll build cm13 with fresh kernel 4.8 tomorrow and test!

My build folder with the ISOs: https://goo.gl/Z6X5tP (change to list view to see the newest)

Answer 9 · 2016-11-18T12:49:58.000Z

Whoops sorry, that kernel file was the default 4.4.30! I'd apparently deleted the compiled file after the issues. Here is the kernel 4.8 built for 6.0.1 with your defconfig:
kernel4.8-mauro_64bit.zip

6.0.1 CM ISO 64-bit with that kernel: https://goo.gl/Si4QZA

Answer 10 · 2016-11-18T17:41:19.000Z

Hi,

so with kernel-4.8 and my android_x86_64_defconfig and
android_x86_defconfig you can boot and you have stability issues,
with marshmallow-x86, but you can boot systematically?

Same question for nougat-x86 were I never had the chance to see a completed
boot?

What about kernel-4.9rc5 with following 490 defconfig files which have the
CONFIG_{drivers requiring memcpy} all disabled,
with these config files you should be able to complete kernel-4.9rc5 build
for both marshmallow-x86 and nougat-x86

What kind of crash are you seeing and do you have logcat/dmesg?

What is your Device Under Test?

I can provide you the full list of issues I know of.
Mauro

2016-11-18 13:49 GMT+01:00 Henri Koivuneva notifications@github.com:

Whoops sorry, that kernel file was the default 4.4.30! I'd apparently
deleted the compiled file after the issues. Here is the kernel 4.8 built
for 6.0.1 with your defconfig:
kernel4.8-mauro_64bit.zip
https://github.com/maurossi/linux/files/599853/kernel4.8-mauro_64bit.zip

6.0.1 CM ISO 64-bit with that kernel: https://goo.gl/Si4QZA

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
#2 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AH9VlHCcR6p4sIy6ThP35gyNCXEr5Rfcks5q_Z73gaJpZM4KL_FK
.

Answer 11 · 2016-12-21T10:28:10.000Z

Hi Mauro! Sorry, I had a very busy time at work and couldn't work on this. Unfortunately I was unable to get any logs out. The device is Surface Pro 3. I'm just now building both nougat and marshmallow with the newest code you guys have pushed to Android-x86 OSDN, kernel-4.9 branch. I didn't do any changes to kernel etc related stuff so I'll test how it goes and will report back! Thanks.

Answer 12 · 2016-12-21T17:50:38.000Z

Hi Henry, I'm writing you about my tests with nougat-x86 7.1.1r4 Due to changes in framework/base the zygote (forked) processes' opened files are checked against a whitelist, this security enforcement is currently incompatible with DEBUG mode (which uses chroot mechanism) The effect is that when booting in DEBUG mode the libandroid_runtime.so library will throw SIGABRT causing a systematic bootloop. Not using DEBUG mode boots fine, but I haven't found an acceptable solution for the problem yet, most probably the chroot mechanism should be avoided with nougat because it may have other issues besides the whitelist for forked processes. Regarding kernel-4.9 of android-x86 repo I cannot build as I see the errors I mentioned, I am currenly building this one https://github.com/maurossi/linux/commits/kernel-4.9 without the 4 wifi drivers causing building error, by changing the android defconfig files: # CONFIG_RTL8192U is not set # CONFIG_RTL8723BS is not set # CONFIG_RTL8723BU is not set # CONFIG_RTL8821AU is not set Mauro PS: I've read on phoronix blog that kernel-4.10 will bring improvements for Surface Pro 3 2016-12-21 11:28 GMT+01:00 Henri Koivuneva <notifications@github.com>:

…

Hi Mauro! Sorry, I had a very busy time at work and couldn't work on this. Unfortunately I was unable to get any logs out. The device is Surface Pro 3. I'm just now building both nougat and marshmallow with the newest code you guys have pushed to Android-x86 OSDN, kernel-4.9 branch. I didn't do any changes to kernel etc related stuff so I'll test how it goes and will report back! Thanks. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#2 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AH9VlI1Yr_dfmpfmo3_kZCrmomrWgwE-ks5rKP87gaJpZM4KL_FK> .

Answer 13 · 2016-12-21T19:00:22.000Z

@maurossi your build 7.1.0 with this patch，https://groups.google.com/d/msg/android-x86/stAxHcoJJP4/jlKrZdiaCAAJ ，debug mode no problem，no bootloops .

Answer 14 · 2016-12-21T19:48:28.000Z

Hi youling257,

the revert that removed on 7.1.0r6 the fd closures and whitelist enforcements was an hack,
I would prefer to add the /android prefixed paths to whitelist or use the patch proposed by Chris Vandomelen in the same thread (or similar),

the problem is that with 7.1.1r4 acting on the whitelist only does not work,
but you're right that applying substantial removal of whitelist enforcement avoids the bootloop.

Mauro

Answer 15 · 2016-12-22T18:24:03.000Z

Thank you! Kernel 4.9 seems to work pretty well on nougat. I wasn't able to build Marshmallow with it as the greybus errors that came earlier are coming again for me. I wasn't able to disable it so I'll build MM with the default 4.4

Userdebug builds boot without issues for me on the SP3. Do they work for you?
https://groups.google.com/d/msg/android-x86/MdaiRnECwtU/B3aadrM1DgAJ
Of course can't get that much information out when debugging is needed haha...

Any user build I try to boot gives the SysRq emergency remount R/O error, weird. Userdebug works well.

Yeah I saw that too! It also sounds like they'll have fixes etc. for Baytrail devices like the Z3735F that's under quite a lot of discussion on the forums. Exciting!

Answer 16 · 2017-05-13T05:13:53.000Z

4.12 upstream rtl8723bs support，no problem .
but 8723bu and 8812au compile error，have to

# CONFIG_RTL8723BU is not set
# CONFIG_RTL8821AU is not set

Answer 17 · 2017-05-13T09:46:50.000Z

I still need to update the defconfig files and perform a few changes to rtl8723bu and rtl8821au patches, in order to have kernel retro compatibility upstreamed in respective githubs, and maybe convince the authors to push to staging in linux-next Setting the CONFIG_ options will work for kernel-4.12 cycle, but I have no hw to test if my changes work Mauro 2017-05-13 7:13 GMT+02:00 youling257 <notifications@github.com>:

…

4.12 upstream rtl8723bs support，no problem . but 8723bu and 8812au compile error，have to CONFIG_RTL8723BU is not set CONFIG_RTL8821AU is not set — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#2 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AH9VlNIeTVq-eQ1qmyFsnbW8qL2u9IrXks5r5TwSgaJpZM4KL_FK> .

Answer 18 · 2017-05-13T11:43:06.000Z

unfortunately，I use cwhuang old 4.9.13 defconfig to get a new 4.11 config，can't boot 4.12 kernel ，I use newest torvalds source.
Now，4.12 not available，still merging，wait for 4.12 rc1 release.

Welcome to Ubuntu Artful Aardvark (development branch) (GNU/Linux 4.8.17-android-x86 i686)

root@localhost:/home/youling257# cd /data/media/0/Download/linux-master/
root@localhost:/data/media/0/Download/linux-master# bash /patch
patching file arch/x86/kernel/cpu/intel.c
Hunk #1 succeeded at 163 (offset 32 lines).
patching file drivers/gpu/drm/drm_ioctl.c
Hunk #1 succeeded at 496 (offset 16 lines).
patching file arch/x86/Makefile
patching file drivers/net/usb/usbnet.c
Hunk #1 succeeded at 1704 (offset 42 lines).
patching file drivers/net/wireless/marvell/mwifiex/main.c
Hunk #1 FAILED at 598.
1 out of 1 hunk FAILED -- saving rejects to file drivers/net/wireless/marvell/mwifiex/main.c.rej
patching file drivers/android/Kconfig
Hunk #1 succeeded at 32 with fuzz 1 (offset 12 lines).
patching file arch/x86/kernel/cpu/intel.c
Hunk #1 succeeded at 163 (offset 32 lines).
patching file arch/x86/entry/vdso/Makefile
patching file arch/x86/kernel/traps.c
Hunk #1 succeeded at 320 (offset 35 lines).
Hunk #2 succeeded at 341 (offset 35 lines).
patching file mm/shmem.c
Hunk #1 succeeded at 3320 (offset 163 lines).
Hunk #2 succeeded at 3343 (offset 163 lines).
patching file drivers/input/touchscreen/silead.c
Hunk #1 succeeded at 65 (offset 1 line).
Hunk #2 succeeded at 121 (offset 2 lines).

root@localhost:/home/youling257# cd /data/media/0/Download/linux-master/

root@localhost:/data/media/0/Download/linux-master#
diff drivers/net/wireless/marvell/mwifiex/main.c.orig drivers/net/wireless/marvell/mwifiex/main.c -u

--- drivers/net/wireless/marvell/mwifiex/main.c.orig 2017-05-13 17:11:54.484971407 +0800
+++ drivers/net/wireless/marvell/mwifiex/main.c 2017-05-13 17:15:54.969967990 +0800
@@ -594,7 +594,7 @@

    rtnl_lock();
    /* Create station interface by default */

- wdev = mwifiex_add_virtual_intf(adapter->wiphy, "mlan%d", NET_NAME_ENUM,
+ wdev = mwifiex_add_virtual_intf(adapter->wiphy, "wlan%d", NET_NAME_ENUM,
NL80211_IFTYPE_STATION, NULL);
if (IS_ERR(wdev)) {
mwifiex_dbg(adapter, ERROR,

root@localhost:/data/media/0/Download/linux-master#cat /patch
patch -p1 < /data/sdext2/patch/0001-x86-intel-force-tsc-to-be-reliable-on-Baytrail.patch
patch -p1 < /data/sdext2/patch/0002-drm-disable-permission-check.patch
patch -p1 < /data/sdext2/patch/0006-merge-x86-build-globally-set-fno-pic.patch
patch -p1 < /data/sdext2/patch/0013-usbnet-TEMP-HACK-force-the-interface-name-to-ethX.patch
patch -p1 < /data/sdext2/patch/0014-mwifiex-change-interface-name-to-wlanX.patch
patch -p1 < /data/sdext2/patch/0015-android-support-32-bit-binder-in-64-bit-kernel.patch
patch -p1 < /data/sdext2/patch/0016-x86-intel-force-tsc-to-be-reliable-on-Cherrytrail.patch
patch -p1 < /data/sdext2/patch/0017-x86-vdso-fix-a-build-break-on-CONFIG_FUNCTION_TRACER.patch
patch -p1 < /data/sdext2/patch/0025-SSSE3-emulation-for-invalid-opcode.patch
patch -p1 < /data/sdext2/patch/0026-shmem-enable-user-xattr-for-tmpfs.patch
patch -p1 <
/data/sdext2/root/silead_add_ts_mod_params.patch
root@localhost:/data/media/0/Download/linux-master#

Answer 19 · 2017-05-13T12:09:19.000Z

Try Bliss kernel until new stuff is merged: https://github.com/BlissRoms-x86/kernel_common/tree/kernel-4.12rc1-master Reportedly working on some devices but had issues with wifi, rotation, audio. BitOBSessiOn [3] did lots of hard work. youling257 kirjoitti 13.5.2017 14:43:

unfortunately，I use cwhuang old 4.9.13 defconfig to get a new 4.11 config，can't boot 4.12 kernel ，I use newest torvalds source. Now，4.12 not available，still merging，wait for 4.12 rc1 release. -- You are receiving this because you commented. Reply to this email directly, view it on GitHub [1], or mute the thread [2].

Links: ------ [1] #2 (comment) [2] https://github.com/notifications/unsubscribe-auth/AAh1qFTtbh27mwzHDId_5xv_8cU_23Uhks5r5ZdLgaJpZM4KL_FK [3] https://github.com/BlissRoms-x86/kernel_common/commits/kernel-4.12rc1-master?author=BitOBSessiOn

Answer 20 · 2017-05-13T12:44:12.000Z

@hkoivuneva ，diff my .config https://github.com/BlissRoms-x86/kernel_common/tree/kernel-4.12rc1-master android-x86_defconfig ，my config enable selinux .
https://pastebin.com/raw/CZXYtYYn
if BitOBSessiOn used cwhuang 4.9 rc6 defconfig get a new 4.11 config，then he edit this new config .
i use cwhuang 4.9.13 defconfig get a new 4.11 config，editor this new config for my tablet .

i download zip from https://github.com/torvalds/linux ，add 0001-x86-intel-force-tsc-to-be-reliable-on-Baytrail.patch, 0002-drm-disable-permission-check.patch, 0006-merge-x86-build-globally-set-fno-pic.patch, 0013-usbnet-TEMP-HACK-force-the-interface-name-to-ethX.patch, 0014-mwifiex-change-interface-name-to-wlanX.patch, 0015-android-support-32-bit-binder-in-64-bit-kernel.patch, 0016-x86-intel-force-tsc-to-be-reliable-on-Cherrytrail.patch, 0017-x86-vdso-fix-a-build-break-on-CONFIG_FUNCTION_TRACER.patch, 0025-SSSE3-emulation-for-invalid-opcode.patch, 0026-shmem-enable-user-xattr-for-tmpfs.patch ，these patches enough for Android x86 .

i download zip from https://github.com/jwrdegoede/linux-sunxi/commits/master ，add 0001-x86-intel-force-tsc-to-be-reliable-on-Baytrail.patch, 0002-drm-disable-permission-check.patch, 0006-merge-x86-build-globally-set-fno-pic.patch, 0013-usbnet-TEMP-HACK-force-the-interface-name-to-ethX.patch, 0014-mwifiex-change-interface-name-to-wlanX.patch, 0015-android-support-32-bit-binder-in-64-bit-kernel.patch, 0016-x86-intel-force-tsc-to-be-reliable-on-Cherrytrail.patch, 0017-x86-vdso-fix-a-build-break-on-CONFIG_FUNCTION_TRACER.patch, 0025-SSSE3-emulation-for-invalid-opcode.patch, 0026-shmem-enable-user-xattr-for-tmpfs.patch these patch，boot Android x86 nougat no problem .

Answer 21 · 2017-05-13T16:54:17.000Z

2017-05-13 7:13 GMT+02:00 youling257 <notifications@github.com>:

4.12 upstream rtl8723bs support，no problem . but 8723bu and 8812au compile error，have to CONFIG_RTL8723BU is not set CONFIG_RTL8821AU is not set

Hi, I re-read your message, but with current kernel-4.12rc1 (well it's actually an rc0) I have: CONFIG_RTL8723BU=m CONFIG_RTL8821AU=m and it is building without errors. What errors are you seeying? Mauro

Answer 22 · 2017-05-13T17:31:44.000Z

@maurossi ，127.0.0.1_20170514012534.txt
127.0.0.1_20170514012534.zip
because 8723bu、8812au compile error，leed to can't compile all modules .o files ，leed to can't compile all modules .
@hkoivuneva ，tried https://github.com/BlissRoms-x86/kernel_common/tree/kernel-4.12rc1-master and BitOBSessiOn defconfig，still can't boot .

Answer 23 · 2017-05-14T15:01:01.000Z

Yes，can't boot，stopped at "random crng init done"，what reason ？
4.12 rc1

Answer 24 · 2017-05-14T16:48:20.000Z

2017-05-14 17:01 GMT+02:00 youling257 <notifications@github.com>:

Yes，can't boot，stopped at "random crng init done"，what reason ？

No idea, I could boot x86 iso image on Core 2 Duo with Nvidia GT210 and x86_64 iso image on Asus H61M-K Celeron G1610 and Sandy Bridge, the problem I see is that DNS_PROBE FAIL but the rtl2x00 driver is ok (passphrase verified and IP address assigned) "No Internet" due to to some DNS client issues Mauro

Answer 25 · 2017-05-14T21:32:28.000Z

I need built in 4.12 kernel 32/64 bits nougat ISO to test .

Answer 26 · 2017-05-23T06:51:21.000Z

I compile 4.12 kernel for marshmallow, boot success, can't boot nought. i need your nought iso, maurossi. 2017年5月15日星期一，maurossi <notifications@github.com> 写道：

…

2017-05-14 17:01 GMT+02:00 youling257 ***@***.*** ***@***.***');>>: > Yes，can't boot，stopped at "random crng init done"，what reason ？ > No idea, I could boot x86 iso image on Core 2 Duo with Nvidia GT210 and x86_64 iso image on Asus H61M-K Celeron G1610 and Sandy Bridge, the problem I see is that DNS_PROBE FAIL but the rtl2x00 driver is ok (passphrase verified and IP address assigned) "No Internet" due to to some DNS client issues Mauro — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/APjKVKLeSZn-KV5iTvMDFYHGqelpJc8Kks5r5zBUgaJpZM4KL_FK> .

Answer 27 · 2017-05-23T07:09:21.000Z

u0_a27@x86_64:/ $ ifconfig eth0 Link encap:Ethernet HWaddr 00:0e:c6:c3:60:ac inet addr:192.168.1.3 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::20e:c6ff:fec3:60ac/64 Scope: Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:7 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1641 TX bytes:1006 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope: Host UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:474 errors:0 dropped:0 overruns:0 frame:0 TX packets:474 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:48390 TX bytes:48390 wlan0 Link encap:Ethernet HWaddr 8c:18:d9:27:71:71 inet addr:192.168.1.11 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::8e18:d9ff:fe27:7171/64 Scope: Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:12 errors:0 dropped:12 overruns:0 frame:0 TX packets:42 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:4722 TX bytes:6866 u0_a27@x86_64:/ $ but，Can not access the Internet ！ 2017年5月23日星期二，youling 257 <youling257@gmail.com> 写道：

…

I compile 4.12 kernel for marshmallow, boot success, can't boot nought. i need your nought iso, maurossi. 2017年5月15日星期一，maurossi ***@***.*** ***@***.***');>> 写道： > 2017-05-14 17:01 GMT+02:00 youling257 ***@***.***>: > > > Yes，can't boot，stopped at "random crng init done"，what reason ？ > > > > No idea, I could boot x86 iso image on Core 2 Duo with Nvidia GT210 > and x86_64 iso image on Asus H61M-K Celeron G1610 and Sandy Bridge, > > the problem I see is that DNS_PROBE FAIL but the rtl2x00 driver is ok > (passphrase verified and IP address assigned) > "No Internet" due to to some DNS client issues > > Mauro > > — > You are receiving this because you authored the thread. > Reply to this email directly, view it on GitHub > <#2 (comment)>, or mute > the thread > <https://github.com/notifications/unsubscribe-auth/APjKVKLeSZn-KV5iTvMDFYHGqelpJc8Kks5r5zBUgaJpZM4KL_FK> > . >

Answer 28 · 2017-05-26T11:50:25.000Z

I use cwhuang 4.9.27 defconfig https://osdn.net/projects/android-x86/scm/git/kernel/tree/kernel-4.9/arch/x86/configs/ to get a new 4.12 .config ，i use whole linux-sunxi source，compile kernel for nougat，i use 4.12 kernel boot success Android-x86/cm_android_x86/android_x86:7.1.1/NOF26W/4f52028002:userdebug/test-keys .

but no Internet，eth0 no Internet，wlan0 no Internet，i don't think dns problem . i tried “Override DNS” app .

Answer 29 · 2017-05-26T11:59:24.000Z

https://github.com/torvalds/linux/commits/master/drivers/net/wireless/marvell/mwifiex/main.c
https://github.com/torvalds/linux/commits/master/drivers/net/usb/usbnet.c
about usbnet-TEMP-HACK-force-the-interface-name-to-ethX.patch、mwifiex-change-interface-name-to-wlanX.patch