Dell Venue 8 Pro 5830 Baytrail Progress

[UniCav]

I wanted to start a fresh thread on this because I'm so close to having this tablet usable that I only need to tweak a few things, and my other threads on building kernels and images are getting too muddled up.

I'm able to build the iso from source, and using the Silvermont config seems to have 90% of everything working. It even seems to have solved the issue of the touch screen no longer responding if I boot with a physical keyboard attached.

Here's what I do not have solved:

1 - Sleep/Wake/Suspend/Resume - The tablet will automatically go to sleep, or I can press the Power/Sleep button and it goes to sleep.

   However, at random the backlight will come back on and run the battery down.

   Or the tablet will refuse to wake back up with the button.

   I either have to force it off with the button and reboot, or at randtom the tablet will reboot itself.

2 - Wifi is buggy, which is an issue with Atheros 6K chips.  It works fine overall, inlcuding wifi indicator.

   However at random if left downloading or something the wifi will lock up or show no longer connected.

   I'm not sure if this is a power-save feature in the wifi or if it's just a buggy wifi but there should be a tweak or two that would fix it.

3 - USB/OTG hotplugging will not work. I can not connect any USB devices after booting up. They are ignored.

   I have to boot with the devices connected, then shut down and disconnect them later.

That's really about all I'm missing right now. Other than that, the image with 4.18 kernel seems pretty stable and really usable. Good speed and response on the tablet.

I just need to fix the above, mainly the Sleep/Wake issue and the wifi dropping, and settle on a good source for apps which is looking like F-Droid except that Chrome and a few others aren't on there.

I was able to install Chome Beta from APK and it seems to work fine.

[DDS Central]

My Baytrail T100TA also suffers from power management issues and somewhat flaky Broadcom Wifi. I only tested regular Ubuntu though. I will do some builds based on the latest kernel as soon as my new SSD arrives.

As for the apps, you can always integrate the Play Store using OpenGapps.

--
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 email to android-x86...@googlegroups.com.
To post to this group, send email to andro...@googlegroups.com.
Visit this group at https://groups.google.com/group/android-x86.
For more options, visit https://groups.google.com/d/optout.

[UniCav]

I briefly tried OpenGApps but the first time I opened it and full screen video ads started hammering me I immediately dropped it

Can you suggest a simple guide on building the source wtih a new kernel? I've tried following a few guides and other's steps but I just can't seem to get an image built or kernel + modules built that I can transplant to the tablet.

[DDS Central]

The proper way is to add a source to manifest and use the repo command.

The easy way is to just replace the kernel directory and copy defconfigs from stock kernel (IIRC, they're in arch/x86/configs).

I will try to make a proper guide, but I do not have a build environment set up on my new rig yet.

[UniCav]

Yes the configs are in arch/x86/configs. I'm about to give it another go that way and see if I can get it to work. I'm still pretty green on Git and Repo so if you get a chance I'd appreciate some pointers on the "right" way

[youling 257]

can you provide dmesg.txt ?

can you ece mem > /sys/power/state ?

can you dmesg > /sdcard/dmesg.txt

what are you did in two month? you just do test iso, what is out of box for you, you never try to find problem reason, you never solved any problem. except someone tell you patch exfield.c

if i buy a dell venue 8 tablet, i can make everything out of box work, but i don't have, so what is you can cooperate with me

在 2018年12月20日星期四 UTC+8上午12:11:57，UniCav写道：

[youling 257]

how to tell you about bay trail suspend to idle s0i3 knowledge ? echo freeze > /sys/power/state

check reach s0ix state, /sys/kernel/debug/pmc_atom/sleep_state

how to tell you Androidx86 nougat sleep.earlysuspend=1 and power.doubleclick=1 knowledge 

在 2018年12月20日星期四 UTC+8上午12:11:57，UniCav写道：

I wanted to start a fresh thread on this because I'm so close to having this tablet usable that I only need to tweak a few things, and my other threads on building kernels and images are getting too muddled up.

[UniCav]

youling_257

what configuration change is needed to fix that on the baytrail/atom tablets?

When I look up SOI3, sleep.earlysuspend and power.doubleclick I fiind conflicting suggestions depending on OS and kernel versions.

I'm running Oreo-x86 with Kernel 4.18 and also trying to build an image with kernel 4.19

[youling 257]

My mode of thinking is not build image.

My mode of thinking is replace kernel and modules to Androidx86/Phoenixos, everything support will different.

在 2018年12月20日星期四 UTC+8上午2:14:04，UniCav写道：

[UniCav]

I've been trying that but I always wind up with a non-functioning system.

running the commands in terminal on the tablet is very difficult.

I remount / as rw, and mount the Android-x86 data directory

I extract your linux-4.19.7-android-x86_64+-x86 files

I copy boot/vmlinuz-4.19.7-android-x86_64+-x86 to "kernel"

I have to delete lib/modules/Linux-4.18.1 because of space

I copy the new lib/modules/4.18.7.android-x86_64+ to the /lib/modules directory.

Tablet reboots, Grub comes up, Goes to black/white border pre-boot screen, then nothing.

I have also tried unpacking system.sfs and replacing the folder in the system.img then repacking system.sfs

and placing that into the system.

Same effect. System will not fully boot up and I can't get far enough to troubleshoot anything.

So something in my system building the image must be different enough that your prebuilt kernel and modules will not function on my tablet.

That's why I keep going back to rebuilding the ISO from source.

I either need to either change some setting in the tablet,

or I need to change the 4.18 config before building the image

or I need to upgrade to 4.19 and config properly and build image

But I have to find a different way to get this working, simply replacing the kernel and modules does not work with this tablet

[UniCav]

After several attempts to build with the kernel 4.19.11 source I still can't get an ISO. Always fails at the point of building the bZimage.

I can build the kernel separately for linux but not for Android-x86

I've tried 3 different targets - android_x86, android_x86-userdebug and android_x86-eng

I can get into menuconfig and edit/save the .config

I've built the defconfig, tried with -j4 and -j1, tried simply /usr/bin/make iso_img which normally will work when nothing else will.

Some examples of the last few attempts

lunch android_x86
/usr/bin/make -j4 iso_img

failed at point of make bZimage

make clean
cd kernel
make mrproper
cd ..
/usr/bin/make -j4 iso_img

failed at same point

make clean
cd kernel
make distclean
make mrproper
cd oreo-x86
/usr/bin/make iso_img

also fails at same point

make clean
. build/envsetup.sh
lunch
/usr/bin/make iso_img

also fails at same point.

End of last soong log

2018/12/19 13:44:06.298115 ui/build/exec.go:57: prebuilts/build-tools/linux-x86/bin/ckati [prebuilts/build-tools/linux-x86/bin/ckati --ninja --ninja_dir=out --ninja_suffix=-android_x86 --regen --ignore_optional_include=out/%.P --detect_android_echo --color_warnings --gen_all_targets --werror_find_emulator -f build/core/main.mk --use_find_emulator BUILDING_WITH_NINJA=true SOONG_ANDROID_MK=out/soong/Android-android_x86.mk SOONG_MAKEVARS_MK=out/soong/make_vars-android_x86.mk]
2018/12/19 13:44:48.991208 ui/build/exec.go:57: prebuilts/build-tools/linux-x86/bin/ninja [prebuilts/build-tools/linux-x86/bin/ninja -d keepdepfile iso_img -j 1 -f out/combined-android_x86.ninja -w dupbuild=err]
2018/12/19 14:54:55.790054 ui/build/exec.go:95: ninja failed with: exit status 1

end of ninja log

4201694    4205376    1545252894    /home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel/.config    798111239621ab24

no failures, errors or warnings in the log

Again I wouldn't be trying to build another kernel or image at all if I could sort out why the screen won't stay asleep or the tablet won't properly sleep/hibernate and won't resume most of the time from sleep. But I find it annoying to have to power off the tablet every time I'm through with it or boot back up every time I want to use it. Really is the straw that breaks the usability of the tablet.

If anyone can suggest another way to get this to build or what I can do on the tablet directly to configure the sleep/wake in a usable state I'd appreciate it.

[UniCav]

Currently trying a new tactic. I'm using  Youling 257's repo of the 4.19 kernel to build the kernel as directed. Got an x86_64 and x_86 kernel both to build.

Burned my Silvermont ISO which worked best so far using Windows, since my Ubuntu VM refuses to burn a USB that can be written to again,

then I'll try swapping the kernel file, editing system.sfs to change the modules directory, and see if I have any luck that way.

[Chih-Wei Huang]

UniCav <uni...@gmail.com> 於 2018年12月22日 週六 上午1:26寫道：

>
> After several attempts to build with the kernel 4.19.11 source I still can't get an ISO. Always fails at the point of building the bZimage.

What kernel source did you use exactly?
If you used vanilla kernel, it won't build
with android toolchain and won't run
unless you know what patches are necessary.
If you don't have the knowledge,
you should build the kernel branch in android-x86 repo
first, once it succeeds, then try to customize it.

> I can build the kernel separately for linux but not for Android-x86
> I've tried 3 different targets - android_x86, android_x86-userdebug and android_x86-eng
> I can get into menuconfig and edit/save the .config
> I've built the defconfig, tried with -j4 and -j1, tried simply /usr/bin/make iso_img which normally will work when nothing else will.
> Some examples of the last few attempts
> lunch android_x86
> /usr/bin/make -j4 iso_img

OH!
If you are building with oreo-x86 (8.1) source,
you must use "make -j4 iso_img" instead of
"/usr/bin/make -j4 iso_img".

THIS IS VERY IMPORTANT!
AOSP has wrapped "make" command
to invoke the soong build system.
If you still use the original "make" command,
it definitely won't build complete.

Note this is only true since Oreo.
Before 8.0 "make" is just "make".

Finally, to check building errors, you have to save
the whole build log into a file to search the
actual errors. The final error message usually
doesn't help.


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

[DDS Central]

Vanilla kernels used to build fine in the past. The only patch I know is needed is disabling those Intel CPU bug workarounds.

What else is needed to build vanilla Linux with Oreo branch ? 

[youling 257]

gcc8 build Vanilla kernels, only need one patch"disable drm permission check", needn't any more patch, make Bzimage and modules to boot any one Androidx86 prebuild image enough, extract .config from any one prebuild image /proc/config.gz

在 2018年12月22日星期六 UTC+8上午11:32:28，Povilas Staniulis写道：

[Chih-Wei Huang]

DDS Central <wdmo...@gmail.com> 於 2018年12月22日 週六 上午11:32寫道：

>
> Vanilla kernels used to build fine in the past. The only patch I know is needed is disabling those Intel CPU bug workarounds.
> What else is needed to build vanilla Linux with Oreo branch ?

You can surely build vanilla kernel with normal Linux toolchain.
However, it can't be built with android toolchain directly.
If you're curious, just try yourself.

Anyway, all patches are in our repo, of course.

> On Sat, Dec 22, 2018, 05:22 Chih-Wei Huang <cwh...@android-x86.org wrote:
>>
>> UniCav <uni...@gmail.com> 於 2018年12月22日 週六 上午1:26寫道：
>> >
>> > After several attempts to build with the kernel 4.19.11 source I still can't get an ISO. Always fails at the point of building the bZimage.
>>
>> What kernel source did you use exactly?
>> If you used vanilla kernel, it won't build
>> with android toolchain and won't run
>> unless you know what patches are necessary.
>> If you don't have the knowledge,
>> you should build the kernel branch in android-x86 repo
>> first, once it succeeds, then try to customize it.

[Povilas Staniulis]

I've done many builds in the past and never had issues just replacing
the kernel directory of normal Android x86 repo with a custom kernel
(with a couple minor patches, like disabling GPU auth).
What has changed in modern kernels so they no longer build with our NDK
toolchain ?

[Povilas Staniulis]

I've done many builds in the past and never had any issues replacing
stock Android x86 kernel with a vanilla one (with a couple minor

patches, like disabling GPU auth).

What has changed so that we can no longer build vanilla Linux with
Android NDK toolchain ? GCC too old ?

On 2018-12-22 11:50, Chih-Wei Huang wrote:

[Kitsune]

I'm also have usually have freezes when use power button to sleep, but echo freeze > /sys/power/state more stable than power button, why?
Need fix this error, sometimes:
dpm_run_callback(): platform_pm_suspend+0x0/0x50 returns -16
[ 82.855532] PM: Device alarmtimer failed to suspend: error -16
?

[ 61.617202] modprobe: /sbin/modprobe net-pf-16-proto-4-type-2-6
[ 62.171182] smpboot: CPU 1 is now offline
[ 62.194913] smpboot: CPU 2 is now offline
[ 62.211623] smpboot: CPU 3 is now offline
[ 62.212501] PM: suspend entry (s2idle)
[ 62.212504] PM: Syncing filesystems ... done.
[ 62.231453] Freezing user space processes ... (elapsed 0.003 seconds) done.
[ 62.235190] OOM killer disabled.
[ 62.235192] Freezing remaining freezable tasks ... (elapsed 0.000 seconds) done.
[ 62.236077] Suspending console(s) (use no_console_suspend to debug)
[ 62.249278] byt_ext_speakers_event, OFF
[ 62.254528] ------------[ cut here ]------------
[ 62.254530] reqid not found
[ 62.254610] WARNING: CPU: 0 PID: 3470 at brcmf_pno_remove_request+0x57/0xc0 [brcmfmac]
[ 62.254612] Modules linked in: tcp_diag inet_diag binfmt_misc rfcomm bnep btusb btrtl gpio_keys soc_button_array coretemp hwmon intel_rapl intel_soc_dts_thermal intel_powerclamp crc32_pclmul crc32c_intel ghash_clmulni_intel int3406_thermal intel_int0002_vgpio pcspkr processor_thermal_device intel_soc_dts_iosf joydev hid_multitouch i2c_hid brcmfmac brcmutil cfg80211 int3403_thermal int340x_thermal_zone snd_soc_sst_bytcr_wm5102 snd_intel_sst_acpi snd_intel_sst_core snd_soc_sst_atom_hifi2_platform snd_soc_acpi_intel_match snd_soc_acpi dw_dmac snd_soc_wm5102 snd_soc_wm_adsp snd_soc_arizona gpio_arizona arizona_spi spi_pxa2xx_platform efi_pstore efivars lsm303d kfifo_buf industrialio int3400_thermal acpi_thermal_rel rfkill_gpio hci_uart btbcm btqca btintel bluetooth ecdh_generic 8250_dw lpc_ich mac_hid
[ 62.254696] i915 drm_kms_helper drm fb_sys_fops syscopyarea sysfillrect sysimgblt i2c_algo_bit
[ 62.254714] CPU: 0 PID: 3470 Comm: kworker/u8:6 Tainted: G W 4.15.18-android-x86_64-ge3e28e88b6de #1
[ 62.254717] Hardware name: LENOVO 60073/INVALID, BIOS 01WT17WW 08/01/2014
[ 62.254726] Workqueue: events_unbound async_run_entry_fn
[ 62.254750] RIP: 0010:brcmf_pno_remove_request+0x57/0xc0 [brcmfmac]
[ 62.254754] RSP: 0000:ffff947586433c88 EFLAGS: 00010286
[ 62.254758] RAX: 0000000000000000 RBX: ffff90852fcad240 RCX: ffffffff93a37038
[ 62.254761] RDX: ffffffff93a37038 RSI: 0000000000000086 RDI: 0000000000000202
[ 62.254763] RBP: ffff90852fcad2c8 R08: 0000000000000000 R09: ffffffff93f966db
[ 62.254766] R10: 00000000000093dc R11: 00000000000003dd R12: 0000000000000000
[ 62.254768] R13: 0000000000000000 R14: 0000000000000000 R15: ffff908534784880
[ 62.254772] FS: 0000000000000000(0000) GS:ffff90853aa00000(0000) knlGS:0000000000000000
[ 62.254775] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 62.254778] CR2: 0000780f4aa01360 CR3: 0000000028121000 CR4: 00000000001006f0
[ 62.254780] Call Trace:
[ 62.254810] brcmf_pno_stop_sched_scan+0x30/0x70 [brcmfmac]
[ 62.254831] brcmf_cfg80211_sched_scan_stop+0x2d/0x70 [brcmfmac]
[ 62.254852] brcmf_cfg80211_suspend+0x338/0x350 [brcmfmac]
[ 62.254907] wiphy_suspend+0x8c/0x340 [cfg80211]
[ 62.254951] ? wiphy_resume+0x1a0/0x1a0 [cfg80211]
[ 62.254979] dpm_run_callback+0x6e/0x1f0
[ 62.254989] __device_suspend+0x17e/0x580
[ 62.254995] ? finish_task_switch+0x90/0x1e0
[ 62.255002] ? sched_clock+0x5/0x10
[ 62.255007] async_suspend+0x1e/0x90
[ 62.255012] async_run_entry_fn+0x31/0x150
[ 62.255020] process_one_work+0x133/0x3c0
[ 62.255026] worker_thread+0x43/0x460
[ 62.255032] ? process_one_work+0x3c0/0x3c0
[ 62.255037] ? process_one_work+0x3c0/0x3c0
[ 62.255042] kthread+0xfa/0x130
[ 62.255048] ? __kthread_init_worker+0xb0/0xb0
[ 62.255053] ? do_syscall_64+0x6d/0x130
[ 62.255059] ? SyS_exit+0x17/0x20
[ 62.255065] ret_from_fork+0x35/0x40
[ 62.255069] Code: 74 43 48 89 da 31 c0 eb 0d 48 8b 4a 10 48 83 c2 08 4c 39 21 74 31 83 c0 01 39 f0 75 ec 48 c7 c7 3a be 86 c0 31 c0 e8 f9 bd 03 d2 <0f> 0b 41 bc fe ff ff ff 48 89 ef e8 d9 8b 87 d2 5b 5d 44 89 e0
[ 62.255150] ---[ end trace 4996921cb5a8b9b7 ]---
[ 62.256245] dpm_run_callback(): platform_pm_suspend+0x0/0x50 returns -16
[ 62.256249] PM: Device alarmtimer failed to suspend: error -16
[ 63.265841] PM: Some devices failed to suspend, or early wake event detected
[ 63.283818] Fennec: byt_wm5102_prepare_and_enable_pll1 - start
[ 63.284523] OOM killer enabled.
[ 63.284524] Restarting tasks ...
[ 63.313191] intel_sst_acpi 80860F28:00: FW sent error response 0x4000e
[ 63.313205] intel_sst_acpi 80860F28:00: free stream returned err -1
[ 63.323959] byt_ext_speakers_event, ON
[ 63.325118] byt_ext_speakers_event, OFF
[ 63.342024] done.
[ 63.371550] PM: suspend exit
[ 63.491823] x86: Booting SMP configuration:
[ 63.491827] smpboot: Booting Node 0 Processor 1 APIC 0x2
[ 63.496725] smpboot: Booting Node 0 Processor 2 APIC 0x4
[ 63.535128] smpboot: Booting Node 0 Processor 3 APIC 0x6
[ 63.569083] mmc0: Problem switching card into high-speed mode!
[ 65.044838] type=1400 audit(1545577815.881:869): avc: denied { map } for pid=1752 comm="putmethod.latin" path="/system/app/LatinIME/LatinIME.apk" dev="loop0" ino=150 scontext=u:r:untrusted_app:s0:c512,c768 tcontext=u:object_r:system_file:s0 tclass=file permissive=1
[ 66.015881] Fennec: byt_wm5102_prepare_and_enable_pll1 - start
[ 66.031061] Fennec: byt_wm5102_prepare_and_enable_pll1 - start
[ 66.068005] byt_ext_speakers_event, ON
[ 67.371400] type=1400 audit(1545577818.207:870): avc: denied { map } for pid=3528 comm="iptables" path="/dev/__properties__/properties_serial" dev="tmpfs" ino=6015 scontext=u:r:netd:s0 tcontext=u:object_r:properties_serial:s0 tclass=file permissive=1
[ 67.407378] modprobe: /sbin/modprobe ipt_IDLETIMER
[ 67.407401] type=1400 audit(1545577818.244:871): avc: denied { map } for pid=3535 comm="modprobe" path="/dev/__properties__/properties_serial" dev="tmpfs" ino=6015 scontext=u:r:init:s0 tcontext=u:object_r:properties_serial:s0 tclass=file permissive=1
[ 67.440698] type=1400 audit(1545577818.277:872): avc: denied { map } for pid=3535 comm="modprobe" path="/dev/__properties__/u:object_r:default_prop:s0" dev="tmpfs" ino=5988 scontext=u:r:init:s0 tcontext=u:object_r:default_prop:s0 tclass=file permissive=1
[ 67.440857] type=1400 audit(1545577818.277:873): avc: denied { map } for pid=3535 comm="modprobe" path="/dev/__properties__/u:object_r:log_tag_prop:s0" dev="tmpfs" ino=6000 scontext=u:r:init:s0 tcontext=u:object_r:log_tag_prop:s0 tclass=file permissive=1
[ 67.440979] type=1400 audit(1545577818.277:874): avc: denied { map } for pid=3535 comm="modprobe" path="/dev/__properties__/u:object_r:logd_prop:s0" dev="tmpfs" ino=5995 scontext=u:r:init:s0 tcontext=u:object_r:logd_prop:s0 tclass=file permissive=1
[ 67.449427] modprobe: /sbin/modprobe ip6t_IDLETIMER
[ 67.483329] init: write_file: Unable to open '/proc/sys/net/ipv4/tcp_default_init_rwnd': No such file or directory
[ 67.483701] init: Starting service 'exec 10 (/system/bin/logwrapper)'...
[ 67.485004] type=1400 audit(1545577818.321:875): avc: denied { execute_no_trans } for pid=3539 comm="init" path="/system/bin/logwrapper" dev="loop0" ino=388 scontext=u:r:init:s0 tcontext=u:object_r:system_file:s0 tclass=file permissive=1
[ 67.485347] type=1400 audit(1545577818.321:876): avc: denied { map } for pid=3539 comm="logwrapper" path="/system/bin/logwrapper" dev="loop0" ino=388 scontext=u:r:init:s0 tcontext=u:object_r:system_file:s0 tclass=file permissive=1
[ 67.508913] type=1400 audit(1545577818.345:877): avc: denied { execute_no_trans } for pid=3542 comm="logwrapper" path="/system/bin/sh" dev="loop0" ino=504 scontext=u:r:init:s0 tcontext=u:object_r:shell_exec:s0 tclass=file permissive=1
[ 67.509662] type=1400 audit(1545577818.346:878): avc: denied { map } for pid=3542 comm="sh" path="/system/bin/sh" dev="loop0" ino=504 scontext=u:r:init:s0 tcontext=u:object_r:shell_exec:s0 tclass=file permissive=1
[ 67.526383] init: Service 'exec 10 (/system/bin/logwrapper)' (pid 3539) exited with status 0
[ 82.402248] type=1400 audit(1545577833.238:893): avc: denied { map } for pid=1758 comm="ndroid.systemui" path="/system/priv-app/SystemUI/SystemUI.apk" dev="loop0" ino=6333 scontext=u:r:platform_app:s0:c512,c768 tcontext=u:object_r:system_file:s0 tclass=file permissive=1
[ 82.812356] PM: suspend entry (s2idle)
[ 82.812366] PM: Syncing filesystems ... done.
[ 82.830163] Freezing user space processes ... (elapsed 0.003 seconds) done.
[ 82.833280] OOM killer disabled.
[ 82.833283] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
[ 82.834612] Suspending console(s) (use no_console_suspend to debug)
[ 82.847565] byt_ext_speakers_event, OFF
[ 82.855527] dpm_run_callback(): platform_pm_suspend+0x0/0x50 returns -16
[ 82.855532] PM: Device alarmtimer failed to suspend: error -16
[ 82.859820] ------------[ cut here ]------------
[ 82.859823] reqid not found
[ 82.859913] WARNING: CPU: 1 PID: 3471 at brcmf_pno_remove_request+0x57/0xc0 [brcmfmac]
[ 82.859915] Modules linked in: tcp_diag inet_diag binfmt_misc rfcomm bnep btusb btrtl gpio_keys soc_button_array coretemp hwmon intel_rapl intel_soc_dts_thermal intel_powerclamp crc32_pclmul crc32c_intel ghash_clmulni_intel int3406_thermal intel_int0002_vgpio pcspkr processor_thermal_device intel_soc_dts_iosf joydev hid_multitouch i2c_hid brcmfmac brcmutil cfg80211 int3403_thermal int340x_thermal_zone snd_soc_sst_bytcr_wm5102 snd_intel_sst_acpi snd_intel_sst_core snd_soc_sst_atom_hifi2_platform snd_soc_acpi_intel_match snd_soc_acpi dw_dmac snd_soc_wm5102 snd_soc_wm_adsp snd_soc_arizona gpio_arizona arizona_spi spi_pxa2xx_platform efi_pstore efivars lsm303d kfifo_buf industrialio int3400_thermal acpi_thermal_rel rfkill_gpio hci_uart btbcm btqca btintel bluetooth ecdh_generic 8250_dw lpc_ich mac_hid
[ 82.860038] i915 drm_kms_helper drm fb_sys_fops syscopyarea sysfillrect sysimgblt i2c_algo_bit
[ 82.860131] CPU: 1 PID: 3471 Comm: kworker/u8:7 Tainted: G W 4.15.18-android-x86_64-ge3e28e88b6de #1
[ 82.860160] Hardware name: LENOVO 60073/INVALID, BIOS 01WT17WW 08/01/2014
[ 82.860171] Workqueue: events_unbound async_run_entry_fn
[ 82.860226] RIP: 0010:brcmf_pno_remove_request+0x57/0xc0 [brcmfmac]
[ 82.860230] RSP: 0018:ffff94758674bc88 EFLAGS: 00010286
[ 82.860235] RAX: 0000000000000000 RBX: ffff90852fcad240 RCX: ffffffff93a37038
[ 82.860238] RDX: ffffffff93a37038 RSI: 0000000000000086 RDI: 0000000000000202
[ 82.860241] RBP: ffff90852fcad2c8 R08: 0000000000000000 R09: ffffffff93f98793
[ 82.860244] R10: 0000000000007324 R11: 0000000000000433 R12: 0000000000000000
[ 82.860247] R13: 0000000000000000 R14: 0000000000000000 R15: ffff908534784880
[ 82.860251] FS: 0000000000000000(0000) GS:ffff90853aa80000(0000) knlGS:0000000000000000
[ 82.860254] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 82.860257] CR2: 0000780f4aa01360 CR3: 000000001c00a000 CR4: 00000000001006e0
[ 82.860260] Call Trace:
[ 82.860293] brcmf_pno_stop_sched_scan+0x30/0x70 [brcmfmac]
[ 82.860317] brcmf_cfg80211_sched_scan_stop+0x2d/0x70 [brcmfmac]
[ 82.860341] brcmf_cfg80211_suspend+0x338/0x350 [brcmfmac]
[ 82.860402] wiphy_suspend+0x8c/0x340 [cfg80211]
[ 82.860452] ? wiphy_resume+0x1a0/0x1a0 [cfg80211]
[ 82.860461] dpm_run_callback+0x6e/0x1f0
[ 82.860469] __device_suspend+0x17e/0x580
[ 82.860477] ? __wake_up_common_lock+0x8c/0xf0
[ 82.860483] async_suspend+0x1e/0x90
[ 82.860488] async_run_entry_fn+0x31/0x150
[ 82.860497] process_one_work+0x133/0x3c0
[ 82.860504] worker_thread+0x43/0x460
[ 82.860511] ? process_one_work+0x3c0/0x3c0
[ 82.860517] ? process_one_work+0x3c0/0x3c0
[ 82.860523] kthread+0xfa/0x130
[ 82.860529] ? __kthread_init_worker+0xb0/0xb0
[ 82.860536] ret_from_fork+0x35/0x40
[ 82.860541] Code: 74 43 48 89 da 31 c0 eb 0d 48 8b 4a 10 48 83 c2 08 4c 39 21 74 31 83 c0 01 39 f0 75 ec 48 c7 c7 3a be 86 c0 31 c0 e8 f9 bd 03 d2 <0f> 0b 41 bc fe ff ff ff 48 89 ef e8 d9 8b 87 d2 5b 5d 44 89 e0
[ 82.860633] ---[ end trace 4996921cb5a8b9b8 ]---
[ 83.874456] PM: Some devices failed to suspend, or early wake event detected
[ 83.894371] OOM killer enabled.
[ 83.894373] Restarting tasks ... done.

среда, 19 декабря 2018 г., 19:05:29 UTC+2 пользователь youling 257 написал:

[Kitsune]

Seems echo freeze > /sys/power/state works very good! How it use on power button?

среда, 19 декабря 2018 г., 19:10:45 UTC+2 пользователь youling 257 написал:

[youling 257]

give root permission to tasker, import this xml.

myself sdcard/suspend.txt is

echo 1 > /sys/class/gpio/gpio339/value

setprop sys.usb.config none

echo mem > /sys/power/state

echo 0 > /sys/class/gpio/gpio339/value

setprop sys.usb.config mtp

在 2018年12月24日星期一 UTC+8上午12:39:45，Kitsune写道：

[spin877]

I would like to report my experiences with intel Baytrail and the 4.20 kernel with the suspend/resume.

after several attempts I managed to start the >4.20 kernel, with Oreo

but after the resumes at times the i2c peripherals lock, giroscope sensors and so on cease to function. 

there have been too many changes have been "cleaned up"

/drivers/i2c/bus/i2c-designWare-baytrail.c

I think the problems with the latest kernels and the baytrail platform are all here.

P.S. 

I also tried that too many WARN_ON () in these functions. it blocked me from booting the kernel.

but maybe I'm wrong. and this change is useless

--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c

+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c

@@ -301,7 +301,7 @@ static void ttm_bo_vm_open(struct vm_area_struct *vma)

        struct ttm_buffer_object *bo =

            (struct ttm_buffer_object *)vma->vm_private_data;

 

-       WARN_ON(bo->bdev->dev_mapping != vma->vm_file->f_mapping);

+//  WARN_ON(bo->bdev->dev_mapping != vma->vm_file->f_mapping);

 

        (void)ttm_bo_reference(bo);

 }

diff --git a/drivers/i2c/busses/i2c-designware-baytrail.c b/drivers/i2c/busses/i2c-designware-baytrail.c

index dbda8c9..2d3c4f2 100644

--- a/drivers/i2c/busses/i2c-designware-baytrail.c

+++ b/drivers/i2c/busses/i2c-designware-baytrail.c

@@ -128,7 +128,7 @@ static int baytrail_i2c_acquire(struct dw_i2c_dev *dev)

        else

                dev_err(dev->dev, "PUNIT SEM: %d\n", sem);

 

-       WARN_ON(1);

+//  WARN_ON(1);

 

        return -ETIMEDOUT;

 }

diff --git a/fs/buffer.c b/fs/buffer.c

index cabc045..3fb0528 100644

--- a/fs/buffer.c

+++ b/fs/buffer.c

@@ -1133,7 +1133,7 @@ void __brelse(struct buffer_head * buf)

                put_bh(buf);

                return;

        }

-       WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n");

+//  WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n");

 }

 EXPORT_SYMBOL(__brelse);

[youling 257]

4.20 kernel only one i2c patch https://github.com/jwrdegoede/linux-sunxi/commit/6aaf09b8662eb36d42272d5a1ba687ca6d09d30c

4.20 kernel everything out of box work for Chinese manufacturer ONDA、teclast、cube、chuwi dual os bay trail tablet, needn't anymore patch from Hans, 

so my work on 4.20、4.21(may be 5.0) is tracking sdcardfs and configfs to mainline kernel, these Chinese manufacturer dual os tablet has gadget hardware support.

android 8.0 compare with 7.0 is functionfs mtp.

4.20 kernel fuse from 128k to 1M, cause when open some app will auto reboot, if used sdcardfs no this problem, sdcardfs can worked on 4.20 kernel.

在 2018年12月24日星期一 UTC+8上午3:07:25，spin877写道：

[UniCav]

Youling_257 I had tried your "android-4.19" repo. I previously had not been able to use the ASOP make because both the "-j" and "-C" options would error out as unknown. I had to use /usr/bin/make to get anything to build.

So at this point I can build the kernel from your source with make targz-pkg -i -j4. I have built both x64 and x86 kernels separately.

But I need to get an ISO built with the new kernels so I can install it. After the image is installed on the tablet it's almost impossible to do any changes on the tablet itself.

I've tried putting the 4.19 kernel into the rest of the build folder but it's failed to build every time.

I'll try again and make sure to use the AOSP make to see if it will build.

[UniCav]

This is what I get every time I try to build the iso_img with anything other than the stock 4.18 kernel in the oreo-x86 repo

make: Entering directory '/home/paradise/android-x86/oreo-x86/kernel'
make[1]: Entering directory '/home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel'
  HOSTCC  scripts/basic/fixdep
  GEN     ./Makefile
  HOSTCC  scripts/kconfig/conf.o
  YACC    scripts/kconfig/zconf.tab.c
  LEX     scripts/kconfig/zconf.lex.c
  HOSTCC  scripts/kconfig/zconf.tab.o
  HOSTLD  scripts/kconfig/conf
scripts/kconfig/conf  --olddefconfig Kconfig
.config:7889:warning: override: reassigning to symbol SECURITY_PATH
.config:7890:warning: override: reassigning to symbol SECURITY_SELINUX_BOOTPARAM
.config:7892:warning: override: reassigning to symbol SECURITY_SELINUX_CHECKREQPROT_VALUE
.config:7893:warning: override: reassigning to symbol DEFAULT_SECURITY_SELINUX
.config:7893:warning: override: DEFAULT_SECURITY_SELINUX changes choice state
.config:7894:warning: override: reassigning to symbol DEFAULT_SECURITY_DAC
.config:7895:warning: override: reassigning to symbol DEFAULT_SECURITY
#
# configuration written to .config
#
make[1]: Leaving directory '/home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel'
make: Leaving directory '/home/paradise/android-x86/oreo-x86/kernel'
[  9% 7406/77576] build /home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel/arch/x86/boot/bzImage
FAILED: /home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel/arch/x86/boot/bzImage
/bin/bash -c "make -j4 -j\$(nproc) -l\$((\$(nproc)+2)) -C kernel O=/home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel ARCH=x86 CROSS_COMPILE=\" /home/paradise/android-x86/oreo-x86/prebuilts/gcc/linux-x86/x86/x86_64-linux-android-4.9/bin/x86_64-linux-android-\"  YACC=/home/paradise/android-x86/oreo-x86/prebuilts/misc/linux-x86/bison/bison LEX=/home/paradise/android-x86/oreo-x86/prebuilts/misc/linux-x86/flex/flex-2.5.39  bzImage modules"
make: Entering directory '/home/paradise/android-x86/oreo-x86/kernel'
make[1]: Entering directory '/home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel'
arch/x86/Makefile:227: *** You are building kernel with non-retpoline compiler, please update your compiler..  Stop.
make[1]: Leaving directory '/home/paradise/android-x86/oreo-x86/out/target/product/x86/obj/kernel'
Makefile:146: recipe for target 'sub-make' failed
make: *** [sub-make] Error 2
make: Leaving directory '/home/paradise/android-x86/oreo-x86/kernel'
[  9% 7408/77576] Build hyb out/target/product/x86/obj/ETC/hyph-hu_...es/hyph-hu.hyb <- external/hyphenation-patterns/hu//hyph-hu.pat.txt
21515 unique nodes, 102669 total
[  9% 7409/77576] target Generated: libmesa_amd_common <= external/mesa/src/amd/common/sid_tables.py
ninja: build stopped: subcommand failed.
10:34:52 ninja failed with: exit status 1

[UniCav]

And same issue as before if I replace the old kernel with new kernel on the USB,

unsquash system.sfs, edit the system.img file and replace the kernel modules, then resquash and copy back to the USB stick

Grub boots but Live or Install both fail. I get an initial white border/black screen and never anything more.

So again either I have to figure out what's wrong and build the image with the newer kernel or just figure out how to fix the issues in the existing 418 kernel in Oreo

[Mauro Rossi]

Hi,

the retpoline problem happens only in kernel 4.20

Just disable retpoline in the kernel config file, as the alternative would be to add recent prebuilt clang binaries,

but I don't know how to update the scripted building rules to use newer compiler.

If you want to give a try the patch to AutoInstall with update, it will update kernel too,

but without all this pain you are experimenting

https://groups.google.com/forum/#!topic/android-x86/St2g8fpypUU

Mauro

[UniCav]

I got the 4.19 kernel from the android-x86 osdn repo and put it in my oreo-x86 folder and it built the image completely.

However it's doing the same thing as Youling's 4.19 source. Grub boots but once I choose Live, debug or Install it gives me the blank, white-bordered screen.

So this is something similar to the old i915 issue that required nomodeset to boot. I did try adding nomodeset in the grub command line but that also didn't work.

I'm wondering if it's actually something in the x86 kernel vs the x86_64 that this tablet doesn't like. The last few images I've actually built were x86 on Mauro's advice to only use x86 for the atom cpu.

The last working image I installed was a fresh download of the 32bit 8.1 RC2 ISO. It did boot and install but still had the battery always charging icon issue (unpatched exfield.c)

I'll try building the x86_64 image and see what I get.

Mauro if I put this back on the original Silvermont image I had that has worked the best so far, where is the auto-update script that I need to patch?

[Mauro Rossi]

Hi,

On Wednesday, December 26, 2018 at 10:28:14 PM UTC+1, UniCav wrote:

I got the 4.19 kernel from the android-x86 osdn repo and put it in my oreo-x86 folder and it built the image completely.

However it's doing the same thing as Youling's 4.19 source. Grub boots but once I choose Live, debug or Install it gives me the blank, white-bordered screen.

So this is something similar to the old i915 issue that required nomodeset to boot. I did try adding nomodeset in the grub command line but that also didn't work.

I'm wondering if it's actually something in the x86 kernel vs the x86_64 that this tablet doesn't like. The last few images I've actually built were x86 on Mauro's advice to only use x86 for the atom cpu.

The usual website CPUboss was givinng erroneous information,

but Z3740D supports SSE4.1 and SSE4.2 so x86_64 should be fine

and you should not need to change build combo to select Silvermont

the stock android_x86_64-user should be fine

http://www.cpu-world.com/CPUs/Atom/Intel-Atom%20Z3740D.html

 

The last working image I installed was a fresh download of the 32bit 8.1 RC2 ISO. It did boot and install but still had the battery always charging icon issue (unpatched exfield.c)

I'll try building the x86_64 image and see what I get.

Mauro if I put this back on the original Silvermont image I had that has worked the best so far, where is the auto-update script that I need to patch?

I am also uploading a couple of kernel-4.20, images x86 and x86_64, with youling mentioned patch,

with latest mesa-dev, gbm_gralloc + drm_framebuffer for intel HD graphics (to avoid issues of Bailtrail, CherryTrail with hwc that will not boot if RGBA_8888 not supported)

and drm_hwc for amdgpu 

In the Beta Testing thread linked folder, in about 6 hours

[youling 257]

choose debug to boot, upload data/log.txt, if you install your 4.19 kernel oreo on ext4 partition, you can easy to found log.txt.

you said it difficult for you do something on your tablet, i guess, your tablet has windows os on ntfs partition, you create ext4 partition only for install test Androidx86,
so your tablet has other one Linux os? your tablet need boot to Linux os to read write ext4 partition, read write Android files.

在 2018年12月27日星期四 UTC+8上午5:28:14，UniCav写道：

[UniCav]

On Wednesday, December 26, 2018 at 8:25:46 PM UTC-6, youling 257 wrote:

choose debug to boot, upload data/log.txt, if you install your 4.19 kernel oreo on ext4 partition, you can easy to found log.txt.

Debug also will not boot.

 

you said it difficult for you do something on your tablet, i guess, your tablet has windows os on ntfs partition, you create ext4 partition only for install test Androidx86,

There is no other OS. Only FAT32 EFI partition and EXT4 partition for Android-X86

I've been testing different builds of the OS to get the most functional one before tweaking and adding apps and starting to use it.

It is difficult to do anything directly on the tablet because it's so small to read and type on with external keyboard.

Doesn't have the power to compile anything on and will lose battery if I do anything that takes a lot of CPU and time.

[youling 257]

debug not boot, but Android data has log.txt, how to you copy log.txt from Android data? you need read write ext4 partition under other os.

do you need learn what is Android debug mode ? that's because data/log.txt.

在 2018年12月27日星期四 UTC+8下午11:48:20，UniCav写道：

[UniCav]

I had no other partition or OS on the tablet, and the tablet will not enable MTP mode so I can't access it from another device.

[UniCav]

Mauro,

I have your K4.20 image installed and running. Looking great so far.

Google allowed me to link to my account, Google Play is updating apps.

Battery and Wifi indicators are correct.

Will see how the Sleep/Wake button works shortly.

Thanks

[UniCav]

It did the usual thing and went to sleep forever. I'm going to dig through here and find the command again that properly suspends it to idle and try that.

If I can get that sorted out and figure out how to convert that to the power button function it'll be pretty much 100%

Oreo-X86, v48, Kernel 4.20, works amazingly smooth on the tablet and so far all apps so far seem to be working great

[UniCav]

give root permission to tasker, import this xml.

 

So I need to purchase and install Tasker? There's no way to do this without buying an app

echo 1 > /sys/class/gpio/gpio339/value
setprop sys.usb.config none
echo mem > /sys/power/state
echo 0 > /sys/class/gpio/gpio339/value

setprop sys.usb.config mtp

I understand I can save this to a file and copy into /sdcard with terminal emulator

So all that including the setprop commands for USB are needed for suspend?

I  have gone into settings and enabled USB Debugging

but when I tell it to set USB mode to MTP it crashes settings

The setting never changes.

I've tried giving the command "setprop.sys.usb.config mtp" in terminal

It doesn't give any error but the mode never changes from Charging

So I'm having to do all this using on-screen keyboard or external.

[youling 257]

open Android terminal

su

echo freeze > /sys/power/state

can you test it ?

在 2018年12月28日星期五 UTC+8上午3:00:09，UniCav写道：

[youling 257]

your /sdcard/suspend.txt only need echo freeze > /sys/power/state.

open Android terminal

su

echo "echo freeze > /sys/power/state" > /sdcard/suspend.txt

sh /sdcard/suspend.txt

在 2018年12月28日星期五 UTC+8上午3:00:09，UniCav写道：

[UniCav]

Yes it works.
Only issue I seem to have is sometimes touchscreen is not responsive or it immediately goes back to sleep then returns. But overall it works well and has not failed to wake.
If Tasker is what I need to reassign the key function it would be worth the purchase

[Povilas Staniulis]

Do you have a repo somewhere with your patches applied (including all
the graphics stuff) ?
I assume the kernel 4.20 is the same as this:
https://github.com/maurossi/linux

I plan to resume doing builds and I'm looking for a base to start with
(don't want to just use the plain OSDN source). I want to add a couple
extra features, namely native kernel drivers for exFAT and NTFS
filesystems (I have access to a Professional version of the Paragon
driver, which works on modern kernels) and perhaps do some additional
tinkering to make Android work better on my T100TA.

> --
> 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 email to android-x86...@googlegroups.com
> <mailto:android-x86...@googlegroups.com>.
> To post to this group, send email to andro...@googlegroups.com
> <mailto:andro...@googlegroups.com>.
> Visit this group at https://groups.google.com/group/android-x86.
> For more options, visit https://groups.google.com/d/optout.

[Povilas Staniulis]

If an app can do this, it should be possible to do without an app as well.
I wonder how apps do this. Is there a hook somewhere in Android's code
which enables apps to remap system button functions ?

[youling 257]

app is freedom for users.

if you like patch PowerManagerService.java

https://github.com/projectceladon/vendor-intel-utils/blob/master/android_p/google_diff/cel_apl/frameworks/base/0018-CELADON-Enabling-suspend-on-IVI-after-clicking.patch

https://github.com/projectceladon/vendor-intel-utils/blob/master/android_p/google_diff/cel_apl/frameworks/base/0019-Changing-usb-config-during-suspend-and-resume.patch

https://github.com/projectceladon/vendor-intel-utils/blob/master/android_p/google_diff/cel_apl/frameworks/base/0021-CELADON-Resuming-display-after-S3.patch

在 2018年12月28日星期五 UTC+8上午10:17:46，Povilas Staniulis写道：

[UniCav]

I'm going to look at the patches today or put Tasker on and get the XML imported and working.

Would like to do something with the top (home) button as well so Tasker would be the simplest solution from everything I've seen.

Testing last night at home, all the basic apps I need work well - Chrome, Youtube, Plex, IVMS and others.

So at this point once I get the Suspend/Resume working the only issues are:

• Retpoline warning - expected. Only shows up at boot. A little scary but eventually we'll get a fix ... maybe?
• Bluetooth - expected? I don't think anyone's ever worked out the firmware for it.
• MTP Mode - so far I haven't found anything that will actually make it switch to MTP. No errors on the commands in Terminal but no change in USB mode.
  It would certainly be nice to be able to plug in and access it through ADB but IIRC I can SSH into it? I just need to find the instructions again.
• OTG - works, and will hotplug active after boot - however it disables the touchscreen and I have to reboot to get it again. Possibly something I can change in settings?
• WIFI goes dead once in awhile. I think there's a glitch with ATH6Kl or Power Saving that needs to be tweaked to prevent this?
  

[youling 257]

study it, https://wiki.linaro.org/LMG/Kernel/AndroidConfigFSGadgets

[UniCav]

I looked at the patches, but I couldn't figure out where the PowerManagerService.java file is on the tablet.

find didn't locate it, Amaze File Explorer crashes whenever trying to search for it. Couldn't find online where it's stored on the system.

Installed Tasker, imported the XML. The suspend.txt file "echo freeze > /sys/power/state" exists.

Gave Tasker root perms. Seemed to be OK. First couple of power on/off tests worked.

Then the tablet went to sleep by itself and now it's unresponsive again, so something else has to be changed for the auto-sleep to use the "freeze" option?

[UniCav]

OK I'm certainly not a developer, but I'm not a complete idiot either when it comes to reading docs and sorting out things on Windows and Linux systems.

But I feel like a complete idiot after digging through Google looking for the PowerManagerService.java info.

I see patches, I see repositories, I see info on creating your own power profiles.

I see the file in the frameworks of the build directory.

But I do not see anything anywhere about being able to patch and replace the file in the system once it's on the tablet.

No "how to's"  or anything about how to work with the file

So I'm guessing that the only way to patch the file is to do it in the source and rebuild the ISO again?

Meanwhile I have also not found any way to fix the problem on the tablet using Tasker.

Sometimes it wakes and sometimes it doesn't.

I woke it up after about 10 minutes fine. Put it back to sleep and now it's dead again.

If it goes to sleep by itself it doesn't wake again.

[youling 257]

if tasker profiles work for you, did you see these from dmesg?

[29252.364237] smpboot: CPU 1 is now offline

[29252.367455] CPU3: update max cpu_capacity 1024

[29252.381359] smpboot: CPU 2 is now offline

[29252.385386] CPU0: update max cpu_capacity 1024

[29252.399243] smpboot: CPU 3 is now offline

[29253.225051] PM: suspend entry (s2idle)

[29253.225058] PM: Syncing filesystems ...

[29253.248536] done.

[29253.248799] Freezing user space processes ... (elapsed 0.003 seconds) done.

[29253.252547] OOM killer disabled.

[29258.442310] OOM killer enabled.

[29258.442320] Restarting tasks ... done.

[29258.452951] PM: suspend exit

[29258.737743] x86: Booting SMP configuration:

[29258.737751] smpboot: Booting Node 0 Processor 1 APIC 0x2

[29258.739084] smpboot: Booting Node 0 Processor 2 APIC 0x4

[29258.742517] smpboot: Booting Node 0 Processor 3 APIC 0x6

[29258.749541] CPU1: update max cpu_capacity 1024

在 2018年12月29日星期六 UTC+8上午4:05:28，UniCav写道：

[UniCav]

Yes here's current dmesg
https://drive.google.com/file/d/10ZcDT2LzDsu5Sm6tTLJr3Ty7IjjxByn6/view?usp=drivesdk

[youling 257]

Forget Android Lock Screen Close Screen, edit tile add tasker quick setting, not use power button to lock/close screen, only click tile to suspend.
https://groups.google.com/forum/#!searchin/android-x86/tasker%7Csort:date/android-x86/zg_brmXGDlU/Nm-1SSWAAgAJ

https://groups.google.com/forum/#!msg/android-x86/VsQ_bhNXqYo/TMYjunvgBAAJ

https://groups.google.com/forum/#!msg/android-x86/hn5YTTCPPM0/JqzZ15ihCwAJ

在 2018年12月29日星期六 UTC+8上午11:09:38，UniCav写道：

[youling 257]

my idea !

if you set Device security Screen lock PIN, edit "suspend to idle" task, run shell

killall com.android.systemui

sleep 5

echo freeze > /sys/power/state

在 2018年12月29日星期六 UTC+8上午11:35:05，youling 257写道：

[UniCav]

That still isn't working. I got the task shortcut working, tested once. Acted odd powering back up. Tried it again, left it for a couple hours, powered back up. It flips back and forth on/off at random. Sometimes goes right back to sleep and I have to hit the button again. Last time used the shortcut and came back a few hours later it won't come back on. So I need to find another solution.

[youling 257]

"killall com.android.systemui" command mean is killer systemui process and will restart systemui process, if set screen lock PIN, when restart systemui process, will locked screen but not close screen.

"sleep 5" command mean is waiting 5 second, then will auto do "echo freeze > sys/power/state" command to suspend close screen, you can leave 5 hours, let your tablet suspend 5 hours, press button to resume open screen.

you must not press power button to close screen, you only can press power button to let tablet resume from suspend.

if charging state change, tablet will auto resume from suspend.

在 2018年12月31日星期一 UTC+8上午10:39:22，UniCav写道：

[UniCav]

That's how I have it set up and I do understand what the script is doing. But it still seems something else is going on.

When I touch the shortcut and tasker runs the script, The desktop background changes to black and the icons stay the same, I'm guessing that's when the systemui is being killed.

Then the screen goes to sleep. When I come back and touch the power button to wake the tablet, it will sometimes do nothing. After several seconds if I hit the power button again, sometimes it will come on.

But many times the screen will come on then immediately go back off or will go to a blank back-lit screen. Sometimes I can hit the power button again and it will wake and let me start using the tablet.

Sometimes the screen will come on but the touchscreen is disabled and I can't do anything to it.

It's very random. I've tried testing it several different ways but it's never consistent.

Once I left it in sleep for about 3 hours. When I touched power I got the on/off reaction. Pressed power again and got it back up and used it awhile.

Then I left it off the rest of the day - about 6 or 8 hours. That night picked it up to use and it was non-responsive until I forced power off and back on then it was fine and battery was still over 80%

I put it to sleep again that night and next day it refused to respond again.

[UniCav]

This morning I powered the tablet back up, went back through the Tasker settings, everything looked correct. Set the PIN code again. Plugged my charger up and used the shortcut to put it to sleep. Left it all day. Came home and pressed the power button, came right up and asked for the PIN. Unlocked fine. Used it for a few minutes and put it back to sleep. Hour later picked it up and no response to any buttons. Forced off and back on and now it comes up to the lock screen but won't take any input from the touchscreen. Just ignores me and goes back to sleep. I really have to try something else