Related
Feb 22nd, 2011 Announcement: Now that Glemsom's auto-build system is back up, I will no longer be posting kernels to this thread. Kernel 1260 is the last that will be posted here. I encourage everyone to use the auto-build service from this point forward. Feel free to continue using this thread to discuss new kernels as they're released.
Feb 14th Update: 2nd kernel today fixes bugs introduced in 1st kernel.
Feb 10th Update: Glemsom's autobuild service appears to be back up, and he's finally gotten around to building the Jan 13th kernel (1253), which he's named 20110210_213850.
Original Dec. 12th, 2010 Announcement: I've received a request to post the latest kernel since the auto-build service was down between Dec. 5th or so and Feb. 10th. Thus I've linked the latest kernel and its modules file below for those who aren't into compiling their own kernels. This kernel was built from the source code available at the linux-msm repo on Gitorious that pretty much everyone (with a rhodium, topaz, raphael, diamond, blackstone & kovsky at least) including Glemsom's auto-build site has been using.
Additional Packages: As of Dec. 26th, compat-wireless & compcache 0.5.4 are fully integrated into these builds, just as they were in Glemsom's builds. 802.11e QoS support has also been enabled in compat-wireless. The modules file for kernel 1245 has been rebuilt to include the modules from these packages. Going forward, all future builds will include these as well. I am considering the possibility of creating a test kernel that includes the "swap free notify" patch and I'm interested in hearing back from anyone who has already looked into this possibility.
Latest Kernel
Glemsom's auto-build site
Older Kernels
Instructions - Note: These instructions are for the kernels I've posted here, not for Glemsom's kernels.
1) Download both the zImage and modules files below if you use wifi, otherwise you only need the zImage.
2) Rename the zImage file to simply: zImage
3) Drop that along with the modules file into your android boot directory (usually /sdcard/andboot).
4) Restart.
Commit# 1260 (includes 1259)
Commit Date: Mon Feb 21 18:19:45 2011
Commit ID: g11945d1
Pushed by: bzo (1259 on behalf of jbruneaux)
Commit Messages:
1260 - Merge branch 'htc-msm-2.6.27' of git://gitorious.org/linux-on-qualcomm-s-msm/linux-msm into htc-msm-2.6.27
1259 - adsp : fix the adsp drivers for camera (missing IOCTL, offsets, modules parameters, IDs fixup for QDSP messages).
F22's Notes: Some, but not all, of the code needed to make the camera work has been committed. This kernel is the same as Glemsom's 20110221_192424.
Kernel: zImage-2.6.27.46-01260-g11945d1
Modules: modules-2.6.27.46-01260-g11945d1.tar.gz
Commit# 1258 (includes 1256 & 1257)
Commit Date: Mon Feb 14 18:22:45 2011
Commit ID: g77a00c3
Pushed by: mweirauch (on behalf of bzo for 1256 & WisTilt2 for 1258)
Commit Messages:
1258 - Rhod Panel collapse and panel detect
1257 - Merge branch 'htc-msm-2.6.27' of gitorious.org:linux-on-qualcomm-s-msm/linux-msm into htc-msm-2.6.27
1256 - merge jbruneaux's amss clock implementation
F22's Notes: Bug Fix for 1255. Adds in missing definition and macro. Also known to fix the screen calibration bug that appeared in 1233 for rhodium only. This kernel is the same as Glemsom's 20110214_200559.
Kernel: zImage-2.6.27.46-01258-gf9e4956
Modules: modules-2.6.27.46-01258-gf9e4956.tar.gz
Commit# 1255 (includes 1254)
Commit Date: Mon Feb 14 06:32:17 2011
Commit ID: g77a00c3
Pushed by: ACL (on behalf of WisTilt2)
Commit Messages:
1255 - Rest of keyboard caps/fn leds
1254 - Rhod Caps/FN keyboard leds
F22's Notes: This kernel fails to compile. The committed code required a constant definition and a macro that weren't included.
Kernel: -
Modules: -
Commit# 1253 (includes 1251 & 1252)
Commit Date: Thu Jan 13 09:12:17 2011
Commit ID: gd53ddf6
Pushed by: ACL
Commit Messages:
1253 - Merge branch 'htc-msm-2.6.27' of [email protected]:linux-on-qualcomm-s-msm/linux-msm into htc-msm-2.6.27
1252 - Enable headphone amp on Rhod for 11pin headset. Still need 3.5mm jack to be enabled. Gpio values should be moved to rhod.h file.
1251 - Correct missing braces and add fix printk.
F22's Notes: snd.force_headset in startup.txt no longer necessary for rhodium due to 1251. Commit #1252 enables/disables extUSB headset jacks properly. 3.5mm jack on CMDA handsets isn't yet addressed. This kernel is the same as Glemsom's 20110210_213850.
Kernel: zImage-2.6.27.46-01253-gd53ddf6
Modules: modules-2.6.27.46-01253-gd53ddf6.tar.gz
Commit# 1250 (includes 1247-1249)
Commit Date: Fri Jan 7 21:02:39 2011
Commit ID: 8f8207c
Pushed by: mweirauch
Commit Messages:
1250 - Merge branch 'topa-panel' into htc-msm-2.6.27
1249 - topa: panel: drop the waits on init
1248 - topa: panel: some more register documentation + drop redundant one
1247 - topa: panel: convert init and deinit sequences to obsolete reg-offset shifting
Kernel: zImage-2.6.27.46-01250-g8f8207c
Modules: modules-2.6.27.46-01250-g8f8207c.tar.gz
Commit# 1246
Commit Date: Wed Jan 7 11:16:28 2011
Commit ID: 5ce0119
Pushed by: ACL (on behalf of WisTilt2)
Commit Messages:
1246 - Keyboard Backlight
F22's Notes: Keyboard Backlights for Rhodium from WisTilt2's test kernel.
Kernel: zImage-2.6.27.46-01246-g5ce0119
Modules: modules-2.6.27.46-01246-g5ce0119.tar.gz
Commit# 1245 (includes 1243 & 1244)
Commit Date: Thu Dec 23 01:58:42 2010
Commit ID: d9d77f2
Pushed by: ACL
Commit Messages:
1245 - Missed a define.
1244 - Merge branch 'htc-msm-2.6.27' of [email protected]:linux-on-qualcomm-s-msm/linux-msm into htc-msm-2.6.27
1243 - <Alex> Update clock code
F22's Notes: This kernel is thought to have introduced an issue that causes issues with some SD cards. See this post for more info.
Kernel: zImage-2.6.27.46-01245-d9d77f2
Modules: modules-2.6.27.46-01245-d9d77f2.tar.gz (Updated Dec. 26th - Now Includes Compat-Wireless & CompCache)
Commit# 1242 (includes 1241)
Commit Date: Tue Dec 22 22:56:16 2010
Commit ID: 51a9760
Pushed by: mweirauch (on behalf of bzo for 1241)
Commit Messages:
1242 - wakelock: fix build w/o CONFIG_WAKELOCK_STAT
1241 - fix for sleep mode of 1 when overclocking PLL2
F22's Notes: 1241 fixes bug that prevents handset from waking up when overclocking if pm.sleep_mode=1.
Kernel: zImage-2.6.27.46-01242-51a9760
Modules: modules-2.6.27.46-01242-51a9760.tar.gz
Commit# 1240
Commit Date: Tue Dec 21 22:43:26 2010
Commit ID: 137e247
Pushed by: mweirauch
Commit Messages:
1240 - topa: panel: move EXIT_SLEEP_MODE and document some registers
Kernel: zImage-2.6.27.46-01240-137e247
Modules: modules-2.6.27.46-01240-137e247.tar.gz
Commit# 1239
Commit Date: Tue Dec 21 06:32:11 2010
Commit ID: acf4596
Pushed by: ACL (on behalf of WisTilt2)
Commit Messages:
1239 - Speakerphone syntax correction
F22's Notes: Fixes hiss introduced in 1238.
Kernel: zImage-2.6.27.46-01238-gacf4596
Modules: modules-2.6.27.46-01238-gacf4596.tar.gz
Commit# 1238
Commit Date: Sat Dec 18 20:44:32 2010
Commit ID: 43f0a96
Pushed by: ACL (on behalf of WisTilt2)
Commit Messages:
1238 - Speakerphone mic fix.
F22's Notes: Speakerphone microphone now works in call.
Kernel: zImage-2.6.27.46-01239-g43f0a96
Modules: modules-2.6.27.46-01239-g43f0a96.tar.gz (Updated Dec. 19th: wlan.ko update)
Commit# 1237 (includes 1236)
Commit Date: Fri Dec 17 22:33:36 2010
Commit ID: 8c0ce38
Pushed by: mweirauch
Commit Messages:
1237 - mddi: novatek: drop vsync setup (by Alex) and update rhod+topa panels
1236 - mddi: simple: bring in line with [android,ca]-msm-2.6.35
Kernel: zImage-2.6.27.46-01237-g8c0ce38
Modules: modules-2.6.27.46-01237-g8c0ce38.tar.gz (Updated Dec. 19th: wlan.ko update)
Commit# 1235 (includes 1234)
Commit Date: Fri Dec 17 20:54:36 2010
Commit ID: 66a01da
Pushed by: mweirauch
Commit Messages:
1235 - mddi: epson: vsync setup to board files (taken from Alex)
1234 - raph+diam: fix toshiba mddi client name and setup vsync gpio (taken from Alex)
F22's Notes: 1234 fixes the Diam/Raph breakage caused by 1233.
Kernel: zImage-2.6.27.46-01235-g66a01da
Modules: modules-2.6.27.46-01235-g66a01da.tar.gz
Commit# 1233
Commit Date: Fri Dec 17 13:28:27 2010
Commit ID: 5b44660
Pushed by: ACL
Commit Message:
1233 - Update Framebuffer to .35
F22's Notes: Known to break Diam/Raph. Fixed in 1234. This kernel also introduced a second bug affecting all xdandroid handsets, causing them to appear to hang during boot if the ts-calibration file isn't already present. See the top of this page for more info.
Kernel: zImage-2.6.27.46-01233-g5b44660
Modules: modules-2.6.27.46-01233-g5b44660.tar.gz
Commit# 1232
Commit Date: Sat Dec 11 12:59:03 2010
Commit ID: b7033df
Pushed by: mweirauch
Commit Message:
1232 - mddi: novatek: improve topa+rhod 2D performance
Kernel: zImage-2.6.27.46-01232-gb7033df
Modules: modules-2.6.27.46-01232-gb7033df.tar.gz (Updates - Dec 14th: File Size Reduced to 806 kB - Dec 12th: TI Wifi Support for Non-Rhodiums Added.)
Pre-Dec. 11th 2010 Kernels
Glemsom's auto-build site
F22 said:
I've received a request to post the latest kernel since the auto-build service is currently down. Thus I've linked the latest kernel and its modules file below for those who aren't into compiling their own kernels. This kernel was built from the source code available at the linux-msm repo on Gitorious that pretty much everyone (with a rhodium, topaz, raphael or diamond at least) including Glemsom's auto-build site has been using.
Instructions
1) Download both the zImage and modules files below if you use wifi, otherwise you only need the zImage.
2) Rename the zImage file to simply: zImage
3) Drop that along with the modules file into your android boot directory (usually /sdcard/andboot).
4) Restart.
Latest Kernel
Commit Date: Sat Dec 11 12:59:03 2010
Commit# 1232
Commit ID: b7033df
Commit Message: mddi: novatek: improve topa+rhod 2D performance
Kernel: zImage-2.6.27.46-01232-gb7033df
Modules: modules-2.6.27.46-01232-gb7033df.tar.gz
Click to expand...
Click to collapse
thank you very much. you awesome. thanks
Thanks for the compile F22
Wifi can't enabled with this kernel on Topaz, says "Error"
It's only for Rhodium?
Hi, what do i use to unpack the .tar file. i tryed winrar but that does not keep de dir structure intackt. Im guessing that there is suposed to be a dir strukture, but i might be wrong. If not where should i put the files from the .tar file?
DmK75 said:
Wifi can't enabled with this kernel on Topaz, says "Error"
It's only for Rhodium?
Click to expand...
Click to collapse
it is normal.this current kernel don't have wlan.ko build into it.Missing wlan module to start wifi
Tar Files
flippy said:
Hi, what do i use to unpack the .tar file. i tryed winrar but that does not keep de dir structure intackt. Im guessing that there is suposed to be a dir strukture, but i might be wrong. If not where should i put the files from the .tar file?
Click to expand...
Click to collapse
There's no need to unpack anything. When you boot android, the init script will automatically untar the contents of modules-2.6.27.46-01232-gb7033df.tar.gz into /data/modules for you. Just leave it as is in your android boot directory (usually /sdcard/andboot).
wlan.ko
xUmaRix said:
DmK75 said:
Wifi can't enabled with this kernel on Topaz, says "Error"
It's only for Rhodium?
Click to expand...
Click to collapse
it is normal.this current kernel don't have wlan.ko build into it.Missing wlan module to start wifi
Click to expand...
Click to collapse
Hmm...as xUmaRix has noted, wlan.ko isn't part of a stock kernel build. It was apparently added in as an extra. I can look into adding it in too for the non-rhodium users. Check back later today.
Thanks so much for posting this! The last few kernels have not treated my Rhod400 so well...I hope this one does the trick!
Hi, just installed the new kernel, everything seems to work fine like before.
I can say the performance in 2d has increased.!!
thx to devs im looking forward..
Additional Modules Now Included for Topaz & Rhobuntu WiFi
F22 said:
Hmm...as xUmaRix has noted, wlan.ko isn't part of a stock kernel build. It was apparently added in as an extra. I can look into adding it in too for the non-rhodium users. Check back later today.
Click to expand...
Click to collapse
I've repackaged the modules file to include an extra 6 drivers that were present in Glemsom's modules file but weren't built in the stock linux-msm kernel compile. They include:
ramzswap.ko
rfkill_backport.ko
wl1251.ko
wl1251_sdio.ko
wlan.ko
xvmalloc.ko
Wifi for topaz should now work. I believe wifi on rhobuntu should work as well. I've updated the link in the 1st post of this thread so that it points to the new modules file.
For me this new kernel is unstable - I have Rhod400 but using in europe (gsm) , phone freezes few minutes after boot, sometimes when I use Opera. Few times there was no service error after boot (waited few minutes - nothing changed).
Just installed the first posted kernel. Working good here on my Rhod400. Boosted my FPS2D to 29 fps. Got 18.7 fps on Neocore. Wifi works fine too.
11.26 driver, 11.29 gralloc, 12.02 build.prop, no overclock.
Stability Issues
Raul^ said:
For me this new kernel is unstable - I have Rhod400 but using in europe (gsm) , phone freezes few minutes after boot, sometimes when I use Opera. Few times there was no service error after boot (waited few minutes - nothing changed).
Click to expand...
Click to collapse
XDANDROID itself isn't entirely stable. On occasion, and especially after you've changed something significant like the kernel, the rootfs, or even just your startup.txt, android's system_server process may go berserk and cause a host of issues including lack of service, timeout messages, sluggishness, etc... A reboot should fix it. If it doesn't, and you're overclocking, lower your overclock by 1 notch (19.2 MHz) and see if that helps.
I had a similar issue too the first time I booted the new kernel, and only the first time. After that it was fine. I've also experienced the same issue while making other tweaks to my handset that had nothing to do with the kernel. So it's a known problem that while extremely annoying, is probably unrelated to the new kernel aside from the fact that system_server can get cranky at times right after you change something.
F22 said:
I've received a request to post the latest kernel since the auto-build service is currently down. Thus I've linked the latest kernel and its modules file below for those who aren't into compiling their own kernels. This kernel was built from the source code available at the linux-msm repo on Gitorious that pretty much everyone (with a rhodium, topaz, raphael, diamond, blackstone & kovsky at least) including Glemsom's auto-build site has been using.
Click to expand...
Click to collapse
Thanks very much! us newbies appreciate it
Everything seems to be working (including wifi). I actually moved the old zimage and modules file to a temp directory that android doesn't even see, so I know I am running the new software, but when I look at the build information, it says:
Android version: 2.2
Kernel version: 2.6.27.46-01323-gb7033df
Build number: full_msm-eng 2.2 MASTER.eng
mjg.20100813.141358 test-keys
I guess I was expecting version 2.3...
Also partially working is the USB connection: When I connect the phone to my computer, I am notified (this is a first!)
Working well so far on my Sprint TP2...although it wasn't in the release notes I was hoping for the "voice call speakerphone mic" fix but mic still does not work...I think I heard that was fixed but just not commited yet?
there is a workaround - start a speakerphone conversation (i.e. voicemail) before launching haret
No Speakerphone Mic Fix
pdawg17 said:
Working well so far on my Sprint TP2...although it wasn't in the release notes I was hoping for the "voice call speakerphone mic" fix but mic still does not work...I think I heard that was fixed but just not commited yet?
Click to expand...
Click to collapse
I know WisTilt2 has speakerphone mic fixed because I helped him test it. I have no idea when it will get committed however. Unfortunately we haven't seen him online since Thanksgiving.
F22 said:
I know WisTilt2 has speakerphone mic fixed because I helped him test it. I have no idea when it will get committed however. Unfortunately we haven't seen him online since Thanksgiving.
Click to expand...
Click to collapse
Hopefully he'll be back at some point...I know about the workaround but sometimes I forget so it'd be nice to not have to think about it
Also partially working is the USB connection: When I connect the phone to my computer, I am notified (this is a first!)
Click to expand...
Click to collapse
Is this supposed to work now?
When I plug the phone into my windows 7 box, I now get an "H" drive. When I click on it, it asks me to insert a disk.
When I click the "Turn on USB storage" button in Android, nothing seems to happen (the spinner icon just goes around and around.)
Hi!
I am experiencing yaffs2 filesystem corruption on the data partition of my Desire.
The errors are always quite similar: Directory entries with empty names that point to the parent directory:
Code:
# ls -lia /data/data/com.android.providers.contacts/lib/databases
total 6
2006 drwxr-xr-x 1 10006 10006 2048 Apr 8 22:30
2006 drwxr-xr-x 1 10006 10006 2048 Apr 8 22:30 .
1737 drwxr-xr-x 1 1000 1000 2048 Apr 9 07:49 ..
It looks very similar to this report.
I've already tried several ROMs (most recently AceSMod007 and RevolutioN v3.0) but both use kernel 2.6.35.14-gingercakes-0.9-avs that maybe doesn't contain the fix.
Can anyone recommend a kernel that includes this patch?
Hello to everybody,
I'm trying to run open VPN in my rooted stock Galaxy Nexus, with ICS 4.0.3.
My goal is to create a bridged interface to get an ip directly from the same DHCP server that serve to all the pcs at my work network instead of assign a special pool by using a tap interface, but i'm not even able to bring up a tun interface to check if openvpn is working.
I keep getting some error when it's trying to bring up the tun0 interface; openvpn quit with value exit with "Linux ifconfig failed: external program fork failed".
If I try to bring up the tun0 from shell, I get "ifconfig: tun0: error fetching interface information: Device not found".
Then I check if the kernel is running with support for tunneling and i found that CONFIG_TUN is correctly set.
Did I miss something? Did I had to set other kernel modules? Also, somewhere in this forum is said that "tun is possible, tap no due to system limitations": someone knows what this limitations are and how can i pass them?
Also, is this a right place to ask or i'm posting to the wrong forum?
Thank you and sorry for my english
R.
Here is a part of logcat showing the errror (whit location masked at line 4)
Code:
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-mgmt( 1959): =============> 0 == 4 resetting dns, leaving dns alone
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-mgmt( 1959): SUCCESS: bytecount interval changed
D/OpenVPNDaemonEnabler( 1959): Received OpenVPN network state changed from Auth to Get Config
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:43 2012 SENT CONTROL [xx.xx.xx.it]: 'PUSH_REQUEST' (status=1)
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 PUSH: Received control message: 'PUSH_REPLY,route 192.168.11.0 255.255.255.0,dhcp-option DNS 10.8.0.1,dhcp-option WINS 192.168.11.2,route 10.8.0.0 255.255.255.0,topology net30,ping 10,ping-restart 120,ifconfig 10.8.0.74 10.8.0.73'
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 OPTIONS IMPORT: timers and/or timeouts modified
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 OPTIONS IMPORT: --ifconfig/up options modified
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 OPTIONS IMPORT: route options modified
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 OPTIONS IMPORT: --ip-win32 and/or --dhcp-option options modified
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 ROUTE default_gateway=2.195.48.1
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 TUN/TAP device tun0 opened
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 TUN/TAP TX queue length set to 100
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 MANAGEMENT: >STATE:1352902484,ASSIGN_IP,,10.8.0.74,
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 /system/xbin/busybox ifconfig tun0 10.8.0.74 pointopoint 10.8.0.73 mtu 1500
V/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-mgmt( 1959): onState(">STATE:1352902484,ASSIGN_IP,,10.8.0.74,")
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 MANAGEMENT: Client disconnected
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 Linux ifconfig failed: external program fork failed
D/OpenVPN-DaemonMonitor[/sdcard/openvpn/Nexus.ovpn]-daemon-stdout( 1959): Wed Nov 14 15:14:44 2012 Exiting
Hey Guys,
i wanted to check what to do, for trying to port a newer Lineage release to this device.
Any hints where to begin?
I luckily compiled the 14.1 branch yesterday.
But what to do next?
Regards for the hints
I've been investigating this recently as well. I started looking at LineageOS 16.0 branch and the Exynos5420 kernal/rom code on github ("https://github.com/exynos5420"), as I am running a build on the tablet from here:-
"https://download.exynos5420.com/LineageOS-14.1-Vanilla/chagalllte/"
BTW, I'm a late/newcomer to android rom development, and I may be off track/incorrect anyways.
I've attempted building branch lineage16.0 with Exynos5420 and currently throws this error :-
bionic/libc/include/bits/fortify/fcntl.h:63:26: error: use of undeclared identifier 'O_TMPFILE'
__clang_error_if(__open_modes_useful(flags), "'open' " __open_too_few_args_error)
The bionic libc wants a definition for O_TMPFILE , I think is to be resolved from the
kernel headers under the kernel device tree at:- kernel/samsung/exynos5420/include/asm-generic/fcntl.h
Of course there is no definition of 'O_TMPFILE' in this header file.
Looks like a later linux kernel version (or patches) to the Exynos5420 kernel code/device is required with headers defining O_TMPFILE.
All I can suggest for porting a newer lineageOs is, get started experimenting/building.
Find/pick up a later device kernel source code tree
For me there is alot to learn and few up-to-date guides
Good luck
bluess57 said:
I've been investigating this recently as well. I started looking at LineageOS 16.0 branch and the Exynos5420 kernal/rom code on github ("https://github.com/exynos5420"), as I am running a build on the tablet from here:-
"https://download.exynos5420.com/LineageOS-14.1-Vanilla/chagalllte/"
BTW, I'm a late/newcomer to android rom development, and I may be off track/incorrect anyways.
I've attempted building branch lineage16.0 with Exynos5420 and currently throws this error :-
bionic/libc/include/bits/fortify/fcntl.h:63:26: error: use of undeclared identifier 'O_TMPFILE'
__clang_error_if(__open_modes_useful(flags), "'open' " __open_too_few_args_error)
The bionic libc wants a definition for O_TMPFILE , I think is to be resolved from the
kernel headers under the kernel device tree at:- kernel/samsung/exynos5420/include/asm-generic/fcntl.h
Of course there is no definition of 'O_TMPFILE' in this header file.
Looks like a later linux kernel version (or patches) to the Exynos5420 kernel code/device is required with headers defining O_TMPFILE.
All I can suggest for porting a newer lineageOs is, get started experimenting/building.
Find/pick up a later device kernel source code tree
For me there is alot to learn and few up-to-date guides
Good luck
Click to expand...
Click to collapse
Add the following lines in the header file kernel/samsung/exynos5420/include/asm-generic/fcntl.h https://github.com/alexenferman/and...mmit/e97a38cefdb63305f85139c4409a597759388081 and you are good to go.
I tried compiling 16.0 for chagallwifi (T800) over the fall and ran into a number of compiler problems, but when I finally got it to compile, the resulting image wouldn't boot. A lot of those problems have now been fixed in gerrit.
I also tried 15.1 last week, but ran into a number of compile problems. Then I noticed a lot of activity on https://github.com/exynos5420 and https://review.exynos5420.com/ wrt to 16.0 so I tried again yesterday and today and I had to make 2 modifications for it to compile, but the result is an image that won't boot. The recovery image won't boot meaning somethings wrong with the boot/kernel stage.
Hopefully the team will make more modifications over the holidays and I will have a bootable 16.0 image?
retiredtab said:
I tried compiling 16.0 for chagallwifi (T800) over the fall and ran into a number of compiler problems, but when I finally got it to compile, the resulting image wouldn't boot. A lot of those problems have now been fixed in gerrit.
I also tried 15.1 last week, but ran into a number of compile problems. Then I noticed a lot of activity on https://github.com/exynos5420 and https://review.exynos5420.com/ wrt to 16.0 so I tried again yesterday and today and I had to make 2 modifications for it to compile, but the result is an image that won't boot. The recovery image won't boot meaning somethings wrong with the boot/kernel stage.
Hopefully the team will make more modifications over the holidays and I will have a bootable 16.0 image?
Click to expand...
Click to collapse
Unpack the recovery and show me the files inside. You might be missing the init files on the recovery, like I did with the Snapdragon Galaxy S3
alexenferman said:
Unpack the recovery and show me the files inside.
Click to expand...
Click to collapse
Code:
$ abootimg -x recovery.img
writing boot image config in bootimg.cfg
extracting kernel in zImage
extracting ramdisk in initrd.img
$ ls -al
total 16000
drwxrwxr-x 2 l l 4096 Dec 22 21:11 .
drwxrwxr-x 14 l l 4096 Dec 22 21:11 ..
-rw-rw-r-- 1 l l 167 Dec 22 21:11 bootimg.cfg
-rw-rw-r-- 1 l l 3336396 Dec 22 21:11 initrd.img
-rw-rw-r-- 1 l l 8183825 Dec 22 21:11 recovery.img
-rw-rw-r-- 1 l l 4842048 Dec 22 21:11 zImage
$ cat bootimg.cfg
bootsize = 0x7ce011
pagesize = 0x800
kerneladdr = 0x10008000
ramdiskaddr = 0x11000000
secondaddr = 0x10f00000
tagsaddr = 0x10000100
name =
cmdline = buildvariant=eng
retiredtab said:
Code:
$ abootimg -x recovery.img
writing boot image config in bootimg.cfg
extracting kernel in zImage
extracting ramdisk in initrd.img
$ ls -al
total 16000
drwxrwxr-x 2 l l 4096 Dec 22 21:11 .
drwxrwxr-x 14 l l 4096 Dec 22 21:11 ..
-rw-rw-r-- 1 l l 167 Dec 22 21:11 bootimg.cfg
-rw-rw-r-- 1 l l 3336396 Dec 22 21:11 initrd.img
-rw-rw-r-- 1 l l 8183825 Dec 22 21:11 recovery.img
-rw-rw-r-- 1 l l 4842048 Dec 22 21:11 zImage
$ cat bootimg.cfg
bootsize = 0x7ce011
pagesize = 0x800
kerneladdr = 0x10008000
ramdiskaddr = 0x11000000
secondaddr = 0x10f00000
tagsaddr = 0x10000100
name =
cmdline = buildvariant=eng
Click to expand...
Click to collapse
Sorry I meant Ramdisk of the recovery
$ ls -al 16/out/target/product/chagallwifi/ram*
-rw-rw-r-- 1 l l 1766995 Dec 22 17:49 ramdisk.img
-rw-rw-r-- 1 l l 9298176 Dec 22 17:49 ramdisk-recovery.cpio
-rw-rw-r-- 1 l l 3336396 Dec 22 17:49 ramdisk-recovery.img
/media/l/16/16/out/target/product/chagallwifi$
Code:
$ ./unpack_ramdisk ramdisk.img
6493 blocks
$ cd ramdisk/
$ ls -al
total 2248
drwxrwxr-x 16 l l 4096 Dec 22 22:01 .
drwxrwxr-x 3 l l 4096 Dec 22 22:01 ..
drwxr-xr-x 2 l l 4096 Dec 22 22:01 acct
lrwxrwxrwx 1 l l 11 Dec 22 22:01 bin -> /system/bin
lrwxrwxrwx 1 l l 50 Dec 22 22:01 bugreports -> /data/user_de/0/com.android.shell/files/bugreports
drwxrwx--- 2 l l 4096 Dec 22 22:01 cache
lrwxrwxrwx 1 l l 13 Dec 22 22:01 charger -> /sbin/charger
dr-xr-xr-x 2 l l 4096 Dec 22 22:01 config
lrwxrwxrwx 1 l l 17 Dec 22 22:01 d -> /sys/kernel/debug
drwxrwx--x 2 l l 4096 Dec 22 22:01 data
-rw------- 1 l l 1334 Dec 22 22:01 default.prop
drwxr-xr-x 2 l l 4096 Dec 22 22:01 dev
lrwxrwxrwx 1 l l 11 Dec 22 22:01 etc -> /system/etc
-rw-r----- 1 l l 2313 Dec 22 22:01 fstab.universal5420
-rwxr-x--- 1 l l 1637644 Dec 22 22:01 init
-rwxr-x--- 1 l l 1154 Dec 22 22:01 init.environ.rc
-rwxr-x--- 1 l l 29431 Dec 22 22:01 init.rc
-rwxr-x--- 1 l l 1524 Dec 22 22:01 init.samsung.rc
-rwxr-x--- 1 l l 2458 Dec 22 22:01 init.target.rc
-rwxr-x--- 1 l l 26408 Dec 22 22:01 init.universal5420.rc
-rwxr-x--- 1 l l 8107 Dec 22 22:01 init.universal5420.usb.rc
-rwxr-x--- 1 l l 2399 Dec 22 22:01 init.universal5420.wifi.rc
-rwxr-x--- 1 l l 7690 Dec 22 22:01 init.usb.configfs.rc
-rwxr-x--- 1 l l 5646 Dec 22 22:01 init.usb.rc
-rwxr-x--- 1 l l 511 Dec 22 22:01 init.zygote32.rc
drwxr-xr-x 2 l l 4096 Dec 22 22:01 mnt
drwxr-xr-x 2 l l 4096 Dec 22 22:01 odm
drwxr-xr-x 2 l l 4096 Dec 22 22:01 oem
-rw-r--r-- 1 l l 25053 Dec 22 22:01 plat_file_contexts
-rw-r--r-- 1 l l 7212 Dec 22 22:01 plat_hwservice_contexts
-rw-r--r-- 1 l l 7102 Dec 22 22:01 plat_property_contexts
-rw-r--r-- 1 l l 1551 Dec 22 22:01 plat_seapp_contexts
-rw-r--r-- 1 l l 14700 Dec 22 22:01 plat_service_contexts
drwxr-xr-x 2 l l 4096 Dec 22 22:01 proc
lrwxrwxrwx 1 l l 15 Dec 22 22:01 product -> /system/product
drwxr-xr-x 3 l l 4096 Dec 22 22:01 res
drwxr-x--- 2 l l 4096 Dec 22 22:01 sbin
lrwxrwxrwx 1 l l 21 Dec 22 22:01 sdcard -> /storage/self/primary
-rw-r--r-- 1 l l 384688 Dec 22 22:01 sepolicy
drwxr-x--x 2 l l 4096 Dec 22 22:01 storage
drwxr-xr-x 2 l l 4096 Dec 22 22:01 sys
drwxr-xr-x 2 l l 4096 Dec 22 22:01 system
-rw-r--r-- 1 l l 5359 Dec 22 22:01 ueventd.rc
-rw-r--r-- 1 l l 3373 Dec 22 22:01 ueventd.universal5420.rc
lrwxrwxrwx 1 l l 14 Dec 22 22:01 vendor -> /system/vendor
-rw-r--r-- 1 l l 7011 Dec 22 22:01 vendor_file_contexts
-rw-r--r-- 1 l l 1989 Dec 22 22:01 vendor_hwservice_contexts
-rw-r--r-- 1 l l 218 Dec 22 22:01 vendor_property_contexts
-rw-r--r-- 1 l l 0 Dec 22 22:01 vendor_seapp_contexts
-rw-r--r-- 1 l l 0 Dec 22 22:01 vendor_service_contexts
-rw-r--r-- 1 l l 65 Dec 22 22:01 vndservice_contexts
alexenferman said:
Sorry I meant Ramdisk of the recovery
Click to expand...
Click to collapse
Or do you mean the initrd.img inside the recovery.img file?
Here is the initrd.img inside recovery.img
Code:
$ xz -dc < initrd.img | cpio -idmv
acct
bin
bugreports
cache
config
d
data
default.prop
dev
etc
etc/mke2fs.conf
etc/mkshrc
etc/recovery.fstab
fstab.universal5420
init
init.rc
mnt
odm
odm/app
odm/bin
odm/firmware
odm/framework
odm/lib
odm/lib64
odm/overlay
odm/priv-app
oem
plat_file_contexts
plat_hwservice_contexts
plat_property_contexts
plat_seapp_contexts
plat_service_contexts
proc
product
prop.default
res
res/images
res/images/erasing_text.png
res/images/error_text.png
res/images/font.png
res/images/font_menu.png
res/images/ic_back.png
res/images/ic_back_sel.png
res/images/ic_factory_reset.png
res/images/ic_factory_reset_sel.png
res/images/ic_options_advanced.png
res/images/ic_options_advanced_sel.png
res/images/ic_reboot.png
res/images/ic_reboot_sel.png
res/images/ic_system_update.png
res/images/ic_system_update_sel.png
res/images/icon_error.png
res/images/installing_security_text.png
res/images/installing_text.png
res/images/logo_image.png
res/images/loop00000.png
res/images/loop00001.png
res/images/loop00002.png
res/images/loop00003.png
res/images/loop00004.png
res/images/loop00005.png
res/images/loop00006.png
res/images/loop00007.png
res/images/loop00008.png
res/images/loop00009.png
res/images/loop00010.png
res/images/loop00011.png
res/images/loop00012.png
res/images/loop00013.png
res/images/loop00014.png
res/images/loop00015.png
res/images/loop00016.png
res/images/loop00017.png
res/images/loop00018.png
res/images/loop00019.png
res/images/loop00020.png
res/images/loop00021.png
res/images/loop00022.png
res/images/loop00023.png
res/images/loop00024.png
res/images/loop00025.png
res/images/loop00026.png
res/images/loop00027.png
res/images/loop00028.png
res/images/loop00029.png
res/images/no_command_text.png
res/images/progress_empty.png
res/images/progress_fill.png
res/images/stage_empty.png
res/images/stage_fill.png
res/keys
sbin
sbin/acpi
sbin/adbd
sbin/awk
sbin/base64
sbin/basename
sbin/blockdev
sbin/bu
sbin/cal
sbin/cat
sbin/chcon
sbin/chgrp
sbin/chmod
sbin/chown
sbin/chroot
sbin/chrt
sbin/cksum
sbin/clear
sbin/cmp
sbin/comm
sbin/cp
sbin/cpio
sbin/cut
sbin/date
sbin/dd
sbin/df
sbin/diff
sbin/dirname
sbin/dmesg
sbin/dos2unix
sbin/du
sbin/e2fsck
sbin/e2fsdroid
sbin/e2fsdroid_static
sbin/echo
sbin/env
sbin/expand
sbin/expr
sbin/fallocate
sbin/false
sbin/file
sbin/find
sbin/flock
sbin/fmt
sbin/free
sbin/fsck.exfat
sbin/fsck.ext4
sbin/fsck.f2fs
sbin/fsck.ntfs
sbin/fsck_msdos
sbin/getenforce
sbin/getprop
sbin/grep
sbin/groups
sbin/gunzip
sbin/gzip
sbin/head
sbin/hostname
sbin/hwclock
sbin/id
sbin/ifconfig
sbin/inotifyd
sbin/insmod
sbin/install
sbin/ionice
sbin/iorenice
sbin/kill
sbin/killall
sbin/ln
sbin/load_policy
sbin/log
sbin/logname
sbin/losetup
sbin/ls
sbin/lsmod
sbin/lsof
sbin/lspci
sbin/lsusb
sbin/md5sum
sbin/microcom
sbin/mkdir
sbin/mke2fs
sbin/mke2fs_static
sbin/mkfifo
sbin/mkfs.exfat
sbin/mkfs.ext4
sbin/mkfs.f2fs
sbin/mkfs.ntfs
sbin/mknod
sbin/mkswap
sbin/mktemp
sbin/modinfo
sbin/modprobe
sbin/more
sbin/mount
sbin/mount.ntfs
sbin/mountpoint
sbin/mv
sbin/netstat
sbin/nice
sbin/nl
sbin/nohup
sbin/od
sbin/paste
sbin/patch
sbin/pgrep
sbin/pidof
sbin/pkill
sbin/pmap
sbin/printenv
sbin/printf
sbin/ps
sbin/pwd
sbin/readlink
sbin/realpath
sbin/reboot
sbin/recovery
sbin/renice
sbin/resize2fs
sbin/restorecon
sbin/rm
sbin/rmdir
sbin/rmmod
sbin/runcon
sbin/sed
sbin/sendevent
sbin/seq
sbin/setenforce
sbin/setprop
sbin/setsid
sbin/sgdisk
sbin/sh
sbin/sha1sum
sbin/sha224sum
sbin/sha256sum
sbin/sha384sum
sbin/sha512sum
sbin/sleep
sbin/sload.f2fs
sbin/sort
sbin/split
sbin/sswap
sbin/start
sbin/stat
sbin/stop
sbin/strings
sbin/stty
sbin/swapoff
sbin/swapon
sbin/sync
sbin/sysctl
sbin/tac
sbin/tail
sbin/tar
sbin/taskset
sbin/tee
sbin/time
sbin/timeout
sbin/top
sbin/touch
sbin/toybox_static
sbin/tr
sbin/true
sbin/truncate
sbin/tty
sbin/tune2fs
sbin/ueventd
sbin/ulimit
sbin/umount
sbin/uname
sbin/uniq
sbin/unix2dos
sbin/unzip
sbin/uptime
sbin/usleep
sbin/uudecode
sbin/uuencode
sbin/vmstat
sbin/watchdogd
sbin/wc
sbin/which
sbin/whoami
sbin/xargs
sbin/xxd
sbin/yes
sbin/zcat
sbin/zip
sdcard
sepolicy
storage
sys
system
tmp
ueventd.rc
ueventd.universal5420.rc
vendor_file_contexts
vendor_hwservice_contexts
vendor_property_contexts
vendor_seapp_contexts
vendor_service_contexts
vndservice_contexts
18161 blocks
I have a 16.0 SM-T800 build booting now. Now I have to go through and see what works and doesn't work. This will take a couple of days.
I will also document what patches I needed to make this work not only for myself, but for other future builders.
Please join the exynos5420 team to contribute to getting LineageOS 16 fully functional.
retiredtab said:
I have a 16.0 SM-T800 build booting now. Now I have to go through and see what works and doesn't work. This will take a couple of days.
I will also document what patches I needed to make this work not only for myself, but for other future builders.
Click to expand...
Click to collapse
What did you have to fix? I am building Los 16 for a Los 14 device too, kernel appears to be working, but it does not boot.
bluess57 said:
Please join the exynos5420 team to contribute to getting LineageOS 16 fully functional.
Click to expand...
Click to collapse
I will submit gerrit patches once the 16.0 code all settles down. There are camera + sensor + hal changes still to be merged. So far all bugs that I have found seem to be due to the pending or merge changes.
Some of the changes made to chagalllte weren't made to chagallwifi so it was relatively easy to copy them.
alexenferman said:
What did you have to fix? I am building Los 16 for a Los 14 device too, kernel appears to be working, but it does not boot.
Click to expand...
Click to collapse
What device are you building for? Are you building for a tab S platform?
If a device doesn't boot, I find adb logcat very helpful.
I also use this utility.
Install logcat-colorize on Ubuntu using the Snap Store | Snapcraft
Get the latest version of logcat-colorize for on Ubuntu - logcat-colorize
snapcraft.io
retiredtab said:
I will submit gerrit patches once the 16.0 code all settles down. There are camera + sensor + hal changes still to be merged. So far all bugs that I have found seem to be due to the pending or merge changes.
Some of the changes made to chagalllte weren't made to chagallwifi so it was relatively easy to copy them.
Click to expand...
Click to collapse
yeah I only have a chagalllte so some of the other devices get overlooked.
As to what's currently not functioning:-
camera video recording
sensors /sensor hal may not be fully functional
RIL
+ whatever else I can't recall atm
alexenferman said:
What did you have to fix? I am building Los 16 for a Los 14 device too, kernel appears to be working, but it does not boot.
Click to expand...
Click to collapse
by does not boot, do you mean it is stuck at the lineageos 3 rings startup animation?
bluess57 said:
As to what's currently not functioning:-
camera video recording
sensors /sensor hal may not be fully functional
RIL
+ whatever else I can't recall atm
Click to expand...
Click to collapse
What I found so far.
Doesn't work
1. home button doesn't wake up tablet. I think I know what is wrong and I'm compiling another new build, but it will take about 1 hour to complete.
2. screen rotation - which I think is due to pending merge hal/sensor changes
3. MTP, this could be due to my eng build for debugging purposes
4. camera - changes not merged. When I try a repopick -t Camera-Bringup, I get cherry-pick errors so I'm going to wait until it's all merged before I test camera.
Works
1. wifi (5 and 2.4)
2. bluetooth
3. brightness
4. external audio
5. audio through headphones
6. GPS - needs this yet to merged patch
https://review.exynos5420.com/c/exynos5420/android_device_samsung_chagalllte/+/7362
I'm not a git expert. Everytime I use it, I struggle with the commands to upload patches. But most of the problems with chagallwifi are because chagalllte commits aren't applied to it which is understandable if the developers have the LTE version.
it will be very good to switch to Los 16 SM-T800 please guys continue
bluess57 said:
yeah I only have a chagalllte so some of the other devices get overlooked.
Click to expand...
Click to collapse
I also have a SM-T700 klimtwifi which I can build and test for. However, I have more than one SM-T800 so I will build for T800 first and get all the bugs worked out and then build T700 later.
Hello,
I want to inspect backup files create by TWRP backup.
The file structure is this:
Code:
$ ll /mnt/HD403LJ/oneplus5t/TWRP/BACKUPS/b94a5888/2019-05-27--00-44-45
.rwxr-xr-x root root 1.0 MB Fri Apr 3 20:13:36 2020 bluetooth.emmc.win
.rwxr-xr-x root root 85 B Fri Apr 3 20:13:36 2020 bluetooth.emmc.win.sha2
.rwxr-xr-x root root 64 MB Fri Apr 3 20:13:42 2020 boot.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:13:42 2020 boot.emmc.win.sha2
.rwxr-xr-x root root 1.5 GB Fri Apr 3 20:16:50 2020 data.ext4.win000
.rwxr-xr-x root root 83 B Fri Apr 3 20:16:50 2020 data.ext4.win000.sha2
.rwxr-xr-x root root 957 MB Fri Apr 3 20:18:36 2020 data.ext4.win001
.rwxr-xr-x root root 83 B Fri Apr 3 20:18:36 2020 data.ext4.win001.sha2
.rwxr-xr-x root root 65 B Fri Apr 3 20:18:36 2020 data.info
.rwxr-xr-x root root 16 MB Fri Apr 3 20:18:38 2020 dsp.emmc.win
.rwxr-xr-x root root 79 B Fri Apr 3 20:18:38 2020 dsp.emmc.win.sha2
.rwxr-xr-x root root 2.0 MB Fri Apr 3 20:18:38 2020 efs1.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:38 2020 efs1.emmc.win.sha2
.rwxr-xr-x root root 2.0 MB Fri Apr 3 20:18:38 2020 efs2.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:38 2020 efs2.emmc.win.sha2
.rwxr-xr-x root root 4.0 KB Fri Apr 3 20:18:38 2020 efsc.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:38 2020 efsc.emmc.win.sha2
.rwxr-xr-x root root 2.0 MB Fri Apr 3 20:18:40 2020 efsg.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:40 2020 efsg.emmc.win.sha2
.rwxr-xr-x root root 110 MB Fri Apr 3 20:18:50 2020 modem.emmc.win
.rwxr-xr-x root root 81 B Fri Apr 3 20:18:50 2020 modem.emmc.win.sha2
.rwxr-xr-x root root 2.1 MB Fri Apr 3 20:18:50 2020 recovery.log
.rwxr-xr-x root root 1.5 GB Fri Apr 3 20:22:04 2020 system.ext4.win000
.rwxr-xr-x root root 85 B Fri Apr 3 20:22:04 2020 system.ext4.win000.sha2
.rwxr-xr-x root root 1.0 GB Fri Apr 3 20:23:50 2020 system.ext4.win001
.rwxr-xr-x root root 85 B Fri Apr 3 20:23:50 2020 system.ext4.win001.sha2
.rwxr-xr-x root root 65 B Fri Apr 3 20:23:50 2020 system.info
.rwxr-xr-x root root 3.0 GB Thu Mar 25 21:30:59 2021 system_image.emmc.win
.rwxr-xr-x root root 88 B Fri Apr 3 20:31:32 2020 system_image.emmc.win.sha2
.rwxr-xr-x root root 594 MB Fri Apr 3 20:32:36 2020 vendor.ext4.win
.rwxr-xr-x root root 82 B Fri Apr 3 20:32:36 2020 vendor.ext4.win.sha2
.rwxr-xr-x root root 64 B Fri Apr 3 20:32:38 2020 vendor.info
.rwxr-xr-x root root 1.0 GB Fri Apr 3 20:34:44 2020 vendor_image.emmc.win
.rwxr-xr-x root root 88 B Fri Apr 3 20:34:46 2020 vendor_image.emmc.win.sha2
Can you please advise how to mount specific files, e.g.
data.ext4.win000
data.ext4.win001
system.ext4.win000
system.ext4.win001
...
?
THX
A quick look at the file gives :
Code:
file system_image.emmc.win
system_image.emmc.win: Linux rev 1.0 ext2 filesystem data (mounted or unclean), UUID=da594c53-9beb-f85c-85c5-cedf76546f7a, volume name "system" (errors) (extents) (large files)
So did you try to mount -t ext2 system_image.emmc.win /mnt/custom ?
PS: I did try and it failed - but I used "compression" while making my backups with TWRP. If you made an uncompressed backup, it's worth a try.
DebianShaka said:
A quick look at the file gives :
Code:
file system_image.emmc.win
system_image.emmc.win: Linux rev 1.0 ext2 filesystem data (mounted or unclean), UUID=da594c53-9beb-f85c-85c5-cedf76546f7a, volume name "system" (errors) (extents) (large files)
So did you try to mount -t ext2 system_image.emmc.win /mnt/custom ?
PS: I did try and it failed - but I used "compression" while making my backups with TWRP. If you made an uncompressed backup, it's worth a try.
Click to expand...
Click to collapse
Didn't work for me either, but I do not remember whether I compressed that backup or not.
If I do
Code:
file system_image.emmc.win
I get
Code:
system_image.emmc.win: Linux rev 1.0 ext2 filesystem data (mounted or unclean), UUID=ca9d8888-a08a-51c7-b8ce-53df5d46f72d (extents) (large files) (huge files)
I might try to create an uncompressed backup later and check if it works.
I'll report back after I did it.
Edit: Nope, still failed.
"Couldn't be mounted in read only mode"
OK, I'm not telling you to execute raw github executables.. but.. while looking for a way to extract image files - using dd or adb pull in TWRP ? - I found a PDF claiming to explain the img format, e.g. "boot.img" is created using mkbootimage and contains a ramdisk : https://2net.co.uk/slides/android-boot-slides-2.0.pdf
Of course if adb pull /dev/mmblk-of-the-partition something.img gives me a fastboot-able something.img, I'll let you know (in this case it can be mounted, by finding the exact fs)
DebianShaka said:
OK, I'm not telling you to execute raw github executables.. but.. while looking for a way to extract image files - using dd or adb pull in TWRP ? - I found a PDF claiming to explain the img format, e.g. "boot.img" is created using mkbootimage and contains a ramdisk : https://2net.co.uk/slides/android-boot-slides-2.0.pdf
Of course if adb pull /dev/mmblk-of-the-partition something.img gives me a fastboot-able something.img, I'll let you know (in this case it can be mounted, by finding the exact fs)
Click to expand...
Click to collapse
I actually opened a boot.img and yes, it is created using mkbootimg and does contain a ramdisk.
There is actually a perl script here on xda which allows you to open boot.img files.
But that's not the same as mounting twrp backups is it?
EDIT: Heres what I get with an boot.img from LOS 17.1 (Android 10):
A folder named "boot.img-ramdisk", a file named "boot.img-kernel.gz", a file named "boot.img-ramdisk.cpio.gz".
EDIT 2: Heres the link: https://forum.xda-developers.com/t/howto-unpack-edit-and-repack-boot-images.443994/
You need to comment 3 lines out though on some boot.img, otherwise it won't work (these 3 lines can be found somewhere in the first 3 pages).
Hoever this is an old script. Maybe you can find another newer method / script if you're interessted.
1. Boot into TWRP
Code:
fastboot boot twrp-3.5.2_9-0-bullhead.img
2. Find the partition you want
Code:
adb shell
ls -alF /dev/block/platform/*/*/by-name/*
exit
In my case, it's /dev/block/mmcblk0p37 for boot
For alternatives, use google, for instance this was useful : https://stackoverflow.com/questions...artitions-of-an-android-device-from-the-shell
3. Try everything
Code:
adb shell
dd if=/dev/block/mmcblk0p37 of=/sdcard/boot_by_dd.img bs=4096
exit
adb pull /sdcard/boot_by_dd.img .
Provides /sdcard/boot_by_dd.img: 1 file pulled, 0 skipped. 8.4 MB/s (33554432 bytes in 3.813s)
Using TWRP, disable compression !, then make a backup of the boot partition
Code:
adb pull /sdcard/TWRP/BACKUPS/xxxserialxxx/2021-04-19--09-04-18_OPM7181205001/boot.emmc.win ./boot_by_twrp.img
Provides /sdcard/TWRP/BACKUPS/xxxserialxxx/2021-04-19--09-04-18_OPM7181205001/boot.emmc.win: 1 file pulled, 0 skipped. 8.0 MB/s (33554432 bytes in 4.017s)
Same size. sounds logical (boot is a 32Mb partition)
Code:
diff -q boot_by_dd.img boot_by_twrp.img
Result is Nothing. Yay !
Basically TWRP - without compression extracts a partition like dd (byte to byte).
4. Compare with flashed partition
Boot has been modified by magisk. So I tried it on vendor partition
Code:
adb shell ls -alF /dev/block/platform/*/*/by-name/* | grep vendor
In my case /dev/block/mmcblk0p39
Code:
adb shell
dd if=/dev/block/mmcblk0p39 of=/sdcard/vendor_by_dd.img bs=4096
exit
adb pull /sdcard/vendor_by_dd.img
Code:
diff -q vendor_by_dd.img StockNexus5X/vendor.img
Tells us Files vendor_by_dd.img and StockNexus5X/vendor.img differ
So basically fastboot format is not TWRP format - and TWRP format is dd format. I'll have a look into source.android.com at some point to look for 1. fastboot format and 2. partition fs format
EDIT : Still wrong. yaffs2 does not seem to be involved in twrp format, according to my recent tests. wrong conclusion below, sorry.
OK, mystery probably solved : see https://source.android.com/devices/bootloader/images
The yaffs2 file system format is (probably) used in "fastboot" format
EDIT: but not in twrp format.
c.monty said:
Hello,
I want to inspect backup files create by TWRP backup.
The file structure is this:
Code:
$ ll /mnt/HD403LJ/oneplus5t/TWRP/BACKUPS/b94a5888/2019-05-27--00-44-45
.rwxr-xr-x root root 1.0 MB Fri Apr 3 20:13:36 2020 bluetooth.emmc.win
.rwxr-xr-x root root 85 B Fri Apr 3 20:13:36 2020 bluetooth.emmc.win.sha2
.rwxr-xr-x root root 64 MB Fri Apr 3 20:13:42 2020 boot.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:13:42 2020 boot.emmc.win.sha2
.rwxr-xr-x root root 1.5 GB Fri Apr 3 20:16:50 2020 data.ext4.win000
.rwxr-xr-x root root 83 B Fri Apr 3 20:16:50 2020 data.ext4.win000.sha2
.rwxr-xr-x root root 957 MB Fri Apr 3 20:18:36 2020 data.ext4.win001
.rwxr-xr-x root root 83 B Fri Apr 3 20:18:36 2020 data.ext4.win001.sha2
.rwxr-xr-x root root 65 B Fri Apr 3 20:18:36 2020 data.info
.rwxr-xr-x root root 16 MB Fri Apr 3 20:18:38 2020 dsp.emmc.win
.rwxr-xr-x root root 79 B Fri Apr 3 20:18:38 2020 dsp.emmc.win.sha2
.rwxr-xr-x root root 2.0 MB Fri Apr 3 20:18:38 2020 efs1.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:38 2020 efs1.emmc.win.sha2
.rwxr-xr-x root root 2.0 MB Fri Apr 3 20:18:38 2020 efs2.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:38 2020 efs2.emmc.win.sha2
.rwxr-xr-x root root 4.0 KB Fri Apr 3 20:18:38 2020 efsc.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:38 2020 efsc.emmc.win.sha2
.rwxr-xr-x root root 2.0 MB Fri Apr 3 20:18:40 2020 efsg.emmc.win
.rwxr-xr-x root root 80 B Fri Apr 3 20:18:40 2020 efsg.emmc.win.sha2
.rwxr-xr-x root root 110 MB Fri Apr 3 20:18:50 2020 modem.emmc.win
.rwxr-xr-x root root 81 B Fri Apr 3 20:18:50 2020 modem.emmc.win.sha2
.rwxr-xr-x root root 2.1 MB Fri Apr 3 20:18:50 2020 recovery.log
.rwxr-xr-x root root 1.5 GB Fri Apr 3 20:22:04 2020 system.ext4.win000
.rwxr-xr-x root root 85 B Fri Apr 3 20:22:04 2020 system.ext4.win000.sha2
.rwxr-xr-x root root 1.0 GB Fri Apr 3 20:23:50 2020 system.ext4.win001
.rwxr-xr-x root root 85 B Fri Apr 3 20:23:50 2020 system.ext4.win001.sha2
.rwxr-xr-x root root 65 B Fri Apr 3 20:23:50 2020 system.info
.rwxr-xr-x root root 3.0 GB Thu Mar 25 21:30:59 2021 system_image.emmc.win
.rwxr-xr-x root root 88 B Fri Apr 3 20:31:32 2020 system_image.emmc.win.sha2
.rwxr-xr-x root root 594 MB Fri Apr 3 20:32:36 2020 vendor.ext4.win
.rwxr-xr-x root root 82 B Fri Apr 3 20:32:36 2020 vendor.ext4.win.sha2
.rwxr-xr-x root root 64 B Fri Apr 3 20:32:38 2020 vendor.info
.rwxr-xr-x root root 1.0 GB Fri Apr 3 20:34:44 2020 vendor_image.emmc.win
.rwxr-xr-x root root 88 B Fri Apr 3 20:34:46 2020 vendor_image.emmc.win.sha2
Can you please advise how to mount specific files, e.g.
data.ext4.win000
data.ext4.win001
system.ext4.win000
system.ext4.win001
...
?
THX
Click to expand...
Click to collapse
Info:
· Were talking about backups without decompress and password applied via twrp.
· We use linux for those operations and the terminal
What we got so far:
System Image partition:
Code:
sudo mount -o loop ./system_image.emmc.win ./mount_point/
Userdata / Data (excl. storage) partition:
Code:
tar -xvf data.ext4.win000
Gives some errors though claiming
tar: ignoring unknown keyword „TWRP.security.fscrypt“ for extended head section
Click to expand...
Click to collapse
(Thats translated. Hope it makes sense, though.)
However I viewed that data and it seems to be all fine.
firmware partition:
Code:
tar -xvf firmware.vfat.win
Works perfectly.
boot partition:
See this: https://forum.xda-developers.com/t/howto-unpack-edit-and-repack-boot-images.443994/
Couldn't figure it out yet for the following:
boot Logo partition:
It's data, but I cannot read it.
EFS partition:
Well, we would have to mount a tim partition...
efs1.emmc.win: TIM image, 24-Bit, Pixel at (4880,50247) Size=65024x31
Click to expand...
Click to collapse
"TIM image" probably a false positive. "file" uses heuristics
DebianShaka said:
"TIM image" probably a false positive. "file" uses heuristics
Click to expand...
Click to collapse
Yes maybe. Maybe we can find something else to determine its type.