Develop Bugs

[Q] HTC Desire HD, partition table broken in internal memory

Hello
I've just recently got a HTC Desire HD. I've rooted it with the help of aahk-11092012.
I've installed CM 7 first and it all worked fine.
A bit later I installed JellyTime ( http://forum.xda-developers.com/showthread.php?t=1778202 ).
This worked fine for about 2 days. Than, suddenly, the phone crashed. The screen went black and I the phone would not accept any key presses anymore. So I removed the battery and started the phone again. It hung on the htc-logo.
I was still able to get into recovery but it complained that it could not access the cache partition.
Upon further investigation i found out, that the partition table of the internal memory is broken. I could not mount or format any of the partitions. Factory reset just complained about that it could not access the partitions.
So i switched off the phone and started adb.
dmesg shows this
Code:
<6>[ 7.630279] rs30000048:00040000 rs30000048:00040000: setting system clock to 2012-11-23 20:51:58 UTC (1353703918)
<4>[ 7.630645] Warning: unable to open an initial console.
<6>[ 7.630828] Freeing init memory: 156K
<6>[ 7.739837] mmc0: new high speed MMC card at address 0001
<6>[ 7.740966] mmcblk0: mmc0:0001 M4G2DE 2.10 GiB
<6>[ 7.741851] mmcblk0: p1 p2 p3 p4 < p5 p6 p7 p8 p9 p10 p11 p12 p13 p14 p15 p16 p17 p18 p19 p20 p21 p22 p23 p24 p25 p26 p27 p28 >
<4>[ 7.767913] mmcblk0: partition table partially beyond EOD, truncated
<4>[ 7.769744] mmcblk0: p28 size 42495 extends beyond EOD, truncated
<3>[ 7.816436] init: Unable to open persistent property directory /data/property errno: 2
And if I try to use parted I only get errors:
Code:
~ # parted /dev/block/mmcblk0
GNU Parted 1.8.8.1.179-aef3
Using /dev/block/mmcblk0
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted) print list
print list
Error: Can't have a partition outside the disk!
(parted)
The same error is show if I try to remove partitions.
When I try to write a new disklabel it just hangs there forever. No error message (also not in dmesg) or anything like this.
How can I fix the partition table of the internal memory and what do I have to do to get it to work again?
I guess that I will have to create a new partition table (and most probably destroy all the data on the internal memory in the process).
What will I have to reinstall in what order afterwards?
hboot?
cwm?
and how do I have to do this from the adb? Or is there a possibility do do this out of recovery?

I've tried several things in the meantime.
Ind the longer I try, the more I think that I have a bricked phone now since I seem to be unable to write anything to the emmc
I've tried to get back to the original firmware. So I copied the File PD98IMG.zip the the sd card, got into hboot and started the installation in hboot.
I could start it and it told me "Updating in progress, do not power off"
But there was no progress bar and after 10 minutes I pulled the battery. The phone is still in the same state as before so I think it did not write anything to the internal memory.
With fastboot, it seems to be the same. I can start fastboot on the phone aqnd it changes to "fastboot usb" when i plug the phone in. But every attempt to flash something to the phone just let's fastboot on the computer stop and hang (no message is shown, it's just sitting there and apparently doing nothing).
My computer system is debian linux btw.
I'm running out of ideas... please help... and it it's just the confirmation, that I have a bricked phone (that is long out of the warranty period anyway)

try with fastboot erase command. do google it a bit before you try

You have a full brick mate the EMMC is fried not amount of commands will fix it.
If you try fastboot getvar all then the try fastboot oem rebootRUU and the phone locks up its a new EMMC chip and jtag session to get it back online.

Google helps sometimes, you can also wipe out sectors on the emmc, because the partition table is corrupt.
http://archive09.linux.com/forums/topic/4033
Sent from my HTC Desire HD using Tapatalk 2

I have tried if he has the Dodgy EMMC doesn't matter what you try its fried and will not work. I have tried everything myself on a Bricked DHD with the Dodgy EMMC chip.

mmcblk0 I/O error – possible hardware problem?

Hello, guys!
This morning I found my phone with black screen and not responding to power button. I pulled the battery out and tried to boot. The phone only showed the "Google" text on the black screen for several minutes (five, may be ten minutes) and did not respond to any buttons. At this point I pulled the battery for the second time and tried the recovery mode. The stock recovery booted after about a minute of "Google" screen. In stock recovery the only option I tried first time was "Wipe cache". It showed a static image of "work-in-progress" with static progress bar for several seconds and then returned to main recovery screen. I tried the "reboot" and received the black screen with no buttons working.
Pulled the battery and tried normal boot, which stuck on black "Google" stage.
Ok, I connected the phone to my laptop, and booted TWRP via fastboot, it took a little longer to boot, but at least I managed to "adb-shell" it and grab dmesg output. This is the relevant part (containing "mmc" word):
Code:
<6>[ 2.828399] mmc0: new high speed MMC card at address 0001
...
<6>[ 2.829101] mmcblk0: mmc0:0001 VYL00M 14.6 GiB
<6>[ 2.829345] mmcblk0boot0: mmc0:0001 VYL00M partition 1 512 KiB
<6>[ 2.829681] mmcblk0boot1: mmc0:0001 VYL00M partition 2 512 KiB
<6>[ 2.835021] Reprogramming LPDDR2 timings to 400000000 Hz
<3>[ 2.836090] Power Management for TI OMAP4.
<6>[ 2.836334] OMAP4 PM: Static dependency added between MPUSS and DUCATI <-> L4_PER/CFG and L3_1.
<6>[ 2.836517] mmcblk0: p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 p13
<4>[ 2.837432] sr_init: No PMIC hook to init smartreflex
<6>[ 2.837799] smartreflex smartreflex.0: omap_sr_probe: SmartReflex driver initialized
<6>[ 2.838073] smartreflex smartreflex.1: omap_sr_probe: SmartReflex driver initialized
<6>[ 2.838317] smartreflex smartreflex.2: omap_sr_probe: SmartReflex driver initialized
<6>[ 2.838592] sr_class1p5_driver_init: OMAP4460: add 10mV margin for class 1.5
<6>[ 2.838928] SmartReflex class 1.5 driver: initialized (86400000ms)
<6>[ 2.839569] mmcblk0boot1: unknown partition table
<6>[ 2.840728] [MODEM_IF] PA EVENT : reset =0, pa=1
<6>[ 2.840972] mmcblk0boot0: unknown partition table
...
<4>[ 3.851135] mmc1: queuing unknown CIS tuple 0x80 (7 bytes)
<4>[ 3.899353] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
<6>[ 3.915679] mmc1: new high speed SDIO card at address 0001
<4>[ 3.917633] F1 signature read @0x18000000=0x16844330
<4>[ 3.927154] DHD: dongle ram size is set to 294912(orig 294912)
...
<3>[ 7.677093] mmcblk0: timed out sending r/w cmd command, card status 0xe00
<3>[ 7.677276] mmcblk0: not retrying timeout
<3>[ 7.677581] end_request: I/O error, dev mmcblk0, sector 1495040
<3>[ 7.677764] Buffer I/O error on device mmcblk0p11, logical block 0
<4>[ 7.678100] lost page write due to I/O error on mmcblk0p11
<6>[ 7.678802] EXT4-fs (mmcblk0p11): recovery complete
<6>[ 7.679138] EXT4-fs (mmcblk0p11): mounted filesystem with ordered data mode. Opts: (null)
<3>[ 8.737213] mmcblk0: timed out sending r/w cmd command, card status 0x400e00
<3>[ 8.737304] mmcblk0: command error, retrying timeout
<3>[ 8.737426] mmcblk0: timed out sending r/w cmd command, card status 0x400e00
<3>[ 8.737548] mmcblk0: command error, retrying timeout
<3>[ 8.737701] mmcblk0: timed out sending r/w cmd command, card status 0x400e00
<3>[ 8.737762] mmcblk0: command error, retrying timeout
<3>[ 8.737915] mmcblk0: timed out sending r/w cmd command, card status 0x400e00
...
and a lot of I/O errors with mmcblk0 block device
Reading from /dev/block/mmcblk0 constantly fails, even the first sectors (dd if=/dev/block/mmcblk0 of=/dev/null).
I pulled the battery out, left the phone and rushed to the work.
With my little knowledge it looks like broken sdcard itself or sdcard controller. Does it make any sense? Is it possible to repair it easily? I realize I cannot simply replace sdcard, as it most likely soldered in the board, but how hard is it to replace if I have the equipment and spare parts?
Another question: as I can boot TWRP recovery via fastboot, is it theoretically possible to find and boot a custom kernel, which does not require internal storage at all, like some stripped kernel which works only in RAM, but still provides the basic functions like phone calling, wifi browsing and something like this?
Any help would be greatly appreciated!

Try restoring your phone to stock using ODIN
http://forum.xda-developers.com/showthread.php?t=1827685
I think the PIT file should repair your partition table if it's just a software problem

Thanks for the tip. I guess I will try odin methods only as as a last resort since it may void my warranty. I am still positive about sending my unit to the service center in January.
At the moment fastboot methods do not work if they try to actually write something. fastboot boot works pretty good.

Looks like a bad partition, ODIN back to factory wont void your warranty. All it does is reset your phone to the stock image, that's probably what they'll do when you send it in for warranty repair. Save yourself the time and trouble.

Bad blocks is something software won't fix. This is a hardware problem with the flash memory and it needs replaced.

I had something like this happen when I was formatting my /data partition in TWRP. I did a factory reset and it was like "wiping /data without wiping /data/media" and just to be on the safe side I wiped /data and it froze and when I got control back it said that it couldn't format it or anything so I ended up restoring the factory image and that fixed everything.

varet tviengt
Once again, just to be on the safe side, I will not try ODIN factory reset just yet. As it may be the hardware problem and I may end up with phone-triangle-computer state, in which case my warranty might be rejected like I messed something up myself blah-blah-blah.
The problem with the warranty is that I should go to my home, and I am now several thousands kilometers away from it on a vacation. So I will just wait for it.

I'm having the exact same problems for the last few weeks. Did you fix it? If so, how?

Please read forum rules before posting
Questions and help issues go in Q&A
Thread moved
Thank you for your cooperation
Friendly Neighborhood Moderator

I have a GNexus 4.2.2 stock and I have reboot problem from the last week.
in the last_kmsg(attached) I read many references to mmc problems, for example
Code:
[ 5213.429595] mmcblk0: timed out sending r/w cmd command, card status 0x900
[ 5213.429779] mmcblk0: not retrying timeout
[ 5213.429901] end_request: I/O error, dev mmcblk0, sector 2450976
[ 5213.430084] end_request: I/O error, dev mmcblk0, sector 2450984
[ 5213.430175] end_request: I/O error, dev mmcblk0, sector 2450992
[ 5213.430328] end_request: I/O error, dev mmcblk0, sector 2451000
[ 5213.430450] end_request: I/O error, dev mmcblk0, sector 2451008
[ 5213.430603] end_request: I/O error, dev mmcblk0, sector 2451016
[ 5213.430694] end_request: I/O error, dev mmcblk0, sector 2451024
[ 5213.430847] end_request: I/O error, dev mmcblk0, sector 2451032
[ 5213.431091] Aborting journal on device mmcblk0p12-8.
I have already restore the Factory Images from Google developer, but I don't solve.
Is damage the memory or is a software problem?

up

Any news here? Since a few days I'm experiencing the same problem .

I have sane issue ! Any info for us ?
Inviato dal mio Galaxy Nexus usando Tapatalk 4

I am experiencing this problem too. Often apps crash when SQLite tries to write something to disk: Disk I/O error (code 1034).
I'm also getting random reboots and corrupted files.
This happens both with CM11-M2 and stock 4.3 ROM, so I don't think it's a software problem.
Do you think the internal storage is dying? Is there something I can do? Thanks!

[SOLVED] I/O Error

Hi all,
I have a Galaxy Nexus with CyanogenMod 10.1 RC2.
Yesterday the battery discharged and the phone turned off.
When at home, I charged and booted it. Short after the phone turned on, it rebooted.
It continues to reboot a few seconds after I turn it on. I wiped everything, and I reinstalled Cyanogenmod, without any effect.
I tried to read the kmesg via ADB. It seems to be a filesysem corruption: pastebin.com/6J2NR4ne
The beginning of the errors is this:
Code:
<6>[ 56.475524] sr_class1p5_calib_work: iva: Calibration complete: Voltage:Nominal=950000,Calib=860920,margin=13000
<4>[ 66.156127] omap_hsi omap_hsi.0: Missed previous CAWAKE falling edge...
<3>[ 66.157775] mmcblk0: timed out sending r/w cmd command, card status 0x900
<3>[ 66.157958] mmcblk0: not retrying timeout
<3>[ 66.158111] end_request: I/O error, dev mmcblk0, sector 2398632
<3>[ 66.158233] end_request: I/O error, dev mmcblk0, sector 2398640
<3>[ 66.158416] end_request: I/O error, dev mmcblk0, sector 2398648
<3>[ 66.158630] end_request: I/O error, dev mmcblk0, sector 2398656
<3>[ 66.158782] Aborting journal on device mmcblk0p12-8.
<0>[ 66.164672] journal commit I/O error
<2>[ 66.170043] EXT4-fs error (device mmcblk0p12): ext4_journal_start_sb:296: Detected aborted journal
<2>[ 66.170715] EXT4-fs (mmcblk0p12): Remounting filesystem read-only
Can anyone help me?

I tried to run e2fsck on /dev/block/mmcblk0p12 while in recovery. It did not solve the problem.
Code:
~ # e2fsck /dev/block/mmcblk0p12
e2fsck 1.41.11 (14-Mar-2010)
/dev/block/mmcblk0p12: recovering journal
/dev/block/mmcblk0p12 contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
/dev/block/mmcblk0p12: ***** FILE SYSTEM WAS MODIFIED *****
/dev/block/mmcblk0p12: 25726/887696 files (4.9% non-contiguous), 1129492/3549691 blocks

SOLVED.
Wiping the /data partition only deleted the files. I flashed userdata.img from the stock image (downloaded from google code), it rewrote /data and it worked

Please report the thread for closing.
Sent from my Nexus

Thread closed per OP's request.

GNex NAND corruption?

A few days ago, I noticed my phone started displaying the SIM unlock menu in the morning. Later, I saw it reboot all of a sudden, and so I left it to observe for a bit more. The random reboots continued, so I rebackuped everything in TiBu, did a factory reset (wipe /data without /sdcard, /cache and dalvik cache) and updated the ROM I was using.
The reboots were still there, so I copied out all the data, suspecting corruption. My phone rebooted halfway through copying the files so I did it again in recovery. Then, I ran fastboot -w and rebooted into TWRP. It asked for an encryption password, and after doing a factory reset and rebooting it still did. Booting normally and then rebooting into TWRP fixed that.
After that, I went back ingo TWRP and used adb to push the sdcard files back in. Halfway through, adb gave me an error:
Code:
failed to copy './DCIM/Camera/IMG_20130121_174846.jpg' to '/sdcard/DCIM/Camera/IMG_20130121_174846.jpg': Read-only file system
I tried again and the same thing happened. I decided to look in dmesg, and lo and behold:
Code:
......
<3>[ 1232.192260] end_request: I/O error, dev mmcblk0, sector 5539768
<3>[ 1232.192443] end_request: I/O error, dev mmcblk0, sector 5539776
<3>[ 1232.192749] end_request: I/O error, dev mmcblk0, sector 5539784
<3>[ 1232.192932] end_request: I/O error, dev mmcblk0, sector 5539792
<3>[ 1232.193267] end_request: I/O error, dev mmcblk0, sector 5539800
<3>[ 1232.193420] end_request: I/O error, dev mmcblk0, sector 5539808
<3>[ 1232.193756] end_request: I/O error, dev mmcblk0, sector 5539816
<3>[ 1232.193939] end_request: I/O error, dev mmcblk0, sector 5539824
<3>[ 1232.194244] end_request: I/O error, dev mmcblk0, sector 5539832
<3>[ 1232.194580] Buffer I/O error on device mmcblk0p12, logical block 394880
<3>[ 1232.194793] Buffer I/O error on device mmcblk0p12, logical block 394881
<3>[ 1232.195129] Buffer I/O error on device mmcblk0p12, logical block 394882
<3>[ 1232.195312] Buffer I/O error on device mmcblk0p12, logical block 394883
<3>[ 1232.195648] Buffer I/O error on device mmcblk0p12, logical block 394884
<3>[ 1232.195861] Buffer I/O error on device mmcblk0p12, logical block 394885
<3>[ 1232.196197] Buffer I/O error on device mmcblk0p12, logical block 394886
<3>[ 1232.196533] Buffer I/O error on device mmcblk0p12, logical block 394887
<3>[ 1232.196716] Buffer I/O error on device mmcblk0p12, logical block 394888
<3>[ 1232.197052] Buffer I/O error on device mmcblk0p12, logical block 394889
......
<3>[ 1275.397430] Buffer I/O error on device mmcblk0p12, logical block 2600935
<3>[ 1275.397491] Buffer I/O error on device mmcblk0p12, logical block 2600936
<4>[ 1275.397613] EXT4-fs warning (device mmcblk0p12): ext4_end_bio:258: I/O error writing to inode 643562 (offset 1572864 size 524288 starting block 2898409)
<4>[ 1275.479003] JBD2: Detected IO errors while flushing file data on mmcblk0p12-8
......
<3>[ 1529.202911] mmcblk0: timed out sending r/w cmd command, card status 0x900
<3>[ 1529.203247] mmcblk0: not retrying timeout
<3>[ 1529.203430] end_request: I/O error, dev mmcblk0, sector 2425376
<3>[ 1529.203765] end_request: I/O error, dev mmcblk0, sector 2425384
<3>[ 1529.203979] end_request: I/O error, dev mmcblk0, sector 2425392
<3>[ 1529.204284] end_request: I/O error, dev mmcblk0, sector 2425400
<3>[ 1529.204620] end_request: I/O error, dev mmcblk0, sector 2425408
<3>[ 1529.204803] end_request: I/O error, dev mmcblk0, sector 2425416
<3>[ 1529.205139] end_request: I/O error, dev mmcblk0, sector 2425424
<3>[ 1529.205322] end_request: I/O error, dev mmcblk0, sector 2425432
<3>[ 1529.205627] end_request: I/O error, dev mmcblk0, sector 2425440
<3>[ 1529.205993] Aborting journal on device mmcblk0p12-8.
<2>[ 1529.206420] EXT4-fs error (device mmcblk0p12) in ext4_reserve_inode_write:5654: Journal has aborted
<2>[ 1529.216522] EXT4-fs (mmcblk0p12): Remounting filesystem read-only
What should I do now?

Seems to me you're posting in the wrong section; and that your internal emmc has fried. Try to flash stock images, pre-erasing and pre-formatting system, userdata and boot partitions.
a manta wrote this.

Happened to me once... Couldn't revive it whatsoever... Sold it eventually.
Sent from my awesome Xperia Z

Boot into fastboot mode, then try to format the partitions (system, userdata, cache, boot) and see if it helps. Otherwise you've deep fried your NAND.

I had my gnex for a while never used it a lot, cause I got a nexus 4 after that and mine reboots several times a day even not overclocked, and I only ever pick it up to listen to music while my 4 is charging
Sent from my Nexus 4 @1.944 GHz on Stock 4.2.2

IRX120 said:
I had my gnex for a while never used it a lot, cause I got a nexus 4 after that and mine reboots several times a day even not overclocked, and I only ever pick it up to listen to music while my 4 is charging
Sent from my Nexus 4 @1.944 GHz on Stock 4.2.2
Click to expand...
Click to collapse
...
Cool story bro; you sure you're not on the wrong thread then?
a maguro wrote this.

beekay201 said:
...
Cool story bro; you sure you're not on the wrong thread then?
a maguro wrote this.
Click to expand...
Click to collapse
I was just trying to explain that my galaxy nexus had a similar issue

IRX120 said:
I was just trying to explain that my galaxy nexus had a similar issue
Click to expand...
Click to collapse
So that would mean all reboots are similar issues? No, whatever may have caused your issue is not necessarily the same issue here.

Sound like brick bug..... If flashing stock doesn't work then your only hope is to send it back to Samsung if you are still under warranty. Else, it looks like there is nothing much you can do as others have mentioned the eemc chip is gone

[Q] 1.5 GB sd-ext Limit

I'm using the CyanoMod007 form michaelm_007. http://forum.xda-developers.com/showthread.php?t=2213033
At the Last Page the user newsoup said the max. sd-ext Patition Limit is 1.5 GB
I've seen it in other ROMs that limit. But honestly, I've never understood why this limitation exists. Also Ext2 could already manage 2 TB. So why does this limit?
I would like to respond directly to the post but I have not yet, unfortunately my 10 post.
If it's just a driver issue then I would also like to fixed it I have some experience with Linux kernel programming.
I would appreciate if someone would tell me this limit
regards
Werner2005

Theres no actual limit...its more of a recommendation. Technically you can make it as big as the file system allow you to, but the reason why most people dont recommend an ext partition >1.5gb is because it will cause issues such as slow downs, crashes, etc.

Chromium_ said:
Theres no actual limit...its more of a recommendation. Technically you can make it as big as the file system allow you to, but the reason why most people dont recommend an ext partition >1.5gb is because it will cause issues such as slow downs, crashes, etc.
Click to expand...
Click to collapse
So just as describe in the thread I have a similar layout. 32 GB SD card and 4 GB sd-ext. I just thought I'm so does not matter enough memory on the SD.
The only problem is I read here repeatedly that how you say it, crashed, etc. are.
I would just interested why these crashes happen.
For example, I observed, at a Such crashes:
Code:
<3>[50587.562805] mmcblk0: error -110 sending read/write command, response 0x0, card status 0x0
<3>[50587.562835] mmcblk0: error -5 transferring data, sector 60169220, nr 69, card status 0x0
<3>[50587.562866] end_request: I/O error, dev mmcblk0, sector 60169220
<3>[50587.563323] mmc1: DMA channel flushed (0x80000004)
<3>[50587.563323] Flush data: 0000c003 3b35f000 00000000 00400040 00080008 00000003
<6>[50587.563385] mmc1: Controller has been re-initialiazed
<3>[50587.569061] mmcblk0: error -110 sending status command
<3>[50587.569091] mmcblk0: error -110 sending read/write command, response 0x0, card status 0x0
<3>[50587.569122] mmcblk0: error -5 transferring data, sector 60169221, nr 68, card status 0x0
<3>[50587.569122] end_request: I/O error, dev mmcblk0, sector 60169221
<3>[50587.569580] mmc1: DMA channel flushed (0x80000004)
<3>[50587.569610] Flush data: 0000c003 3b35f000 00000000 00400040 00080008 00000003
<6>[50587.569671] mmc1: Controller has been re-initialiazed
<3>[50587.575286] mmcblk0: error -110 sending status command
<3>[50587.575286] mmcblk0: error -110 sending read/write command, response 0x0, card status 0x0
<3>[50587.575317] mmcblk0: error -5 transferring data, sector 60169222, nr 67, card status 0x0
<3>[50587.575347] end_request: I/O error, dev mmcblk0, sector 60169222
The rest always looks the same...
It is easy intrsannt it seems as the kernel'd send data to the SD card Contoller. But the stop responding ...
Unfortunately, I've been writing the log to the late Cache was not enough to Figuring out what the first error.

use int2ext...

sohail.patel said:
use int2ext...
Click to expand...
Click to collapse
Take this as a piece of advice
I know you are trying to get 10 posts by being helpful here and there, but posting short answers in old threads doesn't really help anyone. Try referring to the newer threads. You could actually help someone that needs it