Siyah kernel 3.1RC6 and BRICKED GALAXY S2

Breaksomeoff · Apr 26, 2012

Sounds like a terrible brick, you have my sympathy.

Sadly, this is the time we live in. Due to the complexity of our devices, more and more things are getting bricked via unofficial flashing of software. I myself had a £200 device turn into a paperweight about 7 years ago, it's a horrible feeling.

If you update immediately on anything unofficial then you are taking risk, no matter who has done the update. I stayed on 3.1rc5 as I always look at changelog and give it a day. I noticed blnww was removed and didn't want to lose it, but this could of easily happened to anyone and I hope you get a resolution soon.

jchusted · Apr 26, 2012

MrSerious92 said:
Tomorrow i will get my phone back Support repaired it for free Thats why i love germany

That's Nice.

Was to phone totally dead ?
They where not able to see that the phone was rooted ?

sauron96t6 · Apr 26, 2012

MrSerious92 said:
Tomorrow i will get my phone back Support repaired it for free Thats why i love germany

Fast assistance... sounds good!
I'm happy for you!!!

qqqqqq0 · Apr 26, 2012

Stitch! said:
Same experience here btw, hardbricked my phone. I dont consider myself to be a noob, and I tried basically everything that is possible. adb recovery, my jtag, eventually ended up replacing the motherboard. Now my phone is working great again. The main problem was. Samsung warrenty center i send it to initially told me it was not working due to water damage. After getting a second opinion from a friend I knew they where lying about this. No waterdamage whatsoever, just a non responsive motherboard.

Replacing the motherboard is not very hard, but you need a steady hand and some tools. If you are not sure you can do it i would recommend just paying a service center to do it. the 50 euros i saved by doing it myself may not be worth losing your warranty for most people.

be careful

Entropy512 · Apr 26, 2012

pejmany said:
I AM able to get into recovery,
Flashed pit by itself
Flashed KI3 (it failed fullout this time, not just stopped)
Flashed Rogue Recovery 1.1
Did adb, this happened

Rogue recovery - I'm guessing you're on an Epic 4G Touch?

That's interesting - it looks like your partition table is hosed up, it thinks that many of those partitions are 0 length. Cache is good, the rest are screwy. I'll come up with some additional commands to try tonight.

Crescendo Xenomorph said:
Sounds like internal voltage/power to the eMMC chip also get changed with "factory reset" code that can cause eMMC chip broken.
Otherwise its impossible for a software to kill the hardware.

I thought it was the bootloader that got corrupted so the device can't even flash any firmware properly.

Originally gokhan and I were leaning towards voltage regulator stuff, but I think it's something else.

Samsung set MMC_CAP_ERASE in the MMC driver of the kernel for some devices - confirmed for Korean M250* devices. It may be that some eMMC chips don't like this feature and get fried.

sauron96t6 said:
IMHO the Siyah not to blame, it could happen with any other kernel ... unless we take risks we must leave the phone's original factory and we will never have problems. If we want something more than the original product, then we must accept and assume any risks.

In fact it is happening with many other kernels:
Any ICS leak for Epic 4G Touch (Sprint GSII)
Any ICS leak for Galaxy Note
The UCLD3 ICS leak for AT&T Galaxy S II
Apparently Korean M250* kernels (due to language barrier, we often don't hear about these until there's a reason to do serious digging) - These are the only ones where there is source to analyze.

I9100 users need to realize how lucky they are that this is the ONLY device with Exynos that didn't receive dangerous ICS kernels, even in the leak phase.

I think Samsung is aware of the issue and trying to solve it - see: https://github.com/Entropy512/kernel_galaxys2_ics/commit/d069cc5cb9f0fd9409ea10d4037e4c0e68701b8e - they disabled TRIM which is a relative of ERASE. Unfortunately they left ERASE in - I think this is the cause of the issue.

Those who are softbricked:

Code:

cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name

in a shell
Gokhan has VYL00M, as do I in my Note (haven't checked my 777 yet) - he's actually done multiple wipes with the "bad" kernel without issues.

Also, to those talking about repartitioning - you would need to do it using a PIT file in Odin. However I think more data needs to be collected first - I'll have some more commands to run tonight.

Stitch! · Apr 26, 2012

kachrukamble said:
In the end, do you got motherboard under warranty or you paid price for the same?

I replaced it myself, got it for 90 euros through someone in wholesales, since I didn't leave any marks and kept the seal intact, I can still apply for the normal warranty in case my screen breaks or something.

sauron96t6 · Apr 26, 2012

Disassembling S2 - How to

Excellent video step by step to replace the motherboard by yourself!
http://www.youtube.com/watch?v=65pvAINg2ME&feature=related

mckeowngoo · Apr 26, 2012

Is this the JTAG required?

http://www.amazon.co.uk/gp/aw/d/B005ARV26O/ref=redir_mdp_mobile

Doesn't seem that expensive if it is.

Sent from my GT-I9100 using xda premium

Bor_NL · Apr 26, 2012

mckeowngoo said:
Is this the JTAG required?

http://www.amazon.co.uk/gp/aw/d/B005ARV26O/ref=redir_mdp_mobile

Doesn't seem that expensive if it is.

Sent from my GT-I9100 using xda premium

That's a jig to reset your flashcounter

kamesam · Apr 26, 2012

Help with Galaxy S2 i9100 "softbricked"!!

Entropy512 said:
Those who are softbricked:

Code:

cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name

Hi,

I got the "softbrick" I guess. I can flash kernels, bootloaders and firmwares without /data, with .PIT or without it. But when I try to wipe data or flash a firmware that goes through data, then I got stuck (GB or ICS - doesn't matter). Either with Odin or CWM. All buttons still operational so I can still get to download mode or recovery.
Can you please provide more details on
cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name

What exactly it will do?
Do we need to run it through ADB (In this case, can you point me to some references in the forum on how to use it for this purpose)? Do we need to be root?

Thanks in advance,

mckeowngoo · Apr 26, 2012

Bor_NL said:
That's a jig to reset your flashcounter

Don't I feel like the f**king idiot!

So how much does one of these things cost and has anyone got a link please?

Sent from my GT-I9100 using xda premium

Entropy512 · Apr 26, 2012

kamesam said:
Hi,

I got the "softbrick" I guess. I can flash kernels, bootloaders and firmwares without /data, with .PIT or without it. But when I try to wipe data or flash a firmware that goes through data, then I got stuck (GB or ICS - doesn't matter). Either with Odin or CWM. All buttons still operational so I can still get to download mode or recovery.
Can you please provide more details on
cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name

What exactly it will do?
Do we need to run it through ADB (In this case, can you point me to some references in the forum on how to use it for this purpose)? Do we need to be root?

Thanks in advance,

In an ADB shell - and doing that command pulls the identification of your flash chip - I'm wondering of those who softbricked have a different chip than gokhan does.

sauron96t6 · Apr 26, 2012

Riff-Jtag-Box

RIFF-JTAG-BOX
http://www.riffbox.org/downloads/GettingStarted.pdf
http://forum.gsmhosting.com/vbb/f634/riff-jtag-box-important-manuals-1051474/

kachrukamble · Apr 26, 2012

Entropy512 said:
In an ADB shell - and doing that command pulls the identification of your flash chip - I'm wondering of those who softbricked have a different chip than gokhan does.

I am wondering of those soft brick devices had single boot or dualboot.

Sent from my GT-I9000 using Tapatalk

igottu · Apr 27, 2012

Did you know Siyah actually means Black? So your phones are!

KennyLegend · Apr 27, 2012

igottu said:
Did you know Siyah actually means Black? So your phones are!

And????

It was only 1 version of Siyah had issues and not everybody using it got bricked.

Sent from my GT-I9100 using xda premium

kamesam · Apr 27, 2012

Entropy512 said:
In an ADB shell - and doing that command pulls the identification of your flash chip - I'm wondering of those who softbricked have a different chip than gokhan does.

I tried to ADB to the S2 but it's not recognizing the device. It says 'error: device not found'. I am able to ADB to another Samsung device (i5500) though, but not to my S2. I don't know much of ADB. What am I doing wrong here?

Crescendo Xenomorph · Apr 27, 2012

Entropy512 said:
Rogue recovery - I'm guessing you're on an Epic 4G Touch?

That's interesting - it looks like your partition table is hosed up, it thinks that many of those partitions are 0 length. Cache is good, the rest are screwy. I'll come up with some additional commands to try tonight.

Originally gokhan and I were leaning towards voltage regulator stuff, but I think it's something else.

Samsung set MMC_CAP_ERASE in the MMC driver of the kernel for some devices - confirmed for Korean M250* devices. It may be that some eMMC chips don't like this feature and get fried.

In fact it is happening with many other kernels:
Any ICS leak for Epic 4G Touch (Sprint GSII)
Any ICS leak for Galaxy Note
The UCLD3 ICS leak for AT&T Galaxy S II
Apparently Korean M250* kernels (due to language barrier, we often don't hear about these until there's a reason to do serious digging) - These are the only ones where there is source to analyze.

I9100 users need to realize how lucky they are that this is the ONLY device with Exynos that didn't receive dangerous ICS kernels, even in the leak phase.

I think Samsung is aware of the issue and trying to solve it - see: https://github.com/Entropy512/kernel_galaxys2_ics/commit/d069cc5cb9f0fd9409ea10d4037e4c0e68701b8e - they disabled TRIM which is a relative of ERASE. Unfortunately they left ERASE in - I think this is the cause of the issue.

Those who are softbricked:

Code:

cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name

in a shell
Gokhan has VYL00M, as do I in my Note (haven't checked my 777 yet) - he's actually done multiple wipes with the "bad" kernel without issues.

Also, to those talking about repartitioning - you would need to do it using a PIT file in Odin. However I think more data needs to be collected first - I'll have some more commands to run tonight.

I got something on the net, are they the same explanation for those TRIM/ERASE commands that mentioned in your github (https://github.com/Entropy512/kernel_galaxys2_ics/commit/d069cc5cb9f0fd9409ea10d4037e4c0e68701b8e)?

What's the difference between TRIM and ERASE command?
I'm really curious, how come an erase/trim command can fries the chip...

Thanks..

Cheers,
Crescendo

Code:

SDHCI host controller has TRIM/ERASE capability, enable these caps for erasing
purpose.

ERASE command needs R1B response. So for such command with busy signal, before
sending command, SDHCI host driver will simply set a maximum value for timeout
control register which is safe to use.

Signed-off-by: Chuanxiao Dong <chuanxiao.dong@intel.com>

drivers/mmc/host/sdhci.c | 16 +++++++++++--
1 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 655617c..fe7cbd0 100644
a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -41,7 +41,6 @@

static unsigned int debug_quirks = 0;

-static void sdhci_prepare_data(struct sdhci_host *, struct mmc_data *);
static void sdhci_finish_data(struct sdhci_host *);

static void sdhci_send_command(struct sdhci_host *, struct mmc_command *);
@@ -652,16 +651,23 @@ static void sdhci_set_transfer_irqs(struct sdhci_host *host)
sdhci_clear_set_irqs(host, dma_irqs, pio_irqs);
}

-static void sdhci_prepare_data(struct sdhci_host *host, struct mmc_data *data)
+static void sdhci_prepare_data(struct sdhci_host *host, struct mmc_command *cmd)
{
u8 count;
u8 ctrl;
+	struct mmc_data *data = cmd->data;
int ret;

WARN_ON(host->data);

-	if (data == NULL)
+	if (data == NULL) {
+	 /*
+	 * set timeout to be maximum value for command with busy signal.
+	 */
+	 if (cmd->flags & MMC_RSP_BUSY)
+	 sdhci_writeb(host, 0xE, SDHCI_TIMEOUT_CONTROL);
return;
+	}

/* Sanity checks */
BUG_ON(data->blksz * data->blocks > 524288);
@@ -921,7 +927,7 @@ static void sdhci_send_command(struct sdhci_host *host, struct mmc_command *cmd)

host->cmd = cmd;

-	sdhci_prepare_data(host, cmd->data);
+	sdhci_prepare_data(host, cmd);

sdhci_writel(host, cmd->arg, SDHCI_ARGUMENT);

@@ -1939,7 +1945,7 @@ int sdhci_add_host(struct sdhci_host *host)
mmc->f_min = host->max_clk / SDHCI_MAX_DIV_SPEC_200;

mmc->f_max = host->max_clk;
-	mmc->caps |= MMC_CAP_SDIO_IRQ;
+	mmc->caps |= MMC_CAP_SDIO_IRQ | MMC_CAP_ERASE;

/*
* A controller may support 8-bit width, but the board itself
1.6.6.1

drakester09 · Apr 27, 2012

Entropy512 said:
Rogue recovery - I'm guessing you're on an Epic 4G Touch?

That's interesting - it looks like your partition table is hosed up, it thinks that many of those partitions are 0 length. Cache is good, the rest are screwy. I'll come up with some additional commands to try tonight.

Originally gokhan and I were leaning towards voltage regulator stuff, but I think it's something else.

Samsung set MMC_CAP_ERASE in the MMC driver of the kernel for some devices - confirmed for Korean M250* devices. It may be that some eMMC chips don't like this feature and get fried.

In fact it is happening with many other kernels:
Any ICS leak for Epic 4G Touch (Sprint GSII)
Any ICS leak for Galaxy Note
The UCLD3 ICS leak for AT&T Galaxy S II
Apparently Korean M250* kernels (due to language barrier, we often don't hear about these until there's a reason to do serious digging) - These are the only ones where there is source to analyze.

I9100 users need to realize how lucky they are that this is the ONLY device with Exynos that didn't receive dangerous ICS kernels, even in the leak phase.

I think Samsung is aware of the issue and trying to solve it - see: https://github.com/Entropy512/kernel_galaxys2_ics/commit/d069cc5cb9f0fd9409ea10d4037e4c0e68701b8e - they disabled TRIM which is a relative of ERASE. Unfortunately they left ERASE in - I think this is the cause of the issue.

Those who are softbricked:

Code:

cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name

in a shell
Gokhan has VYL00M, as do I in my Note (haven't checked my 777 yet) - he's actually done multiple wipes with the "bad" kernel without issues.

Also, to those talking about repartitioning - you would need to do it using a PIT file in Odin. However I think more data needs to be collected first - I'll have some more commands to run tonight.

Just FYI, I have VYL00M on my i777

Luckily I did not flash the ATT leak nor we had a rc6 for I777.

Sent from my SGH-I777 using Tapatalk 2

pejmany · Apr 27, 2012

Entropy512 said:
Rogue recovery - I'm guessing you're on an Epic 4G Touch?

That's interesting - it looks like your partition table is hosed up, it thinks that many of those partitions are 0 length. Cache is good, the rest are screwy. I'll come up with some additional commands to try tonight.

Originally gokhan and I were leaning towards voltage regulator stuff, but I think it's something else.

Samsung set MMC_CAP_ERASE in the MMC driver of the kernel for some devices - confirmed for Korean M250* devices. It may be that some eMMC chips don't like this feature and get fried.

No, the Bell Galaxy S II (i9100m), white, if it helps.
It seems like the i777 and i9100m may be different, still don't know though.
I tried out the cat command, got

Code:

N:\android-sdk\platform-tools>adb shell
~ # cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name
cat /sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name
cat: can't open '/sys/devices/platform/dw_mmc/mmc_host/mmc0/mmc0:0001/name': No
such file or directory
~ #

Then again, I may have just done it wrong :/ Open to retries

And yeah, in my thread I figured it was partition table, I asked if there's a way to rewrite mbr if I can get the partition data from another gsii
If that ^ makes no sense, I hope you got the idea of what I was going for, cause I'm not too bright definition wise xD

EDIT: should i try reflashing ki3 again before doing the cat?

Siyah kernel 3.1RC6 and BRICKED GALAXY S2

Senior Member

Senior Member

Senior Member

Senior Member

Senior Recognized Developer

Senior Member

Senior Member

Senior Member

Senior Member

New member

Senior Member

Senior Recognized Developer

Senior Member

Senior Member

New member

Senior Member

New member

Senior Member

Senior Member

Senior Member

Similar threads

Top Liked Posts