Bug 107811 - SD card time out too short in tifm_sd.c
Summary: SD card time out too short in tifm_sd.c
Status: NEW
Alias: None
Product: IO/Storage
Classification: Unclassified
Component: Other (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: io_other
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-11-13 12:10 UTC by Bernd Edlinger
Modified: 2016-03-21 18:17 UTC (History)
1 user (show)

See Also:
Kernel Version: 3.13.0
Subsystem:
Regression: No
Bisected commit-id:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Description Bernd Edlinger 2015-11-13 12:10:49 UTC 
Hi,

this happened with a 16GB Sandisk SDHC card, and a TI tifm_7xx1 host if,
originally with 3.13.0 (ubuntu 14.4), but should also happen with latest kernel.


Jul 15 06:54:29 tobias-VGN-N11M-W kernel: [   76.695120] tifm_7xx1 0000:08:03.2: enabling device (0000 -> 0002)
Jul 15 06:54:29 tobias-VGN-N11M-W kernel: [   76.719492] tifm_core: MMC/SD card detected in socket 0:1
Jul 15 06:54:29 tobias-VGN-N11M-W kernel: [   77.029932] mmc0: new SDHC card at address aaaa
Jul 15 06:54:29 tobias-VGN-N11M-W kernel: [   77.039647] mmcblk0: mmc0:aaaa SS16G 14.8 GiB
Jul 15 06:54:29 tobias-VGN-N11M-W kernel: [   77.050276]  mmcblk0: p1 p2
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.336053] tifm_sd0:1 : card failed to respond for a long period of time (19, 1)
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.336273] tifm0 : demand removing card from socket 0:1
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.336388] mmc0: card aaaa removed
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.336479] mmcblk0: error -123 sending status command, retrying
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.336615] mmcblk0: error -123 sending status command, retrying
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.336747] mmcblk0: error -123 sending status command, aborting
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.336879] end_request: I/O error, dev mmcblk0, sector 29362176
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337012] Buffer I/O error on device mmcblk0p2, logical block 0
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337138] lost page write due to I/O error on mmcblk0p2
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337173] end_request: I/O error, dev mmcblk0, sector 29362184
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337303] Buffer I/O error on device mmcblk0p2, logical block 1
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337429] lost page write due to I/O error on mmcblk0p2
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337449] end_request: I/O error, dev mmcblk0, sector 29362192
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337578] Buffer I/O error on device mmcblk0p2, logical block 2
Jul 15 06:55:55 tobias-VGN-N11M-W kernel: [  173.337704] lost page write due to I/O error on mmcblk0p2


the following patch fixed the time out for me:


diff --git a/drivers/mmc/host/tifm_sd.c b/drivers/mmc/host/tifm_sd.c
index 93c4b40..84c3f80 100644
--- a/drivers/mmc/host/tifm_sd.c
+++ b/drivers/mmc/host/tifm_sd.c
@@ -964,7 +964,7 @@ static int tifm_sd_probe(struct tifm_dev *sock)
        host = mmc_priv(mmc);
        tifm_set_drvdata(sock, mmc);
        host->dev = sock;
-       host->timeout_jiffies = msecs_to_jiffies(1000);
+       host->timeout_jiffies = msecs_to_jiffies(10000);
 
        tasklet_init(&host->finish_tasklet, tifm_sd_end_cmd,
                     (unsigned long)host);


An interesting fact is, that it did alwas only happen with cmd19
which is send tuning block.
Note You need to log in before you can comment on or make changes to this bug.