Bugzilla – Bug 328087
SD card reader not working
Last modified: 2007-10-05 08:50:54 UTC
Problem occurs when copying bigger data from card to disk: Sep 24 13:25:02 Staufen kernel: mmc0: Timeout waiting for hardware interrupt. Sep 24 13:25:02 Staufen kernel: mmcblk0: error 1 transferring data Sep 24 13:25:02 Staufen kernel: end_request: I/O error, dev mmcblk0, sector 85720 Sep 24 13:25:12 Staufen kernel: mmc0: Timeout waiting for hardware interrupt. Sep 24 13:25:12 Staufen kernel: mmcblk0: error 1 transferring data Sep 24 13:25:12 Staufen kernel: end_request: I/O error, dev mmcblk0, sector 85728 Sep 24 13:25:22 Staufen kernel: mmc0: Timeout waiting for hardware interrupt. ... lspci: ... 08:09.3 System peripheral: Ricoh Co Ltd R5C843 MMC Host Controller Subsystem: Samsung Electronics Co Ltd Unknown device c024 Flags: medium devsel, IRQ 255 Memory at de002c00 (32-bit, non-prefetchable) [disabled] [size=256] Capabilities: [80] Power Management version 2 08:09.4 System peripheral: Ricoh Co Ltd R5C592 Memory Stick Bus Host Adapter (rev 09) Subsystem: Samsung Electronics Co Ltd Unknown device c024 Flags: medium devsel, IRQ 255 BTW there were also problems at startup with hwclock (boot.clock). I was able to fix this issue by adding --directisa. Memory at de003000 (32-bit, non-prefetchable) [disabled] [size=256] Capabilities: [80] Power Management version 2 08:09.5 System peripheral: Ricoh Co Ltd xD-Picture Card Controller (rev 04) Subsystem: Samsung Electronics Co Ltd Unknown device c024 Flags: medium devsel, IRQ 255 Memory at de003400 (32-bit, non-prefetchable) [disabled] [size=256] Capabilities: [80] Power Management version 2
Tested again: Problem occurs not only with bigger data but almost always when copying or deleting some files on card.
Dirk, do you know whether this worked on older kernels or this has been like this all the time? Thanks.
OK, this is a known issue with this particular Ricoh MMC reader, see http://lkml.org/lkml/2007/10/1/375 I will backport the "disabler" patch to our kernel so that it goes out in the next update.
Jiri, yes it worked well with 2.6.18. Sounds good that this will be fixed soon. Thanks!
Jiri, I tried the patch myself without success, the error remains the same. The new ricoh-mmc module outputs: ricoh-mmc: Main firewire function not found. Cannot disable controller. BTW: Isn't it a little bit confusing that is worked with openSUSE 10.2 (default 2.6.18 kernel)?
Dirk, could you please provide output preceeding this error message? I.e. vendor-id, product-id, and any other related messages. Thanks.
Here you are: ... ACPI: PCI interrupt for device 0000:08:09.2 disabled sdhci: Secure Digital Host Controller Interface driver sdhci: Copyright(c) Pierre Ossman sdhci: SDHCI controller found at 0000:08:09.2 [1180:0822] (rev 18) ACPI: PCI Interrupt 0000:08:09.2[C] -> GSI 22 (level, low) -> IRQ 22 mmc0: SDHCI at 0xde002800 irq 22 DMA mmcblk0: mmc0:815d SD02G 2011136KiB mmcblk0: p1 ACPI: PCI interrupt for device 0000:08:09.2 disabled SFW2-INext-DROP-DEFLT IN=eth1 OUT= MAC=... SRC=... DST=... LEN=36 TOS=0x00 PREC=0x80 TTL=1 ID=31639 DF OPT (94040000) PROTO=2 ricoh-mmc: Ricoh MMC Controller disabling driver ricoh-mmc: Copyright(c) Philip Langdale ricoh-mmc: Ricoh MMC controller found at 0000:08:09.3 [1180:0843] (rev 0) ricoh-mmc: Main firewire function not found. Cannot disable controller. sdhci: Secure Digital Host Controller Interface driver sdhci: Copyright(c) Pierre Ossman sdhci: SDHCI controller found at 0000:08:09.2 [1180:0822] (rev 18) ACPI: PCI Interrupt 0000:08:09.2[C] -> GSI 22 (level, low) -> IRQ 22 mmc0: SDHCI at 0xde002800 irq 22 DMA mmcblk0: mmc0:815d SD02G 2011136KiB mmcblk0: p1 mmc0: Timeout waiting for hardware interrupt. sdhci: ============== REGISTER DUMP ============== sdhci: Sys addr: 0x393f461a | Version: 0x00000200 sdhci: Blk size: 0x00007200 | Blk cnt: 0x00000003 sdhci: Argument: 0x106ba000 | Trn mode: 0x00000033 sdhci: Present: 0x01ff0a06 | Host ctl: 0x00000007 sdhci: Power: 0x0000000f | Blk gap: 0x00000000 sdhci: Wake-up: 0x00000000 | Clock: 0x00000007 sdhci: Timeout: 0x00000007 | Int stat: 0x00000000 sdhci: Int enab: 0x00ff00fb | Sig enab: 0x00ff00fb sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000 sdhci: Caps: 0x01a021a1 | Max curr: 0x00000040 sdhci: =========================================== mmcblk0: error 1 transferring data end_request: I/O error, dev mmcblk0, sector 538064 mmc0: Timeout waiting for hardware interrupt. sdhci: ============== REGISTER DUMP ============== sdhci: Sys addr: 0x393fc006 | Version: 0x00000200 sdhci: Blk size: 0x00007200 | Blk cnt: 0x00000006 sdhci: Argument: 0x106bb000 | Trn mode: 0x00000033 sdhci: Present: 0x01ff0a06 | Host ctl: 0x00000007 sdhci: Power: 0x0000000f | Blk gap: 0x00000000 sdhci: Wake-up: 0x00000000 | Clock: 0x00000007 sdhci: Timeout: 0x00000007 | Int stat: 0x00000000 sdhci: Int enab: 0x00ff00fb | Sig enab: 0x00ff00fb sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000 sdhci: Caps: 0x01a021a1 | Max curr: 0x00000040 sdhci: =========================================== mmcblk0: error 1 transferring data end_request: I/O error, dev mmcblk0, sector 538072 ...
Dirk, as you are able to apply the patches yourself, do you think you could test whether the issue is still present in current vanilla kernel, and if so, that you could find the offending patch using git-bisect (bisecting the patches touching drivers/mmc directory should be enough, so it should require only a few recompile-reboot iterations)? Thanks.
Jiri, I tried kernel-vanilla.rpm from http://ftp.suse.com/pub/projects/kernel/kotd/HEAD/i386/ without any patches: Same problem :-( It doesn't seem to be an openSUSE specific problem but a general kernel problem.
Jiri, I have checked the notebook vendor's page today and found a new firmware version. After updating the firmware the SD card reader works again with the openSUSE standard kernel. Looks as if there was a BIOS (ACPI?) bug which somehow wasn't revealed by openSUSE 10.2 and kernel 2.6.18. Thanks for your support!
Dirk, thanks a lot for letting me know! Closing the bug.