Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
FS#34394 - [linux] 3.8.3-2 - 3.9.x breaks support for RTS5209 PCI Express Card Reader
Attached to Project:
Arch Linux
Opened by Volker Knollmann (foorgol) - Thursday, 21 March 2013, 00:19 GMT
Last edited by Tobias Powalowski (tpowa) - Wednesday, 17 July 2013, 17:36 GMT
Opened by Volker Knollmann (foorgol) - Thursday, 21 March 2013, 00:19 GMT
Last edited by Tobias Powalowski (tpowa) - Wednesday, 17 July 2013, 17:36 GMT
|
DetailsAfter upgrading to linux-3.8.3-2-x86_64 the built-in SD card reader of my laptop did not work anymore. When inserting a SD card, journalctl logged the following kernel message three times in a row:
mmc0: error -110 whilst initialising SD card As a result, no block device for the SD card was created and the card could not be mounted. lspci for the card reader: 0a:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. RTS5209 PCI Express Card Reader (rev 01) The error was reproducible by simply (re-)inserting a SD card. Tried two different cards which both showed the same behavior. A reboot did not solve the issue. I also tried loading all available kernel modules in kernel/drivers/mmc to rule out depmod / udev issues... without success. Google produced no results for error -110. Searching the kernel sources narrowed the origin of the message down to mmc_attach_sd in drivers/mmc/core/sd.c. However, I could not find the header file with the definition of error code 110. The system was fully updated when the error occurred. Downgrading the kernel to linux-3.7.10-1-x86_64 solved the problem. I noticed that the driver module rts_pstor for my card reader is not present in the 3.8.3-2 module tree anymore. But I found some postings saying that this driver has been replaced with a different one. Perhaps this transition to a new driver does not work for the RTS5209 PCI Express Card Reader in my machine... |
This task depends upon
Closed by Tobias Powalowski (tpowa)
Wednesday, 17 July 2013, 17:36 GMT
Reason for closing: Fixed
Additional comments about closing: 3.10.1-1
Wednesday, 17 July 2013, 17:36 GMT
Reason for closing: Fixed
Additional comments about closing: 3.10.1-1
according to this debian bug report (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=702108) it is required to set "CONFIG_MFD_RTSX_PCI" in the kernel config to get this device working again with 3.8.x kernels. i think this option is not set in the arch kernel configs.
CONFIG_MFD_RTSX_PCI=m
It is set.
And watch for dmesg output.
unfortunately I can't confirm that this fixes the bug.
I did a full system update tonight, installing kernel 3.8.11-1-x86_64. After rebooting, lsmod reports the following modules:
root@haktar:/# lsmod | grep rtsx
rtsx_pci_sdmmc 13972 0
mmc_core 91277 1 rtsx_pci_sdmmc
rtsx_pci_ms 7258 0
memstick 7400 1 rtsx_pci_ms
rtsx_pci 21492 2 rtsx_pci_ms,rtsx_pci_sdmmc
However, after inserting a SD card, dmesg still throws a
mmc0: error -110 whilst initialising SD card
The only "rtsx"-relevant dmesg messages are:
[volker@haktar ~]$ dmesg | grep rtsx
[ 15.211821] rtsx_pci 0000:0a:00.0: irq 49 for MSI/MSI-X
[ 15.211838] rtsx_pci 0000:0a:00.0: rtsx_pci_acquire_irq: pcr->msi_en = 1, pci->irq = 49
Looks not really like a bug message to me, but who knows...
I'll gonna downgrade again to 3.7.10-1...
Thanks,
Volker
rtsx_pci is loaded automatically during boot up, logging the same IRQ-related messages I reported yesterday.
Inserting the SD card gives again error -110.
Tonight I updated to 3.10.1-1 and my card reader is working again!
Upon card insertion, the devices mmcblk0 and mmcblk0p1 (for the first partition) are created and even automounting works like a charm!
I guess we can close this issue now...