Bug 44021
Summary: | Internal SD card reader is not working on Dell XPS 17 L502X | ||
---|---|---|---|
Product: | Drivers | Reporter: | Sparhawk (kernel) |
Component: | PCI | Assignee: | drivers_pci (drivers_pci) |
Status: | REOPENED --- | ||
Severity: | normal | CC: | alan, bjorn, chris, deprez.maarten, infinity.probability, szg00000, tails, tobedeprez |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | 3.12 | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: |
dmesg
lspci -vvvnn before rescan lspci -vvvnn after rescan .config file working dmesg dmesg (multiple cycles) dmesg lspci -vvvnn before rescan lspci -vvvnn after rescan |
Description
Sparhawk
2012-07-01 04:26:54 UTC
FWIW, I used the SD card reader recently, using the workaround of "echo 1 > /sys/bus/pci/rescan" as root. Then, I noticed a few entries in dmesg similar to this: [36964.574442] mmc0: Reset 0x4 never completed. [36964.584408] mmc0: Controller never released inhibit bit(s). [36964.683601] mmc0: Reset 0x2 never completed. [36964.782781] mmc0: Reset 0x4 never completed. [36964.792752] mmc0: Controller never released inhibit bit(s). [36964.891997] mmc0: Reset 0x2 never completed. [36964.991203] mmc0: Reset 0x4 never completed. [36965.001173] mmc0: Controller never released inhibit bit(s). [36965.100354] mmc0: Reset 0x2 never completed. [36965.199537] mmc0: Reset 0x4 never completed. [36965.209532] mmc0: Controller never released inhibit bit(s). [36965.308750] mmc0: Reset 0x2 never completed. [36965.407971] mmc0: Reset 0x4 never completed. [36965.418103] mmc0: Controller never released inhibit bit(s). [36965.517340] mmc0: Reset 0x2 never completed. [36965.616556] mmc0: Reset 0x4 never completed. [36965.715913] mmc0: Reset 0x1 never completed. Reassigning to PCI because this looks like a pciehp/acpiphp issue. This bug still affects XPS users - I'm one of them, with exactly the same symptons. Is there any news about it? Is any information needed? Cheers Thanks for chiming in, Ivan. I haven't had a chance to start looking at this, but if it's convenient for you, could you attach two complete dmesg logs: (1) when booting with an SD card already inserted, and (2) when booting with the SD card empty, then inserting an SD card later. I don't know what distro or kernel you're using, but if you can make sure the pciehp driver is linked in statically or loaded before inserting the SD card, that would be great. Created attachment 106863 [details]
dmesg
Created attachment 106864 [details]
lspci -vvvnn before rescan
Created attachment 106865 [details]
lspci -vvvnn after rescan
Created attachment 106866 [details]
.config file
This bug still exists in 3.10. Nothing seems to happen when I insert my SD card (at least there is no message in dmesg). After forcing rescan the SD card is detected and works well. I've attached my dmesg, an lspci before and after rescan and the .config file I used to build the kernel. v3.12-rc1 contains many acpiphp updates, as well as fixes to the code that decides whether to use acpiphp or pciehp. Can you please retest with v3.12-rc1 or later, attach the complete dmesg log, and note whether this issue is resolved? 3.12-rc1 fixes the bug for me. I've attached a (working) dmesg log Created attachment 109101 [details]
working dmesg
dmesg of 3.12-rc1
I can confirm that 3.12.0-031200rc1-generic fixes it for me too. (Thanks for attaching the dmesg log too, Tobe.) Sparhawk's downstream report: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/995743 I couldn't find it quickly for reference, nor post into See Also/URL fields at top of bug page, as either not implimented or permissions too restrictive. Sorry for the noise. Thank you Tobe, Sparhawk, and Christopher. Can any of you test several card insertion/removal cycles? This might be the same issue as reported in https://bugzilla.kernel.org/show_bug.cgi?id=43041, and Thomas reported there that the first card insertion worked correctly, but the card was not recognized after being removed and inserted a second time. I'm not going to close this issue until we have confirmation that it works for many insert/remove cycles. I have tries 10 insertion/removal cycles. The card was recognized every time. Created attachment 109601 [details]
dmesg (multiple cycles)
Great! Thanks, Tobe. I'm closing this. If anybody continues to see the problem, please attach a dmesg log and reopen this (or open a new bug if it seems like it might be a different issue). I'm still having the problem on an XPS 15 with modified firmware using EFI instead of BIOS. Will upload before and after lspcis as soon as i reboot. Created attachment 148211 [details]
dmesg
Created attachment 148221 [details]
lspci -vvvnn before rescan
Created attachment 148231 [details]
lspci -vvvnn after rescan
Hi Maarten, sorry I didn't notice your updates here. Can I trouble you to open a new bugzilla entry with a description of the issue you're seeing? I'm not sure you're seeing the same problem reported in this bug. Please also attach your .config to the new bugzilla. From your dmesg log in comment #20, it looks like you might not have CONFIG_HOTPLUG_PCI_PCIE enabled. That log shows that we enumerated the SD card reader at timestamp 240, but it sounds like maybe you had to manually rescan with something like echo 1 > /sys/bus/pci/rescan That shouldn't be necessary, and if CONFIG_HOTPLUG_PCI_PCIE is enabled, I think the kernel should automatically enumerate new devices. My guess is that your system basically hot-adds the card reader when a card is inserted. Hi Bjorn. CONFIG_HOTPLUG_PCI_PCIE did not resolve the problem. I opened bug #93561 detailing the case. |