Bug 201801

Summary:	SD card reader in Acer Aspire S5 broken after ASPM changes
Product:	Drivers	Reporter:	Rafael J. Wysocki (rjw)
Component:	PCI	Assignee:	drivers_pci (drivers_pci)
Status:	NEW ---
Severity:	normal	CC:	ptalbert
Priority:	P1
Hardware:	All
OS:	Linux
URL:	https://lore.kernel.org/lkml/2960808.4YCFhzuD0k@aspire.rjw.lan/
Kernel Version:	4.20-rc[1-4]	Subsystem:
Regression:	Yes	Bisected commit-id:
Attachments:	dmesg log with pci=earlydump acpidump output Output of 'lspci -vv' from Aspire S5 sample patch for RTS5209

Description Rafael J. Wysocki 2018-11-28 11:59:39 UTC

The SD card reader in my Acer Aspire S5 doesn't work with 4.20-rc.

When it doesn't work, it doesn't generate any interrupts on device insertion
and removal and this seems to be reproducible 100% of the time.

Reverting commit

commit 17c91487364fb33797ed84022564ee7544ac4945
Author: Patrick Talbert <ptalbert@redhat.com>
Date:   Wed Sep 5 09:12:53 2018 +0200

    PCI/ASPM: Do not initialize link state when aspm_disabled is set

makes the problem go away.

Comment 1 Rafael J. Wysocki 2018-11-28 12:08:42 UTC

Created attachment 279695 [details]
dmesg log with pci=earlydump

Comment 2 Rafael J. Wysocki 2018-11-28 12:09:15 UTC

Created attachment 279697 [details]
acpidump output

Comment 3 Rafael J. Wysocki 2018-11-28 12:09:55 UTC

Created attachment 279699 [details]
Output of 'lspci -vv' from Aspire S5

Comment 4 Patrick Talbert 2019-02-06 15:55:12 UTC

Created attachment 281023 [details]
sample patch for RTS5209

This seems to possibly be an issue with the card reader or rtsx driver.

I take it if you boot with pcie_aspm=off the card reader ends up in the same bad state?


What about this silly quirk? Can you test if it works for you?