On Thu, May 20, 2021 at 10:38:12AM -0700, Bhaumik Bhatt wrote: > On 2021-05-20 05:36 AM, Kalle Valo wrote: > > Manivannan Sadhasivam <manivannan.sadhasivam@xxxxxxxxxx> writes: > > > > > On Thu, May 20, 2021 at 12:02:40PM +0200, Greg KH wrote: > > > > On Thu, May 20, 2021 at 12:47:53PM +0300, Kalle Valo wrote: > > > > > Hi, > > > > > > > > > > I got several reports that this mhi commit broke ath11k in v5.12.3: > > > > > > > > > > commit 29b9829718c5e9bd68fc1c652f5e0ba9b9a64fed > > > > > Author: Bhaumik Bhatt <bbhatt@xxxxxxxxxxxxxx> > > > > > Date: Wed Feb 24 15:23:04 2021 -0800 > > > > > > > > > > bus: mhi: core: Process execution environment changes serially > > > > > > > > > > [ Upstream commit ef2126c4e2ea2b92f543fae00a2a0332e4573c48 ] > > > > > > > > > > Here are the reports: > > > > > > > > > > https://bugzilla.kernel.org/show_bug.cgi?id=213055 > > > > > > > > > > https://bugzilla.kernel.org/show_bug.cgi?id=212187 > > > > > > > > > > https://bugs.archlinux.org/task/70849?project=1&string=linux > > > > > > > > > > Interestingly v5.13-rc1 seems to work fine, at least for me, though I > > > > > have not tested v5.12.3 myself. Can someone revert this commit in the > > > > > stable release so that people get their wifi working again, please? > > > > > > > > How does the mhi bus code relate to a ath11k driver? What bus > > > > is that > > > > on? > > > > > > > > > > MHI is the transport used by the ath11k driver to work with the WLAN > > > devices > > > over PCIe. > > > > > > Regarding the bug, I'd suggest to wait for Bhaumik (the author of > > > 29b9829718c5) > > > to comment on the possible commit which needs backporting from > > > mainline. > > > > Ok, but if a quick fix is not available I think we should just revert > > this in the stable releases. I also got a report that v5.11.21 is > > broken: > > > > https://bugzilla.kernel.org/show_bug.cgi?id=213055#c11 > > Please pick [1] as the dependency to [ Upstream commit > ef2126c4e2ea2b92f543fae00a2a0332e4573c48 ]. what is [1]??? What commit do I need to backport, a commit id would be nice... thanks, greg k-h
