Hi -stable kernel team,
Commit 216f8e95aacc8e ("PCI: mvebu: Setup BAR0 in order to fix MSI") fixes ath10k_pci devices on Armada 385 systems, and probably also others using similar PCI hosts. Can you consider applying it to v5.4.x? The commit should apply cleanly to current v5.4.
Technically this bug is not a regression. MSI only worked with vendor provided U-Boot. So I'm not sure it is eligible for -stable. But still, this bug bites users of latest OpenWrt release that is based on kernel v5.4. Oli (on Cc) verified that this commit fixes the issue on OpenWrt with kernel v5.4.
I put PCI/mvebu maintainers to Cc so they can provide their input.
Thanks, baruch
On Mon, Aug 02, 2021 at 01:59:58PM +0300, Baruch Siach wrote:
Hi -stable kernel team,
Commit 216f8e95aacc8e ("PCI: mvebu: Setup BAR0 in order to fix MSI") fixes ath10k_pci devices on Armada 385 systems, and probably also others using similar PCI hosts. Can you consider applying it to v5.4.x? The commit should apply cleanly to current v5.4.
Technically this bug is not a regression. MSI only worked with vendor provided U-Boot. So I'm not sure it is eligible for -stable. But still, this bug bites users of latest OpenWrt release that is based on kernel v5.4. Oli (on Cc) verified that this commit fixes the issue on OpenWrt with kernel v5.4.
Looks obvious enough, now queued up, thanks.
greg k-h
linux-stable-mirror@lists.linaro.org
-
Baruch Siach -
Greg KH