Messages in this thread | | | Date | Fri, 25 Mar 2022 19:36:46 +0200 | Subject | Re: [PATCH 5.10 11/38] swiotlb: rework "fix info leak with DMA_FROM_DEVICE" | From | Thomas Backlund <> |
| |
Den 2022-03-25 kl. 19:08, skrev Linus Torvalds: > On Fri, Mar 25, 2022 at 8:09 AM Greg Kroah-Hartman > <gregkh@linuxfoundation.org> wrote: >> >> From: Halil Pasic <pasic@linux.ibm.com> >> >> commit aa6f8dcbab473f3a3c7454b74caa46d36cdc5d13 upstream. > > This one causes a regression on at least some wireless drivers > (ath9k). There's some active discussion about it, maybe it gets > reverted, maybe there are other options. > > There's an ath9k patch that fixes the problem, so if this patch goes > into the stable tree the ath9k fix will follow. > > But it might be a good idea to simply hold off on this patch a bit, > because that ath9k patch hasn't actually landed yet, there's some > discussion about it all, and it's not clear that other drivers might > not have the same issue. > > So I'm not NAK'ing this patch from stable, but I also don't think it's > timing-critical, and it might be a good idea to delay it for a week or > two to both wait for the ath9k patch and to see if something else > comes up. > > Linus >
As it has already landed in 5.15.29 and 5.16.15 it should IMHO preferably be reverted in those trees too to get them back to working state...
-- Thomas
| |