Bug 214947
Summary: | kernel NULL pointer dereference in snd_pcm_mmap_data for RME HDSP audio interfaces | ||
---|---|---|---|
Product: | Drivers | Reporter: | Till Schäfer (till2.schaefer) |
Component: | Sound(ALSA) | Assignee: | Jaroslav Kysela (perex) |
Status: | RESOLVED CODE_FIX | ||
Severity: | normal | CC: | tiwai |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | 5.14.* | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: |
alsa-info.txt
Fix patch |
Description
Till Schäfer
2021-11-04 23:08:16 UTC
this patch seems to be related: https://www.spinics.net/lists/alsa-devel/msg129824.html (In reply to Till Schäfer from comment #1) > this patch seems to be related: > https://www.spinics.net/lists/alsa-devel/msg129824.html Yes, those three patches should be backported to 5.14.y. The corresponding upstream commits are cbea6e5a7772b7a5b80baa8f98fd77853487fd2a 0899a7a23047f106c06888769d6cd6ff43d7395f 4d9e9153f1c64d91a125c6967bc0bfb0bb653ea0 Could you verify whether they fix the bug? I was able to fix the bug with these two patches atop 5.14.16. ALSA: pcm: Check mmap capability of runtime dma buffer at first ALSA: pci: rme: Set up buffer type properly (the third one was not relevant for me with RME hardware) OK, thanks. I'll ask Greg to cherry-pick those for 5.14.y stable tree. I found another bug in the rme patch 0899a7a23047f106c06888769d6cd6ff43d7395f, and it's not fully backportable to 5.14.y, either. Could you drop those patches again and test the patch below instead? It's the one I submitted and merged to the upstream now. Created attachment 299485 [details]
Fix patch
Works for me! Thanks for quick testing! Then this will be the fix for 5.14.x. It'll be included in the next PR for 5.16-rc1 in this week, then will be backported to 5.14.x. |