Re: [PATCH] net: detect recursive dev_queue_xmit() on RT properly
From: Brian Silverman <hidden>
Date: 2016-11-30 00:35:45
This is already fixed by 0f27b2b421a4 ("net: move xmit_recursion to
per-task variable on -RT"), which is in the various stable-rts too.
Apologies for the noise.
quoted
This also includes "543a589a00e8 net: xmit lock owner cleanup" from v2.6.33.9-rt31 by mingo and tglx, which was just some cleanups to the original patch which make porting to to current upstream easier. Signed-off-by: Mark Beauchemin <redacted> [ ported to latest upstream ] Signed-off-by: Ingo Molnar <redacted> Signed-off-by: Thomas Gleixner <redacted> [ ported to 4.8 and used struct task_struct instead of void ] Signed-off-by: Brian Silverman <redacted>The signed-off-by chain suggest that Mark passed the patch to Ingo, Thomas and to Brian. I doubt that. I can't resolve the sha1 id mentioned here.
It's from the rt-history repository. The original commit in the rt-history repository has that Signed-off-by chain except I added me at the bottom. I guess it would make more sense to remove the other people who just cherry-picked it if I run into this again.
Is this still broken as of v4.8-RT? I remember fixing something similar with bridge devices.
Indeed you did it. It was cherry-picked to v4.1.19-rt22 too, which matches the 4.1 kernel I originally found this on. I thought I checked for the problem one or both of those, but apparently not... My bad.