Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 20 Jan 2022 04:34:14 +0000
From:      bugzilla-noreply@freebsd.org
To:        bugs@FreeBSD.org
Subject:   [Bug 261338] [PATCH] kernel panic "bad pte" under heavy CPU load on 12.2 and 12.3 (i386)
Message-ID:  <bug-261338-227-hGjIH5rqeb@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-261338-227@https.bugs.freebsd.org/bugzilla/>
References:  <bug-261338-227@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D261338

--- Comment #5 from Dmitry K. <thedix@yandex.ru> ---
smp_targeted_tlb_shootdown() is invoked in smp_masked_invl*() and
smp_cache_flush().=20

As I can see smp_masked_invl*() are invoked with correct sched_pin() /
sched_unpin() order in pmap_*().

But smp_cache_flush() is invoked by pmap_invalidate_cache(), which is calle=
d in
multiple places in pmap_*(). Some functions do not have outer sched_pin() /
sched_unpin() guarding when calling pmap_invalidate_cache(), for example, in
pmap_flush_page(). And it leads to the wrong order of pins in
smp_targeted_tlb_shootdown().

So I would suggest to call sched_pin() / sched_unpin() explicitly in
smp_targeted_tlb_shootdown() to make sure (as it was in the previous versio=
n).

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-261338-227-hGjIH5rqeb>