Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable
From: Andrew Morton <akpm@linux-foundation.org>
Date: 2007-03-21 06:09:53
Also in:
lkml, netdev, xen-devel
Possibly related (same subject, not in this thread)
- 2007-04-12 · Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable · David Miller <davem@davemloft.net>
- 2007-03-21 · Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable · Zachary Amsden <hidden>
- 2007-03-21 · Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable · Linus Torvalds <torvalds@linux-foundation.org>
- 2007-03-21 · Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable · Zachary Amsden <hidden>
- 2007-03-21 · Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable · Linus Torvalds <torvalds@linux-foundation.org>
On Tue, 20 Mar 2007 21:23:52 +0100 Andi Kleen [off-list ref] wrote:
Well it causes additional problems. We had some cases where it was really hard to distingush garbage and the true call chain.
yes, for some reason the naive backtraces seem to have got messier and messier over the years and some of them are really quite hard to piece together nowadays. An accurate backtrace would have some value, if we can get it bullet-proof. The fault-injection driver wants it too. And lockdep, I guess.