Messages in this thread | | | Date | Thu, 21 Sep 2000 10:16:11 -0700 (PDT) | From | Linus Torvalds <> | Subject | Re: [PATCH] 2.4.0 i386 watchpoint problems |
| |
On Thu, 21 Sep 2000, James Cownie wrote: > > The problem > http://www.geocrawler.com/archives/3/343/2000/8/0/4140605/ > which is on Alan's list to be looked at for 2.2 remains in > 2.4.0. > > Here is a patch for 2.4.0
I'm very nervous that this patch couldlead to horrible performance issues due to somebody setting debug state to catch some nasty common condition in kernel code - causing tons of traps while we're not even schedulable.
I would suggest an alternate patch, which would be something like
if (SIGTRAP is pending in tsk) goto clear_dr7;
which basically says that we'll send the signal once, regardless of whether the trap happened in user or kernel mode, and we can't get into the nasty "tons of traps" situation because we'll at most get one trap until the signal has been delivered..
Does that work for you?
Linus
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
| |