from what I read from Linus / Roland discussion on this topic (this night) on lkml, Linus only patched 1 out of 3 places which need to be patched (according to Roland). But I don't know yet if Linus will finish the job or not :-( IMO, what Linus proposes heads into the right direction (don't single step in signal handler if the process itself has set the TF flag, but do if the debugger has done it), however I cannot tell whether it's going to fix all issues.
A+
Message du 16/11/04 02:15 De : "Jesse Allen" A : wine-devel@winehq.org Copie à : Objet : Re: Kernel 2.6.9 - Issue Found On Sun, Nov 14, 2004 at 08:45:00PM -0700, Jesse Allen wrote:
On Sun, Nov 14, 2004 at 06:11:33PM -0800, Dan Kegel wrote:
I couldn't find the patch at either of the two links Jesse posted, but I did see something similar here: http://marc.theaimsgroup.com/?l=bk-commits-head&m=110048429304053&w=...
That's a brand new patch today.
"x86: only single-step into signal handlers if the tracer asked for it."
Looks like Linus may have been alerted by the problem and made this patch just in time for 2.6.10-rc2. Is this the "switch-off" we need?
I'll be testing it shortly.
It doesn't seem to fix the copy protection problem. So I'm wondering a few things. Does wine need to be patched? This new patch seems to be related to a chunk of code added in patch #2 mentioned in the original posting. I've tried 2.6.10-rc2 with patch #1 reversed and patch 1+3 reversed. No difference this time.
Any idea will help.
Jesse
On Tue, Nov 16, 2004 at 09:54:26AM +0100, Pouech Eric DMI AEI CAEN wrote:
from what I read from Linus / Roland discussion on this topic (this night) on lkml, Linus only patched 1 out of 3 places which need to be patched (according to Roland). But I don't know yet if Linus will finish the job or not :-( IMO, what Linus proposes heads into the right direction (don't single step in signal handler if the process itself has set the TF flag, but do if the debugger has done it), however I cannot tell whether it's going to fix all issues.
A+
Ok, I noticed the same thing Roland did. So they already know what's going on. I'll keep monitoring the situation as much as I can.
Jesse
Jesse Allen a écrit :
On Tue, Nov 16, 2004 at 09:54:26AM +0100, Pouech Eric DMI AEI CAEN wrote:
from what I read from Linus / Roland discussion on this topic (this night) on lkml, Linus only patched 1 out of 3 places which need to be patched (according to Roland). But I don't know yet if Linus will finish the job or not :-( IMO, what Linus proposes heads into the right direction (don't single step in signal handler if the process itself has set the TF flag, but do if the debugger has done it), however I cannot tell whether it's going to fix all issues.
A+
Ok, I noticed the same thing Roland did. So they already know what's going on. I'll keep monitoring the situation as much as I can.
Jesse
Hi Jesse,
could you try to fix the other spots on 2.6.9 (as Roland suggested), as noone on Linux side seems ready to make the move (that should be rather straightforward IMO). A+
-
Eric Pouech -
Jesse Allen -
Pouech Eric DMI AEI CAEN