On Wed, 2013-02-20 at 17:29 +0100, Oleg Nesterov wrote: > Hi Ben, > > sorry for delay, > > On 02/20, Ben Hutchings wrote: > > > > On Tue, 2013-02-19 at 20:35 +0100, Stefan Priebe wrote: > > > Am 19.02.2013 20:05, schrieb Ben Hutchings: > > > > On Tue, Feb 19, 2013 at 06:51:53AM -0800, Greg KH wrote: > > > >> On Tue, Feb 19, 2013 at 10:18:37AM +0100, Stefan Priebe - Profihost AG > > > >> wrote: > > > >>> Hi, > > > >>> > > > >>> will we see a backport for the above fix for the 3.0.X tree? > > > >> > > > >> Is it applicable to the 3.0.x tree? And if so, someone needs to > > > >> backport it, and provide it to me, can you? > > > > > > > > For 3.2.y I've cherry-picked (in this order): > > > > > > > > 848e8f5f0ad3169560c516fff6471be65f76e69f > > > > 95cf00fa5d5e2a200a2c044c84bde8389a237e02 > > > > 910ffdb18a6408e14febbb6e4b6840fd2c928c82 > > > > 9899d11f654474d2d54ea52ceaa2a1f4db3abd68 > > > > 9067ac85d533651b98c2ff903182a20cbb361fcb > > > > > > > > The last three fix the more recently-discovered race. The first two > > > > are a fix for an x86-specific race in ptrace, made by Oleg back in > > > > August. The fourth at least textually depends on the first two. > > > > > > > > I'm not sure whether 3.0.y would need anything more. > > > at least on 3.0 this picking does not work as > > > 910ffdb18a6408e14febbb6e4b6840fd2c928c82 > > > > > > does not apply. But backports of the last three are here: > > > http://www.mail-archive.com/[email protected]/msg31694.html > > > > Oleg, could you comment on whether 'ptrace/x86: Partly fix > > set_task_blockstep()->update_debugctlmsr() logic' should be backported > > to 3.0.y as well? > > No, I don't think -stable really needs this fix. The fix itself is fine > (I hope ;), but the bug is not serious. > > You can apply 9899d11f654474d2d54ea52ceaa2a1f4db3abd68 without the comment > changes in arch/x86/kernel/step.c
Thanks a lot. Ben. -- Ben Hutchings Sturgeon's Law: Ninety percent of everything is crap.
signature.asc
Description: This is a digitally signed message part
