Messages in this thread | | | From | Linus Torvalds <> | Date | Wed, 22 Feb 2012 19:37:27 -0800 | Subject | Re: [PATCH 0/5] i387: stable kernel backport |
| |
On Wed, Feb 22, 2012 at 6:41 PM, <raphael@buro.asia> wrote: > > The patchset is simply made of: > be98c2cdb15ba26148cd2bd58a857d4f7759ed38 (unmodified) > 5b1cbac37798805c1fee18c8cebe5c0a13975b17 (") > c38e23456278e967f094b08247ffc3711b1029b2 (") > 15d8791cae75dca27bfda8ecfe87dca9379d6bb0 (") > b6c66418dcad0fcf83cd1d0a39482db37bf4fc41 (") > 6d59d7a9f5b723a7ac1925c136e93ec83c0c3043 (") > b3b0870ef3ffed72b92415423da864f440f57ad6 (") > 4903062b5485f0e2c286a23b44c9b59d9b017d53: this one requires a slight > modification: > -#define safe_address (kstat_cpu(0).cpustat.user) > instead of: > -#define safe_address (__get_cpu_var(kernel_cpustat).cpustat[CPUTIME_USER]) > f94edacf998516ac9d849f7bc6949a703977a7f3 (unmodified) > 34ddc81a230b15c0e345b6b253049db731499f7e (")
Oh, Greg, since that series is tested by Raphael on top of 3.2.6 already, let's just make that be the stable backport.
It does mean that the stable backport will contain that whole "rip out fp state preloading and reimplement it" thing, but hey, considering that when I tried to avoid it I clearly screwed something up, maybe that's all for the best. And perhaps staying closer to the development tree is a good idea anyway in case there are any other issues.
Linus
| |