x86 fixes

Linus,

Please pull the latest x86-urgent-for-linus git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git x86-urgent-for-linus

Thanks,

Ingo

------------------>
Avi Kivity (1):
x86: Fix mwait_play_dead() faulting on mwait-incapable cpus

Borislav Petkov (1):
x86 idle: Fix mwait deprecation warning message

 arch/x86/kernel/process.c |    2 +-
 arch/x86/kernel/smpboot.c |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

index 426a5b6..3d83a71 100644
@@ -642,7 +642,7 @@ static int __init idle_setup(char *str)
 		boot_option_idle_override = IDLE_POLL;
 	} else if (!strcmp(str, "mwait")) {
 		boot_option_idle_override = IDLE_FORCE_MWAIT;
-		WARN_ONCE(1, "\idle=mwait\" will be removed in 2012\"\n");
+		WARN_ONCE(1, "\"idle=mwait\" will be removed in 2012\"\n");
 	} else if (!strcmp(str, "halt")) {
 		/*
 		 * When the boot option of idle=halt is added, halt is

index eefd967..33a0c11 100644
@@ -1332,7 +1332,7 @@ static inline void mwait_play_dead(void)
 	void *mwait_ptr;
 	struct cpuinfo_x86 *c = __this_cpu_ptr(&cpu_info);
 
-	if (!this_cpu_has(X86_FEATURE_MWAIT) && mwait_usable(c))
+	if (!(this_cpu_has(X86_FEATURE_MWAIT) && mwait_usable(c)))
 		return;
 	if (!this_cpu_has(X86_FEATURE_CLFLSH))
 		return;

Crap, there's no fixing this string :)

Reportedly, there's another fix from Len:

lkml.kernel.org/r/[email protected]

which is more complete. Ingo, you could drop this one.

Thanks.

Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Gemeinde Aschheim, Landkreis Muenchen
Registergericht Muenchen, HRB Nr. 43632

Well, the primary problem, which needed x86/urgent treatment, was the annoying build warning affecting all users.

The message itself is exceedingly rare, so we can fix it via lower prio pathways.

Len, a small detail: mind resolving the conflict once it occurs by dropping the \" but keeping the \n? Your commit:

8a27ef90237f: x86 idle: fix build warning for WARN_ONCE("idle=mwait")

removes the \n from the end of a printk-alike string - we generally keep the \n even though WARN_*() will accept \n-less strings.

Thanks,

Ingo

Btw, I went and caused it for funsies and I don't think anyone would concentrate on the one-too-many quotes at the end when she/he sees the below :>)

[    0.000000] ------------[ cut here ]------------
[    0.000000] WARNING: at arch/x86/kernel/process.c:645 idle_setup+0x90/0xed()
[    0.000000] "idle=mwait" will be removed in 2012"
[    0.000000] Modules linked in:
[    0.000000] Pid: 0, comm: swapper Not tainted 3.0.0-rc1+ #72
[    0.000000] Call Trace:
[    0.000000]  [<ffffffff8103b63e>] warn_slowpath_common+0x85/0x9d
[    0.000000]  [<ffffffff8103b6f9>] warn_slowpath_fmt+0x46/0x48
[    0.000000]  [<ffffffff81b4ea06>] idle_setup+0x90/0xed
[    0.000000]  [<ffffffff81b48b54>] do_early_param+0x5d/0x89
[    0.000000]  [<ffffffff810559a4>] parse_args+0x17d/0x241
[    0.000000]  [<ffffffff81b48af7>] ? start_kernel+0x2ce/0x2ce
[    0.000000]  [<ffffffff81b484a5>] parse_early_options+0x1e/0x20
[    0.000000]  [<ffffffff81b4881d>] parse_early_param+0x31/0x3d
[    0.000000]  [<ffffffff81b4a9e7>] setup_arch+0x354/0x733
[    0.000000]  [<ffffffff81480999>] ? printk+0x41/0x48
[    0.000000]  [<ffffffff810633b1>] ? clockevents_register_notifier+0x43/0x4f
[    0.000000]  [<ffffffff81b488b8>] start_kernel+0x8f/0x2ce
[    0.000000]  [<ffffffff81b48272>] x86_64_start_reservations+0x82/0x86
[    0.000000]  [<ffffffff81b48363>] x86_64_start_kernel+0xed/0xf4
[    0.000000] ---[ end trace 4eaa2a86a8e2da22 ]---

Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Gemeinde Aschheim, Landkreis Muenchen
Registergericht Muenchen, HRB Nr. 43632

I tried to fix it up in the merge. We call these "evil merges" in git-speak, but I hope the problem was worth doing one ;)

Linus

You have not pushed it out yet, right? [ Or if you have then the evilness is well hidden! :-) ]

Thanks,

Ingo

I'd never noticed one of these before and had to
look myself.

git show af0d6a0a3a30946f7df69c764791f1b0643f7cd6

Oh yeah, looks good: af0d6a0a3a30946f7df69c764791f1b0643f7cd6

Thank you Linus for being evil! :-)

Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Gemeinde Aschheim, Landkreis Muenchen
Registergericht Muenchen, HRB Nr. 43632

Oh, you did push it out, i just forgot to merge it in :)

I don't think it's a truly evil merge if it's declared in the merge commit log - but yeah, it's outside the scope given by the two trees so technically it's evil :)

Thanks,

Ingo