[5/6] powerpc: Disable VSX or current process in giveup_fpu/altivec

From: Greg KH
Date: Thu Jan 14 2010 - 18:11:35 EST

2.6.27-stable review patch. If anyone has any objections, please let us know.


From: Michael Neuling <mikey@xxxxxxxxxxx>

commit 7e875e9dc8af70d126fa632446e967327ac3fdda upstream.

When we call giveup_fpu, we need to need to turn off VSX for the
current process. If we don't, on return to userspace it may execute a
VSX instruction before the next FP instruction, and not have its
register state refreshed correctly from the thread_struct. Ditto for

This caused a bug where an unaligned lfs or stfs results in
fix_alignment calling giveup_fpu so it can use the FPRs (in order to
do a single <-> double conversion), and then returning to userspace
with FP off but VSX on. Then if a VSX instruction is executed, before
another FP instruction, it will proceed without another exception and
hence have the incorrect register state for VSX registers 0-31.

lfs unaligned <- alignment exception turns FP off but leaves VSX on

VSX instruction <- no exception since VSX on, hence we get the
wrong VSX register values for VSX registers 0-31,
which overlap the FPRs.

Signed-off-by: Michael Neuling <mikey@xxxxxxxxxxx>
Signed-off-by: Paul Mackerras <paulus@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxx>

arch/powerpc/kernel/fpu.S | 5 +++++
arch/powerpc/kernel/misc_64.S | 8 ++++++++
2 files changed, 13 insertions(+)

--- a/arch/powerpc/kernel/fpu.S
+++ b/arch/powerpc/kernel/fpu.S
@@ -145,6 +145,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_VSX)
beq 1f
+#ifdef CONFIG_VSX
+ oris r3,r3,MSR_VSX@h
andc r4,r4,r3 /* disable FP for previous task */
--- a/arch/powerpc/kernel/misc_64.S
+++ b/arch/powerpc/kernel/misc_64.S
@@ -493,7 +493,15 @@ _GLOBAL(giveup_altivec)
stvx vr0,r4,r3
beq 1f
+#ifdef CONFIG_VSX
+ lis r3,(MSR_VEC|MSR_VSX)@h
+ lis r3,MSR_VEC@h
lis r3,MSR_VEC@h
andc r4,r4,r3 /* disable FP for previous task */

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/