[PATCH] powerpc/process: Fix flush_all_to_thread for SPE

Subsystems: linux for powerpc (32-bit and 64-bit), the rest

STALE2875d

2 messages, 2 authors, 2018-10-31 · open the first message on its own page

[PATCH] powerpc/process: Fix flush_all_to_thread for SPE

From: DATACOM - Felipe.Rechia <hidden>
Date: 2018-10-25 14:02:09

From: "Felipe Rechia" <redacted>
Date: Wed, 24 Oct 2018 10:57:22 -0300
Subject: [PATCH] powerpc/process: Fix flush_all_to_thread for SPE

Fix a bug introduced by the creation of flush_all_to_thread() for
processors that have SPE (Signal Processing Engine) and use it to
compute floating-point operations.

From userspace perspective, the problem was seen in attempts of
computing floating-point operations which should generate exceptions.
For example:

  fork();
  float x = 0.0 / 0.0;
  isnan(x);           // forked process returns False (should be True)

The operation above also should always cause the SPEFSCR FINV bit to
be set. However, the SPE floating-point exceptions were turned off
after a fork().

Kernel versions prior to the bug used flush_spe_to_thread(), which
first saves SPEFSCR register values in tsk->thread and then calls
giveup_spe(tsk).

After commit 579e633e764e, the save_all() function was called first
to giveup_spe(), and then the SPEFSCR register values were saved in
tsk->thread. This would save the SPEFSCR register values after
disabling SPE for that thread, causing the bug described above.

Fixes 579e633e764e ("powerpc: create flush_all_to_thread()")
Signed-off-by: felipe.rechia <redacted>
---
 arch/powerpc/kernel/process.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index a0c74bb..16eb428 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -566,12 +566,11 @@ void flush_all_to_thread(struct task_struct *tsk)
 	if (tsk->thread.regs) {
 		preempt_disable();
 		BUG_ON(tsk != current);
-		save_all(tsk);
-
 #ifdef CONFIG_SPE
 		if (tsk->thread.regs->msr & MSR_SPE)
 			tsk->thread.spefscr = mfspr(SPRN_SPEFSCR);
 #endif
+		save_all(tsk);
 
 		preempt_enable();
 	}
-- 
2.7.4

Re: powerpc/process: Fix flush_all_to_thread for SPE

From: Michael Ellerman <hidden>
Date: 2018-10-31 05:58:38

On Thu, 2018-10-25 at 12:07:54 UTC, "DATACOM - Felipe.Rechia" wrote:
From: "Felipe Rechia" <redacted>
Date: Wed, 24 Oct 2018 10:57:22 -0300
Subject: [PATCH] powerpc/process: Fix flush_all_to_thread for SPE

Fix a bug introduced by the creation of flush_all_to_thread() for
processors that have SPE (Signal Processing Engine) and use it to
compute floating-point operations.
quoted
From userspace perspective, the problem was seen in attempts of
computing floating-point operations which should generate exceptions.
For example:

  fork();
  float x = 0.0 / 0.0;
  isnan(x);           // forked process returns False (should be True)

The operation above also should always cause the SPEFSCR FINV bit to
be set. However, the SPE floating-point exceptions were turned off
after a fork().

Kernel versions prior to the bug used flush_spe_to_thread(), which
first saves SPEFSCR register values in tsk->thread and then calls
giveup_spe(tsk).

After commit 579e633e764e, the save_all() function was called first
to giveup_spe(), and then the SPEFSCR register values were saved in
tsk->thread. This would save the SPEFSCR register values after
disabling SPE for that thread, causing the bug described above.

Fixes 579e633e764e ("powerpc: create flush_all_to_thread()")
Signed-off-by: felipe.rechia <redacted>
Applied to powerpc next, thanks.

https://git.kernel.org/powerpc/c/e901378578c62202594cba0f6c076f

cheers
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help