diff mbox

[1/2] powerpc: Send SIGBUS on unaligned copy and paste

Message ID 20160609060120.GA26093@distroguy.com (mailing list archive)
State Changes Requested
Headers show

Commit Message

Chris Smart June 9, 2016, 6:01 a.m. UTC
Calling ISA 3.0 instructions copy, copy_first, paste and paste_last
generates an alignment fault when copying or pasting unaligned
data (128 byte). We catch this and send SIGBUS to the userspace
process that caused it.

We do not emulate these because paste may contain additional metadata
when pasting to a co-processor and paste_last is the synchronisation
point for preceding copy/paste sequences.

Thanks to Michael Neuling <mikey@neuling.org> for his help.

Signed-off-by: Chris Smart <chris@distroguy.com>
---
 arch/powerpc/kernel/align.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

Comments

Christophe Leroy June 9, 2016, 6:57 a.m. UTC | #1
Le 09/06/2016 à 08:01, Chris Smart a écrit :
> Calling ISA 3.0 instructions copy, copy_first, paste and paste_last
> generates an alignment fault when copying or pasting unaligned
> data (128 byte). We catch this and send SIGBUS to the userspace
> process that caused it.
>
> We do not emulate these because paste may contain additional metadata
> when pasting to a co-processor and paste_last is the synchronisation
> point for preceding copy/paste sequences.
>
> Thanks to Michael Neuling <mikey@neuling.org> for his help.
>
> Signed-off-by: Chris Smart <chris@distroguy.com>
> ---
> arch/powerpc/kernel/align.c | 15 +++++++++++++++
> 1 file changed, 15 insertions(+)
>
> diff --git a/arch/powerpc/kernel/align.c b/arch/powerpc/kernel/align.c
> index 8e7cb8e2b21a..fcdf09476eec 100644
> --- a/arch/powerpc/kernel/align.c
> +++ b/arch/powerpc/kernel/align.c
> @@ -875,6 +875,21 @@ int fix_alignment(struct pt_regs *regs)
>         return emulate_vsx(addr, reg, areg, regs, flags, nb, elsize);
>     }
> #endif
> +
> +    /*
> +     * ISA 3.0 (such as P9) copy, copy_first, paste and paste_last 
> alignment
> +     * check.
> +     *
> +     * Send a SIGBUS to the process that caused the fault.
> +     *
> +     * We do not emulate these because paste may contain additional 
> metadata
> +     * when pasting to a co-processor. Furthermore, paste_last is the
> +     * synchronisation point for preceding copy/paste sequences.
> +     */
> +    if ((instruction & 0xfc0006fe) == 0x7c00060c){

Can you define corresponding instruction and mask in 
arch/powerpc/include/asm/ppc-opcode.h
and use symbolic names here instead of raw values ?

Christophe
Chris Smart June 14, 2016, 1:50 a.m. UTC | #2
On Thu, Jun 09, 2016 at 08:57:50AM +0200, Christophe Leroy wrote:
>
>Le 09/06/2016 à 08:01, Chris Smart a écrit :
>>Calling ISA 3.0 instructions copy, copy_first, paste and paste_last
>>generates an alignment fault when copying or pasting unaligned
>>data (128 byte). We catch this and send SIGBUS to the userspace
>>process that caused it.
>>

[snip]

>>+    if ((instruction & 0xfc0006fe) == 0x7c00060c){
>
>Can you define corresponding instruction and mask in 
>arch/powerpc/include/asm/ppc-opcode.h
>and use symbolic names here instead of raw values ?
>

Yep, thanks. Will look at doing this for V2.

Cheers,
-c
diff mbox

Patch

diff --git a/arch/powerpc/kernel/align.c b/arch/powerpc/kernel/align.c
index 8e7cb8e2b21a..fcdf09476eec 100644
--- a/arch/powerpc/kernel/align.c
+++ b/arch/powerpc/kernel/align.c
@@ -875,6 +875,21 @@  int fix_alignment(struct pt_regs *regs)
 		return emulate_vsx(addr, reg, areg, regs, flags, nb, elsize);
 	}
 #endif
+
+	/*
+	 * ISA 3.0 (such as P9) copy, copy_first, paste and paste_last alignment
+	 * check.
+	 *
+	 * Send a SIGBUS to the process that caused the fault.
+	 *
+	 * We do not emulate these because paste may contain additional metadata
+	 * when pasting to a co-processor. Furthermore, paste_last is the
+	 * synchronisation point for preceding copy/paste sequences.
+	 */
+	if ((instruction & 0xfc0006fe) == 0x7c00060c){
+		return -EIO;
+	}
+
 	/* A size of 0 indicates an instruction we don't support, with
 	 * the exception of DCBZ which is handled as a special case here
 	 */