From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
Paul Mackerras <paulus@samba.org>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
PowerPC email list <linuxppc-dev@lists.ozlabs.org>,
Anton Blanchard <anton@samba.org>
Subject: Re: [PATCH powerpc] fix unpaired __trace_hcall_entry and __trace_hcall_exit
Date: Tue, 03 Jan 2012 11:54:17 +1100 [thread overview]
Message-ID: <1325552057.23910.5.camel@pasglop> (raw)
In-Reply-To: <1324260407.3006.17.camel@ThinkPad-T61>
On Mon, 2011-12-19 at 10:06 +0800, Li Zhong wrote:
> Unpaired calling of __trace_hcall_entry and __trace_hcall_exit could
> cause incorrect preempt count. And it might happen as the global
> variable hcall_tracepoint_refcount is checked separately before calling
> them.
>
> I don't know much about the powerpc arch. But the idea here is to store
> the hcall_tracepoint_refcount locally, so __trace_hcall_entry and
> __trace_hcall_exit will be called or not called in pair by checking the
> same value.
Can you re-use an existing spot in the stack frame rather than adding
this entry ? There's plenty of unused spots :-) For example the link
editor doubleword is never going to be used in that function, we could
hijack it safely. It would make the patch (and the code) simpler.
Cheers,
Ben.
> Reported-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> Signed-off-by: Li Zhong <zhong@linux.vnet.ibm.com>
> Tested-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> ---
> arch/powerpc/platforms/pseries/hvCall.S | 20 +++++++++++---------
> 1 files changed, 11 insertions(+), 9 deletions(-)
>
> diff --git a/arch/powerpc/platforms/pseries/hvCall.S b/arch/powerpc/platforms/pseries/hvCall.S
> index fd05fde..1240bd2 100644
> --- a/arch/powerpc/platforms/pseries/hvCall.S
> +++ b/arch/powerpc/platforms/pseries/hvCall.S
> @@ -14,6 +14,7 @@
> #include <asm/ptrace.h>
>
> #define STK_PARM(i) (48 + ((i)-3)*8)
> +#define REG_SIZE (2*8)
>
> #ifdef CONFIG_TRACEPOINTS
>
> @@ -32,11 +33,12 @@ hcall_tracepoint_refcount:
> * unconditional cpu feature.
> */
> #define HCALL_INST_PRECALL(FIRST_REG) \
> + std r31,-8(r1); \
> BEGIN_FTR_SECTION; \
> b 1f; \
> END_FTR_SECTION(0, 1); \
> - ld r12,hcall_tracepoint_refcount@toc(r2); \
> - cmpdi r12,0; \
> + ld r31,hcall_tracepoint_refcount@toc(r2); \
> + cmpdi r31,0; \
> beq+ 1f; \
> mflr r0; \
> std r3,STK_PARM(r3)(r1); \
> @@ -49,9 +51,9 @@ END_FTR_SECTION(0, 1); \
> std r10,STK_PARM(r10)(r1); \
> std r0,16(r1); \
> addi r4,r1,STK_PARM(FIRST_REG); \
> - stdu r1,-STACK_FRAME_OVERHEAD(r1); \
> + stdu r1,-STACK_FRAME_OVERHEAD-REG_SIZE(r1); \
> bl .__trace_hcall_entry; \
> - addi r1,r1,STACK_FRAME_OVERHEAD; \
> + addi r1,r1,STACK_FRAME_OVERHEAD+REG_SIZE; \
> ld r0,16(r1); \
> ld r3,STK_PARM(r3)(r1); \
> ld r4,STK_PARM(r4)(r1); \
> @@ -74,8 +76,7 @@ END_FTR_SECTION(0, 1); \
> BEGIN_FTR_SECTION; \
> b 1f; \
> END_FTR_SECTION(0, 1); \
> - ld r12,hcall_tracepoint_refcount@toc(r2); \
> - cmpdi r12,0; \
> + cmpdi r31,0; \
> beq+ 1f; \
> mflr r0; \
> ld r6,STK_PARM(r3)(r1); \
> @@ -83,13 +84,14 @@ END_FTR_SECTION(0, 1); \
> mr r4,r3; \
> mr r3,r6; \
> std r0,16(r1); \
> - stdu r1,-STACK_FRAME_OVERHEAD(r1); \
> + stdu r1,-STACK_FRAME_OVERHEAD-REG_SIZE(r1); \
> bl .__trace_hcall_exit; \
> - addi r1,r1,STACK_FRAME_OVERHEAD; \
> + addi r1,r1,STACK_FRAME_OVERHEAD+REG_SIZE; \
> ld r0,16(r1); \
> ld r3,STK_PARM(r3)(r1); \
> mtlr r0; \
> -1:
> +1: \
> + ld r31,-8(r1);
>
> #define HCALL_INST_POSTCALL_NORETS \
> li r5,0; \
next prev parent reply other threads:[~2012-01-03 0:54 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-12-19 2:06 [PATCH powerpc] fix unpaired __trace_hcall_entry and __trace_hcall_exit Li Zhong
2012-01-03 0:54 ` Benjamin Herrenschmidt [this message]
2012-01-08 3:54 ` [PATCH v2 " Li Zhong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1325552057.23910.5.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=anton@samba.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=paulus@samba.org \
--cc=zhong@linux.vnet.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox