Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1710734 > unrolled thread

Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions

Started byArnd Bergmann <arnd@arndb.de>
First post2017-08-14 11:00 +0200
Last post2017-08-20 05:20 +0200
Articles 6 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions Arnd Bergmann <arnd@arndb.de> - 2017-08-14 11:00 +0200
    Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions Boqun Feng <boqun.feng@gmail.com> - 2017-08-19 01:50 +0200
      Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions Arnd Bergmann <arnd@arndb.de> - 2017-08-19 15:00 +0200
        Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions Arnd Bergmann <arnd@arndb.de> - 2017-08-19 15:40 +0200
          Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions Boqun Feng <boqun.feng@gmail.com> - 2017-08-23 16:50 +0200
        Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions Boqun Feng <boqun.feng@gmail.com> - 2017-08-20 05:20 +0200

#1710734 — Re: [PATCH v8 09/14] lockdep: Apply crossrelease to completions

FromArnd Bergmann <arnd@arndb.de>
Date2017-08-14 11:00 +0200
SubjectRe: [PATCH v8 09/14] lockdep: Apply crossrelease to completions
Message-ID<uejo5-1Xt-11@gated-at.bofh.it>
On Mon, Aug 7, 2017 at 9:12 AM, Byungchul Park <byungchul.park@lge.com> wrote:
> Although wait_for_completion() and its family can cause deadlock, the
> lock correctness validator could not be applied to them until now,
> because things like complete() are usually called in a different context
> from the waiting context, which violates lockdep's assumption.
>
> Thanks to CONFIG_LOCKDEP_CROSSRELEASE, we can now apply the lockdep
> detector to those completion operations. Applied it.
>
> Signed-off-by: Byungchul Park <byungchul.park@lge.com>

This patch introduced a significant growth in kernel stack usage for a small
set of functions. I see two new warnings for functions that get tipped over the
1024 or 2048 byte frame size limit in linux-next (with a few other patches
applied):

Before:

drivers/md/dm-integrity.c: In function 'write_journal':
drivers/md/dm-integrity.c:827:1: error: the frame size of 504 bytes is
larger than xxx bytes [-Werror=frame-larger-than=]
drivers/mmc/core/mmc_test.c: In function 'mmc_test_area_io_seq':
drivers/mmc/core/mmc_test.c:1491:1: error: the frame size of 680 bytes
is larger than 104 bytes [-Werror=frame-larger-than=]

After:

drivers/md/dm-integrity.c: In function 'write_journal':
drivers/md/dm-integrity.c:827:1: error: the frame size of 1280 bytes
is larger than 1024 bytes [-Werror=frame-larger-than=]
drivers/mmc/core/mmc_test.c: In function 'mmc_test_area_io_seq':
drivers/mmc/core/mmc_test.c:1491:1: error: the frame size of 1072
bytes is larger than 1024 bytes [-Werror=frame-larger-than=]

I have not checked in detail why this happens, but I'm guessing that
there is an overall increase in stack usage with
CONFIG_LOCKDEP_COMPLETE in functions using completions,
and I think it would be good to try to come up with a version that doesn't
add as much.

        Arnd

[toc] | [next] | [standalone]


#1715510

FromBoqun Feng <boqun.feng@gmail.com>
Date2017-08-19 01:50 +0200
Message-ID<ufZbz-25E-1@gated-at.bofh.it>
In reply to#1710734
Hi Arnd,

On Mon, Aug 14, 2017 at 10:50:24AM +0200, Arnd Bergmann wrote:
> On Mon, Aug 7, 2017 at 9:12 AM, Byungchul Park <byungchul.park@lge.com> wrote:
> > Although wait_for_completion() and its family can cause deadlock, the
> > lock correctness validator could not be applied to them until now,
> > because things like complete() are usually called in a different context
> > from the waiting context, which violates lockdep's assumption.
> >
> > Thanks to CONFIG_LOCKDEP_CROSSRELEASE, we can now apply the lockdep
> > detector to those completion operations. Applied it.
> >
> > Signed-off-by: Byungchul Park <byungchul.park@lge.com>
> 
> This patch introduced a significant growth in kernel stack usage for a small
> set of functions. I see two new warnings for functions that get tipped over the
> 1024 or 2048 byte frame size limit in linux-next (with a few other patches
> applied):
> 
> Before:
> 
> drivers/md/dm-integrity.c: In function 'write_journal':
> drivers/md/dm-integrity.c:827:1: error: the frame size of 504 bytes is
> larger than xxx bytes [-Werror=frame-larger-than=]
> drivers/mmc/core/mmc_test.c: In function 'mmc_test_area_io_seq':
> drivers/mmc/core/mmc_test.c:1491:1: error: the frame size of 680 bytes
> is larger than 104 bytes [-Werror=frame-larger-than=]
> 
> After:
> 
> drivers/md/dm-integrity.c: In function 'write_journal':
> drivers/md/dm-integrity.c:827:1: error: the frame size of 1280 bytes
> is larger than 1024 bytes [-Werror=frame-larger-than=]
> drivers/mmc/core/mmc_test.c: In function 'mmc_test_area_io_seq':
> drivers/mmc/core/mmc_test.c:1491:1: error: the frame size of 1072
> bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
> 
> I have not checked in detail why this happens, but I'm guessing that
> there is an overall increase in stack usage with
> CONFIG_LOCKDEP_COMPLETE in functions using completions,
> and I think it would be good to try to come up with a version that doesn't
> add as much.
> 

So I have been staring at this for a while, and below is what I found:

(BTW, Arnd, may I know your compiler version? Mine is 7.1.1)

In write_journal(), I can see the code generated like this on x86:

        io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
    2462:       e8 00 00 00 00          callq  2467 <write_journal+0x47>
    2467:       48 8d 85 80 fd ff ff    lea    -0x280(%rbp),%rax
    246e:       48 c7 c6 00 00 00 00    mov    $0x0,%rsi
    2475:       48 c7 c2 00 00 00 00    mov    $0x0,%rdx
        x->done = 0;
    247c:       c7 85 90 fd ff ff 00    movl   $0x0,-0x270(%rbp)
    2483:       00 00 00 
        init_waitqueue_head(&x->wait);
    2486:       48 8d 78 18             lea    0x18(%rax),%rdi
    248a:       e8 00 00 00 00          callq  248f <write_journal+0x6f>
        if (commit_start + commit_sections <= ic->journal_sections) {
    248f:       41 8b 87 a8 00 00 00    mov    0xa8(%r15),%eax
        io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
    2496:       48 8d bd e8 f9 ff ff    lea    -0x618(%rbp),%rdi
    249d:       48 8d b5 90 fd ff ff    lea    -0x270(%rbp),%rsi
    24a4:       b9 17 00 00 00          mov    $0x17,%ecx
    24a9:       f3 48 a5                rep movsq %ds:(%rsi),%es:(%rdi)
        if (commit_start + commit_sections <= ic->journal_sections) {
    24ac:       41 39 c6                cmp    %eax,%r14d
        io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
    24af:       48 8d bd 90 fd ff ff    lea    -0x270(%rbp),%rdi
    24b6:       48 8d b5 e8 f9 ff ff    lea    -0x618(%rbp),%rsi
    24bd:       b9 17 00 00 00          mov    $0x17,%ecx
    24c2:       f3 48 a5                rep movsq %ds:(%rsi),%es:(%rdi)

Those two "rep movsq"s are very suspicious, because
COMPLETION_INITIALIZER_ONSTACK() should initialize the data in-place,
rather than move it to some temporary variable and copy it back.

I tried to reduce the size of completion struct, and the "rep movsq" did
go away, however it seemed the compiler still allocated the memory for
the temporary variables on the stack, because whenever I
increased/decreased  the size of completion, the stack size of
write_journal() got increased/decreased *7* times, but there are only
3 journal_completion structures in write_journal(). So the *4* callsites
of COMPLETION_INITIALIZER_ONSTACK() looked very suspicous.

So I come up with the following patch, trying to teach the compiler not
to do the unnecessary allocation, could you give it a try?

Besides, I could also observe the stack size reduction of
write_journal() even for !LOCKDEP kernel.

-------------------------->8
Subject: [PATCH] completion: Avoid unnecessary stack allocation for
 COMPLETION_INITIALIZER_ONSTACK()

In theory, COMPLETION_INITIALIZER_ONSTACK() should never affect the
stack allocation of the caller. However, on some compilers, a temporary
structure was allocated for the return value of
COMPLETION_INITIALIZER_ONSTACK(), for example in write_journal() with
LOCKDEP_COMPLETIONS=y(gcc is 7.1.1):

		io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
	    2462:       e8 00 00 00 00          callq  2467 <write_journal+0x47>
	    2467:       48 8d 85 80 fd ff ff    lea    -0x280(%rbp),%rax
	    246e:       48 c7 c6 00 00 00 00    mov    $0x0,%rsi
	    2475:       48 c7 c2 00 00 00 00    mov    $0x0,%rdx
		x->done = 0;
	    247c:       c7 85 90 fd ff ff 00    movl   $0x0,-0x270(%rbp)
	    2483:       00 00 00
		init_waitqueue_head(&x->wait);
	    2486:       48 8d 78 18             lea    0x18(%rax),%rdi
	    248a:       e8 00 00 00 00          callq  248f <write_journal+0x6f>
		if (commit_start + commit_sections <= ic->journal_sections) {
	    248f:       41 8b 87 a8 00 00 00    mov    0xa8(%r15),%eax
		io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
	    2496:       48 8d bd e8 f9 ff ff    lea    -0x618(%rbp),%rdi
	    249d:       48 8d b5 90 fd ff ff    lea    -0x270(%rbp),%rsi
	    24a4:       b9 17 00 00 00          mov    $0x17,%ecx
	    24a9:       f3 48 a5                rep movsq %ds:(%rsi),%es:(%rdi)
		if (commit_start + commit_sections <= ic->journal_sections) {
	    24ac:       41 39 c6                cmp    %eax,%r14d
		io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
	    24af:       48 8d bd 90 fd ff ff    lea    -0x270(%rbp),%rdi
	    24b6:       48 8d b5 e8 f9 ff ff    lea    -0x618(%rbp),%rsi
	    24bd:       b9 17 00 00 00          mov    $0x17,%ecx
	    24c2:       f3 48 a5                rep movsq %ds:(%rsi),%es:(%rdi)

We can obviously see the temporary structure allocated, and the compiler
also does two meaningless memcpy with "rep movsq".

To fix this, make the brace block in COMPLETION_INITIALIZER_ONSTACK()
return a pointer and dereference it outside the block rather than return
the whole structure, in this way, we are able to teach the compiler not
to do the unnecessary stack allocation.

This could also reduce the stack size even if !LOCKDEP, for example in
write_journal(), compiled with gcc 7.1.1, the result of command:

	 objdump -d drivers/md/dm-integrity.o | ./scripts/checkstack.pl x86

before:

	0x0000246a write_journal [dm-integrity.o]:              696

after:

	0x00002b7a write_journal [dm-integrity.o]:              296

Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
---
 include/linux/completion.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/completion.h b/include/linux/completion.h
index 791f053f28b7..cae5400022a3 100644
--- a/include/linux/completion.h
+++ b/include/linux/completion.h
@@ -74,7 +74,7 @@ static inline void complete_release_commit(struct completion *x) {}
 #endif
 
 #define COMPLETION_INITIALIZER_ONSTACK(work) \
-	({ init_completion(&work); work; })
+	(*({ init_completion(&work); &work; }))
 
 /**
  * DECLARE_COMPLETION - declare and initialize a completion structure
-- 
2.14.1

[toc] | [prev] | [next] | [standalone]


#1715664

FromArnd Bergmann <arnd@arndb.de>
Date2017-08-19 15:00 +0200
Message-ID<ugbw6-1pj-9@gated-at.bofh.it>
In reply to#1715510
On Sat, Aug 19, 2017 at 1:43 AM, Boqun Feng <boqun.feng@gmail.com> wrote:
> Hi Arnd,
>
> On Mon, Aug 14, 2017 at 10:50:24AM +0200, Arnd Bergmann wrote:
>> On Mon, Aug 7, 2017 at 9:12 AM, Byungchul Park <byungchul.park@lge.com> wrote:
>> > Although wait_for_completion() and its family can cause deadlock, the
>> > lock correctness validator could not be applied to them until now,
>> > because things like complete() are usually called in a different context
>> > from the waiting context, which violates lockdep's assumption.
>> >
>> > Thanks to CONFIG_LOCKDEP_CROSSRELEASE, we can now apply the lockdep
>> > detector to those completion operations. Applied it.
>> >
>> > Signed-off-by: Byungchul Park <byungchul.park@lge.com>
>>
>> This patch introduced a significant growth in kernel stack usage for a small
>> set of functions. I see two new warnings for functions that get tipped over the
>> 1024 or 2048 byte frame size limit in linux-next (with a few other patches
>> applied):
>>
>> Before:
>>
>> drivers/md/dm-integrity.c: In function 'write_journal':
>> drivers/md/dm-integrity.c:827:1: error: the frame size of 504 bytes is
>> larger than xxx bytes [-Werror=frame-larger-than=]
>> drivers/mmc/core/mmc_test.c: In function 'mmc_test_area_io_seq':
>> drivers/mmc/core/mmc_test.c:1491:1: error: the frame size of 680 bytes
>> is larger than 104 bytes [-Werror=frame-larger-than=]
>>
>> After:
>>
>> drivers/md/dm-integrity.c: In function 'write_journal':
>> drivers/md/dm-integrity.c:827:1: error: the frame size of 1280 bytes
>> is larger than 1024 bytes [-Werror=frame-larger-than=]
>> drivers/mmc/core/mmc_test.c: In function 'mmc_test_area_io_seq':
>> drivers/mmc/core/mmc_test.c:1491:1: error: the frame size of 1072
>> bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
>>
>> I have not checked in detail why this happens, but I'm guessing that
>> there is an overall increase in stack usage with
>> CONFIG_LOCKDEP_COMPLETE in functions using completions,
>> and I think it would be good to try to come up with a version that doesn't
>> add as much.
>>
>
> So I have been staring at this for a while, and below is what I found:
>
> (BTW, Arnd, may I know your compiler version? Mine is 7.1.1)

That is what I used as well, on x86, arm32 and arm64.

> In write_journal(), I can see the code generated like this on x86:
>
>         io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
>     2462:       e8 00 00 00 00          callq  2467 <write_journal+0x47>
>     2467:       48 8d 85 80 fd ff ff    lea    -0x280(%rbp),%rax
>     246e:       48 c7 c6 00 00 00 00    mov    $0x0,%rsi
>     2475:       48 c7 c2 00 00 00 00    mov    $0x0,%rdx
>         x->done = 0;
>     247c:       c7 85 90 fd ff ff 00    movl   $0x0,-0x270(%rbp)
>     2483:       00 00 00
>         init_waitqueue_head(&x->wait);
>     2486:       48 8d 78 18             lea    0x18(%rax),%rdi
>     248a:       e8 00 00 00 00          callq  248f <write_journal+0x6f>
>         if (commit_start + commit_sections <= ic->journal_sections) {
>     248f:       41 8b 87 a8 00 00 00    mov    0xa8(%r15),%eax
>         io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
>     2496:       48 8d bd e8 f9 ff ff    lea    -0x618(%rbp),%rdi
>     249d:       48 8d b5 90 fd ff ff    lea    -0x270(%rbp),%rsi
>     24a4:       b9 17 00 00 00          mov    $0x17,%ecx
>     24a9:       f3 48 a5                rep movsq %ds:(%rsi),%es:(%rdi)
>         if (commit_start + commit_sections <= ic->journal_sections) {
>     24ac:       41 39 c6                cmp    %eax,%r14d
>         io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp);
>     24af:       48 8d bd 90 fd ff ff    lea    -0x270(%rbp),%rdi
>     24b6:       48 8d b5 e8 f9 ff ff    lea    -0x618(%rbp),%rsi
>     24bd:       b9 17 00 00 00          mov    $0x17,%ecx
>     24c2:       f3 48 a5                rep movsq %ds:(%rsi),%es:(%rdi)
>
> Those two "rep movsq"s are very suspicious, because
> COMPLETION_INITIALIZER_ONSTACK() should initialize the data in-place,
> rather than move it to some temporary variable and copy it back.

Right. I've seen this behavior before when using c99 compound
literals, but I was surprised to see it here.

I also submitted a patch for the one driver that turned up a new
warning because of this behavior:

https://www.spinics.net/lists/raid/msg58766.html

In case of the mmc driver, the behavior was as expected, it was
just a little too large and I sent the obvious workaround for it

https://patchwork.kernel.org/patch/9902063/

> I tried to reduce the size of completion struct, and the "rep movsq" did
> go away, however it seemed the compiler still allocated the memory for
> the temporary variables on the stack, because whenever I
> increased/decreased  the size of completion, the stack size of
> write_journal() got increased/decreased *7* times, but there are only
> 3 journal_completion structures in write_journal(). So the *4* callsites
> of COMPLETION_INITIALIZER_ONSTACK() looked very suspicous.
>
> So I come up with the following patch, trying to teach the compiler not
> to do the unnecessary allocation, could you give it a try?
>
> Besides, I could also observe the stack size reduction of
> write_journal() even for !LOCKDEP kernel.

Ok.

> -------------------
> Reported-by: Arnd Bergmann <arnd@arndb.de>
> Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
> ---
>  include/linux/completion.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/include/linux/completion.h b/include/linux/completion.h
> index 791f053f28b7..cae5400022a3 100644
> --- a/include/linux/completion.h
> +++ b/include/linux/completion.h
> @@ -74,7 +74,7 @@ static inline void complete_release_commit(struct completion *x) {}
>  #endif
>
>  #define COMPLETION_INITIALIZER_ONSTACK(work) \
> -       ({ init_completion(&work); work; })
> +       (*({ init_completion(&work); &work; }))
>
>  /**
>   * DECLARE_COMPLETION - declare and initialize a completion structure

Nice hack. Any idea why that's different to the compiler?

I've applied that one to my test tree now, and reverted my own patch,
will let you know if anything else shows up. I think we probably want
to merge both patches to mainline.

      Arnd

[toc] | [prev] | [next] | [standalone]


#1715670

FromArnd Bergmann <arnd@arndb.de>
Date2017-08-19 15:40 +0200
Message-ID<ugc8O-1SQ-5@gated-at.bofh.it>
In reply to#1715664
On Sat, Aug 19, 2017 at 2:51 PM, Arnd Bergmann <arnd@arndb.de> wrote:

>> --- a/include/linux/completion.h
>> +++ b/include/linux/completion.h
>> @@ -74,7 +74,7 @@ static inline void complete_release_commit(struct completion *x) {}
>>  #endif
>>
>>  #define COMPLETION_INITIALIZER_ONSTACK(work) \
>> -       ({ init_completion(&work); work; })
>> +       (*({ init_completion(&work); &work; }))
>>
>>  /**
>>   * DECLARE_COMPLETION - declare and initialize a completion structure
>
> Nice hack. Any idea why that's different to the compiler?
>
> I've applied that one to my test tree now, and reverted my own patch,
> will let you know if anything else shows up. I think we probably want
> to merge both patches to mainline.

There is apparently one user of COMPLETION_INITIALIZER_ONSTACK
that causes a regression with the patch above:

drivers/acpi/nfit/core.c: In function 'acpi_nfit_flush_probe':
include/linux/completion.h:77:3: error: value computed is not used
[-Werror=unused-value]
  (*({ init_completion(&work); &work; }))

It would be trivial to convert to init_completion(), which seems to be
what was intended there.

        Arnd

[toc] | [prev] | [next] | [standalone]


#1718426

FromBoqun Feng <boqun.feng@gmail.com>
Date2017-08-23 16:50 +0200
Message-ID<uhF8K-1Mw-17@gated-at.bofh.it>
In reply to#1715670

[Multipart message — attachments visible in raw view] — view raw

On Sat, Aug 19, 2017 at 03:34:01PM +0200, Arnd Bergmann wrote:
> On Sat, Aug 19, 2017 at 2:51 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> 
> >> --- a/include/linux/completion.h
> >> +++ b/include/linux/completion.h
> >> @@ -74,7 +74,7 @@ static inline void complete_release_commit(struct completion *x) {}
> >>  #endif
> >>
> >>  #define COMPLETION_INITIALIZER_ONSTACK(work) \
> >> -       ({ init_completion(&work); work; })
> >> +       (*({ init_completion(&work); &work; }))
> >>
> >>  /**
> >>   * DECLARE_COMPLETION - declare and initialize a completion structure
> >
> > Nice hack. Any idea why that's different to the compiler?
> >

So I find this link:

	https://gcc.gnu.org/onlinedocs/gcc/Statement-Exprs.html

it says:

"In G++, the result value of a statement expression undergoes array and
function pointer decay, and is returned by value to the enclosing
expression. "

I think this is why the temporary variable is constructed(or at least
allocated). Lemme put this in my commit log.

> > I've applied that one to my test tree now, and reverted my own patch,
> > will let you know if anything else shows up. I think we probably want
> > to merge both patches to mainline.
> 
> There is apparently one user of COMPLETION_INITIALIZER_ONSTACK
> that causes a regression with the patch above:
> 
> drivers/acpi/nfit/core.c: In function 'acpi_nfit_flush_probe':
> include/linux/completion.h:77:3: error: value computed is not used
> [-Werror=unused-value]
>   (*({ init_completion(&work); &work; }))
> 
> It would be trivial to convert to init_completion(), which seems to be
> what was intended there.
> 

Thanks. Will send the conversion as a separate patch along with my
patch.

Regards,
Boqun

>         Arnd

[toc] | [prev] | [next] | [standalone]


#1715858

FromBoqun Feng <boqun.feng@gmail.com>
Date2017-08-20 05:20 +0200
Message-ID<ugoWl-1tG-11@gated-at.bofh.it>
In reply to#1715664

[Multipart message — attachments visible in raw view] — view raw

On Sat, Aug 19, 2017 at 02:51:17PM +0200, Arnd Bergmann wrote:
[...]
> > Those two "rep movsq"s are very suspicious, because
> > COMPLETION_INITIALIZER_ONSTACK() should initialize the data in-place,
> > rather than move it to some temporary variable and copy it back.
> 
> Right. I've seen this behavior before when using c99 compound
> literals, but I was surprised to see it here.
> 
> I also submitted a patch for the one driver that turned up a new
> warning because of this behavior:
> 
> https://www.spinics.net/lists/raid/msg58766.html
> 

This solution also came up into my mind but then I found there are
several callsites of COMPLETION_INITIALIZER_ONSTACK(), so I then tried
to find a way to fix the macro itself. But your patch looks good to me
;-)

> In case of the mmc driver, the behavior was as expected, it was
> just a little too large and I sent the obvious workaround for it
> 
> https://patchwork.kernel.org/patch/9902063/
> 

Yep.

> > I tried to reduce the size of completion struct, and the "rep movsq" did
> > go away, however it seemed the compiler still allocated the memory for
> > the temporary variables on the stack, because whenever I
> > increased/decreased  the size of completion, the stack size of
> > write_journal() got increased/decreased *7* times, but there are only
> > 3 journal_completion structures in write_journal(). So the *4* callsites
> > of COMPLETION_INITIALIZER_ONSTACK() looked very suspicous.
> >
> > So I come up with the following patch, trying to teach the compiler not
> > to do the unnecessary allocation, could you give it a try?
> >
> > Besides, I could also observe the stack size reduction of
> > write_journal() even for !LOCKDEP kernel.
> 
> Ok.
> 
> > -------------------
> > Reported-by: Arnd Bergmann <arnd@arndb.de>
> > Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
> > ---
> >  include/linux/completion.h | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/include/linux/completion.h b/include/linux/completion.h
> > index 791f053f28b7..cae5400022a3 100644
> > --- a/include/linux/completion.h
> > +++ b/include/linux/completion.h
> > @@ -74,7 +74,7 @@ static inline void complete_release_commit(struct completion *x) {}
> >  #endif
> >
> >  #define COMPLETION_INITIALIZER_ONSTACK(work) \
> > -       ({ init_completion(&work); work; })
> > +       (*({ init_completion(&work); &work; }))
> >
> >  /**
> >   * DECLARE_COMPLETION - declare and initialize a completion structure
> 
> Nice hack. Any idea why that's different to the compiler?
> 

So *I think* the block {init_completion(&work); &work;} now will return
a pointer rather than a whole structure, and a pointer could fit in a
register, so the compiler won't bother to allocate the memory for it.

> I've applied that one to my test tree now, and reverted my own patch,
> will let you know if anything else shows up. I think we probably want

Thanks ;-)

> to merge both patches to mainline.
> 

Agreed! Unless we want to remove COMPLETION_INITIALIZER_ONSTACK() for
some reason, then my patch is not needed.

Regards,
Boqun

>       Arnd

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web