With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66122
In particular,
with this config: http://busybox.net/~vda/kernel_config
there are more than a thousand copies of tiny
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66122
In particular,
with this config: http://busybox.net/~vda/kernel_config
there are more than a thousand copies of tiny
On Wed, May 13, 2015 at 04:09:18PM +0200, Denys Vlasenko wrote:
> On 05/13/2015 12:43 PM, Ingo Molnar wrote:
> > We only know that the net effect is +70 bytes. Does that come out of:
> >
> > - large fluctuations such as -1000-1000+1000+1070, which happens to
> >net out into a small net
On Wed, May 13, 2015 at 04:09:18PM +0200, Denys Vlasenko wrote:
On 05/13/2015 12:43 PM, Ingo Molnar wrote:
We only know that the net effect is +70 bytes. Does that come out of:
- large fluctuations such as -1000-1000+1000+1070, which happens to
net out into a small net number?
On 05/13/2015 12:43 PM, Ingo Molnar wrote:
>
> * Denys Vlasenko wrote:
>
>> On 05/13/2015 12:17 PM, Ingo Molnar wrote:
> In any case, the interesting measurement would not be -Os comparisons
> (which causes GCC to be too crazy), but to see the size effect of your
> _patch_ that
* Denys Vlasenko wrote:
> On 05/13/2015 12:17 PM, Ingo Molnar wrote:
> >>> In any case, the interesting measurement would not be -Os comparisons
> >>> (which causes GCC to be too crazy), but to see the size effect of your
> >>> _patch_ that always-inlines spinlock ops, on plain defconfig and
On 05/13/2015 12:17 PM, Ingo Molnar wrote:
>>> In any case, the interesting measurement would not be -Os comparisons
>>> (which causes GCC to be too crazy), but to see the size effect of your
>>> _patch_ that always-inlines spinlock ops, on plain defconfig and on
>>> defconfig-Os.
>>
>> Here it
* Denys Vlasenko wrote:
> On 05/12/2015 01:43 PM, Ingo Molnar wrote:
> >>> Furtermore, what is the size win on x86 defconfig with these options
> >>> set?
> >>
> >> CONFIG_OPTIMIZE_INLINING=y is in defconfig.
> >>
> >> Size difference for CC_OPTIMIZE_FOR_SIZE:
> >>
> >> textdata
* Denys Vlasenko dvlas...@redhat.com wrote:
On 05/13/2015 12:17 PM, Ingo Molnar wrote:
In any case, the interesting measurement would not be -Os comparisons
(which causes GCC to be too crazy), but to see the size effect of your
_patch_ that always-inlines spinlock ops, on plain
On 05/13/2015 12:17 PM, Ingo Molnar wrote:
In any case, the interesting measurement would not be -Os comparisons
(which causes GCC to be too crazy), but to see the size effect of your
_patch_ that always-inlines spinlock ops, on plain defconfig and on
defconfig-Os.
Here it is:
text
* Denys Vlasenko dvlas...@redhat.com wrote:
On 05/12/2015 01:43 PM, Ingo Molnar wrote:
Furtermore, what is the size win on x86 defconfig with these options
set?
CONFIG_OPTIMIZE_INLINING=y is in defconfig.
Size difference for CC_OPTIMIZE_FOR_SIZE:
textdata bss
On 05/13/2015 12:43 PM, Ingo Molnar wrote:
* Denys Vlasenko dvlas...@redhat.com wrote:
On 05/13/2015 12:17 PM, Ingo Molnar wrote:
In any case, the interesting measurement would not be -Os comparisons
(which causes GCC to be too crazy), but to see the size effect of your
_patch_ that
On 05/12/2015 01:43 PM, Ingo Molnar wrote:
>>> Furtermore, what is the size win on x86 defconfig with these options
>>> set?
>>
>> CONFIG_OPTIMIZE_INLINING=y is in defconfig.
>>
>> Size difference for CC_OPTIMIZE_FOR_SIZE:
>>
>> textdata bss dechex filename
>> 12335864
* Denys Vlasenko wrote:
> On 05/12/2015 09:44 AM, Ingo Molnar wrote:
> >
> > * Denys Vlasenko wrote:
> >
> >> With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
> >> very small functions we expect to be inlined. In particular,
> >> with this config:
On 05/12/2015 09:44 AM, Ingo Molnar wrote:
>
> * Denys Vlasenko wrote:
>
>> With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
>> very small functions we expect to be inlined. In particular,
>> with this config: http://busybox.net/~vda/kernel_config
>> there are more than
* Denys Vlasenko wrote:
> > Also, the inline_for_broken_gcc definition can be made dependent
> > on particular gcc versions, which will allow us to easily keep an
> > eye on the behaviour of later gcc versions.
>
> I've seen it on gcc-4.7.2 and gcc-4.9.2, so this behavior is not
> limited
On 05/12/2015 12:19 AM, Andrew Morton wrote:
> On Mon, 11 May 2015 19:57:22 +0200 Denys Vlasenko wrote:
>
>> With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
>> very small functions we expect to be inlined. In particular,
>> with this config:
* Andrew Morton | 2015-05-11 15:19:13 [-0700]:
>Presumably Hagen didn't see the issue with spinlock functions. I
>wonder why not.
I think it is a compiler version thing. Not sure why I didn't see it.
>I suppose we should get both these consolidated into a coherent whole.
+1 (let wait for a
* Denys Vlasenko wrote:
> With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
> very small functions we expect to be inlined. In particular,
> with this config: http://busybox.net/~vda/kernel_config
> there are more than a thousand copies of tiny spinlock-related functions:
* Denys Vlasenko dvlas...@redhat.com wrote:
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this config: http://busybox.net/~vda/kernel_config
there are more than a thousand copies of tiny
* Andrew Morton | 2015-05-11 15:19:13 [-0700]:
Presumably Hagen didn't see the issue with spinlock functions. I
wonder why not.
I think it is a compiler version thing. Not sure why I didn't see it.
I suppose we should get both these consolidated into a coherent whole.
+1 (let wait for a
* Denys Vlasenko dvlas...@redhat.com wrote:
Also, the inline_for_broken_gcc definition can be made dependent
on particular gcc versions, which will allow us to easily keep an
eye on the behaviour of later gcc versions.
I've seen it on gcc-4.7.2 and gcc-4.9.2, so this behavior is not
On 05/12/2015 09:44 AM, Ingo Molnar wrote:
* Denys Vlasenko dvlas...@redhat.com wrote:
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this config: http://busybox.net/~vda/kernel_config
there are
* Denys Vlasenko dvlas...@redhat.com wrote:
On 05/12/2015 09:44 AM, Ingo Molnar wrote:
* Denys Vlasenko dvlas...@redhat.com wrote:
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this
On 05/12/2015 12:19 AM, Andrew Morton wrote:
On Mon, 11 May 2015 19:57:22 +0200 Denys Vlasenko dvlas...@redhat.com wrote:
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this config:
On 05/12/2015 01:43 PM, Ingo Molnar wrote:
Furtermore, what is the size win on x86 defconfig with these options
set?
CONFIG_OPTIMIZE_INLINING=y is in defconfig.
Size difference for CC_OPTIMIZE_FOR_SIZE:
textdata bss dechex filename
12335864 1746152 1081344 15163360
On Mon, 11 May 2015 19:57:22 +0200 Denys Vlasenko wrote:
> With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
> very small functions we expect to be inlined. In particular,
> with this config: http://busybox.net/~vda/kernel_config
> there are more than a thousand copies of
On Mon, May 11, 2015 at 07:57:22PM +0200, Denys Vlasenko wrote:
> With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
> very small functions we expect to be inlined. In particular,
> with this config: http://busybox.net/~vda/kernel_config
> there are more than a thousand
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this config: http://busybox.net/~vda/kernel_config
there are more than a thousand copies of tiny spinlock-related functions:
$ nm --size-sort vmlinux | grep
On Mon, May 11, 2015 at 07:57:22PM +0200, Denys Vlasenko wrote:
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this config: http://busybox.net/~vda/kernel_config
there are more than a thousand copies of
On Mon, 11 May 2015 19:57:22 +0200 Denys Vlasenko dvlas...@redhat.com wrote:
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this config: http://busybox.net/~vda/kernel_config
there are more than a
With both gcc 4.7.2 and 4.9.2, sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. In particular,
with this config: http://busybox.net/~vda/kernel_config
there are more than a thousand copies of tiny spinlock-related functions:
$ nm --size-sort vmlinux | grep
32 matches
Mail list logo
