RE: [RFC] change non-atomic bitops method

[Wang, Yalin]

> -----Original Message-----
> From: Andrew Morton [mailto:akpm@xxxxxxxxxxxxxxxxxxxx]
> Sent: Tuesday, February 03, 2015 6:59 PM
> To: Wang, Yalin
> Cc: 'Kirill A. Shutemov'; 'arnd@xxxxxxxx'; 'linux-arch@xxxxxxxxxxxxxxx';
> 'linux-kernel@xxxxxxxxxxxxxxx'; 'linux@xxxxxxxxxxxxxxxx'; 'linux-arm-
> kernel@xxxxxxxxxxxxxxxxxxx'
> Subject: Re: [RFC] change non-atomic bitops method
> 
> On Tue, 3 Feb 2015 16:42:14 +0800 "Wang, Yalin" <Yalin.Wang@xxxxxxxxxxxxxx>
> wrote:
> 
> > I make a change in kernel to test hit/miss ratio:
> 
> Neat, thanks.
> 
> >
> > ...
> >
> > After use the phone some time:
> > root@D5303:/ # cat /proc/meminfo
> > VmallocUsed:       10348 kB
> > VmallocChunk:      75632 kB
> > __set_bit_miss_count:10002 __set_bit_success_count:1096661
> > __clear_bit_miss_count:359484 __clear_bit_success_count:3674617
> > __test_and_set_bit_miss_count:7 __test_and_set_bit_success_count:221
> > __test_and_clear_bit_miss_count:924611
> __test_and_clear_bit_success_count:193
> >
> > __test_and_clear_bit_miss_count has a very high miss rate.
> > In fact, I think set/clear/test_and_set(clear)_bit atomic version can
> also
> > Be investigated to see its miss ratio,
> > I have not tested the atomic version,
> > Because it reside in different architectures.
> 
> Hopefully misses in test_and_X_bit are not a problem.  The CPU
> implementation would be pretty stupid to go and dirty the cacheline
> when it knows it didn't change anything.  But maybe I'm wrong about
> that.
> 
> That we're running clear_bit against a cleared bit 10% of the time is a
> bit alarming.  I wonder where that's coming from.
> 
> The enormous miss count in test_and_clear_bit() might indicate an
> inefficiency somewhere.
I te-test the patch on 3.10 kernel.
The result like this:

VmallocChunk:   251498164 kB
__set_bit_miss_count:11730 __set_bit_success_count:1036316
__clear_bit_miss_count:209640 __clear_bit_success_count:4806556
__test_and_set_bit_miss_count:0 __test_and_set_bit_success_count:121
__test_and_clear_bit_miss_count:0 __test_and_clear_bit_success_count:445

__clear_bit miss rate is a little high,
I check the log, and most miss coming from this code:

<6>[  442.701798] [<ffffffc00021d084>] warn_slowpath_fmt+0x4c/0x58
<6>[  442.701805] [<ffffffc0002461a8>] __clear_bit+0x98/0xa4
<6>[  442.701813] [<ffffffc0003126ac>] __alloc_fd+0xc8/0x124
<6>[  442.701821] [<ffffffc000312768>] get_unused_fd_flags+0x28/0x34
<6>[  442.701828] [<ffffffc0002f9370>] do_sys_open+0x10c/0x1c0
<6>[  442.701835] [<ffffffc0002f9458>] SyS_openat+0xc/0x18
In __clear_close_on_exec(fd, fdt);



<6>[  442.695354] [<ffffffc00021d084>] warn_slowpath_fmt+0x4c/0x58
<6>[  442.695359] [<ffffffc0002461a8>] __clear_bit+0x98/0xa4
<6>[  442.695367] [<ffffffc000312340>] dup_fd+0x1d4/0x280
<6>[  442.695375] [<ffffffc00021b07c>] copy_process.part.56+0x42c/0xe38
<6>[  442.695382] [<ffffffc00021bb9c>] do_fork+0xe0/0x360
<6>[  442.695389] [<ffffffc00021beb4>] SyS_clone+0x10/0x1c
In __clear_open_fd(open_files - i, new_fdt);

Do we need test_bit() before clear_bit()at these 2 place?

Thanks




--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/