2020-04-30 03:31:01

by Zhangshaokun

[permalink] [raw]
Subject: [RFC PATCH] fs: Move @f_count to different cacheline with @f_mode

From: Yuqi Jin <[email protected]>

__fget_files does check the @f_mode with mask variable and will do some
atomic operations on @f_count while both are on the same cacheline.
Many CPU cores do file access and it will cause much conflicts on @f_count.
If we could make the two members into different cachelines, it shall relax
the siutations.

We have tested this on ARM64 and X86, the result is as follows:

Syscall of unixbench has been run on Huawei Kunpeng920 with this patch:
24 x System Call Overhead 1

System Call Overhead 3160841.4 lps (10.0 s, 1 samples)

System Benchmarks Partial Index BASELINE RESULT INDEX
System Call Overhead 15000.0 3160841.4 2107.2
========
System Benchmarks Index Score (Partial Only) 2107.2

Without this patch:
24 x System Call Overhead 1

System Call Overhead 2222456.0 lps (10.0 s, 1 samples)

System Benchmarks Partial Index BASELINE RESULT INDEX
System Call Overhead 15000.0 2222456.0 1481.6
========
System Benchmarks Index Score (Partial Only) 1481.6

And on Intel 6248 platform with this patch:
40 CPUs in system; running 24 parallel copies of tests

System Call Overhead 4288509.1 lps (10.0 s, 1 samples)

System Benchmarks Partial Index BASELINE RESULT INDEX
System Call Overhead 15000.0 4288509.1 2859.0
========
System Benchmarks Index Score (Partial Only) 2859.0

Without this patch:
40 CPUs in system; running 24 parallel copies of tests

System Call Overhead 3666313.0 lps (10.0 s, 1 samples)

System Benchmarks Partial Index BASELINE RESULT INDEX
System Call Overhead 15000.0 3666313.0 2444.2
========
System Benchmarks Index Score (Partial Only) 2444.2

Cc: Alexander Viro <[email protected]>
Signed-off-by: Yuqi Jin <[email protected]>
Signed-off-by: Shaokun Zhang <[email protected]>
---
include/linux/fs.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4f6f59b4f22a..90e76283f0fd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -953,7 +953,6 @@ struct file {
*/
spinlock_t f_lock;
enum rw_hint f_write_hint;
- atomic_long_t f_count;
unsigned int f_flags;
fmode_t f_mode;
struct mutex f_pos_lock;
@@ -976,6 +975,7 @@ struct file {
#endif /* #ifdef CONFIG_EPOLL */
struct address_space *f_mapping;
errseq_t f_wb_err;
+ atomic_long_t f_count;
} __randomize_layout
__attribute__((aligned(4))); /* lest something weird decides that 2 is OK */

--
2.7.4


2020-05-18 02:09:18

by Zhangshaokun

[permalink] [raw]
Subject: Re: [RFC PATCH] fs: Move @f_count to different cacheline with @f_mode

Hi maintainers,

A gentle ping.

Thanks,
Shaokun

On 2020/4/30 11:25, Shaokun Zhang wrote:
> From: Yuqi Jin <[email protected]>
>
> __fget_files does check the @f_mode with mask variable and will do some
> atomic operations on @f_count while both are on the same cacheline.
> Many CPU cores do file access and it will cause much conflicts on @f_count.
> If we could make the two members into different cachelines, it shall relax
> the siutations.
>
> We have tested this on ARM64 and X86, the result is as follows:
>
> Syscall of unixbench has been run on Huawei Kunpeng920 with this patch:
> 24 x System Call Overhead 1
>
> System Call Overhead 3160841.4 lps (10.0 s, 1 samples)
>
> System Benchmarks Partial Index BASELINE RESULT INDEX
> System Call Overhead 15000.0 3160841.4 2107.2
> ========
> System Benchmarks Index Score (Partial Only) 2107.2
>
> Without this patch:
> 24 x System Call Overhead 1
>
> System Call Overhead 2222456.0 lps (10.0 s, 1 samples)
>
> System Benchmarks Partial Index BASELINE RESULT INDEX
> System Call Overhead 15000.0 2222456.0 1481.6
> ========
> System Benchmarks Index Score (Partial Only) 1481.6
>
> And on Intel 6248 platform with this patch:
> 40 CPUs in system; running 24 parallel copies of tests
>
> System Call Overhead 4288509.1 lps (10.0 s, 1 samples)
>
> System Benchmarks Partial Index BASELINE RESULT INDEX
> System Call Overhead 15000.0 4288509.1 2859.0
> ========
> System Benchmarks Index Score (Partial Only) 2859.0
>
> Without this patch:
> 40 CPUs in system; running 24 parallel copies of tests
>
> System Call Overhead 3666313.0 lps (10.0 s, 1 samples)
>
> System Benchmarks Partial Index BASELINE RESULT INDEX
> System Call Overhead 15000.0 3666313.0 2444.2
> ========
> System Benchmarks Index Score (Partial Only) 2444.2
>
> Cc: Alexander Viro <[email protected]>
> Signed-off-by: Yuqi Jin <[email protected]>
> Signed-off-by: Shaokun Zhang <[email protected]>
> ---
> include/linux/fs.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 4f6f59b4f22a..90e76283f0fd 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -953,7 +953,6 @@ struct file {
> */
> spinlock_t f_lock;
> enum rw_hint f_write_hint;
> - atomic_long_t f_count;
> unsigned int f_flags;
> fmode_t f_mode;
> struct mutex f_pos_lock;
> @@ -976,6 +975,7 @@ struct file {
> #endif /* #ifdef CONFIG_EPOLL */
> struct address_space *f_mapping;
> errseq_t f_wb_err;
> + atomic_long_t f_count;
> } __randomize_layout
> __attribute__((aligned(4))); /* lest something weird decides that 2 is OK */
>
>