Skip to contentSkip to conversation

xfs: Use xchg() in xlog_cil_insert_pcp_aggregate()

PATCH Applied Uros Bizjak 4 participants 6 messages Nov 20, 2024 – Nov 28, 2024 XFS filesystem

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 xfs: Use xchg() in xlog_cil_insert_pcp_aggregate()

try_cmpxchg() loop with constant "new" value can be substituted with just xchg() to atomically get and clear the location.

The code on x86_64 improves from:

    1e7f:	48 89 4c 24 10       	mov    %rcx,0x10(%rsp)
    1e84:	48 03 14 c5 00 00 00 	add    0x0(,%rax,8),%rdx
    1e8b:	00
			1e88: R_X86_64_32S	__per_cpu_offset
    1e8c:	8b 02                	mov    (%rdx),%eax
    1e8e:	41 89 c5             	mov    %eax,%r13d
    1e91:	31 c9                	xor    %ecx,%ecx
    1e93:	f0 0f b1 0a          	lock cmpxchg %ecx,(%rdx)
    1e97:	75 f5                	jne    1e8e <xlog_cil_commit+0x84e>
    1e99:	48 8b 4c 24 10       	mov    0x10(%rsp),%rcx
    1e9e:	45 01 e9             	add    %r13d,%r9d

to just:

    1e7f:	48 03 14 cd 00 00 00 	add    0x0(,%rcx,8),%rdx
    1e86:	00
			1e83: R_X86_64_32S	__per_cpu_offset
    1e87:	31 c9                	xor    %ecx,%ecx
    1e89:	87 0a                	xchg   %ecx,(%rdx)
    1e8b:	41 01 cb             	add    %ecx,%r11d

No functional change intended.

Cc Chandan Babu R <[email protected]>
Cc "Darrick J. Wong" <[email protected]>
Cc Christoph Hellwig <[email protected]>
Cc Dave Chinner <[email protected]>
---
 fs/xfs/xfs_log_cil.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
fs/xfs/xfs_log_cil.c +1 -4
index 80da0cf87d7a..9d667be1d909 100644
@@ -171,11 +171,8 @@ xlog_cil_insert_pcp_aggregate(
 	 */
 	for_each_cpu(cpu, &ctx->cil_pcpmask) {
 		struct xlog_cil_pcp	*cilpcp = per_cpu_ptr(cil->xc_pcp, cpu);
-		int			old = READ_ONCE(cilpcp->space_used);
 
-		while (!try_cmpxchg(&cilpcp->space_used, &old, 0))
-			;
-		count += old;
+		count += xchg(&cilpcp->space_used, 0);
 	}
 	atomic_add(count, &ctx->space_used);
 }
Signature · 1 line
2.42.0
#1 lore raw
Quoted 2 lines · Uros Bizjak

try_cmpxchg() loop with constant "new" value can be substituted with just xchg() to atomically get and clear the location.

You're right. With a constant new value (0), there is no need to loop to ensure we get a "stable" update.

Is the READ_ONCE() is still needed?

-Alex

Quoted 51 lines

The code on x86_64 improves from:

     1e7f:	48 89 4c 24 10       	mov    %rcx,0x10(%rsp)
     1e84:	48 03 14 c5 00 00 00 	add    0x0(,%rax,8),%rdx
     1e8b:	00
			1e88: R_X86_64_32S	__per_cpu_offset
     1e8c:	8b 02                	mov    (%rdx),%eax
     1e8e:	41 89 c5             	mov    %eax,%r13d
     1e91:	31 c9                	xor    %ecx,%ecx
     1e93:	f0 0f b1 0a          	lock cmpxchg %ecx,(%rdx)
     1e97:	75 f5                	jne    1e8e <xlog_cil_commit+0x84e>
     1e99:	48 8b 4c 24 10       	mov    0x10(%rsp),%rcx
     1e9e:	45 01 e9             	add    %r13d,%r9d

to just:

     1e7f:	48 03 14 cd 00 00 00 	add    0x0(,%rcx,8),%rdx
     1e86:	00
			1e83: R_X86_64_32S	__per_cpu_offset
     1e87:	31 c9                	xor    %ecx,%ecx
     1e89:	87 0a                	xchg   %ecx,(%rdx)
     1e8b:	41 01 cb             	add    %ecx,%r11d

No functional change intended.

Cc Chandan Babu R <[email protected]>
Cc "Darrick J. Wong" <[email protected]>
Cc Christoph Hellwig <[email protected]>
Cc Dave Chinner <[email protected]>
---
  fs/xfs/xfs_log_cil.c | 5 +----
  1 file changed, 1 insertion(+), 4 deletions(-)
fs/xfs/xfs_log_cil.c +1 -4
index 80da0cf87d7a..9d667be1d909 100644
@@ -171,11 +171,8 @@ xlog_cil_insert_pcp_aggregate(
  	 */
  	for_each_cpu(cpu, &ctx->cil_pcpmask) {
  		struct xlog_cil_pcp	*cilpcp = per_cpu_ptr(cil->xc_pcp, cpu);
-		int			old = READ_ONCE(cilpcp->space_used);
 
-		while (!try_cmpxchg(&cilpcp->space_used, &old, 0))
-			;
-		count += old;
+		count += xchg(&cilpcp->space_used, 0);
  	}
  	atomic_add(count, &ctx->space_used);
  }
#2 lore raw
Quoted 9 lines · Alex Elder
Quoted 2 lines · Uros Bizjak

try_cmpxchg() loop with constant "new" value can be substituted with just xchg() to atomically get and clear the location.

You're right. With a constant new value (0), there is no need to loop to ensure we get a "stable" update.

Is the READ_ONCE() is still needed?

No, xchg() guarantees atomic access on its own.

Uros.

Reviewed-by
#3 lore raw
Quoted 6 lines · Uros Bizjak
Quoted 5 lines · Alex Elder
Quoted 2 lines · Uros Bizjak

try_cmpxchg() loop with constant "new" value can be substituted with just xchg() to atomically get and clear the location.

Quoted 8 lines
Quoted 4 lines

You're right. With a constant new value (0), there is no need to loop to ensure we get a "stable" update.

Is the READ_ONCE() is still needed?

No, xchg() guarantees atomic access on its own.

Uros.

Based on that:

Reviewed-by on v1 1/1
#4 lore raw
In reply to Uros Bizjak
Quoted 44 lines · Uros Bizjak

try_cmpxchg() loop with constant "new" value can be substituted with just xchg() to atomically get and clear the location.

The code on x86_64 improves from:

    1e7f:	48 89 4c 24 10       	mov    %rcx,0x10(%rsp)
    1e84:	48 03 14 c5 00 00 00 	add    0x0(,%rax,8),%rdx
    1e8b:	00
			1e88: R_X86_64_32S	__per_cpu_offset
    1e8c:	8b 02                	mov    (%rdx),%eax
    1e8e:	41 89 c5             	mov    %eax,%r13d
    1e91:	31 c9                	xor    %ecx,%ecx
    1e93:	f0 0f b1 0a          	lock cmpxchg %ecx,(%rdx)
    1e97:	75 f5                	jne    1e8e <xlog_cil_commit+0x84e>
    1e99:	48 8b 4c 24 10       	mov    0x10(%rsp),%rcx
    1e9e:	45 01 e9             	add    %r13d,%r9d

to just:

    1e7f:	48 03 14 cd 00 00 00 	add    0x0(,%rcx,8),%rdx
    1e86:	00
			1e83: R_X86_64_32S	__per_cpu_offset
    1e87:	31 c9                	xor    %ecx,%ecx
    1e89:	87 0a                	xchg   %ecx,(%rdx)
    1e8b:	41 01 cb             	add    %ecx,%r11d

No functional change intended.

Cc Chandan Babu R <[email protected]>
Cc "Darrick J. Wong" <[email protected]>
Cc Christoph Hellwig <[email protected]>
Cc Dave Chinner <[email protected]>
---
 fs/xfs/xfs_log_cil.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
fs/xfs/xfs_log_cil.c +0 -0
index 80da0cf87d7a..9d667be1d909 100644
@@ -171,11 +171,8 @@ xlog_cil_insert_pcp_aggregate(
 	 */
 	for_each_cpu(cpu, &ctx->cil_pcpmask) {

fs/xfs/xfs_log_cil.c xlog_cil_insert_pcp_aggregate( · line 178

Quoted 10 lines
fs/xfs/xfs_log_cil.c +1 -4
 		struct xlog_cil_pcp	*cilpcp = per_cpu_ptr(cil->xc_pcp, cpu);
-		int			old = READ_ONCE(cilpcp->space_used);
 
-		while (!try_cmpxchg(&cilpcp->space_used, &old, 0))
-			;
-		count += old;
+		count += xchg(&cilpcp->space_used, 0);
 	}
 	atomic_add(count, &ctx->space_used);
 }

Looks fine.

Signature · 2 lines
Dave Chinner
[email protected]
maintainer Applied
#5 lore raw
In reply to Uros Bizjak
Quoted 17 lines · Uros Bizjak

try_cmpxchg() loop with constant "new" value can be substituted with just xchg() to atomically get and clear the location.

The code on x86_64 improves from:

    1e7f:	48 89 4c 24 10       	mov    %rcx,0x10(%rsp)
    1e84:	48 03 14 c5 00 00 00 	add    0x0(,%rax,8),%rdx
    1e8b:	00
			1e88: R_X86_64_32S	__per_cpu_offset
    1e8c:	8b 02                	mov    (%rdx),%eax
    1e8e:	41 89 c5             	mov    %eax,%r13d
    1e91:	31 c9                	xor    %ecx,%ecx
    1e93:	f0 0f b1 0a          	lock cmpxchg %ecx,(%rdx)
    1e97:	75 f5                	jne    1e8e <xlog_cil_commit+0x84e>
    1e99:	48 8b 4c 24 10       	mov    0x10(%rsp),%rcx
    1e9e:	45 01 e9             	add    %r13d,%r9d
Quoted 1 line

[...]

Applied to for-next, thanks!

[1/1] xfs: Use xchg() in xlog_cil_insert_pcp_aggregate()

      commit: 214093534f3c046bf5acc9affbf4e6bd9af4538b

Best regards,

Signature · 1 line
Carlos Maiolino <[email protected]>