Skip to contentSkip to conversation

powerpc/perf: Use stack siar instead of mfspr

PATCH Applied Kajol Jain 3 participants 5 messages Aug 18, 2021 – Aug 27, 2021 PowerPC

lore
Share & tools
lore

Kajol Jain

author
#0 lore raw

PATCH v4 1/3 powerpc/perf: Use stack siar instead of mfspr

Minor optimization in the 'perf_instruction_pointer' function code by making use of stack siar instead of mfspr.

Fixes 75382aa72f06 ("powerpc/perf: Move code to select SIAR or pt_regs

into perf_read_regs")

---
 arch/powerpc/perf/core-book3s.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/perf/core-book3s.c +1 -1
index bb0ee716de91..1b464aad29c4 100644
@@ -2260,7 +2260,7 @@ unsigned long perf_instruction_pointer(struct pt_regs *regs)
 		else
 			return regs->nip;
 	} else if (use_siar && siar_valid(regs))
-		return mfspr(SPRN_SIAR) + perf_ip_adjust(regs);
+		return siar + perf_ip_adjust(regs);
 	else if (use_siar)
 		return 0;		// no valid instruction pointer
 	else
Signature · 1 line
2.26.2
author
#1 lore raw

PATCH v4 2/3 powerpc/perf: Drop the case of returning 0 as instruction pointer

Drop the case of returning 0 as instruction pointer since kernel never executes at 0 and userspace almost never does either.

Fixes e6878835ac47 ("powerpc/perf: Sample only if SIAR-Valid

bit is set in P7+")

---
 arch/powerpc/perf/core-book3s.c | 2 --
 1 file changed, 2 deletions(-)
arch/powerpc/perf/core-book3s.c +0 -2
index 1b464aad29c4..23ec89a59893 100644
@@ -2261,8 +2261,6 @@ unsigned long perf_instruction_pointer(struct pt_regs *regs)
 			return regs->nip;
 	} else if (use_siar && siar_valid(regs))
 		return siar + perf_ip_adjust(regs);
-	else if (use_siar)
-		return 0;		// no valid instruction pointer
 	else
 		return regs->nip;
 }
Signature · 1 line
2.26.2
author
#2 lore raw

PATCH v4 3/3 powerpc/perf: Fix the check for SIAR value

In reply to Kajol Jain

Incase of random sampling, there can be scenarios where Sample Instruction Address Register(SIAR) may not latch to the sampled instruction and could result in the value of 0. In these scenarios it is preferred to return regs->nip. These corner cases are seen in the previous generation (p9) also.

Patch adds the check for SIAR value along with regs_use_siar and siar_valid checks so that the function will return regs->nip incase SIAR is zero.

Patch drops the code under PPMU_P10_DD1 flag check which handles SIAR 0 case only for Power10 DD1.

Fixes 2ca13a4cc56c9 ("powerpc/perf: Use regs->nip when SIAR is zero")

---

Changelog:
v3 -> v4
- Remove use_siar variable and directly using regs_use_siar call as suggested by Christophe Leroy

v2 -> v3
- Drop adding new ternary condition to check siar value.
- Remove siar check specific for PPMU_P10_DD1 and add it along with common checks as suggested by Christophe Leroy and Michael Ellermen

 arch/powerpc/perf/core-book3s.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)
arch/powerpc/perf/core-book3s.c +1 -7
index 23ec89a59893..b0a589409039 100644
@@ -2251,15 +2251,9 @@ unsigned long perf_misc_flags(struct pt_regs *regs)
  */
 unsigned long perf_instruction_pointer(struct pt_regs *regs)
 {
-	bool use_siar = regs_use_siar(regs);
 	unsigned long siar = mfspr(SPRN_SIAR);
 
-	if (ppmu && (ppmu->flags & PPMU_P10_DD1)) {
-		if (siar)
-			return siar;
-		else
-			return regs->nip;
-	} else if (use_siar && siar_valid(regs))
+	if (regs_use_siar(regs) && siar_valid(regs) && siar)
 		return siar + perf_ip_adjust(regs);
 	else
 		return regs->nip;
Signature · 1 line
2.26.2
Tested-by on v4 1/3
#3 lore raw
In reply to Kajol Jain
Quoted 6 lines · Kajol Jain

Minor optimization in the 'perf_instruction_pointer' function code by making use of stack siar instead of mfspr.

Fixes 75382aa72f06 ("powerpc/perf: Move code to select SIAR or pt_regs

into perf_read_regs")

Tested this patch series, not seeing any '0' values.

example output:

# perf report -D | grep addr
0 26236879714 0x3dcc8 [0x38]: PERF_RECORD_SAMPLE(IP, 0x1): 1446/1446: 0xc000000000113584 period: 1 addr: 0
0 26236882500 0x3dd00 [0x38]: PERF_RECORD_SAMPLE(IP, 0x1): 1446/1446: 0xc000000000113584 period: 1 addr: 0
0 26236883436 0x3dd38 [0x38]: PERF_RECORD_SAMPLE(IP, 0x1): 1446/1446: 0xc000000000113584 period: 10 addr: 0
...
Quoted 18 lines
---
  arch/powerpc/perf/core-book3s.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/perf/core-book3s.c +1 -1
index bb0ee716de91..1b464aad29c4 100644
@@ -2260,7 +2260,7 @@ unsigned long perf_instruction_pointer(struct pt_regs *regs)
  		else
  			return regs->nip;
  	} else if (use_siar && siar_valid(regs))
-		return mfspr(SPRN_SIAR) + perf_ip_adjust(regs);
+		return siar + perf_ip_adjust(regs);
  	else if (use_siar)
  		return 0;		// no valid instruction pointer
  	else
Signature · 2 lines
Thanks and Regards
R.Nageswara Sastry
#4 lore raw
In reply to Kajol Jain
Quoted 6 lines · Kajol Jain

Minor optimization in the 'perf_instruction_pointer' function code by making use of stack siar instead of mfspr.

Applied to powerpc/next.

[1/3] powerpc/perf: Use stack siar instead of mfspr

      https://git.kernel.org/powerpc/c/b1643084d164cea0c107a39bcdf0119fc52619af
[2/3] powerpc/perf: Drop the case of returning 0 as instruction pointer
      https://git.kernel.org/powerpc/c/cc90c6742ef5b438f4cb86029d7a794bd0a44a06
[3/3] powerpc/perf: Fix the check for SIAR value
      https://git.kernel.org/powerpc/c/3c69a5f22223fa3e312689ec218b5059784d49d7

cheers