perf: arm-spe: Decode SPE source and use for perf c2c
Subsystems (2)
v8 cover perf: arm-spe: Decode SPE source and use for perf c2c
When synthesizing data from SPE, augment the type with source information for Arm Neoverse cores so we can detect situtions like cache line contention and transfers on Arm platforms.
This changes enables future changes to c2c on a system with SPE where lines that are shared among multiple cores show up in perf c2c output.
Changes in v8:
* Report NA for both mem_lvl and mem_lvl_num for stores where we have no information
Changes in v7:
* Minor change requested by Leo Yan
Changes in v6:
* Drop changes to c2c command which will come from Leo Yan
Changes in v5:
* Add a new snooping type to disambiguate cache-to-cache transfers where we don't know if the data is clean or dirty.
* Set snoop flags on all the data-source cases
* Special case stores as we have no information on them
Changes in v4:
* Bring-in the kernel's arch/arm64/include/asm/cputype.h into tools/
* Add neoverse-v1 to the neoverse cores list
Ali Saidi (4):
tools: arm64: Import cputype.h
perf arm-spe: Use SPE data source for neoverse cores perf mem: Support mem_lvl_num in c2c command perf mem: Support HITM for when mem_lvl_num is any
tools/arch/arm64/include/asm/cputype.h | 258 ++++++++++++++++++ .../util/arm-spe-decoder/arm-spe-decoder.c | 1 + .../util/arm-spe-decoder/arm-spe-decoder.h | 12 + tools/perf/util/arm-spe.c | 110 +++++++- tools/perf/util/mem-events.c | 20 +- 5 files changed, 383 insertions(+), 18 deletions(-) create mode 100644 tools/arch/arm64/include/asm/cputype.h
Signature · 1 line
2.32.0
PATCH v8 1/5 perf: Add SNOOP_PEER flag to perf mem data struct
Add a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
--- include/uapi/linux/perf_event.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/uapi/linux/perf_event.h +1 -1
@@ -1310,7 +1310,7 @@ union perf_mem_data_src {#define PERF_MEM_SNOOP_SHIFT 19#define PERF_MEM_SNOOPX_FWD 0x01/* forward */ -/* 1 free */ +#define PERF_MEM_SNOOPX_PEER 0x02/* xfer from peer */ #define PERF_MEM_SNOOPX_SHIFT 38/* locked instruction */
Signature · 1 line
2.32.0
Re: [PATCH v8 1/5] perf: Add SNOOP_PEER flag to perf mem data struct
Quoted 19 lines · Ali Saidi
Add a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
Signed-off-by Ali Saidi <[email protected]>Reviewed-by Leo Yan <[email protected]>--- include/uapi/linux/perf_event.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)include/uapi/linux/perf_event.h +0
-0
include/uapi/linux/perf_event.h union perf_mem_data_src { · line 1316
Quoted 8 lines
include/uapi/linux/perf_event.h +1
-1@@ -1310,7 +1310,7 @@ union perf_mem_data_src {#define PERF_MEM_SNOOP_SHIFT 19#define PERF_MEM_SNOOPX_FWD 0x01/* forward */ -/* 1 free */ +#define PERF_MEM_SNOOPX_PEER 0x02/* xfer from peer */ #define PERF_MEM_SNOOPX_SHIFT 38/* locked instruction */
Patch looks good to me.
Thanks,
Kajol Jain
PATCH v8 2/5 perf tools: sync addition of PERF_MEM_SNOOPX_PEER
In reply to Ali SaidiAdd a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
--- tools/include/uapi/linux/perf_event.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/include/uapi/linux/perf_event.h +1 -1
@@ -1310,7 +1310,7 @@ union perf_mem_data_src {#define PERF_MEM_SNOOP_SHIFT 19#define PERF_MEM_SNOOPX_FWD 0x01/* forward */ -/* 1 free */ +#define PERF_MEM_SNOOPX_PEER 0x02/* xfer from peer */ #define PERF_MEM_SNOOPX_SHIFT 38/* locked instruction */
Signature · 1 line
2.32.0
Arnaldo Carvalho de Melo
maintainerRe: [PATCH v8 2/5] perf tools: sync addition of PERF_MEM_SNOOPX_PEER
Em Wed, May 04, 2022 at 06:48:47PM +0000, Ali Saidi escreveu:
Quoted 10 lines
Add a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
Signed-off-by Ali Saidi <[email protected]>Reviewed-by Leo Yan <[email protected]>
Was this already merged on the ARM kernel tree?
- Arnaldo
Quoted 19 lines
--- tools/include/uapi/linux/perf_event.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)tools/include/uapi/linux/perf_event.h +1
-1@@ -1310,7 +1310,7 @@ union perf_mem_data_src {#define PERF_MEM_SNOOP_SHIFT 19#define PERF_MEM_SNOOPX_FWD 0x01/* forward */ -/* 1 free */ +#define PERF_MEM_SNOOPX_PEER 0x02/* xfer from peer */ #define PERF_MEM_SNOOPX_SHIFT 38/* locked instruction */ Signature · 1 line
2.32.0
Signature · 1 line
- Arnaldo
Re: [PATCH v8 2/5] perf tools: sync addition of PERF_MEM_SNOOPX_PEER
Quoted 9 lines · Arnaldo Carvalho de Melo
Em Wed, May 04, 2022 at 06:48:47PM +0000, Ali Saidi escreveu:
Quoted 8 lines
Add a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
Quoted 4 lines
Quoted 2 lines
Signed-off-by Ali Saidi <[email protected]>Reviewed-by Leo Yan <[email protected]>Was this already merged on the ARM kernel tree?
No, I don't think this patch has been merged on Arm kernel tree. I searched Arm and Arm64 git repos, none of them has merged this patch.
https://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git/log/?qt=author&q=Ali+Saidi
http://git.armlinux.org.uk/cgit/linux-arm.git/log/?qt=author&q=Ali+Saidi
P.s. Ali missed to include German's review tag, see:
https://lore.kernel.org/lkml/[email protected]/
Do you want us to resend the patch set for adding tags?
Thanks,
Leo
Arnaldo Carvalho de Melo
maintainerRe: [PATCH v8 2/5] perf tools: sync addition of PERF_MEM_SNOOPX_PEER
Em Wed, May 11, 2022 at 10:20:04AM +0800, Leo Yan escreveu:
Quoted 18 lines · Arnaldo Carvalho de Melo
Quoted 9 lines
Em Wed, May 04, 2022 at 06:48:47PM +0000, Ali Saidi escreveu:
Quoted 8 lines
Add a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
Quoted 4 lines
Quoted 2 lines
Signed-off-by Ali Saidi <[email protected]>Reviewed-by Leo Yan <[email protected]>Was this already merged on the ARM kernel tree?
No, I don't think this patch has been merged on Arm kernel tree. I searched Arm and Arm64 git repos, none of them has merged this patch.
Quoted 7 lines
https://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git/log/?qt=author&q=Ali+Saidi
http://git.armlinux.org.uk/cgit/linux-arm.git/log/?qt=author&q=Ali+SaidiP.s. Ali missed to include German's review tag, see:
https://lore.kernel.org/lkml/[email protected]/Do you want us to resend the patch set for adding tags?
I use b4 and it should collect Reviewed-by, Acked-by, etc tags, for instance, if I use the message-id in your message:
⬢[acme@toolbox perf]$ b4 am -ctsl --cc-trailers 20220511022004.GA956170@leoy-ThinkPad-X240s Looking up https://lore.kernel.org/r/20220511022004.GA956170%40leoy-ThinkPad-X240s
Grabbing thread from lore.kernel.org/all/20220511022004.GA956170%40leoy-ThinkPad-X240s/t.mbox.gz Checking for newer revisions on https://lore.kernel.org/all/
Analyzing 12 messages in the thread
Checking attestation on all messages, may take a moment...
---
✓ [PATCH v8 1/5] perf: Add SNOOP_PEER flag to perf mem data struct
+ Reviewed-By: Kajol Jain<[email protected]> (✓ DKIM/ibm.com)
+ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
+ Link: https://lore.kernel.org/r/[email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
✓ [PATCH v8 2/5] perf tools: sync addition of PERF_MEM_SNOOPX_PEER
+ Reviewed-By: Kajol Jain<[email protected]> (✓ DKIM/ibm.com)
+ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
+ Link: https://lore.kernel.org/r/[email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
✓ [PATCH v8 3/5] perf mem: Print snoop peer flag
+ Reviewed-By: Kajol Jain<[email protected]> (✓ DKIM/ibm.com)
+ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
+ Link: https://lore.kernel.org/r/[email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
✓ [PATCH v8 4/5] perf arm-spe: Don't set data source if it's not a memory operation
+ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
+ Link: https://lore.kernel.org/r/[email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
✓ [PATCH v8 5/5] perf arm-spe: Use SPE data source for neoverse cores
+ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
+ Link: https://lore.kernel.org/r/[email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
---
✓ Signed: DKIM/amazon.com
---
Total patches: 5
---
Cover: ./v8_20220504_alisaidi_perf_arm_spe_decode_spe_source_and_use_for_perf_c2c.cover
Base: not specified
git am ./v8_20220504_alisaidi_perf_arm_spe_decode_spe_source_and_use_for_perf_c2c.mbx
⬢[acme@toolbox perf]$
Somehow it is not being collected... :-\
Not even when I use:
Quoted 2 lines
P.s. Ali missed to include German's review tag, see:
https://lore.kernel.org/lkml/[email protected]/
Will try updating b4...
- Arnaldo
Arnaldo Carvalho de Melo
maintainerRe: [PATCH v8 2/5] perf tools: sync addition of PERF_MEM_SNOOPX_PEER
Em Wed, May 11, 2022 at 03:28:00PM -0300, Arnaldo Carvalho de Melo escreveu:
Quoted 35 lines
Em Wed, May 11, 2022 at 10:20:04AM +0800, Leo Yan escreveu:
Quoted 18 lines · Arnaldo Carvalho de Melo
Quoted 9 lines
Em Wed, May 04, 2022 at 06:48:47PM +0000, Ali Saidi escreveu:
Quoted 8 lines
Add a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
Quoted 4 lines
Quoted 2 lines
Signed-off-by Ali Saidi <[email protected]>Reviewed-by Leo Yan <[email protected]>Was this already merged on the ARM kernel tree?
No, I don't think this patch has been merged on Arm kernel tree. I searched Arm and Arm64 git repos, none of them has merged this patch.
Quoted 7 lines
https://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git/log/?qt=author&q=Ali+Saidi
http://git.armlinux.org.uk/cgit/linux-arm.git/log/?qt=author&q=Ali+SaidiP.s. Ali missed to include German's review tag, see:
https://lore.kernel.org/lkml/[email protected]/Do you want us to resend the patch set for adding tags?
I use b4 and it should collect Reviewed-by, Acked-by, etc tags, for instance, if I use the message-id in your message:
⬢[acme@toolbox perf]$ b4 am -ctsl --cc-trailers 20220511022004.GA956170@leoy-ThinkPad-X240s Looking up https://lore.kernel.org/r/20220511022004.GA956170%40leoy-ThinkPad-X240s
Grabbing thread from lore.kernel.org/all/20220511022004.GA956170%40leoy-ThinkPad-X240s/t.mbox.gz Checking for newer revisions on https://lore.kernel.org/all/
Analyzing 12 messages in the thread
<SNIP>
Quoted 37 lines
✓ [PATCH v8 5/5] perf arm-spe: Use SPE data source for neoverse cores
+ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
+ Link: https://lore.kernel.org/r/[email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
+ Cc: [email protected]
---
✓ Signed: DKIM/amazon.com
---
Total patches: 5
---
Cover: ./v8_20220504_alisaidi_perf_arm_spe_decode_spe_source_and_use_for_perf_c2c.coverBase: not specified
git am ./v8_20220504_alisaidi_perf_arm_spe_decode_spe_source_and_use_for_perf_c2c.mbx⬢[acme@toolbox perf]$
Somehow it is not being collected... :-\
Not even when I use:
Quoted 7 lines
Quoted 2 lines
P.s. Ali missed to include German's review tag, see:
https://lore.kernel.org/lkml/[email protected]/Will try updating b4...
Didn't help, so please collect the new tags and resubmit.
- Arnaldo
Re: [PATCH v8 2/5] perf tools: sync addition of PERF_MEM_SNOOPX_PEER
In reply to Ali SaidiQuoted 19 lines · Ali Saidi
Add a flag to the perf mem data struct to signal that a request caused a cache-to-cache transfer of a line from a peer of the requestor and wasn't sourced from a lower cache level. The line being moved from one peer cache to another has latency and performance implications. On Arm64 Neoverse systems the data source can indicate a cache-to-cache transfer but not if the line is dirty or clean, so instead of overloading HITM define a new flag that indicates this type of transfer.
Signed-off-by Ali Saidi <[email protected]>Reviewed-by Leo Yan <[email protected]>--- tools/include/uapi/linux/perf_event.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)tools/include/uapi/linux/perf_event.h +0
-0
tools/include/uapi/linux/perf_event.h union perf_mem_data_src { · line 1316
Quoted 8 lines
tools/include/uapi/linux/perf_event.h +1
-1@@ -1310,7 +1310,7 @@ union perf_mem_data_src {#define PERF_MEM_SNOOP_SHIFT 19#define PERF_MEM_SNOOPX_FWD 0x01/* forward */ -/* 1 free */ +#define PERF_MEM_SNOOPX_PEER 0x02/* xfer from peer */ #define PERF_MEM_SNOOPX_SHIFT 38/* locked instruction */
Patch looks good to me.
Thanks,
Kajol Jain
PATCH v8 3/5 perf mem: Print snoop peer flag
In reply to Ali SaidiFrom: Leo Yan <[email protected]>
Since PERF_MEM_SNOOPX_PEER flag is a new snoop type, print this flag if it is set.
Before:
memstress 3603 [020] 122.463754: 1 l1d-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 l1d-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 llc-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 llc-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 tlb-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 memory: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)After:
memstress 3603 [020] 122.463754: 1 l1d-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 l1d-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 llc-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 llc-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 tlb-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
memstress 3603 [020] 122.463754: 1 memory: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)--- tools/perf/util/mem-events.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-)
tools/perf/util/mem-events.c +15 -3
@@ -410,6 +410,11 @@ static const char * const snoop_access[] = { "HitM", }; +static const char *const snoopx_access[] = { +"Fwd" , +"Peer" , +}; +int perf_mem__snp_scnprintf (char *out,size_t sz,struct mem_info *mem_info) {size_t i, l =0 ;
@@ -430,13 +435,20 @@ int perf_mem__snp_scnprintf(char *out, size_t sz, struct mem_info *mem_info) } l +=scnprintf (out + l, sz - l, snoop_access[i]); } -if (mem_info && - (mem_info->data_src.mem_snoopx &PERF_MEM_SNOOPX_FWD )) { + + m =0 ; +if (mem_info) + m = mem_info->data_src.mem_snoopx; + +for (i =0 ; m && i <ARRAY_SIZE (snoopx_access); i++, m >>=1 ) { +if (!(m &0x1 )) +continue ; +if (l) {strcat (out," or " ); l +=4 ; } - l +=scnprintf (out + l, sz - l,"Fwd" ); + l +=scnprintf (out + l, sz - l, snoopx_access[i]); }if (*out =='\0' )
Signature · 1 line
2.32.0
Re: [PATCH v8 3/5] perf mem: Print snoop peer flag
Quoted 22 lines · Ali Saidi
From: Leo Yan <[email protected]>
Since
PERF_MEM_SNOOPX_PEERflag is a new snoop type, print this flag if it is set.Before:
memstress 3603 [020] 122.463754: 1 l1d-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 l1d-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 llc-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 llc-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 tlb-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 memory: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP N/A|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)After:
memstress 3603 [020] 122.463754: 1 l1d-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 l1d-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 llc-miss: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 llc-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 tlb-access: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress) memstress 3603 [020] 122.463754: 1 memory: 8688000842 |OP LOAD|LVL L3 or L3 hit|SNP Peer|TLB Walker hit|LCK No|BLK N/A aaaac17c3e88 [unknown] (/home/ubuntu/memstress)
Quoted 3 lines
Signed-off-by Leo Yan <[email protected]>Reviewed-by Ali Saidi <[email protected]>Tested-by Ali Saidi <[email protected]>
Patch looks good to me.
Thanks,
Kajol Jain
Quoted 44 lines
--- tools/perf/util/mem-events.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-)tools/perf/util/mem-events.c +15
-3@@ -410,6 +410,11 @@ static const char * const snoop_access[] = { "HitM", }; +static const char *const snoopx_access[] = { +"Fwd" , +"Peer" , +}; +int perf_mem__snp_scnprintf (char *out,size_t sz,struct mem_info *mem_info) {size_t i, l =0 ;@@ -430,13 +435,20 @@ int perf_mem__snp_scnprintf(char *out, size_t sz, struct mem_info *mem_info) } l +=scnprintf (out + l, sz - l, snoop_access[i]); } -if (mem_info && - (mem_info->data_src.mem_snoopx &PERF_MEM_SNOOPX_FWD )) { + + m =0 ; +if (mem_info) + m = mem_info->data_src.mem_snoopx; + +for (i =0 ; m && i <ARRAY_SIZE (snoopx_access); i++, m >>=1 ) { +if (!(m &0x1 )) +continue ; +if (l) {strcat (out," or " ); l +=4 ; } - l +=scnprintf (out + l, sz - l,"Fwd" ); + l +=scnprintf (out + l, sz - l, snoopx_access[i]); }if (*out =='\0' )
PATCH v8 4/5 perf arm-spe: Don't set data source if it's not a memory operation
In reply to Ali SaidiFrom: Leo Yan <[email protected]>
Except memory load and store operations, Arm SPE records also can support other operation types, bug when set the data source field the current code assumes a record is a either load operation or store operation, this leads to wrongly synthesize memory samples.
This patch strictly checks the record operation type, it only sets data source only for the operation types ARM_SPE_LD and ARM_SPE_ST, otherwise, returns zero for data source. Therefore, we can synthesize memory samples only when data source is a non-zero value, the function arm_spe__is_memory_event() is useless and removed.
--- tools/perf/util/arm-spe.c | 22 ++++++++-------------- 1 file changed, 8 insertions(+), 14 deletions(-)
tools/perf/util/arm-spe.c +8 -14
@@ -387,26 +387,16 @@ static int arm_spe__synth_instruction_sample(struct arm_spe_queue *speq,return arm_spe_deliver_synth_event (spe, speq, event, &sample); } -#define SPE_MEM_TYPE (ARM_SPE_L1D_ACCESS | ARM_SPE_L1D_MISS | \ - ARM_SPE_LLC_ACCESS | ARM_SPE_LLC_MISS | \ - ARM_SPE_REMOTE_ACCESS) - -static bool arm_spe__is_memory_event (enum arm_spe_sample_type type) -{ -if (type &SPE_MEM_TYPE ) -return true; - -return false; -} -static u64 arm_spe__synth_data_source (const struct arm_spe_record *record) {union perf_mem_data_src data_src = {0 };if (record->op ==ARM_SPE_LD ) data_src.mem_op =PERF_MEM_OP_LOAD ; -else +else if (record->op ==ARM_SPE_ST ) data_src.mem_op =PERF_MEM_OP_STORE ; +else +return 0 ;if (record->type & (ARM_SPE_LLC_ACCESS |ARM_SPE_LLC_MISS )) { data_src.mem_lvl =PERF_MEM_LVL_L3 ;
@@ -510,7 +500,11 @@ static int arm_spe_sample(struct arm_spe_queue *speq)return err; } -if (spe->sample_memory &&arm_spe__is_memory_event (record->type)) { +/* +* When data_src is zero it means the record is not a memory operation, +* skip to synthesize memory sample for this case. +*/ +if (spe->sample_memory && data_src) { err =arm_spe__synth_mem_sample (speq, spe->memory_id, data_src);if (err)return err;
Signature · 1 line
2.32.0
PATCH v8 5/5 perf arm-spe: Use SPE data source for neoverse cores
In reply to Ali SaidiWhen synthesizing data from SPE, augment the type with source information for Arm Neoverse cores. The field is IMPLDEF but the Neoverse cores all use the same encoding. I can't find encoding information for any other SPE implementations to unify their choices with Arm's thus that is left for future work.
This change populates the mem_lvl_num for Neoverse cores as well as the deprecated mem_lvl namespace.
--- .../util/arm-spe-decoder/arm-spe-decoder.c | 1 + .../util/arm-spe-decoder/arm-spe-decoder.h | 12 ++ tools/perf/util/arm-spe.c | 130 +++++++++++++++--- 3 files changed, 127 insertions(+), 16 deletions(-)
tools/perf/util/arm-spe-decoder/arm-spe-decoder.c +1 -0
@@ -220,6 +220,7 @@ static int arm_spe_read_record(struct arm_spe_decoder *decoder)break ;case ARM_SPE_DATA_SOURCE : + decoder->record.source = payload;break ;case ARM_SPE_BAD :break ;
tools/perf/util/arm-spe-decoder/arm-spe-decoder.h +12 -0
@@ -29,6 +29,17 @@ enum arm_spe_op_type { ARM_SPE_ST = 1 << 1, }; +enum arm_spe_neoverse_data_source { +ARM_SPE_NV_L1D =0x0 , +ARM_SPE_NV_L2 =0x8 , +ARM_SPE_NV_PEER_CORE =0x9 , +ARM_SPE_NV_LOCAL_CLUSTER =0xa , +ARM_SPE_NV_SYS_CACHE =0xb , +ARM_SPE_NV_PEER_CLUSTER =0xc , +ARM_SPE_NV_REMOTE =0xd , +ARM_SPE_NV_DRAM =0xe , +}; +struct arm_spe_record {enum arm_spe_sample_type type;int err;
@@ -40,6 +51,7 @@ struct arm_spe_record {u64 virt_addr;u64 phys_addr;u64 context_id; +u16 source; };struct arm_spe_insn ;
tools/perf/util/arm-spe.c +114 -16
180 lines not on this page · raw message
Signature · 1 line
2.32.0
Re: [PATCH v8 5/5] perf arm-spe: Use SPE data source for neoverse cores
Quoted 6 lines · Ali Saidi
When synthesizing data from SPE, augment the type with source information for Arm Neoverse cores. The field is IMPLDEF but the Neoverse cores all use the same encoding. I can't find encoding information for any other SPE implementations to unify their choices with Arm's thus that is left for future work.
Quoted 7 lines
This change populates the
mem_lvl_numfor Neoverse cores as well as the deprecatedmem_lvlnamespace.Signed-off-by Ali Saidi <[email protected]>Reviewed-by German Gomez <[email protected]>Reviewed-by Leo Yan <[email protected]>Tested-by Leo Yan <[email protected]>
Thanks for updating, Ali. It looks good to me.
Since the store operations have been set as PERF_MEM_LVL_NA and PERF_MEM_LVLNUM_NA, this is right thing for me, I will update perf c2c patch set for statistics of store operations with PERF_MEM_LVL_NA.
Thanks,
Leo