Parallelize BTF type info generating of pahole
Subsystems (2)
v4 cover Parallelize BTF type info generating of pahole
Creating an instance of btf for each worker thread allows steal-function provided by pahole to add type info on multiple threads without a lock. The main thread merges the results of worker threads to the primary instance.
Copying data from per-thread btf instances to the primary instance is expensive now. However, there is a patch landed at the bpf-next repository. [1] With the patch for bpf-next and this patch, they drop total runtime to 5.4s from 6.0s with "-j4" on my device to generate BTF for Linux.
V4 includes following changes.
- Fix nits and typos.
- Rollback to calling btf__add_btf() at the main thread to simplify code. The reasons for that are additional lock making it complicated and doing w/o reusing btf_encoder being obvious slower.
[1] https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git/commit/?id=d81283d27266
[v1] https://lore.kernel.org/dwarves/[email protected]/
[v2] https://lore.kernel.org/dwarves/[email protected]/
[v3] https://lore.kernel.org/dwarves/[email protected]/
Kui-Feng Lee (4):
dwarf_loader: Receive per-thread data on worker threads.
dwarf_loader: Prepare and pass per-thread data to worker threads.
pahole: Use per-thread btf instances to avoid mutex locking.
libbpf: Update libbpf to a new revision.
btf_encoder.c | 25 ++++++---- btf_encoder.h | 2 + btf_loader.c | 4 +- ctf_loader.c | 2 +- dwarf_loader.c | 59 ++++++++++++++++++----- dwarves.h | 9 +++- lib/bpf | 2 +- pahole.c | 128 +++++++++++++++++++++++++++++++++++++++++++++---- pdwtags.c | 3 +- pfunct.c | 4 +- 10 files changed, 198 insertions(+), 40 deletions(-)
Signature · 1 line
2.30.2
PATCH v4 1/4 dwarf_loader: Receive per-thread data on worker threads.
Add arguments to steal and thread_exit callbacks of conf_load to receive per-thread data.
--- btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)
btf_loader.c +1 -1
@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);
ctf_loader.c +1 -1
@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);
dwarf_loader.c +2 -2
@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }
@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;
dwarves.h +3 -2
@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;
pahole.c +2 -1
@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;
pdwtags.c +2 -1
@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;
pfunct.c +3 -1
@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {if (function_name) {
Signature · 1 line
2.30.2
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Quoted 6 lines · Kui-Feng Lee
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Quoted 119 lines
btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)btf_loader.c +1
-1@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);ctf_loader.c +1
-1@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);dwarf_loader.c +2
-2@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;dwarves.h +3
-2@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;pahole.c +2
-1@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;pdwtags.c +2
-1@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;pfunct.c +3
-2@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {if (function_name) { --2.30.2
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Quoted 5 lines · Andrii Nakryiko
Quoted 4 lines · Kui-Feng Lee
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.
Quoted 7 lines
Quoted 2 lines
Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Acked-by Andrii Nakryiko <[email protected]>
Sorry for that! It is my first time receiving acks.
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
In reply to Andrii NakryikoEm Wed, Jan 26, 2022 at 11:55:25AM -0800, Andrii Nakryiko escreveu:
Quoted 10 lines · Kui-Feng Lee
Quoted 6 lines
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Yes, please do so.
I'll collect them this time, no need to resend.
- Arnaldo
Quoted 121 lines
Acked-by Andrii Nakryiko <[email protected]>Quoted 119 lines
btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)btf_loader.c +1
-1@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);ctf_loader.c +1
-1@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);dwarf_loader.c +2
-2@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;dwarves.h +3
-2@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;pahole.c +2
-1@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;pdwtags.c +2
-1@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;pfunct.c +3
-2@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {if (function_name) { --2.30.2
Signature · 1 line
- Arnaldo
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Quoted 10 lines · Arnaldo Carvalho de Melo
Em Wed, Jan 26, 2022 at 11:55:25AM -0800, Andrii Nakryiko escreveu:
Quoted 8 lines · Kui-Feng Lee
Quoted 6 lines
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.Signed-off-by Kui-Feng Lee <[email protected]>---
Quoted 7 lines
Quoted 2 lines
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Yes, please do so.
I'll collect them this time, no need to resend.
Hey, Arnaldo!
Any idea when these patches will make it into master branch? I see they are in tmp.master right now.
Quoted 127 lines
- Arnaldo
Quoted 117 lines
Acked-by Andrii Nakryiko <[email protected]>Quoted 115 lines
btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)btf_loader.c +1
-1@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);ctf_loader.c +1
-1@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);dwarf_loader.c +2
-2@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;dwarves.h +3
-2@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;pahole.c +2
-1@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;pdwtags.c +2
-1@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;pfunct.c +3
-1@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {Quoted 4 lines
Quoted 4 lines
pfunct.c +0
-1if (function_name) { --2.30.2
--
- Arnaldo
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Em Tue, Mar 08, 2022 at 03:45:03PM -0800, Andrii Nakryiko escreveu:
Quoted 17 lines
Quoted 15 lines · Arnaldo Carvalho de Melo
Em Wed, Jan 26, 2022 at 11:55:25AM -0800, Andrii Nakryiko escreveu:
Quoted 10 lines · Kui-Feng Lee
Quoted 6 lines
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Yes, please do so.
Quoted 7 lines
Quoted 2 lines
I'll collect them this time, no need to resend.
Hey, Arnaldo!
Any idea when these patches will make it into master branch? I see they are in tmp.master right now.
I did some minor fixups to the cset comment and to the code in the 'pahole --compile' new feature at the head of it and pushed all up, please check.
- Arnaldo
Quoted 127 lines
Quoted 127 lines
- Arnaldo
Quoted 117 lines
Acked-by Andrii Nakryiko <[email protected]>Quoted 115 lines
btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)btf_loader.c +1
-1@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);ctf_loader.c +1
-1@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);dwarf_loader.c +2
-2@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;dwarves.h +3
-2@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;pahole.c +2
-1@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;pdwtags.c +2
-1@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;pfunct.c +3
-1@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {Quoted 4 lines
Quoted 4 lines
pfunct.c +0
-1if (function_name) { --2.30.2
--
- Arnaldo
Signature · 1 line
- Arnaldo
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Quoted 22 lines · Arnaldo Carvalho de Melo
Em Tue, Mar 08, 2022 at 03:45:03PM -0800, Andrii Nakryiko escreveu:
Quoted 20 lines
Quoted 17 lines · Arnaldo Carvalho de Melo
Em Wed, Jan 26, 2022 at 11:55:25AM -0800, Andrii Nakryiko escreveu:
Quoted 10 lines · Kui-Feng Lee
Quoted 6 lines
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Yes, please do so.
I'll collect them this time, no need to resend.
Quoted 9 lines
Quoted 4 lines
Hey, Arnaldo!
Any idea when these patches will make it into master branch? I see they are in tmp.master right now.
I did some minor fixups to the cset comment and to the code in the 'pahole --compile' new feature at the head of it and pushed all up, please check.
I did check locally with latest pahole master, and it seems like something is wrong with generated BTF. I get three selftests failure if I use latest pahole compiled from master.
Kui-Feng, please take a look when you get a chance. Arnaldo, please hold off from releasing a new version for now.
Quoted 133 lines
- Arnaldo
Quoted 119 lines
Quoted 119 lines
- Arnaldo
Quoted 117 lines
Acked-by Andrii Nakryiko <[email protected]>Quoted 115 lines
btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)btf_loader.c +1
-1@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);ctf_loader.c +1
-1@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);dwarf_loader.c +2
-2@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;dwarves.h +3
-2@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;pahole.c +2
-1@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;pdwtags.c +2
-1@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;pfunct.c +3
-1@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
pfunct.c +0
-1if (function_name) { --2.30.2
--
- Arnaldo
--
- Arnaldo
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Quoted 35 lines · Andrii Nakryiko
Quoted 22 lines · Arnaldo Carvalho de Melo
Em Tue, Mar 08, 2022 at 03:45:03PM -0800, Andrii Nakryiko escreveu:
Quoted 20 lines
Quoted 17 lines · Arnaldo Carvalho de Melo
Em Wed, Jan 26, 2022 at 11:55:25AM -0800, Andrii Nakryiko escreveu:
Quoted 10 lines · Kui-Feng Lee
Quoted 6 lines
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Yes, please do so.
I'll collect them this time, no need to resend.
Quoted 9 lines
Quoted 4 lines
Hey, Arnaldo!
Any idea when these patches will make it into master branch? I see they are in tmp.master right now.
I did some minor fixups to the cset comment and to the code in the 'pahole --compile' new feature at the head of it and pushed all up, please check.
Quoted 6 lines
I did check locally with latest pahole master, and it seems like something is wrong with generated BTF. I get three selftests failure if I use latest pahole compiled from master.
Kui-Feng, please take a look when you get a chance. Arnaldo, please hold off from releasing a new version for now.
Specifically:
libbpf: prog 'kfunc_call_test1': BPF program load failed: Permission denied
libbpf: prog 'kfunc_call_test1': -- BEGIN PROG LOAD LOG -- Validating f1() func#1...
2: R1=ctx(off=0,imm=0) R10=fp0
; int __noinline f1(struct __sk_buff *skb)
2: (b4) w7 = -1 ; R7_w=P4294967295
; struct bpf_sock *sk = skb->sk;
3: (79) r1 = *(u64 *)(r1 +168) ;
R1_w=sock_common_or_null(id=1,off=0,imm=0)
; if (!sk)
4: (15) if r1 == 0x0 goto pc+26 ; R1_w=sock_common(off=0,imm=0)
; sk = bpf_sk_fullsock(sk);
5: (85) call bpf_sk_fullsock#95 ; R0_w=sock_or_null(id=2,off=0,imm=0)
6: (bf) r6 = r0 ;
R0_w=sock_or_null(id=2,off=0,imm=0) R6_w=sock_or_null(id=2,off=0,imm=0)
; if (!sk)
7: (15) if r6 == 0x0 goto pc+23 ; R6_w=sock(off=0,imm=0)
; bpf_get_smp_processor_id());
8: (85) call bpf_get_smp_processor_id#8 ; R0=Pscalar()
; active = (int *)bpf_per_cpu_ptr(&bpf_prog_active,
9: (18) r1 = 0x275e0 ; R1_w=rdonly_mem(off=0,imm=0)
11: (bc) w2 = w0 ; R0=Pscalar()
R2_w=Pscalar(umax=4294967295,var_off=(0x0; 0xffffffff))
12: (85) call bpf_per_cpu_ptr#153
R1 type=rdonly_mem expected=percpu_ptr_
processed 10 insns (limit 1000000) max_states_per_insn 0 total_states 1 peak_states 1 mark_read 1
-- END PROG LOAD LOG --
libbpf: failed to load program 'kfunc_call_test1'
libbpf: failed to load object 'kfunc_call_test_subprog'
libbpf: failed to load BPF skeleton 'kfunc_call_test_subprog': -13 test_subprog:FAIL:skel unexpected error: -13
#77/2 kfunc_call/subprog:FAIL
test_subprog_lskel:FAIL:skel unexpected error: -13
#77/3 kfunc_call/subprog_lskel:FAIL
test_ksyms_btf:PASS:btf_exists 0 nsec test_basic:PASS:kallsyms_fopen 0 nsec test_basic:PASS:ksym_find 0 nsec test_basic:PASS:kallsyms_fopen 0 nsec test_basic:PASS:ksym_find 0 nsec
libbpf: prog 'handler': BPF program load failed: Permission denied
libbpf: prog 'handler': -- BEGIN PROG LOAD LOG --
arg#0 reference type('UNKNOWN ') size cannot be determined: -22 0: R1=ctx(off=0,imm=0) R10=fp0
; out__bpf_prog_active_addr = (__u64)&bpf_prog_active;
0: (18) r1 = 0xffffc9000071a008 ; R1_w=map_value(off=8,ks=4,vs=36,imm=0) 2: (18) r2 = 0x275e0 ; R2_w=rdonly_mem(off=0,imm=0) 4: (7b) *(u64 *)(r1 +0) = r2 ; R1_w=map_value(off=8,ks=4,vs=36,imm=0) R2_w=rdonly_mem(off=0,imm=0) ; out__runqueues_addr = (__u64)&runqueues; 5: (18) r1 = 0xffffc9000071a000 ; R1_w=map_value(off=0,ks=4,vs=36,imm=0) 7: (18) r2 = 0x2b140 ; R2_w=ptr_rq(off=0,imm=0) 9: (7b) *(u64 *)(r1 +0) = r2 ;
R1_w=map_value(off=0,ks=4,vs=36,imm=0) R2_w=ptr_rq(off=0,imm=0) ; cpu = bpf_get_smp_processor_id();
10: (85) call bpf_get_smp_processor_id#8 ; R0_w=scalar()
11: (bc) w6 = w0 ; R0_w=scalar()
R6_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff))
; rq = (struct rq *)bpf_per_cpu_ptr(&runqueues, cpu);
12: (18) r1 = 0x2b140 ; R1_w=ptr_rq(off=0,imm=0)
14: (bc) w2 = w6 ;
R2_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff)) R6_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff))
15: (85) call bpf_per_cpu_ptr#153
R1 type=ptr_ expected=percpu_ptr_
processed 11 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0
-- END PROG LOAD LOG --
libbpf: failed to load program 'handler'
libbpf: failed to load object 'test_ksyms_btf'
libbpf: failed to load BPF skeleton 'test_ksyms_btf': -13 test_basic:FAIL:skel_open failed to open and load skeleton #79/1 ksyms_btf/basic:FAIL
#80 ksyms_module:FAIL test_ksyms_module_lskel:FAIL:test_ksyms_module_lskel__open_and_load unexpected error: -13
#80/1 ksyms_module/lskel:FAIL
libbpf: prog 'load': BPF program load failed: Permission denied
libbpf: prog 'load': -- BEGIN PROG LOAD LOG --
0: R1=ctx(off=0,imm=0) R10=fp0
; if (x)
0: (18) r1 = 0xffffc9000109e000 ; R1_w=map_value(off=0,ks=4,vs=4,imm=0)
2: (61) r1 = *(u32 *)(r1 +0) ; R1_w=0
; if (x)
3: (16) if w1 == 0x0 goto pc+1 ; R1_w=P0
; bpf_testmod_test_mod_kfunc(42);
5: (b4) w1 = 42 ; R1_w=42
6: (85) call bpf_testmod_test_mod_kfunc#123710
; out_bpf_testmod_ksym = *(int *)bpf_this_cpu_ptr(&bpf_testmod_ksym_percpu);
7: (18) r1 = 0x2c3e8 ; R1_w=rdonly_mem(off=0,imm=0)
9: (85) call bpf_this_cpu_ptr#154
R1 type=rdonly_mem expected=percpu_ptr_
processed 7 insns (limit 1000000) max_states_per_insn 0 total_states 0 peak_states 0 mark_read 0
-- END PROG LOAD LOG --
libbpf: failed to load program 'load'
libbpf: failed to load object 'test_ksyms_module'
libbpf: failed to load BPF skeleton 'test_ksyms_module': -13 test_ksyms_module_libbpf:FAIL:test_ksyms_module__open unexpected error: -13 #80/2 ksyms_module/libbpf:FAIL
So it seems like something broke around per-CPU variable info generation.
Quoted 135 lines
Quoted 135 lines
- Arnaldo
Quoted 119 lines
Quoted 119 lines
- Arnaldo
Quoted 117 lines
Acked-by Andrii Nakryiko <[email protected]>Quoted 115 lines
btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)btf_loader.c +1
-1@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);ctf_loader.c +1
-1@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);dwarf_loader.c +2
-2@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;dwarves.h +3
-2@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;pahole.c +2
-1@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;pdwtags.c +2
-1@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;pfunct.c +3
-1@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
pfunct.c +0
-1if (function_name) { --2.30.2
--
- Arnaldo
--
- Arnaldo
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Quoted 45 lines · Andrii Nakryiko
Quoted 43 lines · Andrii Nakryiko
Quoted 39 lines · Arnaldo Carvalho de Melo
Em Tue, Mar 08, 2022 at 03:45:03PM -0800, Andrii Nakryiko escreveu:
Quoted 21 lines
Quoted 19 lines · Arnaldo Carvalho de Melo
Em Wed, Jan 26, 2022 at 11:55:25AM -0800, Andrii Nakryiko escreveu:
Quoted 7 lines
Quoted 5 lines · Kui-Feng Lee
Add arguments to steal and
thread_exitcallbacks ofconf_loadto
receive per-thread data.Quoted 6 lines
Quoted 2 lines
Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did
some drastic changes to already acked patches.Yes, please do so.
Quoted 8 lines
Quoted 2 lines
I'll collect them this time, no need to resend.
Hey, Arnaldo!
Any idea when these patches will make it into master branch? I see
they are in tmp.master right now.I did some minor fixups to the cset comment and to the code in the
'pahole --compile' new feature at the head of it and pushed all up,
please check.
Quoted 7 lines
Quoted 7 lines
I did check locally with latest pahole master, and it seems like something is wrong with generated BTF. I get three selftests failure
if I use latest pahole compiled from master.Kui-Feng, please take a look when you get a chance. Arnaldo, please hold off from releasing a new version for now.
I just figure out the root cuase.
It caused by missing info from percpu_secinfo when collecting data from threads. The encoder stores it separatedly from struct btf, and we have separated different encoders for threads. They are not merged together. I will fixed it ASAP.
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Em Mon, Mar 21, 2022 at 07:06:06PM +0000, Kui-Feng Lee escreveu:
Quoted 7 lines · Andrii Nakryiko
Quoted 5 lines · Andrii Nakryiko
Quoted 4 lines
I did check locally with latest pahole master, and it seems like something is wrong with generated BTF. I get three selftests failure
if I use latest pahole compiled from master.
Quoted 8 lines
Quoted 2 lines
Quoted 2 lines
Kui-Feng, please take a look when you get a chance. Arnaldo, please hold off from releasing a new version for now.
I just figure out the root cuase.
It caused by missing info frompercpu_secinfowhen collecting data from threads. The encoder stores it separatedly from struct btf, and we have separated different encoders for threads. They are not merged together. I will fixed it ASAP.
Cool, looking forward to the fix.
- Arnaldo
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
In reply to Andrii NakryikoEm Wed, Mar 09, 2022 at 04:14:40PM -0800, Andrii Nakryiko escreveu:
Quoted 34 lines
Quoted 22 lines · Arnaldo Carvalho de Melo
Em Tue, Mar 08, 2022 at 03:45:03PM -0800, Andrii Nakryiko escreveu:
Quoted 20 lines
Quoted 17 lines · Arnaldo Carvalho de Melo
Em Wed, Jan 26, 2022 at 11:55:25AM -0800, Andrii Nakryiko escreveu:
Quoted 10 lines · Kui-Feng Lee
Quoted 6 lines
Add arguments to steal and
thread_exitcallbacks ofconf_loadto receive per-thread data.Signed-off-by Kui-Feng Lee <[email protected]>---
Please carry over acks you got on previous revisions, unless you did some drastic changes to already acked patches.
Yes, please do so.
I'll collect them this time, no need to resend.
Quoted 9 lines
Quoted 4 lines
Hey, Arnaldo!
Any idea when these patches will make it into master branch? I see they are in tmp.master right now.
I did some minor fixups to the cset comment and to the code in the 'pahole --compile' new feature at the head of it and pushed all up, please check.
Quoted 6 lines
I did check locally with latest pahole master, and it seems like something is wrong with generated BTF. I get three selftests failure if I use latest pahole compiled from master.
Kui-Feng, please take a look when you get a chance. Arnaldo, please hold off from releasing a new version for now.
Sure, will wait.
Also will try to test/fix this as time permits.
- Arnaldo
Quoted 134 lines
Quoted 134 lines
- Arnaldo
Quoted 119 lines
Quoted 119 lines
- Arnaldo
Quoted 117 lines
Acked-by Andrii Nakryiko <[email protected]>Quoted 115 lines
btf_loader.c | 2 +- ctf_loader.c | 2 +- dwarf_loader.c | 4 ++-- dwarves.h | 5 +++-- pahole.c | 3 ++- pdwtags.c | 3 ++- pfunct.c | 4 +++- 7 files changed, 14 insertions(+), 9 deletions(-)btf_loader.c +1
-1@@ -624,7 +624,7 @@ static int cus__load_btf(struct cus *cus, struct conf_load *conf, const char *fi* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);ctf_loader.c +1
-1@@ -722,7 +722,7 @@ int ctf__load_file(struct cus *cus, struct conf_load *conf,* The app stole this cu, possibly deleting it, * so forget about it */ -if (conf && conf->steal && conf->steal (cu, conf)) +if (conf && conf->steal && conf->steal (cu, conf,NULL ))return 0 ;cus__add (cus, cu);dwarf_loader.c +2
-2@@ -2686,7 +2686,7 @@ static int cu__finalize(struct cu *cu, struct conf_load *conf) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf); +return conf->steal (cu, conf,NULL ); }return LSK__KEEPIT ; }@@ -2930,7 +2930,7 @@ static void *dwarf_cus__process_cu_thread(void *arg) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit () !=0 ) +if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;dwarves.h +3
-2@@ -48,8 +48,9 @@ struct conf_fprintf;*/ struct conf_load {enum load_steal_kind (*steal)(struct cu *cu, -struct conf_load *conf); -int (*thread_exit)(void ); +struct conf_load *conf, +void *thr_data); +int (*thread_exit)(struct conf_load *conf,void *thr_data);void *cookie;char *format_path;int nr_jobs;pahole.c +2
-1@@ -2799,7 +2799,8 @@ out:static struct type_instance *header;static enum load_steal_kind pahole_stealer (struct cu *cu, -struct conf_load *conf_load) +struct conf_load *conf_load, +void *thr_data) {int ret =LSK__DELETE ;pdwtags.c +2
-1@@ -72,7 +72,8 @@ static int cu__emit_tags(struct cu *cu) }static enum load_steal_kind pdwtags_stealer (struct cu *cu, -struct conf_load *conf_load__maybe_unused ) +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {cu__emit_tags (cu);return LSK__DELETE ;pfunct.c +3
-1@@ -489,7 +489,9 @@ int elf_symtabs__show(char *filenames[])return EXIT_SUCCESS ; } -static enum load_steal_kind pfunct_stealer (struct cu *cu,struct conf_load *conf_load__maybe_unused ) +static enum load_steal_kind pfunct_stealer (struct cu *cu, +struct conf_load *conf_load__maybe_unused , +void *thr_data__maybe_unused ) {Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
pfunct.c +0
-1if (function_name) { --2.30.2
--
- Arnaldo
--
- Arnaldo
Signature · 1 line
- Arnaldo
Re: [PATCH dwarves v4 1/4] dwarf_loader: Receive per-thread data on worker threads.
Em Mon, Mar 14, 2022 at 12:38:59PM -0300, Arnaldo Carvalho de Melo escreveu:
Quoted 11 lines
Em Wed, Mar 09, 2022 at 04:14:40PM -0800, Andrii Nakryiko escreveu:
Quoted 6 lines
I did check locally with latest pahole master, and it seems like something is wrong with generated BTF. I get three selftests failure if I use latest pahole compiled from master.
Kui-Feng, please take a look when you get a chance. Arnaldo, please hold off from releasing a new version for now.
Sure, will wait.
Also will try to test/fix this as time permits.
The tests at
https://github.com/libbpf/libbpf/actions/workflows/pahole.yml are all happy with what we have, so maybe a new test is needed for checking these three selftests into what is being tested there?
And just so that we're all on the same page, this is the current status in the various branches and git servers:
⬢[acme@toolbox pahole]$ git log --oneline -1065d7273668ded59b (HEAD -> master, quaco/master, quaco/HEAD, github/tmp.master, github/next, github/master, acme/tmp.master, acme/next, acme/master) pahole: Introduce --compile to produce a compilable output 4d004e2314f3252e core: Ditch 'dwarves__active_loader' extern declaration, it was nuked4f332dbfd02072e4 emit: Notice type shadowing, i.e. multiple types with the same name (enum, struct, union, etc)0a82f74ce25a5904 core: Make type->packed_attributes_inferred a one bit memberfac821246c582299 core: type->declaration is just one bit, make it a bitfield member742f04f89da03665 emit: Search for data structures using its type in addition to its name32cc1481721c4b11 fprintf: Consider enumerations without members as forward declarations6afc296eeb180e25 emit: Fix printing typedef of nameless struct/union49a2dd657728675b fprintf: Check if conf->conf_fprintf is not NULL when resolving cacheline_size46cec35ff0411e0f fprintf: Fix division by zero for uninitialized conf_fprintf->cacheline_size field
⬢[acme@toolbox pahole]$
- Arnaldo
PATCH v4 2/4 dwarf_loader: Prepare and pass per-thread data to worker threads.
In reply to Kui-Feng LeeAdd interfaces to allow users of dwarf_loader to prepare and pass per-thread data to steal-functions running on worker threads.
--- dwarf_loader.c | 59 +++++++++++++++++++++++++++++++++++++++----------- dwarves.h | 4 ++++ 2 files changed, 50 insertions(+), 13 deletions(-)
dwarf_loader.c +46 -13
@@ -2682,18 +2682,18 @@ static int class_member__cache_byte_size(struct tag *tag, struct cu *cu,return 0 ; } -static int cu__finalize (struct cu *cu,struct conf_load *conf) +static int cu__finalize (struct cu *cu,struct conf_load *conf,void *thr_data) {cu__for_all_tags (cu, class_member__cache_byte_size, conf);if (conf && conf->steal) { -return conf->steal (cu, conf,NULL ); +return conf->steal (cu, conf, thr_data); }return LSK__KEEPIT ; } -static int cus__finalize (struct cus *cus,struct cu *cu,struct conf_load *conf) +static int cus__finalize (struct cus *cus,struct cu *cu,struct conf_load *conf,void *thr_data) { -int lsk =cu__finalize (cu, conf); +int lsk =cu__finalize (cu, conf, thr_data);switch (lsk) {case LSK__DELETE :cu__delete (cu);
@@ -2862,7 +2862,13 @@ struct dwarf_cus {struct dwarf_cu *type_dcu; }; -static int dwarf_cus__create_and_process_cu (struct dwarf_cus *dcus,Dwarf_Die *cu_die,uint8_t pointer_size) +struct dwarf_thread { +struct dwarf_cus *dcus; +void *data; +}; + +static int dwarf_cus__create_and_process_cu (struct dwarf_cus *dcus,Dwarf_Die *cu_die, +uint8_t pointer_size,void *thr_data) {/* * DW_AT_name in DW_TAG_compile_unit can be NULL, first seen in:
@@ -2884,7 +2890,7 @@ static int dwarf_cus__create_and_process_cu(struct dwarf_cus *dcus, Dwarf_Die *c cu->dfops = &dwarf__ops;if (die__process_and_recode (cu_die, cu, dcus->conf) !=0 || -cus__finalize (dcus->cus, cu, dcus->conf) ==LSK__STOP_LOADING ) +cus__finalize (dcus->cus, cu, dcus->conf, thr_data) ==LSK__STOP_LOADING )return DWARF_CB_ABORT ;return DWARF_CB_OK ;
@@ -2918,7 +2924,8 @@ out_unlock:static void *dwarf_cus__process_cu_thread (void *arg) { -struct dwarf_cus *dcus = arg; +struct dwarf_thread *dthr = arg; +struct dwarf_cus *dcus = dthr->dcus;uint8_t pointer_size, offset_size;Dwarf_Die die_mem, *cu_die;
@@ -2926,11 +2933,13 @@ static void *dwarf_cus__process_cu_thread(void *arg)if (cu_die ==NULL )break ; -if (dwarf_cus__create_and_process_cu (dcus, cu_die, pointer_size) ==DWARF_CB_ABORT ) +if (dwarf_cus__create_and_process_cu (dcus, cu_die, + pointer_size, dthr->data) ==DWARF_CB_ABORT ) goto out_abort; } -if (dcus->conf->thread_exit && dcus->conf->thread_exit (dcus->conf,NULL ) !=0 ) +if (dcus->conf->thread_exit && + dcus->conf->thread_exit (dcus->conf, dthr->data) !=0 ) goto out_abort;return (void *)DWARF_CB_OK ;
@@ -2941,10 +2950,26 @@ out_abort:static int dwarf_cus__threaded_process_cus (struct dwarf_cus *dcus) {pthread_t threads[dcus->conf->nr_jobs]; +struct dwarf_thread dthr[dcus->conf->nr_jobs]; +void *thread_data[dcus->conf->nr_jobs]; +int res;int i; +if (dcus->conf->threads_prepare) { + res = dcus->conf->threads_prepare (dcus->conf, dcus->conf->nr_jobs, thread_data); +if (res !=0 ) +return res; + }else { +memset (thread_data,0 ,sizeof (void *) * dcus->conf->nr_jobs); + } +for (i =0 ; i < dcus->conf->nr_jobs; ++i) { - dcus->error =pthread_create (&threads[i],NULL , dwarf_cus__process_cu_thread, dcus); + dthr[i].dcus = dcus; + dthr[i].data = thread_data[i]; + + dcus->error =pthread_create (&threads[i],NULL , + dwarf_cus__process_cu_thread, + &dthr[i]);if (dcus->error) goto out_join; }
@@ -2960,6 +2985,13 @@ out_join: dcus->error = (long )res; } +if (dcus->conf->threads_collect) { + res = dcus->conf->threads_collect (dcus->conf, dcus->conf->nr_jobs, + thread_data, dcus->error); +if (dcus->error ==0 ) + dcus->error = res; + } +return dcus->error; }
@@ -2976,7 +3008,8 @@ static int __dwarf_cus__process_cus(struct dwarf_cus *dcus)if (cu_die ==NULL )break ; -if (dwarf_cus__create_and_process_cu (dcus, cu_die, pointer_size) ==DWARF_CB_ABORT ) +if (dwarf_cus__create_and_process_cu (dcus, cu_die, + pointer_size,NULL ) ==DWARF_CB_ABORT )return DWARF_CB_ABORT ; dcus->off = noff;
@@ -3070,7 +3103,7 @@ static int cus__merge_and_process_cu(struct cus *cus, struct conf_load *conf,if (cu__resolve_func_ret_types (cu) !=LSK__KEEPIT ) goto out_abort; -if (cus__finalize (cus, cu, conf) ==LSK__STOP_LOADING ) +if (cus__finalize (cus, cu, conf,NULL ) ==LSK__STOP_LOADING ) goto out_abort;return 0 ;
@@ -3102,7 +3135,7 @@ static int cus__load_module(struct cus *cus, struct conf_load *conf, }if (type_cu !=NULL ) { - type_lsk =cu__finalize (type_cu, conf); + type_lsk =cu__finalize (type_cu, conf,NULL );if (type_lsk ==LSK__KEEPIT ) {cus__add (cus, type_cu); }
dwarves.h +4 -0
@@ -71,6 +71,10 @@ struct conf_load {const char *kabi_prefix;struct btf *base_btf;struct conf_fprintf *conf_fprintf; +int (*threads_prepare)(struct conf_load *conf,int nr_threads, +void **thr_data); +int (*threads_collect)(struct conf_load *conf,int nr_threads, +void **thr_data,int error); };/** struct conf_fprintf - hints to the __fprintf routines
Signature · 1 line
2.30.2
Re: [PATCH dwarves v4 2/4] dwarf_loader: Prepare and pass per-thread data to worker threads.
Quoted 6 lines · Kui-Feng Lee
Add interfaces to allow users of
dwarf_loaderto prepare and pass per-thread data to steal-functions running on worker threads.Signed-off-by Kui-Feng Lee <[email protected]>---
Quoted 4 lines
dwarf_loader.c | 59 +++++++++++++++++++++++++++++++++++++++----------- dwarves.h | 4 ++++ 2 files changed, 50 insertions(+), 13 deletions(-)
[...]
PATCH v4 3/4 pahole: Use per-thread btf instances to avoid mutex locking.
In reply to Kui-Feng LeeCreate an instance of btf for each worker thread, and add type info to the local btf instance in the steal-function of pahole without mutex acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.
--- btf_encoder.c | 5 ++ btf_encoder.h | 2 + pahole.c | 125 ++++++++++++++++++++++++++++++++++++++++++++++---- 3 files changed, 124 insertions(+), 8 deletions(-)
btf_encoder.c +5 -0
@@ -1529,3 +1529,8 @@ int btf_encoder__encode_cu(struct btf_encoder *encoder, struct cu *cu) out:return err; } + +struct btf *btf_encoder__btf (struct btf_encoder *encoder) +{ +return encoder->btf; +}
btf_encoder.h +2 -0
@@ -29,4 +29,6 @@ struct btf_encoder *btf_encoders__first(struct list_head *encoders);struct btf_encoder *btf_encoders__next (struct btf_encoder *encoder); +struct btf *btf_encoder__btf (struct btf_encoder *encoder); +#endif /* _BTF_ENCODER_H_ */
pahole.c +117 -8
159 lines not on this page · raw message
Signature · 1 line
2.30.2
Re: [PATCH dwarves v4 3/4] pahole: Use per-thread btf instances to avoid mutex locking.
Quoted 8 lines · Kui-Feng Lee
Create an instance of btf for each worker thread, and add type info to the local btf instance in the steal-function of pahole without mutex acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.
Signed-off-by Kui-Feng Lee <[email protected]>---
There are still unnecessary casts and missing {} in the else branch, but I'll let Arnaldo decide or fix it up.
Once this lands, can you please send kernel patch to use -j if pahole support it during the kernel build? See scripts/pahole-version.sh and scripts/link-vmlinux.sh for how pahole is set up and used in the kernel. Thanks!
Quoted 5 lines
btf_encoder.c | 5 ++ btf_encoder.h | 2 + pahole.c | 125 ++++++++++++++++++++++++++++++++++++++++++++++---- 3 files changed, 124 insertions(+), 8 deletions(-)
[...]
Re: [PATCH dwarves v4 3/4] pahole: Use per-thread btf instances to avoid mutex locking.
Quoted 12 lines · Andrii Nakryiko
Quoted 10 lines · Kui-Feng Lee
Create an instance of btf for each worker thread, and add type info to
the local btf instance in the steal-function of pahole without mutex
acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.Signed-off-by Kui-Feng Lee <[email protected]>---
Quoted 7 lines
There are still unnecessary casts and missing {} in the else branch, but I'll let Arnaldo decide or fix it up.
Once this lands, can you please send kernel patch to use -j if pahole support it during the kernel build? See
scripts/pahole-version.shandscripts/link-vmlinux.shfor how pahole is set up and used in the kernel. Thanks!
Sure!
Re: [PATCH dwarves v4 3/4] pahole: Use per-thread btf instances to avoid mutex locking.
In reply to Andrii NakryikoEm Wed, Jan 26, 2022 at 11:58:27AM -0800, Andrii Nakryiko escreveu:
Quoted 7 lines · Kui-Feng Lee
Quoted 6 lines
Create an instance of btf for each worker thread, and add type info to the local btf instance in the steal-function of pahole without mutex acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.
Quoted 5 lines
Quoted 2 lines
Signed-off-by Kui-Feng Lee <[email protected]>---
There are still unnecessary casts and missing {} in the else branch, but I'll let Arnaldo decide or fix it up.
Yeah, I'll do some sanitization, thanks for all the review!
- Arnaldo
Quoted 14 lines
Once this lands, can you please send kernel patch to use -j if pahole support it during the kernel build? See
scripts/pahole-version.shandscripts/link-vmlinux.shfor how pahole is set up and used in the kernel. Thanks!Acked-by Andrii Nakryiko <[email protected]>Quoted 5 lines
btf_encoder.c | 5 ++ btf_encoder.h | 2 + pahole.c | 125 ++++++++++++++++++++++++++++++++++++++++++++++---- 3 files changed, 124 insertions(+), 8 deletions(-)[...]
Signature · 1 line
- Arnaldo
Arnaldo Carvalho de Melo
AppliedRe: [PATCH dwarves v4 3/4] pahole: Use per-thread btf instances to avoid mutex locking.
In reply to Andrii NakryikoEm Wed, Jan 26, 2022 at 11:58:27AM -0800, Andrii Nakryiko escreveu:
Quoted 10 lines · Kui-Feng Lee
Quoted 8 lines
Create an instance of btf for each worker thread, and add type info to the local btf instance in the steal-function of pahole without mutex acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.
Signed-off-by Kui-Feng Lee <[email protected]>---
Quoted 7 lines
There are still unnecessary casts and missing {} in the else branch, but I'll let Arnaldo decide or fix it up.
Once this lands, can you please send kernel patch to use -j if pahole support it during the kernel build? See
scripts/pahole-version.shandscripts/link-vmlinux.shfor how pahole is set up and used in the kernel. Thanks!
I also tweaked this:
pahole.c +1 -1
@@ -2887,13 +2887,13 @@ static enum load_steal_kind pahole_stealer(struct cu *cu,static pthread_mutex_t btf_lock =PTHREAD_MUTEX_INITIALIZER ;struct btf_encoder *encoder; +pthread_mutex_lock (&btf_lock);/* * FIXME: * * This should be really done at main(), but since in the current codebase only at this * point we'll have cu->elf setup... */ -pthread_mutex_lock (&btf_lock);if (!btf_encoder) {/* * btf_encoder is the primary encoder.
⬢[acme@toolbox pahole]$
As moving it to after that comment will only make the patch a bit larger, changing nothing.
+++ b/pahole.c
@@ -2900,11 +2900,9 @@ static enum load_steal_kind pahole_stealer(struct cu *cu,
* And, it is used by the thread
* create it.
*/
- btf_encoder = btf_encoder__new(cu, detached_btf_filename,
- conf_load->base_btf,
- skip_encoding_btf_vars,
- btf_encode_force,
- btf_gen_floats, global_verbose);
+ btf_encoder = btf_encoder__new(cu, detached_btf_filename, conf_load->base_btf, skip_encoding_btf_vars,
+ btf_encode_force, btf_gen_floats, global_verbose);
+
if (btf_encoder && thr_data) {
struct thread_data *thread = thr_data;⬢[acme@toolbox pahole]$
i.e. cosmetic stuff to make the patch smaller by keeping preexisting lines as-is.
And the missing {} Andrii noticed:
pahole.c +2 -1
@@ -2937,8 +2937,9 @@ static enum load_steal_kind pahole_stealer(struct cu *cu, thread->btf =btf_encoder__btf (thread->encoder); } encoder = thread->encoder; - }else + }else { encoder = btf_encoder; + }if (btf_encoder__encode_cu (encoder, cu)) {fprintf (stderr,"Encountered error while encoding BTF.\n" );
⬢[acme@toolbox pahole]$
I'll look at the needless casts and push it to the 'next' and tmp.master branches so that libbpf's CI can have a chance to test it.
I also added 'perf stat' results for 1.21 (the one in this fedora 34 workstation), 1.23 (parallel DWARF loading) and with your patches + the libbpf update as a committer testing section, please add performance numbers in in future work.
Thanks, applied!
- Arnaldo
Re: [PATCH dwarves v4 3/4] pahole: Use per-thread btf instances to avoid mutex locking.
Em Thu, Jan 27, 2022 at 05:12:02PM -0300, Arnaldo Carvalho de Melo escreveu:
Quoted 9 lines
Em Wed, Jan 26, 2022 at 11:58:27AM -0800, Andrii Nakryiko escreveu:
Quoted 8 lines · Kui-Feng Lee
Quoted 4 lines
Create an instance of btf for each worker thread, and add type info to the local btf instance in the steal-function of pahole without mutex acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.
There are still unnecessary casts and missing {} in the else branch, but I'll let Arnaldo decide or fix it up.
So its just one unneeded cast as thr_data here is just a 'void *':
pahole.c +1 -1
@@ -2924,7 +2924,7 @@ static enum load_steal_kind pahole_stealer(struct cu *cu,* avoids copying the data collected by the first thread. */ if (thr_data) { -struct thread_data *thread = (struct thread_data *)thr_data; +struct thread_data *thread = thr_data;if (thread->encoder ==NULL ) { thread->encoder =
This other is needed as it is a "void **":
@@ -2832,7 +2832,7 @@ static int pahole_thread_exit(struct conf_load *conf, void *thr_data)
static int pahole_threads_collect(struct conf_load *conf, int nr_threads, void **thr_data,
int error)
{
- struct thread_data **threads = (struct thread_data **)thr_data;
+ struct thread_data **threads = thr_data;
int i;
int err = 0;Removing it:
/var/home/acme/git/pahole/pahole.c: In function ‘pahole_threads_collect’:/var/home/acme/git/pahole/pahole.c:2835:40: warning: initialization of ‘struct thread_data **’ from incompatible pointer type ‘void **’ [-Wincompatible-pointer-types]
2835 | struct thread_data **threads = thr_data;
| ^~~~~~~~
And I did some more profiling, now the focus should go to elfutils:
⬢[acme@toolbox pahole]$ perf report --no-children -s dso --call-graph none 2> /dev/null | head -20 # To display the perf.data header info, please use --header/--header-only options. #
#
# Total Lost Samples: 0
#
# Samples: 27K of event 'cycles:u'
# Event count (approx.): 27956766207
#
# Overhead Shared Object
# ........ ...................
#
46.70% libdwarves.so.1.0.0
39.84% libdw-0.186.so
9.70% libc-2.33.so
2.14% libpthread-2.33.so
1.47% [unknown]
0.09% ld-2.33.so
0.06% libelf-0.186.so
0.00% libcrypto.so.1.1.1l
0.00% libk5crypto.so.3.1
⬢[acme@toolbox pahole]$$ perf report -g graph,0.5,2 --stdio --no-children -s dso --dso libdw-0.186.so
# To display the perf.data header info, please use --header/--header-only options. #
# dso: libdw-0.186.so
#
# Total Lost Samples: 0
#
# Samples: 27K of event 'cycles:u'
# Event count (approx.): 27956766207
#
# Overhead Shared Object
# ........ ..............
#
39.84% libdw-0.186.so
|
|--25.66%--__libdw_find_attr
| |
| |--20.96%--__dwarf_attr_internal (inlined)
| | |
| | |--10.94%--attr_numeric
| | | |
| | | |--9.96%--die__process_class
| | | | __die__process_tag
| | | | die__process_unit
| | | | die__process
| | | | dwarf_cus__create_and_process_cu
| | | | dwarf_cus__process_cu_thread
| | | | start_thread
| | | | __GI___clone (inlined)
| | | |
| | | --0.61%--type__init
| | | __die__process_tag
| | | die__process_unit
| | | die__process
| | | dwarf_cus__create_and_process_cu
| | | dwarf_cus__process_cu_thread
| | | start_thread
| | | __GI___clone (inlined)
| | |
| | |--5.43%--attr_type
| | | |
| | | --4.94%--tag__init
| | | |
| | | |--2.60%--die__process_class
| | | | __die__process_tag
| | | | die__process_unit
| | | | die__process
| | | | dwarf_cus__create_and_process_cu
| | | | dwarf_cus__process_cu_thread
| | | | start_thread
| | | | __GI___clone (inlined)
| | | |
| | | |--0.99%--__die__process_tag
| | | | |
| | | | --0.98%--die__process_unit
| | | | die__process
| | | | dwarf_cus__create_and_process_cu
| | | | dwarf_cus__process_cu_thread
| | | | start_thread
| | | | __GI___clone (inlined)
| |
| |--4.01%--__dwarf_siblingof_internal (inlined)
| | |
| | |--1.41%--die__process_class
| | | __die__process_tag
| | | die__process_unit
| | | die__process
| | | dwarf_cus__create_and_process_cu
| | | dwarf_cus__process_cu_thread
| | | start_thread
| | | __GI___clone (inlined)
| | |
| | |--1.02%--die__process
| | | dwarf_cus__create_and_process_cu
| | | dwarf_cus__process_cu_thread
| | | start_thread
| | | __GI___clone (inlined)
|
|--2.38%--__libdw_form_val_compute_len
| __libdw_find_attr
| |
| --1.86%--__dwarf_attr_internal (inlined)
| |
| |--0.94%--attr_numeric
| | |
| | --0.83%--die__process_class
| | __die__process_tag
| | die__process_unit
| | die__process
| | dwarf_cus__create_and_process_cu
| | dwarf_cus__process_cu_thread
| | start_thread
| | __GI___clone (inlined)
| |
| --0.56%--attr_type
| |
| --0.53%--tag__init#
# (Tip: If you have debuginfo enabled, try: perf report -s sym,srcline) #
This find_attr thing needs improvements, its a linear search AFAIK, some hashtable could do wonders, I guess.
Mark, was this considered at some point?
⬢[acme@toolbox pahole]$ rpm -q elfutils-libs
elfutils-libs-0.186-1.fc34.x86_64
Andrii https://github.com/libbpf/libbpf/actions/workflows/pahole.yml is in failure mode for 3 days, and only yesterday I pushed these changes, seems unrelated to pahole:
Tests exit status: 1
Test Results:
bpftool: PASS
test_progs: FAIL (returned 1)
test_progs-no_alu32: FAIL (returned 1)
test_verifier: PASS
shutdown: CLEAN
Error: Process completed with exit code 1.
Can you please check?
- Arnaldo
Re: [PATCH dwarves v4 3/4] pahole: Use per-thread btf instances to avoid mutex locking.
Quoted 76 lines · Arnaldo Carvalho de Melo
Em Thu, Jan 27, 2022 at 05:12:02PM -0300, Arnaldo Carvalho de Melo escreveu:
Quoted 9 lines
Em Wed, Jan 26, 2022 at 11:58:27AM -0800, Andrii Nakryiko escreveu:
Quoted 8 lines · Kui-Feng Lee
Quoted 4 lines
Create an instance of btf for each worker thread, and add type info to the local btf instance in the steal-function of pahole without mutex acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.
There are still unnecessary casts and missing {} in the else branch, but I'll let Arnaldo decide or fix it up.
So its just one unneeded cast as
thr_datahere is just a 'void *':pahole.c +1
-1@@ -2924,7 +2924,7 @@ static enum load_steal_kind pahole_stealer(struct cu *cu,* avoids copying the data collected by the first thread. */ if (thr_data) { -struct thread_data *thread = (struct thread_data *)thr_data; +struct thread_data *thread = thr_data;if (thread->encoder ==NULL ) { thread->encoder =This other is needed as it is a "void **":
pahole.c +1
-1@@ -2832,7 +2832,7 @@ static int pahole_thread_exit(struct conf_load *conf, void *thr_data)static int pahole_threads_collect (struct conf_load *conf,int nr_threads,void **thr_data,int error) { -struct thread_data **threads = (struct thread_data **)thr_data; +struct thread_data **threads = thr_data;int i;int err =0 ;Removing it:
/var/home/acme/git/pahole/pahole.c: In function ‘pahole_threads_collect’:/var/home/acme/git/pahole/pahole.c:2835:40: warning: initialization of ‘structthread_data**’ from incompatible pointer type ‘void **’ [-Wincompatible-pointer-types]
2835 | structthread_data**threads =thr_data;
| ^~~~~~~~And I did some more profiling, now the focus should go to elfutils:
⬢[acme@toolbox pahole]$ perf report --no-children -s dso --call-graph none 2>
/dev/null| head -20 # To display the perf.data header info, please use --header/--header-only options. #
#
# Total Lost Samples: 0
#
# Samples: 27K of event 'cycles:u'
# Event count (approx.): 27956766207
#
# Overhead Shared Object# ........ ................... # 46.70% libdwarves.so.1.0.0 39.84% libdw-0.186.so 9.70% libc-2.33.so 2.14% libpthread-2.33.so 1.47% [unknown] 0.09% ld-2.33.so 0.06% libelf-0.186.so 0.00% libcrypto.so.1.1.1l 0.00% libk5crypto.so.3.1 ⬢[acme@toolbox pahole]$$ perf report -g graph,0.5,2 --stdio --no-children -s dso --dso libdw-0.186.so
[...]
Quoted 9 lines
#
# (Tip: If you have debuginfo enabled, try: perf report -s sym,srcline) #This
find_attrthing needs improvements, its a linear search AFAIK, some hashtable could do wonders, I guess.Mark, was this considered at some point?
This would be a great improvement, yes!
But strange that you didn't see any BTF-related functions, are you sure you profiled the entire DWARF to BTF conversion process? BTF encoding is not dominant, but noticeable anyways (e.g., adding unique strings to BTF is relatively expensive still).
Quoted 17 lines
⬢[acme@toolbox pahole]$ rpm -q elfutils-libs
elfutils-libs-0.186-1.fc34.x86_64Andrii https://github.com/libbpf/libbpf/actions/workflows/pahole.yml is in failure mode for 3 days, and only yesterday I pushed these changes, seems unrelated to pahole:
Tests exit status: 1
Test Results:
bpftool: PASS
test_progs: FAIL (returned 1)
test_progs-no_alu32: FAIL (returned 1)
test_verifier: PASS
shutdown: CLEAN
Error: Process completed with exit code 1.
Quoted 1 line
Can you please check?
Yes, it's not related to pahole. This is the BPF selftests issue which I already fixed last week, but didn't get a chance to sync everything to Github repo before leaving for a short vacation. I'll do another sync tonight and it should be all green again.
Quoted 2 lines
- Arnaldo
Re: [PATCH dwarves v4 3/4] pahole: Use per-thread btf instances to avoid mutex locking.
Em Mon, Jan 31, 2022 at 10:56:16PM -0800, Andrii Nakryiko escreveu:
Quoted 89 lines
Quoted 76 lines · Arnaldo Carvalho de Melo
Em Thu, Jan 27, 2022 at 05:12:02PM -0300, Arnaldo Carvalho de Melo escreveu:
Quoted 9 lines
Em Wed, Jan 26, 2022 at 11:58:27AM -0800, Andrii Nakryiko escreveu:
Quoted 8 lines · Kui-Feng Lee
Quoted 4 lines
Create an instance of btf for each worker thread, and add type info to the local btf instance in the steal-function of pahole without mutex acquiring. Once finished with all worker threads, merge all per-thread btf instances to the primary btf instance.
There are still unnecessary casts and missing {} in the else branch, but I'll let Arnaldo decide or fix it up.
So its just one unneeded cast as
thr_datahere is just a 'void *':pahole.c +1
-1@@ -2924,7 +2924,7 @@ static enum load_steal_kind pahole_stealer(struct cu *cu,* avoids copying the data collected by the first thread. */ if (thr_data) { -struct thread_data *thread = (struct thread_data *)thr_data; +struct thread_data *thread = thr_data;if (thread->encoder ==NULL ) { thread->encoder =This other is needed as it is a "void **":
pahole.c +1
-1@@ -2832,7 +2832,7 @@ static int pahole_thread_exit(struct conf_load *conf, void *thr_data)static int pahole_threads_collect (struct conf_load *conf,int nr_threads,void **thr_data,int error) { -struct thread_data **threads = (struct thread_data **)thr_data; +struct thread_data **threads = thr_data;int i;int err =0 ;Removing it:
/var/home/acme/git/pahole/pahole.c: In function ‘pahole_threads_collect’:/var/home/acme/git/pahole/pahole.c:2835:40: warning: initialization of ‘structthread_data**’ from incompatible pointer type ‘void **’ [-Wincompatible-pointer-types]
2835 | structthread_data**threads =thr_data;
| ^~~~~~~~And I did some more profiling, now the focus should go to elfutils:
⬢[acme@toolbox pahole]$ perf report --no-children -s dso --call-graph none 2>
/dev/null| head -20 # To display the perf.data header info, please use --header/--header-only options. #
#
# Total Lost Samples: 0
#
# Samples: 27K of event 'cycles:u'
# Event count (approx.): 27956766207
#
# Overhead Shared Object# ........ ................... # 46.70% libdwarves.so.1.0.0 39.84% libdw-0.186.so 9.70% libc-2.33.so 2.14% libpthread-2.33.so 1.47% [unknown] 0.09% ld-2.33.so 0.06% libelf-0.186.so 0.00% libcrypto.so.1.1.1l 0.00% libk5crypto.so.3.1 ⬢[acme@toolbox pahole]$$ perf report -g graph,0.5,2 --stdio --no-children -s dso --dso libdw-0.186.so[...]
Quoted 8 lines
#
# (Tip: If you have debuginfo enabled, try: perf report -s sym,srcline) #This
find_attrthing needs improvements, its a linear search AFAIK, some hashtable could do wonders, I guess.
Quoted 8 lines
Quoted 1 line
Mark, was this considered at some point?
This would be a great improvement, yes!
But strange that you didn't see any BTF-related functions, are you sure you profiled the entire DWARF to BTF conversion process? BTF encoding is not dominant, but noticeable anyways (e.g., adding unique strings to BTF is relatively expensive still).
It appears under the
Quoted 1 line
Quoted 1 line
46.70% libdwarves.so.1.0.0
line, since we statically link libbpf. So yeah, it was unfortunate the way I presented the profiling output, I was just trying to point to what I think is the low hanging fruit, i.e. optimizing find_attr routines in libdw-0.186.so.
Quoted 17 lines
Quoted 17 lines
⬢[acme@toolbox pahole]$ rpm -q elfutils-libs
elfutils-libs-0.186-1.fc34.x86_64Andrii https://github.com/libbpf/libbpf/actions/workflows/pahole.yml is in failure mode for 3 days, and only yesterday I pushed these changes, seems unrelated to pahole:
Tests exit status: 1
Test Results:
bpftool: PASS
test_progs: FAIL (returned 1)
test_progs-no_alu32: FAIL (returned 1)
test_verifier: PASS
shutdown: CLEAN
Error: Process completed with exit code 1.
Quoted 6 lines
Quoted 1 line
Can you please check?
Yes, it's not related to pahole. This is the BPF selftests issue which I already fixed last week, but didn't get a chance to sync everything to Github repo before leaving for a short vacation. I'll do another sync tonight and it should be all green again.
I'll check tomorrow.
Going green I'll do some more tests and work towards releasing 1.24.
- Arnaldo
PATCH v4 4/4 libbpf: Update libbpf to a new revision.
In reply to Kui-Feng LeeReplace deprecated APIs with new ones.
--- btf_encoder.c | 20 ++++++++++---------- btf_loader.c | 2 +- lib/bpf | 2 +- 3 files changed, 12 insertions(+), 12 deletions(-)
btf_encoder.c +10 -10
@@ -172,7 +172,7 @@ __attribute ((format (printf, 5, 6)))static void btf__log_err (const struct btf *btf,int kind,const char *name,bool output_cr,const char *fmt, ...) { -fprintf (stderr,"[%u] %s %s" ,btf__get_nr_types (btf) +1 , +fprintf (stderr,"[%u] %s %s" ,btf__type_cnt (btf), btf_kind_str[kind], name ?:"(anon)" );if (fmt && *fmt) {
@@ -203,7 +203,7 @@ static void btf_encoder__log_type(const struct btf_encoder *encoder, const struc out = err ? stderr : stdout;fprintf (out,"[%u] %s %s" , -btf__get_nr_types (btf), btf_kind_str[kind], +btf__type_cnt (btf) -1 , btf_kind_str[kind],btf__printable_name (btf, t->name_off));if (fmt && *fmt) {
@@ -449,10 +449,10 @@ static int btf_encoder__add_field(struct btf_encoder *encoder, const char *name,int err; err =btf__add_field (btf, name, type, offset, bitfield_size); - t =btf__type_by_id (btf,btf__get_nr_types (btf)); + t =btf__type_by_id (btf,btf__type_cnt (btf) -1 );if (err) {fprintf (stderr,"[%u] %s %s's field '%s' offset=%u bit_size=%u type=%u Error emitting field\n" , -btf__get_nr_types (btf), btf_kind_str[btf_kind (t)], +btf__type_cnt (btf) -1 , btf_kind_str[btf_kind (t)],btf__printable_name (btf, t->name_off), name, offset, bitfield_size, type); }else {
@@ -899,9 +899,9 @@ static int btf_encoder__write_raw_file(struct btf_encoder *encoder)const void *raw_btf_data;int fd, err; - raw_btf_data =btf__get_raw_data (encoder->btf, &raw_btf_size); + raw_btf_data =btf__raw_data (encoder->btf, &raw_btf_size);if (raw_btf_data ==NULL ) { -fprintf (stderr,"%s: btf__get_raw_data failed!\n" , __func__); +fprintf (stderr,"%s: btf__raw_data failed!\n" , __func__);return -1 ; }
@@ -976,7 +976,7 @@ static int btf_encoder__write_elf(struct btf_encoder *encoder) } } - raw_btf_data =btf__get_raw_data (btf, &raw_btf_size); + raw_btf_data =btf__raw_data (btf, &raw_btf_size);if (btf_data) {/* Existing .BTF section found */
@@ -1043,10 +1043,10 @@ int btf_encoder__encode(struct btf_encoder *encoder)btf_encoder__add_datasec (encoder,PERCPU_SECTION );/* Empty file, nothing to do, so... done! */ -if (btf__get_nr_types (encoder->btf) ==0 ) +if (btf__type_cnt (encoder->btf) ==1 )return 0 ; -if (btf__dedup (encoder->btf,NULL ,NULL )) { +if (btf__dedup (encoder->btf,NULL )) {fprintf (stderr,"%s: btf__dedup failed!\n" , __func__);return -1 ; }
@@ -1403,7 +1403,7 @@ void btf_encoder__delete(struct btf_encoder *encoder)int btf_encoder__encode_cu (struct btf_encoder *encoder,struct cu *cu) { -uint32_t type_id_off =btf__get_nr_types (encoder->btf); +uint32_t type_id_off =btf__type_cnt (encoder->btf) -1 ;struct llvm_annotation *annot;int btf_type_id, tag_type_id;uint32_t core_id;
btf_loader.c +1 -1
@@ -399,7 +399,7 @@ static int btf__load_types(struct btf *btf, struct cu *cu)uint32_t type_index;int err; -for (type_index =1 ; type_index <=btf__get_nr_types (btf); type_index++) { +for (type_index =1 ; type_index <btf__type_cnt (btf); type_index++) {const struct btf_type *type_ptr =btf__type_by_id (btf, type_index);uint32_t type =btf_kind (type_ptr);
lib/bpf +1 -1
@@ -1 +1 @@ -Subproject commit 94a49850c5ee61ea02dfcbabf48013391e8cecdf +Subproject commit 393a058d061d49d5c3055fa9eefafb4c0c31ccc3
Signature · 1 line
2.30.2
Re: [PATCH dwarves v4 4/4] libbpf: Update libbpf to a new revision.
Quoted 5 lines · Kui-Feng Lee
Replace deprecated APIs with new ones.
Signed-off-by Kui-Feng Lee <[email protected]>---
Quoted 5 lines
btf_encoder.c | 20 ++++++++++---------- btf_loader.c | 2 +- lib/bpf | 2 +- 3 files changed, 12 insertions(+), 12 deletions(-)
[...]