Some different pmu types may have same substring. For example,
on Icelake server, we have pmu types "uncore_imc" and
"uncore_imc_free_running". Both pmu types have substring "uncore_imc".
But the parser would wrongly think they are the same pmu type.
We enable an imc event,
perf stat -e uncore_imc/event=0xe3/ -a -- sleep 1
Perf actually expands the event to:
uncore_imc_0/event=0xe3/
uncore_imc_1/event=0xe3/
uncore_imc_2/event=0xe3/
uncore_imc_3/event=0xe3/
uncore_imc_4/event=0xe3/
uncore_imc_5/event=0xe3/
uncore_imc_6/event=0xe3/
uncore_imc_7/event=0xe3/
uncore_imc_free_running_0/event=0xe3/
uncore_imc_free_running_1/event=0xe3/
uncore_imc_free_running_3/event=0xe3/
uncore_imc_free_running_4/event=0xe3/
That's because the "uncore_imc_free_running" matches the
pattern "uncore_imc*".
Now we check that the last characters of pmu name is
'_<digit>'.
For example, for pattern "uncore_imc*", "uncore_imc_0" is parsed ok,
but "uncore_imc_free_running_0" would be failed.
Fixes: b2b9d3a3f021 ("perf pmu: Support wildcards on pmu name in dynamic pmu events")
Signed-off-by: Jin Yao <[email protected]>
---
tools/perf/util/parse-events.y | 2 +-
tools/perf/util/pmu.c | 36 +++++++++++++++++++++++++++++++++-
tools/perf/util/pmu.h | 1 +
3 files changed, 37 insertions(+), 2 deletions(-)
diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
index aba12a4d488e..9321bd0e2f76 100644
--- a/tools/perf/util/parse-events.y
+++ b/tools/perf/util/parse-events.y
@@ -316,7 +316,7 @@ event_pmu_name opt_pmu_config
if (!strncmp(name, "uncore_", 7) &&
strncmp($1, "uncore_", 7))
name += 7;
- if (!fnmatch(pattern, name, 0)) {
+ if (!perf_pmu__match(pattern, name, $1)) {
if (parse_events_copy_term_list(orig_terms, &terms))
CLEANUP_YYABORT;
if (!parse_events_add_pmu(_parse_state, list, pmu->name, terms, true, false))
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index 88c8ecdc60b0..44b90d638ad5 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf/util/pmu.c
@@ -3,6 +3,7 @@
#include <linux/compiler.h>
#include <linux/string.h>
#include <linux/zalloc.h>
+#include <linux/ctype.h>
#include <subcmd/pager.h>
#include <sys/types.h>
#include <errno.h>
@@ -17,6 +18,7 @@
#include <locale.h>
#include <regex.h>
#include <perf/cpumap.h>
+#include <fnmatch.h>
#include "debug.h"
#include "evsel.h"
#include "pmu.h"
@@ -740,6 +742,27 @@ struct pmu_events_map *__weak pmu_events_map__find(void)
return perf_pmu__find_map(NULL);
}
+static bool perf_pmu__valid_suffix(char *pmu_name, char *tok)
+{
+ char *p;
+
+ if (strncmp(pmu_name, tok, strlen(tok)))
+ return false;
+
+ p = pmu_name + strlen(tok);
+ if (*p == 0)
+ return true;
+
+ if (*p != '_')
+ return false;
+
+ ++p;
+ if (*p == 0 || !isdigit(*p))
+ return false;
+
+ return true;
+}
+
bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
{
char *tmp = NULL, *tok, *str;
@@ -768,7 +791,7 @@ bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
*/
for (; tok; name += strlen(tok), tok = strtok_r(NULL, ",", &tmp)) {
name = strstr(name, tok);
- if (!name) {
+ if (!name || !perf_pmu__valid_suffix((char *)name, tok)) {
res = false;
goto out;
}
@@ -1872,3 +1895,14 @@ bool perf_pmu__has_hybrid(void)
return !list_empty(&perf_pmu__hybrid_pmus);
}
+
+int perf_pmu__match(char *pattern, char *name, char *tok)
+{
+ if (fnmatch(pattern, name, 0))
+ return -1;
+
+ if (tok && !perf_pmu__valid_suffix(name, tok))
+ return -1;
+
+ return 0;
+}
diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h
index a790ef758171..926da483a141 100644
--- a/tools/perf/util/pmu.h
+++ b/tools/perf/util/pmu.h
@@ -133,5 +133,6 @@ void perf_pmu__warn_invalid_config(struct perf_pmu *pmu, __u64 config,
char *name);
bool perf_pmu__has_hybrid(void);
+int perf_pmu__match(char *pattern, char *name, char *tok);
#endif /* __PMU_H */
--
2.17.1
On Thu, Jul 01, 2021 at 02:42:53PM +0800, Jin Yao wrote:
> Some different pmu types may have same substring. For example,
> on Icelake server, we have pmu types "uncore_imc" and
> "uncore_imc_free_running". Both pmu types have substring "uncore_imc".
> But the parser would wrongly think they are the same pmu type.
>
> We enable an imc event,
> perf stat -e uncore_imc/event=0xe3/ -a -- sleep 1
>
> Perf actually expands the event to:
> uncore_imc_0/event=0xe3/
> uncore_imc_1/event=0xe3/
> uncore_imc_2/event=0xe3/
> uncore_imc_3/event=0xe3/
> uncore_imc_4/event=0xe3/
> uncore_imc_5/event=0xe3/
> uncore_imc_6/event=0xe3/
> uncore_imc_7/event=0xe3/
> uncore_imc_free_running_0/event=0xe3/
> uncore_imc_free_running_1/event=0xe3/
> uncore_imc_free_running_3/event=0xe3/
> uncore_imc_free_running_4/event=0xe3/
>
> That's because the "uncore_imc_free_running" matches the
> pattern "uncore_imc*".
>
> Now we check that the last characters of pmu name is
> '_<digit>'.
>
> For example, for pattern "uncore_imc*", "uncore_imc_0" is parsed ok,
> but "uncore_imc_free_running_0" would be failed.
>
> Fixes: b2b9d3a3f021 ("perf pmu: Support wildcards on pmu name in dynamic pmu events")
> Signed-off-by: Jin Yao <[email protected]>
looks good to me, Kan, Andi?
Acked-by: Jiri Olsa <[email protected]>
thanks,
jirka
> ---
> tools/perf/util/parse-events.y | 2 +-
> tools/perf/util/pmu.c | 36 +++++++++++++++++++++++++++++++++-
> tools/perf/util/pmu.h | 1 +
> 3 files changed, 37 insertions(+), 2 deletions(-)
>
> diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
> index aba12a4d488e..9321bd0e2f76 100644
> --- a/tools/perf/util/parse-events.y
> +++ b/tools/perf/util/parse-events.y
> @@ -316,7 +316,7 @@ event_pmu_name opt_pmu_config
> if (!strncmp(name, "uncore_", 7) &&
> strncmp($1, "uncore_", 7))
> name += 7;
> - if (!fnmatch(pattern, name, 0)) {
> + if (!perf_pmu__match(pattern, name, $1)) {
> if (parse_events_copy_term_list(orig_terms, &terms))
> CLEANUP_YYABORT;
> if (!parse_events_add_pmu(_parse_state, list, pmu->name, terms, true, false))
> diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
> index 88c8ecdc60b0..44b90d638ad5 100644
> --- a/tools/perf/util/pmu.c
> +++ b/tools/perf/util/pmu.c
> @@ -3,6 +3,7 @@
> #include <linux/compiler.h>
> #include <linux/string.h>
> #include <linux/zalloc.h>
> +#include <linux/ctype.h>
> #include <subcmd/pager.h>
> #include <sys/types.h>
> #include <errno.h>
> @@ -17,6 +18,7 @@
> #include <locale.h>
> #include <regex.h>
> #include <perf/cpumap.h>
> +#include <fnmatch.h>
> #include "debug.h"
> #include "evsel.h"
> #include "pmu.h"
> @@ -740,6 +742,27 @@ struct pmu_events_map *__weak pmu_events_map__find(void)
> return perf_pmu__find_map(NULL);
> }
>
> +static bool perf_pmu__valid_suffix(char *pmu_name, char *tok)
> +{
> + char *p;
> +
> + if (strncmp(pmu_name, tok, strlen(tok)))
> + return false;
> +
> + p = pmu_name + strlen(tok);
> + if (*p == 0)
> + return true;
> +
> + if (*p != '_')
> + return false;
> +
> + ++p;
> + if (*p == 0 || !isdigit(*p))
> + return false;
> +
> + return true;
> +}
> +
> bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
> {
> char *tmp = NULL, *tok, *str;
> @@ -768,7 +791,7 @@ bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
> */
> for (; tok; name += strlen(tok), tok = strtok_r(NULL, ",", &tmp)) {
> name = strstr(name, tok);
> - if (!name) {
> + if (!name || !perf_pmu__valid_suffix((char *)name, tok)) {
> res = false;
> goto out;
> }
> @@ -1872,3 +1895,14 @@ bool perf_pmu__has_hybrid(void)
>
> return !list_empty(&perf_pmu__hybrid_pmus);
> }
> +
> +int perf_pmu__match(char *pattern, char *name, char *tok)
> +{
> + if (fnmatch(pattern, name, 0))
> + return -1;
> +
> + if (tok && !perf_pmu__valid_suffix(name, tok))
> + return -1;
> +
> + return 0;
> +}
> diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h
> index a790ef758171..926da483a141 100644
> --- a/tools/perf/util/pmu.h
> +++ b/tools/perf/util/pmu.h
> @@ -133,5 +133,6 @@ void perf_pmu__warn_invalid_config(struct perf_pmu *pmu, __u64 config,
> char *name);
>
> bool perf_pmu__has_hybrid(void);
> +int perf_pmu__match(char *pattern, char *name, char *tok);
>
> #endif /* __PMU_H */
> --
> 2.17.1
>
On 7/6/2021 3:44 PM, Jiri Olsa wrote:
> On Thu, Jul 01, 2021 at 02:42:53PM +0800, Jin Yao wrote:
>> Some different pmu types may have same substring. For example,
>> on Icelake server, we have pmu types "uncore_imc" and
>> "uncore_imc_free_running". Both pmu types have substring "uncore_imc".
>> But the parser would wrongly think they are the same pmu type.
>>
>> We enable an imc event,
>> perf stat -e uncore_imc/event=0xe3/ -a -- sleep 1
>>
>> Perf actually expands the event to:
>> uncore_imc_0/event=0xe3/
>> uncore_imc_1/event=0xe3/
>> uncore_imc_2/event=0xe3/
>> uncore_imc_3/event=0xe3/
>> uncore_imc_4/event=0xe3/
>> uncore_imc_5/event=0xe3/
>> uncore_imc_6/event=0xe3/
>> uncore_imc_7/event=0xe3/
>> uncore_imc_free_running_0/event=0xe3/
>> uncore_imc_free_running_1/event=0xe3/
>> uncore_imc_free_running_3/event=0xe3/
>> uncore_imc_free_running_4/event=0xe3/
>>
>> That's because the "uncore_imc_free_running" matches the
>> pattern "uncore_imc*".
>>
>> Now we check that the last characters of pmu name is
>> '_<digit>'.
>>
>> For example, for pattern "uncore_imc*", "uncore_imc_0" is parsed ok,
>> but "uncore_imc_free_running_0" would be failed.
>>
>> Fixes: b2b9d3a3f021 ("perf pmu: Support wildcards on pmu name in dynamic pmu events")
>> Signed-off-by: Jin Yao <[email protected]>
>
> looks good to me, Kan, Andi?
Yes, it looks good to me too.
Reviewed-by: Kan Liang <[email protected]>
BTW: The new uncore patch[1] for the "alias" attribute should be easily
rebase on this patch. I believe Yao has already finished the test as
well. I think he will resend the new "alias" attribute patch later soon.
[1]
https://lore.kernel.org/lkml/[email protected]
Thanks,
Kan
>
> Acked-by: Jiri Olsa <[email protected]>
>
> thanks,
> jirka
>
>> ---
>> tools/perf/util/parse-events.y | 2 +-
>> tools/perf/util/pmu.c | 36 +++++++++++++++++++++++++++++++++-
>> tools/perf/util/pmu.h | 1 +
>> 3 files changed, 37 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
>> index aba12a4d488e..9321bd0e2f76 100644
>> --- a/tools/perf/util/parse-events.y
>> +++ b/tools/perf/util/parse-events.y
>> @@ -316,7 +316,7 @@ event_pmu_name opt_pmu_config
>> if (!strncmp(name, "uncore_", 7) &&
>> strncmp($1, "uncore_", 7))
>> name += 7;
>> - if (!fnmatch(pattern, name, 0)) {
>> + if (!perf_pmu__match(pattern, name, $1)) {
>> if (parse_events_copy_term_list(orig_terms, &terms))
>> CLEANUP_YYABORT;
>> if (!parse_events_add_pmu(_parse_state, list, pmu->name, terms, true, false))
>> diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
>> index 88c8ecdc60b0..44b90d638ad5 100644
>> --- a/tools/perf/util/pmu.c
>> +++ b/tools/perf/util/pmu.c
>> @@ -3,6 +3,7 @@
>> #include <linux/compiler.h>
>> #include <linux/string.h>
>> #include <linux/zalloc.h>
>> +#include <linux/ctype.h>
>> #include <subcmd/pager.h>
>> #include <sys/types.h>
>> #include <errno.h>
>> @@ -17,6 +18,7 @@
>> #include <locale.h>
>> #include <regex.h>
>> #include <perf/cpumap.h>
>> +#include <fnmatch.h>
>> #include "debug.h"
>> #include "evsel.h"
>> #include "pmu.h"
>> @@ -740,6 +742,27 @@ struct pmu_events_map *__weak pmu_events_map__find(void)
>> return perf_pmu__find_map(NULL);
>> }
>>
>> +static bool perf_pmu__valid_suffix(char *pmu_name, char *tok)
>> +{
>> + char *p;
>> +
>> + if (strncmp(pmu_name, tok, strlen(tok)))
>> + return false;
>> +
>> + p = pmu_name + strlen(tok);
>> + if (*p == 0)
>> + return true;
>> +
>> + if (*p != '_')
>> + return false;
>> +
>> + ++p;
>> + if (*p == 0 || !isdigit(*p))
>> + return false;
>> +
>> + return true;
>> +}
>> +
>> bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
>> {
>> char *tmp = NULL, *tok, *str;
>> @@ -768,7 +791,7 @@ bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
>> */
>> for (; tok; name += strlen(tok), tok = strtok_r(NULL, ",", &tmp)) {
>> name = strstr(name, tok);
>> - if (!name) {
>> + if (!name || !perf_pmu__valid_suffix((char *)name, tok)) {
>> res = false;
>> goto out;
>> }
>> @@ -1872,3 +1895,14 @@ bool perf_pmu__has_hybrid(void)
>>
>> return !list_empty(&perf_pmu__hybrid_pmus);
>> }
>> +
>> +int perf_pmu__match(char *pattern, char *name, char *tok)
>> +{
>> + if (fnmatch(pattern, name, 0))
>> + return -1;
>> +
>> + if (tok && !perf_pmu__valid_suffix(name, tok))
>> + return -1;
>> +
>> + return 0;
>> +}
>> diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h
>> index a790ef758171..926da483a141 100644
>> --- a/tools/perf/util/pmu.h
>> +++ b/tools/perf/util/pmu.h
>> @@ -133,5 +133,6 @@ void perf_pmu__warn_invalid_config(struct perf_pmu *pmu, __u64 config,
>> char *name);
>>
>> bool perf_pmu__has_hybrid(void);
>> +int perf_pmu__match(char *pattern, char *name, char *tok);
>>
>> #endif /* __PMU_H */
>> --
>> 2.17.1
>>
>
Em Tue, Jul 06, 2021 at 04:02:24PM -0400, Liang, Kan escreveu:
> On 7/6/2021 3:44 PM, Jiri Olsa wrote:
> > On Thu, Jul 01, 2021 at 02:42:53PM +0800, Jin Yao wrote:
> > > Some different pmu types may have same substring. For example,
> > > on Icelake server, we have pmu types "uncore_imc" and
> > > "uncore_imc_free_running". Both pmu types have substring "uncore_imc".
> > > But the parser would wrongly think they are the same pmu type.
> > > We enable an imc event,
> > > perf stat -e uncore_imc/event=0xe3/ -a -- sleep 1
> > > Perf actually expands the event to:
> > > uncore_imc_0/event=0xe3/
> > > uncore_imc_1/event=0xe3/
> > > uncore_imc_2/event=0xe3/
> > > uncore_imc_3/event=0xe3/
> > > uncore_imc_4/event=0xe3/
> > > uncore_imc_5/event=0xe3/
> > > uncore_imc_6/event=0xe3/
> > > uncore_imc_7/event=0xe3/
> > > uncore_imc_free_running_0/event=0xe3/
> > > uncore_imc_free_running_1/event=0xe3/
> > > uncore_imc_free_running_3/event=0xe3/
> > > uncore_imc_free_running_4/event=0xe3/
> > > That's because the "uncore_imc_free_running" matches the
> > > pattern "uncore_imc*".
> > > Now we check that the last characters of pmu name is
> > > '_<digit>'.
> > >
> > > For example, for pattern "uncore_imc*", "uncore_imc_0" is parsed ok,
> > > but "uncore_imc_free_running_0" would be failed.
> > >
> > > Fixes: b2b9d3a3f021 ("perf pmu: Support wildcards on pmu name in dynamic pmu events")
> > > Signed-off-by: Jin Yao <[email protected]>
> > looks good to me, Kan, Andi?
> Yes, it looks good to me too.
> Reviewed-by: Kan Liang <[email protected]>
Thanks, I'm applying it, Jin, please next time Cc the author of the
patch you're fixing, in this case:
Cc: Agustin Vega-Frias <[email protected]>
I'm adding it to this message as well.
> BTW: The new uncore patch[1] for the "alias" attribute should be easily
> rebase on this patch. I believe Yao has already finished the test as well. I
> think he will resend the new "alias" attribute patch later soon.
> [1] https://lore.kernel.org/lkml/[email protected]
>
> Thanks,
> Kan
>
> >
> > Acked-by: Jiri Olsa <[email protected]>
> >
> > thanks,
> > jirka
> >
> > > ---
> > > tools/perf/util/parse-events.y | 2 +-
> > > tools/perf/util/pmu.c | 36 +++++++++++++++++++++++++++++++++-
> > > tools/perf/util/pmu.h | 1 +
> > > 3 files changed, 37 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y
> > > index aba12a4d488e..9321bd0e2f76 100644
> > > --- a/tools/perf/util/parse-events.y
> > > +++ b/tools/perf/util/parse-events.y
> > > @@ -316,7 +316,7 @@ event_pmu_name opt_pmu_config
> > > if (!strncmp(name, "uncore_", 7) &&
> > > strncmp($1, "uncore_", 7))
> > > name += 7;
> > > - if (!fnmatch(pattern, name, 0)) {
> > > + if (!perf_pmu__match(pattern, name, $1)) {
> > > if (parse_events_copy_term_list(orig_terms, &terms))
> > > CLEANUP_YYABORT;
> > > if (!parse_events_add_pmu(_parse_state, list, pmu->name, terms, true, false))
> > > diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
> > > index 88c8ecdc60b0..44b90d638ad5 100644
> > > --- a/tools/perf/util/pmu.c
> > > +++ b/tools/perf/util/pmu.c
> > > @@ -3,6 +3,7 @@
> > > #include <linux/compiler.h>
> > > #include <linux/string.h>
> > > #include <linux/zalloc.h>
> > > +#include <linux/ctype.h>
> > > #include <subcmd/pager.h>
> > > #include <sys/types.h>
> > > #include <errno.h>
> > > @@ -17,6 +18,7 @@
> > > #include <locale.h>
> > > #include <regex.h>
> > > #include <perf/cpumap.h>
> > > +#include <fnmatch.h>
> > > #include "debug.h"
> > > #include "evsel.h"
> > > #include "pmu.h"
> > > @@ -740,6 +742,27 @@ struct pmu_events_map *__weak pmu_events_map__find(void)
> > > return perf_pmu__find_map(NULL);
> > > }
> > > +static bool perf_pmu__valid_suffix(char *pmu_name, char *tok)
> > > +{
> > > + char *p;
> > > +
> > > + if (strncmp(pmu_name, tok, strlen(tok)))
> > > + return false;
> > > +
> > > + p = pmu_name + strlen(tok);
> > > + if (*p == 0)
> > > + return true;
> > > +
> > > + if (*p != '_')
> > > + return false;
> > > +
> > > + ++p;
> > > + if (*p == 0 || !isdigit(*p))
> > > + return false;
> > > +
> > > + return true;
> > > +}
> > > +
> > > bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
> > > {
> > > char *tmp = NULL, *tok, *str;
> > > @@ -768,7 +791,7 @@ bool pmu_uncore_alias_match(const char *pmu_name, const char *name)
> > > */
> > > for (; tok; name += strlen(tok), tok = strtok_r(NULL, ",", &tmp)) {
> > > name = strstr(name, tok);
> > > - if (!name) {
> > > + if (!name || !perf_pmu__valid_suffix((char *)name, tok)) {
> > > res = false;
> > > goto out;
> > > }
> > > @@ -1872,3 +1895,14 @@ bool perf_pmu__has_hybrid(void)
> > > return !list_empty(&perf_pmu__hybrid_pmus);
> > > }
> > > +
> > > +int perf_pmu__match(char *pattern, char *name, char *tok)
> > > +{
> > > + if (fnmatch(pattern, name, 0))
> > > + return -1;
> > > +
> > > + if (tok && !perf_pmu__valid_suffix(name, tok))
> > > + return -1;
> > > +
> > > + return 0;
> > > +}
> > > diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h
> > > index a790ef758171..926da483a141 100644
> > > --- a/tools/perf/util/pmu.h
> > > +++ b/tools/perf/util/pmu.h
> > > @@ -133,5 +133,6 @@ void perf_pmu__warn_invalid_config(struct perf_pmu *pmu, __u64 config,
> > > char *name);
> > > bool perf_pmu__has_hybrid(void);
> > > +int perf_pmu__match(char *pattern, char *name, char *tok);
> > > #endif /* __PMU_H */
> > > --
> > > 2.17.1
> > >
> >
--
- Arnaldo