perf trace: Format fields with hex specifiers in print_fmt as hexadecimal
PATCH v2 1/1 perf trace: Format fields with hex specifiers in print_fmt as hexadecimal
Currently, when a probe event is added using perf probe --add with a hexadecimal type cast (e.g., var:x64 and var:x32), the kernel traceprobe subsystem sets the field type to "u64" in the event's format file, yet generates a print format specifier of "var=0x%Lx" or "0x%x".
By default, perf trace uses its internal beautifiers which only inspect field->type and field->flags. Because traceprobe events declare primitive integer field types ("u64") without a C pointer asterisk (i.e., "*"), perf trace defaults to printing the field's value as a decimal unsigned long ("%ld"), ignoring the hexadecimal intent.
Introduce field_has_hex_fmt() to inspect the tracepoint's print_fmt format string. If a field's print format assignment uses hexadecimal specifiers (i.e., "=0x", "=%#", or "=%p"), assign SCA_PTR formatting so perf trace displays the value in hexadecimal by default without requiring --libtraceevent. For example:
❯ sudo ./perf probe --add 'enqueue_task_fair rq=rq:x64 p=p:x64 flags=flags:x32'
Added new event:
probe:enqueue_task_fair (on enqueue_task_fair with rq=rq:x64 p=p:x64 flags=flags:x32)
You can now use it in all perf tools, such as:
perf record -e probe:enqueue_task_fair -aR sleep 1
❯ sudo ./perf trace --show-cpu --event probe:enqueue_task_fair --max-event 3
0.000 [000] :0/0 probe:enqueue_task_fair(rq: 0xffff8ac632233180, p: 0xffff8ac2c1240000, flags: 0x100009)
0.006 [004] :0/0 probe:enqueue_task_fair(rq: 0xffff8ac632433180, p: 0xffff8ac624cf8000, flags: 0x9)
0.069 [001] :0/0 probe:enqueue_task_fair(rq: 0xffff8ac6322b3180, p: 0xffff8ac4c4212c40, flags: 0x9)---
Changes since v1:
- Fixed field_has_hex_fmt() to properly handle escaped quotes (\") in format strings
- Link to v1: https://lore.kernel.org/lkml/[email protected]/
--- tools/perf/builtin-trace.c | 46 +++++++++++++++++++++++++++++++++++++- 1 file changed, 45 insertions(+), 1 deletion(-)
tools/perf/builtin-trace.c +45 -1
@@ -2098,6 +2098,49 @@ static bool is_internal_field(struct tep_format_field *field)return !strcmp (field->type,"__data_loc char[]" ); } +static bool field_has_hex_fmt (struct tep_format_field *field,int len) +{ +const char *fmt, *pos, *end =NULL ; + +if (!field || !field->event || !field->event->print_fmt.format) +return false; + + fmt = field->event->print_fmt.format; + +/* NB: Limit scanning strictly to the quoted printf format string */ +if (*fmt =='"' ) { +const char *p = ++fmt; + +while (*p) { +if (*p =='\\' && p[1 ] !='\0' ) { +/* NB: Skip escaped character */ + p +=2 ; + }else if (*p =='"' ) { + end = p; +break ; + }else { + p++; + } + } + }else { + end =strchr (fmt,',' ); + } + +for (pos =strstr (fmt, field->name); pos && (!end || pos < end); + pos =strstr (pos +1 , field->name)) { +if (pos == fmt || !(isalnum (pos[-1 ]) || pos[-1 ] =='_' )) { +const char *after = pos + len; + +if (*after =='=' && (strstarts (after +1 ,"0x" ) || +strstarts (after +1 ,"%#" ) || +strstarts (after +1 ,"%p" ))) +return true; + } + } + +return false; +} +static struct tep_format_field *syscall_arg_fmt__init_array (struct syscall_arg_fmt *arg,struct tep_format_field *field,bool *use_btf)
@@ -2125,7 +2168,8 @@ syscall_arg_fmt__init_array(struct syscall_arg_fmt *arg, struct tep_format_field ((len >= 4 && strcmp(field->name + len - 4, "name") == 0) || strstr(field->name, "path") != NULL)) { arg->scnprintf =SCA_FILENAME ; - } else if ((field->flags & TEP_FIELD_IS_POINTER) || strstr(field->name, "addr")) + } else if ((field->flags & TEP_FIELD_IS_POINTER) || strstr(field->name, "addr") || +field_has_hex_fmt (field, len)) arg->scnprintf = SCA_PTR; else if (strcmp(field->type, "pid_t") == 0) arg->scnprintf = SCA_PID;
Signature · 1 line
2.55.0
Quoted 5 lines · Aaron Tomlin
Currently, when a probe event is added using perf probe --add with a hexadecimal type cast (e.g., var:x64 and var:x32), the kernel traceprobe subsystem sets the field type to "u64" in the event's format file, yet generates a print format specifier of "var=0x%Lx" or "0x%x".
Quoted 7 lines
By default, perf trace uses its internal beautifiers which only inspect
field->typeandfield->flags. Because traceprobe events declare primitive integer field types ("u64") without a C pointer asterisk (i.e., "*"), perf trace defaults to printing the field's value as a decimal unsigned long ("%ld"), ignoring the hexadecimal intent.[...]
Applied to perf-tools-next, thanks!
Best regards,
Namhyung
Quoted 12 lines · Namhyung Kim
Quoted 11 lines · Aaron Tomlin
Currently, when a probe event is added using perf probe --add with a hexadecimal type cast (e.g., var:x64 and var:x32), the kernel traceprobe subsystem sets the field type to "u64" in the event's format file, yet generates a print format specifier of "var=0x%Lx" or "0x%x".
By default, perf trace uses its internal beautifiers which only inspect
field->typeandfield->flags. Because traceprobe events declare primitive integer field types ("u64") without a C pointer asterisk (i.e., "*"), perf trace defaults to printing the field's value as a decimal unsigned long ("%ld"), ignoring the hexadecimal intent.
Quoted 5 lines
Quoted 1 line
[...]
Applied to perf-tools-next, thanks!
Best regards,
Namhyung
Thank you Namhyung!
Signature · 1 line
Aaron Tomlin