Skip to contentSkip to conversation

perf tools: Assorted random fixes and updates

PATCH Applied Namhyung Kim 4 participants 10 messages Sep 23, 2022 – Sep 26, 2022 perf events

lore
Share & tools
lore

Namhyung Kim

author maintainer
#0 lore raw

cover perf tools: Assorted random fixes and updates

Hello,

This is collection of random fixes and updates. And I'm resending them as they seem to be lost in the list.

Thanks,
Namhyung

Namhyung Kim (4):
perf record: Fix a segfault in record__read_lost_samples() perf inject: Clarify build-id options a little bit perf tools: Add 'addr' sort key
perf annotate: Toggle full address <-> offset display

 tools/perf/Documentation/perf-inject.txt |  6 ++--
 tools/perf/Documentation/perf-report.txt |  3 +-
 tools/perf/builtin-record.c              |  6 ++++
 tools/perf/ui/browsers/annotate.c        |  6 +++-
 tools/perf/util/annotate.c               | 19 +++++++++++-
 tools/perf/util/annotate.h               |  4 ++-
 tools/perf/util/hist.c                   |  1 +
 tools/perf/util/hist.h                   |  1 +
 tools/perf/util/sort.c                   | 38 ++++++++++++++++++++++++
 tools/perf/util/sort.h                   |  1 +
 10 files changed, 79 insertions(+), 6 deletions(-)
Signature · 1 line
2.37.3.998.g577e59143f-goog
author maintainer
#1 lore raw

PATCH 1/4 perf record: Fix a segfault in record__read_lost_samples()

When it fails to open events record__open() returns without setting the session->evlist. Then it gets a segfault in the function trying to read lost sample counts. You can easily reproduce it as a normal user like:

  $ perf record -p 1 true
  ...
  perf: Segmentation fault
  ...

Skip the function if it has no evlist. And add more protection for evsels which are not properly initialized.

---
 tools/perf/builtin-record.c | 6 ++++++
 1 file changed, 6 insertions(+)
tools/perf/builtin-record.c +6 -0
index 741e763436ca..f4f1619199e5 100644
@@ -1888,6 +1888,10 @@ static void record__read_lost_samples(struct record *rec)
 	struct perf_record_lost_samples *lost;
 	struct evsel *evsel;
 
+	/* there was an error during record__open */
+	if (session->evlist == NULL)
+		return;
+
 	lost = zalloc(PERF_SAMPLE_MAX_SIZE);
 	if (lost == NULL) {
 		pr_debug("Memory allocation failed\n");
@@ -1899,6 +1903,8 @@ static void record__read_lost_samples(struct record *rec)
 	evlist__for_each_entry(session->evlist, evsel) {
 		struct xyarray *xy = evsel->core.sample_id;
 
+		if (xy == NULL || evsel->core.fd == NULL)
+			continue;
 		if (xyarray__max_x(evsel->core.fd) != xyarray__max_x(xy) ||
 		    xyarray__max_y(evsel->core.fd) != xyarray__max_y(xy)) {
 			pr_debug("Unmatched FD vs. sample ID: skip reading LOST count\n");
Signature · 1 line
2.37.3.998.g577e59143f-goog
reviewer Acked-by on v1 1/4
#2 lore raw

Re: [PATCH 1/4] perf record: Fix a segfault in record__read_lost_samples()

Quoted 10 lines · Namhyung Kim

When it fails to open events record__open() returns without setting the session->evlist. Then it gets a segfault in the function trying to read lost sample counts. You can easily reproduce it as a normal user like:

  $ perf record -p 1 true
  ...
  perf: Segmentation fault
  ...
Quoted 4 lines

Skip the function if it has no evlist. And add more protection for evsels which are not properly initialized.

Thanks,
Ian

Quoted 31 lines
---
 tools/perf/builtin-record.c | 6 ++++++
 1 file changed, 6 insertions(+)
tools/perf/builtin-record.c +6 -1
index 741e763436ca..f4f1619199e5 100644
@@ -1888,6 +1888,10 @@ static void record__read_lost_samples(struct record *rec)
        struct perf_record_lost_samples *lost;
        struct evsel *evsel;
 
+       /* there was an error during record__open */
+       if (session->evlist == NULL)
+               return;
+
        lost = zalloc(PERF_SAMPLE_MAX_SIZE);
        if (lost == NULL) {
                pr_debug("Memory allocation failed\n");
@@ -1899,6 +1903,8 @@ static void record__read_lost_samples(struct record *rec)
        evlist__for_each_entry(session->evlist, evsel) {
                struct xyarray *xy = evsel->core.sample_id;
 
+               if (xy == NULL || evsel->core.fd == NULL)
+                       continue;
                if (xyarray__max_x(evsel->core.fd) != xyarray__max_x(xy) ||
                    xyarray__max_y(evsel->core.fd) != xyarray__max_y(xy)) {
                        pr_debug("Unmatched FD vs. sample ID: skip reading LOST count\n");
--

2.37.3.998.g577e59143f-goog

Tested-by on v1 1/4
#3 lore raw

Re: [PATCH 1/4] perf record: Fix a segfault in record__read_lost_samples()

In reply to Namhyung Kim
Quoted 9 lines · Namhyung Kim

When it fails to open events record__open() returns without setting the session->evlist. Then it gets a segfault in the function trying to read lost sample counts. You can easily reproduce it as a normal user like:

  $ perf record -p 1 true
  ...
  perf: Segmentation fault
  ...
Quoted 4 lines

Skip the function if it has no evlist. And add more protection for evsels which are not properly initialized.

author maintainer
#4 lore raw

PATCH 2/4 perf inject: Clarify build-id options a little bit

In reply to Namhyung Kim

Update the documentation of --build-id and --buildid-all options to clarify the difference between them. The former requires full sample processing to find which DSOs are actually used. While the latter simply injects every DSO's build-id from MMAP{,2} records, skipping SAMPLEs.

---
 tools/perf/Documentation/perf-inject.txt | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
tools/perf/Documentation/perf-inject.txt +4 -2
index 70e2ac3cc91a..c972032f4ca0 100644
@@ -25,10 +25,12 @@ OPTIONS
 -------
 -b::
 --build-ids::
-        Inject build-ids into the output stream
+	Inject build-ids of DSOs hit by samples into the output stream.
+	This means it needs to process all SAMPLE records to find the DSOs.
 
 --buildid-all::
-	Inject build-ids of all DSOs into the output stream
+	Inject build-ids of all DSOs into the output stream regardless of hits
+	and skip SAMPLE processing.
 
 --known-build-ids=::
 	Override build-ids to inject using these comma-separated pairs of
Signature · 1 line
2.37.3.998.g577e59143f-goog
author maintainer
#5 lore raw

PATCH 3/4 perf tools: Add 'addr' sort key

In reply to Namhyung Kim

Sometimes users want to see actual (virtual) address of sampled instructions. Add a new 'addr' sort key to display the raw addresses.

  $ perf record -o- true | perf report -i- -s addr
  # To display the perf.data header info, please use --header/--header-only options.
  #
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.000 MB - ]
  #
  # Total Lost Samples: 0
  #
  # Samples: 12  of event 'cycles:u'
  # Event count (approx.): 252512
  #
  # Overhead  Address
  # ........  ..................
  #
      42.96%  0x7f96f08443d7
      29.55%  0x7f96f0859b50
      14.76%  0x7f96f0852e02
       8.30%  0x7f96f0855028
       4.43%  0xffffffff8de01087

Note that it just compares and displays the sample ip. Each process can have a different memory layout and the ip will be different even if they run the same binary. So this sort key is mostly meaningful for per-process profile data.

Cc Stephane Eranian <[email protected]>
---
 tools/perf/Documentation/perf-report.txt |  3 +-
 tools/perf/util/hist.c                   |  1 +
 tools/perf/util/hist.h                   |  1 +
 tools/perf/util/sort.c                   | 38 ++++++++++++++++++++++++
 tools/perf/util/sort.h                   |  1 +
 5 files changed, 43 insertions(+), 1 deletion(-)
tools/perf/Documentation/perf-report.txt +2 -1
index 24efc0583c93..4533db2ee56b 100644
@@ -73,7 +73,7 @@ OPTIONS
 	Sort histogram entries by given key(s) - multiple keys can be specified
 	in CSV format.  Following sort keys are available:
 	pid, comm, dso, symbol, parent, cpu, socket, srcline, weight,
-	local_weight, cgroup_id.
+	local_weight, cgroup_id, addr.
 
 	Each key has following meaning:
 
@@ -114,6 +114,7 @@ OPTIONS
 	- local_ins_lat: Local instruction latency version
 	- p_stage_cyc: On powerpc, this presents the number of cycles spent in a
 	  pipeline stage. And currently supported only on powerpc.
+	- addr: (Full) virtual address of the sampled instruction
 
 	By default, comm, dso and symbol keys are used.
 	(i.e. --sort comm,dso,symbol)
tools/perf/util/hist.c +1 -0
index 06f5dbf213ad..17a05e943b44 100644
@@ -215,6 +215,7 @@ void hists__calc_col_len(struct hists *hists, struct hist_entry *h)
 	hists__new_col_len(hists, HISTC_GLOBAL_INS_LAT, 13);
 	hists__new_col_len(hists, HISTC_LOCAL_P_STAGE_CYC, 13);
 	hists__new_col_len(hists, HISTC_GLOBAL_P_STAGE_CYC, 13);
+	hists__new_col_len(hists, HISTC_ADDR, BITS_PER_LONG / 4 + 2);
 
 	if (symbol_conf.nanosecs)
 		hists__new_col_len(hists, HISTC_TIME, 16);
tools/perf/util/hist.h +1 -0
index c7a7a3fa0b87..ebd8a8f783ee 100644
@@ -79,6 +79,7 @@ enum hist_column {
 	HISTC_GLOBAL_P_STAGE_CYC,
 	HISTC_ADDR_FROM,
 	HISTC_ADDR_TO,
+	HISTC_ADDR,
 	HISTC_NR_COLS, /* Last entry */
 };
 
tools/perf/util/sort.c +38 -0
index 6d5588e80935..2e7330867e2e 100644
@@ -1948,6 +1948,43 @@ struct sort_entry sort_dso_size = {
 	.se_width_idx	= HISTC_DSO_SIZE,
 };
 
+/* --sort dso_size */
+
+static int64_t
+sort__addr_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+	u64 left_ip = left->ip;
+	u64 right_ip = right->ip;
+	struct map *left_map = left->ms.map;
+	struct map *right_map = right->ms.map;
+
+	if (left_map)
+		left_ip = left_map->unmap_ip(left_map, left_ip);
+	if (right_map)
+		right_ip = right_map->unmap_ip(right_map, right_ip);
+
+	return _sort__addr_cmp(left_ip, right_ip);
+}
+
+static int hist_entry__addr_snprintf(struct hist_entry *he, char *bf,
+				     size_t size, unsigned int width)
+{
+	u64 ip = he->ip;
+	struct map *map = he->ms.map;
+
+	if (map)
+		ip = map->unmap_ip(map, ip);
+
+	return repsep_snprintf(bf, size, "%-#*llx", width, ip);
+}
+
+struct sort_entry sort_addr = {
+	.se_header	= "Address",
+	.se_cmp		= sort__addr_cmp,
+	.se_snprintf	= hist_entry__addr_snprintf,
+	.se_width_idx	= HISTC_ADDR,
+};
+
 
 struct sort_dimension {
 	const char		*name;
@@ -1997,6 +2034,7 @@ static struct sort_dimension common_sort_dimensions[] = {
 	DIM(SORT_GLOBAL_INS_LAT, "ins_lat", sort_global_ins_lat),
 	DIM(SORT_LOCAL_PIPELINE_STAGE_CYC, "local_p_stage_cyc", sort_local_p_stage_cyc),
 	DIM(SORT_GLOBAL_PIPELINE_STAGE_CYC, "p_stage_cyc", sort_global_p_stage_cyc),
+	DIM(SORT_ADDR, "addr", sort_addr),
 };
 
 #undef DIM
tools/perf/util/sort.h +1 -0
index af14eb46c2b6..04ff8b61a2a7 100644
@@ -236,6 +236,7 @@ enum sort_type {
 	SORT_GLOBAL_INS_LAT,
 	SORT_LOCAL_PIPELINE_STAGE_CYC,
 	SORT_GLOBAL_PIPELINE_STAGE_CYC,
+	SORT_ADDR,
 
 	/* branch stack specific sort keys */
 	__SORT_BRANCH_STACK,
Signature · 1 line
2.37.3.998.g577e59143f-goog
reviewer Acked-by on v1 3/4
#6 lore raw

Re: [PATCH 3/4] perf tools: Add 'addr' sort key

Quoted 24 lines · Namhyung Kim

Sometimes users want to see actual (virtual) address of sampled instructions. Add a new 'addr' sort key to display the raw addresses.

  $ perf record -o- true | perf report -i- -s addr
  # To display the perf.data header info, please use --header/--header-only options.
  #
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.000 MB - ]
  #
  # Total Lost Samples: 0
  #
  # Samples: 12  of event 'cycles:u'
  # Event count (approx.): 252512
  #
  # Overhead  Address
  # ........  ..................
  #
      42.96%  0x7f96f08443d7
      29.55%  0x7f96f0859b50
      14.76%  0x7f96f0852e02
       8.30%  0x7f96f0855028
       4.43%  0xffffffff8de01087
Quoted 7 lines

Note that it just compares and displays the sample ip. Each process can have a different memory layout and the ip will be different even if they run the same binary. So this sort key is mostly meaningful for per-process profile data.

Cc Stephane Eranian <[email protected]>

Thanks,
Ian

Quoted 124 lines
---
 tools/perf/Documentation/perf-report.txt |  3 +-
 tools/perf/util/hist.c                   |  1 +
 tools/perf/util/hist.h                   |  1 +
 tools/perf/util/sort.c                   | 38 ++++++++++++++++++++++++
 tools/perf/util/sort.h                   |  1 +
 5 files changed, 43 insertions(+), 1 deletion(-)
tools/perf/Documentation/perf-report.txt +2 -1
index 24efc0583c93..4533db2ee56b 100644
@@ -73,7 +73,7 @@ OPTIONS
        Sort histogram entries by given key(s) - multiple keys can be specified
        in CSV format.  Following sort keys are available:
        pid, comm, dso, symbol, parent, cpu, socket, srcline, weight,
-       local_weight, cgroup_id.
+       local_weight, cgroup_id, addr.
 
        Each key has following meaning:
 
@@ -114,6 +114,7 @@ OPTIONS
        - local_ins_lat: Local instruction latency version
        - p_stage_cyc: On powerpc, this presents the number of cycles spent in a
          pipeline stage. And currently supported only on powerpc.
+       - addr: (Full) virtual address of the sampled instruction
 
        By default, comm, dso and symbol keys are used.
        (i.e. --sort comm,dso,symbol)
tools/perf/util/hist.c +1 -0
index 06f5dbf213ad..17a05e943b44 100644
@@ -215,6 +215,7 @@ void hists__calc_col_len(struct hists *hists, struct hist_entry *h)
        hists__new_col_len(hists, HISTC_GLOBAL_INS_LAT, 13);
        hists__new_col_len(hists, HISTC_LOCAL_P_STAGE_CYC, 13);
        hists__new_col_len(hists, HISTC_GLOBAL_P_STAGE_CYC, 13);
+       hists__new_col_len(hists, HISTC_ADDR, BITS_PER_LONG / 4 + 2);
 
        if (symbol_conf.nanosecs)
                hists__new_col_len(hists, HISTC_TIME, 16);
tools/perf/util/hist.h +1 -0
index c7a7a3fa0b87..ebd8a8f783ee 100644
@@ -79,6 +79,7 @@ enum hist_column {
        HISTC_GLOBAL_P_STAGE_CYC,
        HISTC_ADDR_FROM,
        HISTC_ADDR_TO,
+       HISTC_ADDR,
        HISTC_NR_COLS, /* Last entry */
 };
 
tools/perf/util/sort.c +38 -0
index 6d5588e80935..2e7330867e2e 100644
@@ -1948,6 +1948,43 @@ struct sort_entry sort_dso_size = {
        .se_width_idx   = HISTC_DSO_SIZE,
 };
 
+/* --sort dso_size */
+
+static int64_t
+sort__addr_cmp(struct hist_entry *left, struct hist_entry *right)
+{
+       u64 left_ip = left->ip;
+       u64 right_ip = right->ip;
+       struct map *left_map = left->ms.map;
+       struct map *right_map = right->ms.map;
+
+       if (left_map)
+               left_ip = left_map->unmap_ip(left_map, left_ip);
+       if (right_map)
+               right_ip = right_map->unmap_ip(right_map, right_ip);
+
+       return _sort__addr_cmp(left_ip, right_ip);
+}
+
+static int hist_entry__addr_snprintf(struct hist_entry *he, char *bf,
+                                    size_t size, unsigned int width)
+{
+       u64 ip = he->ip;
+       struct map *map = he->ms.map;
+
+       if (map)
+               ip = map->unmap_ip(map, ip);
+
+       return repsep_snprintf(bf, size, "%-#*llx", width, ip);
+}
+
+struct sort_entry sort_addr = {
+       .se_header      = "Address",
+       .se_cmp         = sort__addr_cmp,
+       .se_snprintf    = hist_entry__addr_snprintf,
+       .se_width_idx   = HISTC_ADDR,
+};
+
 
 struct sort_dimension {
        const char              *name;
@@ -1997,6 +2034,7 @@ static struct sort_dimension common_sort_dimensions[] = {
        DIM(SORT_GLOBAL_INS_LAT, "ins_lat", sort_global_ins_lat),
        DIM(SORT_LOCAL_PIPELINE_STAGE_CYC, "local_p_stage_cyc", sort_local_p_stage_cyc),
        DIM(SORT_GLOBAL_PIPELINE_STAGE_CYC, "p_stage_cyc", sort_global_p_stage_cyc),
+       DIM(SORT_ADDR, "addr", sort_addr),
 };
 
 #undef DIM
tools/perf/util/sort.h +1 -1
index af14eb46c2b6..04ff8b61a2a7 100644
@@ -236,6 +236,7 @@ enum sort_type {
        SORT_GLOBAL_INS_LAT,
        SORT_LOCAL_PIPELINE_STAGE_CYC,
        SORT_GLOBAL_PIPELINE_STAGE_CYC,
+       SORT_ADDR,
 
        /* branch stack specific sort keys */
        __SORT_BRANCH_STACK,
--

2.37.3.998.g577e59143f-goog

author maintainer
#7 lore raw

PATCH 4/4 perf annotate: Toggle full address <-> offset display

In reply to Namhyung Kim

Handle 'f' key to toggle the display offset and full address. Obviously it only works when users set to see disassembler output ('o' key). It'd be useful when users want to see the full virtual address in the TUI annotate browser.

Cc Stephane Eranian <[email protected]>
---
 tools/perf/ui/browsers/annotate.c |  6 +++++-
 tools/perf/util/annotate.c        | 19 ++++++++++++++++++-
 tools/perf/util/annotate.h        |  4 +++-
 3 files changed, 26 insertions(+), 3 deletions(-)
tools/perf/ui/browsers/annotate.c +5 -1
index 9bc1076374ff..725662e21b23 100644
@@ -805,7 +805,8 @@ static int annotate_browser__run(struct annotate_browser *browser,
 		"r             Run available scripts\n"
 		"p             Toggle percent type [local/global]\n"
 		"b             Toggle percent base [period/hits]\n"
-		"?             Search string backwards\n");
+		"?             Search string backwards\n"
+		"f             Toggle showing offsets to full address\n");
 			continue;
 		case 'r':
 			script_browse(NULL, NULL);
@@ -912,6 +913,9 @@ static int annotate_browser__run(struct annotate_browser *browser,
 			hists__scnprintf_title(hists, title, sizeof(title));
 			annotate_browser__show(&browser->b, title, help);
 			continue;
+		case 'f':
+			annotation__toggle_full_addr(notes, ms);
+			continue;
 		case K_LEFT:
 		case K_ESC:
 		case 'q':
tools/perf/util/annotate.c +18 -1
index 5bc63c9e0324..db475e44f42f 100644
@@ -2239,7 +2239,10 @@ int symbol__annotate(struct map_symbol *ms, struct evsel *evsel,
 	}
 
 	args.ms = *ms;
-	notes->start = map__rip_2objdump(ms->map, sym->start);
+	if (notes->options && notes->options->full_addr)
+		notes->start = map__objdump_2mem(ms->map, ms->sym->start);
+	else
+		notes->start = map__rip_2objdump(ms->map, ms->sym->start);
 
 	return symbol__disassemble(sym, &args);
 }
@@ -2762,6 +2765,8 @@ void annotation__update_column_widths(struct annotation *notes)
 {
 	if (notes->options->use_offset)
 		notes->widths.target = notes->widths.min_addr;
+	else if (notes->options->full_addr)
+		notes->widths.target = BITS_PER_LONG / 4;
 	else
 		notes->widths.target = notes->widths.max_addr;
 
@@ -2771,6 +2776,18 @@ void annotation__update_column_widths(struct annotation *notes)
 		notes->widths.addr += notes->widths.jumps + 1;
 }
 
+void annotation__toggle_full_addr(struct annotation *notes, struct map_symbol *ms)
+{
+	notes->options->full_addr = !notes->options->full_addr;
+
+	if (notes->options->full_addr)
+		notes->start = map__objdump_2mem(ms->map, ms->sym->start);
+	else
+		notes->start = map__rip_2objdump(ms->map, ms->sym->start);
+
+	annotation__update_column_widths(notes);
+}
+
 static void annotation__calc_lines(struct annotation *notes, struct map *map,
 				   struct rb_root *root,
 				   struct annotation_options *opts)
tools/perf/util/annotate.h +3 -1
index 3cbd883e4d7a..8934072c39e6 100644
@@ -88,7 +88,8 @@ struct annotation_options {
 	     show_nr_jumps,
 	     show_minmax_cycle,
 	     show_asm_raw,
-	     annotate_src;
+	     annotate_src,
+	     full_addr;
 	u8   offset_level;
 	int  min_pcnt;
 	int  max_lines;
@@ -325,6 +326,7 @@ void annotation__compute_ipc(struct annotation *notes, size_t size);
 void annotation__mark_jump_targets(struct annotation *notes, struct symbol *sym);
 void annotation__update_column_widths(struct annotation *notes);
 void annotation__init_column_widths(struct annotation *notes, struct symbol *sym);
+void annotation__toggle_full_addr(struct annotation *notes, struct map_symbol *ms);
 
 static inline struct sym_hist *annotated_source__histogram(struct annotated_source *src, int idx)
 {
Signature · 1 line
2.37.3.998.g577e59143f-goog
reviewer Acked-by on v1 4/4
#8 lore raw

Re: [PATCH 4/4] perf annotate: Toggle full address <-> offset display

Quoted 8 lines · Namhyung Kim

Handle 'f' key to toggle the display offset and full address. Obviously it only works when users set to see disassembler output ('o' key). It'd be useful when users want to see the full virtual address in the TUI annotate browser.

Cc Stephane Eranian <[email protected]>

Thanks,
Ian

Quoted 99 lines
---
 tools/perf/ui/browsers/annotate.c |  6 +++++-
 tools/perf/util/annotate.c        | 19 ++++++++++++++++++-
 tools/perf/util/annotate.h        |  4 +++-
 3 files changed, 26 insertions(+), 3 deletions(-)
tools/perf/ui/browsers/annotate.c +5 -1
index 9bc1076374ff..725662e21b23 100644
@@ -805,7 +805,8 @@ static int annotate_browser__run(struct annotate_browser *browser,
                "r             Run available scripts\n"
                "p             Toggle percent type [local/global]\n"
                "b             Toggle percent base [period/hits]\n"
-               "?             Search string backwards\n");
+               "?             Search string backwards\n"
+               "f             Toggle showing offsets to full address\n");
                        continue;
                case 'r':
                        script_browse(NULL, NULL);
@@ -912,6 +913,9 @@ static int annotate_browser__run(struct annotate_browser *browser,
                        hists__scnprintf_title(hists, title, sizeof(title));
                        annotate_browser__show(&browser->b, title, help);
                        continue;
+               case 'f':
+                       annotation__toggle_full_addr(notes, ms);
+                       continue;
                case K_LEFT:
                case K_ESC:
                case 'q':
tools/perf/util/annotate.c +18 -1
index 5bc63c9e0324..db475e44f42f 100644
@@ -2239,7 +2239,10 @@ int symbol__annotate(struct map_symbol *ms, struct evsel *evsel,
        }
 
        args.ms = *ms;
-       notes->start = map__rip_2objdump(ms->map, sym->start);
+       if (notes->options && notes->options->full_addr)
+               notes->start = map__objdump_2mem(ms->map, ms->sym->start);
+       else
+               notes->start = map__rip_2objdump(ms->map, ms->sym->start);
 
        return symbol__disassemble(sym, &args);
 }
@@ -2762,6 +2765,8 @@ void annotation__update_column_widths(struct annotation *notes)
 {
        if (notes->options->use_offset)
                notes->widths.target = notes->widths.min_addr;
+       else if (notes->options->full_addr)
+               notes->widths.target = BITS_PER_LONG / 4;
        else
                notes->widths.target = notes->widths.max_addr;
 
@@ -2771,6 +2776,18 @@ void annotation__update_column_widths(struct annotation *notes)
                notes->widths.addr += notes->widths.jumps + 1;
 }
 
+void annotation__toggle_full_addr(struct annotation *notes, struct map_symbol *ms)
+{
+       notes->options->full_addr = !notes->options->full_addr;
+
+       if (notes->options->full_addr)
+               notes->start = map__objdump_2mem(ms->map, ms->sym->start);
+       else
+               notes->start = map__rip_2objdump(ms->map, ms->sym->start);
+
+       annotation__update_column_widths(notes);
+}
+
 static void annotation__calc_lines(struct annotation *notes, struct map *map,
                                   struct rb_root *root,
                                   struct annotation_options *opts)
tools/perf/util/annotate.h +3 -2
index 3cbd883e4d7a..8934072c39e6 100644
@@ -88,7 +88,8 @@ struct annotation_options {
             show_nr_jumps,
             show_minmax_cycle,
             show_asm_raw,
-            annotate_src;
+            annotate_src,
+            full_addr;
        u8   offset_level;
        int  min_pcnt;
        int  max_lines;
@@ -325,6 +326,7 @@ void annotation__compute_ipc(struct annotation *notes, size_t size);
 void annotation__mark_jump_targets(struct annotation *notes, struct symbol *sym);
 void annotation__update_column_widths(struct annotation *notes);
 void annotation__init_column_widths(struct annotation *notes, struct symbol *sym);
+void annotation__toggle_full_addr(struct annotation *notes, struct map_symbol *ms);
 
 static inline struct sym_hist *annotated_source__histogram(struct annotated_source *src, int idx)
 {
--

2.37.3.998.g577e59143f-goog

maintainer Applied
#9 lore raw
In reply to Namhyung Kim

Em Fri, Sep 23, 2022 at 10:31:38AM -0700, Namhyung Kim escreveu:

Quoted 4 lines

Hello,

This is collection of random fixes and updates. And I'm resending them as they seem to be lost in the list.

Thanks, applied.

- Arnaldo

Quoted 24 lines

Thanks,
Namhyung

Namhyung Kim (4):
perf record: Fix a segfault in record__read_lost_samples() perf inject: Clarify build-id options a little bit perf tools: Add 'addr' sort key
perf annotate: Toggle full address <-> offset display

 tools/perf/Documentation/perf-inject.txt |  6 ++--
 tools/perf/Documentation/perf-report.txt |  3 +-
 tools/perf/builtin-record.c              |  6 ++++
 tools/perf/ui/browsers/annotate.c        |  6 +++-
 tools/perf/util/annotate.c               | 19 +++++++++++-
 tools/perf/util/annotate.h               |  4 ++-
 tools/perf/util/hist.c                   |  1 +
 tools/perf/util/hist.h                   |  1 +
 tools/perf/util/sort.c                   | 38 ++++++++++++++++++++++++
 tools/perf/util/sort.h                   |  1 +
 10 files changed, 79 insertions(+), 6 deletions(-)
Signature · 1 line
2.37.3.998.g577e59143f-goog
Signature · 1 line
- Arnaldo