From 8a99255a50c0b4c2a449b96fd8d45fcc8d72c701 Mon Sep 17 00:00:00 2001 From: Jin Yao Date: Thu, 3 Jan 2019 15:40:45 +0800 Subject: perf stat: Fix endless wait for child process We hit a 'perf stat' issue by using following script: #!/bin/bash sleep 1000 & exec perf stat -a -e cycles -I1000 -- sleep 5 Since "perf stat" is launched by exec, the "sleep 1000" would be the child process of "perf stat". The wait4() call will not return because it's waiting for the child process "sleep 1000" to end. So 'perf stat' doesn't return even after 5s passes. This patch lets 'perf stat' return when the specified child process ends (in this case, the specified child process is "sleep 5"). Committer testing: # cat test.sh #!/bin/bash sleep 10 & exec perf stat -a -e cycles -I1000 -- sleep 5 # Before: # time ./test.sh # time counts unit events 1.001113090 108,453,351 cycles 2.002062196 142,075,435 cycles 3.002896194 164,801,068 cycles 4.003731666 107,062,140 cycles 5.002068867 112,241,832 cycles real 0m10.066s user 0m0.016s sys 0m0.101s # After: # time ./test.sh # time counts unit events 1.001016096 91,412,027 cycles 2.002014963 124,063,708 cycles 3.002883964 125,993,929 cycles 4.003706470 120,465,734 cycles 5.002006778 163,560,355 cycles real 0m5.123s user 0m0.014s sys 0m0.105s # Signed-off-by: Jin Yao Reviewed-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Alexander Shishkin Cc: Andi Kleen Cc: Kan Liang Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1546501245-4512-1-git-send-email-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-stat.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools/perf') diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 1410d66192f7..63a3afc7f32b 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -561,7 +561,8 @@ try_again: break; } } - wait4(child_pid, &status, 0, &stat_config.ru_data); + if (child_pid != -1) + wait4(child_pid, &status, 0, &stat_config.ru_data); if (workload_exec_errno) { const char *emsg = str_error_r(workload_exec_errno, msg, sizeof(msg)); -- cgit v1.2.3 From 805e4c8b61bd62f6d3f416c915cb9bb423835fee Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 3 Jan 2019 14:59:11 -0300 Subject: tools beauty: Make the prctl option table generator catch all PR_ options In ba8308856564 ("arm64: add prctl control for resetting ptrauth keys") the PR_PAC_RESET_KEYS prctl option was introduced, get that into the regex in addition to PR_GET_* and PR_SET_*: So just get everything that matches '^#define PR_\w+' this ends up adding these entries: $ tools/perf/trace/beauty/prctl_option.sh > after $ diff -u before after --- before 2019-01-03 14:58:51.541807353 -0300 +++ after 2019-01-03 15:17:05.909583804 -0300 @@ -19,12 +19,18 @@ [20] = "SET_ENDIAN", [21] = "GET_SECCOMP", [22] = "SET_SECCOMP", + [23] = "CAPBSET_READ", + [24] = "CAPBSET_DROP", [25] = "GET_TSC", [26] = "SET_TSC", [27] = "GET_SECUREBITS", [28] = "SET_SECUREBITS", [29] = "SET_TIMERSLACK", [30] = "GET_TIMERSLACK", + [31] = "TASK_PERF_EVENTS_DISABLE", + [32] = "TASK_PERF_EVENTS_ENABLE", + [33] = "MCE_KILL", + [34] = "MCE_KILL_GET", [35] = "SET_MM", [36] = "SET_CHILD_SUBREAPER", [37] = "GET_CHILD_SUBREAPER", @@ -33,8 +39,13 @@ [40] = "GET_TID_ADDRESS", [41] = "SET_THP_DISABLE", [42] = "GET_THP_DISABLE", + [43] = "MPX_ENABLE_MANAGEMENT", + [44] = "MPX_DISABLE_MANAGEMENT", [45] = "SET_FP_MODE", [46] = "GET_FP_MODE", + [47] = "CAP_AMBIENT", + [50] = "SVE_SET_VL", + [51] = "SVE_GET_VL", [52] = "GET_SPECULATION_CTRL", [53] = "SET_SPECULATION_CTRL", [54] = "PAC_RESET_KEYS", $ Cc: Adrian Hunter Cc: Jiri Olsa Cc: Kristina Martsenko Cc: Namhyung Kim Cc: Will Deacon Link: https://lkml.kernel.org/n/tip-sg2pkmtjr5988bhbcp4yp6sw@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/trace/beauty/prctl_option.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools/perf') diff --git a/tools/perf/trace/beauty/prctl_option.sh b/tools/perf/trace/beauty/prctl_option.sh index d32f8f1124af..3109d7b05e11 100755 --- a/tools/perf/trace/beauty/prctl_option.sh +++ b/tools/perf/trace/beauty/prctl_option.sh @@ -4,7 +4,7 @@ [ $# -eq 1 ] && header_dir=$1 || header_dir=tools/include/uapi/linux/ printf "static const char *prctl_options[] = {\n" -regex='^#define[[:space:]]+PR_([GS]ET\w+)[[:space:]]*([[:xdigit:]]+).*' +regex='^#define[[:space:]]+PR_(\w+)[[:space:]]*([[:xdigit:]]+).*' egrep $regex ${header_dir}/prctl.h | grep -v PR_SET_PTRACER | \ sed -r "s/$regex/\2 \1/g" | \ sort -n | xargs printf "\t[%s] = \"%s\",\n" -- cgit v1.2.3 From a3366db06bb656cef2e03f30f780d93059bcc594 Mon Sep 17 00:00:00 2001 From: Jin Yao Date: Fri, 4 Jan 2019 14:10:30 +0800 Subject: perf report: Fix wrong iteration count in --branch-history By calculating the removed loops, we can get the iteration count. But the iteration count could be reported incorrectly, reporting impossibly high counts. That's because previous code uses the number of removed LBR entries for the iteration count. That's not good. Fix this by increasing the iteration count when a loop is detected. When matching the chain, the iteration count would be added up, finally we need to compute the average value when printing out. For example, $ perf report --branch-history --stdio --no-children Before: ---f2 +0 | |--33.62%--f1 +9 (cycles:1) | f1 +0 | main +22 (cycles:1) | main +17 | main +38 (cycles:1) | main +27 | f1 +26 (cycles:1) | f1 +24 | f2 +27 (cycles:7) | f2 +0 | f1 +19 (cycles:1) | f1 +14 | f2 +27 (cycles:11) | f2 +0 | f1 +9 (cycles:1 iter:2968 avg_cycles:3) | f1 +0 | main +22 (cycles:1 iter:2968 avg_cycles:3) | main +17 | main +38 (cycles:1 iter:2968 avg_cycles:3) 2968 is an impossible high iteration count and avg_cycles is too small. After: ---f2 +0 | |--33.62%--f1 +9 (cycles:1) | f1 +0 | main +22 (cycles:1) | main +17 | main +38 (cycles:1) | main +27 | f1 +26 (cycles:1) | f1 +24 | f2 +27 (cycles:7) | f2 +0 | f1 +19 (cycles:1) | f1 +14 | f2 +27 (cycles:11) | f2 +0 | f1 +9 (cycles:1 iter:1 avg_cycles:23) | f1 +0 | main +22 (cycles:1 iter:1 avg_cycles:23) | main +17 | main +38 (cycles:1 iter:1 avg_cycles:23) avg_cycles:23 is the average cycles of this iteration. Fixes: c4ee06251d42 ("perf report: Calculate the average cycles of iterations") Signed-off-by: Jin Yao Cc: Alexander Shishkin Cc: Andi Kleen Cc: Jiri Olsa Cc: Kan Liang Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1546582230-17507-1-git-send-email-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/callchain.c | 32 ++++++++++++++++++++------------ tools/perf/util/callchain.h | 1 + tools/perf/util/machine.c | 2 +- 3 files changed, 22 insertions(+), 13 deletions(-) (limited to 'tools/perf') diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c index 32ef7bdca1cf..dc2212e12184 100644 --- a/tools/perf/util/callchain.c +++ b/tools/perf/util/callchain.c @@ -766,6 +766,7 @@ static enum match_result match_chain(struct callchain_cursor_node *node, cnode->cycles_count += node->branch_flags.cycles; cnode->iter_count += node->nr_loop_iter; cnode->iter_cycles += node->iter_cycles; + cnode->from_count++; } } @@ -1345,10 +1346,10 @@ static int branch_to_str(char *bf, int bfsize, static int branch_from_str(char *bf, int bfsize, u64 branch_count, u64 cycles_count, u64 iter_count, - u64 iter_cycles) + u64 iter_cycles, u64 from_count) { int printed = 0, i = 0; - u64 cycles; + u64 cycles, v = 0; cycles = cycles_count / branch_count; if (cycles) { @@ -1357,14 +1358,16 @@ static int branch_from_str(char *bf, int bfsize, bf + printed, bfsize - printed); } - if (iter_count) { - printed += count_pri64_printf(i++, "iter", - iter_count, - bf + printed, bfsize - printed); + if (iter_count && from_count) { + v = iter_count / from_count; + if (v) { + printed += count_pri64_printf(i++, "iter", + v, bf + printed, bfsize - printed); - printed += count_pri64_printf(i++, "avg_cycles", - iter_cycles / iter_count, - bf + printed, bfsize - printed); + printed += count_pri64_printf(i++, "avg_cycles", + iter_cycles / iter_count, + bf + printed, bfsize - printed); + } } if (i) @@ -1377,6 +1380,7 @@ static int counts_str_build(char *bf, int bfsize, u64 branch_count, u64 predicted_count, u64 abort_count, u64 cycles_count, u64 iter_count, u64 iter_cycles, + u64 from_count, struct branch_type_stat *brtype_stat) { int printed; @@ -1389,7 +1393,8 @@ static int counts_str_build(char *bf, int bfsize, predicted_count, abort_count, brtype_stat); } else { printed = branch_from_str(bf, bfsize, branch_count, - cycles_count, iter_count, iter_cycles); + cycles_count, iter_count, iter_cycles, + from_count); } if (!printed) @@ -1402,13 +1407,14 @@ static int callchain_counts_printf(FILE *fp, char *bf, int bfsize, u64 branch_count, u64 predicted_count, u64 abort_count, u64 cycles_count, u64 iter_count, u64 iter_cycles, + u64 from_count, struct branch_type_stat *brtype_stat) { char str[256]; counts_str_build(str, sizeof(str), branch_count, predicted_count, abort_count, cycles_count, - iter_count, iter_cycles, brtype_stat); + iter_count, iter_cycles, from_count, brtype_stat); if (fp) return fprintf(fp, "%s", str); @@ -1422,6 +1428,7 @@ int callchain_list_counts__printf_value(struct callchain_list *clist, u64 branch_count, predicted_count; u64 abort_count, cycles_count; u64 iter_count, iter_cycles; + u64 from_count; branch_count = clist->branch_count; predicted_count = clist->predicted_count; @@ -1429,11 +1436,12 @@ int callchain_list_counts__printf_value(struct callchain_list *clist, cycles_count = clist->cycles_count; iter_count = clist->iter_count; iter_cycles = clist->iter_cycles; + from_count = clist->from_count; return callchain_counts_printf(fp, bf, bfsize, branch_count, predicted_count, abort_count, cycles_count, iter_count, iter_cycles, - &clist->brtype_stat); + from_count, &clist->brtype_stat); } static void free_callchain_node(struct callchain_node *node) diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h index 154560b1eb65..99d38ac019b8 100644 --- a/tools/perf/util/callchain.h +++ b/tools/perf/util/callchain.h @@ -118,6 +118,7 @@ struct callchain_list { bool has_children; }; u64 branch_count; + u64 from_count; u64 predicted_count; u64 abort_count; u64 cycles_count; diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 6fcb3bce0442..143f7057d581 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -2005,7 +2005,7 @@ static void save_iterations(struct iterations *iter, { int i; - iter->nr_loop_iter = nr; + iter->nr_loop_iter++; iter->cycles = 0; for (i = 0; i < nr; i++) -- cgit v1.2.3 From 442b4eb3af44906fcbb526d98c314b27f8c9acf3 Mon Sep 17 00:00:00 2001 From: Ivan Krylov Date: Sun, 14 Oct 2018 11:18:03 +0300 Subject: perf annotate: Pass filename to objdump via execl The symbol__disassemble() function uses shell to launch objdump and filter its output via grep. Passing filenames by interpolating them into the command line via "%s" may lead to problems if said filenames contain special characters. Instead, pass the filename as a command line argument where it is not subject to any kind of interpretation, then use quoted shell interpolation to build the strings we need safely. Signed-off-by: Ivan Krylov Acked-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Alexander Shishkin Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20181014111803.5d83b806@Tarkus Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/annotate.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'tools/perf') diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c index ac9805e0bc76..70de8f6b3aee 100644 --- a/tools/perf/util/annotate.c +++ b/tools/perf/util/annotate.c @@ -1723,15 +1723,14 @@ static int symbol__disassemble(struct symbol *sym, struct annotate_args *args) err = asprintf(&command, "%s %s%s --start-address=0x%016" PRIx64 " --stop-address=0x%016" PRIx64 - " -l -d %s %s -C \"%s\" 2>/dev/null|grep -v \"%s:\"|expand", + " -l -d %s %s -C \"$1\" 2>/dev/null|grep -v \"$1:\"|expand", opts->objdump_path ?: "objdump", opts->disassembler_style ? "-M " : "", opts->disassembler_style ?: "", map__rip_2objdump(map, sym->start), map__rip_2objdump(map, sym->end), opts->show_asm_raw ? "" : "--no-show-raw", - opts->annotate_src ? "-S" : "", - symfs_filename, symfs_filename); + opts->annotate_src ? "-S" : ""); if (err < 0) { pr_err("Failure allocating memory for the command to run\n"); @@ -1756,7 +1755,8 @@ static int symbol__disassemble(struct symbol *sym, struct annotate_args *args) close(stdout_fd[0]); dup2(stdout_fd[1], 1); close(stdout_fd[1]); - execl("/bin/sh", "sh", "-c", command, NULL); + execl("/bin/sh", "sh", "-c", command, "--", symfs_filename, + NULL); perror(command); exit(-1); } -- cgit v1.2.3 From 099be748865eece21362aee416c350c0b1ae34df Mon Sep 17 00:00:00 2001 From: Mattias Jacobsson Date: Sat, 29 Dec 2018 15:17:50 +0100 Subject: perf strbuf: Remove redundant va_end() in strbuf_addv() Each call to va_copy() should have one, and only one, corresponding call to va_end(). In strbuf_addv() some code paths result in va_end() getting called multiple times. Remove the superfluous va_end(). Signed-off-by: Mattias Jacobsson <2pi@mok.nu> Cc: Jiri Olsa Cc: Peter Zijlstra Cc: Sanskriti Sharma Link: http://lkml.kernel.org/r/20181229141750.16945-1-2pi@mok.nu Fixes: ce49d8436cff ("perf strbuf: Match va_{add,copy} with va_end") Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/strbuf.c | 1 - 1 file changed, 1 deletion(-) (limited to 'tools/perf') diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c index 9005fbe0780e..23092fd6451d 100644 --- a/tools/perf/util/strbuf.c +++ b/tools/perf/util/strbuf.c @@ -109,7 +109,6 @@ static int strbuf_addv(struct strbuf *sb, const char *fmt, va_list ap) return ret; } len = vsnprintf(sb->buf + sb->len, sb->alloc - sb->len, fmt, ap_saved); - va_end(ap_saved); if (len > strbuf_avail(sb)) { pr_debug("this should not happen, your vsnprintf is broken"); va_end(ap_saved); -- cgit v1.2.3 From f712a86c1406abf50b0dcff0500c723e2aca050a Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 3 Jan 2019 16:03:46 -0300 Subject: perf python: Make sure the python binding output directory is in place Instead of doing an unconditional mkdir, use a dummy Makefile variable to check if the directory is there and if not, create it. This is better than what we had and will help with other python bindings that are in development, like one involved with python backtraces. Cc: Adrian Hunter Cc: Jiri Olsa Cc: Namhyung Kim Link: https://lkml.kernel.org/n/tip-iis6us2nocw3y4uuoon9osd7@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Makefile.perf | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'tools/perf') diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf index ff29c3372ec3..2921f829a0f4 100644 --- a/tools/perf/Makefile.perf +++ b/tools/perf/Makefile.perf @@ -524,12 +524,14 @@ $(arch_errno_name_array): $(arch_errno_tbl) all: shell_compatibility_test $(ALL_PROGRAMS) $(LANG_BINDINGS) $(OTHER_PROGRAMS) +# Create python binding output directory if not already present +_dummy := $(shell [ -d '$(OUTPUT)python' ] || mkdir -p '$(OUTPUT)python') + $(OUTPUT)python/perf.so: $(PYTHON_EXT_SRCS) $(PYTHON_EXT_DEPS) $(LIBTRACEEVENT_DYNAMIC_LIST) $(QUIET_GEN)LDSHARED="$(CC) -pthread -shared" \ CFLAGS='$(CFLAGS)' LDFLAGS='$(LDFLAGS) $(LIBTRACEEVENT_DYNAMIC_LIST_LDFLAGS)' \ $(PYTHON_WORD) util/setup.py \ --quiet build_ext; \ - mkdir -p $(OUTPUT)python && \ cp $(PYTHON_EXTBUILD_LIB)perf*.so $(OUTPUT)python/ please_set_SHELL_PATH_to_a_more_modern_shell: -- cgit v1.2.3 From 03fa483821c0b4db7c2b1453d3332f397d82313f Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Fri, 4 Jan 2019 15:10:00 -0300 Subject: perf test shell: Use a fallback to get the pathname in vfs_getname Some kernels, like 4.19.13-300.fc29.x86_64 in fedora 29, fail with the existing probe definition asking for the contents of result->name, working when we ask for the 'filename' variable instead, so add a fallback to that. Now those tests are back working on fedora 29 systems with that kernel: # perf test vfs_getname 65: Use vfs_getname probe to get syscall args filenames : Ok 66: Add vfs_getname probe to get syscall args filenames : Ok 67: Check open filename arg using perf trace + vfs_getname: Ok # Cc: Adrian Hunter Cc: Jiri Olsa Cc: Namhyung Kim Link: https://lkml.kernel.org/n/tip-klt3n0i58dfqttveti09q3fi@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/tests/shell/lib/probe_vfs_getname.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools/perf') diff --git a/tools/perf/tests/shell/lib/probe_vfs_getname.sh b/tools/perf/tests/shell/lib/probe_vfs_getname.sh index 1c16e56cd93e..7cb99b433888 100644 --- a/tools/perf/tests/shell/lib/probe_vfs_getname.sh +++ b/tools/perf/tests/shell/lib/probe_vfs_getname.sh @@ -13,7 +13,8 @@ add_probe_vfs_getname() { local verbose=$1 if [ $had_vfs_getname -eq 1 ] ; then line=$(perf probe -L getname_flags 2>&1 | egrep 'result.*=.*filename;' | sed -r 's/[[:space:]]+([[:digit:]]+)[[:space:]]+result->uptr.*/\1/') - perf probe $verbose "vfs_getname=getname_flags:${line} pathname=result->name:string" + perf probe -q "vfs_getname=getname_flags:${line} pathname=result->name:string" || \ + perf probe $verbose "vfs_getname=getname_flags:${line} pathname=filename:string" fi } -- cgit v1.2.3