BMC/Intel-BMC/linux.git/tools/perf, branch dev-5.14-intel

perf iostat: Fix Segmentation fault from NULL 'struct perf_counts_values *'

2021-10-07T05:53:04+00:00

[ Upstream commit 4da8b121884d84476f3d50d46a471471af1aa9df ] If the 'perf iostat' user specifies two or more iio_root_ports and also specifies the cpu(s) by -C which is not *connected to all* the above iio ports, the iostat_print_metric() will run into trouble: For example: $ perf iostat list S0-uncore_iio_0<0000:16> S1-uncore_iio_0<0000:97> # <--- CPU 1 is located in the socket S0 $ perf iostat 0000:16,0000:97 -C 1 -- ls port Inbound Read(MB) Inbound Write(MB) Outbound Read(MB) Outbound Write(MB) ../perf-iostat: line 12: 104418 Segmentation fault (core dumped) perf stat --iostat$DELIMITER$* The core-dump stack says, in the above corner case, the returned (struct perf_counts_values *) count will be NULL, and the caller iostat_print_metric() apparently doesn't not handle this case. 433 struct perf_counts_values *count = perf_counts(evsel->counts, die, 0); 434 435 if (count->run && count->ena) { (gdb) p count $1 = (struct perf_counts_values *) 0x0 The deeper reason is that there are actually no statistics from the user specified pair "iostat 0000:X, -C (disconnected) Y ", but let's fix it with minimum cost by adding a NULL check in the user space. Fixes: f9ed693e8bc0e7de ("perf stat: Enable iostat mode for x86 platforms") Signed-off-by: Like Xu Cc: Alexander Antonov Cc: Alexander Shishkin Cc: Ian Rogers Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lore.kernel.org/lkml/20210927081115.39568-2-likexu@tencent.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf iostat: Use system-wide mode if the target cpu_list is unspecified

2021-10-07T05:53:04+00:00

[ Upstream commit e4fe5d7349e0b1c0d3da5b6b3e1efce591e85bd2 ] An iostate use case like "perf iostat 0000:16,0000:97 -- ls" should be implemented to work in system-wide mode to ensure that the output from print_header() is consistent with the user documentation perf-iostat.txt, rather than incorrectly assuming that the kernel does not support it: Error: The sys_perf_event_open() syscall returned with 22 (Invalid argument) \ for event (uncore_iio_0/event=0x83,umask=0x04,ch_mask=0xF,fc_mask=0x07/). /bin/dmesg | grep -i perf may provide additional information. This error is easily fixed by assigning system-wide mode by default for IOSTAT_RUN only when the target cpu_list is unspecified. Fixes: f07952b179697771 ("perf stat: Basic support for iostat in perf") Signed-off-by: Like Xu Cc: Alexander Antonov Cc: Alexander Shishkin Cc: Ian Rogers Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lore.kernel.org/lkml/20210927081115.39568-1-likexu@tencent.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf test: Fix DWARF unwind for optimized builds.

2021-10-07T05:53:04+00:00

[ Upstream commit 5c34aea341b16e29fde6e6c8d4b18866cd99754d ] To ensure the stack frames are on the stack tail calls optimizations need to be inhibited. If your compiler supports an attribute use it, otherwise use an asm volatile barrier. The barrier fix was suggested here: https://lore.kernel.org/lkml/20201028081123.GT2628@hirez.programming.kicks-ass.net/ Tested with an optimized clang build and by forcing the asm barrier route with an optimized clang build. A GCC bug tracking a proper disable_tail_calls is: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97831 Fixes: 9ae1e990f1ab ("perf tools: Remove broken __no_tail_call attribute") v2. is a rebase. The original fix patch generated quite a lot of discussion over the right place for the fix: https://lore.kernel.org/lkml/20201114000803.909530-1-irogers@google.com/ The patch reflects my preference of it being near the use, so that future code cleanups don't break this somewhat special usage. Signed-off-by: Ian Rogers Acked-by: Jiri Olsa Cc: Alexander Shishkin Cc: Ard Biesheuvel Cc: Mark Rutland Cc: Miguel Ojeda Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Cc: clang-built-linux@googlegroups.com Link: http://lore.kernel.org/lkml/20210922173812.456348-1-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf tools: Allow build-id with trailing zeros

2021-09-26T12:10:16+00:00

commit 4a86d41404005a3c7e7b6065e8169ac6202887a9 upstream. Currently perf saves a build-id with size but old versions assumes the size of 20. In case the build-id is less than 20 (like for MD5), it'd fill the rest with 0s. I saw a problem when old version of perf record saved a binary in the build-id cache and new version of perf reads the data. The symbols should be read from the build-id cache (as the path no longer has the same binary) but it failed due to mismatch in the build-id. symsrc__init: build id mismatch for /home/namhyung/.debug/.build-id/53/e4c2f42a4c61a2d632d92a72afa08f00000000/elf. The build-id event in the data has 20 byte build-ids, but it saw a different size (16) when it reads the build-id of the elf file in the build-id cache. $ readelf -n ~/.debug/.build-id/53/e4c2f42a4c61a2d632d92a72afa08f00000000/elf Displaying notes found in: .note.gnu.build-id Owner Data size Description GNU 0x00000010 NT_GNU_BUILD_ID (unique build ID bitstring) Build ID: 53e4c2f42a4c61a2d632d92a72afa08f Let's fix this by allowing trailing zeros if the size is different. Fixes: 39be8d0115b321ed ("perf tools: Pass build_id object to dso__build_id_equal()") Signed-off-by: Namhyung Kim Acked-by: Jiri Olsa Cc: Andi Kleen Cc: Ian Rogers Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20210910224630.1084877-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Greg Kroah-Hartman

perf symbol: Look for ImageBase in PE file to compute .text offset

2021-09-26T12:10:16+00:00

commit d2930ede5218be28413a00130a6895d14393c325 upstream. Instead of using the file offset in the debug file. This fixes a regression from 00a3423492bc90be ("perf symbols: Make dso__load_bfd_symbols() load PE files from debug cache only"), causing incorrect symbol resolution when debug file have been stripped from non-debug sections (in which case its .text section is empty and doesn't have any file position). The debug files could also be created with a different file alignment, and have different file positions from the mmap-ed binary, or have the section reordered. This instead looks for the file image base, using the corresponding bfd *ABS* symbols. As PE symbols only have 4 bytes, it also needs to keep .text section vma high bits. Signed-off-by: Remi Bernon Fixes: 00a3423492bc90be ("perf symbols: Make dso__load_bfd_symbols() load PE files from debug cache only") Cc: Alexander Shishkin Cc: Jiri Olsa Cc: Nicholas Fraser Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20210909192637.4139125-1-rbernon@codeweavers.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Greg Kroah-Hartman

perf test: Fix bpf test sample mismatch reporting

2021-09-26T12:10:16+00:00

commit 3e11300cdfd5f1bc13a05dfc6dccf69aca5dd1dc upstream. When the expected sample count in the condition changed, the message needs to be changed too, otherwise we'll get: 0x1001f2091d8: mmap mask[0]: BPF filter result incorrect, expected 56, got 56 samples Fixes: 4b04e0decd2518e5 ("perf test: Fix basic bpf filtering test") Signed-off-by: Michael Petlan Cc: Jiri Olsa Cc: Sumanth Korikkar Link: https //lore.kernel.org/r/20210805160611.5542-1-mpetlan@redhat.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Greg Kroah-Hartman

perf bench inject-buildid: Handle writen() errors

2021-09-22T10:39:30+00:00

[ Upstream commit edf7b4a2d85e37a1ee77156bddaed4aa6af9c5e1 ] The build on fedora:35 and fedora:rawhide with clang is failing with: 49 41.00 fedora:35 : FAIL clang version 13.0.0 (Fedora 13.0.0~rc1-1.fc35) bench/inject-buildid.c:351:6: error: variable 'len' set but not used [-Werror,-Wunused-but-set-variable] u64 len = 0; ^ 1 error generated. make[3]: *** [/git/perf-5.14.0-rc7/tools/build/Makefile.build:139: bench] Error 2 50 41.11 fedora:rawhide : FAIL clang version 13.0.0 (Fedora 13.0.0~rc1-1.fc35) bench/inject-buildid.c:351:6: error: variable 'len' set but not used [-Werror,-Wunused-but-set-variable] u64 len = 0; ^ 1 error generated. make[3]: *** [/git/perf-5.14.0-rc7/tools/build/Makefile.build:139: bench] Error 2 That 'len' variable is not used at all, so just make sure all the synthesize_RECORD() routines return ssize_t to propagate the writen() return, as it may fail, ditch the 'ret' var and bail out if those routines fail. Fixes: 0bf02a0d80427f26 ("perf bench: Add build-id injection benchmark") Acked-by: Namhyung Kim Link: http://lore.kernel.org/lkml/CAM9d7cgEZNSor+B+7Y2C+QYGme_v5aH0Zn0RLfxoQ+Fy83EHrg@mail.gmail.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf unwind: Do not overwrite FEATURE_CHECK_LDFLAGS-libunwind-{x86,aarch64}

2021-09-22T10:39:30+00:00

[ Upstream commit cdf32b44678c382a31dc183d9a767306915cda7b ] When setting LIBUNWIND_DIR, we first set FEATURE_CHECK_LDFLAGS-libunwind-{aarch64,x86} = -L$(LIBUNWIND_DIR)/lib. This happens a bit before, the overwritting, in: libunwind_arch_set_flags = $(eval $(libunwind_arch_set_flags_code)) define libunwind_arch_set_flags_code FEATURE_CHECK_CFLAGS-libunwind-$(1) = -I$(LIBUNWIND_DIR)/include FEATURE_CHECK_LDFLAGS-libunwind-$(1) = -L$(LIBUNWIND_DIR)/lib endef ifdef LIBUNWIND_DIR LIBUNWIND_CFLAGS = -I$(LIBUNWIND_DIR)/include LIBUNWIND_LDFLAGS = -L$(LIBUNWIND_DIR)/lib LIBUNWIND_ARCHS = x86 x86_64 arm aarch64 debug-frame-arm debug-frame-aarch64 $(foreach libunwind_arch,$(LIBUNWIND_ARCHS),$(call libunwind_arch_set_flags,$(libunwind_arch))) endif Look at that 'foreach' on all the LIBUNWIND_ARCHS. After commit 5c4d7c82c0dc ("perf unwind: Do not put libunwind-{x86,aarch64} in FEATURE_TESTS_BASIC"), FEATURE_CHECK_LDFLAGS-libunwind-{x86,aarch64} is overwritten. As a result, the remote libunwind libraries cannot be searched from $(LIBUNWIND_DIR)/lib directory during feature check tests. Fix it with variable appending. Before this patch: perf$ make VF=1 LIBUNWIND_DIR=/opt/libunwind_aarch64 BUILD: Doing 'make -j16' parallel build ... ... libopencsd: [ OFF ] ... libunwind-x86: [ OFF ] ... libunwind-x86_64: [ OFF ] ... libunwind-arm: [ OFF ] ... libunwind-aarch64: [ OFF ] ... libunwind-debug-frame: [ OFF ] ... libunwind-debug-frame-arm: [ OFF ] ... libunwind-debug-frame-aarch64: [ OFF ] ... cxx: [ OFF ] perf$ cat ../build/feature/test-libunwind-aarch64.make.output /usr/bin/ld: cannot find -lunwind-aarch64 /usr/bin/ld: cannot find -lunwind-aarch64 collect2: error: ld returned 1 exit status After this patch: perf$ make VF=1 LIBUNWIND_DIR=/opt/libunwind_aarch64 BUILD: Doing 'make -j16' parallel build ... libopencsd: [ OFF ] ... libunwind-x86: [ OFF ] ... libunwind-x86_64: [ OFF ] ... libunwind-arm: [ OFF ] ... libunwind-aarch64: [ on ] ... libunwind-debug-frame: [ OFF ] ... libunwind-debug-frame-arm: [ OFF ] ... libunwind-debug-frame-aarch64: [ OFF ] ... cxx: [ OFF ] perf$ cat ../build/feature/test-libunwind-aarch64.make.output perf$ ldd ./perf linux-vdso.so.1 (0x00007ffdf07da000) libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f30953dc000) librt.so.1 => /lib/x86_64-linux-gnu/librt.so.1 (0x00007f30951d4000) libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f3094e36000) libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007f3094c32000) libelf.so.1 => /usr/lib/x86_64-linux-gnu/libelf.so.1 (0x00007f3094a18000) libdw.so.1 => /usr/lib/x86_64-linux-gnu/libdw.so.1 (0x00007f30947cc000) libunwind-x86_64.so.8 => /usr/lib/x86_64-linux-gnu/libunwind-x86_64.so.8 (0x00007f30945ad000) libunwind.so.8 => /usr/lib/x86_64-linux-gnu/libunwind.so.8 (0x00007f3094392000) liblzma.so.5 => /lib/x86_64-linux-gnu/liblzma.so.5 (0x00007f309416c000) libunwind-aarch64.so.8 => not found libslang.so.2 => /lib/x86_64-linux-gnu/libslang.so.2 (0x00007f3093c8a000) libpython2.7.so.1.0 => /usr/local/lib/libpython2.7.so.1.0 (0x00007f309386b000) libz.so.1 => /lib/x86_64-linux-gnu/libz.so.1 (0x00007f309364e000) libnuma.so.1 => /usr/lib/x86_64-linux-gnu/libnuma.so.1 (0x00007f3093443000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f3093052000) /lib64/ld-linux-x86-64.so.2 (0x00007f3096097000) libbz2.so.1.0 => /lib/x86_64-linux-gnu/libbz2.so.1.0 (0x00007f3092e42000) libutil.so.1 => /lib/x86_64-linux-gnu/libutil.so.1 (0x00007f3092c3f000) Fixes: 5c4d7c82c0dceccf ("perf unwind: Do not put libunwind-{x86,aarch64} in FEATURE_TESTS_BASIC") Signed-off-by: Li Huafei Cc: Alexander Shishkin Cc: He Kuang Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Zhang Jinhao Link: http://lore.kernel.org/lkml/20210823134340.60955-1-lihuafei1@huawei.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf config: Fix caching and memory leak in perf_home_perfconfig()

2021-09-22T10:39:30+00:00

[ Upstream commit 261f491133aecb943ccfdc3b3794e2d775607a87 ] Acaict, perf_home_perfconfig() is supposed to cache the result of home_perfconfig, which returns the default location of perfconfig for the user, given the HOME environment variable. However, the current implementation calls home_perfconfig every time perf_home_perfconfig() is called (so no caching is actually performed), replacing the previous pointer, thus also causing a memory leak. This patch adds a check of whether either config or failed is set and, in that case, directly returns config without calling home_perfconfig at each invocation. Fixes: f5f03e19ce14fc31 ("perf config: Add perf_home_perfconfig function") Signed-off-by: Riccardo Mancini Cc: Alexander Shishkin Cc: Ian Rogers Cc: Jin Yao Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Song Liu Link: http://lore.kernel.org/lkml/20210820130817.740536-1-rickyman7@gmail.com [ Removed needless double check for the 'failed' variable ] Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf machine: Initialize srcline string member in add_location struct

2021-09-22T10:39:20+00:00

commit 57f0ff059e3daa4e70a811cb1d31a49968262d20 upstream. It's later supposed to be either a correct address or NULL. Without the initialization, it may contain an undefined value which results in the following segmentation fault: # perf top --sort comm -g --ignore-callees=do_idle terminates with: #0 0x00007ffff56b7685 in __strlen_avx2 () from /lib64/libc.so.6 #1 0x00007ffff55e3802 in strdup () from /lib64/libc.so.6 #2 0x00005555558cb139 in hist_entry__init (callchain_size=, sample_self=true, template=0x7fffde7fb110, he=0x7fffd801c250) at util/hist.c:489 #3 hist_entry__new (template=template@entry=0x7fffde7fb110, sample_self=sample_self@entry=true) at util/hist.c:564 #4 0x00005555558cb4ba in hists__findnew_entry (hists=hists@entry=0x5555561d9e38, entry=entry@entry=0x7fffde7fb110, al=al@entry=0x7fffde7fb420, sample_self=sample_self@entry=true) at util/hist.c:657 #5 0x00005555558cba1b in __hists__add_entry (hists=hists@entry=0x5555561d9e38, al=0x7fffde7fb420, sym_parent=, bi=bi@entry=0x0, mi=mi@entry=0x0, sample=sample@entry=0x7fffde7fb4b0, sample_self=true, ops=0x0, block_info=0x0) at util/hist.c:288 #6 0x00005555558cbb70 in hists__add_entry (sample_self=true, sample=0x7fffde7fb4b0, mi=0x0, bi=0x0, sym_parent=, al=, hists=0x5555561d9e38) at util/hist.c:1056 #7 iter_add_single_cumulative_entry (iter=0x7fffde7fb460, al=) at util/hist.c:1056 #8 0x00005555558cc8a4 in hist_entry_iter__add (iter=iter@entry=0x7fffde7fb460, al=al@entry=0x7fffde7fb420, max_stack_depth=, arg=arg@entry=0x7fffffff7db0) at util/hist.c:1231 #9 0x00005555557cdc9a in perf_event__process_sample (machine=, sample=0x7fffde7fb4b0, evsel=, event=, tool=0x7fffffff7db0) at builtin-top.c:842 #10 deliver_event (qe=, qevent=) at builtin-top.c:1202 #11 0x00005555558a9318 in do_flush (show_progress=false, oe=0x7fffffff80e0) at util/ordered-events.c:244 #12 __ordered_events__flush (oe=oe@entry=0x7fffffff80e0, how=how@entry=OE_FLUSH__TOP, timestamp=timestamp@entry=0) at util/ordered-events.c:323 #13 0x00005555558a9789 in __ordered_events__flush (timestamp=, how=, oe=) at util/ordered-events.c:339 #14 ordered_events__flush (how=OE_FLUSH__TOP, oe=0x7fffffff80e0) at util/ordered-events.c:341 #15 ordered_events__flush (oe=oe@entry=0x7fffffff80e0, how=how@entry=OE_FLUSH__TOP) at util/ordered-events.c:339 #16 0x00005555557cd631 in process_thread (arg=0x7fffffff7db0) at builtin-top.c:1114 #17 0x00007ffff7bb817a in start_thread () from /lib64/libpthread.so.0 #18 0x00007ffff5656dc3 in clone () from /lib64/libc.so.6 If you look at the frame #2, the code is: 488 if (he->srcline) { 489 he->srcline = strdup(he->srcline); 490 if (he->srcline == NULL) 491 goto err_rawdata; 492 } If he->srcline is not NULL (it is not NULL if it is uninitialized rubbish), it gets strdupped and strdupping a rubbish random string causes the problem. Also, if you look at the commit 1fb7d06a509e, it adds the srcline property into the struct, but not initializing it everywhere needed. Committer notes: Now I see, when using --ignore-callees=do_idle we end up here at line 2189 in add_callchain_ip(): 2181 if (al.sym != NULL) { 2182 if (perf_hpp_list.parent && !*parent && 2183 symbol__match_regex(al.sym, &parent_regex)) 2184 *parent = al.sym; 2185 else if (have_ignore_callees && root_al && 2186 symbol__match_regex(al.sym, &ignore_callees_regex)) { 2187 /* Treat this symbol as the root, 2188 forgetting its callees. */ 2189 *root_al = al; 2190 callchain_cursor_reset(cursor); 2191 } 2192 } And the al that doesn't have the ->srcline field initialized will be copied to the root_al, so then, back to: 1211 int hist_entry_iter__add(struct hist_entry_iter *iter, struct addr_location *al, 1212 int max_stack_depth, void *arg) 1213 { 1214 int err, err2; 1215 struct map *alm = NULL; 1216 1217 if (al) 1218 alm = map__get(al->map); 1219 1220 err = sample__resolve_callchain(iter->sample, &callchain_cursor, &iter->parent, 1221 iter->evsel, al, max_stack_depth); 1222 if (err) { 1223 map__put(alm); 1224 return err; 1225 } 1226 1227 err = iter->ops->prepare_entry(iter, al); 1228 if (err) 1229 goto out; 1230 1231 err = iter->ops->add_single_entry(iter, al); 1232 if (err) 1233 goto out; 1234 That al at line 1221 is what hist_entry_iter__add() (called from sample__resolve_callchain()) saw as 'root_al', and then: iter->ops->add_single_entry(iter, al); will go on with al->srcline with a bogus value, I'll add the above sequence to the cset and apply, thanks! Signed-off-by: Michael Petlan CC: Milian Wolff Cc: Jiri Olsa Fixes: 1fb7d06a509e ("perf report Use srcline from callchain for hist entries") Link: https //lore.kernel.org/r/20210719145332.29747-1-mpetlan@redhat.com Reported-by: Juri Lelli Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Greg Kroah-Hartman