summaryrefslogtreecommitdiffstats
path: root/tools/build
diff options
context:
space:
mode:
authorSteinar H. Gunderson <sesse@google.com>2024-08-03 17:20:06 +0200
committerArnaldo Carvalho de Melo <acme@redhat.com>2024-09-03 10:15:16 -0300
commitc3f8644c21df9b7db97eb70e08e2826368aaafa0 (patch)
treef4a5801e482dcdfa6bee676cecddb2a5bf8e3283 /tools/build
parent0e7eb23668948585f3f0ea8c6249338f33fde872 (diff)
downloadlinux-stable-c3f8644c21df9b7db97eb70e08e2826368aaafa0.tar.gz
linux-stable-c3f8644c21df9b7db97eb70e08e2826368aaafa0.tar.bz2
linux-stable-c3f8644c21df9b7db97eb70e08e2826368aaafa0.zip
perf report: Support LLVM for addr2line()
In addition to the existing support for libbfd and calling out to an external addr2line command, add support for using libllvm directly. This is both faster than libbfd, and can be enabled in distro builds (the LLVM license has an explicit provision for GPLv2 compatibility). Thus, it is set as the primary choice if available. As an example, running 'perf report' on a medium-size profile with DWARF-based backtraces took 58 seconds with LLVM, 78 seconds with libbfd, 153 seconds with external llvm-addr2line, and I got tired and aborted the test after waiting for 55 minutes with external bfd addr2line (which is the default for perf as compiled by distributions today). Evidently, for this case, the bfd addr2line process needs 18 seconds (on a 5.2 GHz Zen 3) to load the .debug ELF in question, hits the 1-second timeout and gets killed during initialization, getting restarted anew every time. Having an in-process addr2line makes this much more robust. As future extensions, libllvm can be used in many other places where we currently use libbfd or other libraries: - Symbol enumeration (in particular, for PE binaries). - Demangling (including non-Itanium demangling, e.g. Microsoft or Rust). - Disassembling (perf annotate). However, these are much less pressing; most people don't profile PE binaries, and perf has non-bfd paths for ELF. The same with demangling; the default _cxa_demangle path works fine for most users, and while bfd objdump can be slow on large binaries, it is possible to use --objdump=llvm-objdump to get the speed benefits. (It appears LLVM-based demangling is very simple, should we want that.) Tested with LLVM 14, 15, 16, 18 and 19. For some reason, LLVM 12 was not correctly detected using feature_check, and thus was not tested. Committer notes: Added the name and a __maybe_unused to address: 1 13.50 almalinux:8 : FAIL gcc version 8.5.0 20210514 (Red Hat 8.5.0-22) (GCC) util/srcline.c: In function 'dso__free_a2l': util/srcline.c:184:20: error: parameter name omitted void dso__free_a2l(struct dso *) ^~~~~~~~~~~~ make[3]: *** [/git/perf-6.11.0-rc3/tools/build/Makefile.build:158: util] Error 2 Signed-off-by: Steinar H. Gunderson <sesse@google.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Ian Rogers <irogers@google.com> Link: https://lore.kernel.org/r/20240803152008.2818485-1-sesse@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Diffstat (limited to 'tools/build')
-rw-r--r--tools/build/Makefile.feature1
1 files changed, 1 insertions, 0 deletions
diff --git a/tools/build/Makefile.feature b/tools/build/Makefile.feature
index e1900abd44f6..0717e96d6a0e 100644
--- a/tools/build/Makefile.feature
+++ b/tools/build/Makefile.feature
@@ -136,6 +136,7 @@ FEATURE_DISPLAY ?= \
libunwind \
libdw-dwarf-unwind \
libcapstone \
+ llvm \
zlib \
lzma \
get_cpuid \