Summary: | sys-devel/gcc: huge memory usage when building www-client/chromium-113.0.5672.63 (evaluate_prg_hwy.cc) on arm64 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Leonid Kopylov <leonchik1976> |
Component: | Current packages | Assignee: | Gentoo Toolchain Maintainers <toolchain> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | chromium |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | ARM64 | ||
OS: | Linux | ||
See Also: |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109794 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109505 |
||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
build.log.xz
build.log.xz evaluate_prg_hwy.ii.xz evaluate_prg_hwy.ii.xz |
Description
Leonid Kopylov
2023-05-06 12:17:40 UTC
Created attachment 861217 [details]
build.log.xz
{standard input}: Assembler messages: {standard input}:4191: Warning: end of file not at end of a line; newline inserted {standard input}:4231: Error: unknown mnemonic `as' -- `as' g++-12: fatal error: Killed signal terminated program cc1plus compilation terminated. Can you verify you didn't OOM? (In reply to Sam James from comment #2) > {standard input}: Assembler messages: > {standard input}:4191: Warning: end of file not at end of a line; newline > inserted > {standard input}:4231: Error: unknown mnemonic `as' -- `as' > g++-12: fatal error: Killed signal terminated program cc1plus > compilation terminated. > > Can you verify you didn't OOM? 22 GB of ram + 8 GB of swap - shouldn't it be enough? i can try add more swap, but 30GB of free ram should be more than enough for this, no? (In reply to Sam James from comment #2) > {standard input}: Assembler messages: > {standard input}:4191: Warning: end of file not at end of a line; newline > inserted > {standard input}:4231: Error: unknown mnemonic `as' -- `as' > g++-12: fatal error: Killed signal terminated program cc1plus > compilation terminated. > > Can you verify you didn't OOM? * MemTotal: 32817680 kB * SwapTotal: 29360120 kB i increases swap to 30 GB - it didn't helped - so it can't be OOM # free -m total used free shared buff/cache available Mem: 32048 4233 23159 4182 4655 23266 Swap: 28671 6931 21740 Did dmesg include any OOM messages? (In reply to Sam James from comment #6) > Did dmesg include any OOM messages? yes, it's OOO, but at moment of failure: # free -m total used free shared buff/cache available Mem: 32048 26892 526 4214 4629 617 Swap: 44031 43164 867 32GB RAM + 40G of swap - and it's not enough - something very wrong with it. after failure: # free -m total used free shared buff/cache available Mem: 32048 4245 23098 4196 4705 23213 Swap: 44031 7279 36752 so RAM can't be issue. something cause to "eat" all memory - and each time - at exactly same step Created attachment 861396 [details]
build.log.xz
Could be a compiler bug. We had something similar with qtwebengine but that was gcc-10 only. I'll give you some steps in a minute. (In reply to Sam James from comment #9) > Could be a compiler bug. We had something similar with qtwebengine but that > was gcc-10 only. I'll give you some steps in a minute. for now chromium set to gcc-12 Could you try running: /usr/bin/g++-12 -MMD -MF obj/third_party/distributed_point_functions/distributed_point_functions/evaluate_prg_hwy.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_OZONE=1 -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNO_UNWIND_TABLES -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_56 -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_56 -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DGOOGLE_PROTOBUF_INTERNAL_DONATE_STEAL_INLINE=0 -DHAVE_PTHREAD -I../.. -Igen -I../../third_party/distributed_point_functions -I../../third_party/distributed_point_functions/code -Igen/third_party/distributed_point_functions -Igen/shim_headers/zlib_shim -I../../third_party/perfetto/include -Igen/third_party/perfetto/build_config -Igen/third_party/perfetto -Igen/shim_headers/icui18n_shim -Igen/shim_headers/icuuc_shim -I../../third_party/protobuf/src -Igen/protoc_out -I../../third_party/abseil-cpp -I../../third_party/highway/src -I../../third_party/boringssl/src/include -fno-ident -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -fno-unwind-tables -fno-asynchronous-unwind-tables -fPIC -pipe -pthread -fno-omit-frame-pointer -fvisibility=hidden -Wno-unused-local-typedefs -Wno-maybe-uninitialized -Wno-deprecated-declarations -Wno-comments -Wno-packed-not-aligned -Wno-missing-field-initializers -Wno-unused-parameter -Wno-psabi -I/usr/include/glib-2.0 -I/usr/lib64/glib-2.0/include -I/usr/lib64/libffi/include -std=gnu++2a -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -Wno-narrowing -Wno-class-memaccess -O2 -pipe -fno-tree-vectorize -c ../../third_party/distributed_point_functions/code/dpf/internal/evaluate_prg_hwy.cc -o obj/third_party/distributed_point_functions/distributed_point_functions/evaluate_prg_hwy.o in '/var/tmp/portage/www-client/chromium-113.0.5672.63/work/chromium-113.0.5672.63/build/release (I think)? If that OOMs, please let me know, and we can go from there. (In reply to Sam James from comment #11) > Could you try running: > > /usr/bin/g++-12 -MMD -MF > obj/third_party/distributed_point_functions/distributed_point_functions/ > evaluate_prg_hwy.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_GLIB=1 -DUSE_OZONE=1 > -DOFFICIAL_BUILD -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE > -D_LARGEFILE64_SOURCE -DNO_UNWIND_TABLES -DNDEBUG -DNVALGRIND > -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DGLIB_VERSION_MAX_ALLOWED=GLIB_VERSION_2_56 > -DGLIB_VERSION_MIN_REQUIRED=GLIB_VERSION_2_56 -DGOOGLE_PROTOBUF_NO_RTTI > -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER > -DGOOGLE_PROTOBUF_INTERNAL_DONATE_STEAL_INLINE=0 -DHAVE_PTHREAD -I../.. > -Igen -I../../third_party/distributed_point_functions > -I../../third_party/distributed_point_functions/code > -Igen/third_party/distributed_point_functions -Igen/shim_headers/zlib_shim > -I../../third_party/perfetto/include -Igen/third_party/perfetto/build_config > -Igen/third_party/perfetto -Igen/shim_headers/icui18n_shim > -Igen/shim_headers/icuuc_shim -I../../third_party/protobuf/src > -Igen/protoc_out -I../../third_party/abseil-cpp > -I../../third_party/highway/src -I../../third_party/boringssl/src/include > -fno-ident -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector > -fno-unwind-tables -fno-asynchronous-unwind-tables -fPIC -pipe -pthread > -fno-omit-frame-pointer -fvisibility=hidden -Wno-unused-local-typedefs > -Wno-maybe-uninitialized -Wno-deprecated-declarations -Wno-comments > -Wno-packed-not-aligned -Wno-missing-field-initializers > -Wno-unused-parameter -Wno-psabi -I/usr/include/glib-2.0 > -I/usr/lib64/glib-2.0/include -I/usr/lib64/libffi/include -std=gnu++2a > -fno-exceptions -fno-rtti -fvisibility-inlines-hidden -Wno-narrowing > -Wno-class-memaccess -O2 -pipe -fno-tree-vectorize -c > ../../third_party/distributed_point_functions/code/dpf/internal/ > evaluate_prg_hwy.cc -o > obj/third_party/distributed_point_functions/distributed_point_functions/ > evaluate_prg_hwy.o > > in > '/var/tmp/portage/www-client/chromium-113.0.5672.63/work/chromium-113.0.5672. > 63/build/release (I think)? > > If that OOMs, please let me know, and we can go from there. in this directory: /var/tmp/portage/www-client/chromium-113.0.5672.63/work/chromium-113.0.5672.63/build i can't find "release" directory sorry, /var/tmp/portage/www-client/chromium-113.0.5672.63/work/chromium-113.0.5672/out/Release. (In reply to Sam James from comment #13) > sorry, > /var/tmp/portage/www-client/chromium-113.0.5672.63/work/chromium-113.0.5672/ > out/Release. {standard input}: Assembler messages: {standard input}:4191: Warning: end of file not at end of a line; newline inserted {standard input}:4231: Error: unknown mnemonic `as' -- `as' g++-12: fatal error: Killed signal terminated program cc1plus compilation terminated Can you run the same command but with -save-temps at the end (it'll still fail), and then look for evaluate_prg_hwy.ii and upload it? Created attachment 861407 [details]
evaluate_prg_hwy.ii.xz
/var/tmp/portage/www-client/chromium-113.0.5672.63/work/chromium-113.0.5672.63/out/Release/obj/third_party/distributed_point_functions/distributed_point_functions/evaluate_prg_hwy.ii location of file So far this is using 12GB to build (which definitely looks like a memory-hog bug). Can you try gcc-13 and do the same thing for me, giving me the .ii? I'd rather report it upstream against 13 if possible. (In reply to Sam James from comment #18) > So far this is using 12GB to build (which definitely looks like a memory-hog > bug). > > Can you try gcc-13 and do the same thing for me, giving me the .ii? I'd > rather report it upstream against 13 if possible. you mean /usr/bin/g++-13 at the beginning? (In reply to Leonid Kopylov from comment #19) > (In reply to Sam James from comment #18) > > So far this is using 12GB to build (which definitely looks like a memory-hog > > bug). > > > > Can you try gcc-13 and do the same thing for me, giving me the .ii? I'd > > rather report it upstream against 13 if possible. > > you mean /usr/bin/g++-13 at the beginning? yes please Created attachment 861408 [details]
evaluate_prg_hwy.ii.xz
thank you, I've reported this upstream to gcc at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109794. I'll let you know if I need anything more too. There's a patch at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109505#c18 (https://gcc.gnu.org/pipermail/gcc-patches/2023-May/619070.html) but I don't want to add it until it's reviewed because it affects all targets. The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=dbf440b4c0367889ea2a682e0627d77cc4df6568 commit dbf440b4c0367889ea2a682e0627d77cc4df6568 Author: Sam James <sam@gentoo.org> AuthorDate: 2023-05-27 02:09:12 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2023-05-27 02:09:12 +0000 sys-devel/gcc: keyword 12.3.1_p20230526 Closes: https://bugs.gentoo.org/905818 Signed-off-by: Sam James <sam@gentoo.org> sys-devel/gcc/gcc-12.3.1_p20230526.ebuild | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=8cba73661e286989eae65f5b1c26dbc2eb899568 commit 8cba73661e286989eae65f5b1c26dbc2eb899568 Author: Sam James <sam@gentoo.org> AuthorDate: 2023-05-28 16:01:43 +0000 Commit: Sam James <sam@gentoo.org> CommitDate: 2023-05-28 16:01:43 +0000 sys-devel/gcc: keyword 13.1.1_p20230527 Closes: https://bugs.gentoo.org/869536 Closes: https://bugs.gentoo.org/905818 Signed-off-by: Sam James <sam@gentoo.org> sys-devel/gcc/gcc-13.1.1_p20230527.ebuild | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) |