Successfully identified regression in *gcc* in CI configuration tcwg_bmk_gnu_apm/gnu-release-aarch64-spec2k6-Os_LTO. So far, this commit has regressed CI configurations: - tcwg_bmk_gnu_apm/gnu-release-aarch64-spec2k6-Os_LTO

Culprit: <cut> commit ee875b63b22e30a0dcb4b05f7532c2c416ba6cd0 Author: Richard Biener rguenther@suse.de Date: Tue Aug 17 08:38:35 2021 +0200

tree-optimization/101868 - avoid PRE of trapping mems across calls

This backports a fix for the omission of a check of trapping mems when hoisting them across calls that might not return. This was originally done as part of a fix to handle const functions that throw properly.

2021-08-17 Richard Biener rguenther@suse.de

PR tree-optimization/101373 PR tree-optimization/101868 * tree-ssa-pre.c (prune_clobbered_mems): Also prune trapping references when the BB may not return.

* gcc.dg/lto/pr101868_0.c: New testcase. * gcc.dg/lto/pr101868_1.c: Likewise. * gcc.dg/lto/pr101868_2.c: Likewise. * gcc.dg/lto/pr101868_3.c: Likewise. </cut>

Results regressed to (for first_bad == ee875b63b22e30a0dcb4b05f7532c2c416ba6cd0) # reset_artifacts: -10 # build_abe binutils: -9 # build_abe stage1 -- --set gcc_override_configure=--disable-libsanitizer: -8 # build_abe linux: -7 # build_abe glibc: -6 # build_abe stage2 -- --set gcc_override_configure=--disable-libsanitizer: -5 # true: 0 # benchmark -- -Os_LTO artifacts/build-ee875b63b22e30a0dcb4b05f7532c2c416ba6cd0/results_id: 1 # 450.soplex,soplex_base.default regressed by 102

from (for last_good == a0a0499b8bb920fdd98e791804812f001f0b4fe8) # reset_artifacts: -10 # build_abe binutils: -9 # build_abe stage1 -- --set gcc_override_configure=--disable-libsanitizer: -8 # build_abe linux: -7 # build_abe glibc: -6 # build_abe stage2 -- --set gcc_override_configure=--disable-libsanitizer: -5 # true: 0 # benchmark -- -Os_LTO artifacts/build-a0a0499b8bb920fdd98e791804812f001f0b4fe8/results_id: 1

Artifacts of last_good build: https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa... Results ID of last_good: apm_64/tcwg_bmk_gnu_apm/bisect-gnu-release-aarch64-spec2k6-Os_LTO/4497 Artifacts of first_bad build: https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa... Results ID of first_bad: apm_64/tcwg_bmk_gnu_apm/bisect-gnu-release-aarch64-spec2k6-Os_LTO/4482 Build top page/logs: https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa...

Configuration details:

Reproduce builds: <cut> mkdir investigate-gcc-ee875b63b22e30a0dcb4b05f7532c2c416ba6cd0 cd investigate-gcc-ee875b63b22e30a0dcb4b05f7532c2c416ba6cd0

git clone https://git.linaro.org/toolchain/jenkins-scripts

mkdir -p artifacts/manifests curl -o artifacts/manifests/build-baseline.sh https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa... --fail curl -o artifacts/manifests/build-parameters.sh https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa... --fail curl -o artifacts/test.sh https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa... --fail chmod +x artifacts/test.sh

# Reproduce the baseline build (build all pre-requisites) ./jenkins-scripts/tcwg_bmk-build.sh @@ artifacts/manifests/build-baseline.sh

# Save baseline build state (which is then restored in artifacts/test.sh) mkdir -p ./bisect rsync -a --del --delete-excluded --exclude /bisect/ --exclude /artifacts/ --exclude /gcc/ ./ ./bisect/baseline/

cd gcc

# Reproduce first_bad build git checkout --detach ee875b63b22e30a0dcb4b05f7532c2c416ba6cd0 ../artifacts/test.sh

# Reproduce last_good build git checkout --detach a0a0499b8bb920fdd98e791804812f001f0b4fe8 ../artifacts/test.sh

cd .. </cut>

History of pending regressions and results: https://git.linaro.org/toolchain/ci/base-artifacts.git/log/?h=linaro-local/c...

Artifacts: https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa... Build log: https://ci.linaro.org/job/tcwg_bmk_ci_gnu-bisect-tcwg_bmk_apm-gnu-release-aa...

Full commit (up to 1000 lines): <cut> commit ee875b63b22e30a0dcb4b05f7532c2c416ba6cd0 Author: Richard Biener rguenther@suse.de Date: Tue Aug 17 08:38:35 2021 +0200

tree-optimization/101868 - avoid PRE of trapping mems across calls

This backports a fix for the omission of a check of trapping mems when hoisting them across calls that might not return. This was originally done as part of a fix to handle const functions that throw properly.

2021-08-17 Richard Biener rguenther@suse.de

PR tree-optimization/101373 PR tree-optimization/101868 * tree-ssa-pre.c (prune_clobbered_mems): Also prune trapping references when the BB may not return.

* gcc.dg/lto/pr101868_0.c: New testcase. * gcc.dg/lto/pr101868_1.c: Likewise. * gcc.dg/lto/pr101868_2.c: Likewise. * gcc.dg/lto/pr101868_3.c: Likewise. --- gcc/testsuite/gcc.dg/lto/pr101868_0.c | 33 +++++++++++++++++++++++++++++++++ gcc/testsuite/gcc.dg/lto/pr101868_1.c | 23 +++++++++++++++++++++++ gcc/testsuite/gcc.dg/lto/pr101868_2.c | 11 +++++++++++ gcc/testsuite/gcc.dg/lto/pr101868_3.c | 8 ++++++++ gcc/tree-ssa-pre.c | 7 +++++++ 5 files changed, 82 insertions(+)

diff --git a/gcc/testsuite/gcc.dg/lto/pr101868_0.c b/gcc/testsuite/gcc.dg/lto/pr101868_0.c new file mode 100644 index 00000000000..c84d19b0267 --- /dev/null +++ b/gcc/testsuite/gcc.dg/lto/pr101868_0.c @@ -0,0 +1,33 @@ +/* { dg-lto-do run } */ +/* { dg-lto-options { "-O2 -fno-strict-aliasing -flto" } } */ + +typedef unsigned long VALUE; + +__attribute__ ((cold)) +void rb_check_type(VALUE, int); + +static VALUE +repro(VALUE dummy, VALUE hash) +{ + if (hash == 0) { + rb_check_type(hash, 1); + } + else if (*(long *)hash) { + rb_check_type(hash, 1); + } + + + return *(long *)hash; +} + +static VALUE (*that)(VALUE dummy, VALUE hash) = repro; + +int +main(int argc, char **argv) +{ + argc--; + that(0, argc); + + rb_check_type(argc, argc); + +} diff --git a/gcc/testsuite/gcc.dg/lto/pr101868_1.c b/gcc/testsuite/gcc.dg/lto/pr101868_1.c new file mode 100644 index 00000000000..146c14abc76 --- /dev/null +++ b/gcc/testsuite/gcc.dg/lto/pr101868_1.c @@ -0,0 +1,23 @@ +typedef unsigned long VALUE; + + +__attribute__ ((noreturn)) void rexc_raise(VALUE mesg); + +VALUE rb_donothing(VALUE klass); + +static void +funexpected_type(VALUE x, int xt, int t) +{ + rexc_raise(rb_donothing(0)); +} + +__attribute__ ((cold)) +void +rb_check_type(VALUE x, int t) +{ + int xt; + + if (x == 0) { + funexpected_type(x, xt, t); + } +} diff --git a/gcc/testsuite/gcc.dg/lto/pr101868_2.c b/gcc/testsuite/gcc.dg/lto/pr101868_2.c new file mode 100644 index 00000000000..e6f01b23f45 --- /dev/null +++ b/gcc/testsuite/gcc.dg/lto/pr101868_2.c @@ -0,0 +1,11 @@ +typedef unsigned long VALUE; + +static void thing(void) {} +static void (*ptr)(void) = &thing; + +VALUE +rb_donothing(VALUE klass) +{ + ptr(); + return 0; +} diff --git a/gcc/testsuite/gcc.dg/lto/pr101868_3.c b/gcc/testsuite/gcc.dg/lto/pr101868_3.c new file mode 100644 index 00000000000..61217625be7 --- /dev/null +++ b/gcc/testsuite/gcc.dg/lto/pr101868_3.c @@ -0,0 +1,8 @@ +typedef unsigned long VALUE; + +__attribute__((noreturn)) +void +rexc_raise(VALUE mesg) +{ + __builtin_exit(0); +} diff --git a/gcc/tree-ssa-pre.c b/gcc/tree-ssa-pre.c index 04ec4fbaeec..2aedc31e1d7 100644 --- a/gcc/tree-ssa-pre.c +++ b/gcc/tree-ssa-pre.c @@ -2070,6 +2070,13 @@ prune_clobbered_mems (bitmap_set_t set, basic_block block) && value_dies_in_block_x (expr, block)))) to_remove = i; } + /* If the REFERENCE may trap make sure the block does not contain + a possible exit point. + ??? This is overly conservative if we translate AVAIL_OUT + as the available expression might be after the exit point. */ + if (BB_MAY_NOTRETURN (block) + && vn_reference_may_trap (ref)) + to_remove = i; } else if (expr->kind == NARY) { </cut>