https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92264
Bug ID: 92264
Summary: [10 Regression] Compile time hog in 521.wrf_r with
-Ofast -march=znver2 -g since r276318
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Keywords: compile-time-hog
Severity: normal
Priority: P3
Component: rtl-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: marxin at gcc dot gnu.org
CC: rsandifo at gcc dot gnu.org
Blocks: 26163
Target Milestone: ---
Host: x86_64-linux-gnu
Since the mentioned revision I see huge compile time hog:
$ timeout 300 gfortran module_configure.fppized.f90 -c -march=znver2
-std=legacy -fconvert=big-endian -fno-openmp -Ofast -march=znver2 -g
It takes at least 10 minutes to compile now. Before the revision it takes ~200s
with checking compiler on a Ryzen machine.
I see in perf top something like:
62.38% f951 [.] find_base_term
14.27% f951 [.] ix86_find_base_term
5.53% [kernel] [k]
acpi_processor_ffh_cstate_enter
2.90% f951 [.] cselib_sp_based_value_p
2.88% libc-2.30.so [.] _int_malloc
2.44% f951 [.] find_base_term
0.87% libc-2.30.so [.]
__memmove_avx_unaligned_erms
0.72% f951 [.] rtx_equal_for_memref_p
0.54% f951 [.] drop_overlapping_mem_locs
Richard, can you please take a look?
Referenced Bugs:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26163
[Bug 26163] [meta-bug] missed optimization in SPEC (2k17, 2k and 2k6 and 95)
