| Message ID | 20230621091443.1187290-1-richard.henderson@linaro.org |
|---|---|
| State | New |
| Headers | show |
| Series | [v2] tests/plugin: Remove duplicate insn log from libinsn.so | expand |
Richard Henderson <richard.henderson@linaro.org> writes: > This is a perfectly natural occurrence for x86 "rep movb", > where the "rep" prefix forms a counted loop of the one insn. > > During the tests/tcg/multiarch/memory test, this logging is > triggered over 350000 times. Within the context of cross-i386-tci > build, which is already slow by nature, the logging is sufficient > to push the test into timeout. > > Signed-off-by: Richard Henderson <richard.henderson@linaro.org> > --- > tests/tcg/i386/Makefile.softmmu-target | 9 --------- > tests/tcg/i386/Makefile.target | 6 ------ > tests/tcg/x86_64/Makefile.softmmu-target | 9 --------- > 3 files changed, 24 deletions(-) I suspect this wasn't squashed into the changes to insn.c itself.
diff --git a/tests/tcg/i386/Makefile.softmmu-target b/tests/tcg/i386/Makefile.softmmu-target index ed922d59c8..5266f2335a 100644 --- a/tests/tcg/i386/Makefile.softmmu-target +++ b/tests/tcg/i386/Makefile.softmmu-target @@ -33,14 +33,5 @@ EXTRA_RUNS+=$(MULTIARCH_RUNS) memory: CFLAGS+=-DCHECK_UNALIGNED=1 -# non-inline runs will trigger the duplicate instruction heuristics in libinsn.so -run-plugin-%-with-libinsn.so: - $(call run-test, $@, \ - $(QEMU) -monitor none -display none \ - -chardev file$(COMMA)path=$@.out$(COMMA)id=output \ - -plugin ../../plugin/libinsn.so$(COMMA)inline=on \ - -d plugin -D $*-with-libinsn.so.pout \ - $(QEMU_OPTS) $*) - # Running QEMU_OPTS+=-device isa-debugcon,chardev=output -device isa-debug-exit,iobase=0xf4,iosize=0x4 -kernel diff --git a/tests/tcg/i386/Makefile.target b/tests/tcg/i386/Makefile.target index 821822ed0c..f2ee7a4db7 100644 --- a/tests/tcg/i386/Makefile.target +++ b/tests/tcg/i386/Makefile.target @@ -63,12 +63,6 @@ else SKIP_I386_TESTS+=test-i386-fprem endif -# non-inline runs will trigger the duplicate instruction heuristics in libinsn.so -run-plugin-%-with-libinsn.so: - $(call run-test, $@, $(QEMU) $(QEMU_OPTS) \ - -plugin ../../plugin/libinsn.so$(COMMA)inline=on \ - -d plugin -D $*-with-libinsn.so.pout $*) - # Update TESTS I386_TESTS:=$(filter-out $(SKIP_I386_TESTS), $(ALL_X86_TESTS)) TESTS=$(MULTIARCH_TESTS) $(I386_TESTS) diff --git a/tests/tcg/x86_64/Makefile.softmmu-target b/tests/tcg/x86_64/Makefile.softmmu-target index 7207fee94c..1bd763f2e6 100644 --- a/tests/tcg/x86_64/Makefile.softmmu-target +++ b/tests/tcg/x86_64/Makefile.softmmu-target @@ -33,14 +33,5 @@ EXTRA_RUNS+=$(MULTIARCH_RUNS) memory: CFLAGS+=-DCHECK_UNALIGNED=1 -# non-inline runs will trigger the duplicate instruction heuristics in libinsn.so -run-plugin-%-with-libinsn.so: - $(call run-test, $@, \ - $(QEMU) -monitor none -display none \ - -chardev file$(COMMA)path=$@.out$(COMMA)id=output \ - -plugin ../../plugin/libinsn.so$(COMMA)inline=on \ - -d plugin -D $*-with-libinsn.so.pout \ - $(QEMU_OPTS) $*) - # Running QEMU_OPTS+=-device isa-debugcon,chardev=output -device isa-debug-exit,iobase=0xf4,iosize=0x4 -kernel
This is a perfectly natural occurrence for x86 "rep movb", where the "rep" prefix forms a counted loop of the one insn. During the tests/tcg/multiarch/memory test, this logging is triggered over 350000 times. Within the context of cross-i386-tci build, which is already slow by nature, the logging is sufficient to push the test into timeout. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- tests/tcg/i386/Makefile.softmmu-target | 9 --------- tests/tcg/i386/Makefile.target | 6 ------ tests/tcg/x86_64/Makefile.softmmu-target | 9 --------- 3 files changed, 24 deletions(-)