John Reiser ed0665407a More: prefer movzbl over movb fetch to register ...

Avoids hardware failure of writing 0x00 instead of the fetched byte.
1 cycle faster, 1 byte larger (but often recouped later when OR of 32-bits).
Avoid changing win*pe.S for now.

	modified:   ../misc/testsuite/upx_testsuite_1-expected_sha256sums.sh
	modified:   stub/src/amd64-darwin.dylib-entry.S
	modified:   stub/src/amd64-darwin.macho-entry.S
	modified:   stub/src/amd64-expand.S
	modified:   stub/src/amd64-linux.shlib-init.S
	modified:   stub/src/arch/amd64/nrv2b_d.S
	modified:   stub/src/arch/amd64/nrv2d_d.S
	modified:   stub/src/arch/amd64/nrv2e_d.S
	modified:   stub/src/arch/i386/nrv2b_d32-easy.S
	modified:   stub/src/i386-darwin.macho-entry.S
           plus generated *.h *.map *.dump

2024-08-25 15:16:29 -07:00

..

analyze/clang-tidy

all: update for clang-tidy-18

2024-05-06 05:40:22 +02:00

cmake

CI updates

2024-08-13 11:32:29 +02:00

make

CI and cmake updates

2024-08-07 13:37:38 +02:00

podman/rebuild-stubs

CI updates

2024-08-19 07:55:57 +02:00

scripts

CI updates

2024-05-05 09:35:35 +02:00

testsuite

More: prefer movzbl over movb fetch to register

2024-08-25 15:16:29 -07:00

valgrind

all: pacify clang-tidy, CI and cmake updates

2024-02-14 15:14:47 +01:00