]> www.infradead.org Git - users/dwmw2/qemu.git/commit
projects / users / dwmw2 / qemu.git / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 1f7f72b) | patch
target/i386: optimize computation of ZF from CC_OP_DYNAMIC
authorPaolo Bonzini <pbonzini@redhat.com>
Thu, 20 Jun 2024 08:34:28 +0000 (10:34 +0200)
committerPaolo Bonzini <pbonzini@redhat.com>
Thu, 31 Oct 2024 17:28:33 +0000 (18:28 +0100)
commitae14b33de8d329d5497db5446bdc0b0cb6ba756b
treeee8f830177d04ee80dbbe80b2598669ff501887ctree
parent1f7f72bdc4faece6af875503a3abe992e87e776bcommit | diff
target/i386: optimize computation of ZF from CC_OP_DYNAMIC

Most uses of CC_OP_DYNAMIC are for CMP/JB/JE or similar sequences.
We can optimize many of them to avoid computation of the flags.
This eliminates both TCG ops to set up the new cc_op, and helper
instructions because evaluating just ZF is much cheaper.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
target/i386/helper.h diff | blob | history
target/i386/tcg/cc_helper.c diff | blob | history
target/i386/tcg/translate.c diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.