projects / xen.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ca9310b) | patch
x86emul: support AVX512F floating-point conversion insns
authorJan Beulich <jbeulich@suse.com>
Fri, 24 May 2019 08:21:30 +0000 (10:21 +0200)
committerJan Beulich <jbeulich@suse.com>
Fri, 24 May 2019 08:21:30 +0000 (10:21 +0200)
commit4037a62ef93e4becdc8744cf659193328c42bd47
tree58b8e58d48c4c0c0a7fb6787a50959ad34c97008tree | snapshot
parentca9310b24e6205de5387e5982ccd42c35caf89d4commit | diff
x86emul: support AVX512F floating-point conversion insns

VCVTPS2PD, sharing its main opcode with others, needs a "manual"
override of disp8scale.

The simd_size change for twobyte_table[0x5a] is benign to pre-existing
code, but allows decode_disp8scale() to work as is here.

Also correct the comment on an AVX counterpart.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
tools/tests/x86_emulator/evex-disp8.c diff | blob | history
tools/tests/x86_emulator/simd.c diff | blob | history
tools/tests/x86_emulator/simd.h diff | blob | history
tools/tests/x86_emulator/test_x86_emulator.c diff | blob | history
xen/arch/x86/x86_emulate/x86_emulate.c diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.