[PATCH 00/10] x86: operand size handling improvements

[Jan Beulich <jbeulich at suse dot com>]

Not the least in order to make sure the main change here is fine
for both Linux and gcc, but also because of the need / desire to
have sufficient testsuite coverage (which by itself has pointed
out further issues addressed by this series), it has taken me
several years to finally get this set of changes ready. The main
goal is better consistency in the handling of insn operands, i.e.
in particular less unexpected behavior when deducing how things
would behave from observations with one (set of) insn(s) or
operand(s) towards other constructs.

01: x86: drop stray FloatMF
02: x86: drop stray W
03: x86: produce suffix in suffix-always mode for {,V}CVT{,T}S{S,D}2SI
04: x86-64: generalize SIMD test expectations
05: x86: add tests to cover defaulting of operand sizes for ambiguous insns
06: x86: improve handling of insns with ambiguous operand sizes
07: x86: replace adhoc ambiguous operand checking for CRC32
08: x86: move certain MOVSX/MOVZX tests
09: x86: replace adhoc ambiguous operand checking for MOVSX/MOVZX
10: x86: correct VFPCLASSP{S,D} operand size handling

On top of this I'll also send additions to the newly added test
cases that with the series applied still don't work. The question
there is whether (and if so how) we'd want operand size
defaulting to work there.

Because of an as of yet unresolved email issue I'm attaching all
of the patches (in a tarball due to their total size) right here, in
addition to the individual mails that are going to be sent in reply.

Jan

Attachment: binutils.tb2
Description: Binary data