i386: Improve expansion of __builtin_parity

GCC currently hides the shift and xor reduction inside a backend
specific UNSPEC PARITY, making it invisible to the RTL optimizers until
very late during compilation.  It is normally reasonable for the
middle-end to maintain wider mode representations for as long as possible
and split them later, but this only helps if the semantics are visible
at the RTL-level (to combine and other passes), but UNSPECs are black
boxes, so in this case splitting early (during RTL expansion) is a
better strategy.

It turns out that that popcount instruction on modern x86_64 processors
has (almost) made the integer parity flag in the x86 ALU completely
obsolete, especially as POPCOUNT's integer semantics are a much better
fit to RTL.  The one remaining case where these transistors are useful
is where __builtin_parity is immediately tested by a conditional branch,
and therefore the result is wanted in a flags register rather than as
an integer.  This case is captured by two peephole2 optimizations in
the attached patch.

2020-06-07  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog:

	* config/i386/i386.md (paritydi2, paritysi2): Expand reduction
	via shift and xor to an USPEC PARITY matching a parityhi2_cmp.
	(paritydi2_cmp, paritysi2_cmp): Delete these define_insn_and_split.
	(parityhi2, parityqi2): New expanders.
	(parityhi2_cmp): Implement set parity flag with xorb insn.
	(parityqi2_cmp): Implement set parity flag with testb insn.
	New peephole2s to use these insns (UNSPEC PARITY) when appropriate.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/parity-3.c: New test.
	* gcc.target/i386/parity-4.c: Likewise.
	* gcc.target/i386/parity-5.c: Likewise.
	* gcc.target/i386/parity-6.c: Likewise.
	* gcc.target/i386/parity-7.c: Likewise.
	* gcc.target/i386/parity-8.c: Likewise.
	* gcc.target/i386/parity-9.c: Likewise.

0 files changed, 0 insertions, 0 deletions