FreeChainXenon/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

i386: Fine tune STV register conversion costs for -Os.

Browse source 
The eagle-eyed may have spotted that my recent testcases for DImode shifts
on x86_64 included -mno-stv in the dg-options.  This is because the
Scalar-To-Vector (STV) pass currently transforms these shifts to use
SSE vector operations, producing larger code even with -Os.  The issue
is that the compute_convert_gain currently underestimates the size of
instructions required for interunit moves, which is corrected with the
patch below.

For the simple test case:

unsigned long long shl1(unsigned long long x) { return x << 1; }

without this patch, GCC -m32 -Os -mavx2 currently generates:

shl1:	push   %ebp		 // 1 byte
	mov    %esp,%ebp	 // 2 bytes
	vmovq  0x8(%ebp),%xmm0	 // 5 bytes
	pop    %ebp		 // 1 byte
	vpaddq %xmm0,%xmm0,%xmm0 // 4 bytes
	vmovd  %xmm0,%eax	 // 4 bytes
	vpextrd $0x1,%xmm0,%edx  // 6 bytes
	ret			 // 1 byte  = 24 bytes total

with this patch, we now generate the shorter

shl1:	push   %ebp		// 1 byte
	mov    %esp,%ebp	// 2 bytes
	mov    0x8(%ebp),%eax	// 3 bytes
	mov    0xc(%ebp),%edx	// 3 bytes
	pop    %ebp		// 1 byte
	add    %eax,%eax	// 2 bytes
	adc    %edx,%edx	// 2 bytes
	ret			// 1 byte  = 15 bytes total

Benchmarking using CSiBE, shows that this patch saves 1361 bytes
when compiling with -m32 -Os, and saves 172 bytes when compiling
with -Os.

2023-10-24  Roger Sayle  <roger@nextmovesoftware.com>

gcc/ChangeLog
	* config/i386/i386-features.cc (compute_convert_gain): Provide
	more accurate values (sizes) for inter-unit moves with -Os.
This commit is contained in:
Roger Sayle 2023-10-24 16:45:08 +01:00
parent 35f4e95265
commit 99a6c1065d
1 changed files with 27 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

32
gcc/config/i386/i386-features.cc
View file

@ -752,11 +752,33 @@ general_scalar_chain::compute_convert_gain ()
fprintf (dump_file, " Instruction conversion gain: %d\n", gain);
/* Cost the integer to sse and sse to integer moves. */
cost += n_sse_to_integer * ix86_cost->sse_to_integer;
/* ??? integer_to_sse but we only have that in the RA cost table.
Assume sse_to_integer/integer_to_sse are the same which they
are at the moment. */
cost += n_integer_to_sse * ix86_cost->sse_to_integer;
if (!optimize_function_for_size_p (cfun))
{
cost += n_sse_to_integer * ix86_cost->sse_to_integer;
/* ??? integer_to_sse but we only have that in the RA cost table.
Assume sse_to_integer/integer_to_sse are the same which they
are at the moment. */
cost += n_integer_to_sse * ix86_cost->sse_to_integer;
}
else if (TARGET_64BIT || smode == SImode)
{
cost += n_sse_to_integer * COSTS_N_BYTES (4);
cost += n_integer_to_sse * COSTS_N_BYTES (4);
}
else if (TARGET_SSE4_1)
{
/* vmovd (4 bytes) + vpextrd (6 bytes). */
cost += n_sse_to_integer * COSTS_N_BYTES (10);
/* vmovd (4 bytes) + vpinsrd (6 bytes). */
cost += n_integer_to_sse * COSTS_N_BYTES (10);
}
else
{
/* movd (4 bytes) + psrlq (5 bytes) + movd (4 bytes). */
cost += n_sse_to_integer * COSTS_N_BYTES (13);
/* movd (4 bytes) + movd (4 bytes) + unpckldq (4 bytes). */
cost += n_integer_to_sse * COSTS_N_BYTES (12);
}
if (dump_file)
fprintf (dump_file, " Registers conversion cost: %d\n", cost);