Skip to contentSkip to conversation

tcg/i386: Check for shorter instruction sequence for ARITH_AND

PATCH In review Helge Deller 2 participants 3 messages Aug 7, 2023

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 tcg/i386: Check for shorter instruction sequence for ARITH_AND

The tcg uses tgen_arithi(ARITH_AND) during fast CPU TLB lookups, which e.g. translates to:

0x7ff5b011556a: 48 81 e6 00 f0 ff ff andq $0xfffffffffffff000, %rsi

In case the upper 48 bits are all set, the shorter sequence to operate on the lower 16 bits of the target reg (si) can be used, which will then be a 2 bytes shorter instruction sequence:

0x7f4488097b31: 66 81 e6 00 f0 andw $0xf000, %si

---
 tcg/i386/tcg-target.c.inc | 7 +++++++
 1 file changed, 7 insertions(+)
tcg/i386/tcg-target.c.inc +7 -0
index 77482da070..1cb9759c9e 100644
@@ -1342,6 +1342,13 @@ static void tgen_arithi(TCGContext *s, int c, int r0,
                 /* AND with no high bits set can use a 32-bit operation.  */
                 rexw = 0;
             }
+            if ((val & 0xffffffffffff0000) == 0xffffffffffff0000) {
+                /* mask lower 16 bits on 16-bit register */
+                tcg_out8(s, 0x66);
+                tcg_out_modrm(s, OPC_ARITH_EvIz, c, r0);
+                tcg_out16(s, val);
+                return;
+            }
         }
         if (val == 0xffu && (r0 < 4 || TCG_TARGET_REG_BITS == 64)) {
             tcg_out_ext8u(s, r0, r0);

--
2.41.0

#1 lore raw
Quoted 5 lines · Helge Deller

The tcg uses tgen_arithi(ARITH_AND) during fast CPU TLB lookups, which e.g. translates to:

0x7ff5b011556a: 48 81 e6 00 f0 ff ff andq $0xfffffffffffff000, %rsi

Quoted 7 lines

In case the upper 48 bits are all set, the shorter sequence to operate on the lower 16 bits of the target reg (si) can be used, which will then be a 2 bytes shorter instruction sequence:

0x7f4488097b31: 66 81 e6 00 f0 andw $0xf000, %si

Current Intel optimization guidelines

https://www.intel.com/content/www/us/en/content-details/671488/intel-64-and-ia-32-architectures-optimization-reference-manual.html

Section 3.4.2.3, Length Changing Prefixes, suggests that using 16-byte operands slows decode from 1 cycle to 6 cycles.

Section 3.5.2.3, Partial Register Stalls, says that Skylake has fixed the major issues that older microarchitectures had with such stalls, but that these operations have two additional cycles of delay.

So on balance I don't think this is a good tradeoff.

r~

#2 lore raw
Quoted 17 lines · Richard Henderson
Quoted 5 lines · Helge Deller

The tcg uses tgen_arithi(ARITH_AND) during fast CPU TLB lookups, which e.g. translates to:

0x7ff5b011556a:  48 81 e6 00 f0 ff ff     andq     $0xfffffffffffff000, %rsi

Quoted 7 lines

In case the upper 48 bits are all set, the shorter sequence to operate on the lower 16 bits of the target reg (si) can be used, which will then be a 2 bytes shorter instruction sequence:

0x7f4488097b31:  66 81 e6 00 f0           andw     $0xf000, %si

Current Intel optimization guidelines

Quoted 7 lines

https://www.intel.com/content/www/us/en/content-details/671488/intel-64-and-ia-32-architectures-optimization-reference-manual.html

Section 3.4.2.3, Length Changing Prefixes, suggests that using 16-byte operands slows decode from 1 cycle to 6 cycles.

Section 3.5.2.3, Partial Register Stalls, says that Skylake has fixed the major issues that older microarchitectures had with such stalls, but that these operations have two additional cycles of delay.

So on balance I don't think this is a good tradeoff.

Ok. Thanks for the links!

Helge