aboutsummaryrefslogtreecommitdiff
path: root/gcc/config/arm/neon.md
diff options
context:
space:
mode:
authorDennis Zhang <dennis.zhang@arm.com>2020-02-21 15:36:13 +0000
committerDennis Zhang <dennis.zhang@arm.com>2020-02-21 15:36:13 +0000
commit436016f45694c7236e2e9f9db2adb0b4d9bf6b94 (patch)
tree02de5ac2ba0cf5b55459d05ccf280431add435a5 /gcc/config/arm/neon.md
parentb59506cd8b9f92293fc154c1470691534e29ddcf (diff)
downloadgcc-436016f45694c7236e2e9f9db2adb0b4d9bf6b94.zip
gcc-436016f45694c7236e2e9f9db2adb0b4d9bf6b94.tar.gz
gcc-436016f45694c7236e2e9f9db2adb0b4d9bf6b94.tar.bz2
arm: ACLE I8MM multiply-accumulate
This patch adds intrinsics for matrix multiply-accumulate instructions including vmmlaq_s32, vmmlaq_u32, and vusmmlaq_s32. gcc/ChangeLog: 2020-02-21 Dennis Zhang <dennis.zhang@arm.com> * config/arm/arm_neon.h (vmmlaq_s32, vmmlaq_u32, vusmmlaq_s32): New. * config/arm/arm_neon_builtins.def (smmla, ummla, usmmla): New. * config/arm/iterators.md (MATMUL): New iterator. (sup): Add UNSPEC_MATMUL_S, UNSPEC_MATMUL_U, and UNSPEC_MATMUL_US. (mmla_sfx): New attribute. * config/arm/neon.md (neon_<sup>mmlav16qi): New. * config/arm/unspecs.md (UNSPEC_MATMUL_S, UNSPEC_MATMUL_U): New. (UNSPEC_MATMUL_US): New. gcc/testsuite/ChangeLog: 2020-02-21 Dennis Zhang <dennis.zhang@arm.com> * gcc.target/arm/simd/vmmla_1.c: New test.
Diffstat (limited to 'gcc/config/arm/neon.md')
-rw-r--r--gcc/config/arm/neon.md11
1 files changed, 11 insertions, 0 deletions
diff --git a/gcc/config/arm/neon.md b/gcc/config/arm/neon.md
index 5d085dc..039cd90 100644
--- a/gcc/config/arm/neon.md
+++ b/gcc/config/arm/neon.md
@@ -6585,3 +6585,14 @@ if (BYTES_BIG_ENDIAN)
"vabd.<V_if_elem> %<V_reg>0, %<V_reg>1, %<V_reg>2"
[(set_attr "type" "neon_fp_abd_s<q>")]
)
+
+(define_insn "neon_<sup>mmlav16qi"
+ [(set (match_operand:V4SI 0 "register_operand" "=w")
+ (plus:V4SI
+ (unspec:V4SI [(match_operand:V16QI 2 "register_operand" "w")
+ (match_operand:V16QI 3 "register_operand" "w")] MATMUL)
+ (match_operand:V4SI 1 "register_operand" "0")))]
+ "TARGET_I8MM"
+ "v<sup>mmla.<mmla_sfx>\t%q0, %q2, %q3"
+ [(set_attr "type" "neon_mla_s_q")]
+)