Skip to content

Commit b4fc09e

Browse files
authored
Add registers d8 to d15 to clobber lists as the code does not expressly save them
1 parent 8e50b8d commit b4fc09e

1 file changed

Lines changed: 2 additions & 2 deletions

File tree

kernel/arm64/sgemm_direct_arm64_sme1.c

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -58,7 +58,7 @@ void CNAME (BLASLONG M, BLASLONG N, BLASLONG K, float * __restrict A,\
5858
* of reading directly from vector (z) registers.
5959
* */
6060
asm volatile("" : : :"p0", "p1", "p2", "p3", "p4", "p5", "p6", "p7",
61-
"p8", "p9", "p10", "p11", "p12", "p13", "p14", "p15",
61+
"p8", "p9", "p10", "p11", "p12", "p13", "p14", "p15", "d8", "d9", "d10", "d11", "d12", "d13", "d14", "d15",
6262
"z0", "z1", "z2", "z3", "z4", "z5", "z6", "z7",
6363
"z8", "z9", "z10", "z11", "z12", "z13", "z14", "z15",
6464
"z16", "z17", "z18", "z19", "z20", "z21", "z22", "z23",
@@ -74,7 +74,7 @@ fprintf(stderr,"sme direct calling 2x2\n");
7474
SME1_DIRECT2X2(M, K, N, A_mod, B, R);
7575

7676
asm volatile("" : : :"p0", "p1", "p2", "p3", "p4", "p5", "p6", "p7",
77-
"p8", "p9", "p10", "p11", "p12", "p13", "p14", "p15",
77+
"p8", "p9", "p10", "p11", "p12", "p13", "p14", "p15", "d8", "d9", "d10", "d11", "d12", "d13", "d14", "d15",
7878
"z0", "z1", "z2", "z3", "z4", "z5", "z6", "z7",
7979
"z8", "z9", "z10", "z11", "z12", "z13", "z14", "z15",
8080
"z16", "z17", "z18", "z19", "z20", "z21", "z22", "z23",

0 commit comments

Comments
 (0)