QS8 scalar gemm remove bias variables.
PiperOrigin-RevId: 357379347
diff --git a/src/qs8-gemm/MRxNRc4-scalar.c.in b/src/qs8-gemm/MRxNRc4-scalar.c.in
index f117a01..5126016 100644
--- a/src/qs8-gemm/MRxNRc4-scalar.c.in
+++ b/src/qs8-gemm/MRxNRc4-scalar.c.in
@@ -17,6 +17,7 @@
// another thread. We deem this acceptable because the values that are
// read out of bounds do not affect the result, and the the compiler can't know
// about this undefined behavior.
+
void xnn_qs8_gemm_minmax_ukernel_${MR}x${NR}c4__scalar(
size_t mr,
size_t nc,
@@ -56,13 +57,11 @@
// Loop over groups of ${NR} columns.
do {
- // `vaccMN` is the accumulator at row `M`, column `N`.
- // Initialize accumulators with bias. ${NR} bias values are loaded from the
- // weight matrix, at the start of the group of ${NR} columns.
$for N in range(NR):
- int32_t bias${N} = ((const int32_t*)w)[${N}];
- $for M in range(MR):
- int32_t vacc${M}${N} = bias${N};
+ int32_t vacc0${N} = ((const int32_t*) w)[${N}];
+ $for M in range(1, MR):
+ $for N in range(NR):
+ int32_t vacc${M}${N} = vacc0${N};
w = (const void*) ((uintptr_t) w + ${NR} * sizeof(int32_t));
diff --git a/src/qs8-gemm/gen/12x4c4-minmax-scalar.c b/src/qs8-gemm/gen/12x4c4-minmax-scalar.c
index 5ec577a..e82db80 100644
--- a/src/qs8-gemm/gen/12x4c4-minmax-scalar.c
+++ b/src/qs8-gemm/gen/12x4c4-minmax-scalar.c
@@ -21,6 +21,7 @@
// another thread. We deem this acceptable because the values that are
// read out of bounds do not affect the result, and the the compiler can't know
// about this undefined behavior.
+
void xnn_qs8_gemm_minmax_ukernel_12x4c4__scalar(
size_t mr,
size_t nc,
@@ -108,61 +109,54 @@
// Loop over groups of 4 columns.
do {
- // `vaccMN` is the accumulator at row `M`, column `N`.
- // Initialize accumulators with bias. 4 bias values are loaded from the
- // weight matrix, at the start of the group of 4 columns.
- int32_t bias0 = ((const int32_t*)w)[0];
- int32_t vacc00 = bias0;
- int32_t vacc10 = bias0;
- int32_t vacc20 = bias0;
- int32_t vacc30 = bias0;
- int32_t vacc40 = bias0;
- int32_t vacc50 = bias0;
- int32_t vacc60 = bias0;
- int32_t vacc70 = bias0;
- int32_t vacc80 = bias0;
- int32_t vacc90 = bias0;
- int32_t vacc100 = bias0;
- int32_t vacc110 = bias0;
- int32_t bias1 = ((const int32_t*)w)[1];
- int32_t vacc01 = bias1;
- int32_t vacc11 = bias1;
- int32_t vacc21 = bias1;
- int32_t vacc31 = bias1;
- int32_t vacc41 = bias1;
- int32_t vacc51 = bias1;
- int32_t vacc61 = bias1;
- int32_t vacc71 = bias1;
- int32_t vacc81 = bias1;
- int32_t vacc91 = bias1;
- int32_t vacc101 = bias1;
- int32_t vacc111 = bias1;
- int32_t bias2 = ((const int32_t*)w)[2];
- int32_t vacc02 = bias2;
- int32_t vacc12 = bias2;
- int32_t vacc22 = bias2;
- int32_t vacc32 = bias2;
- int32_t vacc42 = bias2;
- int32_t vacc52 = bias2;
- int32_t vacc62 = bias2;
- int32_t vacc72 = bias2;
- int32_t vacc82 = bias2;
- int32_t vacc92 = bias2;
- int32_t vacc102 = bias2;
- int32_t vacc112 = bias2;
- int32_t bias3 = ((const int32_t*)w)[3];
- int32_t vacc03 = bias3;
- int32_t vacc13 = bias3;
- int32_t vacc23 = bias3;
- int32_t vacc33 = bias3;
- int32_t vacc43 = bias3;
- int32_t vacc53 = bias3;
- int32_t vacc63 = bias3;
- int32_t vacc73 = bias3;
- int32_t vacc83 = bias3;
- int32_t vacc93 = bias3;
- int32_t vacc103 = bias3;
- int32_t vacc113 = bias3;
+ int32_t vacc00 = ((const int32_t*) w)[0];
+ int32_t vacc01 = ((const int32_t*) w)[1];
+ int32_t vacc02 = ((const int32_t*) w)[2];
+ int32_t vacc03 = ((const int32_t*) w)[3];
+ int32_t vacc10 = vacc00;
+ int32_t vacc11 = vacc01;
+ int32_t vacc12 = vacc02;
+ int32_t vacc13 = vacc03;
+ int32_t vacc20 = vacc00;
+ int32_t vacc21 = vacc01;
+ int32_t vacc22 = vacc02;
+ int32_t vacc23 = vacc03;
+ int32_t vacc30 = vacc00;
+ int32_t vacc31 = vacc01;
+ int32_t vacc32 = vacc02;
+ int32_t vacc33 = vacc03;
+ int32_t vacc40 = vacc00;
+ int32_t vacc41 = vacc01;
+ int32_t vacc42 = vacc02;
+ int32_t vacc43 = vacc03;
+ int32_t vacc50 = vacc00;
+ int32_t vacc51 = vacc01;
+ int32_t vacc52 = vacc02;
+ int32_t vacc53 = vacc03;
+ int32_t vacc60 = vacc00;
+ int32_t vacc61 = vacc01;
+ int32_t vacc62 = vacc02;
+ int32_t vacc63 = vacc03;
+ int32_t vacc70 = vacc00;
+ int32_t vacc71 = vacc01;
+ int32_t vacc72 = vacc02;
+ int32_t vacc73 = vacc03;
+ int32_t vacc80 = vacc00;
+ int32_t vacc81 = vacc01;
+ int32_t vacc82 = vacc02;
+ int32_t vacc83 = vacc03;
+ int32_t vacc90 = vacc00;
+ int32_t vacc91 = vacc01;
+ int32_t vacc92 = vacc02;
+ int32_t vacc93 = vacc03;
+ int32_t vacc100 = vacc00;
+ int32_t vacc101 = vacc01;
+ int32_t vacc102 = vacc02;
+ int32_t vacc103 = vacc03;
+ int32_t vacc110 = vacc00;
+ int32_t vacc111 = vacc01;
+ int32_t vacc112 = vacc02;
+ int32_t vacc113 = vacc03;
w = (const void*) ((uintptr_t) w + 4 * sizeof(int32_t));
diff --git a/src/qs8-gemm/gen/8x8c4-minmax-scalar.c b/src/qs8-gemm/gen/8x8c4-minmax-scalar.c
index 6c98602..e5719f3 100644
--- a/src/qs8-gemm/gen/8x8c4-minmax-scalar.c
+++ b/src/qs8-gemm/gen/8x8c4-minmax-scalar.c
@@ -21,6 +21,7 @@
// another thread. We deem this acceptable because the values that are
// read out of bounds do not affect the result, and the the compiler can't know
// about this undefined behavior.
+
void xnn_qs8_gemm_minmax_ukernel_8x8c4__scalar(
size_t mr,
size_t nc,
@@ -84,81 +85,70 @@
// Loop over groups of 8 columns.
do {
- // `vaccMN` is the accumulator at row `M`, column `N`.
- // Initialize accumulators with bias. 8 bias values are loaded from the
- // weight matrix, at the start of the group of 8 columns.
- int32_t bias0 = ((const int32_t*)w)[0];
- int32_t vacc00 = bias0;
- int32_t vacc10 = bias0;
- int32_t vacc20 = bias0;
- int32_t vacc30 = bias0;
- int32_t vacc40 = bias0;
- int32_t vacc50 = bias0;
- int32_t vacc60 = bias0;
- int32_t vacc70 = bias0;
- int32_t bias1 = ((const int32_t*)w)[1];
- int32_t vacc01 = bias1;
- int32_t vacc11 = bias1;
- int32_t vacc21 = bias1;
- int32_t vacc31 = bias1;
- int32_t vacc41 = bias1;
- int32_t vacc51 = bias1;
- int32_t vacc61 = bias1;
- int32_t vacc71 = bias1;
- int32_t bias2 = ((const int32_t*)w)[2];
- int32_t vacc02 = bias2;
- int32_t vacc12 = bias2;
- int32_t vacc22 = bias2;
- int32_t vacc32 = bias2;
- int32_t vacc42 = bias2;
- int32_t vacc52 = bias2;
- int32_t vacc62 = bias2;
- int32_t vacc72 = bias2;
- int32_t bias3 = ((const int32_t*)w)[3];
- int32_t vacc03 = bias3;
- int32_t vacc13 = bias3;
- int32_t vacc23 = bias3;
- int32_t vacc33 = bias3;
- int32_t vacc43 = bias3;
- int32_t vacc53 = bias3;
- int32_t vacc63 = bias3;
- int32_t vacc73 = bias3;
- int32_t bias4 = ((const int32_t*)w)[4];
- int32_t vacc04 = bias4;
- int32_t vacc14 = bias4;
- int32_t vacc24 = bias4;
- int32_t vacc34 = bias4;
- int32_t vacc44 = bias4;
- int32_t vacc54 = bias4;
- int32_t vacc64 = bias4;
- int32_t vacc74 = bias4;
- int32_t bias5 = ((const int32_t*)w)[5];
- int32_t vacc05 = bias5;
- int32_t vacc15 = bias5;
- int32_t vacc25 = bias5;
- int32_t vacc35 = bias5;
- int32_t vacc45 = bias5;
- int32_t vacc55 = bias5;
- int32_t vacc65 = bias5;
- int32_t vacc75 = bias5;
- int32_t bias6 = ((const int32_t*)w)[6];
- int32_t vacc06 = bias6;
- int32_t vacc16 = bias6;
- int32_t vacc26 = bias6;
- int32_t vacc36 = bias6;
- int32_t vacc46 = bias6;
- int32_t vacc56 = bias6;
- int32_t vacc66 = bias6;
- int32_t vacc76 = bias6;
- int32_t bias7 = ((const int32_t*)w)[7];
- int32_t vacc07 = bias7;
- int32_t vacc17 = bias7;
- int32_t vacc27 = bias7;
- int32_t vacc37 = bias7;
- int32_t vacc47 = bias7;
- int32_t vacc57 = bias7;
- int32_t vacc67 = bias7;
- int32_t vacc77 = bias7;
+ int32_t vacc00 = ((const int32_t*) w)[0];
+ int32_t vacc01 = ((const int32_t*) w)[1];
+ int32_t vacc02 = ((const int32_t*) w)[2];
+ int32_t vacc03 = ((const int32_t*) w)[3];
+ int32_t vacc04 = ((const int32_t*) w)[4];
+ int32_t vacc05 = ((const int32_t*) w)[5];
+ int32_t vacc06 = ((const int32_t*) w)[6];
+ int32_t vacc07 = ((const int32_t*) w)[7];
+ int32_t vacc10 = vacc00;
+ int32_t vacc11 = vacc01;
+ int32_t vacc12 = vacc02;
+ int32_t vacc13 = vacc03;
+ int32_t vacc14 = vacc04;
+ int32_t vacc15 = vacc05;
+ int32_t vacc16 = vacc06;
+ int32_t vacc17 = vacc07;
+ int32_t vacc20 = vacc00;
+ int32_t vacc21 = vacc01;
+ int32_t vacc22 = vacc02;
+ int32_t vacc23 = vacc03;
+ int32_t vacc24 = vacc04;
+ int32_t vacc25 = vacc05;
+ int32_t vacc26 = vacc06;
+ int32_t vacc27 = vacc07;
+ int32_t vacc30 = vacc00;
+ int32_t vacc31 = vacc01;
+ int32_t vacc32 = vacc02;
+ int32_t vacc33 = vacc03;
+ int32_t vacc34 = vacc04;
+ int32_t vacc35 = vacc05;
+ int32_t vacc36 = vacc06;
+ int32_t vacc37 = vacc07;
+ int32_t vacc40 = vacc00;
+ int32_t vacc41 = vacc01;
+ int32_t vacc42 = vacc02;
+ int32_t vacc43 = vacc03;
+ int32_t vacc44 = vacc04;
+ int32_t vacc45 = vacc05;
+ int32_t vacc46 = vacc06;
+ int32_t vacc47 = vacc07;
+ int32_t vacc50 = vacc00;
+ int32_t vacc51 = vacc01;
+ int32_t vacc52 = vacc02;
+ int32_t vacc53 = vacc03;
+ int32_t vacc54 = vacc04;
+ int32_t vacc55 = vacc05;
+ int32_t vacc56 = vacc06;
+ int32_t vacc57 = vacc07;
+ int32_t vacc60 = vacc00;
+ int32_t vacc61 = vacc01;
+ int32_t vacc62 = vacc02;
+ int32_t vacc63 = vacc03;
+ int32_t vacc64 = vacc04;
+ int32_t vacc65 = vacc05;
+ int32_t vacc66 = vacc06;
+ int32_t vacc67 = vacc07;
+ int32_t vacc70 = vacc00;
+ int32_t vacc71 = vacc01;
+ int32_t vacc72 = vacc02;
+ int32_t vacc73 = vacc03;
+ int32_t vacc74 = vacc04;
+ int32_t vacc75 = vacc05;
+ int32_t vacc76 = vacc06;
+ int32_t vacc77 = vacc07;
w = (const void*) ((uintptr_t) w + 8 * sizeof(int32_t));