Loop Id: 2596 | Module: exec | Source: ams.c:3382-3401 [...] | Coverage: 0.01% |
---|
Loop Id: 2596 | Module: exec | Source: ams.c:3382-3401 [...] | Coverage: 0.01% |
---|
0x4b800a CMP -0x50(%RBP),%R9 |
0x4b800e LEA 0x1(%R9),%R9 |
0x4b8012 JE 4b831d |
0x4b8018 MOV -0x80(%RBP),%RAX |
0x4b801c LEA (%RAX,%R9,1),%RCX |
0x4b8020 MOV -0x30(%RBP),%RAX |
0x4b8024 LEA (%RAX,%RCX,8),%R10 |
0x4b8028 MOVQ $0,(%RAX,%RCX,8) |
0x4b8030 MOV -0x40(%RBP),%RAX |
0x4b8034 MOV (%RAX,%RCX,8),%RDX |
0x4b8038 MOV 0x8(%RAX,%RCX,8),%RDI |
0x4b803d VXORPD %XMM0,%XMM0,%XMM0 |
0x4b8041 MOV %RDI,%RAX |
0x4b8044 SUB %RDX,%RAX |
0x4b8047 JLE 4b81d4 |
0x4b804d LEA (%R12,%RDI,8),%R8 |
0x4b8051 ADD $-0x8,%R8 |
0x4b8055 CMP %R10,%R8 |
0x4b8058 JB 4b8090 |
0x4b805a LEA (%R12,%RDX,8),%R8 |
0x4b805e CMP %R8,%R10 |
0x4b8061 JB 4b8090 |
0x4b8063 NOPW %CS:(%RAX,%RAX,1) |
(2601) 0x4b8070 VMOVSD (%R12,%RDX,8),%XMM1 |
(2601) 0x4b8076 VANDPD %XMM1,%XMM11,%XMM1 |
(2601) 0x4b807a VADDSD %XMM1,%XMM0,%XMM0 |
(2601) 0x4b807e VMOVSD %XMM0,(%R10) |
(2601) 0x4b8083 INC %RDX |
(2601) 0x4b8086 CMP %RDX,%RDI |
(2601) 0x4b8089 JNE 4b8070 |
0x4b808b JMP 4b81d4 |
0x4b8090 CMP $0x8,%RAX |
0x4b8094 JB 4b8129 |
0x4b809a MOV %RAX,%RDI |
0x4b809d SHR $0x3,%RDI |
0x4b80a1 MOV -0xc8(%RBP),%R8 |
0x4b80a8 LEA (%R8,%RDX,8),%R8 |
0x4b80ac NOPL (%RAX) |
(2600) 0x4b80b0 VMOVSD -0x38(%R8),%XMM1 |
(2600) 0x4b80b6 VMOVSD -0x30(%R8),%XMM2 |
(2600) 0x4b80bc VMOVSD -0x28(%R8),%XMM3 |
(2600) 0x4b80c2 VMOVSD -0x20(%R8),%XMM4 |
(2600) 0x4b80c8 VANDPD %XMM1,%XMM12,%XMM1 |
(2600) 0x4b80cc VANDPD %XMM2,%XMM12,%XMM2 |
(2600) 0x4b80d0 VANDPD %XMM3,%XMM12,%XMM3 |
(2600) 0x4b80d4 VANDPD %XMM4,%XMM12,%XMM4 |
(2600) 0x4b80d8 VMOVSD -0x18(%R8),%XMM5 |
(2600) 0x4b80de VANDPD %XMM5,%XMM12,%XMM5 |
(2600) 0x4b80e2 VMOVSD -0x10(%R8),%XMM6 |
(2600) 0x4b80e8 VANDPD %XMM6,%XMM12,%XMM6 |
(2600) 0x4b80ec VMOVSD -0x8(%R8),%XMM7 |
(2600) 0x4b80f2 VANDPD %XMM7,%XMM12,%XMM7 |
(2600) 0x4b80f6 VMOVSD (%R8),%XMM8 |
(2600) 0x4b80fb VANDPD %XMM12,%XMM8,%XMM8 |
(2600) 0x4b8100 VADDSD %XMM7,%XMM8,%XMM7 |
(2600) 0x4b8104 VADDSD %XMM0,%XMM7,%XMM0 |
(2600) 0x4b8108 VADDSD %XMM5,%XMM4,%XMM4 |
(2600) 0x4b810c VADDSD %XMM6,%XMM3,%XMM3 |
(2600) 0x4b8110 VADDSD %XMM4,%XMM3,%XMM3 |
(2600) 0x4b8114 VADDSD %XMM2,%XMM1,%XMM1 |
(2600) 0x4b8118 VADDSD %XMM1,%XMM0,%XMM0 |
(2600) 0x4b811c VADDSD %XMM3,%XMM0,%XMM0 |
(2600) 0x4b8120 ADD $0x40,%R8 |
(2600) 0x4b8124 DEC %RDI |
(2600) 0x4b8127 JNE 4b80b0 |
0x4b8129 MOV %EAX,%EDI |
0x4b812b AND $0x7,%EDI |
0x4b812e DEC %RDI |
0x4b8131 CMP $0x6,%RDI |
0x4b8135 JA 4b81cf |
0x4b81cf VMOVSD %XMM0,(%R10) |
0x4b81d4 MOV -0xa0(%RBP),%RAX |
0x4b81db MOV (%RAX,%RCX,8),%RDX |
0x4b81df MOV 0x8(%RAX,%RCX,8),%RCX |
0x4b81e4 MOV %RCX,%RAX |
0x4b81e7 SUB %RDX,%RAX |
0x4b81ea JLE 4b800a |
0x4b81f0 LEA (%R11,%RCX,8),%RDI |
0x4b81f4 ADD $-0x8,%RDI |
0x4b81f8 CMP %R10,%RDI |
0x4b81fb JB 4b8230 |
0x4b81fd LEA (%R11,%RDX,8),%RDI |
0x4b8201 CMP %RDI,%R10 |
0x4b8204 JB 4b8230 |
0x4b8206 NOPW %CS:(%RAX,%RAX,1) |
(2599) 0x4b8210 VMOVSD (%R11,%RDX,8),%XMM1 |
(2599) 0x4b8216 VANDPD %XMM1,%XMM11,%XMM1 |
(2599) 0x4b821a VADDSD %XMM1,%XMM0,%XMM0 |
(2599) 0x4b821e VMOVSD %XMM0,(%R10) |
(2599) 0x4b8223 INC %RDX |
(2599) 0x4b8226 CMP %RDX,%RCX |
(2599) 0x4b8229 JNE 4b8210 |
0x4b822b JMP 4b800a |
/scratch_na/users/xoserete/qaas_runs/171-172-8218/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c: 3382 - 3401 |
-------------------------------------------------------------------------------- |
3382: for (i = ns; i < ne; i++) |
3383: { |
3384: l1_norm[i] = 0.0; |
3385: if (cf_marker == NULL) |
3386: { |
3387: /* Add the l1 norm of the diag part of the ith row */ |
3388: for (j = A_diag_I[i]; j < A_diag_I[i+1]; j++) |
3389: l1_norm[i] += fabs(A_diag_data[j]); |
3390: /* Add the l1 norm of the offd part of the ith row */ |
3391: if (num_cols_offd) |
3392: { |
3393: for (j = A_offd_I[i]; j < A_offd_I[i+1]; j++) |
3394: l1_norm[i] += fabs(A_offd_data[j]); |
[...] |
3401: for (j = A_diag_I[i]; j < A_diag_I[i+1]; j++) |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.73 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 14.10 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.73 |
Bottlenecks | micro-operation queue, |
Function | hypre_ParCSRComputeL1NormsThreads.extracted |
Source | ams.c:3382-3384,ams.c:3388-3389,ams.c:3393-3394,ams.c:3401-3401 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 8.67 |
CQA cycles if no scalar integer | 5.00 |
CQA cycles if FP arith vectorized | 8.67 |
CQA cycles if fully vectorized | 0.61 |
Front-end cycles | 8.67 |
DIV/SQRT cycles | 5.00 |
P0 cycles | 3.40 |
P1 cycles | 3.33 |
P2 cycles | 3.33 |
P3 cycles | 1.00 |
P4 cycles | 3.40 |
P5 cycles | 5.00 |
P6 cycles | 1.00 |
P7 cycles | 1.00 |
P8 cycles | 1.00 |
P9 cycles | 3.20 |
P10 cycles | 3.33 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | NA |
FE+BE cycles (UFS) | 8.81 |
Stall cycles (UFS) | 0.00 |
Nb insns | 52.00 |
Nb uops | 52.00 |
Nb loads | 10.00 |
Nb stores | 2.00 |
Nb stack references | 6.00 |
FLOP/cycle | 0.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 0.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 11.08 |
Bytes prefetched | 0.00 |
Bytes loaded | 80.00 |
Bytes stored | 16.00 |
Stride 0 | NA |
Stride 1 | NA |
Stride n | NA |
Stride unknown | NA |
Stride indirect | NA |
Vectorization ratio all | 11.11 |
Vectorization ratio load | 0.00 |
Vectorization ratio store | 0.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 14.29 |
Vector-efficiency ratio all | 13.19 |
Vector-efficiency ratio load | 12.50 |
Vector-efficiency ratio store | 9.38 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 14.29 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.73 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 14.10 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.73 |
Bottlenecks | micro-operation queue, |
Function | hypre_ParCSRComputeL1NormsThreads.extracted |
Source | ams.c:3382-3384,ams.c:3388-3389,ams.c:3393-3394,ams.c:3401-3401 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 8.67 |
CQA cycles if no scalar integer | 5.00 |
CQA cycles if FP arith vectorized | 8.67 |
CQA cycles if fully vectorized | 0.61 |
Front-end cycles | 8.67 |
DIV/SQRT cycles | 5.00 |
P0 cycles | 3.40 |
P1 cycles | 3.33 |
P2 cycles | 3.33 |
P3 cycles | 1.00 |
P4 cycles | 3.40 |
P5 cycles | 5.00 |
P6 cycles | 1.00 |
P7 cycles | 1.00 |
P8 cycles | 1.00 |
P9 cycles | 3.20 |
P10 cycles | 3.33 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | NA |
FE+BE cycles (UFS) | 8.81 |
Stall cycles (UFS) | 0.00 |
Nb insns | 52.00 |
Nb uops | 52.00 |
Nb loads | 10.00 |
Nb stores | 2.00 |
Nb stack references | 6.00 |
FLOP/cycle | 0.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 0.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 11.08 |
Bytes prefetched | 0.00 |
Bytes loaded | 80.00 |
Bytes stored | 16.00 |
Stride 0 | NA |
Stride 1 | NA |
Stride n | NA |
Stride unknown | NA |
Stride indirect | NA |
Vectorization ratio all | 11.11 |
Vectorization ratio load | 0.00 |
Vectorization ratio store | 0.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 14.29 |
Vector-efficiency ratio all | 13.19 |
Vector-efficiency ratio load | 12.50 |
Vector-efficiency ratio store | 9.38 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 14.29 |
Path / |
Function | hypre_ParCSRComputeL1NormsThreads.extracted |
Source file and lines | ams.c:3382-3401 |
Module | exec |
nb instructions | 52 |
nb uops | 52 |
loop length | 227 |
used x86 registers | 10 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 6 |
micro-operation queue | 8.67 cycles |
front end | 8.67 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 5.00 | 3.40 | 3.33 | 3.33 | 1.00 | 3.40 | 5.00 | 1.00 | 1.00 | 1.00 | 3.20 | 3.33 |
cycles | 5.00 | 3.40 | 3.33 | 3.33 | 1.00 | 3.40 | 5.00 | 1.00 | 1.00 | 1.00 | 3.20 | 3.33 |
Cycles executing div or sqrt instructions | NA |
FE+BE cycles | 8.81 |
Stall cycles | 0.00 |
Front-end | 8.67 |
Dispatch | 5.00 |
Overall L1 | 8.67 |
all | 0% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 0% |
all | 50% |
load | NA (no load vectorizable/vectorized instructions) |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 11% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 14% |
all | 11% |
load | 12% |
store | 6% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 12% |
all | 18% |
load | NA (no load vectorizable/vectorized instructions) |
store | 12% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 25% |
all | 13% |
load | 12% |
store | 9% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 14% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
CMP -0x50(%RBP),%R9 | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
LEA 0x1(%R9),%R9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
JE 4b831d <hypre_ParCSRComputeL1NormsThreads.extracted+0x1d9d> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV -0x80(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA (%RAX,%R9,1),%RCX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV -0x30(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA (%RAX,%RCX,8),%R10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOVQ $0,(%RAX,%RCX,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0x40(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RAX,%RCX,8),%RDX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV 0x8(%RAX,%RCX,8),%RDI | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
VXORPD %XMM0,%XMM0,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV %RDI,%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
SUB %RDX,%RAX | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JLE 4b81d4 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1c54> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R12,%RDI,8),%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
ADD $-0x8,%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R10,%R8 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8090 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1b10> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R12,%RDX,8),%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMP %R8,%R10 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8090 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1b10> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
NOPW %CS:(%RAX,%RAX,1) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
JMP 4b81d4 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1c54> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.08 |
CMP $0x8,%RAX | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8129 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1ba9> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %RAX,%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
SHR $0x3,%RDI | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0-2 | 0.50 |
MOV -0xc8(%RBP),%R8 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA (%R8,%RDX,8),%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
NOPL (%RAX) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV %EAX,%EDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
AND $0x7,%EDI | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1-2 | 0.20 |
DEC %RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP $0x6,%RDI | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JA 4b81cf <hypre_ParCSRComputeL1NormsThreads.extracted+0x1c4f> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
VMOVSD %XMM0,(%R10) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0xa0(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RAX,%RCX,8),%RDX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV 0x8(%RAX,%RCX,8),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV %RCX,%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
SUB %RDX,%RAX | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JLE 4b800a <hypre_ParCSRComputeL1NormsThreads.extracted+0x1a8a> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R11,%RCX,8),%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
ADD $-0x8,%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R10,%RDI | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8230 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1cb0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R11,%RDX,8),%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMP %RDI,%R10 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8230 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1cb0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
NOPW %CS:(%RAX,%RAX,1) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
JMP 4b800a <hypre_ParCSRComputeL1NormsThreads.extracted+0x1a8a> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.08 |
Function | hypre_ParCSRComputeL1NormsThreads.extracted |
Source file and lines | ams.c:3382-3401 |
Module | exec |
nb instructions | 52 |
nb uops | 52 |
loop length | 227 |
used x86 registers | 10 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 6 |
micro-operation queue | 8.67 cycles |
front end | 8.67 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 5.00 | 3.40 | 3.33 | 3.33 | 1.00 | 3.40 | 5.00 | 1.00 | 1.00 | 1.00 | 3.20 | 3.33 |
cycles | 5.00 | 3.40 | 3.33 | 3.33 | 1.00 | 3.40 | 5.00 | 1.00 | 1.00 | 1.00 | 3.20 | 3.33 |
Cycles executing div or sqrt instructions | NA |
FE+BE cycles | 8.81 |
Stall cycles | 0.00 |
Front-end | 8.67 |
Dispatch | 5.00 |
Overall L1 | 8.67 |
all | 0% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 0% |
all | 50% |
load | NA (no load vectorizable/vectorized instructions) |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 11% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 14% |
all | 11% |
load | 12% |
store | 6% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 12% |
all | 18% |
load | NA (no load vectorizable/vectorized instructions) |
store | 12% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 25% |
all | 13% |
load | 12% |
store | 9% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 14% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
CMP -0x50(%RBP),%R9 | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
LEA 0x1(%R9),%R9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
JE 4b831d <hypre_ParCSRComputeL1NormsThreads.extracted+0x1d9d> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV -0x80(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA (%RAX,%R9,1),%RCX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV -0x30(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA (%RAX,%RCX,8),%R10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOVQ $0,(%RAX,%RCX,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0x40(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RAX,%RCX,8),%RDX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV 0x8(%RAX,%RCX,8),%RDI | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
VXORPD %XMM0,%XMM0,%XMM0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV %RDI,%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
SUB %RDX,%RAX | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JLE 4b81d4 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1c54> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R12,%RDI,8),%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
ADD $-0x8,%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R10,%R8 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8090 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1b10> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R12,%RDX,8),%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMP %R8,%R10 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8090 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1b10> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
NOPW %CS:(%RAX,%RAX,1) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
JMP 4b81d4 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1c54> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.08 |
CMP $0x8,%RAX | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8129 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1ba9> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %RAX,%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
SHR $0x3,%RDI | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0-2 | 0.50 |
MOV -0xc8(%RBP),%R8 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA (%R8,%RDX,8),%R8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
NOPL (%RAX) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV %EAX,%EDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
AND $0x7,%EDI | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1-2 | 0.20 |
DEC %RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP $0x6,%RDI | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JA 4b81cf <hypre_ParCSRComputeL1NormsThreads.extracted+0x1c4f> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
VMOVSD %XMM0,(%R10) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0xa0(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RAX,%RCX,8),%RDX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV 0x8(%RAX,%RCX,8),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV %RCX,%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
SUB %RDX,%RAX | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JLE 4b800a <hypre_ParCSRComputeL1NormsThreads.extracted+0x1a8a> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R11,%RCX,8),%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
ADD $-0x8,%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R10,%RDI | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8230 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1cb0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
LEA (%R11,%RDX,8),%RDI | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMP %RDI,%R10 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JB 4b8230 <hypre_ParCSRComputeL1NormsThreads.extracted+0x1cb0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
NOPW %CS:(%RAX,%RAX,1) | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
JMP 4b800a <hypre_ParCSRComputeL1NormsThreads.extracted+0x1a8a> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2.08 |