Loop Id: 4008 | Module: exec | Source: csr_matvec.c:259-261 | Coverage: 6.55% |
---|
Loop Id: 4008 | Module: exec | Source: csr_matvec.c:259-261 | Coverage: 6.55% |
---|
0x4e9ef0 VMOVUPD (%RBX,%RSI,8),%YMM1 [2] |
0x4e9ef5 KXNORW %K0,%K0,%K1 |
0x4e9ef9 VXORPD %XMM2,%XMM2,%XMM2 |
0x4e9efd VGATHERQPD (%R9,%YMM1,8),%YMM2{%K1} [1] |
0x4e9f04 VFMADD231PD (%RDI,%RSI,8),%YMM2,%YMM0 [3] |
0x4e9f0a ADD $0x4,%RSI |
0x4e9f0e CMP %RCX,%RSI |
0x4e9f11 JBE 4e9ef0 |
/home/kcamus/qaas_runs/169-443-9681/intel/AMG/build/AMG/AMG/seq_mv/csr_matvec.c: 259 - 261 |
-------------------------------------------------------------------------------- |
259: for (jj = A_i[i]; jj < A_i[i+1]; jj++) |
260: { |
261: tempx += A_data[jj] * x_data[A_j[jj]]; |
Coverage (%) | Name | Source Location | Module |
---|---|---|---|
►100.00+ | __kmp_invoke_microtask | libiomp5.so | |
○ | __kmp_fork_call | libiomp5.so | |
○ | __kmpc_fork_call | libiomp5.so | |
○ | hypre_CSRMatrixMatvecOutOfPlac[...] | csr_matvec.c:243 | exec |
○ | hypre_ParCSRMatrixMatvecOutOfP[...] | par_csr_matvec.c:216 | exec |
○ | hypre_PCGSolve | pcg.c:496 | exec |
○ | main | amg.c:419 | exec |
○ | __libc_init_first | libc.so.6 |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.00 |
CQA speedup if FP arith vectorized | 1.33 |
CQA speedup if fully vectorized | 2.67 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.33 |
Bottlenecks | |
Function | hypre_CSRMatrixMatvecOutOfPlace.extracted |
Source | csr_matvec.c:259-261 |
Source loop unroll info | unrolled by 4 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 4 |
CQA cycles | 4.00 |
CQA cycles if no scalar integer | 4.00 |
CQA cycles if FP arith vectorized | 3.00 |
CQA cycles if fully vectorized | 1.50 |
Front-end cycles | 2.75 |
DIV/SQRT cycles | 2.00 |
P0 cycles | 1.50 |
P1 cycles | 3.00 |
P2 cycles | 3.00 |
P3 cycles | 0.00 |
P4 cycles | 1.25 |
P5 cycles | 1.25 |
P6 cycles | 0.00 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 4 |
FE+BE cycles (UFS) | 40.24 |
Stall cycles (UFS) | 37.46 |
Nb insns | 8.00 |
Nb uops | 10.00 |
Nb loads | 3.00 |
Nb stores | 0.00 |
Nb stack references | 0.00 |
FLOP/cycle | 2.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 4.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 24.00 |
Bytes prefetched | 0.00 |
Bytes loaded | 96.00 |
Bytes stored | 0.00 |
Stride 0 | 0.00 |
Stride 1 | 2.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 1.00 |
Vectorization ratio all | 100.00 |
Vectorization ratio load | 100.00 |
Vectorization ratio store | NA |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | 100.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 100.00 |
Vector-efficiency ratio all | 43.75 |
Vector-efficiency ratio load | 50.00 |
Vector-efficiency ratio store | NA |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | 50.00 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 37.50 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.00 |
CQA speedup if FP arith vectorized | 1.33 |
CQA speedup if fully vectorized | 2.67 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.33 |
Bottlenecks | |
Function | hypre_CSRMatrixMatvecOutOfPlace.extracted |
Source | csr_matvec.c:259-261 |
Source loop unroll info | unrolled by 4 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 4 |
CQA cycles | 4.00 |
CQA cycles if no scalar integer | 4.00 |
CQA cycles if FP arith vectorized | 3.00 |
CQA cycles if fully vectorized | 1.50 |
Front-end cycles | 2.75 |
DIV/SQRT cycles | 2.00 |
P0 cycles | 1.50 |
P1 cycles | 3.00 |
P2 cycles | 3.00 |
P3 cycles | 0.00 |
P4 cycles | 1.25 |
P5 cycles | 1.25 |
P6 cycles | 0.00 |
P7 cycles | 0.00 |
Inter-iter dependencies cycles | 4 |
FE+BE cycles (UFS) | 40.24 |
Stall cycles (UFS) | 37.46 |
Nb insns | 8.00 |
Nb uops | 10.00 |
Nb loads | 3.00 |
Nb stores | 0.00 |
Nb stack references | 0.00 |
FLOP/cycle | 2.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 4.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 24.00 |
Bytes prefetched | 0.00 |
Bytes loaded | 96.00 |
Bytes stored | 0.00 |
Stride 0 | 0.00 |
Stride 1 | 2.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 1.00 |
Vectorization ratio all | 100.00 |
Vectorization ratio load | 100.00 |
Vectorization ratio store | NA |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | 100.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 100.00 |
Vector-efficiency ratio all | 43.75 |
Vector-efficiency ratio load | 50.00 |
Vector-efficiency ratio store | NA |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | 50.00 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 37.50 |
Path / |
Function | hypre_CSRMatrixMatvecOutOfPlace.extracted |
Source file and lines | csr_matvec.c:259-261 |
Module | exec |
nb instructions | 8 |
nb uops | 10 |
loop length | 35 |
used x86 registers | 5 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 3 |
used zmm registers | 0 |
nb stack references | 0 |
micro-operation queue | 2.75 cycles |
front end | 2.75 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 2.00 | 1.50 | 3.00 | 3.00 | 0.00 | 1.25 | 1.25 | 0.00 |
cycles | 2.00 | 1.50 | 3.00 | 3.00 | 0.00 | 1.25 | 1.25 | 0.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 4.00 |
FE+BE cycles | 40.24 |
Stall cycles | 37.46 |
LB full (events) | 38.70 |
Front-end | 2.75 |
Dispatch | 3.00 |
Data deps. | 4.00 |
Overall L1 | 4.00 |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 43% |
load | 50% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 50% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 37% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
VMOVUPD (%RBX,%RSI,8),%YMM1 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
KXNORW %K0,%K0,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
VXORPD %XMM2,%XMM2,%XMM2 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VGATHERQPD (%R9,%YMM1,8),%YMM2{%K1} | 4 | 1 | 0 | 2 | 2 | 0 | 1 | 0 | 0 | 20 | 4 |
VFMADD231PD (%RDI,%RSI,8),%YMM2,%YMM0 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
ADD $0x4,%RSI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
CMP %RCX,%RSI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JBE 4e9ef0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
Function | hypre_CSRMatrixMatvecOutOfPlace.extracted |
Source file and lines | csr_matvec.c:259-261 |
Module | exec |
nb instructions | 8 |
nb uops | 10 |
loop length | 35 |
used x86 registers | 5 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 3 |
used zmm registers | 0 |
nb stack references | 0 |
micro-operation queue | 2.75 cycles |
front end | 2.75 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | |
---|---|---|---|---|---|---|---|---|
uops | 2.00 | 1.50 | 3.00 | 3.00 | 0.00 | 1.25 | 1.25 | 0.00 |
cycles | 2.00 | 1.50 | 3.00 | 3.00 | 0.00 | 1.25 | 1.25 | 0.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 4.00 |
FE+BE cycles | 40.24 |
Stall cycles | 37.46 |
LB full (events) | 38.70 |
Front-end | 2.75 |
Dispatch | 3.00 |
Data deps. | 4.00 |
Overall L1 | 4.00 |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 100% |
all | 43% |
load | 50% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 50% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 37% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|
VMOVUPD (%RBX,%RSI,8),%YMM1 | 1 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5-6 | 0.50 |
KXNORW %K0,%K0,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
VXORPD %XMM2,%XMM2,%XMM2 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 |
VGATHERQPD (%R9,%YMM1,8),%YMM2{%K1} | 4 | 1 | 0 | 2 | 2 | 0 | 1 | 0 | 0 | 20 | 4 |
VFMADD231PD (%RDI,%RSI,8),%YMM2,%YMM0 | 1 | 0.50 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
ADD $0x4,%RSI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
CMP %RCX,%RSI | 1 | 0.25 | 0.25 | 0 | 0 | 0 | 0.25 | 0.25 | 0 | 1 | 0.25 |
JBE 4e9ef0 | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50-1 |
Metric | run_0 |
---|---|
Coverage (% app. time) | 6.55 |
Time (s) | 2.48 |
Instance Count | 68591551 |
Iteration Count - min | 2 |
Iteration Count - avg | 5.98 |
Iteration Count - max | 6 |
Cycles per Iteration - min | 0 |
Cycles per Iteration - avg | 0.98 |
Cycles per Iteration - max | 121640 |