Loop Id: 482 | Module: libparcsr_ls.so | Source: ams.c:3677-3680 | Coverage: 0.02% |
---|
Loop Id: 482 | Module: libparcsr_ls.so | Source: ams.c:3677-3680 | Coverage: 0.02% |
---|
0x251b0 VMOVUPD (%RCX,%R9,8),%YMM4 [3] |
0x251b6 VXORPD %XMM5,%XMM5,%XMM5 |
0x251ba KXNORW %K0,%K0,%K1 |
0x251be VGATHERQPD (%RBX,%YMM4,8),%YMM5{%K1} [1] |
0x251c5 VFNMADD231PD (%RAX,%R9,8),%YMM5,%YMM3 [2] |
0x251cb ADD $0x4,%R9 |
0x251cf CMP %R11,%R9 |
0x251d2 JBE 251b0 |
/home/eoseret/qaas_runs_CPU_9468/171-716-5699/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c: 3677 - 3680 |
-------------------------------------------------------------------------------- |
3677: for (jj = A_offd_i[i]; jj < A_offd_i[i+1]; jj++) |
3678: { |
3679: ii = A_offd_j[jj]; |
3680: res -= A_offd_data[jj] * Vext_data[ii]; |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.00 |
CQA speedup if FP arith vectorized | 1.52 |
CQA speedup if fully vectorized | 2.91 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 2.00 |
Bottlenecks | |
Function | hypre_ParCSRRelaxThreads.extracted.57 |
Source | ams.c:3677-3680 |
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 | 2.63 |
CQA cycles if fully vectorized | 1.38 |
Front-end cycles | 2.00 |
DIV/SQRT cycles | 2.00 |
P0 cycles | 2.00 |
P1 cycles | 2.00 |
P2 cycles | 2.00 |
P3 cycles | 0.00 |
P4 cycles | 1.00 |
P5 cycles | 1.00 |
P6 cycles | 0.00 |
P7 cycles | 0.00 |
P8 cycles | 0.00 |
P9 cycles | 0.00 |
P10 cycles | 2.00 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | 4 |
FE+BE cycles (UFS) | 4.19 - 87.54 |
Stall cycles (UFS) | 1.68 - 85.03 |
Nb insns | 8.00 |
Nb uops | 11.00 |
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 | 1.52 |
CQA speedup if fully vectorized | 2.91 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 2.00 |
Bottlenecks | |
Function | hypre_ParCSRRelaxThreads.extracted.57 |
Source | ams.c:3677-3680 |
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 | 2.63 |
CQA cycles if fully vectorized | 1.38 |
Front-end cycles | 2.00 |
DIV/SQRT cycles | 2.00 |
P0 cycles | 2.00 |
P1 cycles | 2.00 |
P2 cycles | 2.00 |
P3 cycles | 0.00 |
P4 cycles | 1.00 |
P5 cycles | 1.00 |
P6 cycles | 0.00 |
P7 cycles | 0.00 |
P8 cycles | 0.00 |
P9 cycles | 0.00 |
P10 cycles | 2.00 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | 4 |
FE+BE cycles (UFS) | 4.19 - 87.54 |
Stall cycles (UFS) | 1.68 - 85.03 |
Nb insns | 8.00 |
Nb uops | 11.00 |
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_ParCSRRelaxThreads.extracted.57 |
Source file and lines | ams.c:3677-3680 |
Module | libparcsr_ls.so |
nb instructions | 8 |
nb uops | 11 |
loop length | 36 |
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 | 2.00 cycles |
front end | 2.00 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 2.00 | 2.00 | 2.00 | 2.00 | 0.00 | 1.00 | 1.00 | 0.00 | 0.00 | 0.00 | 0.00 | 2.00 |
cycles | 2.00 | 2.00 | 2.00 | 2.00 | 0.00 | 1.00 | 1.00 | 0.00 | 0.00 | 0.00 | 0.00 | 2.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 4.00 |
FE+BE cycles | 4.19-87.54 |
Stall cycles | 1.68-85.02 |
ROB full (events) | 2.15-77.63 |
Front-end | 2.00 |
Dispatch | 2.00 |
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 | P8 | P9 | P10 | P11 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
VMOVUPD (%RCX,%R9,8),%YMM4 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0-1 | 0.33 |
VXORPD %XMM5,%XMM5,%XMM5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
KXNORW %K0,%K0,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
VGATHERQPD (%RBX,%YMM4,8),%YMM5{%K1} | 5 | 1 | 1 | 1.33 | 1.33 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1.33 | 0-29 | 2 |
VFNMADD231PD (%RAX,%R9,8),%YMM5,%YMM3 | 1 | 0.50 | 0.50 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 4 | 0.50 |
ADD $0x4,%R9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R11,%R9 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JBE 251b0 <hypre_ParCSRRelaxThreads.extracted.57+0x2f0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
Function | hypre_ParCSRRelaxThreads.extracted.57 |
Source file and lines | ams.c:3677-3680 |
Module | libparcsr_ls.so |
nb instructions | 8 |
nb uops | 11 |
loop length | 36 |
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 | 2.00 cycles |
front end | 2.00 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 2.00 | 2.00 | 2.00 | 2.00 | 0.00 | 1.00 | 1.00 | 0.00 | 0.00 | 0.00 | 0.00 | 2.00 |
cycles | 2.00 | 2.00 | 2.00 | 2.00 | 0.00 | 1.00 | 1.00 | 0.00 | 0.00 | 0.00 | 0.00 | 2.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 4.00 |
FE+BE cycles | 4.19-87.54 |
Stall cycles | 1.68-85.02 |
ROB full (events) | 2.15-77.63 |
Front-end | 2.00 |
Dispatch | 2.00 |
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 | P8 | P9 | P10 | P11 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
VMOVUPD (%RCX,%R9,8),%YMM4 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0-1 | 0.33 |
VXORPD %XMM5,%XMM5,%XMM5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
KXNORW %K0,%K0,%K1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 |
VGATHERQPD (%RBX,%YMM4,8),%YMM5{%K1} | 5 | 1 | 1 | 1.33 | 1.33 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1.33 | 0-29 | 2 |
VFNMADD231PD (%RAX,%R9,8),%YMM5,%YMM3 | 1 | 0.50 | 0.50 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 4 | 0.50 |
ADD $0x4,%R9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R11,%R9 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JBE 251b0 <hypre_ParCSRRelaxThreads.extracted.57+0x2f0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |