| orig_default | aocc_default | icx_2 | aocc_6 |
|---|---|---|---|
[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics. | [ 3 / 3 ] Host configuration allows retrieval of all necessary metrics. | [ 3 / 3 ] Host configuration allows retrieval of all necessary metrics. | [ 3 / 3 ] Host configuration allows retrieval of all necessary metrics. |
[ 0 / 4 ] Application profile is too short (5.37 s) If the overall application profiling time is less than 10 seconds, many of the measurements at function or loop level will very likely be under the measurement quality threshold (0,1 seconds). Rerun to increase runtime duration: for example use a larger dataset or include a repetition loop. | [ 0 / 4 ] Application profile is too short (5.02 s) If the overall application profiling time is less than 10 seconds, many of the measurements at function or loop level will very likely be under the measurement quality threshold (0,1 seconds). Rerun to increase runtime duration: for example use a larger dataset or include a repetition loop. | [ 0 / 4 ] Application profile is too short (5.34 s) If the overall application profiling time is less than 10 seconds, many of the measurements at function or loop level will very likely be under the measurement quality threshold (0,1 seconds). Rerun to increase runtime duration: for example use a larger dataset or include a repetition loop. | [ 0 / 4 ] Application profile is too short (4.98 s) If the overall application profiling time is less than 10 seconds, many of the measurements at function or loop level will very likely be under the measurement quality threshold (0,1 seconds). Rerun to increase runtime duration: for example use a larger dataset or include a repetition loop. |
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time) To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code | [ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.01 % of the execution time) To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code | [ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.00 % of the execution time) To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code | [ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.01 % of the execution time) To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution time in order to analyze as much as possible of the user code |
[ 0 / 9 ] Compilation options are not available Compilation options are an important optimization leverage but ONE-View is not able to analyze them. | [ 0 / 9 ] Compilation options are not available Compilation options are an important optimization leverage but ONE-View is not able to analyze them. | [ 0 / 9 ] Compilation options are not available Compilation options are an important optimization leverage but ONE-View is not able to analyze them. | [ 0 / 9 ] Compilation options are not available Compilation options are an important optimization leverage but ONE-View is not able to analyze them. |
[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated. | [ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated. | [ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated. | [ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated. |
| orig_default | aocc_default | icx_2 | aocc_6 |
|---|---|---|---|
[ 4 / 4 ] CPU activity is good CPU cores are active 93.51% of time | [ 4 / 4 ] CPU activity is good CPU cores are active 92.49% of time | [ 4 / 4 ] CPU activity is good CPU cores are active 93.35% of time | [ 4 / 4 ] CPU activity is good CPU cores are active 92.37% of time |
[ 4 / 4 ] Affinity is good (95.99%) Threads are not migrating to CPU cores: probably successfully pinned | [ 4 / 4 ] Affinity is good (98.64%) Threads are not migrating to CPU cores: probably successfully pinned | [ 4 / 4 ] Affinity is good (95.79%) Threads are not migrating to CPU cores: probably successfully pinned | [ 4 / 4 ] Affinity is good (97.38%) Threads are not migrating to CPU cores: probably successfully pinned |
[ 0 / 3 ] Too many functions do not use all threads Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (11.32%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads | [ 0 / 3 ] Too many functions do not use all threads Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (12.80%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads | [ 0 / 3 ] Too many functions do not use all threads Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (10.33%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads | [ 0 / 3 ] Too many functions do not use all threads Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (12.83%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads |
[ 0 / 3 ] Cumulative Outermost/In between loops coverage (20.61%) greater than cumulative innermost loop coverage (8.78%) Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex | [ 0 / 3 ] Cumulative Outermost/In between loops coverage (36.95%) greater than cumulative innermost loop coverage (4.79%) Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex | [ 3 / 3 ] Cumulative Outermost/In between loops coverage (0.35%) lower than cumulative innermost loop coverage (8.15%) Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex | [ 0 / 3 ] Cumulative Outermost/In between loops coverage (37.32%) greater than cumulative innermost loop coverage (4.88%) Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex |
[ 0 / 4 ] A significant amount of threads are idle (72.15%) On average, more than 10% of observed threads are idle. Such threads are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling. | [ 0 / 4 ] A significant amount of threads are idle (73.90%) On average, more than 10% of observed threads are idle. Such threads are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling. | [ 0 / 4 ] A significant amount of threads are idle (72.16%) On average, more than 10% of observed threads are idle. Such threads are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling. | [ 0 / 4 ] A significant amount of threads are idle (73.62%) On average, more than 10% of observed threads are idle. Such threads are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling. |
[ 2 / 2 ] Less than 10% (0.00%) is spend in BLAS2 operations BLAS2 calls usually could make a poor cache usage and could benefit from inlining. | [ 2 / 2 ] Less than 10% (0.00%) is spend in BLAS2 operations BLAS2 calls usually could make a poor cache usage and could benefit from inlining. | [ 2 / 2 ] Less than 10% (0.00%) is spend in BLAS2 operations BLAS2 calls usually could make a poor cache usage and could benefit from inlining. | [ 2 / 2 ] Less than 10% (0.00%) is spend in BLAS2 operations BLAS2 calls usually could make a poor cache usage and could benefit from inlining. |
[ 0 / 4 ] Too little time of the experiment time spent in analyzed innermost loops (8.78%) If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances. | [ 0 / 4 ] Too little time of the experiment time spent in analyzed innermost loops (4.79%) If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances. | [ 0 / 4 ] Too little time of the experiment time spent in analyzed innermost loops (8.15%) If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances. | [ 0 / 4 ] Too little time of the experiment time spent in analyzed innermost loops (4.88%) If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances. |
[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations It could be more efficient to inline by hand BLAS1 operations | [ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations It could be more efficient to inline by hand BLAS1 operations | [ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations It could be more efficient to inline by hand BLAS1 operations | [ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations It could be more efficient to inline by hand BLAS1 operations |
[ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions) | [ 2 / 2 ] Less than 10% (0.44%) is spend in Libm/SVML (special functions) | [ 2 / 2 ] Less than 10% (0.00%) is spend in Libm/SVML (special functions) | [ 2 / 2 ] Less than 10% (0.35%) is spend in Libm/SVML (special functions) |
[ 4 / 4 ] Loop profile is not flat At least one loop coverage is greater than 4% (20.27%), representing an hotspot for the application | [ 4 / 4 ] Loop profile is not flat At least one loop coverage is greater than 4% (36.64%), representing an hotspot for the application | [ 0 / 4 ] Loop profile is flat No hotspot found in the application (greatest loop coverage is 3.26%), and the twenty hottest loops cumulated coverage is lower than 20% of the application profiled time (8.41%) | [ 4 / 4 ] Loop profile is not flat At least one loop coverage is greater than 4% (37.01%), representing an hotspot for the application |
[ 0 / 4 ] Too little time of the experiment time spent in analyzed loops (29.39%) If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances. | [ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (41.74%) If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances. | [ 0 / 4 ] Too little time of the experiment time spent in analyzed loops (8.50%) If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances. | [ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (42.20%) If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances. |
| Analysis | r0 | r1 | r2 | r3 | |
|---|---|---|---|---|---|
| Loop Computation Issues | Presence of expensive FP instructions | 1 | 1 | 2 | 2 |
| Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA | 1 | 1 | 1 | 0 | |
| Presence of a large number of scalar integer instructions | 3 | 2 | 2 | 3 | |
| Low iteration count | 0 | 0 | 2 | 0 | |
| Control Flow Issues | Presence of calls | 1 | 2 | 1 | 2 |
| Presence of 2 to 4 paths | 1 | 1 | 2 | 2 | |
| Presence of more than 4 paths | 1 | 1 | 1 | 2 | |
| Non-innermost loop | 2 | 2 | 1 | 2 | |
| Low iteration count | 0 | 0 | 2 | 0 | |
| Data Access Issues | Presence of constant non-unit stride data access | 4 | 4 | 4 | 3 |
| Presence of indirect access | 2 | 2 | 2 | 2 | |
| More than 10% of the vector loads instructions are unaligned | 1 | 1 | 2 | 2 | |
| Presence of expensive instructions: scatter/gather | 0 | 0 | 1 | 0 | |
| Presence of special instructions executing on a single port | 1 | 2 | 2 | 3 | |
| More than 20% of the loads are accessing the stack | 2 | 3 | 1 | 3 | |
| Vectorization Roadblocks | Presence of calls | 1 | 2 | 1 | 2 |
| Presence of 2 to 4 paths | 1 | 1 | 2 | 2 | |
| Presence of more than 4 paths | 2 | 2 | 2 | 4 | |
| Non-innermost loop | 2 | 2 | 1 | 2 | |
| Presence of constant non-unit stride data access | 4 | 4 | 4 | 3 | |
| Presence of indirect access | 2 | 2 | 2 | 2 | |
| Inefficient Vectorization | Presence of expensive instructions: scatter/gather | 0 | 0 | 1 | 0 |
| Presence of special instructions executing on a single port | 1 | 2 | 2 | 3 | |
| Use of masked instructions | 0 | 0 | 0 | 1 | |