[ 4 / 4 ] Application profile is long enough (155.99 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 1.93 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 35.71% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.93 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.04 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Application profile is long enough (78.27 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 1.84 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 38.71% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.84 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.04 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Application profile is long enough (39.79 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 1.92 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 36.00% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.92 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.05 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Application profile is long enough (20.84 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 1.89 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 37.04% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.89 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.05 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Application profile is long enough (11.70 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 1.74 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 42.11% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.74 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.05 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 0 / 4 ] Application profile is too short (9.13 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.

[ 1.70 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 43.33% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.70 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.03 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (88.08%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] Threads activity is good

On average, more than 2014.85% of observed threads are actually active

[ 4 / 4 ] CPU activity is good

CPU cores are active 92.99% of time

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (80.95%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (86.81%)

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.

[ 4 / 4 ] Affinity is good (96.36%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 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 (10.84%). 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

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.27%) lower than cumulative innermost loop coverage (86.81%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 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% (0.14%) is spend in Libm/SVML (special functions)

[ 0 / 4 ] Application profile is too short (7.71 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.

[ 2.11 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 29.63% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 2.11 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.03 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (85.93%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] Threads activity is good

On average, more than 2580.71% of observed threads are actually active

[ 4 / 4 ] CPU activity is good

CPU cores are active 90.92% of time

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (83.39%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (84.86%)

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.

[ 4 / 4 ] Affinity is good (95.71%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 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 (13.33%). 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

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.07%) lower than cumulative innermost loop coverage (84.86%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 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% (0.19%) is spend in Libm/SVML (special functions)

[ 0 / 4 ] Application profile is too short (7.27 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.

[ 1.62 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 46.15% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.62 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.03 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (83.29%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] Threads activity is good

On average, more than 3128.86% of observed threads are actually active

[ 3 / 4 ] CPU activity is below 90% (88.82%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (80.93%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (82.22%)

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.

[ 4 / 4 ] Affinity is good (95.43%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 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 (14.87%). 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

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.07%) lower than cumulative innermost loop coverage (82.22%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 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% (0.17%) is spend in Libm/SVML (special functions)

[ 0 / 4 ] Application profile is too short (7.08 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.

[ 1.78 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 40.74% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.78 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.02 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (81.41%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] Threads activity is good

On average, more than 3707.29% of observed threads are actually active

[ 3 / 4 ] CPU activity is below 90% (87.86%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (79.21%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (80.46%)

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.

[ 4 / 4 ] Affinity is good (95.26%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 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 (14.00%). 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

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (0.95%) lower than cumulative innermost loop coverage (80.46%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 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% (0.14%) is spend in Libm/SVML (special functions)

[ 0 / 4 ] Application profile is too short (6.82 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.

[ 1.67 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 44.44% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.67 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (80.09%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] Threads activity is good

On average, more than 4251.69% of observed threads are actually active

[ 3 / 4 ] CPU activity is below 90% (86.53%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (74.39%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (79.08%)

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.

[ 4 / 4 ] Affinity is good (95.05%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 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 (14.27%). 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

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.01%) lower than cumulative innermost loop coverage (79.08%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 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% (0.15%) is spend in Libm/SVML (special functions)

[ 0 / 4 ] Application profile is too short (6.51 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.

[ 1.40 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 53.33% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 3 / 3 ] Optimization level option is correctly used

[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.

Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.

[ 1.40 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.02 % 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

[ 1 / 1 ] Lstopo present. The Topology lstopo report will be generated.

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (79.07%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] Threads activity is good

On average, more than 4707.64% of observed threads are actually active

[ 3 / 4 ] CPU activity is below 90% (84.28%)

CPU cores are idle more than 10% of time. Threads supposed to run on these cores are probably IO/sync waiting. Some hints: use faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (76.85%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (78.04%)

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.

[ 4 / 4 ] Affinity is good (94.80%)

Threads are not migrating to CPU cores: probably successfully pinned

[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 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 (14.40%). 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

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.02%) lower than cumulative innermost loop coverage (78.04%)

Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop optimization more complex

[ 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% (0.16%) is spend in Libm/SVML (special functions)