Profile-guided optimization (PGO), also known as feedback-directed optimization (FDO), is a compiler optimization technique that feeds information (a profile) from representative runs of the application back into to the compiler for the next build of the application, which uses that information to make more informed optimization decisions. For example, the compiler may decide to more aggressively inline functions which the profile indicates are called frequently.
The profiling that is happening in the CI workflow:
|
echo "Running profile workload 1: SFV..." |
|
echo "Running profile workload 2: Validate..." |
|
echo "Running profile workload 3: ZIP..." |
Does not always produce valid profile and breaks the build. I suspect the workload is not big enough. This needs to be improved/fixed.
Profile-guided optimization (PGO), also known as feedback-directed optimization (FDO), is a compiler optimization technique that feeds information (a profile) from representative runs of the application back into to the compiler for the next build of the application, which uses that information to make more informed optimization decisions. For example, the compiler may decide to more aggressively inline functions which the profile indicates are called frequently.
The profiling that is happening in the CI workflow:
sfvbrr/.github/workflows/release.yml
Line 132 in 61e3199
sfvbrr/.github/workflows/release.yml
Line 135 in 61e3199
sfvbrr/.github/workflows/release.yml
Line 156 in 61e3199
Does not always produce valid profile and breaks the build. I suspect the workload is not big enough. This needs to be improved/fixed.