Date
1 - 3 of 3
Split up MdeModulePkg and CryptoPkg CodeQL CI checks
|
Michael D Kinney
Hi Michael,
I am looking at the CI agent execution times since we resolved several issued the last few days. We have CodeQL running in GitHub Actions and the rest of the Checks running in Azure Pipelines. Looking at only CodeQL, I see that the 2 longest jobs are on the MdeModulePkg and CryptoPkg and they build for both IA32 and X64 to get 32-bit and 64-bit coverage. * stuart_ci_build -c .pytool/CISettings.py -p CryptoPkg -t DEBUG -a IA32,X64 TOOL_CHAIN_TAG=VS2019 * stuart_ci_build -c .pytool/CISettings.py -p MdeModulePkg -t DEBUG -a IA32,X64 TOOL_CHAIN_TAG=VS2019 Is it possible to split these two jobs into 4 jobs to check IA32 and X64 independently? * stuart_ci_build -c .pytool/CISettings.py -p CryptoPkg -t DEBUG -a IA32 TOOL_CHAIN_TAG=VS2019 * stuart_ci_build -c .pytool/CISettings.py -p CryptoPkg -t DEBUG -a X64 TOOL_CHAIN_TAG=VS2019 * stuart_ci_build -c .pytool/CISettings.py -p MdeModulePkg -t DEBUG -a IA32 TOOL_CHAIN_TAG=VS2019 * stuart_ci_build -c .pytool/CISettings.py -p MdeModulePkg -t DEBUG -a X64 TOOL_CHAIN_TAG=VS2019 Thanks, Mike |
|
|
|
Michael D Kinney
Hi Michael,
toggle quoted message
Show quoted text
On the Azure Piplines side, the same 2 packages have the longest build time. The DEBUG and RELEASE builds take about the same time each. The NO-TARGET and NOOPT combined are less than DEBUG or RELEASE. stuart_ci_build -c .pytool/CISettings.py -p CryptoPkg -t DEBUG,RELEASE,NO-TARGET,NOOPT -a IA32,X64,ARM,AARCH64,RISCV64,LOONGARCH64 TOOL_CHAIN_TAG=GCC5 Could we split the CryptoPkg and MdeModulePkg jobs into 3 jobs each? * DEBUG * RELEASE * NO-TARGET, NOOPT Alternatively, split into 2 jobs * DEBUG, NO-TARGET * RELEASE, NOOPT Thanks, Mike -----Original Message----- |
|
|
|
Michael Kubacki
Hi Mike,
toggle quoted message
Show quoted text
The changes in this PR look reasonable https://github.com/tianocore/edk2/pull/4233 if you'd like to send a patch. Thanks, Michael On 4/1/2023 6:53 PM, Kinney, Michael D wrote:
Hi Michael, |
|
|