[OS-BUILD PATCH] redhat: configs: Disable TPM 1.2 specific drivers

From: Jerry Snitselaar redhat: configs: Disable TPM 1.2 specific drivers TPM 1.2 is not supported in RHEL9, so disable the TPM 1.2 specific drivers. Signed-off-by: Jerry Snitselaar diff --git a/redhat/configs/ark/generic/CONFIG_TCG_ATMEL b/redhat/configs/ark/generic/CONFIG_TCG_ATMEL new

[OS-BUILD PATCHv3 6/7] redhat/Makefile.variables: Extend git hash length to 15

From: Prarit Bhargava redhat/Makefile.variables: Extend git hash length to 15 While testing the dist-self-test-data target, dzickus noticed that the git hash lengths were 13 and not 12 on his output. After discussion, and to avoid future problems, we have decided to default the hash length to

[OS-BUILD PATCHv3 4/7] redhat/genspec.sh: Rearrange genspec.sh

From: Prarit Bhargava redhat/genspec.sh: Rearrange genspec.sh genspec.sh takes a long time to run when executed to create test examples. The script does three things: creates a changelog, creates a patch file, and creates the specfile. Reorganize the code so that the specfile is created

[OS-BUILD PATCHv3 7/7] redhat/self-tests: Ignore RHGITURL

From: Prarit Bhargava redhat/self-tests: Ignore RHGITURL While testing, dzickus noticed that his RHGITURL was different from the RHGITURL in the default data. This is because RHGITURL is dependent on the user's local git configuration and therefore cannot be guaranteed to be unique across all

[OS-BUILD PATCHv3 0/7] redhat/self-test: Add new tests

From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1706 Making changes to the Makefiles has been difficult and revealed minor issues with even the simplest of changes. Having a test that compares the Makefile variables to a "known good"

[OS-BUILD PATCHv3 1/7] redhat/self-test: Add variable test data

From: Prarit Bhargava redhat/self-test: Add variable test data Add test data for various release types, a script to create the data, and a Makefile target. This data will be used as a baseline for testing of Makefile changes. Note: By default, 'make self-test-data' outputs absolute paths.

[OS-BUILD PATCHv3 5/7] redhat/self-test: Remove changelog from spec files

From: Prarit Bhargava redhat/self-test: Remove changelog from spec files Remove the changelog entries from spec files and replace it with a generic one line entry. We currently do not have changelog tests, and removing them at this time will make testing easier. Signed-off-by: Prarit Bhargava

[OS-BUILD PATCHv3 2/7] redhat/self-test: Add better dist-dump-variables test

From: Prarit Bhargava redhat/self-test: Add better dist-dump-variables test The current dist-dump-variables test is not adequate in that it only tests a small number of fields. Add a test to run create-data.sh and compare that data against the stored "known good" output in self-test/data.

[OS-BUILD PATCHv3 3/7] redhat/self-test: Add spec file data

From: Prarit Bhargava redhat/self-test: Add spec file data Add spec file examples for various release types and modify create-data.sh to create the data. This data will be used as a baseline for testing of Makefile changes. Signed-off-by: Prarit Bhargava diff --git

Re: [OS-BUILD PATCHv2 0/5] redhat/self-test: Add new tests

From: Don Zickus on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1706#note_891877949 I like the spirit of the approach. On the surface things make sense and don't look to intrusive. Unfortunately, playing with your MR caused test failures. The biggest issues were my

Re: Race found with parallel config processing

On Wed, Mar 23, 2022 at 03:56:21PM -0400, Prarit Bhargava wrote: > > Heh, after talking with jforbes about it this is probably just as fast > (sorry for the cut-and-paste) Yep, this for sure will work too :), but it looks like the first config is processed twice, which is IMHO not necessary. >

Re: Race found with parallel config processing

On Tue, Mar 22, 2022 at 06:48:17PM -0400, Prarit Bhargava wrote: > On 3/22/22 17:10, Frantisek Hrbata wrote: > > > > On Tue, Mar 22, 2022, 21:46 Prarit Bhargava > > wrote: > > > > On 3/22/22 15:41, Prarit Bhargava wrote: > > > On 3/22/22 15:37, Prarit Bhargava

Re: Race found with parallel config processing

On Tue, Mar 22, 2022, 21:46 Prarit Bhargava wrote: > On 3/22/22 15:41, Prarit Bhargava wrote: > > On 3/22/22 15:37, Prarit Bhargava wrote: > > > > > >> Thanks, I had thought that too but :( I get an error when running > >> 'make scripts-basic'? > >> > >> [prarit@prarit kernel-ark]$ make V=1

Re: Race found with parallel config processing

On Tue, Mar 22, 2022 at 12:55:48PM -0400, Prarit Bhargava wrote: > Hey everyone, > > This issue was found in centos-stream-9 when I backported the parallel > config processing code into that codebase. I can reproduce it on my desktop > but do not have enough Makefile-fu to figure out why this

[OS-BUILD PATCH 2/2] redhat/configs/process_configs.sh: Avoid race with find

From: Prarit Bhargava redhat/configs/process_configs.sh: Avoid race with find The find command interferes with other commands that get executed and results in errors like "find: './.tmp_6391763917': No such file or directory" being output during process_configs.sh. This can be avoided by only

[OS-BUILD PATCH 0/2] redhat/configs/process_configs.sh: Fix empty output error

From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1712 The CONTINUEONERROR code contained an 'exit 1' which brought the whole script to a halt when a config mismatch was detected. CONTINUEONERROR has no purpose in the parallel

[OS-BUILD PATCH 1/2] redhat/configs/process_configs.sh: Remove CONTINUEONERROR

From: Prarit Bhargava redhat/configs/process_configs.sh: Remove CONTINUEONERROR Back when process_configs.sh was serial, CONTINUEONERROR allowed the process_configs.sh script to continue to the next config even when there was a fatal error with the current analysis. This is no longer needed in

[OS-BUILD PATCHv2 1/5] redhat/self-test: Add variable test data

From: Prarit Bhargava redhat/self-test: Add variable test data Add test data for various release types, a script to create the data, and a Makefile target. This data will be used as a baseline for testing of Makefile changes. Note: By default, 'make self-test-data' outputs absolute paths.

Re: [OS-BUILD PATCHv2 0/5] redhat/self-test: Add new tests

From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1706#note_891772895 New version: Fixed comment about 5.16-rc5 + 2 commits ___ kernel mailing list -- kernel@lists.fedoraproject.org To unsubscribe send an

[OS-BUILD PATCHv2 0/5] redhat/self-test: Add new tests

From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1706 Making changes to the Makefiles has been difficult and revealed minor issues with even the simplest of changes. Having a test that compares the Makefile variables to a "known good"

[OS-BUILD PATCHv2 4/5] redhat/genspec.sh: Rearrange genspec.sh

From: Prarit Bhargava redhat/genspec.sh: Rearrange genspec.sh genspec.sh takes a long time to run when executed to create test examples. The script does three things: creates a changelog, creates a patch file, and creates the specfile. Reorganize the code so that the specfile is created

[OS-BUILD PATCHv2 5/5] redhat/self-test: Remove changelog from spec files

From: Prarit Bhargava redhat/self-test: Remove changelog from spec files Remove the changelog entries from spec files and replace it with a generic one line entry. We currently do not have changelog tests, and removing them at this time will make testing easier. Signed-off-by: Prarit Bhargava

[OS-BUILD PATCHv2 2/5] redhat/self-test: Add better dist-dump-variables test

From: Prarit Bhargava redhat/self-test: Add better dist-dump-variables test The current dist-dump-variables test is not adequate in that it only tests a small number of fields. Add a test to run create-data.sh and compare that data against the stored "known good" output in self-test/data.

[OS-BUILD PATCHv2 3/5] redhat/self-test: Add spec file data

From: Prarit Bhargava redhat/self-test: Add spec file data Add spec file examples for various release types and modify create-data.sh to create the data. This data will be used as a baseline for testing of Makefile changes. Signed-off-by: Prarit Bhargava diff --git

[OS-BUILD PATCH 0/0] [redhat] New configs in lib/Kconfig

From: CKI Gitlab on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1709 NOTE: Truncated patchset due to missing public @redhat.com email address on your GitLab profile at https://gitlab.com/-/profile. Once that is fixed, close and reopen the merge

[OS-BUILD PATCHv2] redhat/configs: Set CONFIG_X86_AMD_PSTATE built-in on Fedora

From: Prarit Bhargava redhat/configs: Set CONFIG_X86_AMD_PSTATE built-in on Fedora In order to use the amd-pstate driver on AMD systems, users must first unload the acpi-cpufreq driver and then load the amd-pstate driver. Another option is to denylist the acpi-cpufreq driver but this also

Re: [OS-BUILD PATCHv6 0/0] Add Partner Supported taint flag

From: Justin M. Forbes on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891563350 That should do it. ___ kernel mailing list -- kernel@lists.fedoraproject.org To unsubscribe send an email to

Re: [OS-BUILD PATCHv6 0/0] Add Partner Supported taint flag

From: Brian Foster on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891548412 Ok. It seems I'm not able to ack this because of unresolved threads, but I don't see a resolve threads button in my view. Did your previous comment create a thread that you're able to

[OS-BUILD PATCH 4/5] redhat/genspec.sh: Rearrange genspec.sh

From: Prarit Bhargava redhat/genspec.sh: Rearrange genspec.sh genspec.sh takes a long time to run when executed to create test examples. The script does three things: creates a changelog, creates a patch file, and creates the specfile. Reorganize the code so that the specfile is created

[OS-BUILD PATCH 3/5] redhat/self-test: Add spec file data

From: Prarit Bhargava redhat/self-test: Add spec file data Add spec file examples for various release types and modify create-data.sh to create the data. This data will be used as a baseline for testing of Makefile changes. Signed-off-by: Prarit Bhargava diff --git

[OS-BUILD PATCH 5/5] redhat/self-test: Remove changelog from spec files

From: Prarit Bhargava redhat/self-test: Remove changelog from spec files Remove the changelog entries from spec files and replace it with a generic one line entry. We currently do not have changelog tests, and removing them at this time will make testing easier. Signed-off-by: Prarit Bhargava

[OS-BUILD PATCH 1/5] redhat/self-test: Add variable test data

From: Prarit Bhargava redhat/self-test: Add variable test data Add test data for various release types, a script to create the data, and a Makefile target. This data will be used as a baseline for testing of Makefile changes. Note: By default, 'make self-test-data' outputs absolute paths.

[OS-BUILD PATCH 2/5] redhat/self-test: Add better dist-dump-variables test

From: Prarit Bhargava redhat/self-test: Add better dist-dump-variables test The current dist-dump-variables test is not adequate in that it only tests a small number of fields. Add a test to run create-data.sh and compare that data against the stored "known good" output in self-test/data.

[OS-BUILD PATCH 0/5] redhat/self-test: Add new tests

From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1706 Making changes to the Makefiles has been difficult and revealed minor issues with even the simplest of changes. Having a test that compares the Makefile variables to a "known good"

Re: [OS-BUILD PATCHv6 0/0] Add Partner Supported taint flag

From: Justin M. Forbes on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891516554 I mean all of the RHELisms are ripped out every time I do a rebased fedora-5.x branch. This ends up being several times a year, and since os-build is rebased, I can't simply keep a

Re: [OS-BUILD PATCHv6 0/0] Add Partner Supported taint flag

From: Brian Foster on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891470421 Can you elaborate? Do you mean that you prefer the ifdef because it makes finding this code easier? Wouldn't we have to scan for callsites anyways given that the empty inlines already

[OS-BUILD PATCHv2] redhat/configs: enable CONFIG_INTEL_ISHTP_ECLITE

From: David Arcari redhat/configs: enable CONFIG_INTEL_ISHTP_ECLITE Enable the Intel ISHTP eclite controller Driver for RHEL. Signed-off-by: David Arcari diff --git a/redhat/configs/common/generic/x86/x86_64/CONFIG_INTEL_ISHTP_ECLITE

[OS-BUILD PATCHv2] redhat/config: Remove obsolete CONFIG_MFD_INTEL_PMT

From: David Arcari redhat/config: Remove obsolete CONFIG_MFD_INTEL_PMT This was replaced by CONFIG_INTEL_VSEC. Signed-off-by: David Arcari diff --git a/redhat/configs/common/generic/CONFIG_MFD_INTEL_PMT b/redhat/configs/common/generic/CONFIG_MFD_INTEL_PMT deleted file mode 100644 index

Re: [OS-BUILD PATCHv6 0/0] Add Partner Supported taint flag

From: Justin M. Forbes on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891415723 I would much rather it not. While things may build and function, it just makes it easier for patches like this to get lost on the Fedora stable rebases, where they are all ripped

Re: [OS-BUILD PATCHv6 0/0] Add Partner Supported taint flag

From: Brian Foster on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/1603#note_891384587 - [2/2] `4254c840c8f3` ("Add Partner Supported taint flag to kAFS") Thanks for the tweaks. One more nit.. the ifdef in this patch can go away now that patch 1 defines the empty