Re: [OS-BUILD PATCH] Add kasan_test to mod-internal.list

2022-10-13 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2092#note_1134900344 I was just testing this exact patch. Thanks :) ___ kernel mailing list -- kernel@lists.fedoraproject.org To unsubscribe send an email to

[OS-BUILD PATCH 3/8] redhat: Move mod-* scripts into their own directory

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Move mod-* scripts into their own directory The redhat/ dir is messy and this is just a cleanup. Move the mod-* scripts into their own directory. Signed-off-by: Prarit Bhargava diff --git a/redhat/Makefile b/redhat/Makefile index blahblah..blahblah 100644 --- a/

[OS-BUILD PATCH 7/8] redhat: Move yaml files into their own directory

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Move yaml files into their own directory The redhat/scripts dir is messy and this is just a cleanup. Move the yaml files into their own directory. Signed-off-by: Prarit Bhargava diff --git a/redhat/Makefile b/redhat/Makefile index blahblah..blahblah 100644 --- a

[OS-BUILD PATCH 8/8] redhat: Move cross-compile scripts into their own directory

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Move cross-compile scripts into their own directory The redhat/scripts dir is messy and this is just a cleanup. Move the cross-compile scripts into their own directory. Signed-off-by: Prarit Bhargava diff --git a/redhat/Makefile.cross b/redhat/Makefile.cross ind

[OS-BUILD PATCH 5/8] redhat: Move kernel-tools scripts into their own directory

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Move kernel-tools scripts into their own directory The redhat/ dir is messy and this is just a cleanup. Move the kernel-tools scripts into their own directory. Signed-off-by: Prarit Bhargava diff --git a/redhat/Makefile b/redhat/Makefile index blahblah..blahblah

[OS-BUILD PATCH 4/8] redhat: Move gen-* scripts into their own directory

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Move gen-* scripts into their own directory The redhat/ dir is messy and this is just a cleanup. Move the gen-* scripts, and the commit_template into their own directory. Signed-off-by: Prarit Bhargava diff --git a/redhat/Makefile b/redhat/Makefile index blahbla

[OS-BUILD PATCH 6/8] redhat: Move update_scripts.sh into redhat/scripts

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Move update_scripts.sh into redhat/scripts The redhat/scripts dir is messy and this is just a cleanup. Move upate_scripts.sh into redhat/scripts. Signed-off-by: Prarit Bhargava diff --git a/redhat/Makefile b/redhat/Makefile index blahblah..blahblah 100644 --- a/

[OS-BUILD PATCH 2/8] Rename rename FORCE_MAX_ZONEORDER to ARCH_FORCE_MAX_ORDER in configs

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Justin M. Forbes Rename rename FORCE_MAX_ZONEORDER to ARCH_FORCE_MAX_ORDER in configs Upstream commit 0192445cb2f7e renamed FORCE_MAX_ZONEORDER to ARCH_FORCE_MAX_ORDER which made it appear as a new config item for us in pending. While I have fixed up pending, this MR just fixes up the pro

[OS-BUILD PATCH 0/8] redhat: Cleanup scripts

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2095 The redhat/ directory contains a bunch of scripts. These can be moved around and cleaned up. Signed-off-by: Prarit Bhargava --- redhat/configs/common/generic/arm/aarch64/CONFIG_A

[OS-BUILD PATCH 1/8] Rename rename FORCE_MAX_ZONEORDER to ARCH_FORCE_MAX_ORDER in pending

2022-10-12 Thread Prarit Bhargava (via Email Bridge)
From: Justin M. Forbes Rename rename FORCE_MAX_ZONEORDER to ARCH_FORCE_MAX_ORDER in pending Upstream commit 0192445cb2f7e renamed FORCE_MAX_ZONEORDER to ARCH_FORCE_MAX_ORDER which appeared as a new config item set to defaults for us. Unfortunately those defaults do not match what is needed. Thi

Re: [OS-BUILD PATCHv2] redhat: Remove parallel_xz.sh

2022-10-11 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2068#note_1131692752 Fixed. ___ kernel mailing list -- kernel@lists.fedoraproject.org To unsubscribe send an email to kernel-le...@lists.fedoraproject.org Fedo

Re: [OS-BUILD PATCHv2] redhat: Remove parallel_xz.sh

2022-10-11 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2068#note_1131692324 Oops. Of course. Fixed. ___ kernel mailing list -- kernel@lists.fedoraproject.org To unsubscribe send an email to kernel-le...@lists.fed

[OS-BUILD PATCHv2] redhat: Remove parallel_xz.sh

2022-10-11 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Remove parallel_xz.sh The parallel_xz.sh script is no longer used. It is defined as Source22 but is never called. The script was removed from execution in ark commit e60c3381ad10 ("Temporarily backout parallel xz script") and never put back into execution. A qu

Re: [OS-BUILD PATCH] redhat/configs: change so watchdog is module versus builtin

2022-10-10 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2044#note_1130702666 Nope, I didn't mean to block this. I think your reply turned it into a discussion. ___ kernel mailing list -- kernel@lists.fedoraproject.

[OS-BUILD PATCH] redhat: Remove parallel_xz.sh

2022-10-07 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Remove parallel_xz.sh The parallel_xz.sh script is no longer used. It is defined as Source22 but is never called. The script was removed from execution in ark commit e60c3381ad10 ("Temporarily backout parallel xz script") and never put back into execution. A qu

Re: [OS-BUILD PATCH] redhat/configs: change so watchdog is module versus builtin

2022-10-06 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2044#note_1127243945 @sfbest, no objection but just a question: The reason the watchdog is typically built in is so that it is immediately active on boot. By making this a module, I'm wondering if this

Re: [OS-BUILD PATCHv2] redhat/configs: Move CONFIG_ARM_SMMU_QCOM_DEBUG to common

2022-09-30 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2035#note_1120881242 Thanks! ___ kernel mailing list -- kernel@lists.fedoraproject.org To unsubscribe send an email to kernel-le...@lists.fedoraproject.org Fed

[OS-BUILD PATCH] redhat/Makefile: Add DIST to git tags for RHEL

2022-09-30 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/Makefile: Add DIST to git tags for RHEL Bugzilla: INTERNAL Upstream Status: RHEL only RHEL repositories may have an extended DIST tag, such as .el9_0. Users looking for builds need a complete tag in dist-git and source-git in order to find a build's source tag. Ad

[OS-BUILD PATCH] RHMAINTAINERS: Update for Sept 30 2022

2022-09-30 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava RHMAINTAINERS: Update for Sept 30 2022 Update RHMAINTAINERS to latest. Signed-off-by: Prarit Bhargava diff --git a/redhat/rhdocs/MAINTAINERS/RHMAINTAINERS b/redhat/rhdocs/MAINTAINERS/RHMAINTAINERS index blahblah..blahblah 100644 --- a/redhat/rhdocs/MAINTAINERS/RHMAINTAI

Re: [OS-BUILD PATCH] redhat/configs: Move CONFIG_ARM_SMMU_QCOM_DEBUG to common

2022-09-29 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2035#note_1119704497 I don't think it's quite right. I think you want to rename it to the common arm directory: redhat/configs/common/generic/arm/CONFIG_ARM_SMMU_QCOM_DEBUG

[OS-BUILD PATCHv3 0/7] redhat/configs: Remove x86_64 directories

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2030 Remove the x86_64 directories and migrate to using x86 as the base directory. Signed-off-by: Prarit Bhargava --- redhat/configs/ark/debug/x86/x86_64/CONFIG_PM_SLEEP_DEBUG => redh

[OS-BUILD PATCHv3 7/7] redhat/configs: Remove x86_64 from priority files

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs: Remove x86_64 from priority files Now that the x86_64 directory has been removed, removed the directory from the priority files. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/priority.common b/redhat/configs/priority.common index blahblah..bl

[OS-BUILD PATCHv3 6/7] redhat/configs/pending-ark: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/pending-ark: Remove x86_64 directory Remove the x86_64 directories. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/pending-ark/debug/x86/x86_64/README b/redhat/configs/pending-ark/debug/x86/README rename from redhat/configs/pending-ark/debug/x

[OS-BUILD PATCHv3 5/7] redhat/configs/pending-fedora: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/pending-fedora: Remove x86_64 directory Remove the x86_64 directories. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/pending-fedora/generic/x86/x86_64/README b/redhat/configs/pending-fedora/generic/x86/README rename from redhat/configs/pendin

[OS-BUILD PATCHv3 4/7] redhat/configs/fedora: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/fedora: Remove x86_64 directory Remove the x86_64 directories. CONFIG_CROS_EC, CONFIG_INTEL_INT0002_VGPIO, CONFIG_INTEL_SOC_PMIC_CHTWC, CONFIG_MEMORY_HOTPLUG, and CONFIG_XO15_EBOOK existed in both x86 and x86_64. The value in the file now reflects the x86 6

[OS-BUILD PATCHv3 3/7] redhat/configs/common: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/common: Remove x86_64 directory Remove the x86_64 directories. CONFIG_EDAC_AMD64 existed in both x86 and x86_64. The value in the file now reflects the x86 64-bit variant. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/common/debug/x86/x86_64

[OS-BUILD PATCHv3 2/7] redhat/configs/ark: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/ark: Remove x86_64 directory Remove the x86_64 directories. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/ark/debug/x86/x86_64/CONFIG_PM_SLEEP_DEBUG b/redhat/configs/ark/debug/x86/CONFIG_PM_SLEEP_DEBUG rename from redhat/configs/ark/debug/x86

[OS-BUILD PATCHv3 1/7] redhat/configs/custom-overrides: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/custom-overrides: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: The READMEs are empty and can be removed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/custom-overrides/debug/x86/x86_64/README b/redhat/configs/cu

[OS-BUILD PATCHv2 8/8] redhat/configs: Remove README files

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs: Remove README files These README files served a purpose when we first moved to the "one config per file" format. They're no longer needed and can be removed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/ark/debug/README b/redhat/configs/ark

[OS-BUILD PATCHv2 7/8] redhat/configs: Remove x86_64 from priority files

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs: Remove x86_64 from priority files Now that the x86_64 directory has been removed, removed the directory from the priority files. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/priority.common b/redhat/configs/priority.common index blahblah..bl

[OS-BUILD PATCHv2 5/8] redhat/configs/pending-fedora: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/pending-fedora: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: Remove READMEs. They served a purpose when originally introduced but no longer are needed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/pending-fedor

[OS-BUILD PATCHv2 6/8] redhat/configs/pending-ark: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/pending-ark: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: Remove READMEs. They served a purpose when originally introduced but no longer are needed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/pending-ark/debu

[OS-BUILD PATCHv2 4/8] redhat/configs/fedora: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/fedora: Remove x86_64 directory Remove the x86_64 directories. CONFIG_CROS_EC, CONFIG_INTEL_INT0002_VGPIO, CONFIG_INTEL_SOC_PMIC_CHTWC, CONFIG_MEMORY_HOTPLUG, and CONFIG_XO15_EBOOK existed in both x86 and x86_64. The value in the file now reflects the x86 6

[OS-BUILD PATCHv2 2/8] redhat/configs/ark: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/ark: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: Remove READMEs. They served a purpose when originally introduced but no longer are needed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/ark/debug/x86/x86_64/CON

[OS-BUILD PATCHv2 3/8] redhat/configs/common: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/common: Remove x86_64 directory Remove the x86_64 directories. CONFIG_EDAC_AMD64 existed in both x86 and x86_64. The value in the file now reflects the x86 64-bit variant. Additional fix: Remove empty README files. Signed-off-by: Prarit Bhargava diff --

[OS-BUILD PATCHv2 1/8] redhat/configs/custom-overrides: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/custom-overrides: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: The READMEs are empty and can be removed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/custom-overrides/debug/x86/x86_64/README b/redhat/configs/cu

[OS-BUILD PATCH 7/8] redhat/configs: Remove x86_64 from priority files

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs: Remove x86_64 from priority files Now that the x86_64 directory has been removed, removed the directory from the priority files. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/priority.common b/redhat/configs/priority.common index blahblah..bl

[OS-BUILD PATCH 8/8] redhat/configs: Remove README files

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs: Remove README files These README files served a purpose when we first moved to the "one config per file" format. They're no longer needed and can be removed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/ark/debug/README b/redhat/configs/ark

[OS-BUILD PATCH 4/8] redhat/configs/fedora: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/fedora: Remove x86_64 directory Remove the x86_64 directories. CONFIG_CROS_EC, CONFIG_INTEL_INT0002_VGPIO, CONFIG_INTEL_SOC_PMIC_CHTWC, CONFIG_MEMORY_HOTPLUG, and CONFIG_XO15_EBOOK existed in both x86 and x86_64. The value in the file now reflects the x86 6

[OS-BUILD PATCH 6/8] redhat/configs/pending-ark: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/pending-ark: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: Remove READMEs. They served a purpose when originally introduced but no longer are needed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/pending-ark/debu

[OS-BUILD PATCH 5/8] redhat/configs/pending-fedora: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/pending-fedora: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: Remove READMEs. They served a purpose when originally introduced but no longer are needed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/pending-fedor

[OS-BUILD PATCH 3/8] redhat/configs/common: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/common: Remove x86_64 directory Remove the x86_64 directories. CONFIG_EDAC_AMD64 existed in both x86 and x86_64. The value in the file now reflects the x86 64-bit variant. Additional fix: Remove empty README files. Signed-off-by: Prarit Bhargava diff --

[OS-BUILD PATCH 1/8] redhat/configs/custom-overrides: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/custom-overrides: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: The READMEs are empty and can be removed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/custom-overrides/debug/x86/x86_64/README b/redhat/configs/cu

[OS-BUILD PATCH 2/8] redhat/configs/ark: Remove x86_64 directory

2022-09-27 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/configs/ark: Remove x86_64 directory Remove the x86_64 directories. Additional cleanup: Remove READMEs. They served a purpose when originally introduced but no longer are needed. Signed-off-by: Prarit Bhargava diff --git a/redhat/configs/ark/debug/x86/x86_64/CON

[OS-BUILD PATCHv5 18/18] redhat: Remove filter-i686.sh.rhel

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Remove filter-i686.sh.rhel This file is no longer used and can be deleted. Suggested by: Herton R. Krzesinski Signed-off-by: Prarit Bhargava diff --git a/redhat/rhel_files/filter-i686.sh.rhel b/redhat/rhel_files/filter-i686.sh.rhel deleted file mode 100644 inde

[OS-BUILD PATCHv5 16/18] redhat: remove GL_DISTGIT_USER, RHDISTGIT and unify dist-git cloning

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: remove GL_DISTGIT_USER, RHDISTGIT and unify dist-git cloning Tested: make DISTRO=centos dist-git-test 1) GL_DISTGIT_USER is used for private fork cloning. There is no reason to use private fork in dist-git cloning. First this would require to keep the fork in

[OS-BUILD PATCHv5 12/18] kabi: add stablelist helpers

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: add stablelist helpers Add helper scripts to update the kabi stablelist checksums and to diff symtypes output. redhat/kabi/diff-kabi computes the current symtypes files and compares it to the reference symtypes stored in the stablelist. redhat/kabi/update-kabi compu

[OS-BUILD PATCHv5 15/18] redhat/Makefile.variables: Add ADD_COMMITID_TO_VERSION

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/Makefile.variables: Add ADD_COMMITID_TO_VERSION As noted in ARK commit 40aae9752b8b ("Fix versioning on stable Fedora"), adding the commit ID to all versions must be optional. Currently, the centos-stream-9 repository does not have this feature and adding it would c

[OS-BUILD PATCHv5 13/18] perf: Require libbpf 0.6.0 or newer

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava perf: Require libbpf 0.6.0 or newer With libbpf 0.5 or older, feature-detection mechanism in tools/build silently fails for some functions (namely bpf_object__next_map() and bpf_object__next_program()), so fallbacks are used and that leads to endless loop. Make the problem

[OS-BUILD PATCHv5 14/18] kernel.spec: disable vmlinux.h generation for s390 zfcpdump config

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kernel.spec: disable vmlinux.h generation for s390 zfcpdump config With the linux upstream commit d9847eb8be3d ("bpf: Make CONFIG_DEBUG_INFO_BTF depend upon CONFIG_BPF_SYSCALL") zfcpdump does not have btf anymore (it does not enable CONFIG_BPF_SYSCALL), so the generation fa

[OS-BUILD PATCHv5 11/18] Makefile: add kabi targets

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava Makefile: add kabi targets Add make wrappers on redhat/kabi/{diff,update}-kabi and redhat/kabi/symtypes: - dist-kabi-update updates the stablelist checksums - dist-symtype-diff diffs the current state against a reference stored in the stablelist - dist-kabi-image-N

[OS-BUILD PATCHv5 9/18] kabi: ignore new stablelist metadata in show-kabi

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: ignore new stablelist metadata in show-kabi Stablelist symbol definition shall be extended to include extra metadata, such as a reference symtypes file, path to a file exporting given symbol, and version history. Reference and path information is not required by show

[OS-BUILD PATCHv5 8/18] redhat/Makefile: add dist-assert-tree-clean target

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/Makefile: add dist-assert-tree-clean target Add dist-assert-tree-clean target that fails whenever staged/unstaged changes or untracked files are present in the git tree. Manual invocation of this target is not expected, thus left out of dist-help. It's intended to a

[OS-BUILD PATCHv5 10/18] kabi: add support for symbol namespaces into check-kabi

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: add support for symbol namespaces into check-kabi Symbol namespaces are emitted to Module.symvers as well as any Module.kabi files that check-kabi consumes. Correctly parse symbol namespaces, if present. Warn on symbol namespace changes, as they constitute an ABI bre

[OS-BUILD PATCHv5 7/18] redhat/kernel.spec.template: Specify vmlinux.h path when building samples/bpf

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: Specify vmlinux.h path when building samples/bpf Bugzilla: https://bugzilla.redhat.com/2041365 Upstream status: RHEL only. Since the backport of commit 384b6b3bbf0d ("samples: bpf: Add vmlinux.h generation support"), building samples/bpf start

[OS-BUILD PATCHv5 2/18] redhat/kernel.spec.template: Link perf with --export-dynamic

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: Link perf with --export-dynamic Bugzilla: INTERNAL commit 00702a636e319b2c3fe913f3b8a8cff51cf96ff6 Author: Michael Petlan Date: Tue Nov 23 11:52:24 2021 +0100 redhat/kernel.spec.template: Link perf with --export-dynamic Bugzilla: h

[OS-BUILD PATCHv5 5/18] redhat/scripts: Update merge-subtrees.sh with new subtree location

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/scripts: Update merge-subtrees.sh with new subtree location Update the maintainer's merge-subtrees.sh script with the location of the new redhat/rhdocs subtree. Signed-off-by: Prarit Bhargava diff --git a/redhat/scripts/merge-subtrees.sh b/redhat/scripts/merge-sub

[OS-BUILD PATCHv5 1/18] redhat: kernel.spec: selftests: abort on build failure

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: kernel.spec: selftests: abort on build failure When a selftest does not compile, abort the whole build. We want to know that we broke selftests by a backport. This commit is forward ported from RHEL 8. It could not have been forward ported to ARK: with rebases to t

[OS-BUILD PATCHv5 4/18] redhat/kernel.spec.template: enable dependencies generation

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: enable dependencies generation Updated kernel-rpm-macros-185-9 package brings various generators for Provides tags that usually present in RHEL kernel packages, including kernel(), which OOT kmod packages rely on. prarit: After discussion with

[OS-BUILD PATCHv5 3/18] redhat: build and include memfd to kernel-selftests-internal

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: build and include memfd to kernel-selftests-internal Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2027506 Upstream Status: RHEL9 only Tested: by reporter This patch adds memfd to the list of selftests built and packaged in kernel-selftests-internal. Signe

[OS-BUILD PATCHv5 6/18] spec: Fix separate tools build

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava spec: Fix separate tools build Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2054579 Upstream Status: RHEL only If we re building only tools without kernel, we need to generate config headers and prepare tree for modules building. The modules_prepare target will co

[OS-BUILD PATCHv5 0/18] redhat: Update with RHEL-only changes

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2021 centos-stream-9 !1308 [1] updates the redhat/ directory in centos-stream-9 to match that of ARK. After that update there are about 40 changes that are not applied to ARK but have bee

[OS-BUILD PATCHv4 17/19] redhat: remove GL_DISTGIT_USER, RHDISTGIT and unify dist-git cloning

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: remove GL_DISTGIT_USER, RHDISTGIT and unify dist-git cloning Tested: make DISTRO=centos dist-git-test 1) GL_DISTGIT_USER is used for private fork cloning. There is no reason to use private fork in dist-git cloning. First this would require to keep the fork in

[OS-BUILD PATCHv4 19/19] redhat: Remove filter-i686.sh.rhel

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Remove filter-i686.sh.rhel This file is no longer used and can be deleted. Suggested by: Herton R. Krzesinski Signed-off-by: Prarit Bhargava diff --git a/redhat/rhel_files/filter-i686.sh.rhel b/redhat/rhel_files/filter-i686.sh.rhel deleted file mode 100644 inde

[OS-BUILD PATCHv4 13/19] kabi: add stablelist helpers

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: add stablelist helpers Add helper scripts to update the kabi stablelist checksums and to diff symtypes output. redhat/kabi/diff-kabi computes the current symtypes files and compares it to the reference symtypes stored in the stablelist. redhat/kabi/update-kabi compu

[OS-BUILD PATCHv4 12/19] Makefile: add kabi targets

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava Makefile: add kabi targets Add make wrappers on redhat/kabi/{diff,update}-kabi and redhat/kabi/symtypes: - dist-kabi-update updates the stablelist checksums - dist-symtype-diff diffs the current state against a reference stored in the stablelist - dist-kabi-image-N

[OS-BUILD PATCHv4 16/19] redhat/Makefile.variables: Add ADD_COMMITID_TO_VERSION

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/Makefile.variables: Add ADD_COMMITID_TO_VERSION As noted in ARK commit 40aae9752b8b ("Fix versioning on stable Fedora"), adding the commit ID to all versions must be optional. Currently, the centos-stream-9 repository does not have this feature and adding it would c

[OS-BUILD PATCHv4 15/19] kernel.spec: disable vmlinux.h generation for s390 zfcpdump config

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kernel.spec: disable vmlinux.h generation for s390 zfcpdump config With the linux upstream commit d9847eb8be3d ("bpf: Make CONFIG_DEBUG_INFO_BTF depend upon CONFIG_BPF_SYSCALL") zfcpdump does not have btf anymore (it does not enable CONFIG_BPF_SYSCALL), so the generation fa

[OS-BUILD PATCHv4 14/19] perf: Require libbpf 0.6.0 or newer

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava perf: Require libbpf 0.6.0 or newer With libbpf 0.5 or older, feature-detection mechanism in tools/build silently fails for some functions (namely bpf_object__next_map() and bpf_object__next_program()), so fallbacks are used and that leads to endless loop. Make the problem

[OS-BUILD PATCHv4 11/19] kabi: add support for symbol namespaces into check-kabi

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: add support for symbol namespaces into check-kabi Symbol namespaces are emitted to Module.symvers as well as any Module.kabi files that check-kabi consumes. Correctly parse symbol namespaces, if present. Warn on symbol namespace changes, as they constitute an ABI bre

[OS-BUILD PATCHv4 9/19] redhat/Makefile: add dist-assert-tree-clean target

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/Makefile: add dist-assert-tree-clean target Add dist-assert-tree-clean target that fails whenever staged/unstaged changes or untracked files are present in the git tree. Manual invocation of this target is not expected, thus left out of dist-help. It's intended to a

[OS-BUILD PATCHv4 10/19] kabi: ignore new stablelist metadata in show-kabi

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: ignore new stablelist metadata in show-kabi Stablelist symbol definition shall be extended to include extra metadata, such as a reference symtypes file, path to a file exporting given symbol, and version history. Reference and path information is not required by show

[OS-BUILD PATCHv4 6/19] redhat/scripts: Update merge-subtrees.sh with new subtree location

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/scripts: Update merge-subtrees.sh with new subtree location Update the maintainer's merge-subtrees.sh script with the location of the new redhat/rhdocs subtree. Signed-off-by: Prarit Bhargava diff --git a/redhat/scripts/merge-subtrees.sh b/redhat/scripts/merge-sub

[OS-BUILD PATCHv4 7/19] spec: Fix separate tools build

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava spec: Fix separate tools build Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2054579 Upstream Status: RHEL only If we re building only tools without kernel, we need to generate config headers and prepare tree for modules building. The modules_prepare target will co

[OS-BUILD PATCHv4 8/19] redhat/kernel.spec.template: Specify vmlinux.h path when building samples/bpf

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: Specify vmlinux.h path when building samples/bpf Bugzilla: https://bugzilla.redhat.com/2041365 Upstream status: RHEL only. Since the backport of commit 384b6b3bbf0d ("samples: bpf: Add vmlinux.h generation support"), building samples/bpf start

[OS-BUILD PATCHv4 5/19] redhat/kernel.spec.template: enable dependencies generation

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: enable dependencies generation Updated kernel-rpm-macros-185-9 package brings various generators for Provides tags that usually present in RHEL kernel packages, including kernel(), which OOT kmod packages rely on. prarit: After discussion with

[OS-BUILD PATCHv4 2/19] redhat: kernel.spec: selftests: abort on build failure

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: kernel.spec: selftests: abort on build failure When a selftest does not compile, abort the whole build. We want to know that we broke selftests by a backport. This commit is forward ported from RHEL 8. It could not have been forward ported to ARK: with rebases to t

[OS-BUILD PATCHv4 3/19] redhat/kernel.spec.template: Link perf with --export-dynamic

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: Link perf with --export-dynamic Bugzilla: INTERNAL commit 00702a636e319b2c3fe913f3b8a8cff51cf96ff6 Author: Michael Petlan Date: Tue Nov 23 11:52:24 2021 +0100 redhat/kernel.spec.template: Link perf with --export-dynamic Bugzilla: h

[OS-BUILD PATCHv4 4/19] redhat: build and include memfd to kernel-selftests-internal

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: build and include memfd to kernel-selftests-internal Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2027506 Upstream Status: RHEL9 only Tested: by reporter This patch adds memfd to the list of selftests built and packaged in kernel-selftests-internal. Signe

[OS-BUILD PATCHv4 1/19] redhat/self-test: Ignore .rhpkg.mk files

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/self-test: Ignore .rhpkg.mk files If a user specifies, for example, a ~/.rhpkg.mk file, the file will be included in the MAKEFILE_LIST variable. This breaks the dist-dump-variables test, so these files should be excluded. Ignore user specified .rhpkg.mk files. Sig

[OS-BUILD PATCHv4 0/19] redhat: Update with RHEL-only changes

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com Merge Request: https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2021 centos-stream-9 !1308 [1] updates the redhat/ directory in centos-stream-9 to match that of ARK. After that update there are about 40 changes that are not applied to ARK but have bee

Re: [OS-BUILD PATCHv3 0/19] redhat: Update with RHEL-only changes

2022-09-22 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2021#note_083138 TBH, I wasn't too keen on the name either :) SPECSELFTESTS_MUST_BUILD. The SPEC must be added to identify it as being used directly in the specfile. Fixed.

Re: [OS-BUILD PATCH] redhat: set LC_ALL=C before sorting config content

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava on gitlab.com https://gitlab.com/cki-project/kernel-ark/-/merge_requests/2011#note_1105800491 Somewhere in this MR it was suggested that we just include a dependency on the locale files. I agree with that approach. It only adds 6M to a 40G buildroot. It's the easiest thing

[OS-BUILD PATCHv3 17/19] redhat: remove GL_DISTGIT_USER, RHDISTGIT and unify dist-git cloning

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: remove GL_DISTGIT_USER, RHDISTGIT and unify dist-git cloning Tested: make DISTRO=centos dist-git-test 1) GL_DISTGIT_USER is used for private fork cloning. There is no reason to use private fork in dist-git cloning. First this would require to keep the fork in

[OS-BUILD PATCHv3 19/19] redhat: Remove filter-i686.sh.rhel

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: Remove filter-i686.sh.rhel This file is no longer used and can be deleted. Suggested by: Herton R. Krzesinski Signed-off-by: Prarit Bhargava diff --git a/redhat/rhel_files/filter-i686.sh.rhel b/redhat/rhel_files/filter-i686.sh.rhel deleted file mode 100644 inde

[OS-BUILD PATCHv3 16/19] redhat/Makefile.variables: Add ADD_COMMITID_TO_VERSION

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/Makefile.variables: Add ADD_COMMITID_TO_VERSION As noted in ARK commit 40aae9752b8b ("Fix versioning on stable Fedora"), adding the commit ID to all versions must be optional. Currently, the centos-stream-9 repository does not have this feature and adding it would c

[OS-BUILD PATCHv3 13/19] kabi: add stablelist helpers

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: add stablelist helpers Add helper scripts to update the kabi stablelist checksums and to diff symtypes output. redhat/kabi/diff-kabi computes the current symtypes files and compares it to the reference symtypes stored in the stablelist. redhat/kabi/update-kabi compu

[OS-BUILD PATCHv3 15/19] kernel.spec: disable vmlinux.h generation for s390 zfcpdump config

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kernel.spec: disable vmlinux.h generation for s390 zfcpdump config With the linux upstream commit d9847eb8be3d ("bpf: Make CONFIG_DEBUG_INFO_BTF depend upon CONFIG_BPF_SYSCALL") zfcpdump does not have btf anymore (it does not enable CONFIG_BPF_SYSCALL), so the generation fa

[OS-BUILD PATCHv3 14/19] perf: Require libbpf 0.6.0 or newer

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava perf: Require libbpf 0.6.0 or newer With libbpf 0.5 or older, feature-detection mechanism in tools/build silently fails for some functions (namely bpf_object__next_map() and bpf_object__next_program()), so fallbacks are used and that leads to endless loop. Make the problem

[OS-BUILD PATCHv3 12/19] Makefile: add kabi targets

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava Makefile: add kabi targets Add make wrappers on redhat/kabi/{diff,update}-kabi and redhat/kabi/symtypes: - dist-kabi-update updates the stablelist checksums - dist-symtype-diff diffs the current state against a reference stored in the stablelist - dist-kabi-image-N

[OS-BUILD PATCHv3 9/19] redhat/Makefile: add dist-assert-tree-clean target

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/Makefile: add dist-assert-tree-clean target Add dist-assert-tree-clean target that fails whenever staged/unstaged changes or untracked files are present in the git tree. Manual invocation of this target is not expected, thus left out of dist-help. It's intended to a

[OS-BUILD PATCHv3 11/19] kabi: add support for symbol namespaces into check-kabi

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: add support for symbol namespaces into check-kabi Symbol namespaces are emitted to Module.symvers as well as any Module.kabi files that check-kabi consumes. Correctly parse symbol namespaces, if present. Warn on symbol namespace changes, as they constitute an ABI bre

[OS-BUILD PATCHv3 10/19] kabi: ignore new stablelist metadata in show-kabi

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava kabi: ignore new stablelist metadata in show-kabi Stablelist symbol definition shall be extended to include extra metadata, such as a reference symtypes file, path to a file exporting given symbol, and version history. Reference and path information is not required by show

[OS-BUILD PATCHv3 8/19] redhat/kernel.spec.template: Specify vmlinux.h path when building samples/bpf

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: Specify vmlinux.h path when building samples/bpf Bugzilla: https://bugzilla.redhat.com/2041365 Upstream status: RHEL only. Since the backport of commit 384b6b3bbf0d ("samples: bpf: Add vmlinux.h generation support"), building samples/bpf start

[OS-BUILD PATCHv3 7/19] spec: Fix separate tools build

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava spec: Fix separate tools build Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2054579 Upstream Status: RHEL only If we re building only tools without kernel, we need to generate config headers and prepare tree for modules building. The modules_prepare target will co

[OS-BUILD PATCHv3 6/19] redhat/scripts: Update merge-subtrees.sh with new subtree location

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/scripts: Update merge-subtrees.sh with new subtree location Update the maintainer's merge-subtrees.sh script with the location of the new redhat/rhdocs subtree. Signed-off-by: Prarit Bhargava diff --git a/redhat/scripts/merge-subtrees.sh b/redhat/scripts/merge-sub

[OS-BUILD PATCHv3 5/19] redhat/kernel.spec.template: enable dependencies generation

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: enable dependencies generation Updated kernel-rpm-macros-185-9 package brings various generators for Provides tags that usually present in RHEL kernel packages, including kernel(), which OOT kmod packages rely on. prarit: After discussion with

[OS-BUILD PATCHv3 3/19] redhat/kernel.spec.template: Link perf with --export-dynamic

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/kernel.spec.template: Link perf with --export-dynamic Bugzilla: INTERNAL commit 00702a636e319b2c3fe913f3b8a8cff51cf96ff6 Author: Michael Petlan Date: Tue Nov 23 11:52:24 2021 +0100 redhat/kernel.spec.template: Link perf with --export-dynamic Bugzilla: h

[OS-BUILD PATCHv3 4/19] redhat: build and include memfd to kernel-selftests-internal

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat: build and include memfd to kernel-selftests-internal Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2027506 Upstream Status: RHEL9 only Tested: by reporter This patch adds memfd to the list of selftests built and packaged in kernel-selftests-internal. Signe

[OS-BUILD PATCHv3 1/19] redhat/self-test: Ignore .rhpkg.mk files

2022-09-19 Thread Prarit Bhargava (via Email Bridge)
From: Prarit Bhargava redhat/self-test: Ignore .rhpkg.mk files If a user specifies, for example, a ~/.rhpkg.mk file, the file will be included in the MAKEFILE_LIST variable. This breaks the dist-dump-variables test, so these files should be excluded. Ignore user specified .rhpkg.mk files. Sig

<    1   2   3   4   5   6   7   8   9   10   >