| Message ID | 20240811-aer-inject-v1-1-9cee7e3bd9b2@gmx.net |
|---|---|
| State | Accepted |
| Headers | show |
| Series | package/aer-inject: Fix build error due to missing basename() prototype | expand |
Hello J, Thanks a lot for your patch, which I have applied. I just have a few very, very minor suggestions to further improve your next patches. First of all, I personally like that the commit title starts with a lower-case letter, i.e: "package/aer-inject: fix ..." rather than "package/aer-inject: Fix ...". On Sun, 11 Aug 2024 12:34:25 +0200 J. Neuschäfer via buildroot <buildroot@buildroot.org> wrote: > Fixes several build failures such as: Please write just "Fixes:". This way, patchwork detects it's a patch fixes something, and it flags it as such. See https://patchwork.ozlabs.org/project/buildroot/list/, where the column A/F/R/T flags the patches that have a Acked-by:, a Fixes:, a Reviewed-by: or a Tested-by:. > http://autobuild.buildroot.net/results/e61/e613fc777051be6325d7e3c088d5f723fab518fa/build-end.log We like to have just: http://autobuild.buildroot.net/results/e613fc777051be6325d7e3c088d5f723fab518fa/ i.e don't point to build-end.log, but to the build result, and drop the extra /e61/ in the URL. > This patch is from Kuppuswamy Sathyanarayanan's fork of aer-inject. An > alternative to including the patch would be to switch to this fork as > Buildroot's idea of upstream. Currently, it is the only change in the > fork, though. I guess at this point, it's fine to keep things as they are. How "official" is each of those two aer-inject Git repos? Thanks a lot! Thomas
On Mon, Aug 12, 2024 at 09:41:24PM +0200, Thomas Petazzoni wrote: > Hello J, Hello Thomas, > Thanks a lot for your patch, which I have applied. I just have a few > very, very minor suggestions to further improve your next patches. > > First of all, I personally like that the commit title starts with a > lower-case letter, i.e: "package/aer-inject: fix ..." rather than > "package/aer-inject: Fix ...". Ah, force of habit… > > On Sun, 11 Aug 2024 12:34:25 +0200 > J. Neuschäfer via buildroot <buildroot@buildroot.org> wrote: > > > Fixes several build failures such as: > > Please write just "Fixes:". This way, patchwork detects it's a patch > fixes something, and it flags it as such. > See https://patchwork.ozlabs.org/project/buildroot/list/, where the column A/F/R/T flags the patches that have a Acked-by:, a Fixes:, a Reviewed-by: or a Tested-by:. > > > http://autobuild.buildroot.net/results/e61/e613fc777051be6325d7e3c088d5f723fab518fa/build-end.log > > We like to have just: > > http://autobuild.buildroot.net/results/e613fc777051be6325d7e3c088d5f723fab518fa/ > > i.e don't point to build-end.log, but to the build result, and drop the > extra /e61/ in the URL. Noted! > > This patch is from Kuppuswamy Sathyanarayanan's fork of aer-inject. An > > alternative to including the patch would be to switch to this fork as > > Buildroot's idea of upstream. Currently, it is the only change in the > > fork, though. > > I guess at this point, it's fine to keep things as they are. How > "official" is each of those two aer-inject Git repos? I took a deeper look right now, and found this on the linux-pci mailing list: https://lore.kernel.org/linux-pci/20240416055035.200085-1-sathyanarayanan.kuppuswamy@linux.intel.com/ Subject: [PATCH v1] PCI/AER: Update aer-inject tool source URL - https://git.kernel.org/cgit/linux/kernel/git/gong.chen/aer-inject.git/ + https://github.com/intel/aer-inject.git This patch was merged, so I think the github.com/intel URL counts as official. -- jn
diff --git a/package/aer-inject/0001-libgen-basename.patch b/package/aer-inject/0001-libgen-basename.patch new file mode 100644 index 0000000000..73c21ad8f4 --- /dev/null +++ b/package/aer-inject/0001-libgen-basename.patch @@ -0,0 +1,35 @@ +From 197f51d29a01f46750fa6928409301aa8f1163d7 Mon Sep 17 00:00:00 2001 +From: Khem Raj <raj.khem@gmail.com> +Date: Mon, 25 Mar 2024 09:04:40 -0700 +Subject: aer-inject: Include libgen.h for explicit basename prototype + +The prototype for basename has been removed from string.h in the latest +versions of musl [1]. This absence of prototype is flagged as an error +by some compilers, such as clang-18. To resolve this, include libgen.h +explicitly, which provides the prototype for basename. + +[1] https://git.musl-libc.org/cgit/musl/commit/?id=725e17ed6dff4d0cd22487bb64470881e86a92e7 +Signed-off-by: Khem Raj <raj.khem@gmail.com> +[sathya: Updated the commit log] +Signed-off-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com> +Upstream: https://git.kernel.org/pub/scm/linux/kernel/git/knsathya/aer-inject.git/commit/?id=197f51d29a01f46750fa6928409301aa8f1163d7 +Signed-off-by: J. Neuschaefer <j.neuschaefer@gmx.net> +--- + aer-inject.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/aer-inject.c b/aer-inject.c +index 74e7f72..eed1211 100644 +--- a/aer-inject.c ++++ b/aer-inject.c +@@ -11,6 +11,7 @@ + */ + + #include <stdio.h> ++#include <libgen.h> + #include <stdlib.h> + #include <string.h> + #include <fcntl.h> +-- +cgit 1.2.3-korg +
Fixes several build failures such as: http://autobuild.buildroot.net/results/e61/e613fc777051be6325d7e3c088d5f723fab518fa/build-end.log Signed-off-by: J. Neuschäfer <j.neuschaefer@gmx.net> --- This patch is from Kuppuswamy Sathyanarayanan's fork of aer-inject. An alternative to including the patch would be to switch to this fork as Buildroot's idea of upstream. Currently, it is the only change in the fork, though. --- package/aer-inject/0001-libgen-basename.patch | 35 +++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) --- base-commit: 74b46d08da7adc49b01f7af7ebb9e1421b8a9362 change-id: 20240811-aer-inject-5ed79f0554fc Best regards, -- J. Neuschäfer <j.neuschaefer@gmx.net>