rpmlint related build fails in KMP
FYI, latest rpmlint fails KMP builds, where their preamble doesn't contain something like: Supplements: (kernel-%1 and %name) BuildRequires: kernel-syms BuildRequires: modutils Before, usually just this was required (if I'm not mistaken): Requires: kernel-%1 As a result, more or less any KMP, that I build, failed to build now. I leave it up to the reader to decide whether this is a good or bad thing. Cheers, Pete
Hi, On Sun, 2021-09-05 at 16:09 +0200, Hans-Peter Jansen wrote:
FYI,
latest rpmlint fails KMP builds, where their preamble doesn't contain something like:
Supplements: (kernel-%1 and %name) BuildRequires: kernel-syms BuildRequires: modutils
Those build requires should be satisfied in the spec file using BuildRequires: %kernel_module_package_buildreqs
Before, usually just this was required (if I'm not mistaken): Requires: kernel-%1
rpmlint does complain (about a lot of things) and recommends the above preamble line, but I never understood why and I usually never do it. IMHO if that is really always needed, the Requires should be provided by the %kernel_module_package macros instead of us dragging around preamble files with our sources.
As a result, more or less any KMP, that I build, failed to build now.
I leave it up to the reader to decide whether this is a good or bad thing.
I don't have any build failures with TW and Leap and have no preamble. https://build.opensuse.org/package/show/home:sbahling:snd-firewire-improve/s... Regards, Scott
Hi Scott, Am Montag, 6. September 2021, 20:46:08 CEST schrieb Scott Bahling:
Hi,
On Sun, 2021-09-05 at 16:09 +0200, Hans-Peter Jansen wrote:
FYI,
latest rpmlint fails KMP builds, where their preamble doesn't contain something like:
Supplements: (kernel-%1 and %name) BuildRequires: kernel-syms BuildRequires: modutils
Those build requires should be satisfied in the spec file using
BuildRequires: %kernel_module_package_buildreqs
Before, usually just this was required (if I'm not mistaken): Requires: kernel-%1
I didn't manage to pass the rpmlint check on TW with a just a preamble like that. In other words, BuildRequires: %kernel_module_package_buildreqs didn't help (neither globally, nor in the KMP section). Only completing the preamble again, as noted above, make the build succeed. [ignoring dirty rpmlintrc measures for now..]
rpmlint does complain (about a lot of things) and recommends the above preamble line, but I never understood why and I usually never do it. IMHO if that is really always needed, the Requires should be provided by the %kernel_module_package macros instead of us dragging around preamble files with our sources.
As a result, more or less any KMP, that I build, failed to build now.
I leave it up to the reader to decide whether this is a good or bad thing.
I don't have any build failures with TW and Leap and have no preamble.
The key seems to not use a preamble in the first place. I needed this fix in lime-kmp, v4l2loopback, and nvidia-gfxG05 with my KMP builds in home:frispete:kernel, virtualbox:kmp worked out of the box, but doesn't use a preamble, and drbd is still broken for 5.14.* for other reasons.
https://build.opensuse.org/package/show/home:sbahling:snd-firewire-improve/s nd-firewire-improve
Best, Pete
On Tue, 2021-09-07 at 17:36 +0200, Hans-Peter Jansen wrote:
Hi Scott,
Am Montag, 6. September 2021, 20:46:08 CEST schrieb Scott Bahling:
Hi,
On Sun, 2021-09-05 at 16:09 +0200, Hans-Peter Jansen wrote:
FYI,
latest rpmlint fails KMP builds, where their preamble doesn't contain something like:
Supplements: (kernel-%1 and %name) BuildRequires: kernel-syms BuildRequires: modutils
Those build requires should be satisfied in the spec file using
BuildRequires: %kernel_module_package_buildreqs
Before, usually just this was required (if I'm not mistaken): Requires: kernel-%1
I didn't manage to pass the rpmlint check on TW with a just a preamble like that. In other words, BuildRequires: %kernel_module_package_buildreqs didn't help (neither globally, nor in the KMP section).
Only completing the preamble again, as noted above, make the build succeed. [ignoring dirty rpmlintrc measures for now..]
This issue was brought up on the factory ML today - apparently there was a bug introduced in rpmlint. Should be fixed now. -Scott
participants (2)
-
Hans-Peter Jansen -
Scott Bahling