ACK/Cmnt: [SRU][J:linux-bluefield][PATCH v1 0/1] mlxbf_gige: add deferred probing to Jammy
Massimiliano Pellizzer
massimiliano.pellizzer at canonical.com
Fri Jul 18 14:41:20 UTC 2025
On Fri, 18 Jul 2025 at 13:01, Kuba Pawlak <kuba.pawlak at canonical.com> wrote:
>
> On 18.07.2025 00:01, David Thompson wrote:
> > BugLink: https://bugs.launchpad.net/bugs/2117197
> >
> > SRU Justification:
> >
> > [Impact]
> > The GIGE driver gets the PHY IRQ resource during its initialization path.
> > If this resource is not available at driver probe(), an error message
> > is emitted to log: "Error getting PHY irq. Use polling instead"
> >
> > [Fix]
> > The fix is to update the GIGE driver to use deferred probing, i.e. return
> > EPROBE_DEFER if the resource is not available. This allows the driver
> > probe() to be re-entered, where the resource acquisition will be tried again.
> > The upstream patch does not apply cleanly, so provide SAUCE patch for now.
> >
> > [Test Case]
> > * Boot BF platform with software containing fix
> > * Check the kernel log via dmesg for above message
> > * Reboot many times and keep checking log
> >
> > [Regression Potential]
> > This fix is upstreamed and tested well.
> >
> > David Thompson (1):
> > UBUNTU: SAUCE: mlxbf_gige: return EPROBE_DEFER if PHY IRQ is not available
> >
> > drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_main.c | 6 ++++--
> > 1 file changed, 4 insertions(+), 2 deletions(-)
> >
> Acked-by: Kuba Pawlak <kuba.pawlak at canonical.com>
> --
> kernel-team mailing list
> kernel-team at lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team
The patch looks good to me, however
instead of using a "SAUCE" patch, I would have backported
e7ea5f5b1858dd mlxbf_gige: return EPROBE_DEFER if PHY IRQ is not available
from mainline.
Also, it would be preferable for all patches to be included in the same thread
if a patch needs to be applied to multiple Ubuntu series.
Acked-by: Massimiliano Pellizzer <massimiliano.pellizzer at canonical.com>
--
Massimiliano Pellizzer
More information about the kernel-team
mailing list