On Wed, 29 May 2024 20:59:19 +0100, "Rob Herring (Arm)" <robh@xxxxxxxxxx> wrote: > > The duplicated parsing continued to bother me, so I've refactored things > to avoid that for parsing the interrupt parent and args in the > interrupt-map. > > It passes testing with unittests on QEMU virt platform, but I don't > think that catches the problematic cases. So please test. > > v1: https://lore.kernel.org/all/20240528164132.2451685-1-maz@xxxxxxxxxx/ > - Refactor existing interrupt-map parsing code and use it for > fw_devlink > > Signed-off-by: Rob Herring (Arm) <robh@xxxxxxxxxx> > --- > Marc Zyngier (1): > of: property: Fix fw_devlink handling of interrupt-map > > Rob Herring (Arm) (1): > of/irq: Factor out parsing of interrupt-map parent phandle+args from of_irq_parse_raw() > > drivers/of/irq.c | 127 +++++++++++++++++++++++++++++------------------- > drivers/of/of_private.h | 3 ++ > drivers/of/property.c | 30 ++++-------- > 3 files changed, 89 insertions(+), 71 deletions(-) I've just gave it a go on an M1 and as a kvmtool guest, and nothing caught fire. Must be perfect. Tested-by: Marc Zyngier <maz@xxxxxxxxxx> M. -- Without deviation from the norm, progress is not possible.
