On Wed, 2025-01-08 at 14:26 +0000, Bernhard Beschow wrote: > > > Am 8. Januar 2025 09:25:28 UTC schrieb Bernhard Beschow <shen...@gmail.com>: > > Signed-off-by: Bernhard Beschow <shen...@gmail.com> > > --- > > include/hw/core/shared-irq.h | 39 ++++++++++++++++ > > hw/core/shared-irq.c | 88 ++++++++++++++++++++++++++++++++++++ > > hw/core/Kconfig | 3 ++ > > hw/core/meson.build | 1 + > > 4 files changed, 131 insertions(+) > > create mode 100644 include/hw/core/shared-irq.h > > create mode 100644 hw/core/shared-irq.c > > > > diff --git a/include/hw/core/shared-irq.h b/include/hw/core/shared-irq.h > > new file mode 100644 > > As pointed out by David, this device is redundant to TYPE_OR_IRQ. I'll drop > it in v2.
Of course, I'd much rather *fix* it to do a reverse callback up the chain when the line is deasserted, to check whether any sources want it to be asserted again. That's the only way we can do VFIO level interrupts properly.
smime.p7s
Description: S/MIME cryptographic signature
