Right now, panthor is one of the rare drivers to signal fences from work items (not even from the threaded IRQ handler). We could move that to the threaded handler, but that would still leave the latency caused by the scheduling of the IRQ thread.
Instead, this patchset moves all the JOB/GPU IRQ processing to the raw IRQ handler, which is fine because what the current code does is demux the interrupts and defer actual handling to sub work items. The only non-trivial thing we keep in the IRQ path is the dma_fence signalling, which should be acceptable in term of CPU cycles burnt in IRQ context. Note that the MMU event handling is left in a threaded handler because it requires acquiring sleepable locks and fixing that is non-trivial. Still very basic testing done, but glmark2 and gfxbench's manhattan test show a ~5% perf improvement on a rk3588 with this patchset applied. Signed-off-by: Boris Brezillon <[email protected]> --- Changes in v2: - Fix commit message in patch 4 - Move devm_kasprintf() before panthor_irq_resume() in patch 3 - Fix erroneous lockdep_assert_held() in patch 6 - Make sure events_lock is held when calling csg_slot_sync_update_locked() in patch 6 - Restore a csg_slot_sync_update_locked() call in patch 7 - Fix a potential deadlock in patch 9 - Drop the IRQ coalescing patch (formerly patch 10) - Change panthor_irq_request() so we don't have to define a dummy threaded handler, and we can let RT kernels move the hard handler to a thread - Add patches to transition GPU event processing to the hard IRQ handler - Link to v1: https://lore.kernel.org/r/[email protected] --- Boris Brezillon (11): drm/panthor: Make panthor_irq::state a non-atomic field drm/panthor: Move the register accessors before the IRQ helpers drm/panthor: Replace the panthor_irq macro machinery by inline helpers drm/panthor: Extend the IRQ logic to allow fast/hard IRQ handlers drm/panthor: Make panthor_fw_{update,toggle}_reqs() callable from IRQ context drm/panthor: Prepare the scheduler logic for FW events in IRQ context drm/panthor: Automate CSG IRQ processing at group unbind time drm/panthor: Automatically enable interrupts in panthor_fw_wait_acks() drm/panthor: Process FW events in IRQ context drm/panthor: Use the irqsave variant of spin_lock in panthor_gpu_irq_handler() drm/panthor: Process GPU events in IRQ context drivers/gpu/drm/panthor/panthor_device.h | 281 +++++++++--------- drivers/gpu/drm/panthor/panthor_fw.c | 76 +++-- drivers/gpu/drm/panthor/panthor_fw.h | 9 +- drivers/gpu/drm/panthor/panthor_gpu.c | 31 +- drivers/gpu/drm/panthor/panthor_mmu.c | 38 +-- drivers/gpu/drm/panthor/panthor_pwr.c | 21 +- drivers/gpu/drm/panthor/panthor_sched.c | 483 ++++++++++++++----------------- 7 files changed, 476 insertions(+), 463 deletions(-) --- base-commit: ac5ac0acf11df04295eb1811066097b7022d6c7f change-id: 20260429-panthor-signal-from-irq-d33684f4d292 Best regards, -- Boris Brezillon <[email protected]>
