From: Gustavo Sousa Date: Tue, 26 Sep 2023 22:19:15 +0000 (-0300) Subject: drm/xe/irq: Clear GFX_MSTR_IRQ as part of IRQ reset X-Git-Url: http://git.maquefel.me/?a=commitdiff_plain;h=51a5d656090e0a865d91f1e6ce0c7a09d71a4b70;p=linux.git drm/xe/irq: Clear GFX_MSTR_IRQ as part of IRQ reset Starting with Xe_LP+, GFX_MSTR_IRQ contains status bits that have W1C behavior. If we do not properly reset them, we would miss delivery of interrupts if a pending bit is set when enabling IRQs. As an example, the display part of our probe routine contains paths where we wait for vblank interrupts. If a display interrupt was already pending when enabling IRQs, we would time out waiting for the vblank. That in fact happened recently when modprobing Xe on a Lunar Lake with a specific configuration; and that's how we found out we were missing this step in the IRQ enabling logic. Fix the issue by clearing GFX_MSTR_IRQ as part of the IRQ reset. v2: - Make resetting GFX_MSTR_IRQ be the last step to avoid bit re-latching. (Ville) v3: - Swap nesting order: guard loop with the IP version check instead of doing the check at each iteration. (Lucas) v4: - Add braces for the "if" statement guarding the loop to make the compiler happy. (Gustavo) BSpec: 50875, 54028, 62357 Cc: Matt Roper Cc: Ville Syrjälä Reviewed-by: Lucas De Marchi Link: https://lore.kernel.org/r/20230926221914.106843-2-gustavo.sousa@intel.com Signed-off-by: Gustavo Sousa Signed-off-by: Rodrigo Vivi --- diff --git a/drivers/gpu/drm/xe/xe_irq.c b/drivers/gpu/drm/xe/xe_irq.c index a91e782e06ebf..4cc5f7086b4ce 100644 --- a/drivers/gpu/drm/xe/xe_irq.c +++ b/drivers/gpu/drm/xe/xe_irq.c @@ -511,6 +511,13 @@ static void dg1_irq_reset(struct xe_tile *tile) mask_and_disable(tile, PCU_IRQ_OFFSET); } +static void dg1_irq_reset_mstr(struct xe_tile *tile) +{ + struct xe_gt *mmio = tile->primary_gt; + + xe_mmio_write32(mmio, GFX_MSTR_IRQ, ~0); +} + static void xe_irq_reset(struct xe_device *xe) { struct xe_tile *tile; @@ -525,6 +532,16 @@ static void xe_irq_reset(struct xe_device *xe) tile = xe_device_get_root_tile(xe); mask_and_disable(tile, GU_MISC_IRQ_OFFSET); + + /* + * The tile's top-level status register should be the last one + * to be reset to avoid possible bit re-latching from lower + * level interrupts. + */ + if (GRAPHICS_VERx100(xe) >= 1210) { + for_each_tile(tile, xe, id) + dg1_irq_reset_mstr(tile); + } } static void xe_irq_postinstall(struct xe_device *xe)