Starting with MTL, the driver needs to not only protect the steering control register from simultaneous software accesses, but also protect against races with hardware/firmware agents. The hardware provides a dedicated locking mechanism to support this via the STEER_SEMAPHORE register. Reading the register acts as a 'trylock' operation; the read will return 0x1 if the lock is acquired or 0x0 if something else is already holding the lock; once acquired, writing 0x1 to the register will release the lock. We'll continue to grab the software lock as well, just so lockdep can track our locking; assuming the hardware lock is behaving properly, there should never be any contention on the software lock in this case. Signed-off-by: Matt Roper <matthew.d.roper@xxxxxxxxx> --- drivers/gpu/drm/i915/gt/intel_gt_mcr.c | 29 +++++++++++++++++++++---- drivers/gpu/drm/i915/gt/intel_gt_regs.h | 1 + 2 files changed, 26 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/gt/intel_gt_mcr.c b/drivers/gpu/drm/i915/gt/intel_gt_mcr.c index f9e722d91904..fe5f5e0affdf 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_mcr.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_mcr.c @@ -345,10 +345,9 @@ static u32 rw_with_mcr_steering(struct intel_gt *gt, * @gt: GT structure * * Performs locking to protect the steering for the duration of an MCR - * operation. Depending on the platform, this may be a software lock - * (gt->mcr_lock) or a hardware lock (i.e., a register that synchronizes - * access not only for the driver, but also for external hardware and - * firmware agents). + * operation. On MTL and beyond, a hardware lock will also be taken to + * serialize access not only for the driver, but also for external hardware and + * firmware agents. * * Context: Takes gt->mcr_lock. uncore->lock should *not* be held when this * function is called, although it may be acquired after this @@ -356,9 +355,28 @@ static u32 rw_with_mcr_steering(struct intel_gt *gt, */ void intel_gt_mcr_lock(struct intel_gt *gt) { + int err = 0; + lockdep_assert_not_held(>->uncore->lock); + /* + * Starting with MTL, we need to coordinate not only with other + * driver threads, but also with hardware/firmware agents. A dedicated + * locking register is used. + */ + if (GRAPHICS_VER(gt->i915) >= IP_VER(12, 70)) + err = wait_for(intel_uncore_read_fw(gt->uncore, + STEER_SEMAPHORE) == 0x1, 1); + + /* + * Even on platforms with a hardware lock, we'll continue to grab + * a software spinlock too for lockdep purposes. If the hardware lock + * was already acquired, there should never be contention on the + * software lock. + */ spin_lock(>->mcr_lock); + + drm_WARN_ON_ONCE(>->i915->drm, err == -ETIMEDOUT); } /** @@ -372,6 +390,9 @@ void intel_gt_mcr_lock(struct intel_gt *gt) void intel_gt_mcr_unlock(struct intel_gt *gt) { spin_unlock(>->mcr_lock); + + if (GRAPHICS_VER(gt->i915) >= IP_VER(12, 70)) + intel_uncore_write_fw(gt->uncore, STEER_SEMAPHORE, 0x1); } /** diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h index 80a979e6f6be..412c0b399ebd 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h @@ -67,6 +67,7 @@ #define GMD_ID_MEDIA _MMIO(MTL_MEDIA_GSI_BASE + 0xd8c) #define MCFG_MCR_SELECTOR _MMIO(0xfd0) +#define STEER_SEMAPHORE _MMIO(0xfd0) #define MTL_MCR_SELECTOR _MMIO(0xfd4) #define SF_MCR_SELECTOR _MMIO(0xfd8) #define GEN8_MCR_SELECTOR _MMIO(0xfdc) -- 2.38.1