From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Static table wastes space for invalid combinations and
engines which are not supported by Gen6 (legacy semaphores).
Replace it with a function devised by Dave Gordon.
I have verified that it generates the same mappings between
mbox selectors and signalling registers.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Cc: Dave Gordon <david.s.gordon@xxxxxxxxx>
Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
---
drivers/gpu/drm/i915/i915_reg.h | 7 ++---
drivers/gpu/drm/i915/intel_engine_cs.c | 48 +++++++++++++++++++++++++++++++++
drivers/gpu/drm/i915/intel_ringbuffer.c | 40 ++-------------------------
drivers/gpu/drm/i915/intel_ringbuffer.h | 3 +++
4 files changed, 57 insertions(+), 41 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 8bfde75789f6..28aa876e2d87 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1604,9 +1604,10 @@ enum skl_disp_power_wells {
#define RING_HEAD(base) _MMIO((base)+0x34)
#define RING_START(base) _MMIO((base)+0x38)
#define RING_CTL(base) _MMIO((base)+0x3c)
-#define RING_SYNC_0(base) _MMIO((base)+0x40)
-#define RING_SYNC_1(base) _MMIO((base)+0x44)
-#define RING_SYNC_2(base) _MMIO((base)+0x48)
+#define RING_SYNC(base, n) _MMIO((base) + 0x40 + (n) * 4)
+#define RING_SYNC_0(base) RING_SYNC(base, 0)
+#define RING_SYNC_1(base) RING_SYNC(base, 1)
+#define RING_SYNC_2(base) RING_SYNC(base, 2)
#define GEN6_RVSYNC (RING_SYNC_0(RENDER_RING_BASE))
#define GEN6_RBSYNC (RING_SYNC_1(RENDER_RING_BASE))
#define GEN6_RVESYNC (RING_SYNC_2(RENDER_RING_BASE))
diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
index f4a35ec78481..9837fddae259 100644
--- a/drivers/gpu/drm/i915/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/intel_engine_cs.c
@@ -209,3 +209,51 @@ int intel_engine_init_common(struct intel_engine_cs *engine)
return i915_cmd_parser_init_ring(engine);
}
+
+#define I915_NUM_GEN6_SEMAPHORE_ENGINES (4)
+
+static int gen6_sem_f(unsigned int x, unsigned int y)
+{
+ if (x == y)
+ return -1;
+
+ x = intel_engines[x].guc_id;
+ y = intel_engines[y].guc_id;
+
+ if (x >= I915_NUM_GEN6_SEMAPHORE_ENGINES ||
+ y >= I915_NUM_GEN6_SEMAPHORE_ENGINES)
+ return -1;
+
+ x -= x >= y;
+ if (y == 1)
+ x = 3 - x;
+ x += y & 1;
+ return x % 3;
+}
+
+u32 gen6_wait_mbox(enum intel_engine_id x, enum intel_engine_id y)
+{
+ int r;
+
+ r = gen6_sem_f(x, y);
+ if (r < 0)
+ return MI_SEMAPHORE_SYNC_INVALID;
+
+ if (r == 1)
+ r = 2;
+ else if (r == 2)
+ r = 1;
+
+ return r << 16;
+}
+
+i915_reg_t gen6_signal_reg(enum intel_engine_id x, enum intel_engine_id y)
+{
+ int r;
+
+ r = gen6_sem_f(x, y);
+ if (r < 0)
+ return GEN6_NOSYNC;
+
+ return RING_SYNC(intel_engines[y].mmio_base, r);
+}
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index b844e6984ae7..049527d381de 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -2731,44 +2731,8 @@ static void intel_ring_init_semaphores(struct drm_i915_private *dev_priv,
* sema between VCS2 and RCS later.
*/
for (i = 0; i < I915_NUM_ENGINES; i++) {
- static const struct {
- u32 wait_mbox;
- i915_reg_t mbox_reg;
- } sem_data[I915_NUM_ENGINES][I915_NUM_ENGINES] = {
- [RCS] = {
- [VCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_RV, .mbox_reg = GEN6_VRSYNC },
- [BCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_RB, .mbox_reg = GEN6_BRSYNC },
- [VECS] = { .wait_mbox = MI_SEMAPHORE_SYNC_RVE, .mbox_reg = GEN6_VERSYNC },
- },
- [VCS] = {
- [RCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_VR, .mbox_reg = GEN6_RVSYNC },
- [BCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_VB, .mbox_reg = GEN6_BVSYNC },
- [VECS] = { .wait_mbox = MI_SEMAPHORE_SYNC_VVE, .mbox_reg = GEN6_VEVSYNC },
- },
- [BCS] = {
- [RCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_BR, .mbox_reg = GEN6_RBSYNC },
- [VCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_BV, .mbox_reg = GEN6_VBSYNC },
- [VECS] = { .wait_mbox = MI_SEMAPHORE_SYNC_BVE, .mbox_reg = GEN6_VEBSYNC },
- },
- [VECS] = {
- [RCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_VER, .mbox_reg = GEN6_RVESYNC },
- [VCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_VEV, .mbox_reg = GEN6_VVESYNC },
- [BCS] = { .wait_mbox = MI_SEMAPHORE_SYNC_VEB, .mbox_reg = GEN6_BVESYNC },
- },
- };
- u32 wait_mbox;
- i915_reg_t mbox_reg;
-
- if (i == engine->id || i == VCS2) {
- wait_mbox = MI_SEMAPHORE_SYNC_INVALID;
- mbox_reg = GEN6_NOSYNC;
- } else {
- wait_mbox = sem_data[engine->id][i].wait_mbox;
- mbox_reg = sem_data[engine->id][i].mbox_reg;
- }
-
- engine->semaphore.mbox.wait[i] = wait_mbox;
- engine->semaphore.mbox.signal[i] = mbox_reg;
+ engine->semaphore.mbox.wait[i] = gen6_wait_mbox(engine->id, i);
+ engine->semaphore.mbox.signal[i] = gen6_signal_reg(engine->id, i);
}
}
}
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 05bab8bda63d..802adcd51569 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -493,6 +493,9 @@ int intel_ring_invalidate_all_caches(struct drm_i915_gem_request *req);
int intel_init_pipe_control(struct intel_engine_cs *engine, int size);
void intel_fini_pipe_control(struct intel_engine_cs *engine);
+u32 gen6_wait_mbox(enum intel_engine_id x, enum intel_engine_id y);
+i915_reg_t gen6_signal_reg(enum intel_engine_id x, enum intel_engine_id y);
+
void intel_engine_setup_common(struct intel_engine_cs *engine);
int intel_engine_init_common(struct intel_engine_cs *engine);
--
1.9.1
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
