drm/i915/perf: implement active wait for noa configurations

NOA configuration take some amount of time to apply. That amount of
time depends on the size of the GT. There is no documented time for
this. For example, past experimentations with powergating
configuration changes seem to indicate a 60~70us delay. We go with
500us as default for now which should be over the required amount of
time (according to HW architects).

v2: Don't forget to save/restore registers used for the wait (Chris)

v3: Name used CS_GPR registers (Chris)
    Fix compile issue due to rebase (Lionel)

v4: Fix save/restore helpers (Umesh)

v5: Move noa_wait from drm_i915_private to i915_perf_stream (Lionel)

v6: Add missing struct declarations in i915_perf.h

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20191012072308.30312-2-chris@chris-wilson.co.uk

1 files changed, 224 insertions, 0 deletions

diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 50f2f972020d..81e8a7934001 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -198,6 +198,7 @@
 #include "gem/i915_gem_context.h"
 #include "gt/intel_engine_pm.h"
 #include "gt/intel_engine_user.h"
+#include "gt/intel_gt.h"
 #include "gt/intel_lrc_reg.h"
 
 #include "i915_drv.h"
@@ -1347,6 +1348,12 @@ free_oa_configs(struct i915_perf_stream *stream)
 		free_oa_config_bo(oa_bo);
 }
 
+static void
+free_noa_wait(struct i915_perf_stream *stream)
+{
+	i915_vma_unpin_and_release(&stream->noa_wait, 0);
+}
+
 static void i915_oa_stream_destroy(struct i915_perf_stream *stream)
 {
 	struct i915_perf *perf = stream->perf;
@@ -1369,6 +1376,7 @@ static void i915_oa_stream_destroy(struct i915_perf_stream *stream)
 		oa_put_render_ctx_id(stream);
 
 	free_oa_configs(stream);
+	free_noa_wait(stream);
 
 	if (perf->spurious_report_rs.missed) {
 		DRM_NOTE("%d spurious OA report notices suppressed due to ratelimiting\n",
@@ -1529,6 +1537,206 @@ err_unref:
 	return ret;
 }
 
+static u32 *save_restore_register(struct i915_perf_stream *stream, u32 *cs,
+				  bool save, i915_reg_t reg, u32 offset,
+				  u32 dword_count)
+{
+	u32 cmd;
+	u32 d;
+
+	cmd = save ? MI_STORE_REGISTER_MEM : MI_LOAD_REGISTER_MEM;
+	if (INTEL_GEN(stream->perf->i915) >= 8)
+		cmd++;
+
+	for (d = 0; d < dword_count; d++) {
+		*cs++ = cmd;
+		*cs++ = i915_mmio_reg_offset(reg) + 4 * d;
+		*cs++ = intel_gt_scratch_offset(stream->engine->gt,
+						offset) + 4 * d;
+		*cs++ = 0;
+	}
+
+	return cs;
+}
+
+static int alloc_noa_wait(struct i915_perf_stream *stream)
+{
+	struct drm_i915_private *i915 = stream->perf->i915;
+	struct drm_i915_gem_object *bo;
+	struct i915_vma *vma;
+	const u64 delay_ticks = 0xffffffffffffffff -
+		DIV64_U64_ROUND_UP(
+			atomic64_read(&stream->perf->noa_programming_delay) *
+			RUNTIME_INFO(i915)->cs_timestamp_frequency_khz,
+			1000000ull);
+	const u32 base = stream->engine->mmio_base;
+#define CS_GPR(x) GEN8_RING_CS_GPR(base, x)
+	u32 *batch, *ts0, *cs, *jump;
+	int ret, i;
+	enum {
+		START_TS,
+		NOW_TS,
+		DELTA_TS,
+		JUMP_PREDICATE,
+		DELTA_TARGET,
+		N_CS_GPR
+	};
+
+	bo = i915_gem_object_create_internal(i915, 4096);
+	if (IS_ERR(bo)) {
+		DRM_ERROR("Failed to allocate NOA wait batchbuffer\n");
+		return PTR_ERR(bo);
+	}
+
+	/*
+	 * We pin in GGTT because we jump into this buffer now because
+	 * multiple OA config BOs will have a jump to this address and it
+	 * needs to be fixed during the lifetime of the i915/perf stream.
+	 */
+	vma = i915_gem_object_ggtt_pin(bo, NULL, 0, 0, PIN_HIGH);
+	if (IS_ERR(vma)) {
+		ret = PTR_ERR(vma);
+		goto err_unref;
+	}
+
+	batch = cs = i915_gem_object_pin_map(bo, I915_MAP_WB);
+	if (IS_ERR(batch)) {
+		ret = PTR_ERR(batch);
+		goto err_unpin;
+	}
+
+	/* Save registers. */
+	for (i = 0; i < N_CS_GPR; i++)
+		cs = save_restore_register(
+			stream, cs, true /* save */, CS_GPR(i),
+			INTEL_GT_SCRATCH_FIELD_PERF_CS_GPR + 8 * i, 2);
+	cs = save_restore_register(
+		stream, cs, true /* save */, MI_PREDICATE_RESULT_1,
+		INTEL_GT_SCRATCH_FIELD_PERF_PREDICATE_RESULT_1, 1);
+
+	/* First timestamp snapshot location. */
+	ts0 = cs;
+
+	/*
+	 * Initial snapshot of the timestamp register to implement the wait.
+	 * We work with 32b values, so clear out the top 32b bits of the
+	 * register because the ALU works 64bits.
+	 */
+	*cs++ = MI_LOAD_REGISTER_IMM(1);
+	*cs++ = i915_mmio_reg_offset(CS_GPR(START_TS)) + 4;
+	*cs++ = 0;
+	*cs++ = MI_LOAD_REGISTER_REG | (3 - 2);
+	*cs++ = i915_mmio_reg_offset(RING_TIMESTAMP(base));
+	*cs++ = i915_mmio_reg_offset(CS_GPR(START_TS));
+
+	/*
+	 * This is the location we're going to jump back into until the
+	 * required amount of time has passed.
+	 */
+	jump = cs;
+
+	/*
+	 * Take another snapshot of the timestamp register. Take care to clear
+	 * up the top 32bits of CS_GPR(1) as we're using it for other
+	 * operations below.
+	 */
+	*cs++ = MI_LOAD_REGISTER_IMM(1);
+	*cs++ = i915_mmio_reg_offset(CS_GPR(NOW_TS)) + 4;
+	*cs++ = 0;
+	*cs++ = MI_LOAD_REGISTER_REG | (3 - 2);
+	*cs++ = i915_mmio_reg_offset(RING_TIMESTAMP(base));
+	*cs++ = i915_mmio_reg_offset(CS_GPR(NOW_TS));
+
+	/*
+	 * Do a diff between the 2 timestamps and store the result back into
+	 * CS_GPR(1).
+	 */
+	*cs++ = MI_MATH(5);
+	*cs++ = MI_MATH_LOAD(MI_MATH_REG_SRCA, MI_MATH_REG(NOW_TS));
+	*cs++ = MI_MATH_LOAD(MI_MATH_REG_SRCB, MI_MATH_REG(START_TS));
+	*cs++ = MI_MATH_SUB;
+	*cs++ = MI_MATH_STORE(MI_MATH_REG(DELTA_TS), MI_MATH_REG_ACCU);
+	*cs++ = MI_MATH_STORE(MI_MATH_REG(JUMP_PREDICATE), MI_MATH_REG_CF);
+
+	/*
+	 * Transfer the carry flag (set to 1 if ts1 < ts0, meaning the
+	 * timestamp have rolled over the 32bits) into the predicate register
+	 * to be used for the predicated jump.
+	 */
+	*cs++ = MI_LOAD_REGISTER_REG | (3 - 2);
+	*cs++ = i915_mmio_reg_offset(CS_GPR(JUMP_PREDICATE));
+	*cs++ = i915_mmio_reg_offset(MI_PREDICATE_RESULT_1);
+
+	/* Restart from the beginning if we had timestamps roll over. */
+	*cs++ = (INTEL_GEN(i915) < 8 ?
+		 MI_BATCH_BUFFER_START :
+		 MI_BATCH_BUFFER_START_GEN8) |
+		MI_BATCH_PREDICATE;
+	*cs++ = i915_ggtt_offset(vma) + (ts0 - batch) * 4;
+	*cs++ = 0;
+
+	/*
+	 * Now add the diff between to previous timestamps and add it to :
+	 *      (((1 * << 64) - 1) - delay_ns)
+	 *
+	 * When the Carry Flag contains 1 this means the elapsed time is
+	 * longer than the expected delay, and we can exit the wait loop.
+	 */
+	*cs++ = MI_LOAD_REGISTER_IMM(2);
+	*cs++ = i915_mmio_reg_offset(CS_GPR(DELTA_TARGET));
+	*cs++ = lower_32_bits(delay_ticks);
+	*cs++ = i915_mmio_reg_offset(CS_GPR(DELTA_TARGET)) + 4;
+	*cs++ = upper_32_bits(delay_ticks);
+
+	*cs++ = MI_MATH(4);
+	*cs++ = MI_MATH_LOAD(MI_MATH_REG_SRCA, MI_MATH_REG(DELTA_TS));
+	*cs++ = MI_MATH_LOAD(MI_MATH_REG_SRCB, MI_MATH_REG(DELTA_TARGET));
+	*cs++ = MI_MATH_ADD;
+	*cs++ = MI_MATH_STOREINV(MI_MATH_REG(JUMP_PREDICATE), MI_MATH_REG_CF);
+
+	/*
+	 * Transfer the result into the predicate register to be used for the
+	 * predicated jump.
+	 */
+	*cs++ = MI_LOAD_REGISTER_REG | (3 - 2);
+	*cs++ = i915_mmio_reg_offset(CS_GPR(JUMP_PREDICATE));
+	*cs++ = i915_mmio_reg_offset(MI_PREDICATE_RESULT_1);
+
+	/* Predicate the jump.  */
+	*cs++ = (INTEL_GEN(i915) < 8 ?
+		 MI_BATCH_BUFFER_START :
+		 MI_BATCH_BUFFER_START_GEN8) |
+		MI_BATCH_PREDICATE;
+	*cs++ = i915_ggtt_offset(vma) + (jump - batch) * 4;
+	*cs++ = 0;
+
+	/* Restore registers. */
+	for (i = 0; i < N_CS_GPR; i++)
+		cs = save_restore_register(
+			stream, cs, false /* restore */, CS_GPR(i),
+			INTEL_GT_SCRATCH_FIELD_PERF_CS_GPR + 8 * i, 2);
+	cs = save_restore_register(
+		stream, cs, false /* restore */, MI_PREDICATE_RESULT_1,
+		INTEL_GT_SCRATCH_FIELD_PERF_PREDICATE_RESULT_1, 1);
+
+	/* And return to the ring. */
+	*cs++ = MI_BATCH_BUFFER_END;
+
+	GEM_BUG_ON(cs - batch > PAGE_SIZE / sizeof(*batch));
+
+	i915_gem_object_flush_map(bo);
+	i915_gem_object_unpin_map(bo);
+
+	stream->noa_wait = vma;
+	return 0;
+
+err_unpin:
+	__i915_vma_unpin(vma);
+err_unref:
+	i915_gem_object_put(bo);
+	return ret;
+}
+
 static void config_oa_regs(struct intel_uncore *uncore,
 			   const struct i915_oa_reg *regs,
 			   u32 n_regs)
@@ -2206,6 +2414,12 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream,
 		}
 	}
 
+	ret = alloc_noa_wait(stream);
+	if (ret) {
+		DRM_DEBUG("Unable to allocate NOA wait batch buffer\n");
+		goto err_noa_wait_alloc;
+	}
+
 	stream->oa_config = i915_perf_get_oa_config(perf, props->metrics_set);
 	if (!stream->oa_config) {
 		DRM_DEBUG("Invalid OA config id=%i\n", props->metrics_set);
@@ -2265,6 +2479,9 @@ err_oa_buf_alloc:
 	intel_engine_pm_put(stream->engine);
 
 err_config:
+	free_noa_wait(stream);
+
+err_noa_wait_alloc:
 	if (stream->ctx)
 		oa_put_render_ctx_id(stream);
 
@@ -3651,6 +3868,9 @@ void i915_perf_init(struct drm_i915_private *i915)
 		ratelimit_set_flags(&perf->spurious_report_rs,
 				    RATELIMIT_MSG_ON_RELEASE);
 
+		atomic64_set(&perf->noa_programming_delay,
+			     500 * 1000 /* 500us */);
+
 		perf->i915 = i915;
 	}
 }
@@ -3680,3 +3900,7 @@ void i915_perf_fini(struct drm_i915_private *i915)
 	memset(&perf->ops, 0, sizeof(perf->ops));
 	perf->i915 = NULL;
 }
+
+#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST)
+#include "selftests/i915_perf.c"
+#endif