#include <drm/drmP.h>
#include <drm/i915_drm.h>
#include "i915_drv.h"
+#include "intel_mocs.h"
#define GEN9_LR_CONTEXT_RENDER_SIZE (22 * PAGE_SIZE)
#define GEN8_LR_CONTEXT_RENDER_SIZE (20 * PAGE_SIZE)
#define GEN8_CTX_ID_SHIFT 32
#define CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x17
-static int intel_lr_context_pin(struct intel_engine_cs *ring,
- struct intel_context *ctx);
+static int intel_lr_context_pin(struct drm_i915_gem_request *rq);
/**
* intel_sanitize_enable_execlists() - sanitize i915.enable_execlists
return lrca >> 12;
}
-static uint64_t execlists_ctx_descriptor(struct intel_engine_cs *ring,
- struct drm_i915_gem_object *ctx_obj)
+static uint64_t execlists_ctx_descriptor(struct drm_i915_gem_request *rq)
{
+ struct intel_engine_cs *ring = rq->ring;
struct drm_device *dev = ring->dev;
+ struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
uint64_t desc;
uint64_t lrca = i915_gem_obj_ggtt_offset(ctx_obj);
return desc;
}
-static void execlists_elsp_write(struct intel_engine_cs *ring,
- struct drm_i915_gem_object *ctx_obj0,
- struct drm_i915_gem_object *ctx_obj1)
+static void execlists_elsp_write(struct drm_i915_gem_request *rq0,
+ struct drm_i915_gem_request *rq1)
{
+
+ struct intel_engine_cs *ring = rq0->ring;
struct drm_device *dev = ring->dev;
struct drm_i915_private *dev_priv = dev->dev_private;
- uint64_t temp = 0;
- uint32_t desc[4];
+ uint64_t desc[2];
- /* XXX: You must always write both descriptors in the order below. */
- if (ctx_obj1)
- temp = execlists_ctx_descriptor(ring, ctx_obj1);
- else
- temp = 0;
- desc[1] = (u32)(temp >> 32);
- desc[0] = (u32)temp;
+ if (rq1) {
+ desc[1] = execlists_ctx_descriptor(rq1);
+ rq1->elsp_submitted++;
+ } else {
+ desc[1] = 0;
+ }
- temp = execlists_ctx_descriptor(ring, ctx_obj0);
- desc[3] = (u32)(temp >> 32);
- desc[2] = (u32)temp;
+ desc[0] = execlists_ctx_descriptor(rq0);
+ rq0->elsp_submitted++;
+ /* You must always write both descriptors in the order below. */
spin_lock(&dev_priv->uncore.lock);
intel_uncore_forcewake_get__locked(dev_priv, FORCEWAKE_ALL);
- I915_WRITE_FW(RING_ELSP(ring), desc[1]);
- I915_WRITE_FW(RING_ELSP(ring), desc[0]);
- I915_WRITE_FW(RING_ELSP(ring), desc[3]);
+ I915_WRITE_FW(RING_ELSP(ring), upper_32_bits(desc[1]));
+ I915_WRITE_FW(RING_ELSP(ring), lower_32_bits(desc[1]));
+ I915_WRITE_FW(RING_ELSP(ring), upper_32_bits(desc[0]));
/* The context is automatically loaded after the following */
- I915_WRITE_FW(RING_ELSP(ring), desc[2]);
+ I915_WRITE_FW(RING_ELSP(ring), lower_32_bits(desc[0]));
- /* ELSP is a wo register, so use another nearby reg for posting instead */
+ /* ELSP is a wo register, use another nearby reg for posting */
POSTING_READ_FW(RING_EXECLIST_STATUS(ring));
intel_uncore_forcewake_put__locked(dev_priv, FORCEWAKE_ALL);
spin_unlock(&dev_priv->uncore.lock);
}
-static int execlists_update_context(struct drm_i915_gem_object *ctx_obj,
- struct drm_i915_gem_object *ring_obj,
- struct i915_hw_ppgtt *ppgtt,
- u32 tail)
+static int execlists_update_context(struct drm_i915_gem_request *rq)
{
+ struct intel_engine_cs *ring = rq->ring;
+ struct i915_hw_ppgtt *ppgtt = rq->ctx->ppgtt;
+ struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
+ struct drm_i915_gem_object *rb_obj = rq->ringbuf->obj;
struct page *page;
uint32_t *reg_state;
+ BUG_ON(!ctx_obj);
+ WARN_ON(!i915_gem_obj_is_pinned(ctx_obj));
+ WARN_ON(!i915_gem_obj_is_pinned(rb_obj));
+
page = i915_gem_object_get_page(ctx_obj, 1);
reg_state = kmap_atomic(page);
- reg_state[CTX_RING_TAIL+1] = tail;
- reg_state[CTX_RING_BUFFER_START+1] = i915_gem_obj_ggtt_offset(ring_obj);
+ reg_state[CTX_RING_TAIL+1] = rq->tail;
+ reg_state[CTX_RING_BUFFER_START+1] = i915_gem_obj_ggtt_offset(rb_obj);
/* True PPGTT with dynamic page allocation: update PDP registers and
* point the unallocated PDPs to the scratch page
return 0;
}
-static void execlists_submit_contexts(struct intel_engine_cs *ring,
- struct intel_context *to0, u32 tail0,
- struct intel_context *to1, u32 tail1)
+static void execlists_submit_requests(struct drm_i915_gem_request *rq0,
+ struct drm_i915_gem_request *rq1)
{
- struct drm_i915_gem_object *ctx_obj0 = to0->engine[ring->id].state;
- struct intel_ringbuffer *ringbuf0 = to0->engine[ring->id].ringbuf;
- struct drm_i915_gem_object *ctx_obj1 = NULL;
- struct intel_ringbuffer *ringbuf1 = NULL;
+ execlists_update_context(rq0);
- BUG_ON(!ctx_obj0);
- WARN_ON(!i915_gem_obj_is_pinned(ctx_obj0));
- WARN_ON(!i915_gem_obj_is_pinned(ringbuf0->obj));
+ if (rq1)
+ execlists_update_context(rq1);
- execlists_update_context(ctx_obj0, ringbuf0->obj, to0->ppgtt, tail0);
-
- if (to1) {
- ringbuf1 = to1->engine[ring->id].ringbuf;
- ctx_obj1 = to1->engine[ring->id].state;
- BUG_ON(!ctx_obj1);
- WARN_ON(!i915_gem_obj_is_pinned(ctx_obj1));
- WARN_ON(!i915_gem_obj_is_pinned(ringbuf1->obj));
-
- execlists_update_context(ctx_obj1, ringbuf1->obj, to1->ppgtt, tail1);
- }
-
- execlists_elsp_write(ring, ctx_obj0, ctx_obj1);
+ execlists_elsp_write(rq0, rq1);
}
static void execlists_context_unqueue(struct intel_engine_cs *ring)
WARN_ON(req1 && req1->elsp_submitted);
- execlists_submit_contexts(ring, req0->ctx, req0->tail,
- req1 ? req1->ctx : NULL,
- req1 ? req1->tail : 0);
-
- req0->elsp_submitted++;
- if (req1)
- req1->elsp_submitted++;
+ execlists_submit_requests(req0, req1);
}
static bool execlists_check_remove_request(struct intel_engine_cs *ring,
int num_elements = 0;
if (request->ctx != ring->default_context)
- intel_lr_context_pin(ring, request->ctx);
+ intel_lr_context_pin(request);
i915_gem_request_reference(request);
{
int ret;
+ request->ringbuf = request->ctx->engine[request->ring->id].ringbuf;
+
if (request->ctx != request->ring->default_context) {
- ret = intel_lr_context_pin(request->ring, request->ctx);
+ ret = intel_lr_context_pin(request);
if (ret)
return ret;
}
- request->ringbuf = request->ctx->engine[request->ring->id].ringbuf;
-
return 0;
}
*
* Return: non-zero if the ringbuffer is not ready to be written to.
*/
-static int intel_logical_ring_begin(struct drm_i915_gem_request *req,
- int num_dwords)
+int intel_logical_ring_begin(struct drm_i915_gem_request *req, int num_dwords)
{
struct drm_i915_private *dev_priv;
int ret;
ctx->engine[ring->id].state;
if (ctx_obj && (ctx != ring->default_context))
- intel_lr_context_unpin(ring, ctx);
+ intel_lr_context_unpin(req);
list_del(&req->execlist_link);
i915_gem_request_unreference(req);
}
return 0;
}
-static int intel_lr_context_pin(struct intel_engine_cs *ring,
- struct intel_context *ctx)
+static int intel_lr_context_pin(struct drm_i915_gem_request *rq)
{
- struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
- struct intel_ringbuffer *ringbuf = ctx->engine[ring->id].ringbuf;
+ struct intel_engine_cs *ring = rq->ring;
+ struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
+ struct intel_ringbuffer *ringbuf = rq->ringbuf;
int ret = 0;
WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
- if (ctx->engine[ring->id].pin_count++ == 0) {
+ if (rq->ctx->engine[ring->id].pin_count++ == 0) {
ret = i915_gem_obj_ggtt_pin(ctx_obj,
GEN8_LR_CONTEXT_ALIGN, 0);
if (ret)
unpin_ctx_obj:
i915_gem_object_ggtt_unpin(ctx_obj);
reset_pin_count:
- ctx->engine[ring->id].pin_count = 0;
+ rq->ctx->engine[ring->id].pin_count = 0;
return ret;
}
-void intel_lr_context_unpin(struct intel_engine_cs *ring,
- struct intel_context *ctx)
+void intel_lr_context_unpin(struct drm_i915_gem_request *rq)
{
- struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
- struct intel_ringbuffer *ringbuf = ctx->engine[ring->id].ringbuf;
+ struct intel_engine_cs *ring = rq->ring;
+ struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
+ struct intel_ringbuffer *ringbuf = rq->ringbuf;
if (ctx_obj) {
WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
- if (--ctx->engine[ring->id].pin_count == 0) {
+ if (--rq->ctx->engine[ring->id].pin_count == 0) {
intel_unpin_ringbuffer_obj(ringbuf);
i915_gem_object_ggtt_unpin(ctx_obj);
}
return 0;
}
-#define wa_ctx_emit(batch, cmd) \
+#define wa_ctx_emit(batch, index, cmd) \
do { \
- if (WARN_ON(index >= (PAGE_SIZE / sizeof(uint32_t)))) { \
+ int __index = (index)++; \
+ if (WARN_ON(__index >= (PAGE_SIZE / sizeof(uint32_t)))) { \
return -ENOSPC; \
} \
- batch[index++] = (cmd); \
+ batch[__index] = (cmd); \
} while (0)
+
+/*
+ * In this WA we need to set GEN8_L3SQCREG4[21:21] and reset it after
+ * PIPE_CONTROL instruction. This is required for the flush to happen correctly
+ * but there is a slight complication as this is applied in WA batch where the
+ * values are only initialized once so we cannot take register value at the
+ * beginning and reuse it further; hence we save its value to memory, upload a
+ * constant value with bit21 set and then we restore it back with the saved value.
+ * To simplify the WA, a constant value is formed by using the default value
+ * of this register. This shouldn't be a problem because we are only modifying
+ * it for a short period and this batch in non-premptible. We can ofcourse
+ * use additional instructions that read the actual value of the register
+ * at that time and set our bit of interest but it makes the WA complicated.
+ *
+ * This WA is also required for Gen9 so extracting as a function avoids
+ * code duplication.
+ */
+static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *ring,
+ uint32_t *const batch,
+ uint32_t index)
+{
+ uint32_t l3sqc4_flush = (0x40400000 | GEN8_LQSC_FLUSH_COHERENT_LINES);
+
+ /*
+ * WaDisableLSQCROPERFforOCL:skl
+ * This WA is implemented in skl_init_clock_gating() but since
+ * this batch updates GEN8_L3SQCREG4 with default value we need to
+ * set this bit here to retain the WA during flush.
+ */
+ if (IS_SKYLAKE(ring->dev) && INTEL_REVID(ring->dev) <= SKL_REVID_E0)
+ l3sqc4_flush |= GEN8_LQSC_RO_PERF_DIS;
+
+ wa_ctx_emit(batch, index, (MI_STORE_REGISTER_MEM_GEN8(1) |
+ MI_SRM_LRM_GLOBAL_GTT));
+ wa_ctx_emit(batch, index, GEN8_L3SQCREG4);
+ wa_ctx_emit(batch, index, ring->scratch.gtt_offset + 256);
+ wa_ctx_emit(batch, index, 0);
+
+ wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(1));
+ wa_ctx_emit(batch, index, GEN8_L3SQCREG4);
+ wa_ctx_emit(batch, index, l3sqc4_flush);
+
+ wa_ctx_emit(batch, index, GFX_OP_PIPE_CONTROL(6));
+ wa_ctx_emit(batch, index, (PIPE_CONTROL_CS_STALL |
+ PIPE_CONTROL_DC_FLUSH_ENABLE));
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+
+ wa_ctx_emit(batch, index, (MI_LOAD_REGISTER_MEM_GEN8(1) |
+ MI_SRM_LRM_GLOBAL_GTT));
+ wa_ctx_emit(batch, index, GEN8_L3SQCREG4);
+ wa_ctx_emit(batch, index, ring->scratch.gtt_offset + 256);
+ wa_ctx_emit(batch, index, 0);
+
+ return index;
+}
+
static inline uint32_t wa_ctx_start(struct i915_wa_ctx_bb *wa_ctx,
uint32_t offset,
uint32_t start_alignment)
uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
/* WaDisableCtxRestoreArbitration:bdw,chv */
- wa_ctx_emit(batch, MI_ARB_ON_OFF | MI_ARB_DISABLE);
+ wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_DISABLE);
/* WaFlushCoherentL3CacheLinesAtContextSwitch:bdw */
if (IS_BROADWELL(ring->dev)) {
- struct drm_i915_private *dev_priv = to_i915(ring->dev);
- uint32_t l3sqc4_flush = (I915_READ(GEN8_L3SQCREG4) |
- GEN8_LQSC_FLUSH_COHERENT_LINES);
-
- wa_ctx_emit(batch, MI_LOAD_REGISTER_IMM(1));
- wa_ctx_emit(batch, GEN8_L3SQCREG4);
- wa_ctx_emit(batch, l3sqc4_flush);
-
- wa_ctx_emit(batch, GFX_OP_PIPE_CONTROL(6));
- wa_ctx_emit(batch, (PIPE_CONTROL_CS_STALL |
- PIPE_CONTROL_DC_FLUSH_ENABLE));
- wa_ctx_emit(batch, 0);
- wa_ctx_emit(batch, 0);
- wa_ctx_emit(batch, 0);
- wa_ctx_emit(batch, 0);
-
- wa_ctx_emit(batch, MI_LOAD_REGISTER_IMM(1));
- wa_ctx_emit(batch, GEN8_L3SQCREG4);
- wa_ctx_emit(batch, l3sqc4_flush & ~GEN8_LQSC_FLUSH_COHERENT_LINES);
+ index = gen8_emit_flush_coherentl3_wa(ring, batch, index);
+ if (index < 0)
+ return index;
}
/* WaClearSlmSpaceAtContextSwitch:bdw,chv */
/* Actual scratch location is at 128 bytes offset */
scratch_addr = ring->scratch.gtt_offset + 2*CACHELINE_BYTES;
- wa_ctx_emit(batch, GFX_OP_PIPE_CONTROL(6));
- wa_ctx_emit(batch, (PIPE_CONTROL_FLUSH_L3 |
- PIPE_CONTROL_GLOBAL_GTT_IVB |
- PIPE_CONTROL_CS_STALL |
- PIPE_CONTROL_QW_WRITE));
- wa_ctx_emit(batch, scratch_addr);
- wa_ctx_emit(batch, 0);
- wa_ctx_emit(batch, 0);
- wa_ctx_emit(batch, 0);
+ wa_ctx_emit(batch, index, GFX_OP_PIPE_CONTROL(6));
+ wa_ctx_emit(batch, index, (PIPE_CONTROL_FLUSH_L3 |
+ PIPE_CONTROL_GLOBAL_GTT_IVB |
+ PIPE_CONTROL_CS_STALL |
+ PIPE_CONTROL_QW_WRITE));
+ wa_ctx_emit(batch, index, scratch_addr);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
+ wa_ctx_emit(batch, index, 0);
/* Pad to end of cacheline */
while (index % CACHELINE_DWORDS)
- wa_ctx_emit(batch, MI_NOOP);
+ wa_ctx_emit(batch, index, MI_NOOP);
/*
* MI_BATCH_BUFFER_END is not required in Indirect ctx BB because
uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
/* WaDisableCtxRestoreArbitration:bdw,chv */
- wa_ctx_emit(batch, MI_ARB_ON_OFF | MI_ARB_ENABLE);
+ wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_ENABLE);
+
+ wa_ctx_emit(batch, index, MI_BATCH_BUFFER_END);
+
+ return wa_ctx_end(wa_ctx, *offset = index, 1);
+}
+
+static int gen9_init_indirectctx_bb(struct intel_engine_cs *ring,
+ struct i915_wa_ctx_bb *wa_ctx,
+ uint32_t *const batch,
+ uint32_t *offset)
+{
+ int ret;
+ struct drm_device *dev = ring->dev;
+ uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
+
+ /* WaDisableCtxRestoreArbitration:skl,bxt */
+ if ((IS_SKYLAKE(dev) && (INTEL_REVID(dev) <= SKL_REVID_D0)) ||
+ (IS_BROXTON(dev) && (INTEL_REVID(dev) == BXT_REVID_A0)))
+ wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_DISABLE);
- wa_ctx_emit(batch, MI_BATCH_BUFFER_END);
+ /* WaFlushCoherentL3CacheLinesAtContextSwitch:skl,bxt */
+ ret = gen8_emit_flush_coherentl3_wa(ring, batch, index);
+ if (ret < 0)
+ return ret;
+ index = ret;
+
+ /* Pad to end of cacheline */
+ while (index % CACHELINE_DWORDS)
+ wa_ctx_emit(batch, index, MI_NOOP);
+
+ return wa_ctx_end(wa_ctx, *offset = index, CACHELINE_DWORDS);
+}
+
+static int gen9_init_perctx_bb(struct intel_engine_cs *ring,
+ struct i915_wa_ctx_bb *wa_ctx,
+ uint32_t *const batch,
+ uint32_t *offset)
+{
+ struct drm_device *dev = ring->dev;
+ uint32_t index = wa_ctx_start(wa_ctx, *offset, CACHELINE_DWORDS);
+
+ /* WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken:skl,bxt */
+ if ((IS_SKYLAKE(dev) && (INTEL_REVID(dev) <= SKL_REVID_B0)) ||
+ (IS_BROXTON(dev) && (INTEL_REVID(dev) == BXT_REVID_A0))) {
+ wa_ctx_emit(batch, index, MI_LOAD_REGISTER_IMM(1));
+ wa_ctx_emit(batch, index, GEN9_SLICE_COMMON_ECO_CHICKEN0);
+ wa_ctx_emit(batch, index,
+ _MASKED_BIT_ENABLE(DISABLE_PIXEL_MASK_CAMMING));
+ wa_ctx_emit(batch, index, MI_NOOP);
+ }
+
+ /* WaDisableCtxRestoreArbitration:skl,bxt */
+ if ((IS_SKYLAKE(dev) && (INTEL_REVID(dev) <= SKL_REVID_D0)) ||
+ (IS_BROXTON(dev) && (INTEL_REVID(dev) == BXT_REVID_A0)))
+ wa_ctx_emit(batch, index, MI_ARB_ON_OFF | MI_ARB_ENABLE);
+
+ wa_ctx_emit(batch, index, MI_BATCH_BUFFER_END);
return wa_ctx_end(wa_ctx, *offset = index, 1);
}
WARN_ON(ring->id != RCS);
/* update this when WA for higher Gen are added */
- if (WARN(INTEL_INFO(ring->dev)->gen > 8,
- "WA batch buffer is not initialized for Gen%d\n",
- INTEL_INFO(ring->dev)->gen))
+ if (INTEL_INFO(ring->dev)->gen > 9) {
+ DRM_ERROR("WA batch buffer is not initialized for Gen%d\n",
+ INTEL_INFO(ring->dev)->gen);
return 0;
+ }
/* some WA perform writes to scratch page, ensure it is valid */
if (ring->scratch.obj == NULL) {
&offset);
if (ret)
goto out;
+ } else if (INTEL_INFO(ring->dev)->gen == 9) {
+ ret = gen9_init_indirectctx_bb(ring,
+ &wa_ctx->indirect_ctx,
+ batch,
+ &offset);
+ if (ret)
+ goto out;
+
+ ret = gen9_init_perctx_bb(ring,
+ &wa_ctx->per_ctx,
+ batch,
+ &offset);
+ if (ret)
+ goto out;
}
out:
I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | ring->irq_keep_mask));
I915_WRITE(RING_HWSTAM(ring->mmio_base), 0xffffffff);
+ if (ring->status_page.obj) {
+ I915_WRITE(RING_HWS_PGA(ring->mmio_base),
+ (u32)ring->status_page.gfx_addr);
+ POSTING_READ(RING_HWS_PGA(ring->mmio_base));
+ }
+
I915_WRITE(RING_MODE_GEN7(ring),
_MASKED_BIT_DISABLE(GFX_REPLAY_MODE) |
_MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE));
return ret;
/* FIXME(BDW): Address space and security selectors. */
- intel_logical_ring_emit(ringbuf, MI_BATCH_BUFFER_START_GEN8 | (ppgtt<<8));
+ intel_logical_ring_emit(ringbuf, MI_BATCH_BUFFER_START_GEN8 |
+ (ppgtt<<8) |
+ (dispatch_flags & I915_DISPATCH_RS ?
+ MI_BATCH_RESOURCE_STREAMER : 0));
intel_logical_ring_emit(ringbuf, lower_32_bits(offset));
intel_logical_ring_emit(ringbuf, upper_32_bits(offset));
intel_logical_ring_emit(ringbuf, MI_NOOP);
if (ret)
return ret;
+ ret = intel_rcs_context_init_mocs(req);
+ /*
+ * Failing to program the MOCS is non-fatal.The system will not
+ * run at peak performance. So generate an error and carry on.
+ */
+ if (ret)
+ DRM_ERROR("MOCS failed to program: expect performance issues.\n");
+
return intel_lr_context_render_state_init(req);
}
reg_state[CTX_CONTEXT_CONTROL] = RING_CONTEXT_CONTROL(ring);
reg_state[CTX_CONTEXT_CONTROL+1] =
_MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH |
- CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT);
+ CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |
+ CTX_CTRL_RS_CTX_ENABLE);
reg_state[CTX_RING_HEAD] = RING_HEAD(ring->mmio_base);
reg_state[CTX_RING_HEAD+1] = 0;
reg_state[CTX_RING_TAIL] = RING_TAIL(ring->mmio_base);