static phys_addr_t __init i865_stolen_base(int num, int slot, int func,
size_t stolen_size)
{
- u16 toud;
+ u16 toud = 0;
- /*
- * FIXME is the graphics stolen memory region
- * always at TOUD? Ie. is it always the last
- * one to be allocated by the BIOS?
- */
toud = read_pci_config_16(0, 0, 0, I865_TOUD);
- return (phys_addr_t)toud << 16;
+ return (phys_addr_t)(toud << 16) + i845_tseg_size();
}
static phys_addr_t __init gen3_stolen_base(int num, int slot, int func,
# Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
subdir-ccflags-$(CONFIG_DRM_I915_WERROR) := -Werror
+subdir-ccflags-y += \
+ $(call as-instr,movntdqa (%eax)$(comma)%xmm0,-DCONFIG_AS_MOVNTDQA)
# Please keep these build lists sorted!
# core driver code
i915-y := i915_drv.o \
i915_irq.o \
+ i915_memcpy.o \
i915_params.o \
i915_pci.o \
i915_suspend.o \
seq_printf(m, "Current sequence (%s): %x\n",
engine->name, intel_engine_get_seqno(engine));
- seq_printf(m, "Current user interrupts (%s): %lx\n",
- engine->name, READ_ONCE(engine->breadcrumbs.irq_wakeups));
spin_lock(&b->lock);
for (rb = rb_first(&b->waiters); rb; rb = rb_next(rb)) {
engine->hangcheck.seqno,
seqno[id],
engine->last_submitted_seqno);
- seq_printf(m, "\twaiters? %d\n",
- intel_engine_has_waiter(engine));
- seq_printf(m, "\tuser interrupts = %lx [current %lx]\n",
- engine->hangcheck.user_interrupts,
- READ_ONCE(engine->breadcrumbs.irq_wakeups));
+ seq_printf(m, "\twaiters? %s, fake irq active? %s\n",
+ yesno(intel_engine_has_waiter(engine)),
+ yesno(test_bit(engine->id,
+ &dev_priv->gpu_error.missed_irq_rings)));
seq_printf(m, "\tACTHD = 0x%08llx [current 0x%08llx]\n",
(long long)engine->hangcheck.acthd,
(long long)acthd[id]);
struct i915_guc_client *client)
{
struct intel_engine_cs *engine;
+ enum intel_engine_id id;
uint64_t tot = 0;
seq_printf(m, "\tPriority %d, GuC ctx index: %u, PD offset 0x%x\n",
client->wq_size, client->wq_offset, client->wq_tail);
seq_printf(m, "\tWork queue full: %u\n", client->no_wq_space);
- seq_printf(m, "\tFailed to queue: %u\n", client->q_fail);
seq_printf(m, "\tFailed doorbell: %u\n", client->b_fail);
seq_printf(m, "\tLast submission result: %d\n", client->retcode);
- for_each_engine(engine, dev_priv) {
+ for_each_engine_id(engine, dev_priv, id) {
+ u64 submissions = client->submissions[id];
+ tot += submissions;
seq_printf(m, "\tSubmissions: %llu %s\n",
- client->submissions[engine->id],
- engine->name);
- tot += client->submissions[engine->id];
+ submissions, engine->name);
}
seq_printf(m, "\tTotal: %llu\n", tot);
}
struct intel_guc guc;
struct i915_guc_client client = {};
struct intel_engine_cs *engine;
+ enum intel_engine_id id;
u64 total = 0;
if (!HAS_GUC_SCHED(dev_priv))
seq_printf(m, "GuC last action error code: %d\n", guc.action_err);
seq_printf(m, "\nGuC submissions:\n");
- for_each_engine(engine, dev_priv) {
+ for_each_engine_id(engine, dev_priv, id) {
+ u64 submissions = guc.submissions[id];
+ total += submissions;
seq_printf(m, "\t%-24s: %10llu, last seqno 0x%08x\n",
- engine->name, guc.submissions[engine->id],
- guc.last_seqno[engine->id]);
- total += guc.submissions[engine->id];
+ engine->name, submissions, guc.last_seqno[id]);
}
seq_printf(m, "\t%s: %llu\n", "Total", total);
struct drm_device *dev = node->minor->dev;
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_engine_cs *engine;
- int num_rings = hweight32(INTEL_INFO(dev)->ring_mask);
+ int num_rings = INTEL_INFO(dev)->num_rings;
enum intel_engine_id id;
int j, ret;
mutex_init(&dev_priv->wm.wm_mutex);
mutex_init(&dev_priv->pps_mutex);
+ i915_memcpy_init_early(dev_priv);
+
ret = i915_workqueues_init(dev_priv);
if (ret < 0)
return ret;
i915_gem_resume(dev);
i915_restore_state(dev);
+ intel_pps_unlock_regs_wa(dev_priv);
intel_opregion_setup(dev_priv);
intel_init_pch_refclk(dev);
u8 gen;
u16 gen_mask;
u8 ring_mask; /* Rings supported by the HW */
+ u8 num_rings;
DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG, SEP_SEMICOLON);
/* Register offsets for the various display pipes and transcoders */
int pipe_offsets[I915_MAX_TRANSCODERS];
struct i915_suspend_saved_registers {
u32 saveDSPARB;
- u32 saveLVDS;
- u32 savePP_ON_DELAYS;
- u32 savePP_OFF_DELAYS;
- u32 savePP_ON;
- u32 savePP_OFF;
- u32 savePP_CONTROL;
- u32 savePP_DIVISOR;
u32 saveFBC_CONTROL;
u32 saveCACHE_MODE_0;
u32 saveMI_ARB_STATE;
uint32_t psr_mmio_base;
+ uint32_t pps_mmio_base;
+
wait_queue_head_t gmbus_wait_queue;
struct pci_dev *bridge_dev;
enum modeset_restore modeset_restore;
struct mutex modeset_restore_lock;
struct drm_atomic_state *modeset_restore_state;
+ struct drm_modeset_acquire_ctx reset_ctx;
struct list_head vm_list; /* Global list of all address spaces */
struct i915_ggtt ggtt; /* VM representing the global address space */
obj->pages_pin_count--;
}
+enum i915_map_type {
+ I915_MAP_WB = 0,
+ I915_MAP_WC,
+};
+
/**
* i915_gem_object_pin_map - return a contiguous mapping of the entire object
* @obj - the object to map into kernel address space
+ * @type - the type of mapping, used to select pgprot_t
*
* Calls i915_gem_object_pin_pages() to prevent reaping of the object's
* pages and then returns a contiguous mapping of the backing storage into
- * the kernel address space.
+ * the kernel address space. Based on the @type of mapping, the PTE will be
+ * set to either WriteBack or WriteCombine (via pgprot_t).
*
* The caller must hold the struct_mutex, and is responsible for calling
* i915_gem_object_unpin_map() when the mapping is no longer required.
* Returns the pointer through which to access the mapped object, or an
* ERR_PTR() on error.
*/
-void *__must_check i915_gem_object_pin_map(struct drm_i915_gem_object *obj);
+void *__must_check i915_gem_object_pin_map(struct drm_i915_gem_object *obj,
+ enum i915_map_type type);
/**
* i915_gem_object_unpin_map - releases an earlier mapping
* is woken.
*/
if (engine->irq_seqno_barrier &&
- READ_ONCE(engine->breadcrumbs.irq_seqno_bh) == current &&
+ rcu_access_pointer(engine->breadcrumbs.irq_seqno_bh) == current &&
cmpxchg_relaxed(&engine->breadcrumbs.irq_posted, 1, 0)) {
struct task_struct *tsk;
* irq_posted == false but we are still running).
*/
rcu_read_lock();
- tsk = READ_ONCE(engine->breadcrumbs.irq_seqno_bh);
+ tsk = rcu_dereference(engine->breadcrumbs.irq_seqno_bh);
if (tsk && tsk != current)
/* Note that if the bottom-half is changed as we
* are sending the wake-up, the new bottom-half will
return false;
}
+void i915_memcpy_init_early(struct drm_i915_private *dev_priv);
+bool i915_memcpy_from_wc(void *dst, const void *src, unsigned long len);
+
+#define ptr_unpack_bits(ptr, bits) ({ \
+ unsigned long __v = (unsigned long)(ptr); \
+ (bits) = __v & ~PAGE_MASK; \
+ (typeof(ptr))(__v & PAGE_MASK); \
+})
+
+#define ptr_pack_bits(ptr, bits) \
+ ((typeof(ptr))((unsigned long)(ptr) | (bits)))
+
#endif
.release = i915_gem_object_release_phys,
};
-int
-i915_gem_object_unbind(struct drm_i915_gem_object *obj)
+int i915_gem_object_unbind(struct drm_i915_gem_object *obj)
{
struct i915_vma *vma;
LIST_HEAD(still_in_list);
int ret;
- /* The vma will only be freed if it is marked as closed, and if we wait
- * upon rendering to the vma, we may unbind anything in the list.
+ lockdep_assert_held(&obj->base.dev->struct_mutex);
+
+ /* Closed vma are removed from the obj->vma_list - but they may
+ * still have an active binding on the object. To remove those we
+ * must wait for all rendering to complete to the object (as unbinding
+ * must anyway), and retire the requests.
*/
+ ret = i915_gem_object_wait_rendering(obj, false);
+ if (ret)
+ return ret;
+
+ i915_gem_retire_requests(to_i915(obj->base.dev));
+
while ((vma = list_first_entry_or_null(&obj->vma_list,
struct i915_vma,
obj_link))) {
list_del(&obj->global_list);
if (obj->mapping) {
+ /* low bits are ignored by is_vmalloc_addr and kmap_to_page */
if (is_vmalloc_addr(obj->mapping))
vunmap(obj->mapping);
else
}
/* The 'mapping' part of i915_gem_object_pin_map() below */
-static void *i915_gem_object_map(const struct drm_i915_gem_object *obj)
+static void *i915_gem_object_map(const struct drm_i915_gem_object *obj,
+ enum i915_map_type type)
{
unsigned long n_pages = obj->base.size >> PAGE_SHIFT;
struct sg_table *sgt = obj->pages;
struct page *stack_pages[32];
struct page **pages = stack_pages;
unsigned long i = 0;
+ pgprot_t pgprot;
void *addr;
/* A single page can always be kmapped */
- if (n_pages == 1)
+ if (n_pages == 1 && type == I915_MAP_WB)
return kmap(sg_page(sgt->sgl));
if (n_pages > ARRAY_SIZE(stack_pages)) {
/* Check that we have the expected number of pages */
GEM_BUG_ON(i != n_pages);
- addr = vmap(pages, n_pages, 0, PAGE_KERNEL);
+ switch (type) {
+ case I915_MAP_WB:
+ pgprot = PAGE_KERNEL;
+ break;
+ case I915_MAP_WC:
+ pgprot = pgprot_writecombine(PAGE_KERNEL_IO);
+ break;
+ }
+ addr = vmap(pages, n_pages, 0, pgprot);
if (pages != stack_pages)
drm_free_large(pages);
}
/* get, pin, and map the pages of the object into kernel space */
-void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj)
+void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj,
+ enum i915_map_type type)
{
+ enum i915_map_type has_type;
+ bool pinned;
+ void *ptr;
int ret;
lockdep_assert_held(&obj->base.dev->struct_mutex);
+ GEM_BUG_ON(!i915_gem_object_has_struct_page(obj));
ret = i915_gem_object_get_pages(obj);
if (ret)
return ERR_PTR(ret);
i915_gem_object_pin_pages(obj);
+ pinned = obj->pages_pin_count > 1;
- if (!obj->mapping) {
- obj->mapping = i915_gem_object_map(obj);
- if (!obj->mapping) {
- i915_gem_object_unpin_pages(obj);
- return ERR_PTR(-ENOMEM);
+ ptr = ptr_unpack_bits(obj->mapping, has_type);
+ if (ptr && has_type != type) {
+ if (pinned) {
+ ret = -EBUSY;
+ goto err;
+ }
+
+ if (is_vmalloc_addr(ptr))
+ vunmap(ptr);
+ else
+ kunmap(kmap_to_page(ptr));
+
+ ptr = obj->mapping = NULL;
+ }
+
+ if (!ptr) {
+ ptr = i915_gem_object_map(obj, type);
+ if (!ptr) {
+ ret = -ENOMEM;
+ goto err;
}
+
+ obj->mapping = ptr_pack_bits(ptr, type);
}
- return obj->mapping;
+ return ptr;
+
+err:
+ i915_gem_object_unpin_pages(obj);
+ return ERR_PTR(ret);
}
static void
struct drm_i915_gem_request *request;
struct intel_ring *ring;
- request = i915_gem_active_peek(&engine->last_request,
- &engine->i915->drm.struct_mutex);
-
/* Mark all pending requests as complete so that any concurrent
* (lockless) lookup doesn't try and wait upon the request as we
* reset it.
*/
- if (request)
- intel_engine_init_seqno(engine, request->fence.seqno);
+ intel_engine_init_seqno(engine, engine->last_submitted_seqno);
/*
* Clear the execlists queue up before freeing the requests, as those
* implicit references on things like e.g. ppgtt address spaces through
* the request.
*/
+ request = i915_gem_active_raw(&engine->last_request,
+ &engine->i915->drm.struct_mutex);
if (request)
i915_gem_request_retire_upto(request);
GEM_BUG_ON(intel_engine_is_active(engine));
container_of(work, typeof(*dev_priv), gt.idle_work.work);
struct drm_device *dev = &dev_priv->drm;
struct intel_engine_cs *engine;
- unsigned int stuck_engines;
bool rearm_hangcheck;
if (!READ_ONCE(dev_priv->gt.awake))
dev_priv->gt.awake = false;
rearm_hangcheck = false;
- /* As we have disabled hangcheck, we need to unstick any waiters still
- * hanging around. However, as we may be racing against the interrupt
- * handler or the waiters themselves, we skip enabling the fake-irq.
- */
- stuck_engines = intel_kick_waiters(dev_priv);
- if (unlikely(stuck_engines))
- DRM_DEBUG_DRIVER("kicked stuck waiters (%x)...missed irq?\n",
- stuck_engines);
-
if (INTEL_GEN(dev_priv) >= 6)
gen6_rps_idle(dev_priv);
intel_runtime_pm_put(dev_priv);
i915_vma_unpin(i915_gem_obj_to_ggtt_view(obj, view));
}
-static __always_inline unsigned __busy_read_flag(unsigned int id)
+static __always_inline unsigned int __busy_read_flag(unsigned int id)
{
/* Note that we could alias engines in the execbuf API, but
* that would be very unwise as it prevents userspace from
static __always_inline unsigned int __busy_write_id(unsigned int id)
{
- return id;
+ /* The uABI guarantees an active writer is also amongst the read
+ * engines. This would be true if we accessed the activity tracking
+ * under the lock, but as we perform the lookup of the object and
+ * its activity locklessly we can not guarantee that the last_write
+ * being active implies that we have set the same engine flag from
+ * last_read - hence we always set both read and write busy for
+ * last_write.
+ */
+ return id | __busy_read_flag(id);
}
-static __always_inline unsigned
+static __always_inline unsigned int
__busy_set_if_active(const struct i915_gem_active *active,
unsigned int (*flag)(unsigned int id))
{
id = request->engine->exec_id;
- /* Check that the pointer wasn't reassigned and overwritten. */
+ /* Check that the pointer wasn't reassigned and overwritten.
+ *
+ * In __i915_gem_active_get_rcu(), we enforce ordering between
+ * the first rcu pointer dereference (imposing a
+ * read-dependency only on access through the pointer) and
+ * the second lockless access through the memory barrier
+ * following a successful atomic_inc_not_zero(). Here there
+ * is no such barrier, and so we must manually insert an
+ * explicit read barrier to ensure that the following
+ * access occurs after all the loads through the first
+ * pointer.
+ *
+ * It is worth comparing this sequence with
+ * raw_write_seqcount_latch() which operates very similarly.
+ * The challenge here is the visibility of the other CPU
+ * writes to the reallocated request vs the local CPU ordering.
+ * Before the other CPU can overwrite the request, it will
+ * have updated our active->request and gone through a wmb.
+ * During the read here, we want to make sure that the values
+ * we see have not been overwritten as we do so - and we do
+ * that by serialising the second pointer check with the writes
+ * on other other CPUs.
+ *
+ * The corresponding write barrier is part of
+ * rcu_assign_pointer().
+ */
+ smp_rmb();
if (request == rcu_access_pointer(active->request))
return flag(id);
} while (1);
}
-static inline unsigned
+static __always_inline unsigned int
busy_check_reader(const struct i915_gem_active *active)
{
return __busy_set_if_active(active, __busy_read_flag);
}
-static inline unsigned
+static __always_inline unsigned int
busy_check_writer(const struct i915_gem_active *active)
{
return __busy_set_if_active(active, __busy_write_id);
args->busy |= busy_check_reader(&obj->last_read[idx]);
/* For ABI sanity, we only care that the write engine is in
- * the set of read engines. This is ensured by the ordering
- * of setting last_read/last_write in i915_vma_move_to_active,
- * and then in reverse in retire.
+ * the set of read engines. This should be ensured by the
+ * ordering of setting last_read/last_write in
+ * i915_vma_move_to_active(), and then in reverse in retire.
+ * However, for good measure, we always report the last_write
+ * request as a busy read as well as being a busy write.
*
* We don't care that the set of active read/write engines
* may change during construction of the result, as it is
const int num_rings =
/* Use an extended w/a on ivb+ if signalling from other rings */
i915.semaphores ?
- hweight32(INTEL_INFO(dev_priv)->ring_mask) - 1 :
+ INTEL_INFO(dev_priv)->num_rings - 1 :
0;
int len, ret;
if (ret)
return ERR_PTR(ret);
- addr = i915_gem_object_pin_map(obj);
+ addr = i915_gem_object_pin_map(obj, I915_MAP_WB);
mutex_unlock(&dev->struct_mutex);
return addr;
goto err_batch_unpin;
}
+ /* Whilst this request exists, batch_obj will be on the
+ * active_list, and so will hold the active reference. Only when this
+ * request is retired will the the batch_obj be moved onto the
+ * inactive_list and lose its active reference. Hence we do not need
+ * to explicitly hold another reference here.
+ */
+ params->request->batch_obj = params->batch->obj;
+
ret = i915_gem_request_add_to_client(params->request, file);
if (ret)
goto err_request;
ret = execbuf_submit(params, args, &eb->vmas);
err_request:
- __i915_add_request(params->request, params->batch->obj, ret == 0);
+ __i915_add_request(params->request, ret == 0);
err_batch_unpin:
/*
}
static struct sg_table *
-intel_rotate_fb_obj_pages(struct intel_rotation_info *rot_info,
+intel_rotate_fb_obj_pages(const struct intel_rotation_info *rot_info,
struct drm_i915_gem_object *obj)
{
const size_t n_pages = obj->base.size / PAGE_SIZE;
- unsigned int size_pages = rot_info->plane[0].width * rot_info->plane[0].height;
- unsigned int size_pages_uv;
+ unsigned int size = intel_rotation_info_size(rot_info);
struct sgt_iter sgt_iter;
dma_addr_t dma_addr;
unsigned long i;
dma_addr_t *page_addr_list;
struct sg_table *st;
- unsigned int uv_start_page;
struct scatterlist *sg;
int ret = -ENOMEM;
if (!page_addr_list)
return ERR_PTR(ret);
- /* Account for UV plane with NV12. */
- if (rot_info->pixel_format == DRM_FORMAT_NV12)
- size_pages_uv = rot_info->plane[1].width * rot_info->plane[1].height;
- else
- size_pages_uv = 0;
-
/* Allocate target SG list. */
st = kmalloc(sizeof(*st), GFP_KERNEL);
if (!st)
goto err_st_alloc;
- ret = sg_alloc_table(st, size_pages + size_pages_uv, GFP_KERNEL);
+ ret = sg_alloc_table(st, size, GFP_KERNEL);
if (ret)
goto err_sg_alloc;
st->nents = 0;
sg = st->sgl;
- /* Rotate the pages. */
- sg = rotate_pages(page_addr_list, 0,
- rot_info->plane[0].width, rot_info->plane[0].height,
- rot_info->plane[0].width,
- st, sg);
-
- /* Append the UV plane if NV12. */
- if (rot_info->pixel_format == DRM_FORMAT_NV12) {
- uv_start_page = size_pages;
-
- /* Check for tile-row un-alignment. */
- if (offset_in_page(rot_info->uv_offset))
- uv_start_page--;
-
- rot_info->uv_start_page = uv_start_page;
-
- sg = rotate_pages(page_addr_list, rot_info->uv_start_page,
- rot_info->plane[1].width, rot_info->plane[1].height,
- rot_info->plane[1].width,
- st, sg);
+ for (i = 0 ; i < ARRAY_SIZE(rot_info->plane); i++) {
+ sg = rotate_pages(page_addr_list, rot_info->plane[i].offset,
+ rot_info->plane[i].width, rot_info->plane[i].height,
+ rot_info->plane[i].stride, st, sg);
}
- DRM_DEBUG_KMS("Created rotated page mapping for object size %zu (%ux%u tiles, %u pages (%u plane 0)).\n",
- obj->base.size, rot_info->plane[0].width,
- rot_info->plane[0].height, size_pages + size_pages_uv,
- size_pages);
+ DRM_DEBUG_KMS("Created rotated page mapping for object size %zu (%ux%u tiles, %u pages)\n",
+ obj->base.size, rot_info->plane[0].width, rot_info->plane[0].height, size);
drm_free_large(page_addr_list);
err_st_alloc:
drm_free_large(page_addr_list);
- DRM_DEBUG_KMS("Failed to create rotated mapping for object size %zu! (%d) (%ux%u tiles, %u pages (%u plane 0))\n",
- obj->base.size, ret, rot_info->plane[0].width,
- rot_info->plane[0].height, size_pages + size_pages_uv,
- size_pages);
+ DRM_DEBUG_KMS("Failed to create rotated mapping for object size %zu! (%ux%u tiles, %u pages)\n",
+ obj->base.size, rot_info->plane[0].width, rot_info->plane[0].height, size);
+
return ERR_PTR(ret);
}
};
struct intel_rotation_info {
- unsigned int uv_offset;
- uint32_t pixel_format;
- unsigned int uv_start_page;
struct {
/* tiles */
- unsigned int width, height;
+ unsigned int width, height, stride, offset;
} plane[2];
};
if (req && i915_gem_request_completed(req))
i915_gem_request_retire(req);
- req = kmem_cache_zalloc(dev_priv->requests, GFP_KERNEL);
+ /* Beware: Dragons be flying overhead.
+ *
+ * We use RCU to look up requests in flight. The lookups may
+ * race with the request being allocated from the slab freelist.
+ * That is the request we are writing to here, may be in the process
+ * of being read by __i915_gem_active_get_rcu(). As such,
+ * we have to be very careful when overwriting the contents. During
+ * the RCU lookup, we change chase the request->engine pointer,
+ * read the request->fence.seqno and increment the reference count.
+ *
+ * The reference count is incremented atomically. If it is zero,
+ * the lookup knows the request is unallocated and complete. Otherwise,
+ * it is either still in use, or has been reallocated and reset
+ * with fence_init(). This increment is safe for release as we check
+ * that the request we have a reference to and matches the active
+ * request.
+ *
+ * Before we increment the refcount, we chase the request->engine
+ * pointer. We must not call kmem_cache_zalloc() or else we set
+ * that pointer to NULL and cause a crash during the lookup. If
+ * we see the request is completed (based on the value of the
+ * old engine and seqno), the lookup is complete and reports NULL.
+ * If we decide the request is not completed (new engine or seqno),
+ * then we grab a reference and double check that it is still the
+ * active request - which it won't be and restart the lookup.
+ *
+ * Do not use kmem_cache_zalloc() here!
+ */
+ req = kmem_cache_alloc(dev_priv->requests, GFP_KERNEL);
if (!req)
return ERR_PTR(-ENOMEM);
req->engine = engine;
req->ctx = i915_gem_context_get(ctx);
+ /* No zalloc, must clear what we need by hand */
+ req->previous_context = NULL;
+ req->file_priv = NULL;
+ req->batch_obj = NULL;
+ req->pid = NULL;
+ req->elsp_submitted = 0;
+
/*
* Reserve space in the ring buffer for all the commands required to
* eventually emit this request. This is to guarantee that the
* request is not being tracked for completion but the work itself is
* going to happen on the hardware. This would be a Bad Thing(tm).
*/
-void __i915_add_request(struct drm_i915_gem_request *request,
- struct drm_i915_gem_object *obj,
- bool flush_caches)
+void __i915_add_request(struct drm_i915_gem_request *request, bool flush_caches)
{
struct intel_engine_cs *engine;
struct intel_ring *ring;
request->head = request_start;
- /* Whilst this request exists, batch_obj will be on the
- * active_list, and so will hold the active reference. Only when this
- * request is retired will the the batch_obj be moved onto the
- * inactive_list and lose its active reference. Hence we do not need
- * to explicitly hold another reference here.
- */
- request->batch_obj = obj;
-
/* Seal the request and mark it as pending execution. Note that
* we may inspect this state, without holding any locks, during
* hangcheck. Hence we apply the barrier to ensure that we do not
* emission time to be associated with the request for tracking how far ahead
* of the GPU the submission is.
*
+ * When modifying this structure be very aware that we perform a lockless
+ * RCU lookup of it that may race against reallocation of the struct
+ * from the slab freelist. We intentionally do not zero the structure on
+ * allocation so that the lookup can use the dangling pointers (and is
+ * cogniscent that those pointers may be wrong). Instead, everything that
+ * needs to be initialised must be done so explicitly.
+ *
* The requests are reference counted.
*/
struct drm_i915_gem_request {
*pdst = src;
}
-void __i915_add_request(struct drm_i915_gem_request *req,
- struct drm_i915_gem_object *batch_obj,
- bool flush_caches);
+void __i915_add_request(struct drm_i915_gem_request *req, bool flush_caches);
#define i915_add_request(req) \
- __i915_add_request(req, NULL, true)
+ __i915_add_request(req, true)
#define i915_add_request_no_flush(req) \
- __i915_add_request(req, NULL, false)
+ __i915_add_request(req, false)
struct intel_rps_client;
#define NO_WAITBOOST ERR_PTR(-1)
}
/**
- * i915_gem_active_peek - report the active request being monitored
+ * i915_gem_active_raw - return the active request
* @active - the active tracker
*
- * i915_gem_active_peek() returns the current request being tracked if
- * still active, or NULL. It does not obtain a reference on the request
- * for the caller, so the caller must hold struct_mutex.
+ * i915_gem_active_raw() returns the current request being tracked, or NULL.
+ * It does not obtain a reference on the request for the caller, so the caller
+ * must hold struct_mutex.
*/
static inline struct drm_i915_gem_request *
-i915_gem_active_peek(const struct i915_gem_active *active, struct mutex *mutex)
+i915_gem_active_raw(const struct i915_gem_active *active, struct mutex *mutex)
{
- struct drm_i915_gem_request *request;
-
- request = rcu_dereference_protected(active->request,
- lockdep_is_held(mutex));
- if (!request || i915_gem_request_completed(request))
- return NULL;
-
- return request;
+ return rcu_dereference_protected(active->request,
+ lockdep_is_held(mutex));
}
/**
- * i915_gem_active_peek_rcu - report the active request being monitored
+ * i915_gem_active_peek - report the active request being monitored
* @active - the active tracker
*
- * i915_gem_active_peek_rcu() returns the current request being tracked if
+ * i915_gem_active_peek() returns the current request being tracked if
* still active, or NULL. It does not obtain a reference on the request
- * for the caller, and inspection of the request is only valid under
- * the RCU lock.
+ * for the caller, so the caller must hold struct_mutex.
*/
static inline struct drm_i915_gem_request *
-i915_gem_active_peek_rcu(const struct i915_gem_active *active)
+i915_gem_active_peek(const struct i915_gem_active *active, struct mutex *mutex)
{
struct drm_i915_gem_request *request;
- request = rcu_dereference(active->request);
+ request = i915_gem_active_raw(active, mutex);
if (!request || i915_gem_request_completed(request))
return NULL;
* just report the active tracker is idle. If the new request is
* incomplete, then we acquire a reference on it and check that
* it remained the active request.
+ *
+ * It is then imperative that we do not zero the request on
+ * reallocation, so that we can chase the dangling pointers!
+ * See i915_gem_request_alloc().
*/
do {
struct drm_i915_gem_request *request;
* incremented) then the following read for rcu_access_pointer()
* must occur after the atomic operation and so confirm
* that this request is the one currently being tracked.
+ *
+ * The corresponding write barrier is part of
+ * rcu_assign_pointer().
*/
if (!request || request == rcu_access_pointer(active->request))
return rcu_pointer_handoff(request);
struct drm_i915_gem_request *request;
int ret;
- request = rcu_dereference_protected(active->request,
- lockdep_is_held(mutex));
+ request = i915_gem_active_raw(active, mutex);
if (!request)
return 0;
base = bsm & INTEL_BSM_MASK;
} else if (IS_I865G(dev)) {
+ u32 tseg_size = 0;
u16 toud = 0;
+ u8 tmp;
+
+ pci_bus_read_config_byte(dev->pdev->bus, PCI_DEVFN(0, 0),
+ I845_ESMRAMC, &tmp);
+
+ if (tmp & TSEG_ENABLE) {
+ switch (tmp & I845_TSEG_SIZE_MASK) {
+ case I845_TSEG_SIZE_512K:
+ tseg_size = KB(512);
+ break;
+ case I845_TSEG_SIZE_1M:
+ tseg_size = MB(1);
+ break;
+ }
+ }
- /*
- * FIXME is the graphics stolen memory region
- * always at TOUD? Ie. is it always the last
- * one to be allocated by the BIOS?
- */
pci_bus_read_config_word(dev->pdev->bus, PCI_DEVFN(0, 0),
I865_TOUD, &toud);
- base = toud << 16;
+ base = (toud << 16) + tseg_size;
} else if (IS_I85X(dev)) {
u32 tseg_size = 0;
u32 tom;
desc.priority = client->priority;
desc.db_id = client->doorbell_id;
- for_each_engine(engine, dev_priv) {
+ for_each_engine_masked(engine, dev_priv, client->engines) {
struct intel_context *ce = &ctx->engine[engine->id];
- struct guc_execlist_context *lrc = &desc.lrc[engine->guc_id];
+ uint32_t guc_engine_id = engine->guc_id;
+ struct guc_execlist_context *lrc = &desc.lrc[guc_engine_id];
struct drm_i915_gem_object *obj;
/* TODO: We have a design issue to be solved here. Only when we
gfx_addr = i915_gem_obj_ggtt_offset(ce->state);
lrc->ring_lcra = gfx_addr + LRC_STATE_PN * PAGE_SIZE;
lrc->context_id = (client->ctx_index << GUC_ELC_CTXID_OFFSET) |
- (engine->guc_id << GUC_ELC_ENGINE_OFFSET);
+ (guc_engine_id << GUC_ELC_ENGINE_OFFSET);
obj = ce->ring->obj;
gfx_addr = i915_gem_obj_ggtt_offset(obj);
lrc->ring_next_free_location = gfx_addr;
lrc->ring_current_tail_pointer_value = 0;
- desc.engines_used |= (1 << engine->guc_id);
+ desc.engines_used |= (1 << guc_engine_id);
}
+ DRM_DEBUG_DRIVER("Host engines 0x%x => GuC engines used 0x%x\n",
+ client->engines, desc.engines_used);
WARN_ON(desc.engines_used == 0);
/*
/* wqi_len is in DWords, and does not include the one-word header */
const size_t wqi_size = sizeof(struct guc_wq_item);
const u32 wqi_len = wqi_size/sizeof(u32) - 1;
+ struct intel_engine_cs *engine = rq->engine;
struct guc_process_desc *desc;
struct guc_wq_item *wqi;
void *base;
/* Now fill in the 4-word work queue item */
wqi->header = WQ_TYPE_INORDER |
(wqi_len << WQ_LEN_SHIFT) |
- (rq->engine->guc_id << WQ_TARGET_SHIFT) |
+ (engine->guc_id << WQ_TARGET_SHIFT) |
WQ_NO_WCFLUSH_WAIT;
/* The GuC wants only the low-order word of the context descriptor */
- wqi->context_desc = (u32)intel_lr_context_descriptor(rq->ctx,
- rq->engine);
+ wqi->context_desc = (u32)intel_lr_context_descriptor(rq->ctx, engine);
wqi->ring_tail = tail << WQ_RING_TAIL_SHIFT;
wqi->fence_id = rq->fence.seqno;
kfree(client);
}
+/* Check that a doorbell register is in the expected state */
+static bool guc_doorbell_check(struct intel_guc *guc, uint16_t db_id)
+{
+ struct drm_i915_private *dev_priv = guc_to_i915(guc);
+ i915_reg_t drbreg = GEN8_DRBREGL(db_id);
+ uint32_t value = I915_READ(drbreg);
+ bool enabled = (value & GUC_DOORBELL_ENABLED) != 0;
+ bool expected = test_bit(db_id, guc->doorbell_bitmap);
+
+ if (enabled == expected)
+ return true;
+
+ DRM_DEBUG_DRIVER("Doorbell %d (reg 0x%x) 0x%x, should be %s\n",
+ db_id, drbreg.reg, value,
+ expected ? "active" : "inactive");
+
+ return false;
+}
+
/*
- * Borrow the first client to set up & tear down every doorbell
+ * Borrow the first client to set up & tear down each unused doorbell
* in turn, to ensure that all doorbell h/w is (re)initialised.
*/
static void guc_init_doorbell_hw(struct intel_guc *guc)
{
- struct drm_i915_private *dev_priv = guc_to_i915(guc);
struct i915_guc_client *client = guc->execbuf_client;
- uint16_t db_id, i;
- int err;
+ uint16_t db_id;
+ int i, err;
+ /* Save client's original doorbell selection */
db_id = client->doorbell_id;
for (i = 0; i < GUC_MAX_DOORBELLS; ++i) {
- i915_reg_t drbreg = GEN8_DRBREGL(i);
- u32 value = I915_READ(drbreg);
+ /* Skip if doorbell is OK */
+ if (guc_doorbell_check(guc, i))
+ continue;
err = guc_update_doorbell_id(guc, client, i);
-
- /* Report update failure or unexpectedly active doorbell */
- if (err || (i != db_id && (value & GUC_DOORBELL_ENABLED)))
- DRM_DEBUG_DRIVER("Doorbell %d (reg 0x%x) was 0x%x, err %d\n",
- i, drbreg.reg, value, err);
+ if (err)
+ DRM_DEBUG_DRIVER("Doorbell %d update failed, err %d\n",
+ i, err);
}
/* Restore to original value */
DRM_ERROR("Failed to restore doorbell to %d, err %d\n",
db_id, err);
- for (i = 0; i < GUC_MAX_DOORBELLS; ++i) {
- i915_reg_t drbreg = GEN8_DRBREGL(i);
- u32 value = I915_READ(drbreg);
-
- if (i != db_id && (value & GUC_DOORBELL_ENABLED))
- DRM_DEBUG_DRIVER("Doorbell %d (reg 0x%x) finally 0x%x\n",
- i, drbreg.reg, value);
-
- }
+ /* Read back & verify all doorbell registers */
+ for (i = 0; i < GUC_MAX_DOORBELLS; ++i)
+ (void)guc_doorbell_check(guc, i);
}
/**
*/
static struct i915_guc_client *
guc_client_alloc(struct drm_i915_private *dev_priv,
+ uint32_t engines,
uint32_t priority,
struct i915_gem_context *ctx)
{
if (!client)
return NULL;
- client->doorbell_id = GUC_INVALID_DOORBELL_ID;
- client->priority = priority;
client->owner = ctx;
client->guc = guc;
+ client->engines = engines;
+ client->priority = priority;
+ client->doorbell_id = GUC_INVALID_DOORBELL_ID;
client->ctx_index = (uint32_t)ida_simple_get(&guc->ctx_ids, 0,
GUC_MAX_GPU_CONTEXTS, GFP_KERNEL);
if (guc_init_doorbell(guc, client, db_id))
goto err;
- DRM_DEBUG_DRIVER("new priority %u client %p: ctx_index %u\n",
- priority, client, client->ctx_index);
+ DRM_DEBUG_DRIVER("new priority %u client %p for engine(s) 0x%x: ctx_index %u\n",
+ priority, client, client->engines, client->ctx_index);
DRM_DEBUG_DRIVER("doorbell id %u, cacheline offset 0x%x\n",
client->doorbell_id, client->doorbell_offset);
/* client for execbuf submission */
client = guc_client_alloc(dev_priv,
+ INTEL_INFO(dev_priv)->ring_mask,
GUC_CTX_PRIORITY_KMD_NORMAL,
dev_priv->kernel_context);
if (!client) {
static void notify_ring(struct intel_engine_cs *engine)
{
smp_store_mb(engine->breadcrumbs.irq_posted, true);
- if (intel_engine_wakeup(engine)) {
+ if (intel_engine_wakeup(engine))
trace_i915_gem_request_notify(engine);
- engine->breadcrumbs.irq_wakeups++;
- }
}
static void vlv_c0_read(struct drm_i915_private *dev_priv,
return HANGCHECK_HUNG;
}
-static unsigned long kick_waiters(struct intel_engine_cs *engine)
-{
- struct drm_i915_private *i915 = engine->i915;
- unsigned long irq_count = READ_ONCE(engine->breadcrumbs.irq_wakeups);
-
- if (engine->hangcheck.user_interrupts == irq_count &&
- !test_and_set_bit(engine->id, &i915->gpu_error.missed_irq_rings)) {
- if (!test_bit(engine->id, &i915->gpu_error.test_irq_rings))
- DRM_ERROR("Hangcheck timer elapsed... %s idle\n",
- engine->name);
-
- intel_engine_enable_fake_irq(engine);
- }
-
- return irq_count;
-}
/*
* This is called when the chip hasn't reported back with completed
* batchbuffers in a long time. We keep track per ring seqno progress and
bool busy = intel_engine_has_waiter(engine);
u64 acthd;
u32 seqno;
- unsigned user_interrupts;
semaphore_clear_deadlocks(dev_priv);
acthd = intel_engine_get_active_head(engine);
seqno = intel_engine_get_seqno(engine);
- /* Reset stuck interrupts between batch advances */
- user_interrupts = 0;
-
if (engine->hangcheck.seqno == seqno) {
if (!intel_engine_is_active(engine)) {
engine->hangcheck.action = HANGCHECK_IDLE;
if (busy) {
/* Safeguard against driver failure */
- user_interrupts = kick_waiters(engine);
engine->hangcheck.score += BUSY;
}
} else {
engine->hangcheck.seqno = seqno;
engine->hangcheck.acthd = acthd;
- engine->hangcheck.user_interrupts = user_interrupts;
busy_count += busy;
}
--- /dev/null
+/*
+ * Copyright © 2016 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <asm/fpu/api.h>
+
+#include "i915_drv.h"
+
+DEFINE_STATIC_KEY_FALSE(has_movntdqa);
+
+#ifdef CONFIG_AS_MOVNTDQA
+static void __memcpy_ntdqa(void *dst, const void *src, unsigned long len)
+{
+ kernel_fpu_begin();
+
+ len >>= 4;
+ while (len >= 4) {
+ asm("movntdqa (%0), %%xmm0\n"
+ "movntdqa 16(%0), %%xmm1\n"
+ "movntdqa 32(%0), %%xmm2\n"
+ "movntdqa 48(%0), %%xmm3\n"
+ "movaps %%xmm0, (%1)\n"
+ "movaps %%xmm1, 16(%1)\n"
+ "movaps %%xmm2, 32(%1)\n"
+ "movaps %%xmm3, 48(%1)\n"
+ :: "r" (src), "r" (dst) : "memory");
+ src += 64;
+ dst += 64;
+ len -= 4;
+ }
+ while (len--) {
+ asm("movntdqa (%0), %%xmm0\n"
+ "movaps %%xmm0, (%1)\n"
+ :: "r" (src), "r" (dst) : "memory");
+ src += 16;
+ dst += 16;
+ }
+
+ kernel_fpu_end();
+}
+#endif
+
+/**
+ * i915_memcpy_from_wc: perform an accelerated *aligned* read from WC
+ * @dst: destination pointer
+ * @src: source pointer
+ * @len: how many bytes to copy
+ *
+ * i915_memcpy_from_wc copies @len bytes from @src to @dst using
+ * non-temporal instructions where available. Note that all arguments
+ * (@src, @dst) must be aligned to 16 bytes and @len must be a multiple
+ * of 16.
+ *
+ * To test whether accelerated reads from WC are supported, use
+ * i915_memcpy_from_wc(NULL, NULL, 0);
+ *
+ * Returns true if the copy was successful, false if the preconditions
+ * are not met.
+ */
+bool i915_memcpy_from_wc(void *dst, const void *src, unsigned long len)
+{
+ if (unlikely(((unsigned long)dst | (unsigned long)src | len) & 15))
+ return false;
+
+#ifdef CONFIG_AS_MOVNTDQA
+ if (static_branch_likely(&has_movntdqa)) {
+ if (likely(len))
+ __memcpy_ntdqa(dst, src, len);
+ return true;
+ }
+#endif
+
+ return false;
+}
+
+void i915_memcpy_init_early(struct drm_i915_private *dev_priv)
+{
+ if (static_cpu_has(X86_FEATURE_XMM4_1))
+ static_branch_enable(&has_movntdqa);
+}
.fastboot = 0,
.prefault_disable = 0,
.load_detect_test = 0,
+ .force_reset_modeset_test = 0,
.reset = true,
.invert_brightness = 0,
.disable_display = 0,
"Force-enable the VGA load detect code for testing (default:false). "
"For developers only.");
+module_param_named_unsafe(force_reset_modeset_test, i915.force_reset_modeset_test, bool, 0600);
+MODULE_PARM_DESC(force_reset_modeset_test,
+ "Force a modeset during gpu reset for testing (default:false). "
+ "For developers only.");
+
module_param_named_unsafe(invert_brightness, i915.invert_brightness, int, 0600);
MODULE_PARM_DESC(invert_brightness,
"Invert backlight brightness "
bool fastboot;
bool prefault_disable;
bool load_detect_test;
+ bool force_reset_modeset_test;
bool reset;
bool disable_display;
bool verbose_state_checks;
#define VIDEO_DIP_ENABLE_SPD_HSW (1 << 0)
/* Panel power sequencing */
-#define PP_STATUS _MMIO(0x61200)
-#define PP_ON (1 << 31)
+#define PPS_BASE 0x61200
+#define VLV_PPS_BASE (VLV_DISPLAY_BASE + PPS_BASE)
+#define PCH_PPS_BASE 0xC7200
+
+#define _MMIO_PPS(pps_idx, reg) _MMIO(dev_priv->pps_mmio_base - \
+ PPS_BASE + (reg) + \
+ (pps_idx) * 0x100)
+
+#define _PP_STATUS 0x61200
+#define PP_STATUS(pps_idx) _MMIO_PPS(pps_idx, _PP_STATUS)
+#define PP_ON (1 << 31)
/*
* Indicates that all dependencies of the panel are on:
*
* - pipe enabled
* - LVDS/DVOB/DVOC on
*/
-#define PP_READY (1 << 30)
-#define PP_SEQUENCE_NONE (0 << 28)
-#define PP_SEQUENCE_POWER_UP (1 << 28)
-#define PP_SEQUENCE_POWER_DOWN (2 << 28)
-#define PP_SEQUENCE_MASK (3 << 28)
-#define PP_SEQUENCE_SHIFT 28
-#define PP_CYCLE_DELAY_ACTIVE (1 << 27)
-#define PP_SEQUENCE_STATE_MASK 0x0000000f
+#define PP_READY (1 << 30)
+#define PP_SEQUENCE_NONE (0 << 28)
+#define PP_SEQUENCE_POWER_UP (1 << 28)
+#define PP_SEQUENCE_POWER_DOWN (2 << 28)
+#define PP_SEQUENCE_MASK (3 << 28)
+#define PP_SEQUENCE_SHIFT 28
+#define PP_CYCLE_DELAY_ACTIVE (1 << 27)
+#define PP_SEQUENCE_STATE_MASK 0x0000000f
#define PP_SEQUENCE_STATE_OFF_IDLE (0x0 << 0)
#define PP_SEQUENCE_STATE_OFF_S0_1 (0x1 << 0)
#define PP_SEQUENCE_STATE_OFF_S0_2 (0x2 << 0)
#define PP_SEQUENCE_STATE_ON_S1_2 (0xa << 0)
#define PP_SEQUENCE_STATE_ON_S1_3 (0xb << 0)
#define PP_SEQUENCE_STATE_RESET (0xf << 0)
-#define PP_CONTROL _MMIO(0x61204)
-#define POWER_TARGET_ON (1 << 0)
-#define PP_ON_DELAYS _MMIO(0x61208)
-#define PP_OFF_DELAYS _MMIO(0x6120c)
-#define PP_DIVISOR _MMIO(0x61210)
+
+#define _PP_CONTROL 0x61204
+#define PP_CONTROL(pps_idx) _MMIO_PPS(pps_idx, _PP_CONTROL)
+#define PANEL_UNLOCK_REGS (0xabcd << 16)
+#define PANEL_UNLOCK_MASK (0xffff << 16)
+#define BXT_POWER_CYCLE_DELAY_MASK 0x1f0
+#define BXT_POWER_CYCLE_DELAY_SHIFT 4
+#define EDP_FORCE_VDD (1 << 3)
+#define EDP_BLC_ENABLE (1 << 2)
+#define PANEL_POWER_RESET (1 << 1)
+#define PANEL_POWER_OFF (0 << 0)
+#define PANEL_POWER_ON (1 << 0)
+
+#define _PP_ON_DELAYS 0x61208
+#define PP_ON_DELAYS(pps_idx) _MMIO_PPS(pps_idx, _PP_ON_DELAYS)
+#define PANEL_PORT_SELECT_SHIFT 30
+#define PANEL_PORT_SELECT_MASK (3 << 30)
+#define PANEL_PORT_SELECT_LVDS (0 << 30)
+#define PANEL_PORT_SELECT_DPA (1 << 30)
+#define PANEL_PORT_SELECT_DPC (2 << 30)
+#define PANEL_PORT_SELECT_DPD (3 << 30)
+#define PANEL_PORT_SELECT_VLV(port) ((port) << 30)
+#define PANEL_POWER_UP_DELAY_MASK 0x1fff0000
+#define PANEL_POWER_UP_DELAY_SHIFT 16
+#define PANEL_LIGHT_ON_DELAY_MASK 0x1fff
+#define PANEL_LIGHT_ON_DELAY_SHIFT 0
+
+#define _PP_OFF_DELAYS 0x6120C
+#define PP_OFF_DELAYS(pps_idx) _MMIO_PPS(pps_idx, _PP_OFF_DELAYS)
+#define PANEL_POWER_DOWN_DELAY_MASK 0x1fff0000
+#define PANEL_POWER_DOWN_DELAY_SHIFT 16
+#define PANEL_LIGHT_OFF_DELAY_MASK 0x1fff
+#define PANEL_LIGHT_OFF_DELAY_SHIFT 0
+
+#define _PP_DIVISOR 0x61210
+#define PP_DIVISOR(pps_idx) _MMIO_PPS(pps_idx, _PP_DIVISOR)
+#define PP_REFERENCE_DIVIDER_MASK 0xffffff00
+#define PP_REFERENCE_DIVIDER_SHIFT 8
+#define PANEL_POWER_CYCLE_DELAY_MASK 0x1f
+#define PANEL_POWER_CYCLE_DELAY_SHIFT 0
/* Panel fitting */
#define PFIT_CONTROL _MMIO(dev_priv->info.display_mmio_offset + 0x61230)
#define PCH_LVDS _MMIO(0xe1180)
#define LVDS_DETECTED (1 << 1)
-/* vlv has 2 sets of panel control regs. */
-#define _PIPEA_PP_STATUS (VLV_DISPLAY_BASE + 0x61200)
-#define _PIPEA_PP_CONTROL (VLV_DISPLAY_BASE + 0x61204)
-#define _PIPEA_PP_ON_DELAYS (VLV_DISPLAY_BASE + 0x61208)
-#define PANEL_PORT_SELECT_VLV(port) ((port) << 30)
-#define _PIPEA_PP_OFF_DELAYS (VLV_DISPLAY_BASE + 0x6120c)
-#define _PIPEA_PP_DIVISOR (VLV_DISPLAY_BASE + 0x61210)
-
-#define _PIPEB_PP_STATUS (VLV_DISPLAY_BASE + 0x61300)
-#define _PIPEB_PP_CONTROL (VLV_DISPLAY_BASE + 0x61304)
-#define _PIPEB_PP_ON_DELAYS (VLV_DISPLAY_BASE + 0x61308)
-#define _PIPEB_PP_OFF_DELAYS (VLV_DISPLAY_BASE + 0x6130c)
-#define _PIPEB_PP_DIVISOR (VLV_DISPLAY_BASE + 0x61310)
-
-#define VLV_PIPE_PP_STATUS(pipe) _MMIO_PIPE(pipe, _PIPEA_PP_STATUS, _PIPEB_PP_STATUS)
-#define VLV_PIPE_PP_CONTROL(pipe) _MMIO_PIPE(pipe, _PIPEA_PP_CONTROL, _PIPEB_PP_CONTROL)
-#define VLV_PIPE_PP_ON_DELAYS(pipe) _MMIO_PIPE(pipe, _PIPEA_PP_ON_DELAYS, _PIPEB_PP_ON_DELAYS)
-#define VLV_PIPE_PP_OFF_DELAYS(pipe) _MMIO_PIPE(pipe, _PIPEA_PP_OFF_DELAYS, _PIPEB_PP_OFF_DELAYS)
-#define VLV_PIPE_PP_DIVISOR(pipe) _MMIO_PIPE(pipe, _PIPEA_PP_DIVISOR, _PIPEB_PP_DIVISOR)
-
-#define _PCH_PP_STATUS 0xc7200
-#define _PCH_PP_CONTROL 0xc7204
-#define PANEL_UNLOCK_REGS (0xabcd << 16)
-#define PANEL_UNLOCK_MASK (0xffff << 16)
-#define BXT_POWER_CYCLE_DELAY_MASK (0x1f0)
-#define BXT_POWER_CYCLE_DELAY_SHIFT 4
-#define EDP_FORCE_VDD (1 << 3)
-#define EDP_BLC_ENABLE (1 << 2)
-#define PANEL_POWER_RESET (1 << 1)
-#define PANEL_POWER_OFF (0 << 0)
-#define PANEL_POWER_ON (1 << 0)
-#define _PCH_PP_ON_DELAYS 0xc7208
-#define PANEL_PORT_SELECT_MASK (3 << 30)
-#define PANEL_PORT_SELECT_LVDS (0 << 30)
-#define PANEL_PORT_SELECT_DPA (1 << 30)
-#define PANEL_PORT_SELECT_DPC (2 << 30)
-#define PANEL_PORT_SELECT_DPD (3 << 30)
-#define PANEL_POWER_UP_DELAY_MASK (0x1fff0000)
-#define PANEL_POWER_UP_DELAY_SHIFT 16
-#define PANEL_LIGHT_ON_DELAY_MASK (0x1fff)
-#define PANEL_LIGHT_ON_DELAY_SHIFT 0
-
-#define _PCH_PP_OFF_DELAYS 0xc720c
-#define PANEL_POWER_DOWN_DELAY_MASK (0x1fff0000)
-#define PANEL_POWER_DOWN_DELAY_SHIFT 16
-#define PANEL_LIGHT_OFF_DELAY_MASK (0x1fff)
-#define PANEL_LIGHT_OFF_DELAY_SHIFT 0
-
-#define _PCH_PP_DIVISOR 0xc7210
-#define PP_REFERENCE_DIVIDER_MASK (0xffffff00)
-#define PP_REFERENCE_DIVIDER_SHIFT 8
-#define PANEL_POWER_CYCLE_DELAY_MASK (0x1f)
-#define PANEL_POWER_CYCLE_DELAY_SHIFT 0
-
-#define PCH_PP_STATUS _MMIO(_PCH_PP_STATUS)
-#define PCH_PP_CONTROL _MMIO(_PCH_PP_CONTROL)
-#define PCH_PP_ON_DELAYS _MMIO(_PCH_PP_ON_DELAYS)
-#define PCH_PP_OFF_DELAYS _MMIO(_PCH_PP_OFF_DELAYS)
-#define PCH_PP_DIVISOR _MMIO(_PCH_PP_DIVISOR)
-
-/* BXT PPS changes - 2nd set of PPS registers */
-#define _BXT_PP_STATUS2 0xc7300
-#define _BXT_PP_CONTROL2 0xc7304
-#define _BXT_PP_ON_DELAYS2 0xc7308
-#define _BXT_PP_OFF_DELAYS2 0xc730c
-
-#define BXT_PP_STATUS(n) _MMIO_PIPE(n, _PCH_PP_STATUS, _BXT_PP_STATUS2)
-#define BXT_PP_CONTROL(n) _MMIO_PIPE(n, _PCH_PP_CONTROL, _BXT_PP_CONTROL2)
-#define BXT_PP_ON_DELAYS(n) _MMIO_PIPE(n, _PCH_PP_ON_DELAYS, _BXT_PP_ON_DELAYS2)
-#define BXT_PP_OFF_DELAYS(n) _MMIO_PIPE(n, _PCH_PP_OFF_DELAYS, _BXT_PP_OFF_DELAYS2)
-
#define _PCH_DP_B 0xe4100
#define PCH_DP_B _MMIO(_PCH_DP_B)
#define _PCH_DPB_AUX_CH_CTL 0xe4110
if (INTEL_INFO(dev)->gen <= 4)
dev_priv->regfile.saveDSPARB = I915_READ(DSPARB);
- /* LVDS state */
- if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
- dev_priv->regfile.saveLVDS = I915_READ(PCH_LVDS);
- else if (INTEL_INFO(dev)->gen <= 4 && IS_MOBILE(dev) && !IS_I830(dev))
- dev_priv->regfile.saveLVDS = I915_READ(LVDS);
-
- /* Panel power sequencer */
- if (HAS_PCH_SPLIT(dev)) {
- dev_priv->regfile.savePP_CONTROL = I915_READ(PCH_PP_CONTROL);
- dev_priv->regfile.savePP_ON_DELAYS = I915_READ(PCH_PP_ON_DELAYS);
- dev_priv->regfile.savePP_OFF_DELAYS = I915_READ(PCH_PP_OFF_DELAYS);
- dev_priv->regfile.savePP_DIVISOR = I915_READ(PCH_PP_DIVISOR);
- } else if (INTEL_INFO(dev)->gen <= 4) {
- dev_priv->regfile.savePP_CONTROL = I915_READ(PP_CONTROL);
- dev_priv->regfile.savePP_ON_DELAYS = I915_READ(PP_ON_DELAYS);
- dev_priv->regfile.savePP_OFF_DELAYS = I915_READ(PP_OFF_DELAYS);
- dev_priv->regfile.savePP_DIVISOR = I915_READ(PP_DIVISOR);
- }
-
/* save FBC interval */
if (HAS_FBC(dev) && INTEL_INFO(dev)->gen <= 4 && !IS_G4X(dev))
dev_priv->regfile.saveFBC_CONTROL = I915_READ(FBC_CONTROL);
static void i915_restore_display(struct drm_device *dev)
{
struct drm_i915_private *dev_priv = to_i915(dev);
- u32 mask = 0xffffffff;
/* Display arbitration */
if (INTEL_INFO(dev)->gen <= 4)
I915_WRITE(DSPARB, dev_priv->regfile.saveDSPARB);
- mask = ~LVDS_PORT_EN;
-
- /* LVDS state */
- if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
- I915_WRITE(PCH_LVDS, dev_priv->regfile.saveLVDS & mask);
- else if (INTEL_INFO(dev)->gen <= 4 && IS_MOBILE(dev) && !IS_I830(dev))
- I915_WRITE(LVDS, dev_priv->regfile.saveLVDS & mask);
-
- /* Panel power sequencer */
- if (HAS_PCH_SPLIT(dev)) {
- I915_WRITE(PCH_PP_ON_DELAYS, dev_priv->regfile.savePP_ON_DELAYS);
- I915_WRITE(PCH_PP_OFF_DELAYS, dev_priv->regfile.savePP_OFF_DELAYS);
- I915_WRITE(PCH_PP_DIVISOR, dev_priv->regfile.savePP_DIVISOR);
- I915_WRITE(PCH_PP_CONTROL, dev_priv->regfile.savePP_CONTROL);
- } else if (INTEL_INFO(dev)->gen <= 4) {
- I915_WRITE(PP_ON_DELAYS, dev_priv->regfile.savePP_ON_DELAYS);
- I915_WRITE(PP_OFF_DELAYS, dev_priv->regfile.savePP_OFF_DELAYS);
- I915_WRITE(PP_DIVISOR, dev_priv->regfile.savePP_DIVISOR);
- I915_WRITE(PP_CONTROL, dev_priv->regfile.savePP_CONTROL);
- }
-
/* only restore FBC info on the platform that supports FBC*/
intel_fbc_global_disable(dev_priv);
#include "i915_drv.h"
+static void intel_breadcrumbs_hangcheck(unsigned long data)
+{
+ struct intel_engine_cs *engine = (struct intel_engine_cs *)data;
+ struct intel_breadcrumbs *b = &engine->breadcrumbs;
+
+ if (!b->irq_enabled)
+ return;
+
+ if (time_before(jiffies, b->timeout)) {
+ mod_timer(&b->hangcheck, b->timeout);
+ return;
+ }
+
+ DRM_DEBUG("Hangcheck timer elapsed... %s idle\n", engine->name);
+ set_bit(engine->id, &engine->i915->gpu_error.missed_irq_rings);
+ mod_timer(&engine->breadcrumbs.fake_irq, jiffies + 1);
+
+ /* Ensure that even if the GPU hangs, we get woken up.
+ *
+ * However, note that if no one is waiting, we never notice
+ * a gpu hang. Eventually, we will have to wait for a resource
+ * held by the GPU and so trigger a hangcheck. In the most
+ * pathological case, this will be upon memory starvation! To
+ * prevent this, we also queue the hangcheck from the retire
+ * worker.
+ */
+ i915_queue_hangcheck(engine->i915);
+}
+
+static unsigned long wait_timeout(void)
+{
+ return round_jiffies_up(jiffies + DRM_I915_HANGCHECK_JIFFIES);
+}
+
static void intel_breadcrumbs_fake_irq(unsigned long data)
{
struct intel_engine_cs *engine = (struct intel_engine_cs *)data;
* every jiffie in order to kick the oldest waiter to do the
* coherent seqno check.
*/
- rcu_read_lock();
if (intel_engine_wakeup(engine))
mod_timer(&engine->breadcrumbs.fake_irq, jiffies + 1);
- rcu_read_unlock();
}
static void irq_enable(struct intel_engine_cs *engine)
*/
engine->breadcrumbs.irq_posted = true;
- /* Make sure the current hangcheck doesn't falsely accuse a just
- * started irq handler from missing an interrupt (because the
- * interrupt count still matches the stale value from when
- * the irq handler was disabled, many hangchecks ago).
- */
- engine->breadcrumbs.irq_wakeups++;
-
spin_lock_irq(&engine->i915->irq_lock);
engine->irq_enable(engine);
spin_unlock_irq(&engine->i915->irq_lock);
}
if (!b->irq_enabled ||
- test_bit(engine->id, &i915->gpu_error.missed_irq_rings))
+ test_bit(engine->id, &i915->gpu_error.missed_irq_rings)) {
mod_timer(&b->fake_irq, jiffies + 1);
-
- /* Ensure that even if the GPU hangs, we get woken up.
- *
- * However, note that if no one is waiting, we never notice
- * a gpu hang. Eventually, we will have to wait for a resource
- * held by the GPU and so trigger a hangcheck. In the most
- * pathological case, this will be upon memory starvation!
- */
- i915_queue_hangcheck(i915);
+ } else {
+ /* Ensure we never sleep indefinitely */
+ GEM_BUG_ON(!time_after(b->timeout, jiffies));
+ mod_timer(&b->hangcheck, b->timeout);
+ }
}
static void __intel_breadcrumbs_disable_irq(struct intel_breadcrumbs *b)
}
rb_link_node(&wait->node, parent, p);
rb_insert_color(&wait->node, &b->waiters);
- GEM_BUG_ON(!first && !b->irq_seqno_bh);
+ GEM_BUG_ON(!first && !rcu_access_pointer(b->irq_seqno_bh));
if (completed) {
struct rb_node *next = rb_next(completed);
GEM_BUG_ON(!next && !first);
if (next && next != &wait->node) {
GEM_BUG_ON(first);
+ b->timeout = wait_timeout();
b->first_wait = to_wait(next);
- smp_store_mb(b->irq_seqno_bh, b->first_wait->tsk);
+ rcu_assign_pointer(b->irq_seqno_bh, b->first_wait->tsk);
/* As there is a delay between reading the current
* seqno, processing the completed tasks and selecting
* the next waiter, we may have missed the interrupt
if (first) {
GEM_BUG_ON(rb_first(&b->waiters) != &wait->node);
+ b->timeout = wait_timeout();
b->first_wait = wait;
- smp_store_mb(b->irq_seqno_bh, wait->tsk);
+ rcu_assign_pointer(b->irq_seqno_bh, wait->tsk);
/* After assigning ourselves as the new bottom-half, we must
* perform a cursory check to prevent a missed interrupt.
* Either we miss the interrupt whilst programming the hardware,
*/
__intel_breadcrumbs_enable_irq(b);
}
- GEM_BUG_ON(!b->irq_seqno_bh);
+ GEM_BUG_ON(!rcu_access_pointer(b->irq_seqno_bh));
GEM_BUG_ON(!b->first_wait);
GEM_BUG_ON(rb_first(&b->waiters) != &b->first_wait->node);
return first;
}
-void intel_engine_enable_fake_irq(struct intel_engine_cs *engine)
-{
- mod_timer(&engine->breadcrumbs.fake_irq, jiffies + 1);
-}
-
static inline bool chain_wakeup(struct rb_node *rb, int priority)
{
return rb && to_wait(rb)->tsk->prio <= priority;
const int priority = wakeup_priority(b, wait->tsk);
struct rb_node *next;
- GEM_BUG_ON(b->irq_seqno_bh != wait->tsk);
+ GEM_BUG_ON(rcu_access_pointer(b->irq_seqno_bh) != wait->tsk);
/* We are the current bottom-half. Find the next candidate,
* the first waiter in the queue on the remaining oldest
* the interrupt, or if we have to handle an
* exception rather than a seqno completion.
*/
+ b->timeout = wait_timeout();
b->first_wait = to_wait(next);
- smp_store_mb(b->irq_seqno_bh, b->first_wait->tsk);
+ rcu_assign_pointer(b->irq_seqno_bh, b->first_wait->tsk);
if (b->first_wait->seqno != wait->seqno)
__intel_breadcrumbs_enable_irq(b);
- wake_up_process(b->irq_seqno_bh);
+ wake_up_process(b->first_wait->tsk);
} else {
b->first_wait = NULL;
- WRITE_ONCE(b->irq_seqno_bh, NULL);
+ rcu_assign_pointer(b->irq_seqno_bh, NULL);
__intel_breadcrumbs_disable_irq(b);
}
} else {
GEM_BUG_ON(b->first_wait == wait);
GEM_BUG_ON(rb_first(&b->waiters) !=
(b->first_wait ? &b->first_wait->node : NULL));
- GEM_BUG_ON(!b->irq_seqno_bh ^ RB_EMPTY_ROOT(&b->waiters));
+ GEM_BUG_ON(!rcu_access_pointer(b->irq_seqno_bh) ^ RB_EMPTY_ROOT(&b->waiters));
spin_unlock(&b->lock);
}
setup_timer(&b->fake_irq,
intel_breadcrumbs_fake_irq,
(unsigned long)engine);
+ setup_timer(&b->hangcheck,
+ intel_breadcrumbs_hangcheck,
+ (unsigned long)engine);
/* Spawn a thread to provide a common bottom-half for all signals.
* As this is an asynchronous interface we cannot steal the current
if (!IS_ERR_OR_NULL(b->signaler))
kthread_stop(b->signaler);
+ del_timer_sync(&b->hangcheck);
del_timer_sync(&b->fake_irq);
}
* RCU lock, i.e. as we call wake_up_process() we must be holding the
* rcu_read_lock().
*/
- rcu_read_lock();
for_each_engine(engine, i915)
if (unlikely(intel_engine_wakeup(engine)))
mask |= intel_engine_flag(engine);
- rcu_read_unlock();
return mask;
}
if (HAS_PCH_SPLIT(dev)) {
u32 port_sel;
- pp_reg = PCH_PP_CONTROL;
- port_sel = I915_READ(PCH_PP_ON_DELAYS) & PANEL_PORT_SELECT_MASK;
+ pp_reg = PP_CONTROL(0);
+ port_sel = I915_READ(PP_ON_DELAYS(0)) & PANEL_PORT_SELECT_MASK;
if (port_sel == PANEL_PORT_SELECT_LVDS &&
I915_READ(PCH_LVDS) & LVDS_PIPEB_SELECT)
/* XXX: else fix for eDP */
} else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
/* presumably write lock depends on pipe, not port select */
- pp_reg = VLV_PIPE_PP_CONTROL(pipe);
+ pp_reg = PP_CONTROL(pipe);
panel_pipe = pipe;
} else {
- pp_reg = PP_CONTROL;
+ pp_reg = PP_CONTROL(0);
if (I915_READ(LVDS) & LVDS_PIPEB_SELECT)
panel_pipe = PIPE_B;
}
* a plane. On ILK+ the pipe PLLs are integrated, so we don't
* need the check.
*/
- if (HAS_GMCH_DISPLAY(dev_priv))
+ if (HAS_GMCH_DISPLAY(dev_priv)) {
if (intel_crtc_has_type(crtc->config, INTEL_OUTPUT_DSI))
assert_dsi_pll_enabled(dev_priv);
else
assert_pll_enabled(dev_priv, pipe);
- else {
+ } else {
if (crtc->config->has_pch_encoder) {
/* if driving the PCH, we need FDI enabled */
assert_fdi_rx_pll_enabled(dev_priv, pch_transcoder);
}
}
-static void
-intel_fill_fb_info(struct drm_i915_private *dev_priv,
- struct drm_framebuffer *fb)
-{
- struct intel_rotation_info *info = &to_intel_framebuffer(fb)->rot_info;
- unsigned int tile_size, tile_width, tile_height, cpp;
-
- tile_size = intel_tile_size(dev_priv);
-
- cpp = drm_format_plane_cpp(fb->pixel_format, 0);
- intel_tile_dims(dev_priv, &tile_width, &tile_height,
- fb->modifier[0], cpp);
-
- info->plane[0].width = DIV_ROUND_UP(fb->pitches[0], tile_width * cpp);
- info->plane[0].height = DIV_ROUND_UP(fb->height, tile_height);
-
- if (info->pixel_format == DRM_FORMAT_NV12) {
- cpp = drm_format_plane_cpp(fb->pixel_format, 1);
- intel_tile_dims(dev_priv, &tile_width, &tile_height,
- fb->modifier[1], cpp);
-
- info->uv_offset = fb->offsets[1];
- info->plane[1].width = DIV_ROUND_UP(fb->pitches[1], tile_width * cpp);
- info->plane[1].height = DIV_ROUND_UP(fb->height / 2, tile_height);
- }
-}
-
static unsigned int intel_linear_alignment(const struct drm_i915_private *dev_priv)
{
if (INTEL_INFO(dev_priv)->gen >= 9)
i915_gem_object_unpin_from_display_plane(obj, &view);
}
+static int intel_fb_pitch(const struct drm_framebuffer *fb, int plane,
+ unsigned int rotation)
+{
+ if (intel_rotation_90_or_270(rotation))
+ return to_intel_framebuffer(fb)->rotated[plane].pitch;
+ else
+ return fb->pitches[plane];
+}
+
+/*
+ * Convert the x/y offsets into a linear offset.
+ * Only valid with 0/180 degree rotation, which is fine since linear
+ * offset is only used with linear buffers on pre-hsw and tiled buffers
+ * with gen2/3, and 90/270 degree rotations isn't supported on any of them.
+ */
+u32 intel_fb_xy_to_linear(int x, int y,
+ const struct intel_plane_state *state,
+ int plane)
+{
+ const struct drm_framebuffer *fb = state->base.fb;
+ unsigned int cpp = drm_format_plane_cpp(fb->pixel_format, plane);
+ unsigned int pitch = fb->pitches[plane];
+
+ return y * pitch + x * cpp;
+}
+
+/*
+ * Add the x/y offsets derived from fb->offsets[] to the user
+ * specified plane src x/y offsets. The resulting x/y offsets
+ * specify the start of scanout from the beginning of the gtt mapping.
+ */
+void intel_add_fb_offsets(int *x, int *y,
+ const struct intel_plane_state *state,
+ int plane)
+
+{
+ const struct intel_framebuffer *intel_fb = to_intel_framebuffer(state->base.fb);
+ unsigned int rotation = state->base.rotation;
+
+ if (intel_rotation_90_or_270(rotation)) {
+ *x += intel_fb->rotated[plane].x;
+ *y += intel_fb->rotated[plane].y;
+ } else {
+ *x += intel_fb->normal[plane].x;
+ *y += intel_fb->normal[plane].y;
+ }
+}
+
/*
- * Adjust the tile offset by moving the difference into
- * the x/y offsets.
- *
* Input tile dimensions and pitch must already be
* rotated to match x and y, and in pixel units.
*/
-static u32 intel_adjust_tile_offset(int *x, int *y,
- unsigned int tile_width,
- unsigned int tile_height,
- unsigned int tile_size,
- unsigned int pitch_tiles,
- u32 old_offset,
- u32 new_offset)
-{
+static u32 _intel_adjust_tile_offset(int *x, int *y,
+ unsigned int tile_width,
+ unsigned int tile_height,
+ unsigned int tile_size,
+ unsigned int pitch_tiles,
+ u32 old_offset,
+ u32 new_offset)
+{
+ unsigned int pitch_pixels = pitch_tiles * tile_width;
unsigned int tiles;
WARN_ON(old_offset & (tile_size - 1));
*y += tiles / pitch_tiles * tile_height;
*x += tiles % pitch_tiles * tile_width;
+ /* minimize x in case it got needlessly big */
+ *y += *x / pitch_pixels * tile_height;
+ *x %= pitch_pixels;
+
+ return new_offset;
+}
+
+/*
+ * Adjust the tile offset by moving the difference into
+ * the x/y offsets.
+ */
+static u32 intel_adjust_tile_offset(int *x, int *y,
+ const struct intel_plane_state *state, int plane,
+ u32 old_offset, u32 new_offset)
+{
+ const struct drm_i915_private *dev_priv = to_i915(state->base.plane->dev);
+ const struct drm_framebuffer *fb = state->base.fb;
+ unsigned int cpp = drm_format_plane_cpp(fb->pixel_format, plane);
+ unsigned int rotation = state->base.rotation;
+ unsigned int pitch = intel_fb_pitch(fb, plane, rotation);
+
+ WARN_ON(new_offset > old_offset);
+
+ if (fb->modifier[plane] != DRM_FORMAT_MOD_NONE) {
+ unsigned int tile_size, tile_width, tile_height;
+ unsigned int pitch_tiles;
+
+ tile_size = intel_tile_size(dev_priv);
+ intel_tile_dims(dev_priv, &tile_width, &tile_height,
+ fb->modifier[plane], cpp);
+
+ if (intel_rotation_90_or_270(rotation)) {
+ pitch_tiles = pitch / tile_height;
+ swap(tile_width, tile_height);
+ } else {
+ pitch_tiles = pitch / (tile_width * cpp);
+ }
+
+ _intel_adjust_tile_offset(x, y, tile_width, tile_height,
+ tile_size, pitch_tiles,
+ old_offset, new_offset);
+ } else {
+ old_offset += *y * pitch + *x * cpp;
+
+ *y = (old_offset - new_offset) / pitch;
+ *x = ((old_offset - new_offset) - *y * pitch) / cpp;
+ }
+
return new_offset;
}
* In the 90/270 rotated case, x and y are assumed
* to be already rotated to match the rotated GTT view, and
* pitch is the tile_height aligned framebuffer height.
+ *
+ * This function is used when computing the derived information
+ * under intel_framebuffer, so using any of that information
+ * here is not allowed. Anything under drm_framebuffer can be
+ * used. This is why the user has to pass in the pitch since it
+ * is specified in the rotated orientation.
*/
-u32 intel_compute_tile_offset(int *x, int *y,
- const struct drm_framebuffer *fb, int plane,
- unsigned int pitch,
- unsigned int rotation)
+static u32 _intel_compute_tile_offset(const struct drm_i915_private *dev_priv,
+ int *x, int *y,
+ const struct drm_framebuffer *fb, int plane,
+ unsigned int pitch,
+ unsigned int rotation,
+ u32 alignment)
{
- const struct drm_i915_private *dev_priv = to_i915(fb->dev);
uint64_t fb_modifier = fb->modifier[plane];
unsigned int cpp = drm_format_plane_cpp(fb->pixel_format, plane);
- u32 offset, offset_aligned, alignment;
+ u32 offset, offset_aligned;
- alignment = intel_surf_alignment(dev_priv, fb_modifier);
if (alignment)
alignment--;
offset = (tile_rows * pitch_tiles + tiles) * tile_size;
offset_aligned = offset & ~alignment;
- intel_adjust_tile_offset(x, y, tile_width, tile_height,
- tile_size, pitch_tiles,
- offset, offset_aligned);
+ _intel_adjust_tile_offset(x, y, tile_width, tile_height,
+ tile_size, pitch_tiles,
+ offset, offset_aligned);
} else {
offset = *y * pitch + *x * cpp;
offset_aligned = offset & ~alignment;
return offset_aligned;
}
+u32 intel_compute_tile_offset(int *x, int *y,
+ const struct intel_plane_state *state,
+ int plane)
+{
+ const struct drm_i915_private *dev_priv = to_i915(state->base.plane->dev);
+ const struct drm_framebuffer *fb = state->base.fb;
+ unsigned int rotation = state->base.rotation;
+ int pitch = intel_fb_pitch(fb, plane, rotation);
+ u32 alignment;
+
+ /* AUX_DIST needs only 4K alignment */
+ if (fb->pixel_format == DRM_FORMAT_NV12 && plane == 1)
+ alignment = 4096;
+ else
+ alignment = intel_surf_alignment(dev_priv, fb->modifier[plane]);
+
+ return _intel_compute_tile_offset(dev_priv, x, y, fb, plane, pitch,
+ rotation, alignment);
+}
+
+/* Convert the fb->offset[] linear offset into x/y offsets */
+static void intel_fb_offset_to_xy(int *x, int *y,
+ const struct drm_framebuffer *fb, int plane)
+{
+ unsigned int cpp = drm_format_plane_cpp(fb->pixel_format, plane);
+ unsigned int pitch = fb->pitches[plane];
+ u32 linear_offset = fb->offsets[plane];
+
+ *y = linear_offset / pitch;
+ *x = linear_offset % pitch / cpp;
+}
+
+static unsigned int intel_fb_modifier_to_tiling(uint64_t fb_modifier)
+{
+ switch (fb_modifier) {
+ case I915_FORMAT_MOD_X_TILED:
+ return I915_TILING_X;
+ case I915_FORMAT_MOD_Y_TILED:
+ return I915_TILING_Y;
+ default:
+ return I915_TILING_NONE;
+ }
+}
+
+static int
+intel_fill_fb_info(struct drm_i915_private *dev_priv,
+ struct drm_framebuffer *fb)
+{
+ struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
+ struct intel_rotation_info *rot_info = &intel_fb->rot_info;
+ u32 gtt_offset_rotated = 0;
+ unsigned int max_size = 0;
+ uint32_t format = fb->pixel_format;
+ int i, num_planes = drm_format_num_planes(format);
+ unsigned int tile_size = intel_tile_size(dev_priv);
+
+ for (i = 0; i < num_planes; i++) {
+ unsigned int width, height;
+ unsigned int cpp, size;
+ u32 offset;
+ int x, y;
+
+ cpp = drm_format_plane_cpp(format, i);
+ width = drm_format_plane_width(fb->width, format, i);
+ height = drm_format_plane_height(fb->height, format, i);
+
+ intel_fb_offset_to_xy(&x, &y, fb, i);
+
+ /*
+ * The fence (if used) is aligned to the start of the object
+ * so having the framebuffer wrap around across the edge of the
+ * fenced region doesn't really work. We have no API to configure
+ * the fence start offset within the object (nor could we probably
+ * on gen2/3). So it's just easier if we just require that the
+ * fb layout agrees with the fence layout. We already check that the
+ * fb stride matches the fence stride elsewhere.
+ */
+ if (i915_gem_object_is_tiled(intel_fb->obj) &&
+ (x + width) * cpp > fb->pitches[i]) {
+ DRM_DEBUG("bad fb plane %d offset: 0x%x\n",
+ i, fb->offsets[i]);
+ return -EINVAL;
+ }
+
+ /*
+ * First pixel of the framebuffer from
+ * the start of the normal gtt mapping.
+ */
+ intel_fb->normal[i].x = x;
+ intel_fb->normal[i].y = y;
+
+ offset = _intel_compute_tile_offset(dev_priv, &x, &y,
+ fb, 0, fb->pitches[i],
+ DRM_ROTATE_0, tile_size);
+ offset /= tile_size;
+
+ if (fb->modifier[i] != DRM_FORMAT_MOD_NONE) {
+ unsigned int tile_width, tile_height;
+ unsigned int pitch_tiles;
+ struct drm_rect r;
+
+ intel_tile_dims(dev_priv, &tile_width, &tile_height,
+ fb->modifier[i], cpp);
+
+ rot_info->plane[i].offset = offset;
+ rot_info->plane[i].stride = DIV_ROUND_UP(fb->pitches[i], tile_width * cpp);
+ rot_info->plane[i].width = DIV_ROUND_UP(x + width, tile_width);
+ rot_info->plane[i].height = DIV_ROUND_UP(y + height, tile_height);
+
+ intel_fb->rotated[i].pitch =
+ rot_info->plane[i].height * tile_height;
+
+ /* how many tiles does this plane need */
+ size = rot_info->plane[i].stride * rot_info->plane[i].height;
+ /*
+ * If the plane isn't horizontally tile aligned,
+ * we need one more tile.
+ */
+ if (x != 0)
+ size++;
+
+ /* rotate the x/y offsets to match the GTT view */
+ r.x1 = x;
+ r.y1 = y;
+ r.x2 = x + width;
+ r.y2 = y + height;
+ drm_rect_rotate(&r,
+ rot_info->plane[i].width * tile_width,
+ rot_info->plane[i].height * tile_height,
+ DRM_ROTATE_270);
+ x = r.x1;
+ y = r.y1;
+
+ /* rotate the tile dimensions to match the GTT view */
+ pitch_tiles = intel_fb->rotated[i].pitch / tile_height;
+ swap(tile_width, tile_height);
+
+ /*
+ * We only keep the x/y offsets, so push all of the
+ * gtt offset into the x/y offsets.
+ */
+ _intel_adjust_tile_offset(&x, &y, tile_size,
+ tile_width, tile_height, pitch_tiles,
+ gtt_offset_rotated * tile_size, 0);
+
+ gtt_offset_rotated += rot_info->plane[i].width * rot_info->plane[i].height;
+
+ /*
+ * First pixel of the framebuffer from
+ * the start of the rotated gtt mapping.
+ */
+ intel_fb->rotated[i].x = x;
+ intel_fb->rotated[i].y = y;
+ } else {
+ size = DIV_ROUND_UP((y + height) * fb->pitches[i] +
+ x * cpp, tile_size);
+ }
+
+ /* how many tiles in total needed in the bo */
+ max_size = max(max_size, offset + size);
+ }
+
+ if (max_size * tile_size > to_intel_framebuffer(fb)->obj->base.size) {
+ DRM_DEBUG("fb too big for bo (need %u bytes, have %zu bytes)\n",
+ max_size * tile_size, to_intel_framebuffer(fb)->obj->base.size);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
static int i9xx_format_to_fourcc(int format)
{
switch (format) {
&obj->frontbuffer_bits);
}
+static int skl_max_plane_width(const struct drm_framebuffer *fb, int plane,
+ unsigned int rotation)
+{
+ int cpp = drm_format_plane_cpp(fb->pixel_format, plane);
+
+ switch (fb->modifier[plane]) {
+ case DRM_FORMAT_MOD_NONE:
+ case I915_FORMAT_MOD_X_TILED:
+ switch (cpp) {
+ case 8:
+ return 4096;
+ case 4:
+ case 2:
+ case 1:
+ return 8192;
+ default:
+ MISSING_CASE(cpp);
+ break;
+ }
+ break;
+ case I915_FORMAT_MOD_Y_TILED:
+ case I915_FORMAT_MOD_Yf_TILED:
+ switch (cpp) {
+ case 8:
+ return 2048;
+ case 4:
+ return 4096;
+ case 2:
+ case 1:
+ return 8192;
+ default:
+ MISSING_CASE(cpp);
+ break;
+ }
+ break;
+ default:
+ MISSING_CASE(fb->modifier[plane]);
+ }
+
+ return 2048;
+}
+
+static int skl_check_main_surface(struct intel_plane_state *plane_state)
+{
+ const struct drm_i915_private *dev_priv = to_i915(plane_state->base.plane->dev);
+ const struct drm_framebuffer *fb = plane_state->base.fb;
+ unsigned int rotation = plane_state->base.rotation;
+ int x = plane_state->base.src.x1 >> 16;
+ int y = plane_state->base.src.y1 >> 16;
+ int w = drm_rect_width(&plane_state->base.src) >> 16;
+ int h = drm_rect_height(&plane_state->base.src) >> 16;
+ int max_width = skl_max_plane_width(fb, 0, rotation);
+ int max_height = 4096;
+ u32 alignment, offset, aux_offset = plane_state->aux.offset;
+
+ if (w > max_width || h > max_height) {
+ DRM_DEBUG_KMS("requested Y/RGB source size %dx%d too big (limit %dx%d)\n",
+ w, h, max_width, max_height);
+ return -EINVAL;
+ }
+
+ intel_add_fb_offsets(&x, &y, plane_state, 0);
+ offset = intel_compute_tile_offset(&x, &y, plane_state, 0);
+
+ alignment = intel_surf_alignment(dev_priv, fb->modifier[0]);
+
+ /*
+ * AUX surface offset is specified as the distance from the
+ * main surface offset, and it must be non-negative. Make
+ * sure that is what we will get.
+ */
+ if (offset > aux_offset)
+ offset = intel_adjust_tile_offset(&x, &y, plane_state, 0,
+ offset, aux_offset & ~(alignment - 1));
+
+ /*
+ * When using an X-tiled surface, the plane blows up
+ * if the x offset + width exceed the stride.
+ *
+ * TODO: linear and Y-tiled seem fine, Yf untested,
+ */
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED) {
+ int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
+
+ while ((x + w) * cpp > fb->pitches[0]) {
+ if (offset == 0) {
+ DRM_DEBUG_KMS("Unable to find suitable display surface offset\n");
+ return -EINVAL;
+ }
+
+ offset = intel_adjust_tile_offset(&x, &y, plane_state, 0,
+ offset, offset - alignment);
+ }
+ }
+
+ plane_state->main.offset = offset;
+ plane_state->main.x = x;
+ plane_state->main.y = y;
+
+ return 0;
+}
+
+static int skl_check_nv12_aux_surface(struct intel_plane_state *plane_state)
+{
+ const struct drm_framebuffer *fb = plane_state->base.fb;
+ unsigned int rotation = plane_state->base.rotation;
+ int max_width = skl_max_plane_width(fb, 1, rotation);
+ int max_height = 4096;
+ int x = plane_state->base.src.x1 >> 17;
+ int y = plane_state->base.src.y1 >> 17;
+ int w = drm_rect_width(&plane_state->base.src) >> 17;
+ int h = drm_rect_height(&plane_state->base.src) >> 17;
+ u32 offset;
+
+ intel_add_fb_offsets(&x, &y, plane_state, 1);
+ offset = intel_compute_tile_offset(&x, &y, plane_state, 1);
+
+ /* FIXME not quite sure how/if these apply to the chroma plane */
+ if (w > max_width || h > max_height) {
+ DRM_DEBUG_KMS("CbCr source size %dx%d too big (limit %dx%d)\n",
+ w, h, max_width, max_height);
+ return -EINVAL;
+ }
+
+ plane_state->aux.offset = offset;
+ plane_state->aux.x = x;
+ plane_state->aux.y = y;
+
+ return 0;
+}
+
+int skl_check_plane_surface(struct intel_plane_state *plane_state)
+{
+ const struct drm_framebuffer *fb = plane_state->base.fb;
+ unsigned int rotation = plane_state->base.rotation;
+ int ret;
+
+ /* Rotate src coordinates to match rotated GTT view */
+ if (intel_rotation_90_or_270(rotation))
+ drm_rect_rotate(&plane_state->base.src,
+ fb->width, fb->height, DRM_ROTATE_270);
+
+ /*
+ * Handle the AUX surface first since
+ * the main surface setup depends on it.
+ */
+ if (fb->pixel_format == DRM_FORMAT_NV12) {
+ ret = skl_check_nv12_aux_surface(plane_state);
+ if (ret)
+ return ret;
+ } else {
+ plane_state->aux.offset = ~0xfff;
+ plane_state->aux.x = 0;
+ plane_state->aux.y = 0;
+ }
+
+ ret = skl_check_main_surface(plane_state);
+ if (ret)
+ return ret;
+
+ return 0;
+}
+
static void i9xx_update_primary_plane(struct drm_plane *primary,
const struct intel_crtc_state *crtc_state,
const struct intel_plane_state *plane_state)
u32 dspcntr;
i915_reg_t reg = DSPCNTR(plane);
unsigned int rotation = plane_state->base.rotation;
- int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
int x = plane_state->base.src.x1 >> 16;
int y = plane_state->base.src.y1 >> 16;
BUG();
}
- if (INTEL_INFO(dev)->gen >= 4 && i915_gem_object_is_tiled(obj))
+ if (INTEL_GEN(dev_priv) >= 4 &&
+ fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
dspcntr |= DISPPLANE_TILED;
if (IS_G4X(dev))
dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
- linear_offset = y * fb->pitches[0] + x * cpp;
+ intel_add_fb_offsets(&x, &y, plane_state, 0);
- if (INTEL_INFO(dev)->gen >= 4) {
+ if (INTEL_INFO(dev)->gen >= 4)
intel_crtc->dspaddr_offset =
- intel_compute_tile_offset(&x, &y, fb, 0,
- fb->pitches[0], rotation);
- linear_offset -= intel_crtc->dspaddr_offset;
- } else {
- intel_crtc->dspaddr_offset = linear_offset;
- }
+ intel_compute_tile_offset(&x, &y, plane_state, 0);
if (rotation == DRM_ROTATE_180) {
dspcntr |= DISPPLANE_ROTATE_180;
x += (crtc_state->pipe_src_w - 1);
y += (crtc_state->pipe_src_h - 1);
-
- /* Finding the last pixel of the last line of the display
- data and adding to linear_offset*/
- linear_offset +=
- (crtc_state->pipe_src_h - 1) * fb->pitches[0] +
- (crtc_state->pipe_src_w - 1) * cpp;
}
+ linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
+
+ if (INTEL_INFO(dev)->gen < 4)
+ intel_crtc->dspaddr_offset = linear_offset;
+
intel_crtc->adjusted_x = x;
intel_crtc->adjusted_y = y;
I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
if (INTEL_INFO(dev)->gen >= 4) {
I915_WRITE(DSPSURF(plane),
- i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+ intel_fb_gtt_offset(fb, rotation) +
+ intel_crtc->dspaddr_offset);
I915_WRITE(DSPTILEOFF(plane), (y << 16) | x);
I915_WRITE(DSPLINOFF(plane), linear_offset);
} else
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_crtc *intel_crtc = to_intel_crtc(crtc_state->base.crtc);
struct drm_framebuffer *fb = plane_state->base.fb;
- struct drm_i915_gem_object *obj = intel_fb_obj(fb);
int plane = intel_crtc->plane;
u32 linear_offset;
u32 dspcntr;
i915_reg_t reg = DSPCNTR(plane);
unsigned int rotation = plane_state->base.rotation;
- int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
int x = plane_state->base.src.x1 >> 16;
int y = plane_state->base.src.y1 >> 16;
BUG();
}
- if (i915_gem_object_is_tiled(obj))
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
dspcntr |= DISPPLANE_TILED;
if (!IS_HASWELL(dev) && !IS_BROADWELL(dev))
dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
- linear_offset = y * fb->pitches[0] + x * cpp;
+ intel_add_fb_offsets(&x, &y, plane_state, 0);
+
intel_crtc->dspaddr_offset =
- intel_compute_tile_offset(&x, &y, fb, 0,
- fb->pitches[0], rotation);
- linear_offset -= intel_crtc->dspaddr_offset;
+ intel_compute_tile_offset(&x, &y, plane_state, 0);
+
if (rotation == DRM_ROTATE_180) {
dspcntr |= DISPPLANE_ROTATE_180;
if (!IS_HASWELL(dev) && !IS_BROADWELL(dev)) {
x += (crtc_state->pipe_src_w - 1);
y += (crtc_state->pipe_src_h - 1);
-
- /* Finding the last pixel of the last line of the display
- data and adding to linear_offset*/
- linear_offset +=
- (crtc_state->pipe_src_h - 1) * fb->pitches[0] +
- (crtc_state->pipe_src_w - 1) * cpp;
}
}
+ linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
+
intel_crtc->adjusted_x = x;
intel_crtc->adjusted_y = y;
I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
I915_WRITE(DSPSURF(plane),
- i915_gem_obj_ggtt_offset(obj) + intel_crtc->dspaddr_offset);
+ intel_fb_gtt_offset(fb, rotation) +
+ intel_crtc->dspaddr_offset);
if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
I915_WRITE(DSPOFFSET(plane), (y << 16) | x);
} else {
}
}
-u32 intel_plane_obj_offset(struct intel_plane *intel_plane,
- struct drm_i915_gem_object *obj,
- unsigned int plane)
+u32 intel_fb_gtt_offset(struct drm_framebuffer *fb,
+ unsigned int rotation)
{
+ struct drm_i915_gem_object *obj = intel_fb_obj(fb);
struct i915_ggtt_view view;
- struct i915_vma *vma;
u64 offset;
- intel_fill_fb_ggtt_view(&view, intel_plane->base.state->fb,
- intel_plane->base.state->rotation);
-
- vma = i915_gem_obj_to_ggtt_view(obj, &view);
- if (WARN(!vma, "ggtt vma for display object not found! (view=%u)\n",
- view.type))
- return -1;
-
- offset = vma->node.start;
+ intel_fill_fb_ggtt_view(&view, fb, rotation);
- if (plane == 1) {
- offset += vma->ggtt_view.params.rotated.uv_start_page *
- PAGE_SIZE;
- }
+ offset = i915_gem_obj_ggtt_offset_view(obj, &view);
WARN_ON(upper_32_bits(offset));
}
}
+u32 skl_plane_stride(const struct drm_framebuffer *fb, int plane,
+ unsigned int rotation)
+{
+ const struct drm_i915_private *dev_priv = to_i915(fb->dev);
+ u32 stride = intel_fb_pitch(fb, plane, rotation);
+
+ /*
+ * The stride is either expressed as a multiple of 64 bytes chunks for
+ * linear buffers or in number of tiles for tiled buffers.
+ */
+ if (intel_rotation_90_or_270(rotation)) {
+ int cpp = drm_format_plane_cpp(fb->pixel_format, plane);
+
+ stride /= intel_tile_height(dev_priv, fb->modifier[0], cpp);
+ } else {
+ stride /= intel_fb_stride_alignment(dev_priv, fb->modifier[0],
+ fb->pixel_format);
+ }
+
+ return stride;
+}
+
u32 skl_plane_ctl_format(uint32_t pixel_format)
{
switch (pixel_format) {
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_crtc *intel_crtc = to_intel_crtc(crtc_state->base.crtc);
struct drm_framebuffer *fb = plane_state->base.fb;
- struct drm_i915_gem_object *obj = intel_fb_obj(fb);
int pipe = intel_crtc->pipe;
- u32 plane_ctl, stride_div, stride;
- u32 tile_height, plane_offset, plane_size;
+ u32 plane_ctl;
unsigned int rotation = plane_state->base.rotation;
- int x_offset, y_offset;
- u32 surf_addr;
+ u32 stride = skl_plane_stride(fb, 0, rotation);
+ u32 surf_addr = plane_state->main.offset;
int scaler_id = plane_state->scaler_id;
- int src_x = plane_state->base.src.x1 >> 16;
- int src_y = plane_state->base.src.y1 >> 16;
+ int src_x = plane_state->main.x;
+ int src_y = plane_state->main.y;
int src_w = drm_rect_width(&plane_state->base.src) >> 16;
int src_h = drm_rect_height(&plane_state->base.src) >> 16;
int dst_x = plane_state->base.dst.x1;
plane_ctl |= PLANE_CTL_PLANE_GAMMA_DISABLE;
plane_ctl |= skl_plane_ctl_rotation(rotation);
- stride_div = intel_fb_stride_alignment(dev_priv, fb->modifier[0],
- fb->pixel_format);
- surf_addr = intel_plane_obj_offset(to_intel_plane(plane), obj, 0);
+ /* Sizes are 0 based */
+ src_w--;
+ src_h--;
+ dst_w--;
+ dst_h--;
- WARN_ON(drm_rect_width(&plane_state->base.src) == 0);
-
- if (intel_rotation_90_or_270(rotation)) {
- int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
-
- /* stride = Surface height in tiles */
- tile_height = intel_tile_height(dev_priv, fb->modifier[0], cpp);
- stride = DIV_ROUND_UP(fb->height, tile_height);
- x_offset = stride * tile_height - src_y - src_h;
- y_offset = src_x;
- plane_size = (src_w - 1) << 16 | (src_h - 1);
- } else {
- stride = fb->pitches[0] / stride_div;
- x_offset = src_x;
- y_offset = src_y;
- plane_size = (src_h - 1) << 16 | (src_w - 1);
- }
- plane_offset = y_offset << 16 | x_offset;
-
- intel_crtc->adjusted_x = x_offset;
- intel_crtc->adjusted_y = y_offset;
+ intel_crtc->adjusted_x = src_x;
+ intel_crtc->adjusted_y = src_y;
I915_WRITE(PLANE_CTL(pipe, 0), plane_ctl);
- I915_WRITE(PLANE_OFFSET(pipe, 0), plane_offset);
- I915_WRITE(PLANE_SIZE(pipe, 0), plane_size);
+ I915_WRITE(PLANE_OFFSET(pipe, 0), (src_y << 16) | src_x);
I915_WRITE(PLANE_STRIDE(pipe, 0), stride);
+ I915_WRITE(PLANE_SIZE(pipe, 0), (src_h << 16) | src_w);
if (scaler_id >= 0) {
uint32_t ps_ctrl = 0;
I915_WRITE(PLANE_POS(pipe, 0), (dst_y << 16) | dst_x);
}
- I915_WRITE(PLANE_SURF(pipe, 0), surf_addr);
+ I915_WRITE(PLANE_SURF(pipe, 0),
+ intel_fb_gtt_offset(fb, rotation) + surf_addr);
POSTING_READ(PLANE_SURF(pipe, 0));
}
for_each_crtc(dev, crtc) {
struct intel_plane *plane = to_intel_plane(crtc->primary);
- struct intel_plane_state *plane_state;
-
- drm_modeset_lock_crtc(crtc, &plane->base);
- plane_state = to_intel_plane_state(plane->base.state);
+ struct intel_plane_state *plane_state =
+ to_intel_plane_state(plane->base.state);
if (plane_state->base.visible)
plane->update_plane(&plane->base,
to_intel_crtc_state(crtc->state),
plane_state);
+ }
+}
+
+static int
+__intel_display_resume(struct drm_device *dev,
+ struct drm_atomic_state *state)
+{
+ struct drm_crtc_state *crtc_state;
+ struct drm_crtc *crtc;
+ int i, ret;
- drm_modeset_unlock_crtc(crtc);
+ intel_modeset_setup_hw_state(dev);
+ i915_redisable_vga(dev);
+
+ if (!state)
+ return 0;
+
+ for_each_crtc_in_state(state, crtc, crtc_state, i) {
+ /*
+ * Force recalculation even if we restore
+ * current state. With fast modeset this may not result
+ * in a modeset when the state is compatible.
+ */
+ crtc_state->mode_changed = true;
}
+
+ /* ignore any reset values/BIOS leftovers in the WM registers */
+ to_intel_atomic_state(state)->skip_intermediate_wm = true;
+
+ ret = drm_atomic_commit(state);
+
+ WARN_ON(ret == -EDEADLK);
+ return ret;
+}
+
+static bool gpu_reset_clobbers_display(struct drm_i915_private *dev_priv)
+{
+ return intel_has_gpu_reset(dev_priv) &&
+ INTEL_GEN(dev_priv) < 5 && !IS_G4X(dev_priv);
}
void intel_prepare_reset(struct drm_i915_private *dev_priv)
{
- /* no reset support for gen2 */
- if (IS_GEN2(dev_priv))
- return;
+ struct drm_device *dev = &dev_priv->drm;
+ struct drm_modeset_acquire_ctx *ctx = &dev_priv->reset_ctx;
+ struct drm_atomic_state *state;
+ int ret;
- /* reset doesn't touch the display */
- if (INTEL_GEN(dev_priv) >= 5 || IS_G4X(dev_priv))
+ /*
+ * Need mode_config.mutex so that we don't
+ * trample ongoing ->detect() and whatnot.
+ */
+ mutex_lock(&dev->mode_config.mutex);
+ drm_modeset_acquire_init(ctx, 0);
+ while (1) {
+ ret = drm_modeset_lock_all_ctx(dev, ctx);
+ if (ret != -EDEADLK)
+ break;
+
+ drm_modeset_backoff(ctx);
+ }
+
+ /* reset doesn't touch the display, but flips might get nuked anyway, */
+ if (!i915.force_reset_modeset_test &&
+ !gpu_reset_clobbers_display(dev_priv))
return;
- drm_modeset_lock_all(&dev_priv->drm);
/*
* Disabling the crtcs gracefully seems nicer. Also the
* g33 docs say we should at least disable all the planes.
*/
- intel_display_suspend(&dev_priv->drm);
+ state = drm_atomic_helper_duplicate_state(dev, ctx);
+ if (IS_ERR(state)) {
+ ret = PTR_ERR(state);
+ state = NULL;
+ DRM_ERROR("Duplicating state failed with %i\n", ret);
+ goto err;
+ }
+
+ ret = drm_atomic_helper_disable_all(dev, ctx);
+ if (ret) {
+ DRM_ERROR("Suspending crtc's failed with %i\n", ret);
+ goto err;
+ }
+
+ dev_priv->modeset_restore_state = state;
+ state->acquire_ctx = ctx;
+ return;
+
+err:
+ drm_atomic_state_free(state);
}
void intel_finish_reset(struct drm_i915_private *dev_priv)
{
+ struct drm_device *dev = &dev_priv->drm;
+ struct drm_modeset_acquire_ctx *ctx = &dev_priv->reset_ctx;
+ struct drm_atomic_state *state = dev_priv->modeset_restore_state;
+ int ret;
+
/*
* Flips in the rings will be nuked by the reset,
* so complete all pending flips so that user space
*/
intel_complete_page_flips(dev_priv);
- /* no reset support for gen2 */
- if (IS_GEN2(dev_priv))
- return;
+ dev_priv->modeset_restore_state = NULL;
/* reset doesn't touch the display */
- if (INTEL_GEN(dev_priv) >= 5 || IS_G4X(dev_priv)) {
+ if (!gpu_reset_clobbers_display(dev_priv)) {
+ if (!state) {
+ /*
+ * Flips in the rings have been nuked by the reset,
+ * so update the base address of all primary
+ * planes to the the last fb to make sure we're
+ * showing the correct fb after a reset.
+ *
+ * FIXME: Atomic will make this obsolete since we won't schedule
+ * CS-based flips (which might get lost in gpu resets) any more.
+ */
+ intel_update_primary_planes(dev);
+ } else {
+ ret = __intel_display_resume(dev, state);
+ if (ret)
+ DRM_ERROR("Restoring old state failed with %i\n", ret);
+ }
+ } else {
/*
- * Flips in the rings have been nuked by the reset,
- * so update the base address of all primary
- * planes to the the last fb to make sure we're
- * showing the correct fb after a reset.
- *
- * FIXME: Atomic will make this obsolete since we won't schedule
- * CS-based flips (which might get lost in gpu resets) any more.
+ * The display has been reset as well,
+ * so need a full re-initialization.
*/
- intel_update_primary_planes(&dev_priv->drm);
- return;
- }
+ intel_runtime_pm_disable_interrupts(dev_priv);
+ intel_runtime_pm_enable_interrupts(dev_priv);
- /*
- * The display has been reset as well,
- * so need a full re-initialization.
- */
- intel_runtime_pm_disable_interrupts(dev_priv);
- intel_runtime_pm_enable_interrupts(dev_priv);
-
- intel_modeset_init_hw(&dev_priv->drm);
+ intel_modeset_init_hw(dev);
- spin_lock_irq(&dev_priv->irq_lock);
- if (dev_priv->display.hpd_irq_setup)
- dev_priv->display.hpd_irq_setup(dev_priv);
- spin_unlock_irq(&dev_priv->irq_lock);
+ spin_lock_irq(&dev_priv->irq_lock);
+ if (dev_priv->display.hpd_irq_setup)
+ dev_priv->display.hpd_irq_setup(dev_priv);
+ spin_unlock_irq(&dev_priv->irq_lock);
- intel_display_resume(&dev_priv->drm);
+ ret = __intel_display_resume(dev, state);
+ if (ret)
+ DRM_ERROR("Restoring old state failed with %i\n", ret);
- intel_hpd_init(dev_priv);
+ intel_hpd_init(dev_priv);
+ }
- drm_modeset_unlock_all(&dev_priv->drm);
+ drm_modeset_drop_locks(ctx);
+ drm_modeset_acquire_fini(ctx);
+ mutex_unlock(&dev->mode_config.mutex);
}
static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
I915_STATE_WARN(I915_READ(SPLL_CTL) & SPLL_PLL_ENABLE, "SPLL enabled\n");
I915_STATE_WARN(I915_READ(WRPLL_CTL(0)) & WRPLL_PLL_ENABLE, "WRPLL1 enabled\n");
I915_STATE_WARN(I915_READ(WRPLL_CTL(1)) & WRPLL_PLL_ENABLE, "WRPLL2 enabled\n");
- I915_STATE_WARN(I915_READ(PCH_PP_STATUS) & PP_ON, "Panel power on\n");
+ I915_STATE_WARN(I915_READ(PP_STATUS(0)) & PP_ON, "Panel power on\n");
I915_STATE_WARN(I915_READ(BLC_PWM_CPU_CTL2) & BLM_PWM_ENABLE,
"CPU PWM1 enabled\n");
if (IS_HASWELL(dev))
MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
intel_ring_emit(ring, fb->pitches[0]);
intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset |
- i915_gem_object_get_tiling(obj));
+ intel_fb_modifier_to_tiling(fb->modifier[0]));
/* XXX Enabling the panel-fitter across page-flip is so far
* untested on non-native modes, so ignore it for now.
intel_ring_emit(ring, MI_DISPLAY_FLIP |
MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
- intel_ring_emit(ring, fb->pitches[0] | i915_gem_object_get_tiling(obj));
+ intel_ring_emit(ring, fb->pitches[0] |
+ intel_fb_modifier_to_tiling(fb->modifier[0]));
intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset);
/* Contrary to the suggestions in the documentation,
}
intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit);
- intel_ring_emit(ring, fb->pitches[0] | i915_gem_object_get_tiling(obj));
+ intel_ring_emit(ring, fb->pitches[0] |
+ intel_fb_modifier_to_tiling(fb->modifier[0]));
intel_ring_emit(ring, intel_crtc->flip_work->gtt_offset);
intel_ring_emit(ring, (MI_NOOP));
struct drm_i915_private *dev_priv = to_i915(dev);
struct drm_framebuffer *fb = intel_crtc->base.primary->fb;
const enum pipe pipe = intel_crtc->pipe;
- u32 ctl, stride, tile_height;
+ u32 ctl, stride = skl_plane_stride(fb, 0, rotation);
ctl = I915_READ(PLANE_CTL(pipe, 0));
ctl &= ~PLANE_CTL_TILED_MASK;
}
/*
- * The stride is either expressed as a multiple of 64 bytes chunks for
- * linear buffers or in number of tiles for tiled buffers.
- */
- if (intel_rotation_90_or_270(rotation)) {
- /* stride = Surface height in tiles */
- tile_height = intel_tile_height(dev_priv, fb->modifier[0], 0);
- stride = DIV_ROUND_UP(fb->height, tile_height);
- } else {
- stride = fb->pitches[0] /
- intel_fb_stride_alignment(dev_priv, fb->modifier[0],
- fb->pixel_format);
- }
-
- /*
* Both PLANE_CTL and PLANE_STRIDE are not updated on vblank but on
* PLANE_SURF updates, the update is then guaranteed to be atomic.
*/
{
struct drm_device *dev = intel_crtc->base.dev;
struct drm_i915_private *dev_priv = to_i915(dev);
- struct intel_framebuffer *intel_fb =
- to_intel_framebuffer(intel_crtc->base.primary->fb);
- struct drm_i915_gem_object *obj = intel_fb->obj;
+ struct drm_framebuffer *fb = intel_crtc->base.primary->fb;
i915_reg_t reg = DSPCNTR(intel_crtc->plane);
u32 dspcntr;
dspcntr = I915_READ(reg);
- if (i915_gem_object_is_tiled(obj))
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
dspcntr |= DISPPLANE_TILED;
else
dspcntr &= ~DISPPLANE_TILED;
if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
engine = &dev_priv->engine[BCS];
- if (i915_gem_object_get_tiling(obj) !=
- i915_gem_object_get_tiling(intel_fb_obj(work->old_fb)))
+ if (fb->modifier[0] != old_fb->modifier[0])
/* vlv: DISPLAY_FLIP fails to change tiling */
engine = NULL;
} else if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
if (ret)
goto cleanup_pending;
- work->gtt_offset = intel_plane_obj_offset(to_intel_plane(primary),
- obj, 0);
+ work->gtt_offset = intel_fb_gtt_offset(fb, primary->state->rotation);
work->gtt_offset += intel_crtc->dspaddr_offset;
work->rotation = crtc->primary->state->rotation;
struct intel_crtc_state *crtc_state,
struct intel_plane_state *state)
{
+ struct drm_i915_private *dev_priv = to_i915(plane->dev);
struct drm_crtc *crtc = state->base.crtc;
int min_scale = DRM_PLANE_HELPER_NO_SCALING;
int max_scale = DRM_PLANE_HELPER_NO_SCALING;
bool can_position = false;
+ int ret;
- if (INTEL_INFO(plane->dev)->gen >= 9) {
+ if (INTEL_GEN(dev_priv) >= 9) {
/* use scaler when colorkey is not required */
if (state->ckey.flags == I915_SET_COLORKEY_NONE) {
min_scale = 1;
can_position = true;
}
- return drm_plane_helper_check_state(&state->base,
- &state->clip,
- min_scale, max_scale,
- can_position, true);
+ ret = drm_plane_helper_check_state(&state->base,
+ &state->clip,
+ min_scale, max_scale,
+ can_position, true);
+ if (ret)
+ return ret;
+
+ if (!state->base.fb)
+ return 0;
+
+ if (INTEL_GEN(dev_priv) >= 9) {
+ ret = skl_check_plane_surface(state);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
}
static void intel_begin_crtc_commit(struct drm_crtc *crtc,
return true;
}
+void intel_pps_unlock_regs_wa(struct drm_i915_private *dev_priv)
+{
+ int pps_num;
+ int pps_idx;
+
+ if (HAS_DDI(dev_priv))
+ return;
+ /*
+ * This w/a is needed at least on CPT/PPT, but to be sure apply it
+ * everywhere where registers can be write protected.
+ */
+ if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+ pps_num = 2;
+ else
+ pps_num = 1;
+
+ for (pps_idx = 0; pps_idx < pps_num; pps_idx++) {
+ u32 val = I915_READ(PP_CONTROL(pps_idx));
+
+ val = (val & ~PANEL_UNLOCK_MASK) | PANEL_UNLOCK_REGS;
+ I915_WRITE(PP_CONTROL(pps_idx), val);
+ }
+}
+
+static void intel_pps_init(struct drm_i915_private *dev_priv)
+{
+ if (HAS_PCH_SPLIT(dev_priv) || IS_BROXTON(dev_priv))
+ dev_priv->pps_mmio_base = PCH_PPS_BASE;
+ else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+ dev_priv->pps_mmio_base = VLV_PPS_BASE;
+ else
+ dev_priv->pps_mmio_base = PPS_BASE;
+
+ intel_pps_unlock_regs_wa(dev_priv);
+}
+
static void intel_setup_outputs(struct drm_device *dev)
{
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_encoder *encoder;
bool dpd_is_edp = false;
+ intel_pps_init(dev_priv);
+
/*
* intel_edp_init_connector() depends on this completing first, to
* prevent the registeration of both eDP and LVDS and the incorrect
struct drm_i915_gem_object *obj)
{
struct drm_i915_private *dev_priv = to_i915(dev);
- unsigned int aligned_height;
+ unsigned int tiling = i915_gem_object_get_tiling(obj);
int ret;
u32 pitch_limit, stride_alignment;
WARN_ON(!mutex_is_locked(&dev->struct_mutex));
if (mode_cmd->flags & DRM_MODE_FB_MODIFIERS) {
- /* Enforce that fb modifier and tiling mode match, but only for
- * X-tiled. This is needed for FBC. */
- if (!!(i915_gem_object_get_tiling(obj) == I915_TILING_X) !=
- !!(mode_cmd->modifier[0] == I915_FORMAT_MOD_X_TILED)) {
+ /*
+ * If there's a fence, enforce that
+ * the fb modifier and tiling mode match.
+ */
+ if (tiling != I915_TILING_NONE &&
+ tiling != intel_fb_modifier_to_tiling(mode_cmd->modifier[0])) {
DRM_DEBUG("tiling_mode doesn't match fb modifier\n");
return -EINVAL;
}
} else {
- if (i915_gem_object_get_tiling(obj) == I915_TILING_X)
+ if (tiling == I915_TILING_X) {
mode_cmd->modifier[0] = I915_FORMAT_MOD_X_TILED;
- else if (i915_gem_object_get_tiling(obj) == I915_TILING_Y) {
+ } else if (tiling == I915_TILING_Y) {
DRM_DEBUG("No Y tiling for legacy addfb\n");
return -EINVAL;
}
return -EINVAL;
}
+ /*
+ * gen2/3 display engine uses the fence if present,
+ * so the tiling mode must match the fb modifier exactly.
+ */
+ if (INTEL_INFO(dev_priv)->gen < 4 &&
+ tiling != intel_fb_modifier_to_tiling(mode_cmd->modifier[0])) {
+ DRM_DEBUG("tiling_mode must match fb modifier exactly on gen2/3\n");
+ return -EINVAL;
+ }
+
stride_alignment = intel_fb_stride_alignment(dev_priv,
mode_cmd->modifier[0],
mode_cmd->pixel_format);
return -EINVAL;
}
- if (mode_cmd->modifier[0] == I915_FORMAT_MOD_X_TILED &&
+ /*
+ * If there's a fence, enforce that
+ * the fb pitch and fence stride match.
+ */
+ if (tiling != I915_TILING_NONE &&
mode_cmd->pitches[0] != i915_gem_object_get_stride(obj)) {
DRM_DEBUG("pitch (%d) must match tiling stride (%d)\n",
mode_cmd->pitches[0],
if (mode_cmd->offsets[0] != 0)
return -EINVAL;
- aligned_height = intel_fb_align_height(dev, mode_cmd->height,
- mode_cmd->pixel_format,
- mode_cmd->modifier[0]);
- /* FIXME drm helper for size checks (especially planar formats)? */
- if (obj->base.size < aligned_height * mode_cmd->pitches[0])
- return -EINVAL;
-
drm_helper_mode_fill_fb_struct(&intel_fb->base, mode_cmd);
intel_fb->obj = obj;
- intel_fill_fb_info(dev_priv, &intel_fb->base);
+ ret = intel_fill_fb_info(dev_priv, &intel_fb->base);
+ if (ret)
+ return ret;
ret = drm_framebuffer_init(dev, &intel_fb->base, &intel_fb_funcs);
if (ret) {
return false;
}
+static bool has_pch_trancoder(struct drm_i915_private *dev_priv,
+ enum transcoder pch_transcoder)
+{
+ return HAS_PCH_IBX(dev_priv) || HAS_PCH_CPT(dev_priv) ||
+ (HAS_PCH_LPT_H(dev_priv) && pch_transcoder == TRANSCODER_A);
+}
+
static void intel_sanitize_crtc(struct intel_crtc *crtc)
{
struct drm_device *dev = crtc->base.dev;
* worst a fifo underrun happens which also sets this to false.
*/
crtc->cpu_fifo_underrun_disabled = true;
- crtc->pch_fifo_underrun_disabled = true;
+ /*
+ * We track the PCH trancoder underrun reporting state
+ * within the crtc. With crtc for pipe A housing the underrun
+ * reporting state for PCH transcoder A, crtc for pipe B housing
+ * it for PCH transcoder B, etc. LPT-H has only PCH transcoder A,
+ * and marking underrun reporting as disabled for the non-existing
+ * PCH transcoders B and C would prevent enabling the south
+ * error interrupt (see cpt_can_enable_serr_int()).
+ */
+ if (has_pch_trancoder(dev_priv, (enum transcoder)crtc->pipe))
+ crtc->pch_fifo_underrun_disabled = true;
}
}
struct drm_atomic_state *state = dev_priv->modeset_restore_state;
struct drm_modeset_acquire_ctx ctx;
int ret;
- bool setup = false;
dev_priv->modeset_restore_state = NULL;
+ if (state)
+ state->acquire_ctx = &ctx;
/*
* This is a cludge because with real atomic modeset mode_config.mutex
mutex_lock(&dev->mode_config.mutex);
drm_modeset_acquire_init(&ctx, 0);
-retry:
- ret = drm_modeset_lock_all_ctx(dev, &ctx);
-
- if (ret == 0 && !setup) {
- setup = true;
-
- intel_modeset_setup_hw_state(dev);
- i915_redisable_vga(dev);
- }
-
- if (ret == 0 && state) {
- struct drm_crtc_state *crtc_state;
- struct drm_crtc *crtc;
- int i;
-
- state->acquire_ctx = &ctx;
-
- /* ignore any reset values/BIOS leftovers in the WM registers */
- to_intel_atomic_state(state)->skip_intermediate_wm = true;
-
- for_each_crtc_in_state(state, crtc, crtc_state, i) {
- /*
- * Force recalculation even if we restore
- * current state. With fast modeset this may not result
- * in a modeset when the state is compatible.
- */
- crtc_state->mode_changed = true;
- }
-
- ret = drm_atomic_commit(state);
- }
+ while (1) {
+ ret = drm_modeset_lock_all_ctx(dev, &ctx);
+ if (ret != -EDEADLK)
+ break;
- if (ret == -EDEADLK) {
drm_modeset_backoff(&ctx);
- goto retry;
}
+ if (!ret)
+ ret = __intel_display_resume(dev, state);
+
drm_modeset_drop_locks(&ctx);
drm_modeset_acquire_fini(&ctx);
mutex_unlock(&dev->mode_config.mutex);
static void
intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev,
struct intel_dp *intel_dp);
+static void
+intel_dp_pps_init(struct drm_device *dev, struct intel_dp *intel_dp);
static void pps_lock(struct intel_dp *intel_dp)
{
static bool vlv_pipe_has_pp_on(struct drm_i915_private *dev_priv,
enum pipe pipe)
{
- return I915_READ(VLV_PIPE_PP_STATUS(pipe)) & PP_ON;
+ return I915_READ(PP_STATUS(pipe)) & PP_ON;
}
static bool vlv_pipe_has_vdd_on(struct drm_i915_private *dev_priv,
enum pipe pipe)
{
- return I915_READ(VLV_PIPE_PP_CONTROL(pipe)) & EDP_FORCE_VDD;
+ return I915_READ(PP_CONTROL(pipe)) & EDP_FORCE_VDD;
}
static bool vlv_pipe_any(struct drm_i915_private *dev_priv,
enum pipe pipe;
for (pipe = PIPE_A; pipe <= PIPE_B; pipe++) {
- u32 port_sel = I915_READ(VLV_PIPE_PP_ON_DELAYS(pipe)) &
+ u32 port_sel = I915_READ(PP_ON_DELAYS(pipe)) &
PANEL_PORT_SELECT_MASK;
if (port_sel != PANEL_PORT_SELECT_VLV(port))
struct intel_dp *intel_dp,
struct pps_registers *regs)
{
+ int pps_idx = 0;
+
memset(regs, 0, sizeof(*regs));
- if (IS_BROXTON(dev_priv)) {
- int idx = bxt_power_sequencer_idx(intel_dp);
-
- regs->pp_ctrl = BXT_PP_CONTROL(idx);
- regs->pp_stat = BXT_PP_STATUS(idx);
- regs->pp_on = BXT_PP_ON_DELAYS(idx);
- regs->pp_off = BXT_PP_OFF_DELAYS(idx);
- } else if (HAS_PCH_SPLIT(dev_priv)) {
- regs->pp_ctrl = PCH_PP_CONTROL;
- regs->pp_stat = PCH_PP_STATUS;
- regs->pp_on = PCH_PP_ON_DELAYS;
- regs->pp_off = PCH_PP_OFF_DELAYS;
- regs->pp_div = PCH_PP_DIVISOR;
- } else {
- enum pipe pipe = vlv_power_sequencer_pipe(intel_dp);
+ if (IS_BROXTON(dev_priv))
+ pps_idx = bxt_power_sequencer_idx(intel_dp);
+ else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
+ pps_idx = vlv_power_sequencer_pipe(intel_dp);
- regs->pp_ctrl = VLV_PIPE_PP_CONTROL(pipe);
- regs->pp_stat = VLV_PIPE_PP_STATUS(pipe);
- regs->pp_on = VLV_PIPE_PP_ON_DELAYS(pipe);
- regs->pp_off = VLV_PIPE_PP_OFF_DELAYS(pipe);
- regs->pp_div = VLV_PIPE_PP_DIVISOR(pipe);
- }
+ regs->pp_ctrl = PP_CONTROL(pps_idx);
+ regs->pp_stat = PP_STATUS(pps_idx);
+ regs->pp_on = PP_ON_DELAYS(pps_idx);
+ regs->pp_off = PP_OFF_DELAYS(pps_idx);
+ if (!IS_BROXTON(dev_priv))
+ regs->pp_div = PP_DIVISOR(pps_idx);
}
static i915_reg_t
i915_reg_t pp_ctrl_reg, pp_div_reg;
u32 pp_div;
- pp_ctrl_reg = VLV_PIPE_PP_CONTROL(pipe);
- pp_div_reg = VLV_PIPE_PP_DIVISOR(pipe);
+ pp_ctrl_reg = PP_CONTROL(pipe);
+ pp_div_reg = PP_DIVISOR(pipe);
pp_div = I915_READ(pp_div_reg);
pp_div &= PP_REFERENCE_DIVIDER_MASK;
lockdep_assert_held(&dev_priv->pps_mutex);
control = I915_READ(_pp_ctrl_reg(intel_dp));
- if (!IS_BROXTON(dev)) {
+ if (WARN_ON(!HAS_DDI(dev_priv) &&
+ (control & PANEL_UNLOCK_MASK) != PANEL_UNLOCK_REGS)) {
control &= ~PANEL_UNLOCK_MASK;
control |= PANEL_UNLOCK_REGS;
}
DRM_DEBUG_KMS("PP_STATUS: 0x%08x PP_CONTROL: 0x%08x\n",
I915_READ(pp_stat_reg), I915_READ(pp_ctrl_reg));
- if ((pp & POWER_TARGET_ON) == 0)
+ if ((pp & PANEL_POWER_ON) == 0)
intel_dp->panel_power_off_time = ktime_get_boottime();
power_domain = intel_display_port_aux_power_domain(intel_encoder);
POSTING_READ(pp_ctrl_reg);
}
- pp |= POWER_TARGET_ON;
+ pp |= PANEL_POWER_ON;
if (!IS_GEN5(dev))
pp |= PANEL_POWER_RESET;
pp = ironlake_get_pp_control(intel_dp);
/* We need to switch off panel power _and_ force vdd, for otherwise some
* panels get very unhappy and cease to work. */
- pp &= ~(POWER_TARGET_ON | PANEL_POWER_RESET | EDP_FORCE_VDD |
+ pp &= ~(PANEL_POWER_ON | PANEL_POWER_RESET | EDP_FORCE_VDD |
EDP_BLC_ENABLE);
pp_ctrl_reg = _pp_ctrl_reg(intel_dp);
struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
struct drm_i915_private *dev_priv = to_i915(intel_dig_port->base.base.dev);
enum pipe pipe = intel_dp->pps_pipe;
- i915_reg_t pp_on_reg = VLV_PIPE_PP_ON_DELAYS(pipe);
+ i915_reg_t pp_on_reg = PP_ON_DELAYS(pipe);
edp_panel_vdd_off_sync(intel_dp);
pps_lock(intel_dp);
- /*
- * Read out the current power sequencer assignment,
- * in case the BIOS did something with it.
- */
- if (IS_VALLEYVIEW(encoder->dev) || IS_CHERRYVIEW(encoder->dev))
- vlv_initial_power_sequencer_setup(intel_dp);
-
+ /* Reinit the power sequencer, in case BIOS did something with it. */
+ intel_dp_pps_init(encoder->dev, intel_dp);
intel_edp_panel_vdd_sanitize(intel_dp);
pps_unlock(intel_dp);
I915_READ(regs.pp_div));
}
+static void intel_dp_pps_init(struct drm_device *dev,
+ struct intel_dp *intel_dp)
+{
+ if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
+ vlv_initial_power_sequencer_setup(intel_dp);
+ } else {
+ intel_dp_init_panel_power_sequencer(dev, intel_dp);
+ intel_dp_init_panel_power_sequencer_registers(dev, intel_dp);
+ }
+}
+
/**
* intel_dp_set_drrs_state - program registers for RR switch to take effect
* @dev: DRM device
pps_lock(intel_dp);
intel_dp_init_panel_power_timestamps(intel_dp);
-
- if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
- vlv_initial_power_sequencer_setup(intel_dp);
- } else {
- intel_dp_init_panel_power_sequencer(dev, intel_dp);
- intel_dp_init_panel_power_sequencer_registers(dev, intel_dp);
- }
-
+ intel_dp_pps_init(dev, intel_dp);
intel_edp_panel_vdd_sanitize(intel_dp);
pps_unlock(intel_dp);
struct drm_framebuffer base;
struct drm_i915_gem_object *obj;
struct intel_rotation_info rot_info;
+
+ /* for each plane in the normal GTT view */
+ struct {
+ unsigned int x, y;
+ } normal[2];
+ /* for each plane in the rotated GTT view */
+ struct {
+ unsigned int x, y;
+ unsigned int pitch; /* pixels */
+ } rotated[2];
};
struct intel_fbdev {
struct drm_plane_state base;
struct drm_rect clip;
+ struct {
+ u32 offset;
+ int x, y;
+ } main;
+ struct {
+ u32 offset;
+ int x, y;
+ } aux;
+
/*
* scaler_id
* = -1 : not using a scaler
const char *name, u32 reg, int ref_freq);
extern const struct drm_plane_funcs intel_plane_funcs;
void intel_init_display_hooks(struct drm_i915_private *dev_priv);
+unsigned int intel_fb_xy_to_linear(int x, int y,
+ const struct intel_plane_state *state,
+ int plane);
+void intel_add_fb_offsets(int *x, int *y,
+ const struct intel_plane_state *state, int plane);
unsigned int intel_rotation_info_size(const struct intel_rotation_info *rot_info);
bool intel_has_pending_fb_unpin(struct drm_device *dev);
void intel_mark_busy(struct drm_i915_private *dev_priv);
void intel_mark_idle(struct drm_i915_private *dev_priv);
void intel_crtc_restore_mode(struct drm_crtc *crtc);
int intel_display_suspend(struct drm_device *dev);
+void intel_pps_unlock_regs_wa(struct drm_i915_private *dev_priv);
void intel_encoder_destroy(struct drm_encoder *encoder);
int intel_connector_init(struct intel_connector *);
struct intel_connector *intel_connector_alloc(void);
#define assert_pipe_enabled(d, p) assert_pipe(d, p, true)
#define assert_pipe_disabled(d, p) assert_pipe(d, p, false)
u32 intel_compute_tile_offset(int *x, int *y,
- const struct drm_framebuffer *fb, int plane,
- unsigned int pitch,
- unsigned int rotation);
+ const struct intel_plane_state *state, int plane);
void intel_prepare_reset(struct drm_i915_private *dev_priv);
void intel_finish_reset(struct drm_i915_private *dev_priv);
void hsw_enable_pc8(struct drm_i915_private *dev_priv);
int skl_update_scaler_crtc(struct intel_crtc_state *crtc_state);
int skl_max_scale(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state);
-u32 intel_plane_obj_offset(struct intel_plane *intel_plane,
- struct drm_i915_gem_object *obj,
- unsigned int plane);
+u32 intel_fb_gtt_offset(struct drm_framebuffer *fb, unsigned int rotation);
u32 skl_plane_ctl_format(uint32_t pixel_format);
u32 skl_plane_ctl_tiling(uint64_t fb_modifier);
u32 skl_plane_ctl_rotation(unsigned int rotation);
+u32 skl_plane_stride(const struct drm_framebuffer *fb, int plane,
+ unsigned int rotation);
+int skl_check_plane_surface(struct intel_plane_state *plane_state);
/* intel_csr.c */
void intel_csr_ucode_init(struct drm_i915_private *);
int intel_engines_init(struct drm_device *dev)
{
struct drm_i915_private *dev_priv = to_i915(dev);
+ struct intel_device_info *device_info = mkwrite_device_info(dev_priv);
unsigned int mask = 0;
int (*init)(struct intel_engine_cs *engine);
unsigned int i;
* are added to the driver by a warning and disabling the forgotten
* engines.
*/
- if (WARN_ON(mask != INTEL_INFO(dev_priv)->ring_mask)) {
- struct intel_device_info *info =
- (struct intel_device_info *)&dev_priv->info;
- info->ring_mask = mask;
- }
+ if (WARN_ON(mask != INTEL_INFO(dev_priv)->ring_mask))
+ device_info->ring_mask = mask;
+
+ device_info->num_rings = hweight32(mask);
return 0;
void intel_engine_init_hangcheck(struct intel_engine_cs *engine)
{
memset(&engine->hangcheck, 0, sizeof(engine->hangcheck));
+ clear_bit(engine->id, &engine->i915->gpu_error.missed_irq_rings);
}
static void intel_engine_init_requests(struct intel_engine_cs *engine)
void *client_base; /* first page (only) of above */
struct i915_gem_context *owner;
struct intel_guc *guc;
+
+ uint32_t engines; /* bitmap of (host) engine ids */
uint32_t priority;
uint32_t ctx_index;
-
uint32_t proc_desc_offset;
+
uint32_t doorbell_offset;
uint32_t cookie;
uint16_t doorbell_id;
- uint16_t padding; /* Maintain alignment */
+ uint16_t padding[3]; /* Maintain alignment */
uint32_t wq_offset;
uint32_t wq_size;
uint32_t wq_tail;
- uint32_t unused; /* Was 'wq_head' */
-
uint32_t no_wq_space;
- uint32_t q_fail; /* No longer used */
uint32_t b_fail;
int retcode;
*
*/
-#define I915_SKL_GUC_UCODE "i915/skl_guc_ver6_1.bin"
+#define SKL_FW_MAJOR 6
+#define SKL_FW_MINOR 1
+
+#define BXT_FW_MAJOR 8
+#define BXT_FW_MINOR 7
+
+#define KBL_FW_MAJOR 9
+#define KBL_FW_MINOR 14
+
+#define GUC_FW_PATH(platform, major, minor) \
+ "i915/" __stringify(platform) "_guc_ver" __stringify(major) "_" __stringify(minor) ".bin"
+
+#define I915_SKL_GUC_UCODE GUC_FW_PATH(skl, SKL_FW_MAJOR, SKL_FW_MINOR)
MODULE_FIRMWARE(I915_SKL_GUC_UCODE);
-#define I915_BXT_GUC_UCODE "i915/bxt_guc_ver8_7.bin"
+#define I915_BXT_GUC_UCODE GUC_FW_PATH(bxt, BXT_FW_MAJOR, BXT_FW_MINOR)
MODULE_FIRMWARE(I915_BXT_GUC_UCODE);
-#define I915_KBL_GUC_UCODE "i915/kbl_guc_ver9_14.bin"
+#define I915_KBL_GUC_UCODE GUC_FW_PATH(kbl, KBL_FW_MAJOR, KBL_FW_MINOR)
MODULE_FIRMWARE(I915_KBL_GUC_UCODE);
/* User-friendly representation of an enum */
fw_path = NULL;
} else if (IS_SKYLAKE(dev)) {
fw_path = I915_SKL_GUC_UCODE;
- guc_fw->guc_fw_major_wanted = 6;
- guc_fw->guc_fw_minor_wanted = 1;
+ guc_fw->guc_fw_major_wanted = SKL_FW_MAJOR;
+ guc_fw->guc_fw_minor_wanted = SKL_FW_MINOR;
} else if (IS_BROXTON(dev)) {
fw_path = I915_BXT_GUC_UCODE;
- guc_fw->guc_fw_major_wanted = 8;
- guc_fw->guc_fw_minor_wanted = 7;
+ guc_fw->guc_fw_major_wanted = BXT_FW_MAJOR;
+ guc_fw->guc_fw_minor_wanted = BXT_FW_MINOR;
} else if (IS_KABYLAKE(dev)) {
fw_path = I915_KBL_GUC_UCODE;
- guc_fw->guc_fw_major_wanted = 9;
- guc_fw->guc_fw_minor_wanted = 14;
+ guc_fw->guc_fw_major_wanted = KBL_FW_MAJOR;
+ guc_fw->guc_fw_minor_wanted = KBL_FW_MINOR;
} else {
fw_path = ""; /* unknown device */
}
if (ret)
goto err;
- vaddr = i915_gem_object_pin_map(ce->state);
+ vaddr = i915_gem_object_pin_map(ce->state, I915_MAP_WB);
if (IS_ERR(vaddr)) {
ret = PTR_ERR(vaddr);
goto unpin_ctx_obj;
}
ret = i915_gem_object_ggtt_pin(engine->wa_ctx.obj, NULL,
- 0, PAGE_SIZE, 0);
+ 0, PAGE_SIZE, PIN_HIGH);
if (ret) {
DRM_DEBUG_DRIVER("pin LRC WA ctx backing obj failed: %d\n",
ret);
/* The HWSP is part of the default context object in LRC mode. */
engine->status_page.gfx_addr = i915_gem_obj_ggtt_offset(dctx_obj) +
LRC_PPHWSP_PN * PAGE_SIZE;
- hws = i915_gem_object_pin_map(dctx_obj);
+ hws = i915_gem_object_pin_map(dctx_obj, I915_MAP_WB);
if (IS_ERR(hws))
return PTR_ERR(hws);
engine->status_page.page_addr = hws + LRC_PPHWSP_PN * PAGE_SIZE;
return ret;
}
- vaddr = i915_gem_object_pin_map(ctx_obj);
+ vaddr = i915_gem_object_pin_map(ctx_obj, I915_MAP_WB);
if (IS_ERR(vaddr)) {
ret = PTR_ERR(vaddr);
DRM_DEBUG_DRIVER("Could not map object pages! (%d)\n", ret);
if (!ctx_obj)
continue;
- vaddr = i915_gem_object_pin_map(ctx_obj);
+ vaddr = i915_gem_object_pin_map(ctx_obj, I915_MAP_WB);
if (WARN_ON(IS_ERR(vaddr)))
continue;
struct notifier_block lid_notifier;
};
+struct intel_lvds_pps {
+ /* 100us units */
+ int t1_t2;
+ int t3;
+ int t4;
+ int t5;
+ int tx;
+
+ int divider;
+
+ int port;
+ bool powerdown_on_reset;
+};
+
struct intel_lvds_encoder {
struct intel_encoder base;
i915_reg_t reg;
u32 a3_power;
+ struct intel_lvds_pps init_pps;
+ u32 init_lvds_val;
+
struct intel_lvds_connector *attached_connector;
};
pipe_config->base.adjusted_mode.crtc_clock = pipe_config->port_clock;
}
+static void intel_lvds_pps_get_hw_state(struct drm_i915_private *dev_priv,
+ struct intel_lvds_pps *pps)
+{
+ u32 val;
+
+ pps->powerdown_on_reset = I915_READ(PP_CONTROL(0)) & PANEL_POWER_RESET;
+
+ val = I915_READ(PP_ON_DELAYS(0));
+ pps->port = (val & PANEL_PORT_SELECT_MASK) >>
+ PANEL_PORT_SELECT_SHIFT;
+ pps->t1_t2 = (val & PANEL_POWER_UP_DELAY_MASK) >>
+ PANEL_POWER_UP_DELAY_SHIFT;
+ pps->t5 = (val & PANEL_LIGHT_ON_DELAY_MASK) >>
+ PANEL_LIGHT_ON_DELAY_SHIFT;
+
+ val = I915_READ(PP_OFF_DELAYS(0));
+ pps->t3 = (val & PANEL_POWER_DOWN_DELAY_MASK) >>
+ PANEL_POWER_DOWN_DELAY_SHIFT;
+ pps->tx = (val & PANEL_LIGHT_OFF_DELAY_MASK) >>
+ PANEL_LIGHT_OFF_DELAY_SHIFT;
+
+ val = I915_READ(PP_DIVISOR(0));
+ pps->divider = (val & PP_REFERENCE_DIVIDER_MASK) >>
+ PP_REFERENCE_DIVIDER_SHIFT;
+ val = (val & PANEL_POWER_CYCLE_DELAY_MASK) >>
+ PANEL_POWER_CYCLE_DELAY_SHIFT;
+ /*
+ * Remove the BSpec specified +1 (100ms) offset that accounts for a
+ * too short power-cycle delay due to the asynchronous programming of
+ * the register.
+ */
+ if (val)
+ val--;
+ /* Convert from 100ms to 100us units */
+ pps->t4 = val * 1000;
+
+ if (INTEL_INFO(dev_priv)->gen <= 4 &&
+ pps->t1_t2 == 0 && pps->t5 == 0 && pps->t3 == 0 && pps->tx == 0) {
+ DRM_DEBUG_KMS("Panel power timings uninitialized, "
+ "setting defaults\n");
+ /* Set T2 to 40ms and T5 to 200ms in 100 usec units */
+ pps->t1_t2 = 40 * 10;
+ pps->t5 = 200 * 10;
+ /* Set T3 to 35ms and Tx to 200ms in 100 usec units */
+ pps->t3 = 35 * 10;
+ pps->tx = 200 * 10;
+ }
+
+ DRM_DEBUG_DRIVER("LVDS PPS:t1+t2 %d t3 %d t4 %d t5 %d tx %d "
+ "divider %d port %d powerdown_on_reset %d\n",
+ pps->t1_t2, pps->t3, pps->t4, pps->t5, pps->tx,
+ pps->divider, pps->port, pps->powerdown_on_reset);
+}
+
+static void intel_lvds_pps_init_hw(struct drm_i915_private *dev_priv,
+ struct intel_lvds_pps *pps)
+{
+ u32 val;
+
+ val = I915_READ(PP_CONTROL(0));
+ WARN_ON((val & PANEL_UNLOCK_MASK) != PANEL_UNLOCK_REGS);
+ if (pps->powerdown_on_reset)
+ val |= PANEL_POWER_RESET;
+ I915_WRITE(PP_CONTROL(0), val);
+
+ I915_WRITE(PP_ON_DELAYS(0), (pps->port << PANEL_PORT_SELECT_SHIFT) |
+ (pps->t1_t2 << PANEL_POWER_UP_DELAY_SHIFT) |
+ (pps->t5 << PANEL_LIGHT_ON_DELAY_SHIFT));
+ I915_WRITE(PP_OFF_DELAYS(0), (pps->t3 << PANEL_POWER_DOWN_DELAY_SHIFT) |
+ (pps->tx << PANEL_LIGHT_OFF_DELAY_SHIFT));
+
+ val = pps->divider << PP_REFERENCE_DIVIDER_SHIFT;
+ val |= (DIV_ROUND_UP(pps->t4, 1000) + 1) <<
+ PANEL_POWER_CYCLE_DELAY_SHIFT;
+ I915_WRITE(PP_DIVISOR(0), val);
+}
+
static void intel_pre_enable_lvds(struct intel_encoder *encoder)
{
struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
assert_pll_disabled(dev_priv, pipe);
}
- temp = I915_READ(lvds_encoder->reg);
+ intel_lvds_pps_init_hw(dev_priv, &lvds_encoder->init_pps);
+
+ temp = lvds_encoder->init_lvds_val;
temp |= LVDS_PORT_EN | LVDS_A0A2_CLKA_POWER_UP;
if (HAS_PCH_CPT(dev)) {
struct intel_connector *intel_connector =
&lvds_encoder->attached_connector->base;
struct drm_i915_private *dev_priv = to_i915(dev);
- i915_reg_t ctl_reg, stat_reg;
-
- if (HAS_PCH_SPLIT(dev)) {
- ctl_reg = PCH_PP_CONTROL;
- stat_reg = PCH_PP_STATUS;
- } else {
- ctl_reg = PP_CONTROL;
- stat_reg = PP_STATUS;
- }
I915_WRITE(lvds_encoder->reg, I915_READ(lvds_encoder->reg) | LVDS_PORT_EN);
- I915_WRITE(ctl_reg, I915_READ(ctl_reg) | POWER_TARGET_ON);
+ I915_WRITE(PP_CONTROL(0), I915_READ(PP_CONTROL(0)) | PANEL_POWER_ON);
POSTING_READ(lvds_encoder->reg);
- if (intel_wait_for_register(dev_priv, stat_reg, PP_ON, PP_ON, 1000))
+ if (intel_wait_for_register(dev_priv, PP_STATUS(0), PP_ON, PP_ON, 1000))
DRM_ERROR("timed out waiting for panel to power on\n");
intel_panel_enable_backlight(intel_connector);
struct drm_device *dev = encoder->base.dev;
struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
struct drm_i915_private *dev_priv = to_i915(dev);
- i915_reg_t ctl_reg, stat_reg;
-
- if (HAS_PCH_SPLIT(dev)) {
- ctl_reg = PCH_PP_CONTROL;
- stat_reg = PCH_PP_STATUS;
- } else {
- ctl_reg = PP_CONTROL;
- stat_reg = PP_STATUS;
- }
- I915_WRITE(ctl_reg, I915_READ(ctl_reg) & ~POWER_TARGET_ON);
- if (intel_wait_for_register(dev_priv, stat_reg, PP_ON, 0, 1000))
+ I915_WRITE(PP_CONTROL(0), I915_READ(PP_CONTROL(0)) & ~PANEL_POWER_ON);
+ if (intel_wait_for_register(dev_priv, PP_STATUS(0), PP_ON, 0, 1000))
DRM_ERROR("timed out waiting for panel to power off\n");
I915_WRITE(lvds_encoder->reg, I915_READ(lvds_encoder->reg) & ~LVDS_PORT_EN);
int pipe;
u8 pin;
- /*
- * Unlock registers and just leave them unlocked. Do this before
- * checking quirk lists to avoid bogus WARNINGs.
- */
- if (HAS_PCH_SPLIT(dev)) {
- I915_WRITE(PCH_PP_CONTROL,
- I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS);
- } else if (INTEL_INFO(dev_priv)->gen < 5) {
- I915_WRITE(PP_CONTROL,
- I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS);
- }
if (!intel_lvds_supported(dev))
return;
DRM_DEBUG_KMS("LVDS is not present in VBT, but enabled anyway\n");
}
- /* Set the Panel Power On/Off timings if uninitialized. */
- if (INTEL_INFO(dev_priv)->gen < 5 &&
- I915_READ(PP_ON_DELAYS) == 0 && I915_READ(PP_OFF_DELAYS) == 0) {
- /* Set T2 to 40ms and T5 to 200ms */
- I915_WRITE(PP_ON_DELAYS, 0x019007d0);
-
- /* Set T3 to 35ms and Tx to 200ms */
- I915_WRITE(PP_OFF_DELAYS, 0x015e07d0);
-
- DRM_DEBUG_KMS("Panel power timings uninitialized, setting defaults\n");
- }
-
lvds_encoder = kzalloc(sizeof(*lvds_encoder), GFP_KERNEL);
if (!lvds_encoder)
return;
dev->mode_config.scaling_mode_property,
DRM_MODE_SCALE_ASPECT);
intel_connector->panel.fitting_mode = DRM_MODE_SCALE_ASPECT;
+
+ intel_lvds_pps_get_hw_state(dev_priv, &lvds_encoder->init_pps);
+ lvds_encoder->init_lvds_val = lvds;
+
/*
* LVDS discovery:
* 1) check for EDID on DDC
u32 pcbr;
int pctx_size = 24*1024;
- mutex_lock(&dev_priv->drm.struct_mutex);
-
pcbr = I915_READ(VLV_PCBR);
if (pcbr) {
/* BIOS set it up already, grab the pre-alloc'd space */
out:
DRM_DEBUG_DRIVER("PCBR: 0x%08x\n", I915_READ(VLV_PCBR));
dev_priv->vlv_pctx = pctx;
- mutex_unlock(&dev_priv->drm.struct_mutex);
}
static void valleyview_cleanup_pctx(struct drm_i915_private *dev_priv)
intel_runtime_pm_get(dev_priv);
}
+ mutex_lock(&dev_priv->drm.struct_mutex);
mutex_lock(&dev_priv->rps.hw_lock);
/* Initialize RPS limits (for userspace) */
dev_priv->rps.boost_freq = dev_priv->rps.max_freq;
mutex_unlock(&dev_priv->rps.hw_lock);
+ mutex_unlock(&dev_priv->drm.struct_mutex);
intel_autoenable_gt_powersave(dev_priv);
}
enum intel_engine_id id;
int ret, num_rings;
- num_rings = hweight32(INTEL_INFO(dev_priv)->ring_mask);
+ num_rings = INTEL_INFO(dev_priv)->num_rings;
ret = intel_ring_begin(req, (num_rings-1) * 8);
if (ret)
return ret;
enum intel_engine_id id;
int ret, num_rings;
- num_rings = hweight32(INTEL_INFO(dev_priv)->ring_mask);
+ num_rings = INTEL_INFO(dev_priv)->num_rings;
ret = intel_ring_begin(req, (num_rings-1) * 6);
if (ret)
return ret;
enum intel_engine_id id;
int ret, num_rings;
- num_rings = hweight32(INTEL_INFO(dev_priv)->ring_mask);
+ num_rings = INTEL_INFO(dev_priv)->num_rings;
ret = intel_ring_begin(req, round_up((num_rings-1) * 3, 2));
if (ret)
return ret;
if (ret)
goto err_unpin;
- addr = i915_gem_object_pin_map(obj);
+ addr = i915_gem_object_pin_map(obj, I915_MAP_WB);
if (IS_ERR(addr)) {
ret = PTR_ERR(addr);
goto err_unpin;
if (ce->state) {
ret = i915_gem_object_ggtt_pin(ce->state, NULL, 0,
- ctx->ggtt_alignment, 0);
+ ctx->ggtt_alignment, PIN_HIGH);
if (ret)
goto error;
}
/* After manually advancing the seqno, fake the interrupt in case
* there are any waiters for that seqno.
*/
- rcu_read_lock();
intel_engine_wakeup(engine);
- rcu_read_unlock();
}
static void gen6_bsd_submit_request(struct drm_i915_gem_request *request)
i915.semaphores = 0;
} else {
i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
- ret = i915_gem_object_ggtt_pin(obj, NULL, 0, 0, 0);
+ ret = i915_gem_object_ggtt_pin(obj, NULL,
+ 0, 0, PIN_HIGH);
if (ret != 0) {
i915_gem_object_put(obj);
DRM_ERROR("Failed to pin semaphore bo. Disabling semaphores\n");
struct intel_engine_hangcheck {
u64 acthd;
- unsigned long user_interrupts;
u32 seqno;
int score;
enum intel_engine_hangcheck_action action;
* the overhead of waking that client is much preferred.
*/
struct intel_breadcrumbs {
- struct task_struct *irq_seqno_bh; /* bh for user interrupts */
- unsigned long irq_wakeups;
+ struct task_struct __rcu *irq_seqno_bh; /* bh for interrupts */
bool irq_posted;
spinlock_t lock; /* protects the lists of requests */
struct task_struct *signaler; /* used for fence signalling */
struct drm_i915_gem_request *first_signal;
struct timer_list fake_irq; /* used after a missed interrupt */
+ struct timer_list hangcheck; /* detect missed interrupts */
+
+ unsigned long timeout;
bool irq_enabled : 1;
bool rpm_wakelock : 1;
/* An RCU guarded pointer to the last request. No reference is
* held to the request, users must carefully acquire a reference to
- * the request using i915_gem_active_get_request_rcu(), or hold the
+ * the request using i915_gem_active_get_rcu(), or hold the
* struct_mutex.
*/
struct i915_gem_active last_request;
struct intel_wait *wait);
void intel_engine_enable_signaling(struct drm_i915_gem_request *request);
-static inline bool intel_engine_has_waiter(struct intel_engine_cs *engine)
+static inline bool intel_engine_has_waiter(const struct intel_engine_cs *engine)
{
- return READ_ONCE(engine->breadcrumbs.irq_seqno_bh);
+ return rcu_access_pointer(engine->breadcrumbs.irq_seqno_bh);
}
-static inline bool intel_engine_wakeup(struct intel_engine_cs *engine)
+static inline bool intel_engine_wakeup(const struct intel_engine_cs *engine)
{
bool wakeup = false;
- struct task_struct *tsk = READ_ONCE(engine->breadcrumbs.irq_seqno_bh);
+
/* Note that for this not to dangerously chase a dangling pointer,
- * the caller is responsible for ensure that the task remain valid for
- * wake_up_process() i.e. that the RCU grace period cannot expire.
+ * we must hold the rcu_read_lock here.
*
* Also note that tsk is likely to be in !TASK_RUNNING state so an
* early test for tsk->state != TASK_RUNNING before wake_up_process()
* is unlikely to be beneficial.
*/
- if (tsk)
- wakeup = wake_up_process(tsk);
+ if (intel_engine_has_waiter(engine)) {
+ struct task_struct *tsk;
+
+ rcu_read_lock();
+ tsk = rcu_dereference(engine->breadcrumbs.irq_seqno_bh);
+ if (tsk)
+ wakeup = wake_up_process(tsk);
+ rcu_read_unlock();
+ }
+
return wakeup;
}
-void intel_engine_enable_fake_irq(struct intel_engine_cs *engine);
void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine);
unsigned int intel_kick_waiters(struct drm_i915_private *i915);
unsigned int intel_kick_signalers(struct drm_i915_private *i915);
DRM_DEBUG_KMS("Disabling DC9\n");
gen9_set_dc_state(dev_priv, DC_STATE_DISABLE);
+
+ intel_pps_unlock_regs_wa(dev_priv);
}
static void assert_csr_loaded(struct drm_i915_private *dev_priv)
}
i915_redisable_vga_power_on(&dev_priv->drm);
+
+ intel_pps_unlock_regs_wa(dev_priv);
}
static void vlv_display_power_well_deinit(struct drm_i915_private *dev_priv)
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_plane *intel_plane = to_intel_plane(drm_plane);
struct drm_framebuffer *fb = plane_state->base.fb;
- struct drm_i915_gem_object *obj = intel_fb_obj(fb);
const int pipe = intel_plane->pipe;
const int plane = intel_plane->plane + 1;
- u32 plane_ctl, stride_div, stride;
+ u32 plane_ctl;
const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
- u32 surf_addr;
- u32 tile_height, plane_offset, plane_size;
+ u32 surf_addr = plane_state->main.offset;
unsigned int rotation = plane_state->base.rotation;
- int x_offset, y_offset;
+ u32 stride = skl_plane_stride(fb, 0, rotation);
int crtc_x = plane_state->base.dst.x1;
int crtc_y = plane_state->base.dst.y1;
uint32_t crtc_w = drm_rect_width(&plane_state->base.dst);
uint32_t crtc_h = drm_rect_height(&plane_state->base.dst);
- uint32_t x = plane_state->base.src.x1 >> 16;
- uint32_t y = plane_state->base.src.y1 >> 16;
+ uint32_t x = plane_state->main.x;
+ uint32_t y = plane_state->main.y;
uint32_t src_w = drm_rect_width(&plane_state->base.src) >> 16;
uint32_t src_h = drm_rect_height(&plane_state->base.src) >> 16;
plane_ctl |= skl_plane_ctl_rotation(rotation);
- stride_div = intel_fb_stride_alignment(dev_priv, fb->modifier[0],
- fb->pixel_format);
-
- /* Sizes are 0 based */
- src_w--;
- src_h--;
- crtc_w--;
- crtc_h--;
-
if (key->flags) {
I915_WRITE(PLANE_KEYVAL(pipe, plane), key->min_value);
I915_WRITE(PLANE_KEYMAX(pipe, plane), key->max_value);
else if (key->flags & I915_SET_COLORKEY_SOURCE)
plane_ctl |= PLANE_CTL_KEY_ENABLE_SOURCE;
- surf_addr = intel_plane_obj_offset(intel_plane, obj, 0);
-
- if (intel_rotation_90_or_270(rotation)) {
- int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
-
- /* stride: Surface height in tiles */
- tile_height = intel_tile_height(dev_priv, fb->modifier[0], cpp);
- stride = DIV_ROUND_UP(fb->height, tile_height);
- plane_size = (src_w << 16) | src_h;
- x_offset = stride * tile_height - y - (src_h + 1);
- y_offset = x;
- } else {
- stride = fb->pitches[0] / stride_div;
- plane_size = (src_h << 16) | src_w;
- x_offset = x;
- y_offset = y;
- }
- plane_offset = y_offset << 16 | x_offset;
+ /* Sizes are 0 based */
+ src_w--;
+ src_h--;
+ crtc_w--;
+ crtc_h--;
- I915_WRITE(PLANE_OFFSET(pipe, plane), plane_offset);
+ I915_WRITE(PLANE_OFFSET(pipe, plane), (y << 16) | x);
I915_WRITE(PLANE_STRIDE(pipe, plane), stride);
- I915_WRITE(PLANE_SIZE(pipe, plane), plane_size);
+ I915_WRITE(PLANE_SIZE(pipe, plane), (src_h << 16) | src_w);
/* program plane scaler */
if (plane_state->scaler_id >= 0) {
}
I915_WRITE(PLANE_CTL(pipe, plane), plane_ctl);
- I915_WRITE(PLANE_SURF(pipe, plane), surf_addr);
+ I915_WRITE(PLANE_SURF(pipe, plane),
+ intel_fb_gtt_offset(fb, rotation) + surf_addr);
POSTING_READ(PLANE_SURF(pipe, plane));
}
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_plane *intel_plane = to_intel_plane(dplane);
struct drm_framebuffer *fb = plane_state->base.fb;
- struct drm_i915_gem_object *obj = intel_fb_obj(fb);
int pipe = intel_plane->pipe;
int plane = intel_plane->plane;
u32 sprctl;
u32 sprsurf_offset, linear_offset;
unsigned int rotation = dplane->state->rotation;
- int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
int crtc_x = plane_state->base.dst.x1;
int crtc_y = plane_state->base.dst.y1;
*/
sprctl |= SP_GAMMA_ENABLE;
- if (i915_gem_object_is_tiled(obj))
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
sprctl |= SP_TILED;
/* Sizes are 0 based */
crtc_w--;
crtc_h--;
- linear_offset = y * fb->pitches[0] + x * cpp;
- sprsurf_offset = intel_compute_tile_offset(&x, &y, fb, 0,
- fb->pitches[0], rotation);
- linear_offset -= sprsurf_offset;
+ intel_add_fb_offsets(&x, &y, plane_state, 0);
+ sprsurf_offset = intel_compute_tile_offset(&x, &y, plane_state, 0);
if (rotation == DRM_ROTATE_180) {
sprctl |= SP_ROTATE_180;
x += src_w;
y += src_h;
- linear_offset += src_h * fb->pitches[0] + src_w * cpp;
}
+ linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
+
if (key->flags) {
I915_WRITE(SPKEYMINVAL(pipe, plane), key->min_value);
I915_WRITE(SPKEYMAXVAL(pipe, plane), key->max_value);
I915_WRITE(SPSTRIDE(pipe, plane), fb->pitches[0]);
I915_WRITE(SPPOS(pipe, plane), (crtc_y << 16) | crtc_x);
- if (i915_gem_object_is_tiled(obj))
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
I915_WRITE(SPTILEOFF(pipe, plane), (y << 16) | x);
else
I915_WRITE(SPLINOFF(pipe, plane), linear_offset);
I915_WRITE(SPSIZE(pipe, plane), (crtc_h << 16) | crtc_w);
I915_WRITE(SPCNTR(pipe, plane), sprctl);
- I915_WRITE(SPSURF(pipe, plane), i915_gem_obj_ggtt_offset(obj) +
- sprsurf_offset);
+ I915_WRITE(SPSURF(pipe, plane),
+ intel_fb_gtt_offset(fb, rotation) + sprsurf_offset);
POSTING_READ(SPSURF(pipe, plane));
}
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_plane *intel_plane = to_intel_plane(plane);
struct drm_framebuffer *fb = plane_state->base.fb;
- struct drm_i915_gem_object *obj = intel_fb_obj(fb);
enum pipe pipe = intel_plane->pipe;
u32 sprctl, sprscale = 0;
u32 sprsurf_offset, linear_offset;
unsigned int rotation = plane_state->base.rotation;
- int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
int crtc_x = plane_state->base.dst.x1;
int crtc_y = plane_state->base.dst.y1;
*/
sprctl |= SPRITE_GAMMA_ENABLE;
- if (i915_gem_object_is_tiled(obj))
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
sprctl |= SPRITE_TILED;
if (IS_HASWELL(dev) || IS_BROADWELL(dev))
if (crtc_w != src_w || crtc_h != src_h)
sprscale = SPRITE_SCALE_ENABLE | (src_w << 16) | src_h;
- linear_offset = y * fb->pitches[0] + x * cpp;
- sprsurf_offset = intel_compute_tile_offset(&x, &y, fb, 0,
- fb->pitches[0], rotation);
- linear_offset -= sprsurf_offset;
+ intel_add_fb_offsets(&x, &y, plane_state, 0);
+ sprsurf_offset = intel_compute_tile_offset(&x, &y, plane_state, 0);
if (rotation == DRM_ROTATE_180) {
sprctl |= SPRITE_ROTATE_180;
if (!IS_HASWELL(dev) && !IS_BROADWELL(dev)) {
x += src_w;
y += src_h;
- linear_offset += src_h * fb->pitches[0] + src_w * cpp;
}
}
+ linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
+
if (key->flags) {
I915_WRITE(SPRKEYVAL(pipe), key->min_value);
I915_WRITE(SPRKEYMAX(pipe), key->max_value);
* register */
if (IS_HASWELL(dev) || IS_BROADWELL(dev))
I915_WRITE(SPROFFSET(pipe), (y << 16) | x);
- else if (i915_gem_object_is_tiled(obj))
+ else if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
I915_WRITE(SPRTILEOFF(pipe), (y << 16) | x);
else
I915_WRITE(SPRLINOFF(pipe), linear_offset);
I915_WRITE(SPRSCALE(pipe), sprscale);
I915_WRITE(SPRCTL(pipe), sprctl);
I915_WRITE(SPRSURF(pipe),
- i915_gem_obj_ggtt_offset(obj) + sprsurf_offset);
+ intel_fb_gtt_offset(fb, rotation) + sprsurf_offset);
POSTING_READ(SPRSURF(pipe));
}
struct drm_i915_private *dev_priv = to_i915(dev);
struct intel_plane *intel_plane = to_intel_plane(plane);
struct drm_framebuffer *fb = plane_state->base.fb;
- struct drm_i915_gem_object *obj = intel_fb_obj(fb);
int pipe = intel_plane->pipe;
u32 dvscntr, dvsscale;
u32 dvssurf_offset, linear_offset;
unsigned int rotation = plane_state->base.rotation;
- int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
int crtc_x = plane_state->base.dst.x1;
int crtc_y = plane_state->base.dst.y1;
*/
dvscntr |= DVS_GAMMA_ENABLE;
- if (i915_gem_object_is_tiled(obj))
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
dvscntr |= DVS_TILED;
if (IS_GEN6(dev))
if (crtc_w != src_w || crtc_h != src_h)
dvsscale = DVS_SCALE_ENABLE | (src_w << 16) | src_h;
- linear_offset = y * fb->pitches[0] + x * cpp;
- dvssurf_offset = intel_compute_tile_offset(&x, &y, fb, 0,
- fb->pitches[0], rotation);
- linear_offset -= dvssurf_offset;
+ intel_add_fb_offsets(&x, &y, plane_state, 0);
+ dvssurf_offset = intel_compute_tile_offset(&x, &y, plane_state, 0);
if (rotation == DRM_ROTATE_180) {
dvscntr |= DVS_ROTATE_180;
x += src_w;
y += src_h;
- linear_offset += src_h * fb->pitches[0] + src_w * cpp;
}
+ linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
+
if (key->flags) {
I915_WRITE(DVSKEYVAL(pipe), key->min_value);
I915_WRITE(DVSKEYMAX(pipe), key->max_value);
I915_WRITE(DVSSTRIDE(pipe), fb->pitches[0]);
I915_WRITE(DVSPOS(pipe), (crtc_y << 16) | crtc_x);
- if (i915_gem_object_is_tiled(obj))
+ if (fb->modifier[0] == I915_FORMAT_MOD_X_TILED)
I915_WRITE(DVSTILEOFF(pipe), (y << 16) | x);
else
I915_WRITE(DVSLINOFF(pipe), linear_offset);
I915_WRITE(DVSSCALE(pipe), dvsscale);
I915_WRITE(DVSCNTR(pipe), dvscntr);
I915_WRITE(DVSSURF(pipe),
- i915_gem_obj_ggtt_offset(obj) + dvssurf_offset);
+ intel_fb_gtt_offset(fb, rotation) + dvssurf_offset);
POSTING_READ(DVSSURF(pipe));
}
int hscale, vscale;
int max_scale, min_scale;
bool can_scale;
+ int ret;
src->x1 = state->base.src_x;
src->y1 = state->base.src_y;
dst->y1 = crtc_y;
dst->y2 = crtc_y + crtc_h;
+ if (INTEL_GEN(dev) >= 9) {
+ ret = skl_check_plane_surface(state);
+ if (ret)
+ return ret;
+ }
+
return 0;
}
#define I845_TSEG_SIZE_1M (3 << 1)
#define INTEL_BSM 0x5c
-#define INTEL_BSM_MASK (0xFFFF << 20)
+#define INTEL_BSM_MASK (-(1u << 20))
#endif /* _I915_DRM_H_ */
projects / platform / kernel / linux-starfive.git / commitdiff