UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity

drm/i915: Make own struct for execlist items 

Engine's execlist related items have been increasing to
a point where a separate struct is warranted. Carve execlist
specific items to a dedicated struct to add clarity.

v2: add kerneldoc and fix whitespace (Joonas, Chris)
v3: csb_mmio changes, rebase
v4: s/\b(el|execlist)\b/execlists/ (Joonas)

Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> (v3)
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v3)
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-1-mika.kuoppala@intel.com
This commit is contained in:
Mika Kuoppala 2017-09-22 15:43:03 +03:00 committed by Mika Kuoppala
parent d27ffc1d00
commit b620e87021
8 changed files with 167 additions and 99 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
drivers/gpu/drm/i915/i915_debugfs.c
View File

@ -3323,7 +3323,7 @@ static int i915_engine_info(struct seq_file *m, void *unused)
read = GEN8_CSB_READ_PTR(ptr); read = GEN8_CSB_READ_PTR(ptr);
write = GEN8_CSB_WRITE_PTR(ptr); write = GEN8_CSB_WRITE_PTR(ptr);
seq_printf(m, "\tExeclist CSB read %d [%d cached], write %d [%d from hws], interrupt posted? %s\n", seq_printf(m, "\tExeclist CSB read %d [%d cached], write %d [%d from hws], interrupt posted? %s\n",
read, engine->csb_head, read, engine->execlists.csb_head,
write, write,
intel_read_status_page(engine, intel_hws_csb_write_index(engine->i915)), intel_read_status_page(engine, intel_hws_csb_write_index(engine->i915)),
yesno(test_bit(ENGINE_IRQ_EXECLIST, yesno(test_bit(ENGINE_IRQ_EXECLIST,
@ -3345,10 +3345,10 @@ static int i915_engine_info(struct seq_file *m, void *unused)
} }
rcu_read_lock(); rcu_read_lock();
for (idx = 0; idx < ARRAY_SIZE(engine->execlist_port); idx++) { for (idx = 0; idx < ARRAY_SIZE(engine->execlists.port); idx++) {
unsigned int count; unsigned int count;
rq = port_unpack(&engine->execlist_port[idx], rq = port_unpack(&engine->execlists.port[idx],
&count); &count);
if (rq) { if (rq) {
seq_printf(m, "\t\tELSP[%d] count=%d, ", seq_printf(m, "\t\tELSP[%d] count=%d, ",
@ -3362,7 +3362,7 @@ static int i915_engine_info(struct seq_file *m, void *unused)
rcu_read_unlock(); rcu_read_unlock();
spin_lock_irq(&engine->timeline->lock); spin_lock_irq(&engine->timeline->lock);
for (rb = engine->execlist_first; rb; rb = rb_next(rb)){ for (rb = engine->execlists.first; rb; rb = rb_next(rb)) {
struct i915_priolist *p = struct i915_priolist *p =
rb_entry(rb, typeof(*p), node); rb_entry(rb, typeof(*p), node);

6
drivers/gpu/drm/i915/i915_gem.c
View File

@ -2815,8 +2815,8 @@ i915_gem_reset_prepare_engine(struct intel_engine_cs *engine)
* Turning off the engine->irq_tasklet until the reset is over * Turning off the engine->irq_tasklet until the reset is over
* prevents the race. * prevents the race.
*/ */
tasklet_kill(&engine->irq_tasklet); tasklet_kill(&engine->execlists.irq_tasklet);
tasklet_disable(&engine->irq_tasklet); tasklet_disable(&engine->execlists.irq_tasklet);
if (engine->irq_seqno_barrier) if (engine->irq_seqno_barrier)
engine->irq_seqno_barrier(engine); engine->irq_seqno_barrier(engine);
@ -2995,7 +2995,7 @@ void i915_gem_reset(struct drm_i915_private *dev_priv)
void i915_gem_reset_finish_engine(struct intel_engine_cs *engine) void i915_gem_reset_finish_engine(struct intel_engine_cs *engine)
{ {
tasklet_enable(&engine->irq_tasklet); tasklet_enable(&engine->execlists.irq_tasklet);
kthread_unpark(engine->breadcrumbs.signaler); kthread_unpark(engine->breadcrumbs.signaler);
} }

4
drivers/gpu/drm/i915/i915_gpu_error.c
View File

@ -1327,10 +1327,10 @@ static void engine_record_requests(struct intel_engine_cs *engine,
static void error_record_engine_execlists(struct intel_engine_cs *engine, static void error_record_engine_execlists(struct intel_engine_cs *engine,
struct drm_i915_error_engine *ee) struct drm_i915_error_engine *ee)
{ {
const struct execlist_port *port = engine->execlist_port; const struct execlist_port *port = engine->execlists.port;
unsigned int n; unsigned int n;
for (n = 0; n < ARRAY_SIZE(engine->execlist_port); n++) { for (n = 0; n < ARRAY_SIZE(engine->execlists.port); n++) {
struct drm_i915_gem_request *rq = port_request(&port[n]); struct drm_i915_gem_request *rq = port_request(&port[n]);
if (!rq) if (!rq)

31
drivers/gpu/drm/i915/i915_guc_submission.c
View File

@ -494,11 +494,12 @@ static void i915_guc_submit(struct intel_engine_cs *engine)
struct drm_i915_private *dev_priv = engine->i915; struct drm_i915_private *dev_priv = engine->i915;
struct intel_guc *guc = &dev_priv->guc; struct intel_guc *guc = &dev_priv->guc;
struct i915_guc_client *client = guc->execbuf_client; struct i915_guc_client *client = guc->execbuf_client;
struct execlist_port *port = engine->execlist_port; struct intel_engine_execlists * const execlists = &engine->execlists;
unsigned int engine_id = engine->id; struct execlist_port *port = execlists->port;
const unsigned int engine_id = engine->id;
unsigned int n; unsigned int n;
for (n = 0; n < ARRAY_SIZE(engine->execlist_port); n++) { for (n = 0; n < ARRAY_SIZE(execlists->port); n++) {
struct drm_i915_gem_request *rq; struct drm_i915_gem_request *rq;
unsigned int count; unsigned int count;
@ -558,7 +559,8 @@ static void port_assign(struct execlist_port *port,
static void i915_guc_dequeue(struct intel_engine_cs *engine) static void i915_guc_dequeue(struct intel_engine_cs *engine)
{ {
struct execlist_port *port = engine->execlist_port; struct intel_engine_execlists * const execlists = &engine->execlists;
struct execlist_port *port = execlists->port;
struct drm_i915_gem_request *last = NULL; struct drm_i915_gem_request *last = NULL;
bool submit = false; bool submit = false;
struct rb_node *rb; struct rb_node *rb;
@ -567,15 +569,15 @@ static void i915_guc_dequeue(struct intel_engine_cs *engine)
port++; port++;
spin_lock_irq(&engine->timeline->lock); spin_lock_irq(&engine->timeline->lock);
rb = engine->execlist_first; rb = execlists->first;
GEM_BUG_ON(rb_first(&engine->execlist_queue) != rb); GEM_BUG_ON(rb_first(&execlists->queue) != rb);
while (rb) { while (rb) {
struct i915_priolist *p = rb_entry(rb, typeof(*p), node); struct i915_priolist *p = rb_entry(rb, typeof(*p), node);
struct drm_i915_gem_request *rq, *rn; struct drm_i915_gem_request *rq, *rn;
list_for_each_entry_safe(rq, rn, &p->requests, priotree.link) { list_for_each_entry_safe(rq, rn, &p->requests, priotree.link) {
if (last && rq->ctx != last->ctx) { if (last && rq->ctx != last->ctx) {
if (port != engine->execlist_port) { if (port != execlists->port) {
__list_del_many(&p->requests, __list_del_many(&p->requests,
&rq->priotree.link); &rq->priotree.link);
goto done; goto done;
@ -596,13 +598,13 @@ static void i915_guc_dequeue(struct intel_engine_cs *engine)
} }
rb = rb_next(rb); rb = rb_next(rb);
rb_erase(&p->node, &engine->execlist_queue); rb_erase(&p->node, &execlists->queue);
INIT_LIST_HEAD(&p->requests); INIT_LIST_HEAD(&p->requests);
if (p->priority != I915_PRIORITY_NORMAL) if (p->priority != I915_PRIORITY_NORMAL)
kmem_cache_free(engine->i915->priorities, p); kmem_cache_free(engine->i915->priorities, p);
} }
done: done:
engine->execlist_first = rb; execlists->first = rb;
if (submit) { if (submit) {
port_assign(port, last); port_assign(port, last);
i915_guc_submit(engine); i915_guc_submit(engine);
@ -612,8 +614,8 @@ done:
static void i915_guc_irq_handler(unsigned long data) static void i915_guc_irq_handler(unsigned long data)
{ {
struct intel_engine_cs *engine = (struct intel_engine_cs *)data; struct intel_engine_cs * const engine = (struct intel_engine_cs *)data;
struct execlist_port *port = engine->execlist_port; struct execlist_port *port = engine->execlists.port;
struct drm_i915_gem_request *rq; struct drm_i915_gem_request *rq;
rq = port_request(&port[0]); rq = port_request(&port[0]);
@ -1144,7 +1146,7 @@ int i915_guc_submission_enable(struct drm_i915_private *dev_priv)
* and it is guaranteed that it will remove the work item from the * and it is guaranteed that it will remove the work item from the
* queue before our request is completed. * queue before our request is completed.
*/ */
BUILD_BUG_ON(ARRAY_SIZE(engine->execlist_port) * BUILD_BUG_ON(ARRAY_SIZE(engine->execlists.port) *
sizeof(struct guc_wq_item) * sizeof(struct guc_wq_item) *
I915_NUM_ENGINES > GUC_WQ_SIZE); I915_NUM_ENGINES > GUC_WQ_SIZE);
@ -1175,14 +1177,15 @@ int i915_guc_submission_enable(struct drm_i915_private *dev_priv)
guc_interrupts_capture(dev_priv); guc_interrupts_capture(dev_priv);
for_each_engine(engine, dev_priv, id) { for_each_engine(engine, dev_priv, id) {
struct intel_engine_execlists * const execlists = &engine->execlists;
/* The tasklet was initialised by execlists, and may be in /* The tasklet was initialised by execlists, and may be in
* a state of flux (across a reset) and so we just want to * a state of flux (across a reset) and so we just want to
* take over the callback without changing any other state * take over the callback without changing any other state
* in the tasklet. * in the tasklet.
*/ */
engine->irq_tasklet.func = i915_guc_irq_handler; execlists->irq_tasklet.func = i915_guc_irq_handler;
clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted);
tasklet_schedule(&engine->irq_tasklet); tasklet_schedule(&execlists->irq_tasklet);
} }
return 0; return 0;

5
drivers/gpu/drm/i915/i915_irq.c
View File

@ -1346,10 +1346,11 @@ static void snb_gt_irq_handler(struct drm_i915_private *dev_priv,
static void static void
gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir, int test_shift) gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir, int test_shift)
{ {
struct intel_engine_execlists * const execlists = &engine->execlists;
bool tasklet = false; bool tasklet = false;
if (iir & (GT_CONTEXT_SWITCH_INTERRUPT << test_shift)) { if (iir & (GT_CONTEXT_SWITCH_INTERRUPT << test_shift)) {
if (port_count(&engine->execlist_port[0])) { if (port_count(&execlists->port[0])) {
__set_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); __set_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted);
tasklet = true; tasklet = true;
} }
@ -1361,7 +1362,7 @@ gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir, int test_shift)
} }
if (tasklet) if (tasklet)
tasklet_hi_schedule(&engine->irq_tasklet); tasklet_hi_schedule(&execlists->irq_tasklet);
} }
static irqreturn_t gen8_gt_irq_ack(struct drm_i915_private *dev_priv, static irqreturn_t gen8_gt_irq_ack(struct drm_i915_private *dev_priv,

12
drivers/gpu/drm/i915/intel_engine_cs.c
View File

@ -393,8 +393,8 @@ static void intel_engine_init_timeline(struct intel_engine_cs *engine)
*/ */
void intel_engine_setup_common(struct intel_engine_cs *engine) void intel_engine_setup_common(struct intel_engine_cs *engine)
{ {
engine->execlist_queue = RB_ROOT; engine->execlists.queue = RB_ROOT;
engine->execlist_first = NULL; engine->execlists.first = NULL;
intel_engine_init_timeline(engine); intel_engine_init_timeline(engine);
intel_engine_init_hangcheck(engine); intel_engine_init_hangcheck(engine);
@ -1475,11 +1475,11 @@ bool intel_engine_is_idle(struct intel_engine_cs *engine)
return false; return false;
/* Both ports drained, no more ELSP submission? */ /* Both ports drained, no more ELSP submission? */
if (port_request(&engine->execlist_port[0])) if (port_request(&engine->execlists.port[0]))
return false; return false;
/* ELSP is empty, but there are ready requests? */ /* ELSP is empty, but there are ready requests? */
if (READ_ONCE(engine->execlist_first)) if (READ_ONCE(engine->execlists.first))
return false; return false;
/* Ring stopped? */ /* Ring stopped? */
@ -1528,8 +1528,8 @@ void intel_engines_mark_idle(struct drm_i915_private *i915)
for_each_engine(engine, i915, id) { for_each_engine(engine, i915, id) {
intel_engine_disarm_breadcrumbs(engine); intel_engine_disarm_breadcrumbs(engine);
i915_gem_batch_pool_fini(&engine->batch_pool); i915_gem_batch_pool_fini(&engine->batch_pool);
tasklet_kill(&engine->irq_tasklet); tasklet_kill(&engine->execlists.irq_tasklet);
engine->no_priolist = false; engine->execlists.no_priolist = false;
} }
} }

100
drivers/gpu/drm/i915/intel_lrc.c
View File

@ -291,17 +291,18 @@ lookup_priolist(struct intel_engine_cs *engine,
struct i915_priotree *pt, struct i915_priotree *pt,
int prio) int prio)
{ {
struct intel_engine_execlists * const execlists = &engine->execlists;
struct i915_priolist *p; struct i915_priolist *p;
struct rb_node **parent, *rb; struct rb_node **parent, *rb;
bool first = true; bool first = true;
if (unlikely(engine->no_priolist)) if (unlikely(execlists->no_priolist))
prio = I915_PRIORITY_NORMAL; prio = I915_PRIORITY_NORMAL;
find_priolist: find_priolist:
/* most positive priority is scheduled first, equal priorities fifo */ /* most positive priority is scheduled first, equal priorities fifo */
rb = NULL; rb = NULL;
parent = &engine->execlist_queue.rb_node; parent = &execlists->queue.rb_node;
while (*parent) { while (*parent) {
rb = *parent; rb = *parent;
p = rb_entry(rb, typeof(*p), node); p = rb_entry(rb, typeof(*p), node);
@ -316,7 +317,7 @@ find_priolist:
} }
if (prio == I915_PRIORITY_NORMAL) { if (prio == I915_PRIORITY_NORMAL) {
p = &engine->default_priolist; p = &execlists->default_priolist;
} else { } else {
p = kmem_cache_alloc(engine->i915->priorities, GFP_ATOMIC); p = kmem_cache_alloc(engine->i915->priorities, GFP_ATOMIC);
/* Convert an allocation failure to a priority bump */ /* Convert an allocation failure to a priority bump */
@ -331,7 +332,7 @@ find_priolist:
* requests, so if userspace lied about their * requests, so if userspace lied about their
* dependencies that reordering may be visible. * dependencies that reordering may be visible.
*/ */
engine->no_priolist = true; execlists->no_priolist = true;
goto find_priolist; goto find_priolist;
} }
} }
@ -339,10 +340,10 @@ find_priolist:
p->priority = prio; p->priority = prio;
INIT_LIST_HEAD(&p->requests); INIT_LIST_HEAD(&p->requests);
rb_link_node(&p->node, rb, parent); rb_link_node(&p->node, rb, parent);
rb_insert_color(&p->node, &engine->execlist_queue); rb_insert_color(&p->node, &execlists->queue);
if (first) if (first)
engine->execlist_first = &p->node; execlists->first = &p->node;
return ptr_pack_bits(p, first, 1); return ptr_pack_bits(p, first, 1);
} }
@ -393,12 +394,12 @@ static u64 execlists_update_context(struct drm_i915_gem_request *rq)
static void execlists_submit_ports(struct intel_engine_cs *engine) static void execlists_submit_ports(struct intel_engine_cs *engine)
{ {
struct execlist_port *port = engine->execlist_port; struct execlist_port *port = engine->execlists.port;
u32 __iomem *elsp = u32 __iomem *elsp =
engine->i915->regs + i915_mmio_reg_offset(RING_ELSP(engine)); engine->i915->regs + i915_mmio_reg_offset(RING_ELSP(engine));
unsigned int n; unsigned int n;
for (n = ARRAY_SIZE(engine->execlist_port); n--; ) { for (n = ARRAY_SIZE(engine->execlists.port); n--; ) {
struct drm_i915_gem_request *rq; struct drm_i915_gem_request *rq;
unsigned int count; unsigned int count;
u64 desc; u64 desc;
@ -453,7 +454,7 @@ static void port_assign(struct execlist_port *port,
static void execlists_dequeue(struct intel_engine_cs *engine) static void execlists_dequeue(struct intel_engine_cs *engine)
{ {
struct drm_i915_gem_request *last; struct drm_i915_gem_request *last;
struct execlist_port *port = engine->execlist_port; struct execlist_port *port = engine->execlists.port;
struct rb_node *rb; struct rb_node *rb;
bool submit = false; bool submit = false;
@ -491,8 +492,8 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
*/ */
spin_lock_irq(&engine->timeline->lock); spin_lock_irq(&engine->timeline->lock);
rb = engine->execlist_first; rb = engine->execlists.first;
GEM_BUG_ON(rb_first(&engine->execlist_queue) != rb); GEM_BUG_ON(rb_first(&engine->execlists.queue) != rb);
while (rb) { while (rb) {
struct i915_priolist *p = rb_entry(rb, typeof(*p), node); struct i915_priolist *p = rb_entry(rb, typeof(*p), node);
struct drm_i915_gem_request *rq, *rn; struct drm_i915_gem_request *rq, *rn;
@ -515,7 +516,7 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
* combine this request with the last, then we * combine this request with the last, then we
* are done. * are done.
*/ */
if (port != engine->execlist_port) { if (port != engine->execlists.port) {
__list_del_many(&p->requests, __list_del_many(&p->requests,
&rq->priotree.link); &rq->priotree.link);
goto done; goto done;
@ -552,13 +553,13 @@ static void execlists_dequeue(struct intel_engine_cs *engine)
} }
rb = rb_next(rb); rb = rb_next(rb);
rb_erase(&p->node, &engine->execlist_queue); rb_erase(&p->node, &engine->execlists.queue);
INIT_LIST_HEAD(&p->requests); INIT_LIST_HEAD(&p->requests);
if (p->priority != I915_PRIORITY_NORMAL) if (p->priority != I915_PRIORITY_NORMAL)
kmem_cache_free(engine->i915->priorities, p); kmem_cache_free(engine->i915->priorities, p);
} }
done: done:
engine->execlist_first = rb; engine->execlists.first = rb;
if (submit) if (submit)
port_assign(port, last); port_assign(port, last);
spin_unlock_irq(&engine->timeline->lock); spin_unlock_irq(&engine->timeline->lock);
@ -569,7 +570,8 @@ done:
static void execlists_cancel_requests(struct intel_engine_cs *engine) static void execlists_cancel_requests(struct intel_engine_cs *engine)
{ {
struct execlist_port *port = engine->execlist_port; struct intel_engine_execlists * const execlists = &engine->execlists;
struct execlist_port *port = execlists->port;
struct drm_i915_gem_request *rq, *rn; struct drm_i915_gem_request *rq, *rn;
struct rb_node *rb; struct rb_node *rb;
unsigned long flags; unsigned long flags;
@ -578,9 +580,9 @@ static void execlists_cancel_requests(struct intel_engine_cs *engine)
spin_lock_irqsave(&engine->timeline->lock, flags); spin_lock_irqsave(&engine->timeline->lock, flags);
/* Cancel the requests on the HW and clear the ELSP tracker. */ /* Cancel the requests on the HW and clear the ELSP tracker. */
for (n = 0; n < ARRAY_SIZE(engine->execlist_port); n++) for (n = 0; n < ARRAY_SIZE(execlists->port); n++)
i915_gem_request_put(port_request(&port[n])); i915_gem_request_put(port_request(&port[n]));
memset(engine->execlist_port, 0, sizeof(engine->execlist_port)); memset(execlists->port, 0, sizeof(execlists->port));
/* Mark all executing requests as skipped. */ /* Mark all executing requests as skipped. */
list_for_each_entry(rq, &engine->timeline->requests, link) { list_for_each_entry(rq, &engine->timeline->requests, link) {
@ -590,7 +592,7 @@ static void execlists_cancel_requests(struct intel_engine_cs *engine)
} }
/* Flush the queued requests to the timeline list (for retiring). */ /* Flush the queued requests to the timeline list (for retiring). */
rb = engine->execlist_first; rb = execlists->first;
while (rb) { while (rb) {
struct i915_priolist *p = rb_entry(rb, typeof(*p), node); struct i915_priolist *p = rb_entry(rb, typeof(*p), node);
@ -603,7 +605,7 @@ static void execlists_cancel_requests(struct intel_engine_cs *engine)
} }
rb = rb_next(rb); rb = rb_next(rb);
rb_erase(&p->node, &engine->execlist_queue); rb_erase(&p->node, &execlists->queue);
INIT_LIST_HEAD(&p->requests); INIT_LIST_HEAD(&p->requests);
if (p->priority != I915_PRIORITY_NORMAL) if (p->priority != I915_PRIORITY_NORMAL)
kmem_cache_free(engine->i915->priorities, p); kmem_cache_free(engine->i915->priorities, p);
@ -611,8 +613,8 @@ static void execlists_cancel_requests(struct intel_engine_cs *engine)
/* Remaining _unready_ requests will be nop'ed when submitted */ /* Remaining _unready_ requests will be nop'ed when submitted */
engine->execlist_queue = RB_ROOT; execlists->queue = RB_ROOT;
engine->execlist_first = NULL; execlists->first = NULL;
GEM_BUG_ON(port_isset(&port[0])); GEM_BUG_ON(port_isset(&port[0]));
/* /*
@ -628,7 +630,7 @@ static void execlists_cancel_requests(struct intel_engine_cs *engine)
static bool execlists_elsp_ready(const struct intel_engine_cs *engine) static bool execlists_elsp_ready(const struct intel_engine_cs *engine)
{ {
const struct execlist_port *port = engine->execlist_port; const struct execlist_port *port = engine->execlists.port;
return port_count(&port[0]) + port_count(&port[1]) < 2; return port_count(&port[0]) + port_count(&port[1]) < 2;
} }
@ -639,8 +641,9 @@ static bool execlists_elsp_ready(const struct intel_engine_cs *engine)
*/ */
static void intel_lrc_irq_handler(unsigned long data) static void intel_lrc_irq_handler(unsigned long data)
{ {
struct intel_engine_cs *engine = (struct intel_engine_cs *)data; struct intel_engine_cs * const engine = (struct intel_engine_cs *)data;
struct execlist_port *port = engine->execlist_port; struct intel_engine_execlists * const execlists = &engine->execlists;
struct execlist_port *port = execlists->port;
struct drm_i915_private *dev_priv = engine->i915; struct drm_i915_private *dev_priv = engine->i915;
/* We can skip acquiring intel_runtime_pm_get() here as it was taken /* We can skip acquiring intel_runtime_pm_get() here as it was taken
@ -652,7 +655,7 @@ static void intel_lrc_irq_handler(unsigned long data)
*/ */
GEM_BUG_ON(!dev_priv->gt.awake); GEM_BUG_ON(!dev_priv->gt.awake);
intel_uncore_forcewake_get(dev_priv, engine->fw_domains); intel_uncore_forcewake_get(dev_priv, execlists->fw_domains);
/* Prefer doing test_and_clear_bit() as a two stage operation to avoid /* Prefer doing test_and_clear_bit() as a two stage operation to avoid
* imposing the cost of a locked atomic transaction when submitting a * imposing the cost of a locked atomic transaction when submitting a
@ -665,10 +668,10 @@ static void intel_lrc_irq_handler(unsigned long data)
unsigned int head, tail; unsigned int head, tail;
/* However GVT emulation depends upon intercepting CSB mmio */ /* However GVT emulation depends upon intercepting CSB mmio */
if (unlikely(engine->csb_use_mmio)) { if (unlikely(execlists->csb_use_mmio)) {
buf = (u32 * __force) buf = (u32 * __force)
(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_BUF_LO(engine, 0))); (dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_BUF_LO(engine, 0)));
engine->csb_head = -1; /* force mmio read of CSB ptrs */ execlists->csb_head = -1; /* force mmio read of CSB ptrs */
} }
/* The write will be ordered by the uncached read (itself /* The write will be ordered by the uncached read (itself
@ -682,19 +685,20 @@ static void intel_lrc_irq_handler(unsigned long data)
* is set and we do a new loop. * is set and we do a new loop.
*/ */
__clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); __clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted);
if (unlikely(engine->csb_head == -1)) { /* following a reset */ if (unlikely(execlists->csb_head == -1)) { /* following a reset */
head = readl(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine))); head = readl(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine)));
tail = GEN8_CSB_WRITE_PTR(head); tail = GEN8_CSB_WRITE_PTR(head);
head = GEN8_CSB_READ_PTR(head); head = GEN8_CSB_READ_PTR(head);
engine->csb_head = head; execlists->csb_head = head;
} else { } else {
const int write_idx = const int write_idx =
intel_hws_csb_write_index(dev_priv) - intel_hws_csb_write_index(dev_priv) -
I915_HWS_CSB_BUF0_INDEX; I915_HWS_CSB_BUF0_INDEX;
head = engine->csb_head; head = execlists->csb_head;
tail = READ_ONCE(buf[write_idx]); tail = READ_ONCE(buf[write_idx]);
} }
while (head != tail) { while (head != tail) {
struct drm_i915_gem_request *rq; struct drm_i915_gem_request *rq;
unsigned int status; unsigned int status;
@ -748,8 +752,8 @@ static void intel_lrc_irq_handler(unsigned long data)
!(status & GEN8_CTX_STATUS_ACTIVE_IDLE)); !(status & GEN8_CTX_STATUS_ACTIVE_IDLE));
} }
if (head != engine->csb_head) { if (head != execlists->csb_head) {
engine->csb_head = head; execlists->csb_head = head;
writel(_MASKED_FIELD(GEN8_CSB_READ_PTR_MASK, head << 8), writel(_MASKED_FIELD(GEN8_CSB_READ_PTR_MASK, head << 8),
dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine))); dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine)));
} }
@ -758,7 +762,7 @@ static void intel_lrc_irq_handler(unsigned long data)
if (execlists_elsp_ready(engine)) if (execlists_elsp_ready(engine))
execlists_dequeue(engine); execlists_dequeue(engine);
intel_uncore_forcewake_put(dev_priv, engine->fw_domains); intel_uncore_forcewake_put(dev_priv, execlists->fw_domains);
} }
static void insert_request(struct intel_engine_cs *engine, static void insert_request(struct intel_engine_cs *engine,
@ -769,7 +773,7 @@ static void insert_request(struct intel_engine_cs *engine,
list_add_tail(&pt->link, &ptr_mask_bits(p, 1)->requests); list_add_tail(&pt->link, &ptr_mask_bits(p, 1)->requests);
if (ptr_unmask_bits(p, 1) && execlists_elsp_ready(engine)) if (ptr_unmask_bits(p, 1) && execlists_elsp_ready(engine))
tasklet_hi_schedule(&engine->irq_tasklet); tasklet_hi_schedule(&engine->execlists.irq_tasklet);
} }
static void execlists_submit_request(struct drm_i915_gem_request *request) static void execlists_submit_request(struct drm_i915_gem_request *request)
@ -782,7 +786,7 @@ static void execlists_submit_request(struct drm_i915_gem_request *request)
insert_request(engine, &request->priotree, request->priotree.priority); insert_request(engine, &request->priotree, request->priotree.priority);
GEM_BUG_ON(!engine->execlist_first); GEM_BUG_ON(!engine->execlists.first);
GEM_BUG_ON(list_empty(&request->priotree.link)); GEM_BUG_ON(list_empty(&request->priotree.link));
spin_unlock_irqrestore(&engine->timeline->lock, flags); spin_unlock_irqrestore(&engine->timeline->lock, flags);
@ -1289,6 +1293,7 @@ static u8 gtiir[] = {
static int gen8_init_common_ring(struct intel_engine_cs *engine) static int gen8_init_common_ring(struct intel_engine_cs *engine)
{ {
struct drm_i915_private *dev_priv = engine->i915; struct drm_i915_private *dev_priv = engine->i915;
struct intel_engine_execlists * const execlists = &engine->execlists;
int ret; int ret;
ret = intel_mocs_init_engine(engine); ret = intel_mocs_init_engine(engine);
@ -1321,11 +1326,11 @@ static int gen8_init_common_ring(struct intel_engine_cs *engine)
I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]), I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]),
GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift); GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift);
clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted);
engine->csb_head = -1; execlists->csb_head = -1;
/* After a GPU reset, we may have requests to replay */ /* After a GPU reset, we may have requests to replay */
if (!i915_modparams.enable_guc_submission && engine->execlist_first) if (!i915_modparams.enable_guc_submission && execlists->first)
tasklet_schedule(&engine->irq_tasklet); tasklet_schedule(&execlists->irq_tasklet);
return 0; return 0;
} }
@ -1366,7 +1371,8 @@ static int gen9_init_render_ring(struct intel_engine_cs *engine)
static void reset_common_ring(struct intel_engine_cs *engine, static void reset_common_ring(struct intel_engine_cs *engine,
struct drm_i915_gem_request *request) struct drm_i915_gem_request *request)
{ {
struct execlist_port *port = engine->execlist_port; struct intel_engine_execlists * const execlists = &engine->execlists;
struct execlist_port *port = execlists->port;
struct drm_i915_gem_request *rq, *rn; struct drm_i915_gem_request *rq, *rn;
struct intel_context *ce; struct intel_context *ce;
unsigned long flags; unsigned long flags;
@ -1383,9 +1389,9 @@ static void reset_common_ring(struct intel_engine_cs *engine,
* guessing the missed context-switch events by looking at what * guessing the missed context-switch events by looking at what
* requests were completed. * requests were completed.
*/ */
for (n = 0; n < ARRAY_SIZE(engine->execlist_port); n++) for (n = 0; n < ARRAY_SIZE(execlists->port); n++)
i915_gem_request_put(port_request(&port[n])); i915_gem_request_put(port_request(&port[n]));
memset(engine->execlist_port, 0, sizeof(engine->execlist_port)); memset(execlists->port, 0, sizeof(execlists->port));
/* Push back any incomplete requests for replay after the reset. */ /* Push back any incomplete requests for replay after the reset. */
list_for_each_entry_safe_reverse(rq, rn, list_for_each_entry_safe_reverse(rq, rn,
@ -1719,8 +1725,8 @@ void intel_logical_ring_cleanup(struct intel_engine_cs *engine)
* Tasklet cannot be active at this point due intel_mark_active/idle * Tasklet cannot be active at this point due intel_mark_active/idle
* so this is just for documentation. * so this is just for documentation.
*/ */
if (WARN_ON(test_bit(TASKLET_STATE_SCHED, &engine->irq_tasklet.state))) if (WARN_ON(test_bit(TASKLET_STATE_SCHED, &engine->execlists.irq_tasklet.state)))
tasklet_kill(&engine->irq_tasklet); tasklet_kill(&engine->execlists.irq_tasklet);
dev_priv = engine->i915; dev_priv = engine->i915;
@ -1744,7 +1750,7 @@ static void execlists_set_default_submission(struct intel_engine_cs *engine)
engine->submit_request = execlists_submit_request; engine->submit_request = execlists_submit_request;
engine->cancel_requests = execlists_cancel_requests; engine->cancel_requests = execlists_cancel_requests;
engine->schedule = execlists_schedule; engine->schedule = execlists_schedule;
engine->irq_tasklet.func = intel_lrc_irq_handler; engine->execlists.irq_tasklet.func = intel_lrc_irq_handler;
} }
static void static void
@ -1806,7 +1812,7 @@ logical_ring_setup(struct intel_engine_cs *engine)
/* Intentionally left blank. */ /* Intentionally left blank. */
engine->buffer = NULL; engine->buffer = NULL;
engine->csb_use_mmio = irq_handler_force_mmio(dev_priv); engine->execlists.csb_use_mmio = irq_handler_force_mmio(dev_priv);
fw_domains = intel_uncore_forcewake_for_reg(dev_priv, fw_domains = intel_uncore_forcewake_for_reg(dev_priv,
RING_ELSP(engine), RING_ELSP(engine),
@ -1820,9 +1826,9 @@ logical_ring_setup(struct intel_engine_cs *engine)
RING_CONTEXT_STATUS_BUF_BASE(engine), RING_CONTEXT_STATUS_BUF_BASE(engine),
FW_REG_READ); FW_REG_READ);
engine->fw_domains = fw_domains; engine->execlists.fw_domains = fw_domains;
tasklet_init(&engine->irq_tasklet, tasklet_init(&engine->execlists.irq_tasklet,
intel_lrc_irq_handler, (unsigned long)engine); intel_lrc_irq_handler, (unsigned long)engine);
logical_ring_default_vfuncs(engine); logical_ring_default_vfuncs(engine);

100
drivers/gpu/drm/i915/intel_ringbuffer.h
View File

@ -184,6 +184,84 @@ struct i915_priolist {
int priority; int priority;
}; };
/**
* struct intel_engine_execlists - execlist submission queue and port state
*
* The struct intel_engine_execlists represents the combined logical state of
* driver and the hardware state for execlist mode of submission.
*/
struct intel_engine_execlists {
/**
* @irq_tasklet: softirq tasklet for bottom handler
*/
struct tasklet_struct irq_tasklet;
/**
* @default_priolist: priority list for I915_PRIORITY_NORMAL
*/
struct i915_priolist default_priolist;
/**
* @no_priolist: priority lists disabled
*/
bool no_priolist;
/**
* @port: execlist port states
*
* For each hardware ELSP (ExecList Submission Port) we keep
* track of the last request and the number of times we submitted
* that port to hw. We then count the number of times the hw reports
* a context completion or preemption. As only one context can
* be active on hw, we limit resubmission of context to port[0]. This
* is called Lite Restore, of the context.
*/
struct execlist_port {
/**
* @request_count: combined request and submission count
*/
struct drm_i915_gem_request *request_count;
#define EXECLIST_COUNT_BITS 2
#define port_request(p) ptr_mask_bits((p)->request_count, EXECLIST_COUNT_BITS)
#define port_count(p) ptr_unmask_bits((p)->request_count, EXECLIST_COUNT_BITS)
#define port_pack(rq, count) ptr_pack_bits(rq, count, EXECLIST_COUNT_BITS)
#define port_unpack(p, count) ptr_unpack_bits((p)->request_count, count, EXECLIST_COUNT_BITS)
#define port_set(p, packed) ((p)->request_count = (packed))
#define port_isset(p) ((p)->request_count)
#define port_index(p, e) ((p) - (e)->execlists.port)
/**
* @context_id: context ID for port
*/
GEM_DEBUG_DECL(u32 context_id);
} port[2];
/**
* @queue: queue of requests, in priority lists
*/
struct rb_root queue;
/**
* @first: leftmost level in priority @queue
*/
struct rb_node *first;
/**
* @fw_domains: forcewake domains for irq tasklet
*/
unsigned int fw_domains;
/**
* @csb_head: context status buffer head
*/
unsigned int csb_head;
/**
* @csb_use_mmio: access csb through mmio, instead of hwsp
*/
bool csb_use_mmio;
};
#define INTEL_ENGINE_CS_MAX_NAME 8 #define INTEL_ENGINE_CS_MAX_NAME 8
struct intel_engine_cs { struct intel_engine_cs {
@ -380,27 +458,7 @@ struct intel_engine_cs {
u32 *(*signal)(struct drm_i915_gem_request *req, u32 *cs); u32 *(*signal)(struct drm_i915_gem_request *req, u32 *cs);
} semaphore; } semaphore;
/* Execlists */ struct intel_engine_execlists execlists;
struct tasklet_struct irq_tasklet;
struct i915_priolist default_priolist;
bool no_priolist;
struct execlist_port {
struct drm_i915_gem_request *request_count;
#define EXECLIST_COUNT_BITS 2
#define port_request(p) ptr_mask_bits((p)->request_count, EXECLIST_COUNT_BITS)
#define port_count(p) ptr_unmask_bits((p)->request_count, EXECLIST_COUNT_BITS)
#define port_pack(rq, count) ptr_pack_bits(rq, count, EXECLIST_COUNT_BITS)
#define port_unpack(p, count) ptr_unpack_bits((p)->request_count, count, EXECLIST_COUNT_BITS)
#define port_set(p, packed) ((p)->request_count = (packed))
#define port_isset(p) ((p)->request_count)
#define port_index(p, e) ((p) - (e)->execlist_port)
GEM_DEBUG_DECL(u32 context_id);
} execlist_port[2];
struct rb_root execlist_queue;
struct rb_node *execlist_first;
unsigned int fw_domains;
unsigned int csb_head;
bool csb_use_mmio;
/* Contexts are pinned whilst they are active on the GPU. The last /* Contexts are pinned whilst they are active on the GPU. The last
* context executed remains active whilst the GPU is idle - the * context executed remains active whilst the GPU is idle - the