Commit fca6af6a authored by Joerg Roedel's avatar Joerg Roedel

iommu/amd: Add per-domain timer to flush per-cpu queues

Add a timer to each dma_ops domain so that we flush unused
IOTLB entries regularily, even if the queues don't get full
all the time.
Signed-off-by: default avatarJoerg Roedel <jroedel@suse.de>
parent a6e3f6f0
...@@ -182,6 +182,13 @@ struct dma_ops_domain { ...@@ -182,6 +182,13 @@ struct dma_ops_domain {
* flushed out of the IOTLB of the domain. * flushed out of the IOTLB of the domain.
*/ */
atomic64_t flush_finish_cnt; atomic64_t flush_finish_cnt;
/*
* Timer to make sure we don't keep IOVAs around unflushed
* for too long
*/
struct timer_list flush_timer;
atomic_t flush_timer_on;
}; };
static struct iova_domain reserved_iova_ranges; static struct iova_domain reserved_iova_ranges;
...@@ -1834,6 +1841,14 @@ static int dma_ops_domain_alloc_flush_queue(struct dma_ops_domain *dom) ...@@ -1834,6 +1841,14 @@ static int dma_ops_domain_alloc_flush_queue(struct dma_ops_domain *dom)
return 0; return 0;
} }
static void dma_ops_domain_flush_tlb(struct dma_ops_domain *dom)
{
atomic64_inc(&dom->flush_start_cnt);
domain_flush_tlb(&dom->domain);
domain_flush_complete(&dom->domain);
atomic64_inc(&dom->flush_finish_cnt);
}
static inline bool queue_ring_full(struct flush_queue *queue) static inline bool queue_ring_full(struct flush_queue *queue)
{ {
assert_spin_locked(&queue->lock); assert_spin_locked(&queue->lock);
...@@ -1875,22 +1890,12 @@ static inline void queue_ring_remove_head(struct flush_queue *queue) ...@@ -1875,22 +1890,12 @@ static inline void queue_ring_remove_head(struct flush_queue *queue)
queue->head = (queue->head + 1) % FLUSH_QUEUE_SIZE; queue->head = (queue->head + 1) % FLUSH_QUEUE_SIZE;
} }
static void queue_add(struct dma_ops_domain *dom, static void queue_ring_free_flushed(struct dma_ops_domain *dom,
unsigned long address, unsigned long pages) struct flush_queue *queue)
{ {
struct flush_queue *queue; u64 counter = atomic64_read(&dom->flush_finish_cnt);
unsigned long flags;
u64 counter;
int idx; int idx;
pages = __roundup_pow_of_two(pages);
address >>= PAGE_SHIFT;
counter = atomic64_read(&dom->flush_finish_cnt);
queue = get_cpu_ptr(dom->flush_queue);
spin_lock_irqsave(&queue->lock, flags);
queue_ring_for_each(idx, queue) { queue_ring_for_each(idx, queue) {
/* /*
* This assumes that counter values in the ring-buffer are * This assumes that counter values in the ring-buffer are
...@@ -1905,12 +1910,25 @@ static void queue_add(struct dma_ops_domain *dom, ...@@ -1905,12 +1910,25 @@ static void queue_add(struct dma_ops_domain *dom,
queue_ring_remove_head(queue); queue_ring_remove_head(queue);
} }
}
static void queue_add(struct dma_ops_domain *dom,
unsigned long address, unsigned long pages)
{
struct flush_queue *queue;
unsigned long flags;
int idx;
pages = __roundup_pow_of_two(pages);
address >>= PAGE_SHIFT;
queue = get_cpu_ptr(dom->flush_queue);
spin_lock_irqsave(&queue->lock, flags);
queue_ring_free_flushed(dom, queue);
if (queue_ring_full(queue)) { if (queue_ring_full(queue)) {
atomic64_inc(&dom->flush_start_cnt); dma_ops_domain_flush_tlb(dom);
domain_flush_tlb(&dom->domain);
domain_flush_complete(&dom->domain);
atomic64_inc(&dom->flush_finish_cnt);
queue_release(dom, queue); queue_release(dom, queue);
} }
...@@ -1921,9 +1939,33 @@ static void queue_add(struct dma_ops_domain *dom, ...@@ -1921,9 +1939,33 @@ static void queue_add(struct dma_ops_domain *dom,
queue->entries[idx].counter = atomic64_read(&dom->flush_start_cnt); queue->entries[idx].counter = atomic64_read(&dom->flush_start_cnt);
spin_unlock_irqrestore(&queue->lock, flags); spin_unlock_irqrestore(&queue->lock, flags);
if (atomic_cmpxchg(&dom->flush_timer_on, 0, 1) == 0)
mod_timer(&dom->flush_timer, jiffies + msecs_to_jiffies(10));
put_cpu_ptr(dom->flush_queue); put_cpu_ptr(dom->flush_queue);
} }
static void queue_flush_timeout(unsigned long data)
{
struct dma_ops_domain *dom = (struct dma_ops_domain *)data;
int cpu;
atomic_set(&dom->flush_timer_on, 0);
dma_ops_domain_flush_tlb(dom);
for_each_possible_cpu(cpu) {
struct flush_queue *queue;
unsigned long flags;
queue = per_cpu_ptr(dom->flush_queue, cpu);
spin_lock_irqsave(&queue->lock, flags);
queue_ring_free_flushed(dom, queue);
spin_unlock_irqrestore(&queue->lock, flags);
}
}
/* /*
* Free a domain, only used if something went wrong in the * Free a domain, only used if something went wrong in the
* allocation path and we need to free an already allocated page table * allocation path and we need to free an already allocated page table
...@@ -1935,6 +1977,9 @@ static void dma_ops_domain_free(struct dma_ops_domain *dom) ...@@ -1935,6 +1977,9 @@ static void dma_ops_domain_free(struct dma_ops_domain *dom)
del_domain_from_list(&dom->domain); del_domain_from_list(&dom->domain);
if (timer_pending(&dom->flush_timer))
del_timer(&dom->flush_timer);
dma_ops_domain_free_flush_queue(dom); dma_ops_domain_free_flush_queue(dom);
put_iova_domain(&dom->iovad); put_iova_domain(&dom->iovad);
...@@ -1978,6 +2023,11 @@ static struct dma_ops_domain *dma_ops_domain_alloc(void) ...@@ -1978,6 +2023,11 @@ static struct dma_ops_domain *dma_ops_domain_alloc(void)
if (dma_ops_domain_alloc_flush_queue(dma_dom)) if (dma_ops_domain_alloc_flush_queue(dma_dom))
goto free_dma_dom; goto free_dma_dom;
setup_timer(&dma_dom->flush_timer, queue_flush_timeout,
(unsigned long)dma_dom);
atomic_set(&dma_dom->flush_timer_on, 0);
add_domain_to_list(&dma_dom->domain); add_domain_to_list(&dma_dom->domain);
return dma_dom; return dma_dom;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment