From: Joerg Roedel <joro@8bytes.org>
To: iommu@lists.linux-foundation.org
Cc: linux-kernel@vger.kernel.org,
Tom Lendacky <thomas.lendacky@amd.com>,
Arindam Nath <arindam.nath@amd.com>,
Joerg Roedel <jroedel@suse.de>
Subject: [PATCH 4/7] iommu/amd: Add locking to per-domain flush-queue
Date: Wed, 7 Jun 2017 16:58:17 +0200 [thread overview]
Message-ID: <1496847500-6781-5-git-send-email-joro@8bytes.org> (raw)
In-Reply-To: <1496847500-6781-1-git-send-email-joro@8bytes.org>
From: Joerg Roedel <jroedel@suse.de>
With locking we can safely access the flush-queues of other
cpus.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
---
drivers/iommu/amd_iommu.c | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index 6a5c858..9aa2735 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -144,6 +144,7 @@ struct flush_queue_entry {
struct flush_queue {
struct flush_queue_entry *entries;
unsigned head, tail;
+ spinlock_t lock;
};
/*
@@ -1773,6 +1774,8 @@ static int dma_ops_domain_alloc_flush_queue(struct dma_ops_domain *dom)
dma_ops_domain_free_flush_queue(dom);
return -ENOMEM;
}
+
+ spin_lock_init(&queue->lock);
}
return 0;
@@ -1780,6 +1783,8 @@ static int dma_ops_domain_alloc_flush_queue(struct dma_ops_domain *dom)
static inline bool queue_ring_full(struct flush_queue *queue)
{
+ assert_spin_locked(&queue->lock);
+
return (((queue->tail + 1) % FLUSH_QUEUE_SIZE) == queue->head);
}
@@ -1791,6 +1796,8 @@ static void queue_release(struct dma_ops_domain *dom,
{
unsigned i;
+ assert_spin_locked(&queue->lock);
+
queue_ring_for_each(i, queue)
free_iova_fast(&dom->iovad,
queue->entries[i].iova_pfn,
@@ -1803,6 +1810,7 @@ static inline unsigned queue_ring_add(struct flush_queue *queue)
{
unsigned idx = queue->tail;
+ assert_spin_locked(&queue->lock);
queue->tail = (idx + 1) % FLUSH_QUEUE_SIZE;
return idx;
@@ -1812,12 +1820,14 @@ 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);
if (queue_ring_full(queue)) {
domain_flush_tlb(&dom->domain);
@@ -1830,6 +1840,7 @@ static void queue_add(struct dma_ops_domain *dom,
queue->entries[idx].iova_pfn = address;
queue->entries[idx].pages = pages;
+ spin_unlock_irqrestore(&queue->lock, flags);
put_cpu_ptr(dom->flush_queue);
}
--
2.7.4
next prev parent reply other threads:[~2017-06-07 14:59 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-07 14:58 [PATCH 0/7] iommu/amd: Optimize iova queue flushing Joerg Roedel
2017-06-07 14:58 ` [PATCH 1/7] iommu/amd: Rip out old queue flushing code Joerg Roedel
2017-06-07 14:58 ` [PATCH 2/7] iommu/amd: Add per-domain flush-queue data structures Joerg Roedel
2017-06-07 14:58 ` [PATCH 3/7] iommu/amd: Make use of the per-domain flush queue Joerg Roedel
2017-06-07 14:58 ` Joerg Roedel [this message]
2017-06-07 14:58 ` [PATCH 5/7] iommu/amd: Add flush counters to struct dma_ops_domain Joerg Roedel
2017-06-07 14:58 ` [PATCH 6/7] iommu/amd: Add per-domain timer to flush per-cpu queues Joerg Roedel
2017-06-07 14:58 ` [PATCH 7/7] iommu/amd: Remove queue_release() function Joerg Roedel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1496847500-6781-5-git-send-email-joro@8bytes.org \
--to=joro@8bytes.org \
--cc=arindam.nath@amd.com \
--cc=iommu@lists.linux-foundation.org \
--cc=jroedel@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=thomas.lendacky@amd.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox