iommu.lists.linux-foundation.org archive mirror
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: John Garry <john.garry@huawei.com>
Cc: trivial@kernel.org, maz@kernel.org, linux-kernel@vger.kernel.org,
	linuxarm@huawei.com, iommu@lists.linux-foundation.org,
	robin.murphy@arm.com, linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH 0/4] iommu/arm-smmu-v3: Improve cmdq lock efficiency
Date: Thu, 16 Jul 2020 11:28:15 +0100	[thread overview]
Message-ID: <20200716102814.GD7036@willie-the-truck> (raw)
In-Reply-To: <20200716102233.GC7036@willie-the-truck>

On Thu, Jul 16, 2020 at 11:22:33AM +0100, Will Deacon wrote:
> On Thu, Jul 16, 2020 at 11:19:41AM +0100, Will Deacon wrote:
> > On Tue, Jun 23, 2020 at 01:28:36AM +0800, John Garry wrote:
> > > As mentioned in [0], the CPU may consume many cycles processing
> > > arm_smmu_cmdq_issue_cmdlist(). One issue we find is the cmpxchg() loop to
> > > get space on the queue takes approx 25% of the cycles for this function.
> > > 
> > > This series removes that cmpxchg().
> > 
> > How about something much simpler like the diff below?
> 
> Ah, scratch that, I don't drop the lock if we fail the cas with it held.
> Let me hack it some more (I have no hardware so I can only build-test this).

Right, second attempt...

Will

--->8

diff --git a/drivers/iommu/arm-smmu-v3.c b/drivers/iommu/arm-smmu-v3.c
index f578677a5c41..e6bcddd6ef69 100644
--- a/drivers/iommu/arm-smmu-v3.c
+++ b/drivers/iommu/arm-smmu-v3.c
@@ -560,6 +560,7 @@ struct arm_smmu_cmdq {
 	atomic_long_t			*valid_map;
 	atomic_t			owner_prod;
 	atomic_t			lock;
+	spinlock_t			slock;
 };
 
 struct arm_smmu_cmdq_batch {
@@ -1378,7 +1379,7 @@ static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
 	u64 cmd_sync[CMDQ_ENT_DWORDS];
 	u32 prod;
 	unsigned long flags;
-	bool owner;
+	bool owner, locked = false;
 	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
 	struct arm_smmu_ll_queue llq = {
 		.max_n_shift = cmdq->q.llq.max_n_shift,
@@ -1387,27 +1388,38 @@ static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
 
 	/* 1. Allocate some space in the queue */
 	local_irq_save(flags);
-	llq.val = READ_ONCE(cmdq->q.llq.val);
 	do {
 		u64 old;
+		llq.val = READ_ONCE(cmdq->q.llq.val);
 
-		while (!queue_has_space(&llq, n + sync)) {
+		if (queue_has_space(&llq, n + sync))
+			goto try_cas;
+
+		if (locked)
+			spin_unlock(&cmdq->slock);
+
+		do {
 			local_irq_restore(flags);
 			if (arm_smmu_cmdq_poll_until_not_full(smmu, &llq))
 				dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
 			local_irq_save(flags);
-		}
+		} while (!queue_has_space(&llq, n + sync));
 
+try_cas:
 		head.cons = llq.cons;
 		head.prod = queue_inc_prod_n(&llq, n + sync) |
 					     CMDQ_PROD_OWNED_FLAG;
 
 		old = cmpxchg_relaxed(&cmdq->q.llq.val, llq.val, head.val);
-		if (old == llq.val)
+		if (old != llq.val)
 			break;
 
-		llq.val = old;
+		if (!locked) {
+			spin_lock(&cmdq->slock);
+			locked = true;
+		}
 	} while (1);
+
 	owner = !(llq.prod & CMDQ_PROD_OWNED_FLAG);
 	head.prod &= ~CMDQ_PROD_OWNED_FLAG;
 	llq.prod &= ~CMDQ_PROD_OWNED_FLAG;
@@ -3192,6 +3204,7 @@ static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu)
 
 	atomic_set(&cmdq->owner_prod, 0);
 	atomic_set(&cmdq->lock, 0);
+	spin_lock_init(&cmdq->slock);
 
 	bitmap = (atomic_long_t *)bitmap_zalloc(nents, GFP_KERNEL);
 	if (!bitmap) {
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

  reply	other threads:[~2020-07-16 10:28 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-22 17:28 [PATCH 0/4] iommu/arm-smmu-v3: Improve cmdq lock efficiency John Garry
2020-06-22 17:28 ` [PATCH 1/4] iommu/arm-smmu-v3: Fix trivial typo John Garry
2020-06-22 17:28 ` [PATCH 2/4] iommu/arm-smmu-v3: Calculate bits for prod and owner John Garry
2020-06-22 17:28 ` [PATCH 3/4] iommu/arm-smmu-v3: Always issue a CMD_SYNC per batch John Garry
2020-06-22 17:28 ` [PATCH 4/4] iommu/arm-smmu-v3: Remove cmpxchg() in arm_smmu_cmdq_issue_cmdlist() John Garry
2020-06-23  1:07   ` kernel test robot
2020-06-23  9:21     ` John Garry
2020-06-23  9:35       ` Rikard Falkeborn
2020-06-23 10:19         ` John Garry
2020-06-23 13:55           ` Rikard Falkeborn
2020-06-26 10:05             ` John Garry
2020-06-23 16:22       ` Robin Murphy
2020-06-24  8:15         ` John Garry
2020-07-16 10:20   ` Will Deacon
2020-07-16 10:26     ` John Garry
2020-07-08 13:00 ` [PATCH 0/4] iommu/arm-smmu-v3: Improve cmdq lock efficiency John Garry
2020-07-16 10:19 ` Will Deacon
2020-07-16 10:22   ` Will Deacon
2020-07-16 10:28     ` Will Deacon [this message]
2020-07-16 10:56       ` John Garry
2020-07-16 11:22         ` Robin Murphy
2020-07-16 11:30           ` John Garry
2020-07-16 11:32           ` Will Deacon
2020-07-16 16:50             ` John Garry
2020-07-16 13:31       ` John Garry

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=20200716102814.GD7036@willie-the-truck \
    --to=will@kernel.org \
    --cc=iommu@lists.linux-foundation.org \
    --cc=john.garry@huawei.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=maz@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=trivial@kernel.org \
    /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;
as well as URLs for NNTP newsgroup(s).