[PATCH 1/3] rbtree: include rcu.h because we use it

[PATCH 1/3] rbtree: include rcu.h because we use it

[Sebastian Andrzej Siewior]

Since commit c1adf20052d8 ("Introduce rb_replace_node_rcu()")
rbtree_augmented.h uses RCU related data structures but does not include
them. It works as long as gets somehow included before that and fails
otherwise.
Noticed during -RT rebase.

Cc: David Howells <dhowells@redhat.com
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 include/linux/rbtree_augmented.h | 1 +
 include/linux/rbtree_latch.h     | 1 +
 2 files changed, 2 insertions(+)

diff --git a/include/linux/rbtree_augmented.h b/include/linux/rbtree_augmented.h
index 9702b6e183bc..110e14060dc9 100644
--- a/include/linux/rbtree_augmented.h
+++ b/include/linux/rbtree_augmented.h
@@ -26,6 +26,7 @@
 
 #include <linux/compiler.h>
 #include <linux/rbtree.h>
+#include <linux/rcupdate.h>
 
 /*
  * Please note - only struct rb_augment_callbacks and the prototypes for
diff --git a/include/linux/rbtree_latch.h b/include/linux/rbtree_latch.h
index 4f3432c61d12..0a7762cf04f7 100644
--- a/include/linux/rbtree_latch.h
+++ b/include/linux/rbtree_latch.h
@@ -34,6 +34,7 @@
 
 #include <linux/rbtree.h>
 #include <linux/seqlock.h>
+#include <linux/rcupdate.h>
 
 struct latch_tree_node {
 	struct rb_node node[2];
-- 
2.13.1

[PATCH 2/3] iommu/iova: don't disable preempt around this_cpu_ptr()

[Sebastian Andrzej Siewior]

Commit 583248e6620a ("iommu/iova: Disable preemption around use of
this_cpu_ptr()") disables preemption while accessing a per-CPU variable.
This does keep lockdep quiet. However I don't see the point why it is
bad if we get migrated after its access to another CPU.
__iova_rcache_insert() and __iova_rcache_get() immediately locks the
variable after obtaining it - before accessing its members.
_If_ we get migrated away after retrieving the address of cpu_rcache
before taking the lock then the *other* task on the same CPU will
retrieve the same address of cpu_rcache and will spin on the lock.

alloc_iova_fast() disables preemption while invoking
free_cpu_cached_iovas() on each CPU. The function itself uses
per_cpu_ptr() which does not trigger a warning (like this_cpu_ptr()
does). It _could_ make sense to use get_online_cpus() instead but the we
have a hotplug notifier for CPU down (and none for up) so we are good.

Cc: Joerg Roedel <joro@8bytes.org>
Cc: iommu@lists.linux-foundation.org
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 drivers/iommu/iova.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index 5c88ba70e4e0..f0ff0aa04081 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -22,6 +22,7 @@
 #include <linux/slab.h>
 #include <linux/smp.h>
 #include <linux/bitops.h>
+#include <linux/cpu.h>
 
 static bool iova_rcache_insert(struct iova_domain *iovad,
 			       unsigned long pfn,
@@ -398,10 +399,8 @@ alloc_iova_fast(struct iova_domain *iovad, unsigned long size,
 
 		/* Try replenishing IOVAs by flushing rcache. */
 		flushed_rcache = true;
-		preempt_disable();
 		for_each_online_cpu(cpu)
 			free_cpu_cached_iovas(cpu, iovad);
-		preempt_enable();
 		goto retry;
 	}
 
@@ -729,7 +728,7 @@ static bool __iova_rcache_insert(struct iova_domain *iovad,
 	bool can_insert = false;
 	unsigned long flags;
 
-	cpu_rcache = get_cpu_ptr(rcache->cpu_rcaches);
+	cpu_rcache = raw_cpu_ptr(rcache->cpu_rcaches);
 	spin_lock_irqsave(&cpu_rcache->lock, flags);
 
 	if (!iova_magazine_full(cpu_rcache->loaded)) {
@@ -759,7 +758,6 @@ static bool __iova_rcache_insert(struct iova_domain *iovad,
 		iova_magazine_push(cpu_rcache->loaded, iova_pfn);
 
 	spin_unlock_irqrestore(&cpu_rcache->lock, flags);
-	put_cpu_ptr(rcache->cpu_rcaches);
 
 	if (mag_to_free) {
 		iova_magazine_free_pfns(mag_to_free, iovad);
@@ -793,7 +791,7 @@ static unsigned long __iova_rcache_get(struct iova_rcache *rcache,
 	bool has_pfn = false;
 	unsigned long flags;
 
-	cpu_rcache = get_cpu_ptr(rcache->cpu_rcaches);
+	cpu_rcache = raw_cpu_ptr(rcache->cpu_rcaches);
 	spin_lock_irqsave(&cpu_rcache->lock, flags);
 
 	if (!iova_magazine_empty(cpu_rcache->loaded)) {
@@ -815,7 +813,6 @@ static unsigned long __iova_rcache_get(struct iova_rcache *rcache,
 		iova_pfn = iova_magazine_pop(cpu_rcache->loaded, limit_pfn);
 
 	spin_unlock_irqrestore(&cpu_rcache->lock, flags);
-	put_cpu_ptr(rcache->cpu_rcaches);
 
 	return iova_pfn;
 }
-- 
2.13.1

Re: [PATCH 2/3] iommu/iova: don't disable preempt around this_cpu_ptr()

[Joerg Roedel]

On Tue, Jun 27, 2017 at 06:16:47PM +0200, Sebastian Andrzej Siewior wrote:
> Commit 583248e6620a ("iommu/iova: Disable preemption around use of
> this_cpu_ptr()") disables preemption while accessing a per-CPU variable.
> This does keep lockdep quiet. However I don't see the point why it is
> bad if we get migrated after its access to another CPU.
> __iova_rcache_insert() and __iova_rcache_get() immediately locks the
> variable after obtaining it - before accessing its members.
> _If_ we get migrated away after retrieving the address of cpu_rcache
> before taking the lock then the *other* task on the same CPU will
> retrieve the same address of cpu_rcache and will spin on the lock.
> 
> alloc_iova_fast() disables preemption while invoking
> free_cpu_cached_iovas() on each CPU. The function itself uses
> per_cpu_ptr() which does not trigger a warning (like this_cpu_ptr()
> does). It _could_ make sense to use get_online_cpus() instead but the we
> have a hotplug notifier for CPU down (and none for up) so we are good.

Does that really matter? The spin_lock disables irqs and thus avoids
preemption too. We also can't get rid of the irqsave lock here because
these locks are taken in the dma-api path which is used from interrupt
context.



	Joerg

Re: [PATCH 2/3] iommu/iova: don't disable preempt around this_cpu_ptr()

[Sebastian Andrzej Siewior]

On 2017-06-28 11:22:05 [+0200], Joerg Roedel wrote:
> On Tue, Jun 27, 2017 at 06:16:47PM +0200, Sebastian Andrzej Siewior wrote:
> > Commit 583248e6620a ("iommu/iova: Disable preemption around use of
> > this_cpu_ptr()") disables preemption while accessing a per-CPU variable.
> > This does keep lockdep quiet. However I don't see the point why it is
> > bad if we get migrated after its access to another CPU.
> > __iova_rcache_insert() and __iova_rcache_get() immediately locks the
> > variable after obtaining it - before accessing its members.
> > _If_ we get migrated away after retrieving the address of cpu_rcache
> > before taking the lock then the *other* task on the same CPU will
> > retrieve the same address of cpu_rcache and will spin on the lock.
> > 
> > alloc_iova_fast() disables preemption while invoking
> > free_cpu_cached_iovas() on each CPU. The function itself uses
> > per_cpu_ptr() which does not trigger a warning (like this_cpu_ptr()
> > does). It _could_ make sense to use get_online_cpus() instead but the we
> > have a hotplug notifier for CPU down (and none for up) so we are good.
> 
> Does that really matter? The spin_lock disables irqs and thus avoids
> preemption too. We also can't get rid of the irqsave lock here because
> these locks are taken in the dma-api path which is used from interrupt
> context.

It really does. The spin_lock() does disable preemption but this is not
the problem. The thing is that the preempt_disable() is superfluous and
it hurts Preempt-RT (and this is how I noticed it). Also the
get_cpu_ptr() is not requited and was only added to keep lockdep quiet
(according to the history).
Everything else here can stay as-is, I am just asking for the removal of
the redundant preempt_disable() where it is not required.

> 	Joerg

Sebastian

Re: [PATCH 2/3] iommu/iova: don't disable preempt around this_cpu_ptr()

[Joerg Roedel]

On Wed, Jun 28, 2017 at 11:31:55AM +0200, Sebastian Andrzej Siewior wrote:
> It really does. The spin_lock() does disable preemption but this is not
> the problem. The thing is that the preempt_disable() is superfluous and
> it hurts Preempt-RT (and this is how I noticed it). Also the
> get_cpu_ptr() is not requited and was only added to keep lockdep quiet
> (according to the history).
> Everything else here can stay as-is, I am just asking for the removal of
> the redundant preempt_disable() where it is not required.

Okay, makes sense, I applied both patches.


Thanks,

	Joerg

[PATCH 3/3] iommu/vt-d: don't disable preemption while accessing deferred_flush()

[Sebastian Andrzej Siewior]

get_cpu() disables preemption and returns the current CPU number. The
CPU number is only used once while retrieving the address of the local's
CPU deferred_flush pointer.
We can instead use raw_cpu_ptr() while we remain preemptible. The worst
thing that can happen is that flush_unmaps_timeout() is invoked multiple
times: once by taskA after seeing HIGH_WATER_MARK and then preempted to
another CPU and then by taskB which saw HIGH_WATER_MARK on the same CPU
as taskA. It is also likely that ->size got from HIGH_WATER_MARK to 0
right after its read because another CPU invoked flush_unmaps_timeout()
for this CPU.
The access to flush_data is protected by a spinlock so even if we get
migrated to another CPU or preempted - the data structure is protected.

While at it, I marked deferred_flush static since I can't find a
reference to it outside of this file.

Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: iommu@lists.linux-foundation.org
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 drivers/iommu/intel-iommu.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 8500deda9175..1c7118d1525e 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -481,7 +481,7 @@ struct deferred_flush_data {
 	struct deferred_flush_table *tables;
 };
 
-DEFINE_PER_CPU(struct deferred_flush_data, deferred_flush);
+static DEFINE_PER_CPU(struct deferred_flush_data, deferred_flush);
 
 /* bitmap for indexing intel_iommus */
 static int g_num_of_iommus;
@@ -3725,10 +3725,8 @@ static void add_unmap(struct dmar_domain *dom, unsigned long iova_pfn,
 	struct intel_iommu *iommu;
 	struct deferred_flush_entry *entry;
 	struct deferred_flush_data *flush_data;
-	unsigned int cpuid;
 
-	cpuid = get_cpu();
-	flush_data = per_cpu_ptr(&deferred_flush, cpuid);
+	flush_data = raw_cpu_ptr(&deferred_flush);
 
 	/* Flush all CPUs' entries to avoid deferring too much.  If
 	 * this becomes a bottleneck, can just flush us, and rely on
@@ -3761,8 +3759,6 @@ static void add_unmap(struct dmar_domain *dom, unsigned long iova_pfn,
 	}
 	flush_data->size++;
 	spin_unlock_irqrestore(&flush_data->lock, flags);
-
-	put_cpu();
 }
 
 static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size)
-- 
2.13.1