From: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>
To: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org,
will.deacon-5wv7dgnIgG8@public.gmane.org,
catalin.marinas-5wv7dgnIgG8@public.gmane.org
Cc: laurent.pinchart+renesas-ryLnwIuWjnjg/C1BVhZhaw@public.gmane.org,
arnd-r2nGTMty4D4@public.gmane.org,
djkurtz-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
thunder.leizhen-hv44wF8Li93QT0dZR+AlfA@public.gmane.org,
yingjoe.chen-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org,
treding-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org,
David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
yong.wu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org
Subject: [PATCH v2 1/4] iommu/iova: Avoid over-allocating when size-aligned
Date: Thu, 11 Jun 2015 16:54:21 +0100 [thread overview]
Message-ID: <0c0dd852b83cb7ae8104884b117f9c17cf6d440e.1434036796.git.robin.murphy@arm.com> (raw)
In-Reply-To: <cover.1434036796.git.robin.murphy-5wv7dgnIgG8@public.gmane.org>
Currently, allocating a size-aligned IOVA region quietly adjusts the
actual allocation size in the process, returning a rounded-up
power-of-two-sized allocation. This results in mismatched behaviour in
the IOMMU driver if the original size was not a power of two, where the
original size is mapped, but the rounded-up IOVA size is unmapped.
Whilst some IOMMUs will happily unmap already-unmapped pages, others
consider this an error, so fix it by computing the necessary alignment
padding without altering the actual allocation size. Also clean up by
making pad_size unsigned, since negative padding makes no sense.
CC: David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Signed-off-by: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>
---
drivers/iommu/intel-iommu.c | 2 ++
drivers/iommu/iova.c | 23 ++++++-----------------
2 files changed, 8 insertions(+), 17 deletions(-)
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 68d43be..ada1fb9 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -2910,6 +2910,8 @@ static struct iova *intel_alloc_iova(struct device *dev,
/* Restrict dma_mask to the width that the iommu can handle */
dma_mask = min_t(uint64_t, DOMAIN_MAX_ADDR(domain->gaw), dma_mask);
+ /* Ensure we reserve the whole size-aligned region */
+ nrpages = __roundup_pow_of_two(nrpages);
if (!dmar_forcedac && dma_mask > DMA_BIT_MASK(32)) {
/*
diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index 9dd8208..2c4a51b 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -120,19 +120,14 @@ __cached_rbnode_delete_update(struct iova_domain *iovad, struct iova *free)
}
}
-/* Computes the padding size required, to make the
- * the start address naturally aligned on its size
+/*
+ * Computes the padding size required, to make the start address
+ * naturally aligned on the power-of-two order of its size
*/
-static int
-iova_get_pad_size(int size, unsigned int limit_pfn)
+static unsigned int
+iova_get_pad_size(unsigned int size, unsigned int limit_pfn)
{
- unsigned int pad_size = 0;
- unsigned int order = ilog2(size);
-
- if (order)
- pad_size = (limit_pfn + 1) % (1 << order);
-
- return pad_size;
+ return (limit_pfn + 1 - size) & (__roundup_pow_of_two(size) - 1);
}
static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
@@ -264,12 +259,6 @@ alloc_iova(struct iova_domain *iovad, unsigned long size,
if (!new_iova)
return NULL;
- /* If size aligned is set then round the size to
- * to next power of two.
- */
- if (size_aligned)
- size = __roundup_pow_of_two(size);
-
ret = __alloc_and_insert_iova_range(iovad, size, limit_pfn,
new_iova, size_aligned);
--
1.9.1
WARNING: multiple messages have this Message-ID (diff)
From: robin.murphy@arm.com (Robin Murphy)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v2 1/4] iommu/iova: Avoid over-allocating when size-aligned
Date: Thu, 11 Jun 2015 16:54:21 +0100 [thread overview]
Message-ID: <0c0dd852b83cb7ae8104884b117f9c17cf6d440e.1434036796.git.robin.murphy@arm.com> (raw)
In-Reply-To: <cover.1434036796.git.robin.murphy@arm.com>
Currently, allocating a size-aligned IOVA region quietly adjusts the
actual allocation size in the process, returning a rounded-up
power-of-two-sized allocation. This results in mismatched behaviour in
the IOMMU driver if the original size was not a power of two, where the
original size is mapped, but the rounded-up IOVA size is unmapped.
Whilst some IOMMUs will happily unmap already-unmapped pages, others
consider this an error, so fix it by computing the necessary alignment
padding without altering the actual allocation size. Also clean up by
making pad_size unsigned, since negative padding makes no sense.
CC: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
---
drivers/iommu/intel-iommu.c | 2 ++
drivers/iommu/iova.c | 23 ++++++-----------------
2 files changed, 8 insertions(+), 17 deletions(-)
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 68d43be..ada1fb9 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -2910,6 +2910,8 @@ static struct iova *intel_alloc_iova(struct device *dev,
/* Restrict dma_mask to the width that the iommu can handle */
dma_mask = min_t(uint64_t, DOMAIN_MAX_ADDR(domain->gaw), dma_mask);
+ /* Ensure we reserve the whole size-aligned region */
+ nrpages = __roundup_pow_of_two(nrpages);
if (!dmar_forcedac && dma_mask > DMA_BIT_MASK(32)) {
/*
diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index 9dd8208..2c4a51b 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -120,19 +120,14 @@ __cached_rbnode_delete_update(struct iova_domain *iovad, struct iova *free)
}
}
-/* Computes the padding size required, to make the
- * the start address naturally aligned on its size
+/*
+ * Computes the padding size required, to make the start address
+ * naturally aligned on the power-of-two order of its size
*/
-static int
-iova_get_pad_size(int size, unsigned int limit_pfn)
+static unsigned int
+iova_get_pad_size(unsigned int size, unsigned int limit_pfn)
{
- unsigned int pad_size = 0;
- unsigned int order = ilog2(size);
-
- if (order)
- pad_size = (limit_pfn + 1) % (1 << order);
-
- return pad_size;
+ return (limit_pfn + 1 - size) & (__roundup_pow_of_two(size) - 1);
}
static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
@@ -264,12 +259,6 @@ alloc_iova(struct iova_domain *iovad, unsigned long size,
if (!new_iova)
return NULL;
- /* If size aligned is set then round the size to
- * to next power of two.
- */
- if (size_aligned)
- size = __roundup_pow_of_two(size);
-
ret = __alloc_and_insert_iova_range(iovad, size, limit_pfn,
new_iova, size_aligned);
--
1.9.1
next prev parent reply other threads:[~2015-06-11 15:54 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-11 15:54 [PATCH v2 0/4] arm64: IOMMU-backed DMA mapping Robin Murphy
2015-06-11 15:54 ` Robin Murphy
[not found] ` <cover.1434036796.git.robin.murphy-5wv7dgnIgG8@public.gmane.org>
2015-06-11 15:54 ` Robin Murphy [this message]
2015-06-11 15:54 ` [PATCH v2 1/4] iommu/iova: Avoid over-allocating when size-aligned Robin Murphy
2015-06-11 15:54 ` [PATCH v2 2/4] iommu: Implement common IOMMU ops for DMA mapping Robin Murphy
2015-06-11 15:54 ` Robin Murphy
[not found] ` <bf63d80c36533aa4dcc401b5a4b312ba00237e1c.1434036796.git.robin.murphy-5wv7dgnIgG8@public.gmane.org>
2015-06-18 15:00 ` Yong Wu
2015-06-18 15:00 ` Yong Wu
2015-06-18 16:54 ` Robin Murphy
2015-06-18 16:54 ` Robin Murphy
[not found] ` <5582F7C5.8060901-5wv7dgnIgG8@public.gmane.org>
2015-06-19 1:06 ` Yong Wu
2015-06-19 1:06 ` Yong Wu
2015-07-03 9:27 ` Yong Wu
2015-07-03 9:27 ` Yong Wu
2015-07-03 16:44 ` Robin Murphy
2015-07-03 16:44 ` Robin Murphy
[not found] ` <5596BBD3.1040708-5wv7dgnIgG8@public.gmane.org>
2015-07-07 7:37 ` Yong Wu
2015-07-07 7:37 ` Yong Wu
2015-06-11 15:54 ` [PATCH v2 3/4] arm64: Add IOMMU dma_ops Robin Murphy
2015-06-11 15:54 ` Robin Murphy
[not found] ` <4e298538ccedbe6f593d8edc40eb3a08ad9e4df5.1434036796.git.robin.murphy-5wv7dgnIgG8@public.gmane.org>
2015-06-18 15:00 ` Yong Wu
2015-06-18 15:00 ` Yong Wu
2015-06-11 15:54 ` [PATCH v2 4/4] arm64: Hook up " Robin Murphy
2015-06-11 15:54 ` Robin Murphy
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=0c0dd852b83cb7ae8104884b117f9c17cf6d440e.1434036796.git.robin.murphy@arm.com \
--to=robin.murphy-5wv7dgnigg8@public.gmane.org \
--cc=arnd-r2nGTMty4D4@public.gmane.org \
--cc=catalin.marinas-5wv7dgnIgG8@public.gmane.org \
--cc=djkurtz-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org \
--cc=laurent.pinchart+renesas-ryLnwIuWjnjg/C1BVhZhaw@public.gmane.org \
--cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
--cc=thunder.leizhen-hv44wF8Li93QT0dZR+AlfA@public.gmane.org \
--cc=treding-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org \
--cc=will.deacon-5wv7dgnIgG8@public.gmane.org \
--cc=yingjoe.chen-NuS5LvNUpcJWk0Htik3J/w@public.gmane.org \
--cc=yong.wu-NuS5LvNUpcJWk0Htik3J/w@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.