From: Niklas Schnelle <schnelle@linux.ibm.com>
To: Joerg Roedel <joro@8bytes.org>, Will Deacon <will@kernel.org>,
Robin Murphy <robin.murphy@arm.com>,
Jason Gunthorpe <jgg@nvidia.com>,
Wenjia Zhang <wenjia@linux.ibm.com>
Cc: Matthew Rosato <mjrosato@linux.ibm.com>,
Gerd Bayer <gbayer@linux.ibm.com>,
Pierre Morel <pmorel@linux.ibm.com>,
iommu@lists.linux.dev, linux-s390@vger.kernel.org,
borntraeger@linux.ibm.com, hca@linux.ibm.com, gor@linux.ibm.com,
gerald.schaefer@linux.ibm.com, agordeev@linux.ibm.com,
svens@linux.ibm.com, linux-kernel@vger.kernel.org,
Julian Ruess <julianr@linux.ibm.com>
Subject: [PATCH v4 7/7] iommu/dma: Add IOMMU op to choose lazy domain type
Date: Wed, 4 Jan 2023 13:05:43 +0100 [thread overview]
Message-ID: <20230104120543.308933-8-schnelle@linux.ibm.com> (raw)
In-Reply-To: <20230104120543.308933-1-schnelle@linux.ibm.com>
With two flush queue variants add an IOMMU operation that allows the
IOMMU driver to choose its preferred flush queue variant on a per device
basis. For s390 use this callback to choose the single queue variant
whenever the device requires explicit IOTLB flushes on map indicating
that we're running in a paged memory guest with expensive IOTLB flushes.
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
---
drivers/iommu/iommu.c | 13 +++++++++++++
drivers/iommu/s390-iommu.c | 11 +++++++++++
include/linux/iommu.h | 5 +++++
3 files changed, 29 insertions(+)
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 7ae2ff35b88e..c4699a0e5feb 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -1616,6 +1616,16 @@ static int iommu_get_def_domain_type(struct device *dev)
return 0;
}
+static int iommu_get_lazy_domain_type(struct device *dev)
+{
+ const struct iommu_ops *ops = dev_iommu_ops(dev);
+
+ if (ops->lazy_domain_type)
+ return ops->lazy_domain_type(dev);
+
+ return 0;
+}
+
static int iommu_group_alloc_default_domain(struct bus_type *bus,
struct iommu_group *group,
unsigned int type)
@@ -1649,6 +1659,9 @@ static int iommu_alloc_default_domain(struct iommu_group *group,
type = iommu_get_def_domain_type(dev) ? : iommu_def_domain_type;
+ if (!!(type & __IOMMU_DOMAIN_DMA_LAZY))
+ type = iommu_get_lazy_domain_type(dev) ? : type;
+
return iommu_group_alloc_default_domain(dev->bus, group, type);
}
diff --git a/drivers/iommu/s390-iommu.c b/drivers/iommu/s390-iommu.c
index ff73b75be886..b8aab37e8b15 100644
--- a/drivers/iommu/s390-iommu.c
+++ b/drivers/iommu/s390-iommu.c
@@ -459,6 +459,16 @@ static void s390_iommu_get_resv_regions(struct device *dev,
}
}
+static int s390_iommu_lazy_domain_type(struct device *dev)
+{
+ struct zpci_dev *zdev = to_zpci_dev(dev);
+
+ if (zdev->tlb_refresh)
+ return IOMMU_DOMAIN_DMA_SQ;
+
+ return IOMMU_DOMAIN_DMA_FQ;
+}
+
static struct iommu_device *s390_iommu_probe_device(struct device *dev)
{
struct zpci_dev *zdev;
@@ -798,6 +808,7 @@ static const struct iommu_ops s390_iommu_ops = {
.device_group = generic_device_group,
.pgsize_bitmap = SZ_4K,
.get_resv_regions = s390_iommu_get_resv_regions,
+ .lazy_domain_type = s390_iommu_lazy_domain_type,
.default_domain_ops = &(const struct iommu_domain_ops) {
.attach_dev = s390_iommu_attach_device,
.detach_dev = s390_iommu_detach_device,
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index 74cee59516aa..aec895087f63 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -250,6 +250,10 @@ struct iommu_iotlb_gather {
* - IOMMU_DOMAIN_IDENTITY: must use an identity domain
* - IOMMU_DOMAIN_DMA: must use a dma domain
* - 0: use the default setting
+ * @lazy_domain_type: Domain type for lazy TLB invalidation, return value:
+ * - IOMMU_DOMAIN_DMA_FQ: Use per-CPU flush queue
+ * - IOMMU_DOMAIN_DMA_SQ: Use single flush queue
+ * - 0: use the default setting
* @default_domain_ops: the default ops for domains
* @remove_dev_pasid: Remove any translation configurations of a specific
* pasid, so that any DMA transactions with this pasid
@@ -283,6 +287,7 @@ struct iommu_ops {
struct iommu_page_response *msg);
int (*def_domain_type)(struct device *dev);
+ int (*lazy_domain_type)(struct device *dev);
void (*remove_dev_pasid)(struct device *dev, ioasid_t pasid);
const struct iommu_domain_ops *default_domain_ops;
--
2.34.1
prev parent reply other threads:[~2023-01-04 12:06 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-04 12:05 [PATCH v4 0/7] iommu/dma: s390 DMA API conversion and optimized IOTLB flushing Niklas Schnelle
2023-01-04 12:05 ` [PATCH v4 1/7] s390/ism: Set DMA coherent mask Niklas Schnelle
2023-01-17 15:09 ` Matthew Rosato
2023-01-04 12:05 ` [PATCH v4 2/7] iommu: Allow .iotlb_sync_map to fail and handle s390's -ENOMEM return Niklas Schnelle
2023-01-05 7:15 ` Baolu Lu
2023-01-17 15:10 ` Matthew Rosato
2023-01-19 11:31 ` Niklas Schnelle
2023-01-04 12:05 ` [PATCH v4 3/7] s390/pci: prepare is_passed_through() for dma-iommu Niklas Schnelle
2023-01-17 15:10 ` Matthew Rosato
2023-01-04 12:05 ` [PATCH v4 4/7] s390/pci: Use dma-iommu layer Niklas Schnelle
2023-01-17 15:09 ` Matthew Rosato
2023-01-19 11:03 ` Niklas Schnelle
2023-01-19 15:59 ` Matthew Rosato
2023-01-19 16:04 ` Niklas Schnelle
2023-01-19 16:12 ` Matthew Rosato
2023-01-19 16:33 ` Niklas Schnelle
2023-01-19 16:40 ` Matthew Rosato
2023-01-04 12:05 ` [PATCH v4 5/7] iommu/dma: Allow a single FQ in addition to per-CPU FQs Niklas Schnelle
2023-01-19 15:55 ` Niklas Schnelle
2023-01-19 16:16 ` Niklas Schnelle
2023-01-04 12:05 ` [PATCH v4 6/7] iommu/dma: Enable variable queue size and use larger single queue Niklas Schnelle
2023-01-04 12:05 ` Niklas Schnelle [this message]
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=20230104120543.308933-8-schnelle@linux.ibm.com \
--to=schnelle@linux.ibm.com \
--cc=agordeev@linux.ibm.com \
--cc=borntraeger@linux.ibm.com \
--cc=gbayer@linux.ibm.com \
--cc=gerald.schaefer@linux.ibm.com \
--cc=gor@linux.ibm.com \
--cc=hca@linux.ibm.com \
--cc=iommu@lists.linux.dev \
--cc=jgg@nvidia.com \
--cc=joro@8bytes.org \
--cc=julianr@linux.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=mjrosato@linux.ibm.com \
--cc=pmorel@linux.ibm.com \
--cc=robin.murphy@arm.com \
--cc=svens@linux.ibm.com \
--cc=wenjia@linux.ibm.com \
--cc=will@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