There are lots of mismatches between comments and codes, this
patch do these comment fixes.
Signed-off-by: Jiang Biao <benbjiang@tencent.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
'desc' should be freed before leaving from err handing path.
Fixes: 1ce9e6055f ("virtio_ring: introduce packed ring support")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
stable@vger.kernel.org
vring_create_virtqueue() allows the caller to specify via the
may_reduce_num parameter whether the vring code is allowed to
allocate a smaller ring than specified.
However, the split ring allocation code tries to allocate a
smaller ring on allocation failure regardless of what the
caller specified. This may cause trouble for e.g. virtio-pci
in legacy mode, which does not support ring resizing. (The
packed ring code does not resize in any case.)
Let's fix this by bailing out immediately in the split ring code
if the requested size cannot be allocated and may_reduce_num has
not been specified.
While at it, fix a typo in the usage instructions.
Fixes: 2a2d1382fe ("virtio: Add improved queue allocation API")
Cc: stable@vger.kernel.org # v4.6+
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Halil Pasic <pasic@linux.ibm.com>
Reviewed-by: Jens Freimann <jfreimann@redhat.com>
This function returns the maximum segment size for a single
dma transaction of a virtio device. The possible limit comes
from the SWIOTLB implementation in the Linux kernel, that
has an upper limit of (currently) 256kb of contiguous
memory it can map. Other DMA-API implementations might also
have limits.
Use the new dma_max_mapping_size() function to determine the
maximum mapping size when DMA-API is in use for virtio.
Cc: stable@vger.kernel.org
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
There's no reason to expose struct vring_packed in UAPI - if we do we
won't be able to change or drop it, and it's not part of any interface.
Let's move it to virtio_ring.c
Cc: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch introduces the support for VIRTIO_F_ORDER_PLATFORM.
If this feature is negotiated, the driver must use the barriers
suitable for hardware devices. Otherwise, the device and driver
are assumed to be implemented in software, that is they can be
assumed to run on identical CPUs in an SMP configuration. Thus
a weaker form of memory barriers is sufficient to yield better
performance.
It is recommended that an add-in card based PCI device offers
this feature for portability. The device will fail to operate
further or will operate in a slower emulation mode if this
feature is offered but not accepted.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Leverage the EVENT_IDX feature in packed ring to suppress
events when it's available.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce the packed ring support. Packed ring can only be
created by vring_create_virtqueue() and each chunk of packed
ring will be allocated individually. Packed ring can not be
created on preallocated memory by vring_new_virtqueue() or
the likes currently.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cache whether we will use DMA API, instead of doing the
check every time. We are going to check whether DMA API
is used more often in packed ring.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce a specific function to create the split ring.
And also move the DMA allocation and size information to
the .split sub-structure.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Put the split ring's desc state into the .split sub-structure,
and allocate desc state for split ring separately, this makes
the code more readable and more consistent with what we will
do for packed ring.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce a helper to check whether we will use indirect
feature. It will be used by packed ring too.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce debug helpers for last_add_time update, check and
invalid. They will be used by packed ring too.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Put the split ring specific fields in a sub-struct named
as "split" to avoid misuse after introducing packed ring.
There is no functional change.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Put the xxx_split() functions together to make the
code more readable and avoid misuse after introducing
the packed ring. There is no functional change.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add _split suffix for split ring specific functions. This
is a preparation for introducing the packed ring support.
There is no functional change.
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently <linux/slab.h> #includes <linux/kmemleak.h> for no obvious
reason. It looks like it's only a convenience, so remove kmemleak.h
from slab.h and add <linux/kmemleak.h> to any users of kmemleak_* that
don't already #include it. Also remove <linux/kmemleak.h> from source
files that do not use it.
This is tested on i386 allmodconfig and x86_64 allmodconfig. It would
be good to run it through the 0day bot for other $ARCHes. I have
neither the horsepower nor the storage space for the other $ARCHes.
Update: This patch has been extensively build-tested by both the 0day
bot & kisskb/ozlabs build farms. Both of them reported 2 build failures
for which patches are included here (in v2).
[ slab.h is the second most used header file after module.h; kernel.h is
right there with slab.h. There could be some minor error in the
counting due to some #includes having comments after them and I didn't
combine all of those. ]
[akpm@linux-foundation.org: security/keys/big_key.c needs vmalloc.h, per sfr]
Link: http://lkml.kernel.org/r/e4309f98-3749-93e1-4bb7-d9501a39d015@infradead.org
Link: http://kisskb.ellerman.id.au/kisskb/head/13396/
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Reported-by: Michael Ellerman <mpe@ellerman.id.au> [2 build failures]
Reported-by: Fengguang Wu <fengguang.wu@intel.com> [2 build failures]
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Wei Yongjun <weiyongjun1@huawei.com>
Cc: Luis R. Rodriguez <mcgrof@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: John Johansen <john.johansen@canonical.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The vq->vq.num_free hasn't been changed when error happens,
so it shouldn't be changed when handling the error.
Fixes: 780bc7903a ("virtio_ring: Support DMA APIs")
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This is mostly updates of the usual suspects: lpfc, qla2xxx, hisi_sas, megaraid_sas, zfcp and a host of minor updates.
The major driver change here is the elimination of the block based
cciss driver in favour of the SCSI based hpsa driver (which now drives
all the legacy cases cciss used to be required for). Plus a reset
handler clean up and the redo of the SAS SMP handler to use bsg lib.
Signed-off-by: James E.J. Bottomley <jejb@linux.vnet.ibm.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=nQUG
-----END PGP SIGNATURE-----
Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI updates from James Bottomley:
"This is mostly updates of the usual suspects: lpfc, qla2xxx, hisi_sas,
megaraid_sas, zfcp and a host of minor updates.
The major driver change here is the elimination of the block based
cciss driver in favour of the SCSI based hpsa driver (which now drives
all the legacy cases cciss used to be required for). Plus a reset
handler clean up and the redo of the SAS SMP handler to use bsg lib"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (279 commits)
scsi: scsi-mq: Always unprepare before requeuing a request
scsi: Show .retries and .jiffies_at_alloc in debugfs
scsi: Improve requeuing behavior
scsi: Call scsi_initialize_rq() for filesystem requests
scsi: qla2xxx: Reset the logo flag, after target re-login.
scsi: qla2xxx: Fix slow mem alloc behind lock
scsi: qla2xxx: Clear fc4f_nvme flag
scsi: qla2xxx: add missing includes for qla_isr
scsi: qla2xxx: Fix an integer overflow in sysfs code
scsi: aacraid: report -ENOMEM to upper layer from aac_convert_sgraw2()
scsi: aacraid: get rid of one level of indentation
scsi: aacraid: fix indentation errors
scsi: storvsc: fix memory leak on ring buffer busy
scsi: scsi_transport_sas: switch to bsg-lib for SMP passthrough
scsi: smartpqi: remove the smp_handler stub
scsi: hpsa: remove the smp_handler stub
scsi: bsg-lib: pass the release callback through bsg_setup_queue
scsi: Rework handling of scsi_device.vpd_pg8[03]
scsi: Rework the code for caching Vital Product Data (VPD)
scsi: rcu: Introduce rcu_swap_protected()
...
If using indirect descriptors, you can make the total_sg as large as you
want. If not, BUG is too serious because the function later returns
-ENOSPC.
Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Allow zero to be store as a ctx, with this we could store e.g zero
value which could be meaningful for the case of storing headroom
through ctx.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow extra context per descriptor. To avoid slow down for data path,
this disables use of indirect descriptors for this vq.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Allows maintaining extra context per vq. For ease of use, passing in
NULL is legal and disables the feature for all vqs.
Includes fixes by Christian for s390, acked by Cornelia.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This reverts commit c7070619f3.
This has been shown to regress on some ARM systems:
by forcing on DMA API usage for ARM systems, we have inadvertently
kicked open a hornets' nest in terms of cache-coherency. Namely that
unless the virtio device is explicitly described as capable of coherent
DMA by firmware, the DMA APIs on ARM and other DT-based platforms will
assume it is non-coherent. This turns out to cause a big problem for the
likes of QEMU and kvmtool, which generate virtio-mmio devices in their
guest DTs but neglect to add the often-overlooked "dma-coherent"
property; as a result, we end up with the guest making non-cacheable
accesses to the vring, the host doing so cacheably, both talking past
each other and things going horribly wrong.
We are working on a safer work-around.
Fixes: c7070619f3 ("vring: Force use of DMA API for ARM-based systems with legacy devices")
Reported-by: Robin Murphy <robin.murphy@arm.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Booting Linux on an ARM fastmodel containing an SMMU emulation results
in an unexpected I/O page fault from the legacy virtio-blk PCI device:
[ 1.211721] arm-smmu-v3 2b400000.smmu: event 0x10 received:
[ 1.211800] arm-smmu-v3 2b400000.smmu: 0x00000000fffff010
[ 1.211880] arm-smmu-v3 2b400000.smmu: 0x0000020800000000
[ 1.211959] arm-smmu-v3 2b400000.smmu: 0x00000008fa081002
[ 1.212075] arm-smmu-v3 2b400000.smmu: 0x0000000000000000
[ 1.212155] arm-smmu-v3 2b400000.smmu: event 0x10 received:
[ 1.212234] arm-smmu-v3 2b400000.smmu: 0x00000000fffff010
[ 1.212314] arm-smmu-v3 2b400000.smmu: 0x0000020800000000
[ 1.212394] arm-smmu-v3 2b400000.smmu: 0x00000008fa081000
[ 1.212471] arm-smmu-v3 2b400000.smmu: 0x0000000000000000
<system hangs failing to read partition table>
This is because the legacy virtio-blk device is behind an SMMU, so we
have consequently swizzled its DMA ops and configured the SMMU to
translate accesses. This then requires the vring code to use the DMA API
to establish translations, otherwise all transactions will result in
fatal faults and termination.
Given that ARM-based systems only see an SMMU if one is really present
(the topology is all described by firmware tables such as device-tree or
IORT), then we can safely use the DMA API for all legacy virtio devices.
Modern devices can advertise the prescense of an IOMMU using the
VIRTIO_F_IOMMU_PLATFORM feature flag.
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: <stable@vger.kernel.org>
Fixes: 876945dbf6 ("arm64: Hook up IOMMU dma_ops")
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
The device (not the driver) populates the used ring and includes the len
of how much data was written.
Signed-off-by: Felipe Franciosi <felipe@nutanix.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# make C=2 CF="-D__CHECK_ENDIAN__" ./drivers/virtio/
drivers/virtio/virtio_ring.c:423:19: warning: incorrect type in assignment (different base types)
drivers/virtio/virtio_ring.c:423:19: expected unsigned int [unsigned] [assigned] i
drivers/virtio/virtio_ring.c:423:19: got restricted __virtio16 [usertype] next
drivers/virtio/virtio_ring.c:423:19: warning: incorrect type in assignment (different base types)
drivers/virtio/virtio_ring.c:423:19: expected unsigned int [unsigned] [assigned] i
drivers/virtio/virtio_ring.c:423:19: got restricted __virtio16 [usertype] next
drivers/virtio/virtio_ring.c:423:19: warning: incorrect type in assignment (different base types)
drivers/virtio/virtio_ring.c:423:19: expected unsigned int [unsigned] [assigned] i
drivers/virtio/virtio_ring.c:423:19: got restricted __virtio16 [usertype] next
drivers/virtio/virtio_ring.c:604:39: warning: incorrect type in initializer (different base types)
drivers/virtio/virtio_ring.c:604:39: expected unsigned short [unsigned] [usertype] nextflag
drivers/virtio/virtio_ring.c:604:39: got restricted __virtio16
drivers/virtio/virtio_ring.c:612:33: warning: restricted __virtio16 degrades to integer
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This inline function is unused on configurations
where dma_map/unmap are empty macros.
Make the function inline to avoid gcc errors because
of an unused static function.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
According to the spec, if the VIRTIO_RING_F_EVENT_IDX feature bit is
negotiated the driver MUST set flags to 0. Not dirtying the available
ring in virtqueue_disable_cb also has a minor positive performance
impact, improving L1 dcache load missed by ~0.5% in vring_bench.
Writes to the used event field (vring_used_event) are still unconditional.
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: <stable@vger.kernel.org> # f277ec4 virtio_ring: shadow available
Cc: <stable@vger.kernel.org>
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We get 1 warning when building kernel with W=1:
drivers/virtio/virtio_ring.c:170:16: warning: no previous prototype for 'vring_dma_dev' [-Wmissing-prototypes]
In fact, this function is only used in the file in which it is
declared and don't need a declaration, but can be made static.
so this patch marks this function with 'static'.
Signed-off-by: Baoyou Xie <baoyou.xie@linaro.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
On error, virtqueue_add calls START_USE but not
END_USE. Thankfully that's normally empty anyway,
but might not be when debugging. Fix it up.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
When using the indirect buffers feature, 'desc' is allocated in
virtqueue_add() but isn't freed before leaving on a ring full error,
causing a memory leak.
For example, it seems rather clear that this can trigger
with virtio net if mergeable buffers are not used.
Cc: stable@vger.kernel.org
Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The interaction between virtio and IOMMUs is messy.
On most systems with virtio, physical addresses match bus addresses,
and it doesn't particularly matter which one we use to program
the device.
On some systems, including Xen and any system with a physical device
that speaks virtio behind a physical IOMMU, we must program the IOMMU
for virtio DMA to work at all.
On other systems, including SPARC and PPC64, virtio-pci devices are
enumerated as though they are behind an IOMMU, but the virtio host
ignores the IOMMU, so we must either pretend that the IOMMU isn't
there or somehow map everything as the identity.
Add a feature bit to detect that quirk: VIRTIO_F_IOMMU_PLATFORM.
Any device with this feature bit set to 0 needs a quirk and has to be
passed physical addresses (as opposed to bus addresses) even though
the device is behind an IOMMU.
Note: it has to be a per-device quirk because for example, there could
be a mix of passed-through and virtual virtio devices. As another
example, some devices could be implemented by an out of process
hypervisor backend (in case of qemu vhost, or vhost-user) and so support
for an IOMMU needs to be coded up separately.
It would be cleanest to handle this in IOMMU core code, but that needs
per-device DMA ops. While we are waiting for that to be implemented, use
a work-around in virtio core.
Note: a "noiommu" feature is a quirk - add a wrapper to make
that clear.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Smatch complains that we might not initialize "queue". The issue is
callers like setup_vq() from virtio_pci_modern.c where "num" could be
something like 2 and "vring_align" is 64. In that case, vring_size() is
less than PAGE_SIZE. It won't happen in real life, but we're getting
the value of "num" from a register so it's not really possible to tell
what value it holds with static analysis.
Let's just silence the warning.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Wei Liu <wei.liu2@citrix.com>
This leaves vring_new_virtqueue alone for compatbility, but it
adds two new improved APIs:
vring_create_virtqueue: Creates a virtqueue backed by automatically
allocated coherent memory. (Some day it this could be extended to
support non-coherent memory, too, if there ends up being a platform
on which it's worthwhile.)
__vring_new_virtqueue: Creates a virtqueue with a manually-specified
layout. This should allow mic_virtio to work much more cleanly.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
virtio_ring currently sends the device (usually a hypervisor)
physical addresses of its I/O buffers. This is okay when DMA
addresses and physical addresses are the same thing, but this isn't
always the case. For example, this never works on Xen guests, and
it is likely to fail if a physical "virtio" device ever ends up
behind an IOMMU or swiotlb.
The immediate use case for me is to enable virtio on Xen guests.
For that to work, we need vring to support DMA address translation
as well as a corresponding change to virtio_pci or to another
driver.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This is a kludge, but no one has come up with a a better idea yet.
We'll introduce DMA API support guarded by vring_use_dma_api().
Eventually we may be able to return true on more and more systems,
and hopefully we can get rid of vring_use_dma_api() entirely some
day.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We need a full barrier after writing out event index, using
virt_store_mb there seems better than open-coding. As usual, we need a
wrapper to account for strong barriers.
It's tempting to use this in vhost as well, for that, we'll
need a variant of smp_store_mb that works on __user pointers.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Improves cacheline transfer flow of available ring header.
Virtqueues are implemented as a pair of rings, one producer->consumer
avail ring and one consumer->producer used ring; preceding the
avail ring in memory are two contiguous u16 fields -- avail->flags
and avail->idx. A producer posts work by writing to avail->idx and
a consumer reads avail->idx.
The flags and idx fields only need to be written by a producer CPU
and only read by a consumer CPU; when the producer and consumer are
running on different CPUs and the virtio_ring code is structured to
only have source writes/sink reads, we can continuously transfer the
avail header cacheline between 'M' states between cores. This flow
optimizes core -> core bandwidth on certain CPUs.
(see: "Software Optimization Guide for AMD Family 15h Processors",
Section 11.6; similar language appears in the 10h guide and should
apply to CPUs w/ exclusive caches, using LLC as a transfer cache)
Unfortunately the existing virtio_ring code issued reads to the
avail->idx and read-modify-writes to avail->flags on the producer.
This change shadows the flags and index fields in producer memory;
the vring code now reads from the shadows and only ever writes to
avail->flags and avail->idx, allowing the cacheline to transfer
core -> core optimally.
In a concurrent version of vring_bench, the time required for
10,000,000 buffer checkout/returns was reduced by ~2% (average
across many runs) on an AMD Piledriver (15h) CPU:
(w/o shadowing):
Performance counter stats for './vring_bench':
5,451,082,016 L1-dcache-loads
...
2.221477739 seconds time elapsed
(w/ shadowing):
Performance counter stats for './vring_bench':
5,405,701,361 L1-dcache-loads
...
2.168405376 seconds time elapsed
The further away (in a NUMA sense) virtio producers and consumers are
from each other, the more we expect to benefit. Physical implementations
of virtio devices and implementations of virtio where the consumer polls
vring avail indexes (vhost) should also benefit.
Signed-off-by: Venkatesh Srinivas <venkateshs@google.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
b92b1b89a3 ("virtio: force vring descriptors to be allocated from
lowmem") tried to exclude highmem pages for descriptors so it cleared
__GFP_HIGHMEM from a given gfp mask. The patch also cleared __GFP_HIGH
which doesn't make much sense for this fix because __GFP_HIGH only
controls access to memory reserves and it doesn't have any influence
on the zone selection. Some of the call paths use GFP_ATOMIC and
dropping __GFP_HIGH will reduce their changes for success because the
lack of access to memory reserves.
Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Reviewed-by: Mel Gorman <mgorman@techsingularity.net>
The virtqueue_add() calls START_USE() upon entry. The virtqueue_kick() is
called if vq->num_added == (1 << 16) - 1 before calling END_USE().
The virtqueue_kick_prepare() called via virtqueue_kick() calls START_USE()
upon entry, and will call panic() if DEBUG is enabled.
Move this virtqueue_kick() call to after END_USE() call.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Most of our code has
struct foo {
}
Fix one instances where ring is inconsistent.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
For virtio-1, we can theoretically have a more complex virtqueue
layout with avail and used buffers not on a contiguous memory area
with the descriptor table. For now, it's fine for a transport driver
to stay with the old layout: It needs, however, a way to access
the locations of the avail/used rings so it can register them with
the host.
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Use virtioXX_to_cpu and friends for access to
all multibyte structures in memory.
Note: this is intentionally mechanical.
A follow-up patch will split long lines etc.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
It seemed like a good idea to use bitmap for features
in struct virtio_device, but it's actually a pain,
and seems to become even more painful when we get more
than 32 feature bits. Just change it to a u32 for now.
Based on patch by Rusty.
Suggested-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
virtqueue_add() populates the virtqueue descriptor table from the sgs
given. If it uses an indirect descriptor table, then it puts a single
descriptor in the descriptor table pointing to the kmalloc'ed indirect
table where the sg is populated.
Previously vring_add_indirect() did the allocation and the simple
linear layout. We replace that with alloc_indirect() which allocates
the indirect table then chains it like the normal descriptor table so
we can reuse the core logic.
This slows down pktgen by less than 1/2 a percent (which uses direct
descriptors), as well as vring_bench, but it's far neater.
vring_bench before:
1061485790-1104800648(1.08254e+09+/-6.6e+06)ns
vring_bench after:
1125610268-1183528965(1.14172e+09+/-8e+06)ns
pktgen before:
787781-796334(793165+/-2.4e+03)pps 365-369(367.5+/-1.2)Mb/sec (365530384-369498976(3.68028e+08+/-1.1e+06)bps) errors: 0
pktgen after:
779988-790404(786391+/-2.5e+03)pps 361-366(364.35+/-1.3)Mb/sec (361914432-366747456(3.64885e+08+/-1.2e+06)bps) errors: 0
Now, if we make force indirect descriptors by turning off any_header_sg
in virtio_net.c:
pktgen before:
713773-721062(718374+/-2.1e+03)pps 331-334(332.95+/-0.92)Mb/sec (331190672-334572768(3.33325e+08+/-9.6e+05)bps) errors: 0
pktgen after:
710542-719195(714898+/-2.4e+03)pps 329-333(331.15+/-1.1)Mb/sec (329691488-333706480(3.31713e+08+/-1.1e+06)bps) errors: 0
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>