linux/drivers/virtio, branch v4.4

virtio_ring: shadow available ring flags & index

2015-12-07T15:28:11Z

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 Signed-off-by: Michael S. Tsirkin

virtio: Do not drop __GFP_HIGH in alloc_indirect

2015-12-07T15:28:11Z

b92b1b89a33c ("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 Signed-off-by: Michael S. Tsirkin Acked-by: Will Deacon Reviewed-by: Mel Gorman

virtio: fix memory leak of virtio ida cache layers

2015-12-07T15:28:01Z

The virtio core uses a static ida named virtio_index_ida for assigning index numbers to virtio devices during registration. The ida core may allocate some internal idr cache layers and an ida bitmap upon any ida allocation, and all these layers are truely freed only upon the ida destruction. The virtio_index_ida is not destroyed at present, leading to a memory leak when using the virtio core as a module and atleast one virtio device is registered and unregistered. Fix this by invoking ida_destroy() in the virtio core module exit. Cc: stable@vger.kernel.org Signed-off-by: Suman Anna Signed-off-by: Michael S. Tsirkin

virtio_balloon: do not change memory amount visible via /proc/meminfo

2015-09-08T10:32:11Z

Balloon device is frequently used as a mean of cooperative memory control in between guest and host to manage memory overcommitment. This is the typical case for any hosting workload when KVM guest is provided for end-user. Though there is a problem in this setup. The end-user and hosting provider have signed SLA agreement in which some amount of memory is guaranted for the guest. The good thing is that this memory will be given to the guest when the guest will really need it (f.e. with OOM in guest and with VIRTIO_BALLOON_F_DEFLATE_ON_OOM configuration flag set). The bad thing is that end-user does not know this. Balloon by default reduce the amount of memory exposed to the end-user each time when the page is stolen from guest or returned back by using adjust_managed_page_count and thus /proc/meminfo shows reduced amount of memory. Fortunately the solution is simple, we should just avoid to call adjust_managed_page_count with VIRTIO_BALLOON_F_DEFLATE_ON_OOM set. Signed-off-by: Denis V. Lunev CC: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

virtio_ballon: change stub of release_pages_by_pfn

2015-09-08T10:31:51Z

and rename it to release_pages_balloon. The function originally takes arrays of pfns and now it takes pointer to struct virtio_ballon. This change is necessary to conditionally call adjust_managed_page_count in the next patch. Signed-off-by: Denis V. Lunev CC: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

virtio_mmio: add ACPI probing

2015-09-08T10:30:28Z

Added the match table and pointers for ACPI probing to the driver. This uses the same identifier for virt devices as being used for qemu ARM64 ACPI support. http://git.linaro.org/people/shannon.zhao/qemu.git/commit/d0bf1955a3ecbab4b51d46f8c5dda02b7e14a17e Signed-off-by: Graeme Gregory Signed-off-by: Michael S. Tsirkin

virtio-input: reset device and detach unused during remove

2015-08-06T07:40:35Z

Spec requires a device reset during cleanup, so do it and avoid warn in virtio core. And detach unused buffers to avoid memory leak. Signed-off-by: Jason Wang Signed-off-by: Michael S. Tsirkin

Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost

2015-07-03T23:02:25Z

Pull virtio/vhost cross endian support from Michael Tsirkin: "I have just queued some more bugfix patches today but none fix regressions and none are related to these ones, so it looks like a good time for a merge for -rc1. The motivation for this is support for legacy BE guests on the new LE hosts. There are two redeeming properties that made me merge this: - It's a trivial amount of code: since we wrap host/guest accesses anyway, almost all of it is well hidden from drivers. - Sane platforms would never set flags like VHOST_CROSS_ENDIAN_LEGACY, and when it's clear, there's zero overhead (as some point it was tested by compiling with and without the patches, got the same stripped binary). Maybe we could create a Kconfig symbol to enforce the second point: prevent people from enabling it eg on x86. I will look into this" * tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost: virtio-pci: alloc only resources actually used. macvtap/tun: cross-endian support for little-endian hosts vhost: cross-endian support for legacy devices virtio: add explicit big-endian support to memory accessors vhost: introduce vhost_is_little_endian() helper vringh: introduce vringh_is_little_endian() helper macvtap: introduce macvtap_is_little_endian() helper tun: add tun_is_little_endian() helper virtio: introduce virtio_is_little_endian() helper

Merge tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux

2015-07-01T17:49:25Z

Pull module updates from Rusty Russell: "Main excitement here is Peter Zijlstra's lockless rbtree optimization to speed module address lookup. He found some abusers of the module lock doing that too. A little bit of parameter work here too; including Dan Streetman's breaking up the big param mutex so writing a parameter can load another module (yeah, really). Unfortunately that broke the usual suspects, !CONFIG_MODULES and !CONFIG_SYSFS, so those fixes were appended too" * tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux: (26 commits) modules: only use mod->param_lock if CONFIG_MODULES param: fix module param locks when !CONFIG_SYSFS. rcu: merge fix for Convert ACCESS_ONCE() to READ_ONCE() and WRITE_ONCE() module: add per-module param_lock module: make perm const params: suppress unused variable error, warn once just in case code changes. modules: clarify CONFIG_MODULE_COMPRESS help, suggest 'N'. kernel/module.c: avoid ifdefs for sig_enforce declaration kernel/workqueue.c: remove ifdefs over wq_power_efficient kernel/params.c: export param_ops_bool_enable_only kernel/params.c: generalize bool_enable_only kernel/module.c: use generic module param operaters for sig_enforce kernel/params: constify struct kernel_param_ops uses sysfs: tightened sysfs permission checks module: Rework module_addr_{min,max} module: Use __module_address() for module_address_lookup() module: Make the mod_tree stuff conditional on PERF_EVENTS || TRACING module: Optimize __module_address() using a latched RB-tree rbtree: Implement generic latch_tree seqlock: Introduce raw_read_seqcount_latch() ...

virtio-pci: alloc only resources actually used.

2015-06-24T06:15:09Z

Move resource allocation from common code to legacy and modern code. Only request resources actually used, i.e. bar0 in legacy mode and the bar(s) specified by capabilities in modern mode. Signed-off-by: Gerd Hoffmann Signed-off-by: Michael S. Tsirkin