- 15 Dec, 2016 40 commits
-
-
Jason Wang authored
When event index was enabled, we need to fetch used event from userspace memory each time. This userspace fetch (with memory barrier) could be saved sometime when 1) caching used event and 2) if used event is ahead of new and old to new updating does not cross it, we're sure there's no need to notify guest. This will be useful for heavy tx load e.g guest pktgen test with Linux driver shows ~3.5% improvement. Signed-off-by:
Jason Wang <jasowang@redhat.com> Signed-off-by:
Michael S. Tsirkin <mst@redhat.com>
-
Gao feng authored
Multi vsocks may setup the same cid at the same time. Signed-off-by:
Gao feng <omarapazanadi@gmail.com> Signed-off-by:
Stefan Hajnoczi <stefanha@redhat.com>
-
Christoph Hellwig authored
There is basically no shared logic between the INTx and MSI-X case in vp_try_to_find_vqs, so split the function into two and clean them up a little bit. Also remove the fairly pointless vp_request_intx wrapper while we're at it. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
vp_request_msix_vectors is only called by vp_try_to_find_vqs, which already calls vp_free_vectors through vp_del_vqs in the failure case. Signed-off-by:
-
Christoph Hellwig authored
This avoids the separate allocation for the msix_entries structures, and instead allows us to use pci_irq_vector to find a given IRQ vector. Signed-off-by:
-
Michael S. Tsirkin authored
We call del_vqs twice when request_irq fails, this makes no sense. Signed-off-by: -
Michael S. Tsirkin authored
These fields are 64 bit, using le32_to_cpu and friends on these will not do the right thing. Fix this up. Cc: stable@vger.kernel.org Signed-off-by: -
Michael S. Tsirkin authored
virtio_transport_alloc_pkt is only used locally, make it static. Signed-off-by: -
Michael S. Tsirkin authored
guest cid is read from config space, therefore it's in little endian format and is treated as such, annotate it accordingly. Signed-off-by: -
Michael S. Tsirkin authored
Several vhost functions were missing __user annotations on pointers, causing sparse warnings. Fix this up. sparse also warns about vhost_process_iotlb_msg which is local and should be static. Fix that up as well. Signed-off-by: -
Michael S. Tsirkin authored
vhost_umem_interval_tree is only used locally within vhost.c, mark it static. As some functions generated go unused, this triggers warnings unless we also mark it inline. Signed-off-by: -
Michael S. Tsirkin authored
virtio_gpu_queue_ctrl_buffer_locked is called with ctrlq.qlock taken, it releases and acquires this lock. This causes a sparse warning. Add appropriate annotations for sparse context checking. Signed-off-by: -
Michael S. Tsirkin authored
When virtio_gpu_free_vbufs exits due to list empty, it does not drop the free_vbufs lock that it took. list empty is not expected to happen anyway, but it can't hurt to fix this and drop the lock. Signed-off-by: -
Michael S. Tsirkin authored
virtio_gpu_cmd_transfer_to_host_2d expects x and y parameters in LE, but virtio_gpu_primary_plane_update passes in the CPU format instead. Signed-off-by: -
Michael S. Tsirkin authored
struct ports_device includes a config field including the whole virtio_console_config, but only max_nr_ports in there is ever updated or used. The rest is unused and in fact does not even mirror the device config. Drop everything except max_nr_ports, saving some memory. Signed-off-by:
-
Gonglei authored
# 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:
-
Gonglei authored
drivers/virtio/virtio_pci_modern.c:66:40: warning: incorrect type in argument 2 (different base types) drivers/virtio/virtio_pci_modern.c:66:40: expected unsigned int [noderef] [usertype] <asn:2>*addr drivers/virtio/virtio_pci_modern.c:66:40: got restricted __le32 [noderef] [usertype] <asn:2>*lo drivers/virtio/virtio_pci_modern.c:67:33: warning: incorrect type in argument 2 (different base types) drivers/virtio/virtio_pci_modern.c:67:33: expected unsigned int [noderef] [usertype] <asn:2>*addr drivers/virtio/virtio_pci_modern.c:67:33: got restricted __le32 [noderef] [usertype] <asn:2>*hi drivers/virtio/virtio_pci_modern.c:150:32: warning: incorrect type in argument 2 (different base types) drivers/virtio/virtio_pci_modern.c:150:32: expected unsigned int [noderef] [usertype] <asn:2>*addr drivers/virtio/virtio_pci_modern.c:150:32: got restricted __le32 [noderef] <asn:2>*<noident> drivers/virtio/virtio_pci_modern.c:151:39: warning: incorrect type in argument 1 (different base types) drivers/virtio/virtio_pci_modern.c:151:39: expected unsigned int [noderef] [usertype] <asn:2>*addr drivers/virtio/virtio_pci_modern.c:151:39: got restricted __le32 [noderef] <asn:2>*<noident> drivers/virtio/virtio_pci_modern.c:152:32: warning: incorrect type in argument 2 (different base types) drivers/virtio/virtio_pci_modern.c:152:32: expected unsigned int [noderef] [usertype] <asn:2>*addr Signed-off-by:
-
Linus Torvalds authored
Merge more updates from Andrew Morton: - a few misc things - kexec updates - DMA-mapping updates to better support networking DMA operations - IPC updates - various MM changes to improve DAX fault handling - lots of radix-tree changes, mainly to the test suite. All leading up to reimplementing the IDA/IDR code to be a wrapper layer over the radix-tree. However the final trigger-pulling patch is held off for 4.11. * emailed patches from Andrew Morton <akpm@linux-foundation.org>: (114 commits) radix tree test suite: delete unused rcupdate.c radix tree test suite: add new tag check radix-tree: ensure counts are initialised radix tree test suite: cache recently freed objects radix tree test suite: add some more functionality idr: reduce the number of bits per level from 8 to 6 rxrpc: abstract away knowledge of IDR internals tpm: use idr_find(), not idr_find_slowpath() idr: add ida_is_empty radix tree test suite: check multiorder iteration radix-tree: fix replacement for multiorder entries radix-tree: add radix_tree_split_preload() radix-tree: add radix_tree_split radix-tree: add radix_tree_join radix-tree: delete radix_tree_range_tag_if_tagged() radix-tree: delete radix_tree_locate_item() radix-tree: improve multiorder iterators btrfs: fix race in btrfs_free_dummy_fs_info() radix-tree: improve dump output radix-tree: make radix_tree_find_next_bit more useful ...
-
git://git.kernel.dk/linux-blockLinus Torvalds authored
Pull block IO fixes from Jens Axboe: "A few fixes that I collected as post-merge. I was going to wait a bit with sending this out, but the O_DIRECT fix should really go in sooner rather than later" * 'for-linus' of git://git.kernel.dk/linux-block: blk-mq: Fix failed allocation path when mapping queues blk-mq: Avoid memory reclaim when remapping queues block_dev: don't update file access position for sync direct IO nvme/pci: Log PCI_STATUS when the controller dies block_dev: don't test bdev->bd_contains when it is not stable
-
git://git.kernel.dk/linux-blockLinus Torvalds authored
Pull fs meta data unmap optimization from Jens Axboe: "A series from Jan Kara, providing a more efficient way for unmapping meta data from in the buffer cache than doing it block-by-block. Provide a general helper that existing callers can use" * 'for-4.10/fs-unmap' of git://git.kernel.dk/linux-block: fs: Remove unmap_underlying_metadata fs: Add helper to clean bdev aliases under a bh and use it ext2: Use clean_bdev_aliases() instead of iteration ext4: Use clean_bdev_aliases() instead of iteration direct-io: Use clean_bdev_aliases() instead of handmade iteration fs: Provide function to unmap metadata for a range of blocks
-
Linus Torvalds authored
Jaegeuk Kim reports that the debian kernel package build gets confused by the lack of Documentation/Changes file. We also refer to that path name in ver_linux and various how-to files and Kconfig files. The file got renamed away in commit 186128f7 ("docs-rst: add documents to development-process"), and as Jaegeuk Kim points out, the commit message for that change says "use symlinks instead of renames", but then the commit itself actually does renames after all. Maybe we should do the other files too, but for now this just adds the minimal symlink back to the historical name, so that people looking for Documentation/Changes will actually find what they are looking for, and the debian scripts continue to work. Reported-by:
Jaegeuk Kim <jaegeuk@kernel.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Mauro Carvalho Chehab <mchehab@s-opensource.com> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Matthew Wilcox authored
This file was used to implement call_rcu() before liburcu implemented that function. It hasn't even been compiled since before the test suite was added to the kernel. Remove it to reduce confusion. Link: http://lkml.kernel.org/r/1481667692-14500-5-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
Matthew Wilcox <mawilcox@microsoft.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Konstantin Khlebnikov <koct9i@gmail.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Matthew Wilcox authored
We have a check that setting a tag on a single entry at root succeeds, but we were missing a check that clearing a tag on that same entry also succeeds. Link: http://lkml.kernel.org/r/1481667692-14500-4-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
radix_tree_join() was freeing nodes with a non-zero ->exceptional count, and radix_tree_split() wasn't zeroing ->exceptional when it allocated the new node. Fix this by making all callers of radix_tree_node_alloc() pass in the new counts (and some other always-initialised fields), which will prevent the problem recurring if in future we decide to do something similar. Link: http://lkml.kernel.org/r/1481667692-14500-3-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
The kmem_cache_alloc implementation simply allocates new memory from malloc() and calls the ctor, which zeroes out the entire object. This means it cannot spot bugs where the object isn't properly reinitialised before being freed. Add a small (11 objects) cache before freeing objects back to malloc. This is enough to let us write a test to catch it, although the memory allocator is now aware of the structure of the radix tree node, since it chains free objects through ->private_data (like the percpu cache does). Link: http://lkml.kernel.org/r/1481667692-14500-2-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
IDR needs more functionality from the kernel: kmalloc()/kfree(), and xchg(). Link: http://lkml.kernel.org/r/1480369871-5271-67-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
Matthew Wilcox <willy@linux.intel.com> Tested-by:
Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Konstantin Khlebnikov <koct9i@gmail.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Cc: Matthew Wilcox <mawilcox@microsoft.com> Signed-off-by:
-
Matthew Wilcox authored
In preparation for merging the IDR and radix tree, reduce the fanout at each level from 256 to 64. If this causes a performance problem then a bisect will point to this commit, and we'll have a better idea about what we might do to fix it. Link: http://lkml.kernel.org/r/1480369871-5271-66-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
Add idr_get_cursor() / idr_set_cursor() APIs, and remove the reference to IDR_SIZE. Link: http://lkml.kernel.org/r/1480369871-5271-65-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
David Howells <dhowells@redhat.com> Tested-by:
-
Matthew Wilcox authored
idr_find_slowpath() is not intended to be part of the public API, it's an implementation detail. There's no reason to skip straight to the slowpath here. Link: http://lkml.kernel.org/r/1480369871-5271-64-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
Two of the USB Gadgets were poking around in the internals of struct ida in order to determine if it is empty. Add the appropriate abstraction. Link: http://lkml.kernel.org/r/1480369871-5271-63-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
Konstantin Khlebnikov <koct9i@gmail.com> Tested-by:
-
Matthew Wilcox authored
The random iteration test only inserts order-0 entries currently. Update it to insert entries of order between 7 and 0. Also make the maximum index configurable, make some variables static, make the test duration variable, remove some useless spinning, and add a fifth thread which calls tag_tagged_items(). Link: http://lkml.kernel.org/r/1480369871-5271-62-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
When replacing an entry with NULL, we need to delete any sibling entries. Also account deleting exceptional entries properly. Also fix a bug with radix_tree_iter_replace() where we would fail to remove entirely freed nodes. Also fix accounting bug when switching between normal and exceptional entries with replace_slot. Also add testcases for all these bugs. Link: http://lkml.kernel.org/r/1480369871-5271-61-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
Calculate how many nodes we need to allocate to split an old_order entry into multiple entries, each of size new_order. The test suite checks that we allocated exactly the right number of nodes; neither too many (checked by rtp->nr == 0), nor too few (checked by comparing nr_allocated before and after the call to radix_tree_split()). Link: http://lkml.kernel.org/r/1480369871-5271-60-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
This new function splits a larger multiorder entry into smaller entries (potentially multi-order entries). These entries are initialised to RADIX_TREE_RETRY to ensure that RCU walkers who see this state aren't confused. The caller should then call radix_tree_for_each_slot() and radix_tree_replace_slot() in order to turn these retry entries into the intended new entries. Tags are replicated from the original multiorder entry into each new entry. Link: http://lkml.kernel.org/r/1480369871-5271-59-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
This new function allows for the replacement of many smaller entries in the radix tree with one larger multiorder entry. From the point of view of an RCU walker, they may see a mixture of the smaller entries and the large entry during the same walk, but they will never see NULL for an index which was populated before the join. Link: http://lkml.kernel.org/r/1480369871-5271-58-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
This is an exceptionally complicated function with just one caller (tag_pages_for_writeback). We devote a large portion of the runtime of the test suite to testing this one function which has one caller. By introducing the new function radix_tree_iter_tag_set(), we can eliminate all of the complexity while keeping the performance. The caller can now use a fairly standard radix_tree_for_each() loop, and it doesn't need to worry about tricksy things like 'start' wrapping. The test suite continues to spend a large amount of time investigating this function, but now it's testing the underlying primitives such as radix_tree_iter_resume() and the radix_tree_for_each_tagged() iterator which are also used by other parts of the kernel. Link: http://lkml.kernel.org/r/1480369871-5271-57-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
Matthew Wilcox <willy@infradead.org> Tested-by:
-
Matthew Wilcox authored
This rather complicated function can be better implemented as an iterator. It has only one caller, so move the functionality to the only place that needs it. Update the test suite to follow the same pattern. Link: http://lkml.kernel.org/r/1480369871-5271-56-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
This fixes several interlinked problems with the iterators in the presence of multiorder entries. 1. radix_tree_iter_next() would only advance by one slot, which would result in the iterators returning the same entry more than once if there were sibling entries. 2. radix_tree_next_slot() could return an internal pointer instead of a user pointer if a tagged multiorder entry was immediately followed by an entry of lower order. 3. radix_tree_next_slot() expanded to a lot more code than it used to when multiorder support was compiled in. And I wasn't comfortable with entry_to_node() being in a header file. Fixing radix_tree_iter_next() for the presence of sibling entries necessarily involves examining the contents of the radix tree, so we now need to pass 'slot' to radix_tree_iter_next(), and we need to change the calling convention so it is called *before* dropping the lock which protects the tree. Also rename it to radix_tree_iter_resume(), as some people thought it was necessary to call radix_tree_iter_next() each time around the loop. radix_tree_next_slot() becomes closer to how it looked before multiorder support was introduced. It only checks to see if the next entry in the chunk is a sibling entry or a pointer to a node; this should be rare enough that handling this case out of line is not a performance impact (and such impact is amortised by the fact that the entry we just processed was a multiorder entry). Also, radix_tree_next_slot() used to force a new chunk lookup for untagged entries, which is more expensive than the out of line sibling entry skipping. Link: http://lkml.kernel.org/r/1480369871-5271-55-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
Matthew Wilcox authored
We drop the lock which protects the radix tree, so we must call radix_tree_iter_next() in order to avoid a modification to the tree invalidating the iterator state. Link: http://lkml.kernel.org/r/1480369871-5271-54-git-send-email-mawilcox@linuxonhyperv.comSigned-off-by:
-
