- 26 4月, 2013 2 次提交
-
-
由 Liu Yuan 提交于
Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
由 Liu Yuan 提交于
The 'TRIM' command from VM that is to release underlying data storage for better thin-provision is already supported by the Sheepdog. This patch adds the TRIM support at QEMU part. For older Sheepdog that doesn't support it, we return 0(success) to upper layer. Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 22 4月, 2013 2 次提交
-
-
由 Kevin Wolf 提交于
It is unused now in all block drivers. Signed-off-by: NKevin Wolf <kwolf@redhat.com> Reviewed-by: NEric Blake <eblake@redhat.com>
-
由 Kevin Wolf 提交于
This is only to convert the internal interface that is used for passing the "filename" to be parsed, but converting to actual fine grained options is left for another day, as it doesn't look trivial. Signed-off-by: NKevin Wolf <kwolf@redhat.com> Reviewed-by: NEric Blake <eblake@redhat.com>
-
- 15 4月, 2013 1 次提交
-
-
由 Kevin Wolf 提交于
Signed-off-by: NKevin Wolf <kwolf@redhat.com> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 02 4月, 2013 1 次提交
-
-
由 Stefan Hajnoczi 提交于
The fcntl(fd, F_SETFL, O_NONBLOCK) flag is not specific to sockets. Rename to qemu_set_nonblock() just like qemu_set_cloexec(). Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com> Reviewed-by: NEric Blake <eblake@redhat.com> Signed-off-by: NLuiz Capitulino <lcapitulino@redhat.com>
-
- 23 3月, 2013 1 次提交
-
-
由 Kevin Wolf 提交于
The new parameter is unused yet. Signed-off-by: NKevin Wolf <kwolf@redhat.com> Reviewed-by: NEric Blake <eblake@redhat.com>
-
- 19 3月, 2013 1 次提交
-
-
由 Liu Yuan 提交于
Sheepdog (neither quorum nor unsafe mode) will refuse to serve IO requests when number of alive nodes is less than that of copies specified by users. This will return 0x19 to QEMU client which currently doesn't recognize it. This patch adds an error description when QEMU client receives it, other than plainly printing 'Invalid error code' Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com> Reviewed-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 15 3月, 2013 2 次提交
-
-
由 MORITA Kazutaka 提交于
If an io_flush handler is not set, qemu_aio_wait doesn't invoke callbacks. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
由 MORITA Kazutaka 提交于
Using a blocking socket in the coroutine context reduces the chance of switching to other work. This patch makes the sheepdog driver use a non-blocking fd always. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 04 3月, 2013 4 次提交
-
-
由 MORITA Kazutaka 提交于
This patch adds support for a unix domain socket for a connection between qemu and local sheepdog server. You can use the unix domain socket with the following syntax: $ qemu sheepdog+unix:///<vdiname>?socket=<socket path>[#snapid] Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
由 MORITA Kazutaka 提交于
This uses the form "<host>:<port>" for the representation of the sheepdog server to use inet_connect. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
由 MORITA Kazutaka 提交于
The URI syntax is consistent with the NBD and Gluster syntax. The syntax is sheepdog[+tcp]://[host:port]/vdiname[#snapid|#tag] Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
由 MORITA Kazutaka 提交于
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 01 2月, 2013 1 次提交
-
-
由 Liu Yuan 提交于
Sheep daemon needs vdi_id to identify which vdi is closed to release resources such as object cache. Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Reviewed-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 15 1月, 2013 2 次提交
-
-
由 Liu Yuan 提交于
The last two parameters of sd_aio_setup() are never used, so remove them. Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
由 Liu Yuan 提交于
This will reduce sockfds connected to the sheep server to one, which simply the future hacks. Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 14 1月, 2013 1 次提交
-
-
由 Liu Yuan 提交于
Sheepdog supports both writeback/writethrough write but has not yet supported DIRECTIO semantics which bypass the cache completely even if Sheepdog daemon is set up with cache enabled. Suppose cache is enabled on Sheepdog daemon size, the new cache control is cache=writeback # enable the writeback semantics for write cache=writethrough # enable the emulated writethrough semantics for write cache=directsync # disable cache competely Guest WCE toggling on the run time to toggle writeback/writethrough is also supported. Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@gmail.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com> Reviewed-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 02 1月, 2013 2 次提交
-
-
由 Liu Yuan 提交于
Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Reviewed-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
由 Liu Yuan 提交于
For the error case such as SD_RES_NO_SPACE, we shouldn't update the inode bitmap to avoid the scenario that the object is allocated but wasn't created at the server side. This will result in VM's IO error on the failed object. Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Cc: Kevin Wolf <kwolf@redhat.com> Signed-off-by: NLiu Yuan <tailai.ly@taobao.com> Reviewed-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 19 12月, 2012 2 次提交
-
-
由 Paolo Bonzini 提交于
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
-
由 Paolo Bonzini 提交于
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
-
- 15 11月, 2012 1 次提交
-
-
由 Stefan Hajnoczi 提交于
Now that AIOPool no longer keeps a freelist, it isn't really a "pool" anymore. Rename it to AIOCBInfo and make it const since it no longer needs to be modified. Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 12 10月, 2012 1 次提交
-
-
由 MORITA Kazutaka 提交于
This improves readability. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
-
- 05 10月, 2012 1 次提交
-
-
由 Jim Meyering 提交于
* parse_vdiname: Use pstrcpy, not strncpy, when the destination buffer must be NUL-terminated. * sd_open: Likewise, avoid buffer overrun. * do_sd_create: Likewise. Leave the preceding memset, since pstrcpy does not NUL-fill, and filename needs that. * sd_snapshot_create: Add a comment/question. * find_vdi_name: Remove a useless memset. * sd_snapshot_goto: Remove a useless memset. Use pstrcpy to NUL-terminate, because find_vdi_name requires that its vdi arg (filename parameter) be NUL-terminated. It seems ok not to NUL-fill the buffer. Do the same for snapid: remove useless memset-0 (instead, zero tag[0]). Use pstrcpy, not strncpy. * sd_snapshot_list: Use pstrcpy, not strncpy to write into the ->name member. Each must be NUL-terminated. Acked-by: NKevin Wolf <kwolf@redhat.com> Acked-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NJim Meyering <meyering@redhat.com> Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>
-
- 24 9月, 2012 1 次提交
-
-
由 Jeff Cody 提交于
Block drivers should ignore BDRV_O_CACHE_WB in .bdrv_open flags, and in the bs->open_flags. This patch removes the code, leaving the behaviour behind as if BDRV_O_CACHE_WB was set. Signed-off-by: NJeff Cody <jcody@redhat.com> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 12 9月, 2012 1 次提交
-
-
由 MORITA Kazutaka 提交于
This patch sets data to be sent to Sheepdog correctly and fixes savevm and loadvm operations on a Sheepdog image. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 22 8月, 2012 1 次提交
-
-
由 Jim Meyering 提交于
Signed-off-by: NJim Meyering <meyering@redhat.com> Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>
-
- 17 7月, 2012 2 次提交
-
-
由 Christoph Hellwig 提交于
Only buffers that map to unallocated blocks need to be zeroed. Signed-off-by: NChristoph Hellwig <hch@lst.de> Acked-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
由 MORITA Kazutaka 提交于
This reduces some code duplication. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 09 7月, 2012 6 次提交
-
-
由 MORITA Kazutaka 提交于
The pending list can be modified in other coroutine context sd_co_rw_vector, so we need to traverse the list from the first again after we send the pending request. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
由 MORITA Kazutaka 提交于
outstanding_list_head is used for both pending and inflight requests. This patch splits it and improves readability. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
由 MORITA Kazutaka 提交于
This patch increments the pending counter before sending requests, and make sures that aiocb is not freed while sending them. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
由 MORITA Kazutaka 提交于
This removes blocking network I/Os in coroutine context. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
由 MORITA Kazutaka 提交于
Currently, no one reenters the yielded coroutine. This fixes it. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
由 MORITA Kazutaka 提交于
This fixes warnings about dprintf format in debug mode. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 15 6月, 2012 1 次提交
-
-
由 MORITA Kazutaka 提交于
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Reviewed-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 12 6月, 2012 1 次提交
-
-
由 Michael Tokarev 提交于
The same as for non-coroutine versions in previous patches: rename arguments to be more obvious, change type of arguments from int to size_t where appropriate, and use common code for send and receive paths (with one extra argument) since these are exactly the same. Use common iov_send_recv() directly. qemu_co_sendv(), qemu_co_recvv(), and qemu_co_recv() are now trivial #define's merely adding one extra arg. qemu_co_sendv() and qemu_co_recvv() callers are converted to different argument order and extra `iov_cnt' argument. Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>
-
- 30 5月, 2012 1 次提交
-
-
由 MORITA Kazutaka 提交于
bdrv_save_vmstate and bdrv_load_vmstate should return the vmstate size on success, and -errno on error. Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-
- 26 5月, 2012 1 次提交
-
-
由 MORITA Kazutaka 提交于
bdrv_create() is called in coroutine context now, so we cannot use more stack than 1 MB in the function if we use ucontext coroutine. This patch allocates BDRVSheepdogState, whose size is 4 MB, on the heap in sd_create(). Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp> Signed-off-by: NKevin Wolf <kwolf@redhat.com>
-