aboutsummaryrefslogtreecommitdiff
path: root/block
AgeCommit message (Collapse)Author
2015-10-02block/raw-posix: Open file descriptor O_RDWR to work around glibc ↵Richard W.M. Jones
posix_fallocate emulation issue. https://bugzilla.redhat.com/show_bug.cgi?id=1265196 The following command fails on an NFS mountpoint: $ qemu-img create -f qcow2 -o preallocation=falloc disk.img 262144 Formatting 'disk.img', fmt=qcow2 size=262144 encryption=off cluster_size=65536 preallocation='falloc' lazy_refcounts=off qemu-img: disk.img: Could not preallocate data for the new file: Bad file descriptor The reason turns out to be because NFS doesn't support the posix_fallocate call. glibc emulates it instead. However glibc's emulation involves using the pread(2) syscall. The pread syscall fails with EBADF if the file descriptor is opened without the read open-flag (ie. open (..., O_WRONLY)). I contacted glibc upstream about this, and their response is here: https://bugzilla.redhat.com/show_bug.cgi?id=1265196#c9 There are two possible fixes: Use Linux fallocate directly, or (this fix) work around the problem in qemu by opening the file with O_RDWR instead of O_WRONLY. Signed-off-by: Richard W.M. Jones <rjones@redhat.com> BZ: https://bugzilla.redhat.com/show_bug.cgi?id=1265196 Reviewed-by: Jeff Cody <jcody@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-10-02raw-win32: Fix write request error handlingKevin Wolf
aio_worker() wrote the return code to the wrong variable. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Tested-by: Guangmu Zhu <guangmuzhu@gmail.com> Reviewed-by: Eric Blake <eblake@redhat.com>
2015-10-01block: mirror - fix full sync mode when target does not support zero initJeff Cody
During mirror, if the target device does not support zero init, a mirror may result in a corrupted image for sync="full" mode. This is due to how the initial dirty bitmap is set up prior to copying data - we did not mark sectors as dirty that are unallocated. This means those unallocated sectors are skipped over on the target, and for a device without zero init, invalid data may reside in those holes. If both of the following conditions are true, then we will explicitly mark all sectors as dirty: 1.) sync = "full" 2.) bdrv_has_zero_init(target) == false If the target does support zero init, but a target image is passed in with data already present (i.e. an "existing" image), it is assumed the data present in the existing image is valid data for those sectors. Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Message-id: 91ed4bc5bda7e2b09eb508b07c83f4071fe0b3c9.1443705220.git.jcody@redhat.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into stagingPeter Maydell
* First batch of MAINTAINERS updates * IOAPIC fixes (to pass kvm-unit-tests with -machine kernel_irqchip=off) * NBD API upgrades from Daniel * strtosz fixes from Marc-André * improved support for readonly=on on scsi-generic devices * new "info ioapic" and "info lapic" monitor commands * Peter Crosthwaite's ELF_MACHINE cleanups * docs patches from Thomas and Daniel # gpg: Signature made Fri 25 Sep 2015 11:20:52 BST using RSA key ID 78C7AE83 # gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" # gpg: aka "Paolo Bonzini <pbonzini@redhat.com>" * remotes/bonzini/tags/for-upstream: (52 commits) doc: Refresh URLs in the qemu-tech documentation docs: describe the QEMU build system structure / design typedef: add typedef for QemuOpts i386: interrupt poll processing i386: partial revert of interrupt poll fix ppc: Rename ELF_MACHINE to be PPC specific i386: Rename ELF_MACHINE to be x86 specific alpha: Remove ELF_MACHINE from cpu.h mips: Remove ELF_MACHINE from cpu.h sparc: Remove ELF_MACHINE from cpu.h s390: Remove ELF_MACHINE from cpu.h sh4: Remove ELF_MACHINE from cpu.h xtensa: Remove ELF_MACHINE from cpu.h tricore: Remove ELF_MACHINE from cpu.h or32: Remove ELF_MACHINE from cpu.h lm32: Remove ELF_MACHINE from cpu.h unicore: Remove ELF_MACHINE from cpu.h moxie: Remove ELF_MACHINE from cpu.h cris: Remove ELF_MACHINE from cpu.h m68k: Remove ELF_MACHINE from cpu.h ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-09-25sheepdog: refine discard supportHitoshi Mitake
This patch refines discard support of the sheepdog driver. The existing discard mechanism was implemented on SD_OP_DISCARD_OBJ, which was introduced before fine grained reference counting on newer sheepdog. It doesn't care about relations of snapshots and clones and discards objects unconditionally. With this patch, the driver just updates an inode object for updating reference. Removing the object is done in sheep process side. Cc: Teruaki Ishizaki <ishizaki.teruaki@lab.ntt.co.jp> Cc: Vasiliy Tolstov <v.tolstov@selfip.ru> Cc: Jeff Cody <jcody@redhat.com> Signed-off-by: Hitoshi Mitake <mitake.hitoshi@lab.ntt.co.jp> Tested-by: Vasiliy Tolstov <v.tolstov@selfip.ru> Message-id: 1441076590-8015-3-git-send-email-mitake.hitoshi@lab.ntt.co.jp Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25sheepdog: use per AIOCB dirty indexes for non overlapping requestsHitoshi Mitake
In the commit 96b14ff85acf, requests for overlapping areas are serialized. However, it cannot handle a case of non overlapping requests. In such a case, min_dirty_data_idx and max_dirty_data_idx can be overwritten by the requests and invalid inode update can happen e.g. a case like create(1, 2) and create(3, 4) are issued in parallel. This patch lets SheepdogAIOCB have dirty data indexes instead of BDRVSheepdogState for avoiding the above situation. This patch also does trivial renaming for better description: overwrapping -> overlapping Cc: Teruaki Ishizaki <ishizaki.teruaki@lab.ntt.co.jp> Cc: Vasiliy Tolstov <v.tolstov@selfip.ru> Cc: Jeff Cody <jcody@redhat.com> Signed-off-by: Hitoshi Mitake <mitake.hitoshi@lab.ntt.co.jp> Tested-by: Vasiliy Tolstov <v.tolstov@selfip.ru> Message-id: 1441076590-8015-2-git-send-email-mitake.hitoshi@lab.ntt.co.jp Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25Backup: don't do copy-on-read in before_write_notifierWen Congyang
We will copy data in before_write_notifier to do backup. It is a nested I/O request, so we cannot do copy-on-read. The steps to reproduce it: 1. -drive copy-on-read=on,... // qemu option 2. drive_backup -f disk0 /path_to_backup.img // monitor command Signed-off-by: Wen Congyang <wency@cn.fujitsu.com> Tested-by: Jeff Cody <jcody@redhat.com> Message-id: 1441682913-14320-3-git-send-email-wency@cn.fujitsu.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25block: Introduce a new API bdrv_co_no_copy_on_readv()Wen Congyang
In some cases, we need to disable copy-on-read, and just read the data. Signed-off-by: Wen Congyang <wency@cn.fujitsu.com> Message-id: 1441682913-14320-2-git-send-email-wency@cn.fujitsu.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25sheepdog: add reopen supportLiu Yuan
With reopen supported, block-commit (and offline commit) is now supported for image files whose base image uses the Sheepdog protocol driver. Cc: qemu-devel@nongnu.org Cc: Jeff Cody <jcody@redhat.com> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Liu Yuan <liuyuan@cmss.chinamobile.com> Message-id: 1440730438-24676-1-git-send-email-namei.unix@gmail.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25block/nfs: cache allocated filesize for read-only filesPeter Lieven
If the file is readonly its not expected to grow so save the blocking call to nfs_fstat_async and use the value saved at connection time. Also important the monitor (and thus the main loop) will not hang if block device info is queried and the NFS share is unresponsive. Signed-off-by: Peter Lieven <pl@kamp.de> Reviewed-by: Jeff Cody <jcody@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-id: 1440671441-7978-1-git-send-email-pl@kamp.de Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25block/nfs: fix calculation of allocated file sizePeter Lieven
st.st_blocks is always counted in 512 byte units. Do not use st.st_blksize as multiplicator which may be larger. Cc: qemu-stable@nongnu.org Signed-off-by: Peter Lieven <pl@kamp.de> Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Jeff Cody <jcody@redhat.com> Message-id: 1440067607-14547-1-git-send-email-pl@kamp.de Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-09-25nbd: convert to use the QAPI SocketAddress objectDaniel P. Berrange
The nbd block driver currently uses a QemuOpts object when setting up sockets. Switch it over to use the QAPI SocketAddress object instead. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-Id: <1442411543-28513-2-git-send-email-berrange@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-09-14Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into stagingPeter Maydell
Block layer patches (v2) # gpg: Signature made Mon 14 Sep 2015 15:56:54 BST using RSA key ID C88F2FD6 # gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" * remotes/kevin/tags/for-upstream: (23 commits) qcow2: Make qcow2_alloc_bytes() more explicit vmdk: Fix next_cluster_sector for compressed write iotests: Add test for checking large image files qcow2: Make size_to_clusters() return uint64_t qemu-iotests: More qcow2 reopen tests qemu-iotests: Reopen qcow2 with lazy-refcounts change qcow2: Support updating driver-specific options in reopen qcow2: Make qcow2_update_options() suitable for transactions qcow2: Fix memory leak in qcow2_update_options() error path qcow2: Leave s unchanged on qcow2_update_options() failure qcow2: Move rest of option handling to qcow2_update_options() qcow2: Move qcow2_update_options() call up qcow2: Factor out qcow2_update_options() qcow2: Improve error message qemu-io: Add command 'reopen' qemu-io: Remove duplicate 'open' error message block: Allow specifying driver-specific options to reopen qcow2: Rename BDRVQcowState to BDRVQcow2State block: Drop bdrv_find_whitelisted_format() block: Drop drv parameter from bdrv_fill_options() ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-09-14Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into stagingPeter Maydell
* Support for jemalloc * qemu_mutex_lock_iothread "No such process" fix * cutils: qemu_strto* wrappers * iohandler.c simplification * Many other fixes and misc patches. And some MTTCG work (with Emilio's fixes squashed): * Signal-free TCG kick * Removing spinlock in favor of QemuMutex * User-mode emulation multi-threading fixes/docs # gpg: Signature made Thu 10 Sep 2015 09:03:07 BST using RSA key ID 78C7AE83 # gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" # gpg: aka "Paolo Bonzini <pbonzini@redhat.com>" * remotes/bonzini/tags/for-upstream: (44 commits) cutils: work around platform differences in strto{l,ul,ll,ull} cpu-exec: fix lock hierarchy for user-mode emulation exec: make mmap_lock/mmap_unlock globally available tcg: comment on which functions have to be called with mmap_lock held tcg: add memory barriers in page_find_alloc accesses remove unused spinlock. replace spinlock by QemuMutex. cpus: remove tcg_halt_cond and tcg_cpu_thread globals cpus: protect work list with work_mutex scripts/dump-guest-memory.py: fix after RAMBlock change configure: Add support for jemalloc add macro file for coccinelle configure: factor out adding disas configure vhost-scsi: fix wrong vhost-scsi firmware path checkpatch: remove tests that are not relevant outside the kernel checkpatch: adapt some tests to QEMU CODING_STYLE: update mixed declaration rules qmp: Add example usage of strto*l() qemu wrapper cutils: Add qemu_strtoull() wrapper cutils: Add qemu_strtoll() wrapper ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-09-14qcow2: Make qcow2_alloc_bytes() more explicitMax Reitz
In case of -EAGAIN returned by update_refcount(), we should discard the cluster offset we were trying to allocate and request a new one, because in theory that old offset might now be taken by a refcount block. In practice, this was not the case due to update_refcount() generally returning strictly monotonic increasing cluster offsets. However, this behavior is not set in stone, and it is also not obvious when looking at qcow2_alloc_bytes() alone, so we should not rely on it. Reported-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-09-14vmdk: Fix next_cluster_sector for compressed writeRadoslav Gerganov
When the VMDK is streamOptimized (or compressed), the next_cluster_sector must not be incremented by a fixed number of sectors. Instead of this, it must be rounded up to the next consecutive sector. Fixing this results in much smaller compressed images. Signed-off-by: Radoslav Gerganov <rgerganov@vmware.com> Reviewed-by: Fam Zheng <famz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-09-14qcow2: Make size_to_clusters() return uint64_tMax Reitz
Sadly, some images may have more clusters than what can be represented using a plain int. We should be prepared for that case (in qcow2_check_refcounts() we actually were trying to catch that case, but since size_to_clusters() truncated the returned value, that check never did anything useful). Cc: qemu-stable <qemu-stable@nongnu.org> Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-09-14qcow2: Support updating driver-specific options in reopenKevin Wolf
For updating the cache sizes, disabling lazy refcounts and updating the clean_cache_timer there is a bit more to do than just changing the variables, but otherwise we're all set for changing options during bdrv_reopen(). Just implement the missing pieces and hook the functions up in bdrv_reopen(). Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Make qcow2_update_options() suitable for transactionsKevin Wolf
Before we can allow updating options at runtime with bdrv_reopen(), we need to split the function into prepare/commit/abort parts. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Fix memory leak in qcow2_update_options() error pathKevin Wolf
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Leave s unchanged on qcow2_update_options() failureKevin Wolf
On return, either all new options should be applied to BDRVQcowState (on success), or all of the old settings should be preserved (on failure). Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Move rest of option handling to qcow2_update_options()Kevin Wolf
With this commit, the handling of driver-specific options in qcow2_open() is completely separated out into qcow2_update_options(). Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Move qcow2_update_options() call upKevin Wolf
qcow2_update_options() only updates some variables in BDRVQcowState and doesn't really depend on other parts of it being initialised yet, so it can be moved so that it immediately follows the other half of option handling code in qcow2_open(). Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Factor out qcow2_update_options()Kevin Wolf
Eventually we want to be able to change options at runtime. As a first step towards that goal, separate some option handling code from the general initialisation code in qcow2_open(). Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Improve error messageKevin Wolf
Eric says that "any" sounds better than "either", and my non-native feeling says the same, so let's change it. Suggested-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14block: Allow specifying driver-specific options to reopenKevin Wolf
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
2015-09-14qcow2: Rename BDRVQcowState to BDRVQcow2StateKevin Wolf
BDRVQcowState is already used by qcow1, and gdb is always confused which one to use. Rename the qcow2 one so they can be distinguished. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Alberto Garcia <berto@igalia.com>
2015-09-14block: Drop drv parameter from bdrv_open()Max Reitz
Now that this parameter is effectively unused, we can drop it and just pass NULL on to bdrv_open_inherit(). Signed-off-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-09-14block: Always pass NULL as drv for bdrv_open()Max Reitz
Change all callers of bdrv_open() to pass the driver name in the options QDict instead of passing its BlockDriver pointer. Signed-off-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-09-11maint: remove / fix many doubled wordsDaniel P. Berrange
Many source files have doubled words (eg "the the", "to to", and so on). Most of these can simply be removed, but a couple were actual mis-spellings (eg "to to" instead of "to do"). There was even one triple word score "to to to" :-) Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
2015-09-11maint: remove double semicolons in many filesDaniel P. Berrange
A number of source files have statements accidentally terminated by a double semicolon - eg 'foo = bar;;'. This is harmless but a mistake none the less. The tcg/ia64/tcg-target.c file is whitelisted because it has valid use of ';;' in a comment containing assembly code. Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
2015-09-07block/iscsi: validate block size returned from targetPeter Lieven
It has been reported that at least tgtd returns a block size of 0 for LUN 0. To avoid running into divide by zero later on and protect against other problematic block sizes validate the block size right at connection time. Cc: qemu-stable@nongnu.org Reported-by: Andrey Korolyov <andrey@xdel.ru> Signed-off-by: Peter Lieven <pl@kamp.de> Message-Id: <1439552016-8557-1-git-send-email-pl@kamp.de> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-09-04quorum: validate vote threshold against num_children even if read-pattern is ↵Wen Congyang
fifo We need to use threshold to check if too many write operation fails. If threshold is larger than num children, we always get write error event even if all write operations success. Signed-off-by: Wen Congyang <wency@cn.fujitsu.com> Message-id: 55962F72.3060003@cn.fujitsu.com Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com>
2015-09-04qcow2: reorder fields in Qcow2CachedTable to reduce paddingAlberto Garcia
Changing the current ordering saves 8 bytes per cache entry in x86_64. Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Message-id: 0bd55291211df3dfb514d0e7d2031dd5c4f9f807.1438690126.git.berto@igalia.com Signed-off-by: Max Reitz <mreitz@redhat.com>
2015-09-04qcow2: add option to clean unused cache entries after some timeAlberto Garcia
This adds a new 'cache-clean-interval' option that cleans all qcow2 cache entries that haven't been used in a certain interval, given in seconds. This allows setting a large L2 cache size so it can handle scenarios with lots of I/O and at the same time use little memory during periods of inactivity. This feature currently relies on MADV_DONTNEED to free that memory, so it is not useful in systems that don't follow that behavior. Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-id: a70d12da60433df9360ada648b3f34b8f6f354ce.1438690126.git.berto@igalia.com Signed-off-by: Max Reitz <mreitz@redhat.com>
2015-09-04qcow2: mark the memory as no longer needed after qcow2_cache_empty()Alberto Garcia
After having emptied the cache, the data in the cache tables is no longer useful, so we can tell the kernel that we are done with it. In Linux this frees the resources associated with it. The effect of this can be seen in the HMP commit operation: it moves data from the top to the base image (and fills both caches), then it empties the top image. At this point the data in that cache is no longer needed so it's just wasting memory. Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Max Reitz <mreitz@redhat.com> Message-id: 08538b098e1faf6c92496477cf9b47a20e5aacea.1438690126.git.berto@igalia.com Signed-off-by: Max Reitz <mreitz@redhat.com>
2015-09-04block/raw-posix: Use raw_normalize_devicepath()Max Reitz
The filename given to qemu_open() in block/raw-posix.c should generally have been processed by raw_normalize_devicepath(); unless we are only probing (in which case the caller often checks whether the file is a block device or not, and this property will be changed by raw_normalize_devicepath() on NetBSD) or it is about a deprecated device (i.e. floppy). Signed-off-by: Max Reitz <mreitz@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-09-02block: more check for replaced nodeWen Congyang
We use mirror+replace to fix quorum's broken child. bs/s->common.bs is quorum, and to_replace is the broken child. The new child is target_bs. Without this patch, the replace node can be any node, and it can be top BDS with BB, or another quorum's child. We just check if the broken child is part of the quorum BDS in this patch. Signed-off-by: Wen Congyang <wency@cn.fujitsu.com> Message-id: 55A86486.1000404@cn.fujitsu.com Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2015-08-14mirror: Fix coroutine reentranceKevin Wolf
This fixes a regression introduced by commit dcfb3beb ("mirror: Do zero write on target if sectors not allocated"), which was reported to cause aborts with the message "Co-routine re-entered recursively". The cause for this bug is the following code in mirror_iteration_done(): if (s->common.busy) { qemu_coroutine_enter(s->common.co, NULL); } This has always been ugly because - unlike most places that reenter - it doesn't have a specific yield that it pairs with, but is more uncontrolled. What we really mean here is "reenter the coroutine if it's in one of the four explicit yields in mirror.c". This used to be equivalent with s->common.busy because neither mirror_run() nor mirror_iteration() call any function that could yield. However since commit dcfb3beb this doesn't hold true any more: bdrv_get_block_status_above() can yield. So what happens is that bdrv_get_block_status_above() wants to take a lock that is already held, so it adds itself to the queue of waiting coroutines and yields. Instead of being woken up by the unlock function, however, it gets woken up by mirror_iteration_done(), which is obviously wrong. In most cases the code actually happens to cope fairly well with such cases, but in this specific case, the unlock must already have scheduled the coroutine for wakeup when mirror_iteration_done() reentered it. And then the coroutine happened to process the scheduled restarts and tried to reenter itself recursively. This patch fixes the problem by pairing the reenter in mirror_iteration_done() with specific yields instead of abusing s->common.busy. Cc: qemu-stable@nongnu.org Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Jeff Cody <jcody@redhat.com> Message-id: 1439455310-11263-1-git-send-email-kwolf@redhat.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-08-06block/mirror: limit qiov to IOV_MAX elementsStefan Hajnoczi
If mirror has more free buffers than IOV_MAX, preadv(2)/pwritev(2) EINVAL failures may be encountered. It is possible to trigger this by setting granularity to a low value like 8192. This patch stops appending chunks once IOV_MAX is reached. The spurious EINVAL failure can be reproduced with a qcow2 image file and the following QMP invocation: qmp.command('drive-mirror', device='virtio0', target='/tmp/r7.s1', granularity=8192, sync='full', mode='absolute-paths', format='raw') While the guest is running dd if=/dev/zero of=/var/tmp/foo oflag=direct bs=4k. Cc: Jeff Cody <jcody@redhat.com> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Message-id: 1435761950-26714-1-git-send-email-stefanha@redhat.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-08-05block: don't register quorum driver if SHA256 support is unavailableSascha Silbe
Commit 488981a4 [block: convert quorum blockdrv to use crypto APIs] broke qemu-iotest 041 on hosts with GnuTLS < 2.10.0. It converted a compile-time check to a run-time check at device open time. The result is that we now advertise a feature (the quorum block driver) that will never work (on those hosts). There's no way (short of parsing human-readable error messages) for qemu-iotests or any other API consumer to recognise that the quorum block driver isn't _actually_ available and shouldn't be used or tested. Move the run-time check to bdrv_quorum_init() to avoid registering the quorum block driver if we know it cannot work. This way API consumers can recognise it's unavailable. Fixes: 488981a4af396551a3178d032cc2b41d9553ada2 Signed-off-by: Sascha Silbe <silbe@linux.vnet.ibm.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Message-id: 1438699705-21761-1-git-send-email-silbe@linux.vnet.ibm.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-07-28Merge remote-tracking branch ↵Peter Maydell
'remotes/cody/tags/jtc-for-upstream-pull-request' into staging # gpg: Signature made Tue Jul 28 05:22:29 2015 BST using RSA key ID C0DE3057 # gpg: Good signature from "Jeffrey Cody <jcody@redhat.com>" # gpg: aka "Jeffrey Cody <jeff@codyprime.org>" # gpg: aka "Jeffrey Cody <codyprime@gmail.com>" # gpg: WARNING: This key is not certified with sufficiently trusted signatures! # gpg: It is not certain that the signature belongs to the owner. # Primary key fingerprint: 9957 4B4D 3474 90E7 9D98 D624 BDBE 7B27 C0DE 3057 * remotes/cody/tags/jtc-for-upstream-pull-request: block/ssh: Avoid segfault if inet_connect doesn't set errno. sheepdog: serialize requests to overwrapping area Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-07-28block/ssh: Avoid segfault if inet_connect doesn't set errno.Richard W.M. Jones
On some (but not all) systems: $ qemu-img create -f qcow2 overlay -b ssh://xen/ Segmentation fault It turns out this happens when inet_connect returns -1 in the following code, but errno == 0. s->sock = inet_connect(s->hostport, errp); if (s->sock < 0) { ret = -errno; goto err; } In the test case above, no host called "xen" exists, so getaddrinfo fails. On Fedora 22, getaddrinfo happens to set errno = ENOENT (although it is *not* documented to do that), so it doesn't segfault. On RHEL 7, errno is not set by the failing getaddrinfo, so ret = -errno = 0, so the caller doesn't know there was an error and continues with a half-initialized BDRVSSHState struct, and everything goes south from there, eventually resulting in a segfault. Fix this by setting ret to -EIO (same as block/nbd.c and block/sheepdog.c). The real error is saved in the Error** errp struct, so it is printed correctly: $ ./qemu-img create -f qcow2 overlay -b ssh://xen/ qemu-img: overlay: address resolution failed for xen:22: No address associated with hostname Signed-off-by: Richard W.M. Jones <rjones@redhat.com> Reported-by: Jun Li BZ: https://bugzilla.redhat.com/show_bug.cgi?id=1147343 Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-07-28sheepdog: serialize requests to overwrapping areaHitoshi Mitake
Current sheepdog driver only serializes create requests in oid unit. This mechanism isn't enough for handling requests to overwrapping area spanning multiple oids, so it can result bugs like below: https://bugs.launchpad.net/sheepdog-project/+bug/1456421 This patch adds a new serialization mechanism for the problem. The difference from the old one is: 1. serialize entire aiocb if their targetting areas overwrap 2. serialize all requests (read, write, and discard), not only creates This patch also removes the old mechanism because the new one can be an alternative. Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Cc: Teruaki Ishizaki <ishizaki.teruaki@lab.ntt.co.jp> Cc: Vasiliy Tolstov <v.tolstov@selfip.ru> Signed-off-by: Hitoshi Mitake <mitake.hitoshi@lab.ntt.co.jp> Tested-by: Vasiliy Tolstov <v.tolstov@selfip.ru> Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-07-27block: vpc - prevent overflow if max_table_entries >= 0x40000000Jeff Cody
When we allocate the pagetable based on max_table_entries, we multiply the max table entry value by 4 to accomodate a table of 32-bit integers. However, max_table_entries is a uint32_t, and the VPC driver accepts ranges for that entry over 0x40000000. So during this allocation: s->pagetable = qemu_try_blockalign(bs->file, s->max_table_entries * 4); The size arg overflows, allocating significantly less memory than expected. Since qemu_try_blockalign() size argument is size_t, cast the multiplication correctly to prevent overflow. The value of "max_table_entries * 4" is used elsewhere in the code as well, so store the correct value for use in all those cases. We also check the Max Tables Entries value, to make sure that it is < SIZE_MAX / 4, so we know the pagetable size will fit in size_t. Cc: qemu-stable@nongnu.org Reported-by: Richard W.M. Jones <rjones@redhat.com> Signed-off-by: Jeff Cody <jcody@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-07-22mirror: Speed up bitmap initial scanningFam Zheng
Limiting to sectors_per_chunk for each bdrv_is_allocated_above is slow, because the underlying protocol driver would issue much more queries than necessary. We should coalesce the query. Signed-off-by: Fam Zheng <famz@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Message-id: <1436413678-7114-4-git-send-email-famz@redhat.com> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2015-07-14block/curl: Don't lose original error when a connection fails.Richard W.M. Jones
Currently if qemu is connected to a curl source (eg. web server), and the web server fails / times out / dies, you always see a bogus EIO "Input/output error". For example, choose a large file located on any local webserver which you control: $ qemu-img convert -p http://example.com/large.iso /tmp/test Once it starts copying the file, stop the webserver and you will see qemu-img fail with: qemu-img: error while reading sector 61440: Input/output error This patch does two things: Firstly print the actual error from curl so it doesn't get lost. Secondly, change EIO to EPROTO. EPROTO is a POSIX.1 compatible errno which more accurately reflects that there was a protocol error, rather than some kind of hardware failure. After this patch is applied, the error changes to: $ qemu-img convert -p http://example.com/large.iso /tmp/test qemu-img: curl: transfer closed with 469989 bytes remaining to read qemu-img: error while reading sector 16384: Protocol error Signed-off-by: Richard W.M. Jones <rjones@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-07-14mirror: correct buf_sizeWen Congyang
If bus_size is less than 0, the command fails. If buf_size is 0, use DEFAULT_MIRROR_BUF_SIZE. If buf_size % granularity is not 0, mirror_free_init() will do dangerous things. Signed-off-by: Wen Congyang <wency@cn.fujitsu.com> Reviewed-by: Fam Zheng <famz@redhat.com> Message-id: 5555A588.3080907@cn.fujitsu.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-07-14block: keep bitmap if incremental backup job is cancelledStefan Hajnoczi
Reclaim the dirty bitmap if an incremental backup block job is cancelled. The ret variable may be 0 when the job is cancelled so it's not enough to check ret < 0. Reviewed-by: John Snow <jsnow@redhat.com> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Message-id: 1434380534-7680-1-git-send-email-stefanha@redhat.com Signed-off-by: Jeff Cody <jcody@redhat.com>
2015-07-14block/mirror: Sleep periodically during bitmap scanningFam Zheng
Before, we only yield after initializing dirty bitmap, where the QMP command would return. That may take very long, and guest IO will be blocked. Add sleep points like the later mirror iterations. Signed-off-by: Fam Zheng <famz@redhat.com> Reviewed-by: Wen Congyang <wency@cn.fujitsu.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Message-id: 1431486673-19280-1-git-send-email-famz@redhat.com Signed-off-by: Jeff Cody <jcody@redhat.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-27 16:05:32 +0000