aboutsummaryrefslogtreecommitdiff
path: root/migration/rdma.c
AgeCommit message (Collapse)Author
2020-06-01migration/rdma: fix potential nullptr access in rdma_start_incoming_migrationPan Nengyuan
'rdma' is NULL when taking the first error branch in rdma_start_incoming_migration. And it will cause a null pointer access in label 'err'. Fix that. Fixes: 59c59c67ee6b0327ae932deb303caa47919aeb1e Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com> Message-Id: <20200508100755.7875-2-pannengyuan@huawei.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Note this is CID 1428762
2020-05-07migration/rdma: fix a memleak on error path in rdma_start_incoming_migrationPan Nengyuan
'rdma->host' is malloced in qemu_rdma_data_init, but forgot to free on the error path in rdma_start_incoming_migration(), this patch fix that. The leak stack: Direct leak of 2 byte(s) in 1 object(s) allocated from: #0 0x7fb7add18ae8 in __interceptor_malloc (/lib64/libasan.so.5+0xefae8) #1 0x7fb7ad0df1d5 in g_malloc (/lib64/libglib-2.0.so.0+0x531d5) #2 0x7fb7ad0f8b32 in g_strdup (/lib64/libglib-2.0.so.0+0x6cb32) #3 0x55a0464a0f6f in qemu_rdma_data_init /mnt/sdb/qemu/migration/rdma.c:2647 #4 0x55a0464b0e76 in rdma_start_incoming_migration /mnt/sdb/qemu/migration/rdma.c:4020 #5 0x55a0463f898a in qemu_start_incoming_migration /mnt/sdb/qemu/migration/migration.c:365 #6 0x55a0458c75d3 in qemu_init /mnt/sdb/qemu/softmmu/vl.c:4438 #7 0x55a046a3d811 in main /mnt/sdb/qemu/softmmu/main.c:48 #8 0x7fb7a8417872 in __libc_start_main (/lib64/libc.so.6+0x23872) #9 0x55a04536b26d in _start (/mnt/sdb/qemu/build/x86_64-softmmu/qemu-system-x86_64+0x286926d) Reported-by: Euler Robot <euler.robot@huawei.com> Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com> Message-Id: <20200420102727.17339-1-pannengyuan@huawei.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2020-02-13migration/rdma: rdma_accept_incoming_migration fix error handlingDr. David Alan Gilbert
rdma_accept_incoming_migration is called from an fd handler and can't return an Error * anywhere. Currently it's leaking Error's in errp/local_err - there's no point putting them in there unless we can report them. Turn most into fprintf's, and the last into an error_reportf_err where it's coming up from another function. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2020-01-29multifd: Make multifd_load_setup() get an Error parameterJuan Quintela
We need to change the full chain to pass the Error parameter. Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2019-10-11migration: Use automatic rcu_read unlock in rdma.cDr. David Alan Gilbert
Use the automatic read unlocker in migration/rdma.c. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Message-Id: <20191007143642.301445-5-dgilbert@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2019-09-25migration/rdma.c: Swap synchronize_rcu for call_rcuDr. David Alan Gilbert
This fixes a deadlock that can occur on the migration source after a failed RDMA migration; as the source tries to cleanup it clears a pair of pointers and uses synchronize_rcu to wait; this is happening on the main thread. With the CPUs running a CPU thread can be an rcu reader and attempt to grab the main lock (kvm_handle_io->address_space_write->flatview_write->flatview_write_continue-> prepare_mmio_access->qemu_mutex_lock_iothread_impl) Replace the synchronize_rcu with a call_rcu to postpone the freeing. Fixes: 74637e6f08fceda98806 ("migration: implement bi-directional RDMA QIOChannel") ( https://bugzilla.redhat.com/show_bug.cgi?id=1746787 ) Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-Id: <20190913163507.1403-3-dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2019-09-25migration/rdma: Don't moan about disconnects at the endDr. David Alan Gilbert
If we've already finished the migration or something has already gone wrong, don't moan about the migration stream disconnecting. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-Id: <20190913163507.1403-2-dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2019-08-16Merge remote-tracking branch ↵Peter Maydell
'remotes/armbru/tags/pull-include-2019-08-13-v2' into staging Header cleanup patches for 2019-08-13 # gpg: Signature made Fri 16 Aug 2019 12:39:12 BST # gpg: using RSA key 354BC8B3D7EB2A6B68674E5F3870B400EB918653 # gpg: issuer "armbru@redhat.com" # gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" [full] # gpg: aka "Markus Armbruster <armbru@pond.sub.org>" [full] # Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867 4E5F 3870 B400 EB91 8653 * remotes/armbru/tags/pull-include-2019-08-13-v2: (29 commits) sysemu: Split sysemu/runstate.h off sysemu/sysemu.h sysemu: Move the VMChangeStateEntry typedef to qemu/typedefs.h Include sysemu/sysemu.h a lot less Clean up inclusion of sysemu/sysemu.h numa: Move remaining NUMA declarations from sysemu.h to numa.h Include sysemu/hostmem.h less numa: Don't include hw/boards.h into sysemu/numa.h Include hw/boards.h a bit less Include hw/qdev-properties.h less Include qemu/main-loop.h less Include qemu/queue.h slightly less Include hw/hw.h exactly where needed Include qom/object.h slightly less Include exec/memory.h slightly less Include migration/vmstate.h less migration: Move the VMStateDescription typedef to typedefs.h Clean up inclusion of exec/cpu-common.h Include hw/irq.h a lot less typedefs: Separate incomplete types and function types ide: Include hw/ide/internal a bit less outside hw/ide/ ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2019-08-16Include exec/memory.h slightly lessMarkus Armbruster
Drop unnecessary inclusions from headers. Downgrade a few more to exec/hwaddr.h. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-Id: <20190812052359.30071-17-armbru@redhat.com>
2019-08-14migration: use migration_in_postcopy() to check POSTCOPY_ACTIVEWei Yang
Use common helper function to check the state. Signed-off-by: Wei Yang <richardw.yang@linux.intel.com> Message-Id: <20190719071129.11880-1-richardw.yang@linux.intel.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2019-07-04migration: move port_attr inside CONFIG_LINUXAlex Bennée
Otherwise the FreeBSD compiler complains about an unused variable. Signed-off-by: Alex Bennée <alex.bennee@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
2019-06-12Include qemu/module.h where needed, drop it from qemu-common.hMarkus Armbruster
Signed-off-by: Markus Armbruster <armbru@redhat.com> Message-Id: <20190523143508.25387-4-armbru@redhat.com> [Rebased with conflicts resolved automatically, except for hw/usb/dev-hub.c hw/misc/exynos4210_rng.c hw/misc/bcm2835_rng.c hw/misc/aspeed_scu.c hw/display/virtio-vga.c hw/arm/stm32f205_soc.c; ui/cocoa.m fixed up]
2019-03-25migration/rdma: Check qemu_rdma_init_one_blockDr. David Alan Gilbert
Actually it can't fail at the moment, but Coverity moans that it's the only place it's not checked, and it's an easy check. Reported-by: Coverity (CID 1399413) Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2019-03-06migration: Add an ability to ignore shared RAM blocksYury Kotov
If ignore-shared capability is set then skip shared RAMBlocks during the RAM migration. Also, move qemu_ram_foreach_migratable_block (and rename) to the migration code, because it requires access to the migration capabilities. Signed-off-by: Yury Kotov <yury-kotov@yandex-team.ru> Message-Id: <20190215174548.2630-4-yury-kotov@yandex-team.ru> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2019-03-06exec: Change RAMBlockIterFunc definitionYury Kotov
Currently, qemu_ram_foreach_* calls RAMBlockIterFunc with many block-specific arguments. But often iter func needs RAMBlock*. This refactoring is needed for fast access to RAMBlock flags from qemu_ram_foreach_block's callback. The only way to achieve this now is to call qemu_ram_block_from_host (which also enumerates blocks). So, this patch reduces complexity of qemu_ram_foreach_block() -> cb() -> qemu_ram_block_from_host() from O(n^2) to O(n). Fix RAMBlockIterFunc definition and add some functions to read RAMBlock* fields witch were passed. Signed-off-by: Yury Kotov <yury-kotov@yandex-team.ru> Message-Id: <20190215174548.2630-2-yury-kotov@yandex-team.ru> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2019-03-06migration/rdma: clang compilation fixMarcel Apfelbaum
Configuring QEMU with: ../configure --cc=clang --enable-rdma Leads to compilation error: CC migration/rdma.o CC migration/block.o qemu/migration/rdma.c:3615:58: error: taking address of packed member 'rkey' of class or structure 'RDMARegisterResult' may result in an unaligned pointer value [-Werror,-Waddress-of-packed-member] (uintptr_t)host_addr, NULL, &reg_result->rkey, ^~~~~~~~~~~~~~~~ Fix it by using a temp local variable. Signed-off-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com> Message-Id: <20190304184923.24215-1-marcel.apfelbaum@gmail.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
2019-03-06migration/rdma: Fix qemu_rdma_cleanup null checkDr. David Alan Gilbert
If the migration fails before the channel is open (e.g. a bad address) we end up in the cleanup with rdma->channel==NULL. Spotted by Coverity: CID 1398634 Fixes: fbbaacab2758cb3f32a0 Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-Id: <20190214185351.5927-1-dgilbert@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
2019-01-23migration/rdma: unregister fd handlerDr. David Alan Gilbert
Unregister the fd handler before we destroy the channel, otherwise we've got a race where we might land in the fd handler just as we're closing the device. (The race is quite data dependent, you just have to have the right set of devices for it to trigger). Corresponds to RH bz: https://bugzilla.redhat.com/show_bug.cgi?id=1666601 Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-Id: <20190122173111.29821-1-dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2018-09-26migration/rdma: Fix uninitialised rdma_return_pathDr. David Alan Gilbert
Clang correctly errors out moaning that rdma_return_path is used uninitialised in the earlier error paths. Make it NULL so that the error path ignores it. Fixes: 55cc1b5937a8e709e4c102e74b206281073aab82 Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reported-by: Cornelia Huck <cohuck@redhat.com> Message-Id: <20180830173657.22939-1-dgilbert@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2018-08-22migration: poll the cm event for destination qemuLidong Chen
The destination qemu only poll the comp_channel->fd in qemu_rdma_wait_comp_channel. But when source qemu disconnnect the rdma connection, the destination qemu should be notified. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-08-22migration: implement the shutdown for RDMA QIOChannelLidong Chen
Because RDMA QIOChannel not implement shutdown function, If the to_dst_file was set error, the return path thread will wait forever. and the migration thread will wait return path thread exit. the backtrace of return path thread is: (gdb) bt #0 0x00007f372a76bb0f in ppoll () from /lib64/libc.so.6 #1 0x000000000071dc24 in qemu_poll_ns (fds=0x7ef7091d0580, nfds=2, timeout=100000000) at qemu-timer.c:325 #2 0x00000000006b2fba in qemu_rdma_wait_comp_channel (rdma=0xd424000) at migration/rdma.c:1501 #3 0x00000000006b3191 in qemu_rdma_block_for_wrid (rdma=0xd424000, wrid_requested=4000, byte_len=0x7ef7091d0640) at migration/rdma.c:1580 #4 0x00000000006b3638 in qemu_rdma_exchange_get_response (rdma=0xd424000, head=0x7ef7091d0720, expecting=3, idx=0) at migration/rdma.c:1726 #5 0x00000000006b3ad6 in qemu_rdma_exchange_recv (rdma=0xd424000, head=0x7ef7091d0720, expecting=3) at migration/rdma.c:1903 #6 0x00000000006b5d03 in qemu_rdma_get_buffer (opaque=0x6a57dc0, buf=0x5c80030 "", pos=8, size=32768) at migration/rdma.c:2714 #7 0x00000000006a9635 in qemu_fill_buffer (f=0x5c80000) at migration/qemu-file.c:232 #8 0x00000000006a9ecd in qemu_peek_byte (f=0x5c80000, offset=0) at migration/qemu-file.c:502 #9 0x00000000006a9f1f in qemu_get_byte (f=0x5c80000) at migration/qemu-file.c:515 #10 0x00000000006aa162 in qemu_get_be16 (f=0x5c80000) at migration/qemu-file.c:591 #11 0x00000000006a46d3 in source_return_path_thread ( opaque=0xd826a0 <current_migration.37100>) at migration/migration.c:1331 #12 0x00007f372aa49e25 in start_thread () from /lib64/libpthread.so.0 #13 0x00007f372a77635d in clone () from /lib64/libc.so.6 the backtrace of migration thread is: (gdb) bt #0 0x00007f372aa4af57 in pthread_join () from /lib64/libpthread.so.0 #1 0x00000000007d5711 in qemu_thread_join (thread=0xd826f8 <current_migration.37100+88>) at util/qemu-thread-posix.c:504 #2 0x00000000006a4bc5 in await_return_path_close_on_source ( ms=0xd826a0 <current_migration.37100>) at migration/migration.c:1460 #3 0x00000000006a53e4 in migration_completion (s=0xd826a0 <current_migration.37100>, current_active_state=4, old_vm_running=0x7ef7089cf976, start_time=0x7ef7089cf980) at migration/migration.c:1695 #4 0x00000000006a5c54 in migration_thread (opaque=0xd826a0 <current_migration.37100>) at migration/migration.c:1837 #5 0x00007f372aa49e25 in start_thread () from /lib64/libpthread.so.0 #6 0x00007f372a77635d in clone () from /lib64/libc.so.6 Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-08-22migration: poll the cm event while wait RDMA work request completionLidong Chen
If the peer qemu is crashed, the qemu_rdma_wait_comp_channel function maybe loop forever. so we should also poll the cm event fd, and when receive RDMA_CM_EVENT_DISCONNECTED and RDMA_CM_EVENT_DEVICE_REMOVAL, we consider some error happened. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Signed-off-by: Gal Shachaf <galsha@mellanox.com> Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-08-22migration: implement io_set_aio_fd_handler function for RDMA QIOChannelLidong Chen
if qio_channel_rdma_readv return QIO_CHANNEL_ERR_BLOCK, the destination qemu crash. The backtrace is: (gdb) bt #0 0x0000000000000000 in ?? () #1 0x00000000008db50e in qio_channel_set_aio_fd_handler (ioc=0x38111e0, ctx=0x3726080, io_read=0x8db841 <qio_channel_restart_read>, io_write=0x0, opaque=0x38111e0) at io/channel.c: #2 0x00000000008db952 in qio_channel_set_aio_fd_handlers (ioc=0x38111e0) at io/channel.c:438 #3 0x00000000008dbab4 in qio_channel_yield (ioc=0x38111e0, condition=G_IO_IN) at io/channel.c:47 #4 0x00000000007a870b in channel_get_buffer (opaque=0x38111e0, buf=0x440c038 "", pos=0, size=327 at migration/qemu-file-channel.c:83 #5 0x00000000007a70f6 in qemu_fill_buffer (f=0x440c000) at migration/qemu-file.c:299 #6 0x00000000007a79d0 in qemu_peek_byte (f=0x440c000, offset=0) at migration/qemu-file.c:562 #7 0x00000000007a7a22 in qemu_get_byte (f=0x440c000) at migration/qemu-file.c:575 #8 0x00000000007a7c78 in qemu_get_be32 (f=0x440c000) at migration/qemu-file.c:655 #9 0x00000000007a0508 in qemu_loadvm_state (f=0x440c000) at migration/savevm.c:2126 #10 0x0000000000794141 in process_incoming_migration_co (opaque=0x0) at migration/migration.c:366 #11 0x000000000095c598 in coroutine_trampoline (i0=84033984, i1=0) at util/coroutine-ucontext.c:1 #12 0x00007f9c0db56d40 in ?? () from /lib64/libc.so.6 #13 0x00007f96fe858760 in ?? () #14 0x0000000000000000 in ?? () RDMA QIOChannel not implement io_set_aio_fd_handler. so qio_channel_set_aio_fd_handler will access NULL pointer. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-08-22migration: Stop rdma yielding during incoming postcopyLidong Chen
During incoming postcopy, the destination qemu will invoke qemu_rdma_wait_comp_channel in a seprate thread. So does not use rdma yield, and poll the completion channel fd instead. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-08-22migration: implement bi-directional RDMA QIOChannelLidong Chen
This patch implements bi-directional RDMA QIOChannel. Because different threads may access RDMAQIOChannel currently, this patch use RCU to protect it. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-08-22migration: create a dedicated connection for rdma return pathLidong Chen
If start a RDMA migration with postcopy enabled, the source qemu establish a dedicated connection for return path. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-08-22migration: disable RDMA WRITE after postcopy startedLidong Chen
RDMA WRITE operations are performed with no notification to the destination qemu, then the destination qemu can not wakeup. This patch disable RDMA WRITE after postcopy started. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-06-15migration: Fixes for non-migratable RAMBlocksDr. David Alan Gilbert
There are still a few cases where migration code is using the macros and functions that do all RAMBlocks rather than just the migratable blocks; fix those up. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-Id: <20180605162545.80778-2-dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Cédric Le Goater <clg@kaod.org> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2018-06-04migration: not wait RDMA_CM_EVENT_DISCONNECTED event after rdma_disconnectLidong Chen
When cancel migration during RDMA precopy, the source qemu main thread hangs sometime. The backtrace is: (gdb) bt #0 0x00007f249eabd43d in write () from /lib64/libpthread.so.0 #1 0x00007f24a1ce98e4 in rdma_get_cm_event (channel=0x4675d10, event=0x7ffe2f643dd0) at src/cma.c:2189 #2 0x00000000007b6166 in qemu_rdma_cleanup (rdma=0x6784000) at migration/rdma.c:2296 #3 0x00000000007b7cae in qio_channel_rdma_close (ioc=0x3bfcc30, errp=0x0) at migration/rdma.c:2999 #4 0x00000000008db60e in qio_channel_close (ioc=0x3bfcc30, errp=0x0) at io/channel.c:273 #5 0x00000000007a8765 in channel_close (opaque=0x3bfcc30) at migration/qemu-file-channel.c:98 #6 0x00000000007a71f9 in qemu_fclose (f=0x527c000) at migration/qemu-file.c:334 #7 0x0000000000795b96 in migrate_fd_cleanup (opaque=0x3b46280) at migration/migration.c:1162 #8 0x000000000093a71b in aio_bh_call (bh=0x3db7a20) at util/async.c:90 #9 0x000000000093a7b2 in aio_bh_poll (ctx=0x3b121c0) at util/async.c:118 #10 0x000000000093f2ad in aio_dispatch (ctx=0x3b121c0) at util/aio-posix.c:436 #11 0x000000000093ab41 in aio_ctx_dispatch (source=0x3b121c0, callback=0x0, user_data=0x0) at util/async.c:261 #12 0x00007f249f73c7aa in g_main_context_dispatch () from /lib64/libglib-2.0.so.0 #13 0x000000000093dc5e in glib_pollfds_poll () at util/main-loop.c:215 #14 0x000000000093dd4e in os_host_main_loop_wait (timeout=28000000) at util/main-loop.c:263 #15 0x000000000093de05 in main_loop_wait (nonblocking=0) at util/main-loop.c:522 #16 0x00000000005bc6a5 in main_loop () at vl.c:1944 #17 0x00000000005c39b5 in main (argc=56, argv=0x7ffe2f6443f8, envp=0x3ad0030) at vl.c:4752 It does not get the RDMA_CM_EVENT_DISCONNECTED event after rdma_disconnect sometime. According to IB Spec once active side send DREQ message, it should wait for DREP message and only once it arrived it should trigger a DISCONNECT event. DREP message can be dropped due to network issues. For that case the spec defines a DREP_timeout state in the CM state machine, if the DREP is dropped we should get a timeout and a TIMEWAIT_EXIT event will be trigger. Unfortunately the current kernel CM implementation doesn't include the DREP_timeout state and in above scenario we will not get DISCONNECT or TIMEWAIT_EXIT events. So it should not invoke rdma_get_cm_event which may hang forever, and the event channel is also destroyed in qemu_rdma_cleanup. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2018-06-04migration: remove unnecessary variables len in QIOChannelRDMALidong Chen
Because qio_channel_rdma_writev and qio_channel_rdma_readv maybe invoked by different threads concurrently, this patch removes unnecessary variables len in QIOChannelRDMA and use local variable instead. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Lidong Chen <jemmy858585@gmail.com>
2018-05-15migration: update index field when delete or qsort RDMALocalBlockLidong Chen
rdma_delete_block function deletes RDMALocalBlock base on index field, but not update the index field. So when next time invoke rdma_delete_block, it will not work correctly. If start and cancel migration repeatedly, some RDMALocalBlock not invoke ibv_dereg_mr to decrease kernel mm_struct vmpin. When vmpin is large than max locked memory limitation, ibv_reg_mr will failed, and migration can not start successfully again. Signed-off-by: Lidong Chen <lidongchen@tencent.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-Id: <1525618499-1560-1-git-send-email-lidongchen@tencent.com> Signed-off-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Lidong Chen <jemmy858585@gmail.com>
2018-02-06migration: Allow migrate_fd_connect to take an Error *Dr. David Alan Gilbert
Allow whatever is performing the connection to pass migrate_fd_connect an error to indicate there was a problem during connection, an allow us to clean up. The caller must free the error. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2018-01-16maint: Fix macros with broken 'do/while(0); ' usageEric Blake
The point of writing a macro embedded in a 'do { ... } while (0)' loop (particularly if the macro has multiple statements or would otherwise end with an 'if' statement) is so that the macro can be used as a drop-in statement with the caller supplying the trailing ';'. Although our coding style frowns on brace-less 'if': if (cond) statement; else something else; that is the classic case where failure to use do/while(0) wrapping would cause the 'else' to pair with any embedded 'if' in the macro rather than the intended outer 'if'. But conversely, if the macro includes an embedded ';', then the same brace-less coding style would now have two statements, making the 'else' a syntax error rather than pairing with the outer 'if'. Thus, even though our coding style with required braces is not impacted, ending a macro with ';' makes our code harder to port to projects that use brace-less styles. The change should have no semantic impact. I was not able to fully compile-test all of the changes (as some of them are examples of the ugly bit-rotting debug print statements that are completely elided by default, and I didn't want to recompile with the necessary -D witnesses - cleaning those up is left as a bite-sized task for another day); I did, however, audit that for all files touched, all callers of the changed macros DID supply a trailing ';' at the callsite, and did not appear to be used as part of a brace-less conditional. Found mechanically via: $ git grep -B1 'while (0);' | grep -A1 \\\\ Signed-off-by: Eric Blake <eblake@redhat.com> Acked-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Message-Id: <20171201232433.25193-7-eblake@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-07-18migration/rdma: Send error during cancellingDr. David Alan Gilbert
When we issue a cancel and clean up the RDMA channel send a CONTROL_ERROR to get the destination to quit. The rdma_cleanup code waits for the event to come back from the rdma_disconnect; but that wont happen until the destination quits and there's currently nothing to force it. Note this makes the case of a cancel work while the destination is alive, and it already works if the destination is truly dead. Note it doesn't fix the case where the destination is hung (we get stuck waiting for the rdma_disconnect event). Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Message-Id: <20170717110936.23314-7-dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2017-07-18migration/rdma: Safely convert control typesDr. David Alan Gilbert
control_desc[] is an array of strings that correspond to a series of message types; they're used only for error messages, but if the message type is seriously broken then we could go off the end of the array. Convert the array to a function control_desc() that bound checks. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Message-Id: <20170717110936.23314-6-dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2017-07-18migration/rdma: Allow cancelling while waiting for wridDr. David Alan Gilbert
When waiting for a WRID, if the other side dies we end up waiting for ever with no way to cancel the migration. Cure this by poll()ing the fd first with a timeout and checking error flags and migration state. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Message-Id: <20170717110936.23314-5-dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2017-07-18migration/rdma: fix qemu_rdma_block_for_wrid error pathsDr. David Alan Gilbert
The two places that 'goto err_block_for_wrid' weren't setting ret and so would end up returning 0 even though we've failed. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Message-Id: <20170717110936.23314-4-dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2017-07-18migration/rdma: Fix race on sourceDr. David Alan Gilbert
Fix a race where the destination might try and send the source a WRID_READY before the source has done a post-recv for it. rdma_post_recv has to happen after the qp exists, and we're OK since we've already called qemu_rdma_source_init that calls qemu_alloc_qp. This corresponds to: https://bugzilla.redhat.com/show_bug.cgi?id=1285044 The race can be triggered by adding a few ms wait before this post_recv_control (which was originally due to me turning on loads of debug). Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Message-Id: <20170717110936.23314-2-dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2017-06-13migration: Move migration.h to migration/Juan Quintela
Nothing uses it outside of migration.h Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Laurent Vivier <lvivier@redhat.com>
2017-06-01migration: Export ram.c functions in its own fileJuan Quintela
All functions are internal except for ram_mig_init(). Create migration/misc.h for this kind of functions. Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2017-06-01migration: Export rdma.c functions in its own fileJuan Quintela
Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2017-06-01migration: Split qemu-file.hJuan Quintela
Split the file into public and internal interfaces. I have to rename the external one because we can't have two include files with the same name in the same directory. Build system gets confused. The only exported functions are the ones that handle basic types. Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2017-05-18migration: Export qemu-file-channel.c functions in its own fileJuan Quintela
Signed-off-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
2017-05-09sockets: Prepare inet_parse() for flattened SocketAddressMarkus Armbruster
I'm going to flatten SocketAddress: rename SocketAddress to SocketAddressLegacy, SocketAddressFlat to SocketAddress, eliminate SocketAddressLegacy except in external interfaces. inet_parse() returns a newly allocated InetSocketAddress. Lift the allocation from inet_parse() into its caller socket_parse() to prepare for flattening SocketAddress. Signed-off-by: Markus Armbruster <armbru@redhat.com> Message-Id: <1493192202-3184-3-git-send-email-armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> [Straightforward rebase]
2017-04-24migration: Make errp the last parameter of local functionsFam Zheng
Signed-off-by: Fam Zheng <famz@redhat.com> Message-Id: <20170421122710.15373-13-famz@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-10-13migration/rdma: Don't flag an error when we've been told about oneDr. David Alan Gilbert
If the other side tells us there's been an error and we fail the migration, we don't need to signal that failure to the other side because it already knew. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Michael R. Hines <michael@hinespot.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2016-10-13migration/rdma: Pass qemu_file errors across linkDr. David Alan Gilbert
If we fail for some reason (e.g. a mismatched RAMBlock) and it's set the qemu_file error flag, pass that error back to the peer so it can clean up rather than waiting for some higher level progress. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Reviewed-by: Michael R. Hines <michael@hinespot.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
2016-09-13cutils: Export only buffer_is_zeroRichard Henderson
Since the two users don't make use of the returned offset, beyond ensuring that the entire buffer is zero, consider the can_use_buffer_find_nonzero_offset and buffer_find_nonzero_offset functions internal. Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Richard Henderson <rth@twiddle.net> Message-Id: <1472496380-19706-4-git-send-email-rth@twiddle.net> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2016-06-16migration: rename functions to starting migrationsDaniel P. Berrange
Apply the following renames for starting incoming migration: process_incoming_migration -> migration_fd_process_incoming migration_set_incoming_channel -> migration_channel_process_incoming migration_tls_set_incoming_channel -> migration_tls_channel_process_incoming and for starting outgoing migration: migration_set_outgoing_channel -> migration_channel_connect migration_tls_set_outgoing_channel -> migration_tls_channel_connect Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Message-id: 1464776234-9910-3-git-send-email-berrange@redhat.com Message-Id: <1464776234-9910-3-git-send-email-berrange@redhat.com> Signed-off-by: Amit Shah <amit.shah@redhat.com>
2016-05-26migration: convert RDMA to use QIOChannel interfaceDaniel P. Berrange
This converts the RDMA code to provide a subclass of QIOChannel that uses RDMA for the data transport. This implementation of RDMA does not correctly handle non-blocking mode. Reads might block if there was not already some pending data and writes will block until all data is sent. This flawed behaviour was already present in the existing impl, so appears to not be a critical problem at this time. It should be on the list of things to fix in the future though. The RDMA code would be much better off it it could be split up in a generic RDMA layer, a QIOChannel impl based on RMDA, and then the RMDA migration glue. This is left as a future exercise for the brave. Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Message-Id: <1461751518-12128-18-git-send-email-berrange@redhat.com> Signed-off-by: Amit Shah <amit.shah@redhat.com>
generated by cgit v1.2.3 (git 2.26.2) at 2024-12-28 02:03:44 +0000