We use errno after calling Libibverbs functions that are not
documented to set errno (manual page does not mention errno), or where
the documentation is unclear ("returns [...] the value of errno on
failure"). While this could be read as "sets errno and returns it",
a glance at the source code[*] kills that hope:
static inline int ibv_post_send(struct ibv_qp *qp, struct ibv_send_wr *wr,
struct ibv_send_wr **bad_wr)
{
return qp->context->ops.post_send(qp, wr, bad_wr);
}
The callback can be
static int mana_post_send(struct ibv_qp *ibqp, struct ibv_send_wr *wr,
struct ibv_send_wr **bad)
{
/* This version of driver supports RAW QP only.
* Posting WR is done directly in the application.
*/
return EOPNOTSUPP;
}
Neither of them touches errno.
One of these errno uses is easy to fix, so do that now. Several more
will go away later in the series; add temporary FIXME commments.
Three will remain; add TODO comments. TODO, not FIXME, because the
bug might be in Libibverbs documentation.
[*] https://github.com/linux-rdma/rdma-core.git
commit 55fa316b4b18f258d8ac1ceb4aa5a7a35b094dcf
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-17-armbru@redhat.com>
@error_reported and @received_error are flags. The latter is even
assigned bool true. Change them from int to bool.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-16-armbru@redhat.com>
qemu_rdma_buffer_mergeable() is semantically a predicate. It returns
int 0 or 1. Return bool instead, and fix the function name's
spelling.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-15-armbru@redhat.com>
qemu_rdma_search_ram_block() can't fail. Return void, and drop the
unreachable error handling.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-14-armbru@redhat.com>
rdma_add_block() can't fail. Return void, and drop the unreachable
error handling.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-13-armbru@redhat.com>
When all we do with an Error we receive into a local variable is
propagating to somewhere else, we can just as well receive it there
right away.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-12-armbru@redhat.com>
include/qapi/error.h demands:
* - Functions that use Error to report errors have an Error **errp
* parameter. It should be the last parameter, except for functions
* taking variable arguments.
qemu_rdma_connect() does not conform. Clean it up.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-11-armbru@redhat.com>
qemu_rdma_accept() returns 0 in some cases even when it didn't
complete its job due to errors. Impact is not obvious. I figure the
caller will soon fail again with a misleading error message.
Fix it to return -1 on any failure.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-10-armbru@redhat.com>
qemu_rdma_exchange_get_response() compares int parameter @expecting
with uint32_t head->type. Actual arguments are non-negative
enumeration constants, RDMAControlHeader uint32_t member type, or
qemu_rdma_exchange_recv() int parameter expecting. Actual arguments
for the latter are non-negative enumeration constants. Change both
parameters to uint32_t.
In qio_channel_rdma_readv(), loop control variable @i is ssize_t, and
counts from 0 up to @niov, which is size_t. Change @i to size_t.
While there, make qio_channel_rdma_readv() and
qio_channel_rdma_writev() more consistent: change the former's @done
to ssize_t, and delete the latter's useless initialization of @len.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-8-armbru@redhat.com>
qio_channel_rdma_readv() assigns the size_t value of qemu_rdma_fill()
to an int variable before it adds it to @done / subtracts it from
@want, both size_t. Truncation when qemu_rdma_fill() copies more than
INT_MAX bytes. Seems vanishingly unlikely, but needs fixing all the
same.
Fixes: 6ddd2d76ca (migration: convert RDMA to use QIOChannel interface)
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-7-armbru@redhat.com>
We use int instead of uint64_t in a few places. Change them to
uint64_t.
This cleans up a comparison of signed qemu_rdma_block_for_wrid()
parameter @wrid_requested with unsigned @wr_id. Harmless, because the
actual arguments are non-negative enumeration constants.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-6-armbru@redhat.com>
wrid_desc[] uses 4001 pointers to map four integer values to strings.
print_wrid() accesses wrid_desc[] out of bounds when passed a negative
argument. It returns null for values 2..1999 and 2001..3999.
qemu_rdma_poll() and qemu_rdma_block_for_wrid() print wrid_desc[wr_id]
and passes print_wrid(wr_id) to tracepoints. Could conceivably crash
trying to format a null string. I believe access out of bounds is not
possible.
Not worth cleaning up. Dumb down to show just numeric wr_id.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-5-armbru@redhat.com>
rdma_delete_block() always returns 0, which its only caller ignores.
Return void instead.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-4-armbru@redhat.com>
qemu_rdma_data_init() return type is void *. It actually returns
RDMAContext *, and all its callers assign the value to an
RDMAContext *. Unclean.
Return RDMAContext * instead.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-3-armbru@redhat.com>
qemu_rdma_poll()'s return type is uint64_t, even though it returns 0,
-1, or @ret, which is int. Its callers assign the return value to int
variables, then check whether it's negative. Unclean.
Return int instead.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230928132019.2544702-2-armbru@redhat.com>
Previously, we got a confusion error that complains
the RDMAControlHeader.repeat:
qemu-system-x86_64: rdma: Too many requests in this message (3638950032).Bailing.
Actually, it's caused by an unexpected RDMAControlHeader.type.
After this patch, error will become:
qemu-system-x86_64: Unknown control message QEMU FILE
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-ID: <20230926100103.201564-2-lizhijian@fujitsu.com>
When we sent a page through QEMUFile hooks (RDMA) there are three
posiblities:
- We are not using RDMA. return RAM_SAVE_CONTROL_DELAYED and
control_save_page() returns false to let anything else to proceed.
- There is one error but we are using RDMA. Then we return a negative
value, control_save_page() needs to return true.
- Everything goes well and RDMA start the sent of the page
asynchronously. It returns RAM_SAVE_CONTROL_DELAYED and we need to
return 1 for ram_save_page_legacy.
Clear?
I know, I know, the interface is as bad as it gets. I think that now
it is a bit clearer, but this needs to be done some other way.
Reviewed-by: Leonardo Bras <leobras@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20230515195709.63843-16-quintela@redhat.com>
Reviewed-by: Leonardo Bras <leobras@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20230515195709.63843-13-quintela@redhat.com>
Just create a variable for it, the same way that multifd does. This
way it is safe to use for other thread, etc, etc.
Reviewed-by: Leonardo Bras <leobras@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20230515195709.63843-11-quintela@redhat.com>
Local variables shadowing other local variables or parameters make the
code needlessly hard to understand. Tracked down with -Wshadow=local.
Clean up: delete inner declarations when they are actually redundant,
else rename variables.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Message-ID: <20230921121312.1301864-3-armbru@redhat.com>
qemu_rdma_save_page() reports polling error with error_report(), then
succeeds anyway. This is because the variable holding the polling
status *shadows* the variable the function returns. The latter
remains zero.
Broken since day one, and duplicated more recently.
Fixes: 2da776db48 (rdma: core logic)
Fixes: b390afd8c5 (migration/rdma: Fix out of order wrid)
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Message-ID: <20230921121312.1301864-2-armbru@redhat.com>
The ongoing QEMU multi-queue block layer effort makes it possible for multiple
threads to process I/O in parallel. The nbd block driver is not compatible with
the multi-queue block layer yet because QIOChannel cannot be used easily from
coroutines running in multiple threads. This series changes the QIOChannel API
to make that possible.
In the current API, calling qio_channel_attach_aio_context() sets the
AioContext where qio_channel_yield() installs an fd handler prior to yielding:
qio_channel_attach_aio_context(ioc, my_ctx);
...
qio_channel_yield(ioc); // my_ctx is used here
...
qio_channel_detach_aio_context(ioc);
This API design has limitations: reading and writing must be done in the same
AioContext and moving between AioContexts involves a cumbersome sequence of API
calls that is not suitable for doing on a per-request basis.
There is no fundamental reason why a QIOChannel needs to run within the
same AioContext every time qio_channel_yield() is called. QIOChannel
only uses the AioContext while inside qio_channel_yield(). The rest of
the time, QIOChannel is independent of any AioContext.
In the new API, qio_channel_yield() queries the AioContext from the current
coroutine using qemu_coroutine_get_aio_context(). There is no need to
explicitly attach/detach AioContexts anymore and
qio_channel_attach_aio_context() and qio_channel_detach_aio_context() are gone.
One coroutine can read from the QIOChannel while another coroutine writes from
a different AioContext.
This API change allows the nbd block driver to use QIOChannel from any thread.
It's important to keep in mind that the block driver already synchronizes
QIOChannel access and ensures that two coroutines never read simultaneously or
write simultaneously.
This patch updates all users of qio_channel_attach_aio_context() to the
new API. Most conversions are simple, but vhost-user-server requires a
new qemu_coroutine_yield() call to quiesce the vu_client_trip()
coroutine when not attached to any AioContext.
While the API is has become simpler, there is one wart: QIOChannel has a
special case for the iohandler AioContext (used for handlers that must not run
in nested event loops). I didn't find an elegant way preserve that behavior, so
I added a new API called qio_channel_set_follow_coroutine_ctx(ioc, true|false)
for opting in to the new AioContext model. By default QIOChannel uses the
iohandler AioHandler. Code that formerly called
qio_channel_attach_aio_context() now calls
qio_channel_set_follow_coroutine_ctx(ioc, true) once after the QIOChannel is
created.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20230830224802.493686-5-stefanha@redhat.com>
[eblake: also fix migration/rdma.c]
Signed-off-by: Eric Blake <eblake@redhat.com>
This is how everything else in QEMUFile is structured.
As a bonus they are three less lines of code.
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-ID: <20230530183941.7223-17-quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
All callers now pass is_external=false to aio_set_fd_handler() and
aio_set_event_notifier(). The aio_disable_external() API that
temporarily disables fd handlers that were registered is_external=true
is therefore dead code.
Remove aio_disable_external(), aio_enable_external(), and the
is_external arguments to aio_set_fd_handler() and
aio_set_event_notifier().
The entire test-fdmon-epoll test is removed because its sole purpose was
testing aio_disable_external().
Parts of this patch were generated using the following coccinelle
(https://coccinelle.lip6.fr/) semantic patch:
@@
expression ctx, fd, is_external, io_read, io_write, io_poll, io_poll_ready, opaque;
@@
- aio_set_fd_handler(ctx, fd, is_external, io_read, io_write, io_poll, io_poll_ready, opaque)
+ aio_set_fd_handler(ctx, fd, io_read, io_write, io_poll, io_poll_ready, opaque)
@@
expression ctx, notifier, is_external, io_read, io_poll, io_poll_ready;
@@
- aio_set_event_notifier(ctx, notifier, is_external, io_read, io_poll, io_poll_ready)
+ aio_set_event_notifier(ctx, notifier, io_read, io_poll, io_poll_ready)
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20230516190238.8401-21-stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Originally, migration_incoming_co was introduced by
25d0c16f62
"migration: Switch to COLO process after finishing loadvm"
to be able to enter from COLO code to one specific yield point, added
by 25d0c16f62.
Later in 923709896b
"migration: poll the cm event for destination qemu"
we reused this variable to wake the migration incoming coroutine from
RDMA code.
That was doubtful idea. Entering coroutines is a very fragile thing:
you should be absolutely sure which yield point you are going to enter.
I don't know how much is it safe to enter during qemu_loadvm_state()
which I think what RDMA want to do. But for sure RDMA shouldn't enter
the special COLO-related yield-point. As well, COLO code doesn't want
to enter during qemu_loadvm_state(), it want to enter it's own specific
yield-point.
As well, when in 8e48ac9586
"COLO: Add block replication into colo process" we added
bdrv_invalidate_cache_all() call (now it's called activate_all())
it became possible to enter the migration incoming coroutine during
that call which is wrong too.
So, let't make these things separate and disjoint: loadvm_co for RDMA,
non-NULL during qemu_loadvm_state(), and colo_incoming_co for COLO,
non-NULL only around specific yield.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20230515130640.46035-3-vsementsov@yandex-team.ru>
Signed-off-by: Juan Quintela <quintela@redhat.com>
It makes no sense first try to see if there is an rdma error and then
do nothing on postcopy stage. Change it so we check we are in
postcopy before doing anything.
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230504114443.23891-6-quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230504114443.23891-4-quintela@redhat.com>
We can calculate it from the QEMUFile like the caller.
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230503131847.11603-6-quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Now that we have atomic counters, we can do it on the place that we
need it, no need to do it inside ram.c.
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Lukas Straub <lukasstraub2@web.de>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
---
Fixed missing space after comma (fabiano)
Once that we are there, we rename the function to migrate_return_path()
to be consistent with all other capabilities.
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
It is clear from the context what that means, and such a long name
with the extra long names of the capabilities make very difficilut to
stay inside the 80 columns limit.
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
It's no longer needed since commit
44bcfd45e9 ("migration/rdma: destination: create the return patch after the first accept")
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
The RDMA code has return-path handling code, but it's only enabled
if postcopy is enabled; if the 'return-path' migration capability
is enabled, the return path is NOT setup but the core migration
code still tries to use it and breaks.
Enable the RDMA return path if either postcopy or the return-path
capability is enabled.
bz: https://bugzilla.redhat.com/show_bug.cgi?id=2063615
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
MSG_PEEK peeks at the channel, The data is treated as unread and
the next read shall still return this data. This support is
currently added only for socket class. Extra parameter 'flags'
is added to io_readv calls to pass extra read flags like MSG_PEEK.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Suggested-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: manish.mishra <manish.mishra@nutanix.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
upon errors. As the documentation in include/io/channel.h states, only
-1 and QIO_CHANNEL_ERR_BLOCK should be returned upon error. Other
values have the potential to confuse the call sites.
error_setg is used rather than error_setg_errno, because there are
certain code paths where -1 (as a non-errno) is propagated up (e.g.
starting from qemu_rdma_block_for_wrid or qemu_rdma_post_recv_control)
all the way to qio_channel_rdma_{readv,writev}.
Similar to a216ec85b7 ("migration/channel-block: fix return value for
qio_channel_block_{readv,writev}").
Suggested-by: Zhang Chen <chen.zhang@intel.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Now that all QEMUFile callbacks are removed, the entire concept can be
deleted.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-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: Dr. David Alan Gilbert <dgilbert@redhat.com>
The only user of the hooks is RDMA which provides a QIOChannel backed
impl of QEMUFile. It can thus use the qemu_file_get_ioc() method.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-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: Dr. David Alan Gilbert <dgilbert@redhat.com>
The QEMUFile 'save_hook' callback has a 'size_t size' parameter.
The RDMA impl of this has logic that takes different actions
depending on whether the value is zero or non-zero. It has
commented out logic that would have taken further actions
if the value was negative.
The only place where the 'save_hook' callback is invoked is
the ram_control_save_page() method, which passes 'size'
through from its caller. The only caller of this method is
in turn control_save_page(). This method unconditionally
passes the 'TARGET_PAGE_SIZE' constant for the 'size' parameter.
IOW, the only scenario for 'size' that can execute in the
qemu_rdma_save_page method is 'size > 0'. The remaining code
has been unreachable since RDMA support was first introduced
9 years ago.
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-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: Dr. David Alan Gilbert <dgilbert@redhat.com>
Nobody has ever showed up to unregister individual pages, and another
set of patches written by Daniel P. Berrangé <berrange@redhat.com>
just remove qemu_rdma_signal_unregister() function needed here.
Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Add flags to io_writev and introduce io_flush as optional callback to
QIOChannelClass, allowing the implementation of zero copy writes by
subclasses.
How to use them:
- Write data using qio_channel_writev*(...,QIO_CHANNEL_WRITE_FLAG_ZERO_COPY),
- Wait write completion with qio_channel_flush().
Notes:
As some zero copy write implementations work asynchronously, it's
recommended to keep the write buffer untouched until the return of
qio_channel_flush(), to avoid the risk of sending an updated buffer
instead of the buffer state during write.
As io_flush callback is optional, if a subclass does not implement it, then:
- io_flush will return 0 without changing anything.
Also, some functions like qio_channel_writev_full_all() were adapted to
receive a flag parameter. That allows shared code between zero copy and
non-zero copy writev, and also an easier implementation on new flags.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20220513062836.965425-3-leobras@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
We hit following error during testing RDMA transport:
in case of migration error, mgmt daemon pick one migration port,
incoming rdma:[::]:8089: RDMA ERROR: Error: could not rdma_bind_addr
Then try another -incoming rdma:[::]:8103, sometime it worked,
sometimes need another try with other ports number.
Set the REUSEADDR option for destination, This allow address could
be reused to avoid rdma_bind_addr error out.
Signed-off-by: Jack Wang <jinpu.wang@ionos.com>
Message-Id: <20220208085640.19702-1-jinpu.wang@ionos.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@ionos.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
dgilbert: Fixed up some tabs
Adaptive polling measures the execution time of the polling check plus
handlers called when a polled event becomes ready. Handlers can take a
significant amount of time, making it look like polling was running for
a long time when in fact the event handler was running for a long time.
For example, on Linux the io_submit(2) syscall invoked when a virtio-blk
device's virtqueue becomes ready can take 10s of microseconds. This
can exceed the default polling interval (32 microseconds) and cause
adaptive polling to stop polling.
By excluding the handler's execution time from the polling check we make
the adaptive polling calculation more accurate. As a result, the event
loop now stays in polling mode where previously it would have fallen
back to file descriptor monitoring.
The following data was collected with virtio-blk num-queues=2
event_idx=off using an IOThread. Before:
168k IOPS, IOThread syscalls:
9837.115 ( 0.020 ms): IO iothread1/620155 io_submit(ctx_id: 140512552468480, nr: 16, iocbpp: 0x7fcb9f937db0) = 16
9837.158 ( 0.002 ms): IO iothread1/620155 write(fd: 103, buf: 0x556a2ef71b88, count: 8) = 8
9837.161 ( 0.001 ms): IO iothread1/620155 write(fd: 104, buf: 0x556a2ef71b88, count: 8) = 8
9837.163 ( 0.001 ms): IO iothread1/620155 ppoll(ufds: 0x7fcb90002800, nfds: 4, tsp: 0x7fcb9f1342d0, sigsetsize: 8) = 3
9837.164 ( 0.001 ms): IO iothread1/620155 read(fd: 107, buf: 0x7fcb9f939cc0, count: 512) = 8
9837.174 ( 0.001 ms): IO iothread1/620155 read(fd: 105, buf: 0x7fcb9f939cc0, count: 512) = 8
9837.176 ( 0.001 ms): IO iothread1/620155 read(fd: 106, buf: 0x7fcb9f939cc0, count: 512) = 8
9837.209 ( 0.035 ms): IO iothread1/620155 io_submit(ctx_id: 140512552468480, nr: 32, iocbpp: 0x7fca7d0cebe0) = 32
174k IOPS (+3.6%), IOThread syscalls:
9809.566 ( 0.036 ms): IO iothread1/623061 io_submit(ctx_id: 140539805028352, nr: 32, iocbpp: 0x7fd0cdd62be0) = 32
9809.625 ( 0.001 ms): IO iothread1/623061 write(fd: 103, buf: 0x5647cfba5f58, count: 8) = 8
9809.627 ( 0.002 ms): IO iothread1/623061 write(fd: 104, buf: 0x5647cfba5f58, count: 8) = 8
9809.663 ( 0.036 ms): IO iothread1/623061 io_submit(ctx_id: 140539805028352, nr: 32, iocbpp: 0x7fd0d0388b50) = 32
Notice that ppoll(2) and eventfd read(2) syscalls are eliminated because
the IOThread stays in polling mode instead of falling back to file
descriptor monitoring.
As usual, polling is not implemented on Windows so this patch ignores
the new io_poll_read() callback in aio-win32.c.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20211207132336.36627-2-stefanha@redhat.com
[Fixed up aio_set_event_notifier() calls in
tests/unit/test-fdmon-epoll.c added after this series was queued.
--Stefan]
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
destination:
../qemu/build/qemu-system-x86_64 -enable-kvm -netdev tap,id=hn0,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown -device e1000,netdev=hn0,mac=50:52:54:00:11:22 -boot c -drive if=none,file=./Fedora-rdma-server-migration.qcow2,id=drive-virtio-disk0 -device virtio-blk-pci,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0 -m 2048 -smp 2 -device piix3-usb-uhci -device usb-tablet -monitor stdio -vga qxl -spice streaming-video=filter,port=5902,disable-ticketing -incoming rdma:192.168.22.23:8888
qemu-system-x86_64: -spice streaming-video=filter,port=5902,disable-ticketing: warning: short-form boolean option 'disable-ticketing' deprecated
Please use disable-ticketing=on instead
QEMU 6.0.50 monitor - type 'help' for more information
(qemu) trace-event qemu_rdma_block_for_wrid_miss on
(qemu) dest_init RDMA Device opened: kernel name rxe_eth0 uverbs device name uverbs2, infiniband_verbs class device path /sys/class/infiniband_verbs/uverbs2, infiniband class device path /sys/class/infiniband/rxe_eth0, transport: (2) Ethernet
qemu_rdma_block_for_wrid_miss A Wanted wrid CONTROL SEND (2000) but got CONTROL RECV (4000)
source:
../qemu/build/qemu-system-x86_64 -enable-kvm -netdev tap,id=hn0,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown -device e1000,netdev=hn0,mac=50:52:54:00:11:22 -boot c -drive if=none,file=./Fedora-rdma-server.qcow2,id=drive-virtio-disk0 -device virtio-blk-pci,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0 -m 2048 -smp 2 -device piix3-usb-uhci -device usb-tablet -monitor stdio -vga qxl -spice streaming-video=filter,port=5901,disable-ticketing -S
qemu-system-x86_64: -spice streaming-video=filter,port=5901,disable-ticketing: warning: short-form boolean option 'disable-ticketing' deprecated
Please use disable-ticketing=on instead
QEMU 6.0.50 monitor - type 'help' for more information
(qemu)
(qemu) trace-event qemu_rdma_block_for_wrid_miss on
(qemu) migrate -d rdma:192.168.22.23:8888
source_resolve_host RDMA Device opened: kernel name rxe_eth0 uverbs device name uverbs2, infiniband_verbs class device path /sys/class/infiniband_verbs/uverbs2, infiniband class device path /sys/class/infiniband/rxe_eth0, transport: (2) Ethernet
(qemu) qemu_rdma_block_for_wrid_miss A Wanted wrid WRITE RDMA (1) but got CONTROL RECV (4000)
NOTE: we use soft RoCE as the rdma device.
[root@iaas-rpma images]# rdma link show rxe_eth0/1
link rxe_eth0/1 state ACTIVE physical_state LINK_UP netdev eth0
This migration could not be completed when out of order(OOO) CQ event occurs.
The send queue and receive queue shared a same completion queue, and
qemu_rdma_block_for_wrid() will drop the CQs it's not interested in. But
the dropped CQs by qemu_rdma_block_for_wrid() could be later CQs it wants.
So in this case, qemu_rdma_block_for_wrid() will block forever.
OOO cases will occur in both source side and destination side. And a
forever blocking happens on only SEND and RECV are out of order. OOO between
'WRITE RDMA' and 'RECV' doesn't matter.
below the OOO sequence:
source destination
rdma_write_one() qemu_rdma_registration_handle()
1. S1: post_recv X D1: post_recv Y
2. wait for recv CQ event X
3. D2: post_send X ---------------+
4. wait for send CQ send event X (D2) |
5. recv CQ event X reaches (D2) |
6. +-S2: post_send Y |
7. | wait for send CQ event Y |
8. | recv CQ event Y (S2) (drop it) |
9. +-send CQ event Y reaches (S2) |
10. send CQ event X reaches (D2) -----+
11. wait recv CQ event Y (dropped by (8))
Although a hardware IB works fine in my a hundred of runs, the IB specification
doesn't guaratee the CQ order in such case.
Here we introduce a independent send completion queue to distinguish
ibv_post_send completion queue from the original mixed completion queue.
It helps us to poll the specific CQE we are really interested in.
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
The responder mr registering with ODP will sent RNR NAK back to
the requester in the face of the page fault.
---------
ibv_poll_cq wc.status=13 RNR retry counter exceeded!
ibv_poll_cq wrid=WRITE RDMA!
---------
ibv_advise_mr(3) helps to make pages present before the actual IO is
conducted so that the responder does page fault as little as possible.
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Previously, for the fsdax mem-backend-file, it will register failed with
Operation not supported. In this case, we can try to register it with
On-Demand Paging[1] like what rpma_mr_reg() does on rpma[2].
[1]: https://community.mellanox.com/s/article/understanding-on-demand-paging--odp-x
[2]: http://pmem.io/rpma/manpages/v0.9.0/rpma_mr_reg.3
CC: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
backtrace:
'0x00007ffff5f44ec2 in __ibv_dereg_mr_1_1 (mr=0x7fff1007d390) at /home/lizhijian/rdma-core/libibverbs/verbs.c:478
478 void *addr = mr->addr;
(gdb) bt
#0 0x00007ffff5f44ec2 in __ibv_dereg_mr_1_1 (mr=0x7fff1007d390) at /home/lizhijian/rdma-core/libibverbs/verbs.c:478
#1 0x0000555555891fcc in rdma_delete_block (block=<optimized out>, rdma=0x7fff38176010) at ../migration/rdma.c:691
#2 qemu_rdma_cleanup (rdma=0x7fff38176010) at ../migration/rdma.c:2365
#3 0x00005555558925b0 in qio_channel_rdma_close_rcu (rcu=0x555556b8b6c0) at ../migration/rdma.c:3073
#4 0x0000555555d652a3 in call_rcu_thread (opaque=opaque@entry=0x0) at ../util/rcu.c:281
#5 0x0000555555d5edf9 in qemu_thread_start (args=0x7fffe88bb4d0) at ../util/qemu-thread-posix.c:541
#6 0x00007ffff54c73f9 in start_thread () at /lib64/libpthread.so.0
#7 0x00007ffff53f3b03 in clone () at /lib64/libc.so.6 '
Signed-off-by: Li Zhijian <lizhijian@cn.fujitsu.com>
Message-Id: <20210708144521.1959614-1-lizhijian@cn.fujitsu.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Markus Armbruster
Li Zhijian
Stefan Hajnoczi
Juan Quintela
Vladimir Sementsov-Ogievskiy
Dr. David Alan Gilbert
manish.mishra
Fiona Ebner
Daniel P. Berrangé
Leonardo Bras
Jack Wang
Li Zhijian