rdma: export yield_until_fd_readable()
The RDMA event channel can be made non-blocking just like a TCPsocket. Exporting this function allows us to yield so that theQEMU monitor remains available.
Reviewed-by: Juan Quintela <quintela@redhat.com>Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>...
rdma: export throughput w/ MigrationStats QMP
This exposes throughput (in megabits/sec) through QMP.
Reviewed-by: Juan Quintela <quintela@redhat.com>Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>Reviewed-by: Chegu Vinod <chegu_vinod@hp.com>Tested-by: Chegu Vinod <chegu_vinod@hp.com>...
rdma: introduce qemu_file_mode_is_not_valid()
QEMUFileRDMA also has read and write modes. This function is nowshared to reduce code duplication.
Reviewed-by: Juan Quintela <quintela@redhat.com>Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>Reviewed-by: Chegu Vinod <chegu_vinod@hp.com>...
rdma: export qemu_fflush()
RDMA uses this to flush the control channel before sending itsown message to handle page registrations.
rdma: add documentation
docs/rdma.txt contains full documentation,wiki links, github url and contact information.
rdma: introduce qemu_update_position()
RDMA writes happen asynchronously, and thus the performance accountingalso needs to be able to occur asynchronously. This allows anybodyto call into savevm.c to update both f->pos as well as into arch_init.cto update the acct_info structure with up-to-date values when...
Revert "migration: do not sent zero pages in bulk stage"
Not sending zero pages breaks migration if a page is zeroat the source but not at the destination. This can e.g. happenif different BIOS versions are used at source and destination.It has also been reported that migration on pseries is completely...
migration: do not overwrite zero pages
on incoming migration do not memset pages to zero if they already read as zero.this will allocate a new zero page and consume memory unnecessarily. evenif we madvise a MADV_DONTNEED later this will only deallocate the memory...
arch_init/ram_load: add error message for block length mismatch
Makes it easier to debug situations where the source and target havedifferent ram blocks in a device and migration fails due to that, forinstance a BAR size change on a PCI device.
Signed-off-by: Alon Levy <alevy@redhat.com>...
Merge remote-tracking branch 'pmaydell/arm-devs.for-upstream' into staging
View all revisions | View revisions
Also available in: Atom