History | View | Annotate | Download (104.9 kB)
rdma: clean up of qemu_rdma_cleanup()
- It can't be determined by RDMAContext::cm_id != NULL if the connection is established or not.- RDMAContext::cm_id is leaked and not destroyed because it is set to NULL too early.- RDMAContext::qp is created by rdma_create_qp() so that it should be destroyed...
rdma: constify ram_chunk_{index, start, end}
Signed-off-by: Isaku Yamahata <yamahata@private.email.ne.jp>Signed-off-by: Juan Quintela <quintela@redhat.com>
rdma: silly ipv6 bugfix
My bad - but it's very important for us to warn the user thatIPv6 is broken on RoCE in linux right now, until linux releasesa fixed version.
Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
misc: Fix some typos in names and comments
Most typos were found using a modified version of codespell:
accross -> acrossissueing -> issuingTICNT_THRESHHOLD -> TICNT_THRESHOLDbandwith -> bandwidthVCARD_7816_PROPIETARY -> VCARD_7816_PROPRIETARYoccured -> occurred...
rdma: proper getaddrinfo() handling
getaddrinfo() already knows what it's doing,but it can potentially return multiple addresses.We need to handle that...
Reviewed-by: Orit Wasserman <owasserm@redhat.com>Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>...
rdma: IPv6 over Ethernet (RoCE) is broken in linux - workaround
We've gotten reports from multiple testers (including Frank Yangjieand myself) that RDMA IPv6 support over RocE (Ethernet) is brokenin linux.
A patch to Linux is still in review:
http://comments.gmane.org/gmane.linux.drivers.rdma/16448...
rdma: use resp.len after validation in qemu_rdma_registration_stop
resp.len is given from remote host. So should be validated before use.Otherwise memcpy can access beyond the buffer.
Cc: Michael R. Hines <mrhines@us.ibm.com>Reviewed-by: Orit Wasserman <owasserm@redhat.com>...
rdma: validate RDMAControlHeader::len
RMDAControlHeader::len is provided from remote, so validate it.
Reviewed-by: Orit Wasserman <owasserm@redhat.com>Reviewed-by: Michael R. Hines <mrhines@us.ibm.com>Signed-off-by: Isaku Yamahata <yamahata@private.email.ne.jp>...
rdma: check if RDMAControlHeader::len match transferred byte
RDMAControlHeader::len is provided from remote, so check if the valuematch the actual transferred byte_len.
Reviewed-by: Orit Wasserman <owasserm@redhat.com>Reviewed-by: Michael R. Hines <mrhines@us.ibm.com>...
rdma: correct newlines in error statements
Don't print newlines on the error_setg() function,but still allow newlines on fprintf().
Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>Message-id: 1375584894-9917-4-git-send-email-mrhines@linux.vnet.ibm.com...
rdma: don't use negative index to array
Reviewed-by: Michael R. Hines <mrhines@us.ibm.com>Signed-off-by: Isaku Yamahata <yamahata@private.email.ne.jp>Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>Message-id: 1375584894-9917-5-git-send-email-mrhines@linux.vnet.ibm.com...
rdma: qemu_rdma_post_send_control uses wrongly RDMA_WRID_MAX
RDMA_WRID_CONTROL should be used. And remove related work around.
Reviewed-by: Michael R. Hines <mrhines@us.ibm.com>Signed-off-by: Isaku Yamahata <yamahata@private.email.ne.jp>Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>...
rdma: use RDMA_WRID_READY
Reviewed-by: Michael R. Hines <mrhines@us.ibm.com>Signed-off-by: Isaku Yamahata <yamahata@private.email.ne.jp>Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>Message-id: 1375584894-9917-7-git-send-email-mrhines@linux.vnet.ibm.com...
rdma: memory leak RDMAContext::host
It is allocated by g_strdup(), so needs to be freed.
rdma: bugfix: make IPv6 support work
RDMA does not use sockets, so we cannot use many of the sockethelper functions, but we do use inet_parse() which givesRDMA all the necessary details of the connection parameters.
However, when testing with libvirt, a simple IPv6 migration test failed...
rdma: forgot to turn off the debugging flag
Ooops. We forgot to turn off the flag.
Signed-off-by: Michael R. Hines <mrhines@us.ibm.com>Message-id: 1375584894-9917-3-git-send-email-mrhines@linux.vnet.ibm.comSigned-off-by: Anthony Liguori <aliguori@us.ibm.com>
misc: Fix new typos in comments and strings
All these typos were found by codespell.
sould -> shouldemperical -> empiricalintialization -> initializationsuccessfuly -> successfullygaurantee -> guarantee
Fix also another error (before before) in the same context....
rdma: core logic
Code that does need to be visible is keptwell contained inside this file and this is the onlynew additional file to the entire patch.
This file includes the entire protocol and interfacesrequired to perform RDMA migration.
Also, the configure and Makefile modifications to link...