aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6Linus Torvalds2008-12-28
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6: (31 commits) [CIFS] Remove redundant test [CIFS] make sure that DFS pathnames are properly formed Remove an already-checked error condition in SendReceiveBlockingLock Streamline SendReceiveBlockingLock: Use "goto out:" in an error condition Streamline SendReceiveBlockingLock: Use "goto out:" in an error condition [CIFS] Streamline SendReceive[2] by using "goto out:" in an error condition Slightly streamline SendReceive[2] Check the return value of cifs_sign_smb[2] [CIFS] Cleanup: Move the check for too large R/W requests [CIFS] Slightly simplify wait_for_free_request(), remove an unnecessary "else" branch Simplify allocate_mid() slightly: Remove some unnecessary "else" branches [CIFS] In SendReceive, move consistency check out of the mutexed region cifs: store password in tcon cifs: have calc_lanman_hash take more granular args cifs: zero out session password before freeing it cifs: fix wait_for_response to time out sleeping processes correctly [CIFS] Can not mount with prefixpath if root directory of share is inaccessible [CIFS] various minor cleanups pointed out by checkpatch script [CIFS] fix typo [CIFS] remove sparse warning ... Fix trivial conflict in fs/cifs/cifs_fs_sb.h due to comment changes for the CIFS_MOUNT_xyz bit definitions between cifs updates and security updates.
| * [CIFS] Remove redundant testJulia Lawall2008-12-25
| | | | | | | | | | | | | | | | | | | | | | In fs/cifs/cifssmb.c, pLockData is tested for being NULL at the beginning of the function, and not reassigned subsequently. A simplified version of the semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] make sure that DFS pathnames are properly formedSteve French2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The paths in a DFS request are supposed to only have a single preceding backslash, but we are sending them with a double backslash. This is exposing a bug in Windows where it also sends a path in the response that has a double backslash. The existing code that builds the mount option string however expects a double backslash prefix in a couple of places when it tries to use the path returned by build_path_from_dentry. Fix compose_mount_options to expect properly formed DFS paths (single backslash at front). Also clean up error handling in that function. There was a possible NULL pointer dereference and situations where a partially built option string would be returned. Tested against Samba 3.0.28-ish server and Samba 3.3 and Win2k8. CC: Stable <stable@kernel.org> Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * Remove an already-checked error condition in SendReceiveBlockingLockVolker Lendecke2008-12-25
| | | | | | | | | | | | | | Remove an already-checked error condition in SendReceiveBlockingLock Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * Streamline SendReceiveBlockingLock: Use "goto out:" in an error conditionVolker Lendecke2008-12-25
| | | | | | | | | | | | | | Streamline SendReceiveBlockingLock: Use "goto out:" in an error condition Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * Streamline SendReceiveBlockingLock: Use "goto out:" in an error conditionVolker Lendecke2008-12-25
| | | | | | | | | | | | | | Streamline SendReceiveBlockingLock: Use "goto out:" in an error condition Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] Streamline SendReceive[2] by using "goto out:" in an error conditionSteve French2008-12-25
| | | | | | | | | | Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * Slightly streamline SendReceive[2]Volker Lendecke2008-12-25
| | | | | | | | | | | | | | | | | | Slightly streamline SendReceive[2] Remove an else branch by naming the error condition what it is Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * Check the return value of cifs_sign_smb[2]Volker Lendecke2008-12-25
| | | | | | | | | | | | | | Check the return value of cifs_sign_smb[2] Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] Cleanup: Move the check for too large R/W requestsSteve French2008-12-25
| | | | | | | | | | | | | | This avoids an unnecessary else branch Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] Slightly simplify wait_for_free_request(), remove an unnecessary ↵Volker Lendecke2008-12-25
| | | | | | | | | | | | | | | | | | | | "else" branch This is no functional change, because in the "if" branch we do an early "return 0;". Signed-off-by: Volker Lendecke <vl@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * Simplify allocate_mid() slightly: Remove some unnecessary "else" branchesVolker Lendecke2008-12-25
| | | | | | | | | | | | | | | | Simplify allocate_mid() slightly: Remove some unnecessary "else" branches Signed-off-by: Volker Lendecke <vl@samba.org> Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] In SendReceive, move consistency check out of the mutexed regionVolker Lendecke2008-12-25
| | | | | | | | | | | | | | | | | | inbuf->smb_buf_length does not change in in wait_for_free_request() or in allocate_mid(), so we can check it early. Signed-off-by: Volker Lendecke <vl@samba.org> Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: store password in tconJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | cifs: store password in tcon Each tcon has its own password for share-level security. Store it in the tcon and wipe it clean and free it when freeing the tcon. When doing the tree connect with share-level security, use the tcon password instead of the session password. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: have calc_lanman_hash take more granular argsJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | cifs: have calc_lanman_hash take more granular args We need to use this routine to encrypt passwords associated with the tcon too. Don't assume that the password will be attached to the smb_session. Also, make some of the values in the lower encryption functions const since they aren't changed. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: zero out session password before freeing itJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | cifs: zero out session password before freeing it ...just to be on the safe side. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: fix wait_for_response to time out sleeping processes correctlyJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | cifs: fix wait_for_response to time out sleeping processes correctly The current scheme that CIFS uses to sleep and wait for a response is not quite what we want. After sending a request, wait_for_response puts the task to sleep with wait_event(). One of the conditions for wait_event is a timeout (using time_after()). The problem with this is that there is no guarantee that the process will ever be woken back up. If the server stops sending data, then cifs_demultiplex_thread will leave its response queue sleeping. I think the only thing that saves us here is the fact that cifs_dnotify_thread periodically (every 15s) wakes up sleeping processes on all response_q's that have calls in flight. This makes for unnecessary wakeups of some processes. It also means large variability in the timeouts since they're all woken up at once. Instead of this, put the tasks to sleep with wait_event_timeout. This makes them wake up on their own if they time out. With this change, cifs_dnotify_thread should no longer be needed. I've been testing this in conjunction with some other patches that I'm working on. It doesn't seem to affect performance at all with with heavy I/O. Identical iozone -ac runs complete in almost exactly the same time (<1% difference in times). Thanks to Wasrshi Nimara for initially pointing this out. Wasrshi, it would be nice to know whether this patch also helps your testcase. Signed-off-by: Jeff Layton <jlayton@redhat.com> Cc: Wasrshi Nimara <warshinimara@gmail.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] Can not mount with prefixpath if root directory of share is inaccessibleSteve French2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Windows allows you to deny access to the top of a share, but permit access to a directory lower in the path. With the prefixpath feature of cifs (ie mounting \\server\share\directory\subdirectory\etc.) this should have worked if the user specified a prefixpath which put the root of the mount at a directory to which he had access, but we still were doing a lookup on the root of the share (null path) when we should have been doing it on the prefixpath subdirectory. This fixes Samba bug # 5925 Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] various minor cleanups pointed out by checkpatch scriptSteve French2008-12-25
| | | | | | | | Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] fix typoSteve French2008-12-25
| | | | | | | | Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] remove sparse warningSteve French2008-12-25
| | | | | | | | Signed-off-by: Steve French <sfrench@us.ibm.com>
| * [CIFS] add mount option to send mandatory rather than advisory locksSteve French2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some applications/subsystems require mandatory byte range locks (as is used for Windows/DOS/OS2 etc). Sending advisory (posix style) byte range lock requests (instead of mandatory byte range locks) can lead to problems for these applications (which expect that other clients be prevented from writing to portions of the file which they have locked and are updating). This mount option allows mounting cifs with the new mount option "forcemand" (or "forcemandatorylock") in order to have the cifs client use mandatory byte range locks (ie SMB/CIFS/Windows/NTFS style locks) rather than posix byte range lock requests, even if the server would support posix byte range lock requests. This has no effect if the server does not support the CIFS Unix Extensions (since posix style locks require support for the CIFS Unix Extensions), but for mounts to Samba servers this can be helpful for Wine and applications that require mandatory byte range locks. Acked-by: Jeff Layton <jlayton@redhat.com> CC: Alexander Bokovoy <ab@samba.org> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: make ipv6_connect take a TCP_Server_Info argJeff Layton2008-12-25
| | | | | | | | | | Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: make ipv4_connect take a TCP_Server_Info argJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | In order to unify the smb_send routines, we need to reorganize the routines that connect the sockets. Have ipv4_connect take a TCP_Server_Info pointer and get the necessary fields from that. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: don't declare smb_vol info on the stackJeff Layton2008-12-25
| | | | | | | | | | | | | | struct smb_vol is fairly large, it's probably best to kzalloc it... Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: move allocation of new TCP_Server_Info into separate functionJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | Clean up cifs_mount a bit by moving the code that creates new TCP sessions into a separate function. Have that function search for an existing socket and then create a new one if one isn't found. Also reorganize the initializion of TCP_Server_Info a bit to prepare for cleanup of the socket connection code. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: account for IPv6 in ses->serverName and clean up netbios name handlingJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current code for setting the session serverName is IPv4-specific. Allow it to be an IPv6 address as well. Use NIP* macros to set the format. This also entails increasing the length of the serverName field, so declare a new macro for RFC1001 name length and use it in the appropriate places. Finally, drop the unicode_server_Name field from TCP_Server_Info since it's not used. We can add it back later if needed, but for now it just wastes memory. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: make dnotify thread experimental codeJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | Now that tasks sleeping in wait_for_response will time out on their own, we're not reliant on the dnotify thread to do this. Mark it as experimental code for now. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: convert tcpSem to a mutexJeff Layton2008-12-25
| | | | | | | | | | | | | | Mutexes are preferred for single-holder semaphores... Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: take module reference when starting cifsdJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | cifsd can outlive the last cifs mount. We need to hold a module reference until it exits to prevent someone from unplugging the module until we're ready. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: display addr and prefixpath options in /proc/mountsJeff Layton2008-12-25
| | | | | | | | | | | | | | | | | | Have cifs_show_options display the addr and prefixpath options in /proc/mounts. Reduce struct dereferencing by adding some local variables. Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
| * cifs: remove unused SMB session pointer from struct mid_q_entryJeff Layton2008-12-25
| | | | | | | | | | Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve French <sfrench@us.ibm.com>
* | Merge branch 'for-linus' of ↵Linus Torvalds2008-12-28
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (26 commits) IB/mlx4: Set ownership bit correctly when copying CQEs during CQ resize RDMA/nes: Remove tx_free_list RDMA/cma: Add IPv6 support RDMA/addr: Add support for translating IPv6 addresses mlx4_core: Delete incorrect comment mlx4_core: Add support for multiple completion event vectors IB/iser: Avoid recv buffer exhaustion caused by unexpected PDUs IB/ehca: Remove redundant test of vpage IB/ehca: Replace modulus operations in flush error completion path IB/ipath: Add locking for interrupt use of ipath_pd contexts vs free IB/ipath: Fix spi_pioindex value IB/ipath: Only do 1X workaround on rev1 chips IB/ipath: Don't count IB symbol and link errors unless link is UP IB/ipath: Check return value of dma_map_single() IB/ipath: Fix PSN of send WQEs after an RDMA read resend RDMA/nes: Cleanup warnings RDMA/nes: Add loopback check to make_cm_node() RDMA/nes: Check cqp_avail_reqs is empty after locking the list RDMA/nes: Fix TCP compliance test failures RDMA/nes: Forward packets for a new connection with stale APBVT entry ...
| | \
| | \
| | \
| | \
| | \
| | \
| | \
| | \
| *-------. \ Merge branches 'cma', 'ehca', 'ipath', 'iser', 'mlx4' and 'nes' into for-nextRoland Dreier2008-12-24
| |\ \ \ \ \ \
| | | | | | * | RDMA/nes: Remove tx_free_listFaisal Latif2008-12-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is no lock protecting tx_free_list thus causing a system crash when skb_dequeue() is called and the list is empty. Since it did not give any performance boost under heavy load, remove it to simplify the code. Replace get_free_pkt() with dev_alloc_skb() to allocate MAX_CM_BUFFER skb for connection establishment/teardown as well as MPA request/response. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Cleanup warningsChien Tung2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Wrap NES_DEBUG and assert macros with do while (0) to avoid ambiguous else. No one is using sk_buff * returned from form_cm_frame(), so drop the return. drop_packet() should not be incrementing reset counter on receiving a FIN. Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Add loopback check to make_cm_node()Chien Tung2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Check for loopback connection in make_cm_node(). Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Check cqp_avail_reqs is empty after locking the listFaisal Latif2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Between the first empty list check and locking the list, the list can change. Check it again after it is locked to make sure the list is still not empty. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Fix TCP compliance test failuresFaisal Latif2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ANVL testing showed we are not handling all cm_node states during connection establishment. Add missing state handlers and fix sequence number send reset in handle_tcp_options(). Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Forward packets for a new connection with stale APBVT entryFaisal Latif2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Under heavy traffic, there is a small windows when an APBVT entry is not yet removed and a new connection is established. Packets for the new connection are dropped until APBVT entry is removed. This patch will forward the packets instead of dropping them. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Avoid race between MPA request and reset event to rdma_cmFaisal Latif2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In passive open, after indicating MPA request to rdma_cm, an incoming RST would fire a reset event to rdma_cm causing it to crash, since the current state is not connected. The solution is to wait for nes_accept() or nes_reject() before firing the reset event. If nes_accept() or nes_reject() is already done, then the reset event will be fired when RST is processed. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Lock down connected_nodes list while processing itFaisal Latif2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | While processing connected_nodes list, we would release the lock when we need to send reset to remote partner. That created a window where the list can be modified. Change this into a two step process: place nodes that need processing on a local list then process the local list. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * | RDMA/nes: Cleanup cqp_request list usageFaisal Latif2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use nes_free_cqp_request() instead of open coding. Change some continue to break in nes_cm_timer_tick, because send_entry used to be a list processed in a loop (so continue went to the next item). Now it is a single item, so using break is correct. Signed-off-by: Faisal Latif <faisal.latif@intel.com> Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | * | | IB/mlx4: Set ownership bit correctly when copying CQEs during CQ resizeJack Morgenstein2008-12-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When resizing a CQ, when copying over unpolled CQEs from the old CQE buffer to the new buffer, the ownership bit must be set appropriately for the new buffer, or the ownership bit in the new buffer gets corrupted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | * | | mlx4_core: Delete incorrect commentRoland Dreier2008-12-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The comment about a "Conditional on hca_type" was cut-and-pasted from the mthca driver, and doesn't apply to mlx4 (since only one type of HCA is handled by mlx4). So just delete it. Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | * | | mlx4_core: Add support for multiple completion event vectorsYevgeny Petrilin2008-12-22
| | | | | |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | * / / IB/iser: Avoid recv buffer exhaustion caused by unexpected PDUsDavid Disseldorp2008-12-21
| | | | |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | iSCSI/iSER targets may send PDUs without a prior request from the initiator. RFC 5046 refers to these PDUs as "unexpected". NOP-In PDUs with itt=RESERVED and Asynchronous Message PDUs occupy this category. The amount of active "unexpected" PDU's an iSER target may have at any time is governed by the MaxOutstandingUnexpectedPDUs key, which is not yet supported. Currently when an iSER target sends an "unexpected" PDU, the initiators recv buffer consumed by the PDU is not replaced. If over initial_post_recv_bufs_num "unexpected" PDUs are received then the receive queue will run out of receive work requests entirely. This patch ensures recv buffers consumed by "unexpected" PDUs are replaced in the next iser_post_receive_control() call. Signed-off-by: David Disseldorp <ddiss@sgi.com> Signed-off-by: Ken Sandars <ksandars@sgi.com> Acked-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | * | | IB/ipath: Add locking for interrupt use of ipath_pd contexts vs freeDave Olson2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes timing race resulting in panic. Not a performance sensitive path. Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | * | | IB/ipath: Fix spi_pioindex valueDave Olson2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ipath_piobufbase was a single value offset, but is multiple values on newer chips, so use only the 32 bits for the 2K buffers (4K buffers are currently used only by the driver). Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | * | | IB/ipath: Only do 1X workaround on rev1 chipsDave Olson2008-12-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Dave Olson <dave.olson@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-11-20 05:34:21 -0500