- 05 Mar, 2005 40 commits
-
-
Roland Dreier authored
Factor the allocation and freeing of completion queue buffers into mthca_alloc_cq_buf() and mthca_free_cq_buf(). This makes the code more readable and will eventually make handling userspace CQs simpler (the kernel doesn't have to allocate a buffer at all). Signed-off-by:
Roland Dreier <roland@topspin.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Roland Dreier authored
Add a mthca_write_db_rec() to wrap writing doorbell records. On 64-bit archs, this is just a 64-bit write, while on 32-bit archs it splits the write into two 32-bit writes with a memory barrier to make sure the two halves of the record are written in the correct order. Signed-off-by:
-
Roland Dreier authored
Mem-free mode requires the driver to allocate additional doorbell pages for each user access region. Add support for this in mthca_memfree.c, and have the driver allocate a table in db_tab for kernel use. Signed-off-by:
-
Roland Dreier authored
Have MAP_ICM_page() firmware command map assume pages are always the HCA-native 4K size rather than using the kernel's page size. This will make handling doorbell pages for mem-free mode simpler. Signed-off-by:
-
Roland Dreier authored
Slightly improve debugging output for UNMAP_ICM and MODIFY_QP firmware commands. Signed-off-by:
-
Roland Dreier authored
Update interrupt handling code to handle mem-free mode. While we're at it, improve the Tavor interrupt handling to avoid an extra MMIO read of the event cause register. Signed-off-by:
-
Roland Dreier authored
Add code to initialize EQ context properly in both Tavor and mem-free mode. Signed-off-by:
-
Roland Dreier authored
Add support for mem-free mode to memory region code. This mostly amounts to properly munging between keys and indices. Signed-off-by:
-
Roland Dreier authored
Add support for mapping more memory into HCA's context to cover context tables when new objects are allocated. Pass the object size into mthca_alloc_icm_table(), reference count the ICM chunks, and add new mthca_table_get() and mthca_table_put() functions to handle mapping memory when allocating or destroying objects. Signed-off-by:
-
Roland Dreier authored
Add support for allocating user access regions (UARs). Use this to allocate a region for kernel at driver init instead using hard-coded MTHCA_KAR_PAGE index. Signed-off-by:
-
Roland Dreier authored
Move the request/ioremap of regions related to event handling into mthca_eq.c. Map the correct regions depending on whether we're in Tavor or native mem-free mode. Signed-off-by:
-
Michael S. Tsirkin authored
Remove support for unsignaled receive requests. This is a non-standard extension to the IB spec that is not used by any known applications or protocols, and is not supported by newer hardware. Signed-off-by:
Michael S. Tsirkin <mst@mellanox.co.il> Signed-off-by:
-
Roland Dreier authored
Simplify some of the code for CQ handling slightly. Signed-off-by:
-
Michael S. Tsirkin authored
Locking during the poll cq operation can be reduced by locking the cq while qp is being removed from the qp array. This also avoids an extra atomic operation for reference counting. Signed-off-by:
-
Michael S. Tsirkin authored
Avoid taking the CQ table lock in the fast path path by using synchronize_irq() after removing a CQ from the table to make sure that no completion events are still in progress. This gets a nice speedup (about 4%) in IP over IB on my hardware. Signed-off-by:
-
Michael S. Tsirkin authored
Clean up CQ code so that we only calculate the address of a CQ entry once when using it. Signed-off-by:
-
Sean Hefty authored
Fix ib_find_cached_gid() to return the correct port number relative to the port numbering used by the device. Signed-off-by:
Sean Hefty <sean.hefty@intel.com> Signed-off-by:
-
Roland Dreier authored
Fix up locking for IPoIB path table. Make sure that destruction of address handles, neighbour info and path structs is locked properly to avoid races and deadlocks. (Problem originally diagnosed by Shirley Ma) Signed-off-by:
-
Shirley Ma authored
ipoib_put_ah() may call ipoib_free_ah(), which might take the device's lock. Therefore we need to make sure we don't call ipoib_put_ah() when holding the lock already. Signed-off-by:
Shirley Ma <xma@us.ibm.com> Signed-off-by:
-
Shirley Ma authored
IPoIB small fixes: Initialize path->ah to NULL, and fix dereference after free of neigh in error path of neigh_add_path(). Signed-off-by:
-
Roland Dreier authored
Make IPoIB data_debug_level module parameter static to the single file where it is used. Also Rename IPoIB module parameter variable from "debug_level" to "ipoib_debug_level". This avoids possible name clashes if IPoIB is built into the kernel. We use module_param_named so that the user-visible parameter names remain the same. Signed-off-by:
Tom Duffy <tduffy@sun.com> Signed-off-by:
-
Shirley Ma authored
Change uses of list_for_each_entry() where the loop variable is freed inside the loop to list_for_each_entry_safe(). Signed-off-by:
-
Roland Dreier authored
Fix memory leak when posting a receive buffer (pointed out by Shirley Ma). Signed-off-by:
-
Roland Dreier authored
MTHCA_RESET_VALUE must always be swapped, since the HCA expects to see it in big-endian order and we write it with writel. This means on little-endian systems we have to swap it to big-endian order before writing, and on big-endian systems we need to swap it to make up for the additional swap that writel will do. This fixes resetting the HCA on big-endian machines. Signed-off-by:
-
Roland Dreier authored
Add missing break statements in switch in mthca_profile.c (pointed out by Michael Tsirkin). Signed-off-by:
-
Tom Duffy authored
Fix some sparse warnings by making sure we have appropriate "extern" declarations visible. Signed-off-by:
-
shaharf@voltaire.com authored
Fix bug when deregistering a vendor class MAD agent. Signed-off-by:
-
Hal Rosenstock authored
Remove unneeded MAD agent registration by using a single agent for both directed-route and LID-routed MADs. Signed-off-by:
Hal Rosenstock <halr@voltaire.com> Signed-off-by:
-
Yoichi Yuasa authored
CC arch/mips/kernel/ptrace.o arch/mips/kernel/ptrace.c: In function 'do_syscall_trace': arch/mips/kernel/ptrace.c:310: warning: implicit declaration of function 'audit_syscall_entry' arch/mips/kernel/ptrace.c:310: error: 'struct pt_regs' has no member named 'orig_eax' arch/mips/kernel/ptrace.c:314: warning: implicit declaration of function 'audit_syscall_exit' Signed-off-by:
Yoichi Yuasa <yuasa@hh.iij4u.or.jp> Signed-off-by:
-
Randy Dunlap authored
u64's are not longlongs on sparc64 and ppc64. Signed-off-by:
Randy Dunlap <rddunlap@osdl.org> Signed-off-by:
-
Neil Brown authored
This patch should fix a problem that has been experienced on at-least one busy NFS server, but it has not had lots of testing yet. If -mm could provide that ..... The rpc auth cache currently differentiates between a reference due to being in a hash chain (signalled by CACHE_HASHED flag) and any other reference (counted in refcnt). This is an artificial difference due to an historical accident, and it makes cache_put unsafe. This patch removes the distinction so now existance in a hash chain is counted just like any other reference. Thus a race window in cache_put is closed. Signed-off-by:
Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by:
-
Neil Brown authored
also kmalloc the cache one entry at a time, instead of in one big slab. Signed-off-by:
-
Neil Brown authored
The patch below contains the following cleanups: - make some needlessly global code static Signed-off-by:
Adrian Bunk <bunk@stusta.de> Signed-off-by:
-
Neil Brown authored
The pg_authenticate (now pg_authenticate_obsolete) callback was only being used by the nfs4 client callback code to circumvent the svcauth_unix code's insistence on checking all requests against the export table. With that problem solved, we no longer need it. Signed-off-by:
J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by:
-
Neil Brown authored
On lockd callbacks, we're a client, and the source address is that of a server, so we shouldn't be trying to match the source address of the callback request against our export table. Signed-off-by:
-
Neil Brown authored
Use new pg_authenticate method to simplify nfs4 callback authentication. This also has the effect of changing the error return from rejectedcred to badcred. I believe the change is correct. Signed-off-by:
-
Neil Brown authored
svcauth_null_accept() and svcauth_unix_accept() are currently hard-wired to check the source ip address on an incoming request against the export table, which make sense for nfsd but not necessarily for other rpc-based services. So instead we have the accept() method call a program-specific pg_authenticate() method. We also move the call to this method into svc_process instead of calling it from the flavor-specific accept() routines. Signed-off-by:
-
Neil Brown authored
Later patches remove pg_authenticate and use the name for a different purpose; so rename it to pg_authenticate_obsolete for now. Signed-off-by:
-
Neil Brown authored
Add a set_client method to the server rpc auth_ops struct, used to set the client (for the purposes of nfsd export authorization) using flavor-specific information. Signed-off-by:
-
Neil Brown authored
find_exported_dentry() declares char nbuf[NAME_MAX+1]; in 2 separate places, and gcc allocates space on the stack for both of them. Having just one of them will suffice, if we can put put with its scope. Reduces function stack usage on x86-32 from 0x230 to 0x130. Signed-off-by:
-
