- 23 Sep, 2005 40 commits
-
-
Chuck Lever authored
Add a generic mechanism for skipping over transport-specific headers when constructing an RPC request. This removes another "xprt->stream" dependency. Test-plan: Write-intensive workload on a single mount point (try both UDP and TCP). Signed-off-by:
Chuck Lever <cel@netapp.com> Signed-off-by:
Trond Myklebust <Trond.Myklebust@netapp.com>
-
Chuck Lever authored
Split the RPC client's main socket write path into a TCP version and a UDP version to eliminate another dependency on the "xprt->stream" variable. Compiler optimization removes unneeded code from xs_sendpages, as this function is now called with some constant arguments. We can now cleanly perform transport protocol-specific return code testing and error recovery in each path. Test-plan: Millions of fsx operations. Performance characterization such as "sio" or "iozone". Examine oprofile results for any changes before and after this patch is applied. Version: Thu, 11 Aug 2005 16:08:46 -0400 Signed-off-by:
-
Chuck Lever authored
Create separate connection worker functions for managing UDP and TCP transport sockets. This eliminates several dependencies on "xprt->stream". Test-plan: Destructive testing (unplugging the network temporarily). Connectathon with v2, v3, and v4. Version: Thu, 11 Aug 2005 16:08:18 -0400 Signed-off-by:
-
Chuck Lever authored
Split the socket write space callback function into a TCP version and UDP version, eliminating one dependence on the "xprt->stream" variable. Keep the common pieces of this path in xprt.c so other transports can use it too. Test-plan: Write-intensive workload on a single mount point. Version: Thu, 11 Aug 2005 16:07:51 -0400 Signed-off-by:
-
Chuck Lever authored
Clean-up: change some comments to reflect the realities of the new RPC transport switch mechanism. Get rid of unused xprt_receive() prototype. Also, organize function prototypes in xprt.h by usage and scope. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:07:21 -0400 Signed-off-by:
-
Chuck Lever authored
Clean-up: remove only reference to xprt->pending from the socket transport implementation. This makes a cleaner interface for other transport implementations as well. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:06:52 -0400 Signed-off-by:
-
Chuck Lever authored
Clean-up: get rid of unnecessary socket.h and in.h includes in the generic parts of the RPC client. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:06:23 -0400 Signed-off-by:
-
Chuck Lever authored
Clean-up: get rid of a name reference to sockets in the generic parts of the RPC client by renaming the sockstate field in the rpc_xprt structure. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:05:53 -0400 Signed-off-by:
-
Chuck Lever authored
Clean-up: Replace the xprt_lock with something more aptly named. This lock single-threads the XID and request slot reservation process. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:05:26 -0400 Signed-off-by:
-
Chuck Lever authored
Clean-up: replace a name reference to sockets in the generic parts of the RPC client by renaming sock_lock in the rpc_xprt structure. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:05:00 -0400 Signed-off-by:
-
Chuck Lever authored
Reduce stack utilization of the RPC socket transport's send path. A couple of unlikely()s are added to ensure the compiler places the tail processing at the end of the csect. Test-plan: Millions of fsx operations. Performance characterization such as "sio" or "iozone". Version: Thu, 11 Aug 2005 16:04:30 -0400 Signed-off-by:
-
Chuck Lever authored
Introduce block header comments and a function naming convention to the socket transport implementation. Provide a debug setting for transports that is separate from RPCDBG_XPRT. Eliminate xprt_default_timeout(). Provide block comments for exposed interfaces in xprt.c, and eliminate the useless obvious comments. Convert printk's to dprintk's. Test-plan: Compile kernel with CONFIG_NFS enabled. Version: Thu, 11 Aug 2005 16:04:04 -0400 Signed-off-by:
-
Chuck Lever authored
Move the bulk of client-side socket-specific code into a separate source file, net/sunrpc/xprtsock.c. Test-plan: Millions of fsx operations. Performance characterization such as "sio" or "iozone". Destructive testing (unplugging the network temporarily, server reboots). Connectathon with v2, v3, and v4. Version: Thu, 11 Aug 2005 16:03:38 -0400 Signed-off-by:
-
Chuck Lever authored
Clean-up: Move some code that is common to both RPC client- and server-side socket transports into its own source file, net/sunrpc/socklib.c. Test-plan: Compile kernel with CONFIG_NFS enabled. Millions of fsx operations over UDP, client and server. Connectathon over UDP. Version: Thu, 11 Aug 2005 16:03:09 -0400 Signed-off-by:
-
Chuck Lever authored
The in-kernel portmapper does not require a reserved port for making bind queries. Test-plan: Tens of runs of the Connectathon locking suite with TCP and UDP against several other NFS server implementations using NFSv3, not NFSv4 (which doesn't require rpcbind). Version: Thu, 11 Aug 2005 16:02:43 -0400 Signed-off-by:
-
Chuck Lever authored
Implement a best practice: don't use exponential backoff when computing retransmit timeout values on TCP connections, but simply retransmit at regular intervals. This also fixes a bug introduced when xprt_reset_majortimeo() was added. Test-plan: Enable RPC debugging and watch timeout behavior on a NFS/TCP mount. Version: Thu, 11 Aug 2005 16:02:19 -0400 Signed-off-by:
-
Chuck Lever authored
Implement a best practice: for soft mounts, an rpcbind timeout should cause an RPC request to fail. This also provides an FSM hook for retrying an rpcbind with a different rpcbind protocol version. We'll use this later to try multiple rpcbind protocol versions when binding. To enable this, expose the RPC error code returned during a portmap request to the FSM so it can make some decision about how to report, retry, or fail the request. Test-plan: Hundreds of passes with connectathon NFSv3 locking suite, on the client and server. Version: Thu, 11 Aug 2005 16:01:53 -0400 Signed-off-by:
-
Chuck Lever authored
Fix up xprt_connect_status: the soft timeout logic was clobbering tk_status, so TCP connect errors were not properly reported on soft mounts. Test-plan: Destructive testing (unplugging the network temporarily). Connectathon with UDP and TCP. Version: Thu, 11 Aug 2005 16:01:28 -0400 Signed-off-by:
-
Trond Myklebust authored
Fixes a condition whereby the kernel is returning the non-POSIX error EBADCOOKIE to userspace. Signed-off-by: -
Olaf Kirch authored
[PATCH] Fix miscompare in __posix_lock_file If an application requests the same lock twice, the kernel should just leave the existing lock in place. Currently, it will install a second lock of the same type. Signed-off-by:
Olaf Kirch <okir@suse.de> Signed-off-by:
-
Trond Myklebust authored
When doing a rename on top of an existing file that is not in use, the inode of the overwritten file will remain in the icache. The fix is to decrement i_nlink of the overwritten inode, like we do for unlink, rmdir etc already. Problem diagnosed by Olaf Kirch. This patch is a slight variation on his fix. Signed-off-by: -
Ivan Kokshaysky authored
I believe the change that broke things is introduction of pci_fixup_parent_subordinate_busnr(). The patch here does two things: - hunk #1 should fix the problems you've seen when you boot without additional "pci" kernel options; - hunk #2 supposedly fixes boot with "pci=assign-busses" option which otherwise hangs Acer TM81xx machines as reported. Please try this with and without "pci=assign-busses". If it boots, I'd like to see 'lspci -vvx' for both cases. Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Linus Torvalds authored
-
Sridhar Samudrala authored
Fix to allow SCTP_SHUTDOWN notifications to be received on 1-1 style SCTP SOCK_STREAM sockets. Add SCTP_SHUTDOWN notification to the receive queue before updating the state of the association. Signed-off-by:
Sridhar Samudrala <sri@us.ibm.com> Signed-off-by:
Neil Horman <nhorman@tuxdriver.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Harald Welte authored
This patch fixes a number of bugs. It cannot be reasonably split up in multiple fixes, since all bugs interact with each other and affect the same function: Bug #1: The event cache code cannot be called while a lock is held. Therefore, the call to ip_conntrack_event_cache() within ip_ct_refresh_acct() needs to be moved outside of the locked section. This fixes a number of 2.6.14-rcX oops and deadlock reports. Bug #2: We used to call ct_add_counters() for unconfirmed connections without holding a lock. Since the add operations are not atomic, we could race with another CPU. Bug #3: ip_ct_refresh_acct() lost REFRESH events in some cases where refresh (and the corresponding event) are desired, but no accounting shall be performed. Both, evenst and accounting implicitly depended on the skb parameter bein non-null. We now re-introduce a non-accounting "ip_ct_refresh()" variant to explicitly state the desired behaviour. Signed-off-by:
Harald Welte <laforge@netfilter.org> Signed-off-by:
-
Harald Welte authored
Signed-off-by:
-
Alexey Dobriyan authored
Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
Harald Welte authored
As noted by Alexey Dobriyan, the DEBUGP statement prints the wrong callID. Signed-off-by:
-
Vlad Drukker authored
Signed-off-by:
Vlad Drukker <vlad@storewiz.com> Acked-by:
Stephen Hemminger <shemminger@osdl.org> Signed-off-by:
-
Herbert Xu authored
Since the introduction of TSO pcount a year ago, it has been possible for tcp_fragment() to cause packets_out to decrease. Prior to that, tcp_retrans_try_collapse() was the only way for that to happen on the retransmission path. When this happens with Reno, it is possible for sasked_out to become invalid because it is only an estimate and not tied to any particular packet on the retransmission queue. Therefore we need to adjust sacked_out as well as left_out in the Reno case. The following patch does exactly that. This bug is pretty difficult to trigger in practice though since you need a SACKless peer with a retransmission that occurs just as the cached MTU value expires. Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by:
-
Paolo 'Blaisorblade' Giarrusso authored
I've recently discovered the real functionality of device-mapper snapshots, and since they are not well known, I've decided to write some docs for them. Signed-off-by:
Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Signed-off-by:
Alasdair G Kergon <agk@redhat.com> Signed-off-by:
-
Nick Wilson authored
nfs_readpage_release() causes an oops while accessing a file with NFS debugging turned on (echo 32767 > /proc/sys/sunrpc/nfs_debug) and a kernel built with CONFIG_DEBUG_SLAB. This patch moves the debugging statement above nfs_release_request() to avoid accessing freed memory. Signed-off-by:
Nick Wilson <njw@osdl.org> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by:
-
Rob Landley authored
Problem: In some circumstances, bd_claim() is returning the wrong error code. If we try to swapon an unused block device that isn't swap formatted, we get -EINVAL. But if that same block device is already mounted, we instead get -EBUSY, even though it still isn't a valid swap device. This issue came up on the busybox list trying to get the error message from "swapon -a" right. If a swap device is already enabled, we get -EBUSY, and we shouldn't report this as an error. But we can't distinguish the two -EBUSY conditions, which are very different errors. In the code, bd_claim() returns either 0 or -EBUSY, but in this case busy means "somebody other than sys_swapon has already claimed this", and _that_ means this block device can't be a valid swap device. So return -EINVAL there. Signed-off-by:
Rob Landley <rob@landley.net> Signed-off-by:
-
Glauber de Oliveira Costa authored
Fix some warnings and a build error when EXT3_DEBUG is enabled. Signed-off-by:
-
Clemens Buchacher authored
It's deprecated. Use "%s", __FUNCTION__ instead. Signed-off-by:
Clemens Buchacher <drizzd@aon.at> Signed-off-by:
Maximilian Attems <janitor@sternwelten.at> Signed-off-by:
Domen Puncer <domen@coderock.org> Signed-off-by:
-
Chris Zankel authored
Remove io_remap_page_range() from all of Linux 2.6.x (as requested and suggested by Randy Dunlap) and minor clean-ups. Signed-off-by:
Chris Zankel <chris@zankel.net> Signed-off-by:
-
Paolo 'Blaisorblade' Giarrusso authored
User get *a lot* confused when consoles don't work but we don't report anything. And, as reported in the comment, using printk to report "your console doesn't work" isn't likely to go that far. Fix the problem on the base of this: stack consumption by host printf(). Use kernel sprintf() and os_write_file, using a wild guess that one page will be enough for the message, to preallocate the buffer with kmalloc(). Signed-off-by:
-
Paolo 'Blaisorblade' Giarrusso authored
setup_initial_poll is only called with sigio_lock() held, so use appropriate allocation. Also, parse_chan() can also be called when holding a spinlock (see line_open() -> parse_chan_pair()). I have sporadic problems (spinlock taken twice, with spinlock debugging on UP) which could be caused by a sequence like "take spinlock, alloc and go to sleep, take again the spinlock in the other thread". Signed-off-by:
-
Paolo 'Blaisorblade' Giarrusso authored
GFP_ATOMIC | GFP_KERNEL is meaningless and won't work. Actually it never worked, even in 2.4. Signed-off-by:
-
Paolo 'Blaisorblade' Giarrusso authored
Following i386, we should maybe refuse trying to fault in pages when we're doing atomic operations, because to handle the fault we could need to take already taken spinlocks. Also, if we're doing an atomic operation (in the sense of in_atomic()) we're surely in kernel mode and we're surely going to handle adequately the failed fault, so it's safe to behave this way. Currently, on UML SMP is rarely used, and we don't support PREEMPT, so this is unlikely to create problems right now, but it might in the future. Signed-off-by:
-
