- 27 Apr, 2012 39 commits
-
-
Trond Myklebust authored
If the allocation of nfs_write_header fails, the list of nfs_pages that needs to be cleaned up is still on desc->pg_list... Reported-by:
Bryan Schumaker <bjschuma@netapp.com> Signed-off-by:
Trond Myklebust <Trond.Myklebust@netapp.com> Acked-by:
Fred Isaman <iisaman@netapp.com>
-
Bryan Schumaker authored
Now that I'm doing secinfo automatically in the v4 code this extra argument isn't needed. Signed-off-by:
-
Bryan Schumaker authored
This simplifies the code for v2 and v3 and gives v4 a chance to decide on referrals without needing to modify the generic client. Signed-off-by:
-
Bryan Schumaker authored
And also remove the unneeded rpc_op. Signed-off-by:
-
Fred Isaman authored
This is a bugfix that applies on top of the previous directio patches, that fixes a bug introduced in "NFS: create struct nfs_commit_info". Signed-off-by:
-
Fred Isaman authored
This also has the advantage that it allows directio to use pnfs. Signed-off-by:
-
Fred Isaman authored
Signed-off-by:
-
Fred Isaman authored
Need this to pass into nfs_commitdata_init, in order to keep data->dreq accurate. Signed-off-by:
-
Fred Isaman authored
Factors out the code that needs to change when directio starts using these code paths. Signed-off-by:
-
Fred Isaman authored
It is COMMIT that is handled the most differently between the paged and direct paths. Create a structure that encapsulates everything either path needs to know about the commit state. We could use void to hide some of the layout driver stuff, but Trond suggests pulling it out to ensure type checking, given the huge changes being made, and the fact that it doesn't interfere with other drivers. Signed-off-by:
-
Fred Isaman authored
Simple refactoring. Signed-off-by:
-
Fred Isaman authored
This also has the advantage that it allows directio to use pnfs. Signed-off-by:
-
Fred Isaman authored
The coalesce code made assumptions that will no longer be true once non-page aligned io occurs. This introduces no change in current behavior, but allows for more general situations to come. Signed-off-by:
-
Fred Isaman authored
Signed-off-by:
-
Fred Isaman authored
Factors out the code that will need to change when directio starts using these code paths. This will allow directio to use the generic pagein and flush routines Signed-off-by:
-
Fred Isaman authored
Decouple nfs_pgio_header and nfs_write_data, and have (possibly multiple) nfs_write_datas each take a refcount on nfs_pgio_header. For the moment keeps nfs_write_header as a way to preallocate a single nfs_write_data with the nfs_pgio_header. The code doesn't need this, and would be prettier without, but given the amount of churn I am already introducing I didn't want to play with tuning new mempools. This also fixes bug in pnfs_ld_handle_write_error. In the case of desc->pg_bsize < PAGE_CACHE_SIZE, the pages list was empty, causing replay attempt to do nothing. Signed-off-by:
-
Fred Isaman authored
Decouple nfs_pgio_header and nfs_read_data, and have (possibly multiple) nfs_read_datas each take a refcount on nfs_pgio_header. For the moment keeps nfs_read_header as a way to preallocate a single nfs_read_data with the nfs_pgio_header. The code doesn't need this, and would be prettier without, but given the amount of churn I am already introducing I didn't want to play with tuning new mempools. This also fixes bug in pnfs_ld_handle_read_error. In the case of desc->pg_bsize < PAGE_CACHE_SIZE, the pages list was empty, causing replay attempt to do nothing. Signed-off-by:
-
Fred Isaman authored
Both nfs_read_data and nfs_write_data devote several fields which can be combined into a single shared struct. Signed-off-by:
-
Fred Isaman authored
In order to avoid duplicating all the data in nfs_read_data whenever we split it up into multiple RPC calls (either due to a short read result or due to rsize < PAGE_SIZE), we split out the bits that are the same per RPC call into a separate "header" structure. The goal this patch moves towards is to have a single header refcounted by several rpc_data structures. Thus, want to always refer from rpc_data to the header, and not the other way. This patch comes close to that ideal, but the directio code currently needs some special casing, isolated in the nfs_direct_[read_write]hdr_release() functions. This will be dealt with in a future patch. Signed-off-by:
-
Fred Isaman authored
Signed-off-by:
-
Fred Isaman authored
Signed-off-by:
-
Fred Isaman authored
Make it consistent with nfs_initiate_commit. Signed-off-by:
-
Fred Isaman authored
Signed-off-by:
-
Fred Isaman authored
Commits don't need the vectors of pages, etc. that writes do. Split out a separate structure for the commit operation. Signed-off-by:
-
Fred Isaman authored
Also create a commit_info structure to hold the bucket array and push it up from the lseg to the layout where it really belongs. While we are at it, fix a refcounting bug due to an (incorrect) implicit assumption that filelayout_scan_ds_commit_list always completely emptied the src list. This clarifies refcounting, removes the ugly find_only_write_lseg functions, and pushes the file layout commit code along on the path to supporting multiple lsegs. Signed-off-by:
-
Fred Isaman authored
The two functions had diverged quite a bit, with the write function being a bit more robust than the read. However, these still break badly in the desc->pg_bsize < PAGE_CACHE_SIZE case, as then there is nothing hanging on the data->pages list, and the resend ends up doing nothing. This will be fixed in a patch later in the series. Signed-off-by:
-
Fred Isaman authored
Signed-off-by:
-
Bryan Schumaker authored
This fixes a compiler warning. Signed-off-by:
-
Bryan Schumaker authored
The authflavor is set in an nfs_clone_mount structure and passed to the xdev_mount() functions where it was promptly ignored. Instead, use it to initialize an rpc_clnt for the cloned server. Signed-off-by:
-
Bryan Schumaker authored
I create a new proc_lookup_mountpoint() to use when submounting an NFS v4 share. This function returns an rpc_clnt to use for performing an fs_locations() call on a referral's mountpoint. Signed-off-by:
-
Bryan Schumaker authored
Whenever lookup sees wrongsec do a secinfo and retry the lookup to find attributes of the file or directory, such as "is this a referral mountpoint?". This also allows me to remove handling -NFS4ERR_WRONSEC as part of getattr xdr decoding. Signed-off-by:
-
Bryan Schumaker authored
We don't want to return -NFS4ERR_WRONGSEC to the VFS because it could cause the kernel to oops. Signed-off-by:
-
Bryan Schumaker authored
I was using the same decoder function for SECINFO and SECINFO_NO_NAME, so it was returning an error when it tried to decode an OP_SECINFO_NO_NAME header as OP_SECINFO. Signed-off-by:
-
Stanislav Kinsbursky authored
v2: recursion was replaced by loop If client is a clone, then it's parent can not be in the list. But parent's Pipefs dentries have to be created and destroyed. Note: event skip helper for clients introduced Signed-off-by:
Stanislav Kinsbursky <skinsbursky@parallels.com> Signed-off-by:
-
Stanislav Kinsbursky authored
There can be a case, when on MOUNT event RPC client (after it's dentries were created) is not longer hold by anyone except notification callback. I.e. on release this client will be destoroyed. And it's dentries have to be destroyed as well. Which in turn requires per-net PipeFS superblock to be set. Signed-off-by:
-
Stanislav Kinsbursky authored
1) This is sane. 2) Otherwise there will be soft lockup: do { rpc_get_client_for_event (clnt->cl_dentry == NULL ==> choose) __rpc_pipefs_event (clnt->cl_program->pipe_dir_name == NULL ==> return) } while (1) Signed-off-by: -
Stanislav Kinsbursky authored
These clients can't be safely dereferenced if their counter in 0. Signed-off-by:
-
Sachin Prabhu authored
When attempting to cache ACLs returned from the server, if the bitmap size + the ACL size is greater than a PAGE_SIZE but the ACL size itself is smaller than a PAGE_SIZE, we can read past the buffer page boundary. Signed-off-by:
Sachin Prabhu <sprabhu@redhat.com> Reported-by:
Jian Li <jiali@redhat.com> Signed-off-by:
-
Sachin Prabhu authored
Bug noticed in commit bf118a34 When calling GETACL, if the size of the bitmap array, the length attribute and the acl returned by the server is greater than the allocated buffer(args.acl_len), we can Oops with a General Protection fault at _copy_from_pages() when we attempt to read past the pages allocated. This patch allocates an extra PAGE for the bitmap and checks to see that the bitmap + attribute_length + ACLs don't exceed the buffer space allocated to it. Signed-off-by:
-
- 26 Apr, 2012 1 commit
-
-
Jim Rees authored
Signed-off-by:
Jim Rees <rees@umich.edu> Suggested-by:
Andy Adamson <andros@netapp.com> Suggested-by:
-
