xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[PATCH] NFS: Introduce the use of inode->i_lock to protect fields in nfsi

Browse Source 
Down the road we want to eliminate the use of the global kernel lock entirely
from the NFS client.  To do this, we need to protect the fields in the
nfs_inode structure adequately.  Start by serializing updates to the
"cache_validity" field.

Note this change addresses an SMP hang found by njw@osdl.org, where processes
deadlock because nfs_end_data_update and nfs_revalidate_mapping update the
"cache_validity" field without proper serialization.

Test plan:
 Millions of fsx ops on SMP clients.  Run Nick Wilson's breaknfs program on
 large SMP clients.

Signed-off-by: Chuck Lever <cel@netapp.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This commit is contained in:
Chuck Lever
2005-08-18 11:24:12 -07:00
committed by Linus Torvalds
parent 412d582ec1
commit dc59250c6e
5 changed files with 48 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fs/nfs/read.c
View File

@@ -140,7 +140,9 @@ static int nfs_readpage_sync(struct nfs_open_context *ctx, struct inode *inode,
if (rdata->res.eof != 0 || result == 0)
break;
} while (count);
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME;
spin_unlock(&inode->i_lock);
if (count)
memclear_highpage_flush(page, rdata->args.pgbase, count);
@@ -473,7 +475,9 @@ void nfs_readpage_result(struct rpc_task *task)
}
task->tk_status = -EIO;
}
spin_lock(&data->inode->i_lock);
NFS_I(data->inode)->cache_validity |= NFS_INO_INVALID_ATIME;
spin_unlock(&data->inode->i_lock);
data->complete(data, status);
}