The following patch seems to cure the problem of sigmasks not being
properly set in NFS write requests in 2.1.117-ac1 that I reported
earlier today. It protects the synchronous call that is made when the
server is congested. Is this fix correct Alan?
Cheers,
Trond
--- linux-2.1.117-ac1/fs/nfs/write.c-2.1.117-ac1 Mon Aug 24 17:54:21 1998
+++ linux-2.1.117-ac1/fs/nfs/write.c Mon Aug 24 17:58:22 1998
@@ -442,11 +442,15 @@
sync = 1;
if (sync) {
+ sigset_t oldmask;
+ struct rpc_clnt *clnt = NFS_CLIENT(inode);
dprintk("NFS: %4d schedule_write_request (sync)\n",
task->tk_pid);
/* Page is already locked */
req->wb_flags |= NFS_WRITE_LOCKED;
+ rpc_clnt_sigmask(clnt, &oldmask);
rpc_execute(task);
+ rpc_clnt_sigunmask(clnt, &oldmask);
} else {
dprintk("NFS: %4d schedule_write_request (async)\n",
task->tk_pid);
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html