* [PATCH] 9p: don't use __getname/__putname for uname/aname
@ 2012-09-07 11:55 Jeff Layton
0 siblings, 0 replies; only message in thread
From: Jeff Layton @ 2012-09-07 11:55 UTC (permalink / raw)
To: v9fs-developer; +Cc: linux-fsdevel, linux-kernel
These are generally very small strings. We don't need an entire 4k
allocation for each. Instead, just free and reallocate them on an
as-needed basis.
Note: This patch is untested since I don't have a 9p server available at
the moment. It's mainly something I noticed while doing some
getname/putname cleanup work.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
---
fs/9p/v9fs.c | 30 +++++++++++++++++++-----------
1 file changed, 19 insertions(+), 11 deletions(-)
diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c
index b85efa7..b20af74 100644
--- a/fs/9p/v9fs.c
+++ b/fs/9p/v9fs.c
@@ -184,10 +184,20 @@ static int v9fs_parse_options(struct v9fs_session_info *v9ses, char *opts)
v9ses->afid = option;
break;
case Opt_uname:
- match_strlcpy(v9ses->uname, &args[0], PATH_MAX);
+ kfree(v9ses->uname);
+ v9ses->uname = match_strdup(&args[0]);
+ if (!v9ses->uname) {
+ ret = -ENOMEM;
+ goto free_and_return;
+ }
break;
case Opt_remotename:
- match_strlcpy(v9ses->aname, &args[0], PATH_MAX);
+ kfree(v9ses->aname);
+ v9ses->aname = match_strdup(&args[0]);
+ if (!v9ses->aname) {
+ ret = -ENOMEM;
+ goto free_and_return;
+ }
break;
case Opt_nodevmap:
v9ses->nodev = 1;
@@ -287,21 +297,21 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
struct p9_fid *fid;
int rc;
- v9ses->uname = __getname();
+ v9ses->uname = kstrdup(V9FS_DEFUSER, GFP_KERNEL);
if (!v9ses->uname)
return ERR_PTR(-ENOMEM);
- v9ses->aname = __getname();
+ v9ses->aname = kstrdup(V9FS_DEFANAME, GFP_KERNEL);
if (!v9ses->aname) {
- __putname(v9ses->uname);
+ kfree(v9ses->uname);
return ERR_PTR(-ENOMEM);
}
init_rwsem(&v9ses->rename_sem);
rc = bdi_setup_and_register(&v9ses->bdi, "9p", BDI_CAP_MAP_COPY);
if (rc) {
- __putname(v9ses->aname);
- __putname(v9ses->uname);
+ kfree(v9ses->aname);
+ kfree(v9ses->uname);
return ERR_PTR(rc);
}
@@ -309,8 +319,6 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
list_add(&v9ses->slist, &v9fs_sessionlist);
spin_unlock(&v9fs_sessionlist_lock);
- strcpy(v9ses->uname, V9FS_DEFUSER);
- strcpy(v9ses->aname, V9FS_DEFANAME);
v9ses->uid = ~0;
v9ses->dfltuid = V9FS_DEFUID;
v9ses->dfltgid = V9FS_DEFGID;
@@ -412,8 +420,8 @@ void v9fs_session_close(struct v9fs_session_info *v9ses)
kfree(v9ses->cachetag);
}
#endif
- __putname(v9ses->uname);
- __putname(v9ses->aname);
+ kfree(v9ses->uname);
+ kfree(v9ses->aname);
bdi_destroy(&v9ses->bdi);
--
1.7.11.4
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2012-09-07 11:55 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-09-07 11:55 [PATCH] 9p: don't use __getname/__putname for uname/aname Jeff Layton
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).