[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <tip-55e858c8483af427144f33b42b818b30612b82b0@git.kernel.org>
Date: Wed, 14 Oct 2009 15:51:46 GMT
From: tip-bot for John Kacur <jkacur@...hat.com>
To: linux-tip-commits@...r.kernel.org
Cc: linux-kernel@...r.kernel.org, hpa@...or.com, mingo@...hat.com,
jkacur@...hat.com, airlied@...ux.ie, tglx@...utronix.de
Subject: [tip:bkl/drivers] agp: Remove the BKL from agp_open
Commit-ID: 55e858c8483af427144f33b42b818b30612b82b0
Gitweb: http://git.kernel.org/tip/55e858c8483af427144f33b42b818b30612b82b0
Author: John Kacur <jkacur@...hat.com>
AuthorDate: Sun, 11 Oct 2009 22:24:25 +0200
Committer: Thomas Gleixner <tglx@...utronix.de>
CommitDate: Wed, 14 Oct 2009 17:36:54 +0200
agp: Remove the BKL from agp_open
- Remove the BKL from agp_open
- Perform a few clean-ups.
Analysis:
---------
int minor is local to the function.
The following are protected by agp_fe.agp_mutex
struct agp_file_private *priv;
struct agp_client *client;
Call-outs:
kzalloc should be safe to call under the mutex_lock
agp_find_client_by_pid:
- agp_mmap calls that under agp_fe.agp_mutex which we hold in agp_open
- agpioc_reserve_wrap calls it without any locking what-so-ever.
- Is that an error? Or is that okay because it has pid that is
a unique handle?
agp_insert_file_private:
- This function only manipulates struct agp_file_private, once again
while agp_fe.agp_mutex is held
Signed-off-by: John Kacur <jkacur@...hat.com>
Acked-by: David Airlie <airlied@...ux.ie>
LKML-Reference: <alpine.LFD.2.00.0910112216060.12574@...alhost.localdomain>
Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
---
drivers/char/agp/frontend.c | 28 +++++++++++-----------------
1 files changed, 11 insertions(+), 17 deletions(-)
diff --git a/drivers/char/agp/frontend.c b/drivers/char/agp/frontend.c
index a96f319..43412c0 100644
--- a/drivers/char/agp/frontend.c
+++ b/drivers/char/agp/frontend.c
@@ -676,25 +676,25 @@ static int agp_open(struct inode *inode, struct file *file)
int minor = iminor(inode);
struct agp_file_private *priv;
struct agp_client *client;
- int rc = -ENXIO;
-
- lock_kernel();
- mutex_lock(&(agp_fe.agp_mutex));
if (minor != AGPGART_MINOR)
- goto err_out;
+ return -ENXIO;
+
+ mutex_lock(&(agp_fe.agp_mutex));
priv = kzalloc(sizeof(struct agp_file_private), GFP_KERNEL);
- if (priv == NULL)
- goto err_out_nomem;
+ if (priv == NULL) {
+ mutex_unlock(&(agp_fe.agp_mutex));
+ return -ENOMEM;
+ }
set_bit(AGP_FF_ALLOW_CLIENT, &priv->access_flags);
priv->my_pid = current->pid;
- if (capable(CAP_SYS_RAWIO)) {
+ if (capable(CAP_SYS_RAWIO))
/* Root priv, can be controller */
set_bit(AGP_FF_ALLOW_CONTROLLER, &priv->access_flags);
- }
+
client = agp_find_client_by_pid(current->pid);
if (client != NULL) {
@@ -704,16 +704,10 @@ static int agp_open(struct inode *inode, struct file *file)
file->private_data = (void *) priv;
agp_insert_file_private(priv);
DBG("private=%p, client=%p", priv, client);
- mutex_unlock(&(agp_fe.agp_mutex));
- unlock_kernel();
- return 0;
-err_out_nomem:
- rc = -ENOMEM;
-err_out:
mutex_unlock(&(agp_fe.agp_mutex));
- unlock_kernel();
- return rc;
+
+ return 0;
}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists