Further PDE->data assignments cleanups
Subsystems (10)
cover Further PDE->data assignments cleanups
This set of patches replaces proc_create/etc and further PDE->data assignment mostly with proc_data_create.
PATCH 1/12 sunrpc: assign PDE->data before gluing PDE into /proc tree
Simply replace proc_create and further data assigned with proc_create_data.
--- net/sunrpc/cache.c | 15 ++++++--------- net/sunrpc/stats.c | 8 +------- 2 files changed, 7 insertions(+), 16 deletions(-)
net/sunrpc/cache.c +6 -9
@@ -316,31 +316,28 @@ static int create_cache_proc_entries(struct cache_detail *cd) cd->proc_ent->owner = cd->owner; cd->channel_ent = cd->content_ent =NULL ; - p =proc_create ("flush" ,S_IFREG |S_IRUSR |S_IWUSR , - cd->proc_ent, &cache_flush_operations); + p =proc_create_data ("flush" ,S_IFREG |S_IRUSR |S_IWUSR , + cd->proc_ent, &cache_flush_operations, cd); cd->flush_ent = p;if (p ==NULL ) goto out_nomem; p->owner = cd->owner; - p->data = cd;if (cd->cache_request || cd->cache_parse) { - p =proc_create ("channel" ,S_IFREG |S_IRUSR |S_IWUSR , - cd->proc_ent, &cache_file_operations); + p =proc_create_data ("channel" ,S_IFREG |S_IRUSR |S_IWUSR , + cd->proc_ent, &cache_file_operations, cd); cd->channel_ent = p;if (p ==NULL ) goto out_nomem; p->owner = cd->owner; - p->data = cd; }if (cd->cache_show) { - p =proc_create ("content" ,S_IFREG |S_IRUSR |S_IWUSR , - cd->proc_ent, &content_file_operations); + p =proc_create_data ("content" ,S_IFREG |S_IRUSR |S_IWUSR , + cd->proc_ent, &content_file_operations, cd); cd->content_ent = p;if (p ==NULL ) goto out_nomem; p->owner = cd->owner; - p->data = cd; }return 0 ; out_nomem:
net/sunrpc/stats.c +1 -7
@@ -224,16 +224,10 @@ EXPORT_SYMBOL_GPL(rpc_print_iostats);static inline struct proc_dir_entry *do_register (const char *name,void *data,const struct file_operations *fops) { -struct proc_dir_entry *ent; -rpc_proc_init ();dprintk ("RPC: registering /proc/net/rpc/%s\n" , name); - ent =proc_create (name,0 , proc_net_rpc, fops); -if (ent) { - ent->data = data; - } -return ent; +return proc_create_data (name,0 , proc_net_rpc, fops, data); } struct proc_dir_entry *
Signature · 1 line
1.5.3.rc5
Re: [PATCH 1/12] sunrpc: assign PDE->data before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:11 +0400
Quoted 3 lines
Simply replace
proc_createand further data assigned withproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>
Applied.
PATCH 2/12 netfilter: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevSimply replace proc_create and further data assigned with proc_create_data.
--- net/ipv4/netfilter/ipt_CLUSTERIP.c | 6 +++--- net/netfilter/xt_hashlimit.c | 12 ++++++------ 2 files changed, 9 insertions(+), 9 deletions(-)
net/ipv4/netfilter/ipt_CLUSTERIP.c +3 -3
@@ -169,14 +169,14 @@ clusterip_config_init(const struct ipt_clusterip_tgt_info *i, __be32 ip,/* create proc dir entry */ sprintf (buffer,"%u.%u.%u.%u" ,NIPQUAD (ip)); - c->pde =proc_create (buffer,S_IWUSR |S_IRUSR , - clusterip_procdir, &clusterip_proc_fops); + c->pde =proc_create_data (buffer,S_IWUSR |S_IRUSR , + clusterip_procdir, + &clusterip_proc_fops, c);if (!c->pde) {kfree (c);return NULL ; } } - c->pde->data = c;#endif write_lock_bh (&clusterip_lock);
net/netfilter/xt_hashlimit.c +6 -6
@@ -237,15 +237,15 @@ static int htable_create_v0(struct xt_hashlimit_info *minfo, int family) hinfo->family = family; hinfo->rnd_initialized =0 ;spin_lock_init (&hinfo->lock); - hinfo->pde =proc_create (minfo->name,0 , + hinfo->pde = +proc_create_data (minfo->name,0 , family ==AF_INET ? hashlimit_procdir4 : hashlimit_procdir6, - &dl_file_ops); + &dl_file_ops, hinfo);if (!hinfo->pde) {vfree (hinfo);return -1 ; } - hinfo->pde->data = hinfo;setup_timer (&hinfo->timer, htable_gc, (unsigned long )hinfo); hinfo->timer.expires = jiffies +msecs_to_jiffies (hinfo->cfg.gc_interval);
@@ -301,15 +301,15 @@ static int htable_create(struct xt_hashlimit_mtinfo1 *minfo, hinfo->rnd_initialized =0 ;spin_lock_init (&hinfo->lock); - hinfo->pde =proc_create (minfo->name,0 , + hinfo->pde = +proc_create_data (minfo->name,0 , family ==AF_INET ? hashlimit_procdir4 : hashlimit_procdir6, - &dl_file_ops); + &dl_file_ops, hinfo);if (hinfo->pde ==NULL ) {vfree (hinfo);return -1 ; } - hinfo->pde->data = hinfo;setup_timer (&hinfo->timer, htable_gc, (unsigned long )hinfo); hinfo->timer.expires = jiffies +msecs_to_jiffies (hinfo->cfg.gc_interval);
Signature · 1 line
1.5.3.rc5
Re: [PATCH 2/12] netfilter: assign PDE->data before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:12 +0400
Quoted 6 lines
Simply replace
proc_createand further data assigned withproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc Patrick McHardy <[email protected]>
Applied, with Patrick's ACK.
PATCH 3/12 net: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevSimply replace proc_create and further data assigned with proc_create_data. Additionally, there is no need to assign NULL to PDE->data after creation, /proc generic has already done this for us.
--- net/core/neighbour.c | 5 ++--- net/core/pktgen.c | 14 +++++--------- 2 files changed, 7 insertions(+), 12 deletions(-)
net/core/neighbour.c +2 -3
@@ -1430,11 +1430,10 @@ void neigh_table_init_no_netlink(struct neigh_table *tbl)panic ("cannot create neighbour cache statistics" );#ifdef CONFIG_PROC_FS - tbl->pde =proc_create (tbl->id,0 , init_net.proc_net_stat, - &neigh_stat_seq_fops); + tbl->pde =proc_create_data (tbl->id,0 , init_net.proc_net_stat, + &neigh_stat_seq_fops, tbl);if (!tbl->pde)panic ("cannot create neighbour proc dir entry" ); - tbl->pde->data = tbl;#endif tbl->hash_mask =1 ;
net/core/pktgen.c +4 -8
@@ -3570,15 +3570,14 @@ static int pktgen_add_device(struct pktgen_thread *t, const char *ifname)if (err) goto out1; - pkt_dev->entry =proc_create (ifname,0600 , - pg_proc_dir, &pktgen_if_fops); + pkt_dev->entry =proc_create_data (ifname,0600 , pg_proc_dir, + &pktgen_if_fops, pkt_dev);if (!pkt_dev->entry) {printk (KERN_ERR "pktgen: cannot create %s/%s procfs entry.\n" ,PG_PROC_DIR , ifname); err = -EINVAL ; goto out2; } - pkt_dev->entry->data = pkt_dev;#ifdef CONFIG_XFRM pkt_dev->ipsmode =XFRM_MODE_TRANSPORT ; pkt_dev->ipsproto =IPPROTO_ESP ;
@@ -3628,7 +3627,8 @@ static int __init pktgen_create_thread(int cpu)kthread_bind (p, cpu); t->tsk = p; - pe =proc_create (t->tsk->comm,0600 , pg_proc_dir, &pktgen_thread_fops); + pe =proc_create_data (t->tsk->comm,0600 , pg_proc_dir, + &pktgen_thread_fops, t);if (!pe) {printk (KERN_ERR "pktgen: cannot create %s/%s procfs entry.\n" ,PG_PROC_DIR , t->tsk->comm);
@@ -3638,8 +3638,6 @@ static int __init pktgen_create_thread(int cpu)return -EINVAL ; } - pe->data = t; -wake_up_process (p);return 0 ;
@@ -3716,8 +3714,6 @@ static int __init pg_init(void)return -EINVAL ; } - pe->data =NULL ; -/* Register us to receive netdevice events */ register_netdevice_notifier (&pktgen_notifier_block);
Signature · 1 line
1.5.3.rc5
Re: [PATCH 3/12] net: assign PDE->data before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:13 +0400
Quoted 5 lines
Simply replace
proc_createand further data assigned withproc_create_data. Additionally, there is no need to assign NULL toPDE->dataafter creation, /proc generic has already done this for us.Signed-off-by Denis V. Lunev <[email protected]>
Applied, thanks.
PATCH 4/12 ipv6: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevSimply replace proc_create and further data assigned with proc_create_data.
--- net/ipv6/proc.c | 6 ++---- 1 files changed, 2 insertions(+), 4 deletions(-)
net/ipv6/proc.c +2 -4
@@ -247,13 +247,11 @@ int snmp6_register_dev(struct inet6_dev *idev)if (!proc_net_devsnmp6)return -ENOENT ; - p =proc_create (idev->dev->name,S_IRUGO , - proc_net_devsnmp6, &snmp6_seq_fops); + p =proc_create_data (idev->dev->name,S_IRUGO , + proc_net_devsnmp6, &snmp6_seq_fops, idev);if (!p)return -ENOMEM ; - p->data = idev; - idev->stats.proc_dir_entry = p;return 0 ; }
Signature · 1 line
1.5.3.rc5
Re: [PATCH 4/12] ipv6: assign PDE->data before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:14 +0400
Quoted 3 lines
Simply replace
proc_createand further data assigned withproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>
Applied, thanks.
PATCH 5/12 atm: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevSimply replace proc_create and further data assigned with proc_create_data. proc_atm_dev_ops holds proper referrence.
--- net/atm/proc.c | 6 ++---- 1 files changed, 2 insertions(+), 4 deletions(-)
net/atm/proc.c +2 -4
@@ -417,12 +417,10 @@ int atm_proc_dev_register(struct atm_dev *dev) goto err_out;sprintf (dev->proc_name,"%s:%d" ,dev->type, dev->number); - dev->proc_entry =proc_create (dev->proc_name,0 , atm_proc_root, - &proc_atm_dev_ops); + dev->proc_entry =proc_create_data (dev->proc_name,0 , atm_proc_root, + &proc_atm_dev_ops, dev);if (!dev->proc_entry) goto err_free_name; - dev->proc_entry->data = dev; - dev->proc_entry->owner =THIS_MODULE ;return 0 ; err_free_name:kfree (dev->proc_name);
Signature · 1 line
1.5.3.rc5
Re: [PATCH 5/12] atm: assign PDE->data before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:15 +0400
Quoted 4 lines
Simply replace
proc_createand further data assigned withproc_create_data.proc_atm_dev_opsholds proper referrence.Signed-off-by Denis V. Lunev <[email protected]>
Applied.
PATCH 6/12 vlan: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevSimply replace proc_create and further data assigned with proc_create_data.
--- net/8021q/vlanproc.c | 7 +++---- 1 files changed, 3 insertions(+), 4 deletions(-)
net/8021q/vlanproc.c +3 -4
@@ -176,12 +176,11 @@ int vlan_proc_add_dev(struct net_device *vlandev)struct vlan_dev_info *dev_info =vlan_dev_info (vlandev);struct vlan_net *vn =net_generic (dev_net (vlandev), vlan_net_id); - dev_info->dent =proc_create (vlandev->name,S_IFREG |S_IRUSR |S_IWUSR , - vn->proc_vlan_dir, &vlandev_fops); + dev_info->dent = +proc_create_data (vlandev->name,S_IFREG |S_IRUSR |S_IWUSR , + vn->proc_vlan_dir, &vlandev_fops, vlandev);if (!dev_info->dent)return -ENOBUFS ; - - dev_info->dent->data = vlandev;return 0 ; }
Signature · 1 line
1.5.3.rc5
Re: [PATCH 6/12] vlan: assign PDE->data before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:16 +0400
Quoted 6 lines
Simply replace
proc_createand further data assigned withproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc Patrick McHardy <[email protected]>
Applied, with Patrick's ACK.
Thanks.
PATCH 7/12 cciss: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevSimply replace proc_create and further data assigned with proc_create_data.
--- drivers/block/cciss.c | 8 ++------ 1 files changed, 2 insertions(+), 6 deletions(-)
drivers/block/cciss.c +2 -6
@@ -428,13 +428,9 @@ static void __devinit cciss_procinit(int i) proc_cciss =proc_mkdir ("driver/cciss" ,NULL );if (!proc_cciss)return ; - pde =proc_create (hba[i]->devname,S_IWUSR |S_IRUSR |S_IRGRP | + pde =proc_create_data (hba[i]->devname,S_IWUSR |S_IRUSR |S_IRGRP |S_IROTH , proc_cciss, - &cciss_proc_fops); -if (!pde) -return ; - - pde->data = hba[i]; + &cciss_proc_fops, hba[i]); }#endif /* CONFIG_PROC_FS */
Signature · 1 line
1.5.3.rc5
RE: [PATCH 7/12] cciss: assign PDE->data before gluing PDE into /proc tree
Quoted 11 lines
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Denis V. Lunev
Sent: Tuesday, April 29, 2008 6:13 AM
To: [email protected]
Cc: [email protected]; Denis V. Lunev; Alexey Dobriyan; Eric W. Biederman; ISS StorageDev; Greg Kroah-Hartman
Subject: [PATCH 7/12] cciss: assignPDE->databefore gluing PDE into /proc tree
Quoted 8 lines
Simply replace
proc_createand further data assigned withproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc <[email protected]>Cc Greg Kroah-Hartman <[email protected]>
Quoted 30 lines
--- drivers/block/cciss.c | 8 ++------ 1 files changed, 2 insertions(+), 6 deletions(-)drivers/block/cciss.c +0
-1@@ -428,13 +428,9 @@ static void __devinit cciss_procinit(int i) proc_cciss =proc_mkdir ("driver/cciss" ,NULL );if (!proc_cciss)return ; - pde = proc_create(hba[i]->devname, S_IWUSR | S_IRUSR|
S_IRGRP|drivers/block/cciss.c +3
-6+ pde =proc_create_data (hba[i]->devname,S_IWUSR |S_IRUSR | +S_IRGRP |S_IROTH , proc_cciss, - &cciss_proc_fops); -if (!pde) -return ; - - pde->data = hba[i]; + &cciss_proc_fops, hba[i]); }#endif /* CONFIG_PROC_FS */ --1.5.3.rc5
PATCH 8/12 powerpc: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevSimply replace proc_create and further data assigned with proc_create_data. No need to check for data!=NULL after that.
--- arch/powerpc/platforms/pseries/scanlog.c | 19 ++----------------- 1 files changed, 2 insertions(+), 17 deletions(-)
arch/powerpc/platforms/pseries/scanlog.c +2 -17
@@ -57,11 +57,6 @@ static ssize_t scanlog_read(struct file *file, char __user *buf, dp =PDE (inode); data = (unsigned int *)dp->data; -if (!data) { -printk (KERN_ERR "scanlog: read failed no data\n" ); -return -EIO ; - } -if (count >RTAS_DATA_BUF_SIZE ) count =RTAS_DATA_BUF_SIZE ;
@@ -153,11 +148,6 @@ static int scanlog_open(struct inode * inode, struct file * file)struct proc_dir_entry *dp =PDE (inode);unsigned int *data = (unsigned int *)dp->data; -if (!data) { -printk (KERN_ERR "scanlog: open failed no data\n" ); -return -EIO ; - } -if (data[0 ] !=0 ) {/* This imperfect test stops a second copy of the * data (or a reset while data is being copied)
@@ -175,10 +165,6 @@ static int scanlog_release(struct inode * inode, struct file * file)struct proc_dir_entry *dp =PDE (inode);unsigned int *data = (unsigned int *)dp->data; -if (!data) { -printk (KERN_ERR "scanlog: release failed no data\n" ); -return -EIO ; - } data[0 ] =0 ;return 0 ;
@@ -207,12 +193,11 @@ static int __init scanlog_init(void)if (!data) goto err; - ent =proc_create ("ppc64/rtas/scan-log-dump" ,S_IRUSR ,NULL , - &scanlog_fops); + ent =proc_create_data ("ppc64/rtas/scan-log-dump" ,S_IRUSR ,NULL , + &scanlog_fops, data);if (!ent) goto err; - ent->data = data; proc_ppc64_scan_log_dump = ent;return 0 ;
Signature · 1 line
1.5.3.rc5
PATCH 9/12 ipv4: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevThe check for PDE->data != NULL becomes useless after the replacement of proc_net_fops_create with proc_create_data.
--- net/ipv4/tcp_ipv4.c | 10 +++------- net/ipv4/udp.c | 7 +++---- 2 files changed, 6 insertions(+), 11 deletions(-)
net/ipv4/tcp_ipv4.c +3 -7
@@ -2214,9 +2214,6 @@ static int tcp_seq_open(struct inode *inode, struct file *file)struct tcp_iter_state *s;int err; -if (unlikely (afinfo ==NULL )) -return -EINVAL ; - err =seq_open_net (inode, file, &afinfo->seq_ops,sizeof (struct tcp_iter_state ));if (err <0 )
@@ -2241,10 +2238,9 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) afinfo->seq_ops.next = tcp_seq_next; afinfo->seq_ops.stop = tcp_seq_stop; - p =proc_net_fops_create (net, afinfo->name,S_IRUGO , &afinfo->seq_fops); -if (p) - p->data = afinfo; -else + p =proc_create_data (afinfo->name,S_IRUGO , net->proc_net, + &afinfo->seq_fops, afinfo); +if (!p) rc = -ENOMEM ;return rc; }
net/ipv4/udp.c +3 -4
@@ -1605,10 +1605,9 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo) afinfo->seq_ops.next = udp_seq_next; afinfo->seq_ops.stop = udp_seq_stop; - p =proc_net_fops_create (net, afinfo->name,S_IRUGO , &afinfo->seq_fops); -if (p) - p->data = afinfo; -else + p =proc_create_data (afinfo->name,S_IRUGO , net->proc_net, + &afinfo->seq_fops, afinfo); +if (!p) rc = -ENOMEM ;return rc; }
Signature · 1 line
1.5.3.rc5
Re: [PATCH 9/12] ipv4: assign PDE->data before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:19 +0400
Quoted 4 lines
The check for
PDE->data!= NULL becomes useless after the replacement ofproc_net_fops_createwithproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>
Applied.
Re: [PATCH 9/12] ipv4: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevFirst off, sorry to bring such an old email back but I can seem to get a bad feeling when looking back over it.
Quoted 17 lines · Denis V. Lunev
The check for
PDE->data!= NULL becomes useless after the replacement ofproc_net_fops_createwithproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc David S. Miller <[email protected]>--- net/ipv4/tcp_ipv4.c | 10 +++------- net/ipv4/udp.c | 7 +++---- 2 files changed, 6 insertions(+), 11 deletions(-)net/ipv4/tcp_ipv4.c +0
-0
net/ipv4/tcp_ipv4.c static int tcp_seq_open(struct inode *inode, struct file *file) · line 2218
Quoted 5 lines
net/ipv4/tcp_ipv4.c +0
-2@@ -2214,9 +2214,6 @@ static int tcp_seq_open(struct inode *inode, struct file *file)struct tcp_iter_state *s;int err; -if (unlikely (afinfo ==NULL )) -return -EINVAL ;
I think that this check needs to stay in some form, reason below.
net/ipv4/tcp_ipv4.c int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) · line 2241
Quoted 13 lines
net/ipv4/tcp_ipv4.c +1
-5- err =seq_open_net (inode, file, &afinfo->seq_ops,sizeof (struct tcp_iter_state ));if (err <0 )@@ -2241,10 +2238,9 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) afinfo->seq_ops.next = tcp_seq_next; afinfo->seq_ops.stop = tcp_seq_stop; - p =proc_net_fops_create (net, afinfo->name,S_IRUGO , &afinfo->seq_fops); -if (p) - p->data = afinfo; -else + p = proc_create_data(afinfo->name, S_IRUGO, net->proc_net,
When you try to pass in afinfo->name (and also the seq_fops) you are assuming that afinfo is not null meaning in the unlikely(as shown above) even that it is null you get a very bad null pointer problem. If I am just way off do let me know because this just seams to me like a bad idea. This is also still present in 2.6.26-rc9.
-Stoyan G
Quoted 32 lines
net/ipv4/tcp_ipv4.c +2
-0+ &afinfo->seq_fops, afinfo); +if (!p) rc = -ENOMEM ;return rc; }net/ipv4/udp.c +3
-5@@ -1605,10 +1605,9 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo) afinfo->seq_ops.next = udp_seq_next; afinfo->seq_ops.stop = udp_seq_stop; - p =proc_net_fops_create (net, afinfo->name,S_IRUGO , &afinfo->seq_fops); -if (p) - p->data = afinfo; -else + p =proc_create_data (afinfo->name,S_IRUGO , net->proc_net, + &afinfo->seq_fops, afinfo); +if (!p) rc = -ENOMEM ;return rc; } --1.5.3.rc5
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 9/12] ipv4: assign PDE->data before gluing PDE into /proc tree
Quoted 43 lines · Stoyan Gaydarov
First off, sorry to bring such an old email back but I can seem to get a bad feeling when looking back over it.
net/ipv4/tcp_ipv4.c static int tcp_seq_open(struct inode *inode, struct file
Quoted 17 lines · Denis V. Lunev
The check for
PDE->data!= NULL becomes useless after the replacement ofproc_net_fops_createwithproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc David S. Miller <[email protected]>--- net/ipv4/tcp_ipv4.c | 10 +++------- net/ipv4/udp.c | 7 +++---- 2 files changed, 6 insertions(+), 11 deletions(-)net/ipv4/tcp_ipv4.c +0
-0@@ -2214,9 +2214,6 @@ static int tcp_seq_open(struct inode *inode, struct file*file)
net/ipv4/tcp_ipv4.c static int tcp_seq_open(struct inode *inode, struct file *file) · line 2218
Quoted 5 lines
net/ipv4/tcp_ipv4.c +0
-2struct tcp_iter_state *s;int err; -if (unlikely (afinfo ==NULL )) -return -EINVAL ;I think that this check needs to stay in some form, reason below.
net/ipv4/tcp_ipv4.c int tcp_proc_register(struct net *net, struct
Quoted 5 lines
net/ipv4/tcp_ipv4.c +0
-1- err =seq_open_net (inode, file, &afinfo->seq_ops,sizeof (struct tcp_iter_state ));if (err <0 )@@ -2241,10 +2238,9 @@ int tcp_proc_register(struct net *net, struct
tcp_seq_afinfo*afinfo)net/ipv4/tcp_ipv4.c int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) · line 2241
Quoted 8 lines
net/ipv4/tcp_ipv4.c +1
-4afinfo->seq_ops.next = tcp_seq_next; afinfo->seq_ops.stop = tcp_seq_stop; - p =proc_net_fops_create (net, afinfo->name,S_IRUGO , &afinfo->seq_fops); -if (p) - p->data = afinfo; -else + p = proc_create_data(afinfo->name, S_IRUGO, net->proc_net,
Quoted 5 lines
When you try to pass in
afinfo->name(and also theseq_fops) you are assuming that afinfo is not null meaning in the unlikely(as shown above) even that it is null you get a very bad null pointer problem. If I am just way off do let me know because this just seams to me like a bad idea. This is also still present in 2.6.26-rc9.
It appears you are getting things confused. The original window is that tcp_seq_open (which is what get called when you open the proc file) had a small race that p->data could be read before it was set.
With proc_create_data that race was closed.
You are saying that it is a problem for tcp_seq_open to be passed a NULL afinfo. It is. That has nothing to do with the original race (as that is a very different part of the code). Feel free to audit all of the callers if you like. That problem however is not subtle or racy.
So I see nothing wrong with this patch unless you can find a problem with proc_create_data.
Eric
Re: [PATCH 9/12] ipv4: assign PDE->data before gluing PDE into /proc tree
In reply to Stoyan GaydarovQuoted 35 lines · Stoyan Gaydarov
First off, sorry to bring such an old email back but I can seem to get a bad feeling when looking back over it.
Quoted 17 lines · Denis V. Lunev
The check for
PDE->data!= NULL becomes useless after the replacement ofproc_net_fops_createwithproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc David S. Miller <[email protected]>--- net/ipv4/tcp_ipv4.c | 10 +++------- net/ipv4/udp.c | 7 +++---- 2 files changed, 6 insertions(+), 11 deletions(-)net/ipv4/tcp_ipv4.c +0
-0net/ipv4/tcp_ipv4.c static int tcp_seq_open(struct inode *inode, struct file *file) · line 2218
Quoted 5 lines
net/ipv4/tcp_ipv4.c +0
-2@@ -2214,9 +2214,6 @@ static int tcp_seq_open(struct inode *inode, struct file *file)struct tcp_iter_state *s;int err; -if (unlikely (afinfo ==NULL )) -return -EINVAL ;I think that this check needs to stay in some form, reason below.
Quoted 8 lines
net/ipv4/tcp_ipv4.c +0
-1- err =seq_open_net (inode, file, &afinfo->seq_ops,sizeof (struct tcp_iter_state ));if (err <0 )@@ -2241,10 +2238,9 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) afinfo->seq_ops.next = tcp_seq_next; afinfo->seq_ops.stop = tcp_seq_stop;
Quoted 11 lines
net/ipv4/tcp_ipv4.c int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) · line 2241
Quoted 5 lines
net/ipv4/tcp_ipv4.c +1
-4- p =proc_net_fops_create (net, afinfo->name,S_IRUGO , &afinfo->seq_fops); -if (p) - p->data = afinfo; -else + p = proc_create_data(afinfo->name, S_IRUGO, net->proc_net,When you try to pass in
afinfo->name(and also theseq_fops) you are assuming that afinfo is not null meaning in the unlikely(as shown above) even that it is null you get a very bad null pointer problem. If I am just way off do let me know because this just seams to me like a bad idea. This is also still present in 2.6.26-rc9.
The reason to remove the check is simple - afinfo comes in the form of the static pointer during init time. It is impossible to face NULL pointer the problem in the reality. It can come as the programmer mistake, but the OOPS will be immediate and straightforward. Namely, the kernel will not boot/work at all.
Regards,
Den
PATCH 10/12 netfilter: assign PDE->fops before gluing PDE into /proc tree
In reply to Denis V. LunevReplace create_proc_entry with specially created for this purpose proc_create.
--- net/netfilter/nf_conntrack_standalone.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-)
net/netfilter/nf_conntrack_standalone.c +3 -3
@@ -296,11 +296,11 @@ static int nf_conntrack_standalone_init_proc(void) pde =proc_net_fops_create (&init_net,"nf_conntrack" ,0440 , &ct_file_ops);if (!pde) goto out_nf_conntrack; - pde =create_proc_entry ("nf_conntrack" ,S_IRUGO , init_net.proc_net_stat); + + pde =proc_create ("nf_conntrack" ,S_IRUGO , init_net.proc_net_stat, + &ct_cpu_seq_fops);if (!pde) goto out_stat_nf_conntrack; - pde->proc_fops = &ct_cpu_seq_fops; - pde->owner =THIS_MODULE ;return 0 ; out_stat_nf_conntrack:
Signature · 1 line
1.5.3.rc5
Re: [PATCH 10/12] netfilter: assign PDE->fops before gluing PDE into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:20 +0400
Quoted 6 lines
Replace
create_proc_entrywith specially created for this purposeproc_create.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc Patrick McHardy <[email protected]>
Applied, with Patrick's ACK.
PATCH 11/12 netfilter: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevReplace proc_net_fops_create with proc_create_data.
--- net/netfilter/x_tables.c | 13 ++++++------- 1 files changed, 6 insertions(+), 7 deletions(-)
net/netfilter/x_tables.c +6 -7
@@ -936,25 +936,24 @@ int xt_proto_init(struct net *net, int af)#ifdef CONFIG_PROC_FS strlcpy (buf, xt_prefix[af],sizeof (buf));strlcat (buf,FORMAT_TABLES ,sizeof (buf)); - proc =proc_net_fops_create (net, buf,0440 , &xt_table_ops); + proc =proc_create_data (buf,0440 , net->proc_net, &xt_table_ops, + (void *)(unsigned long )af);if (!proc) goto out; - proc->data = (void *)(unsigned long )af; -strlcpy (buf, xt_prefix[af],sizeof (buf));strlcat (buf,FORMAT_MATCHES ,sizeof (buf)); - proc =proc_net_fops_create (net, buf,0440 , &xt_match_ops); + proc =proc_create_data (buf,0440 , net->proc_net, &xt_match_ops, + (void *)(unsigned long )af);if (!proc) goto out_remove_tables; - proc->data = (void *)(unsigned long )af;strlcpy (buf, xt_prefix[af],sizeof (buf));strlcat (buf,FORMAT_TARGETS ,sizeof (buf)); - proc =proc_net_fops_create (net, buf,0440 , &xt_target_ops); + proc =proc_create_data (buf,0440 , net->proc_net, &xt_target_ops, + (void *)(unsigned long )af);if (!proc) goto out_remove_matches; - proc->data = (void *)(unsigned long )af;#endif return 0 ;
Signature · 1 line
1.5.3.rc5
Re: [PATCH 11/12] netfilter: assign PDE->data before gluing PDE into /proc tree
Quoted 6 lines · Denis V. Lunev
Replace
proc_net_fops_createwithproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc Patrick McHardy <[email protected]>
For the VLAN and netfilter patches:
Re: [PATCH 11/12] netfilter: assign PDE->data before gluing PDE into /proc tree
In reply to Denis V. LunevFrom: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:21 +0400
Quoted 6 lines
Replace
proc_net_fops_createwithproc_create_data.Signed-off-by Denis V. Lunev <[email protected]>Cc Alexey Dobriyan <[email protected]>Cc Eric W. Biederman <[email protected]>Cc Patrick McHardy <[email protected]>
Applied, with Patrick's ACK.
PATCH 12/12 netns: assign PDE->data before gluing entry into /proc tree
In reply to Denis V. LunevIn this unfortunate case, proc_mkdir_mode wrapper can't be used anymore and this is no way to reuse proc_create_data due to nlinks assignment. So, copy the code from proc_mkdir and assign PDE->data at the appropriate moment.
--- fs/proc/generic.c | 17 +++++++++++++++++ fs/proc/proc_net.c | 11 ----------- 2 files changed, 17 insertions(+), 11 deletions(-)
fs/proc/generic.c +17 -0
@@ -641,6 +641,23 @@ struct proc_dir_entry *proc_mkdir_mode(const char *name, mode_t mode,return ent; } +struct proc_dir_entry *proc_net_mkdir (struct net *net,const char *name, +struct proc_dir_entry *parent) +{ +struct proc_dir_entry *ent; + + ent =__proc_create (&parent, name,S_IFDIR |S_IRUGO |S_IXUGO ,2 ); +if (ent) { + ent->data = net; +if (proc_register (parent, ent) <0 ) { +kfree (ent); + ent =NULL ; + } + } +return ent; +} +EXPORT_SYMBOL_GPL (proc_net_mkdir); +struct proc_dir_entry *proc_mkdir (const char *name,struct proc_dir_entry *parent) {
fs/proc/proc_net.c +0 -11
@@ -159,17 +159,6 @@ struct net *get_proc_net(const struct inode *inode) }EXPORT_SYMBOL_GPL (get_proc_net); -struct proc_dir_entry *proc_net_mkdir (struct net *net,const char *name, -struct proc_dir_entry *parent) -{ -struct proc_dir_entry *pde; - pde =proc_mkdir_mode (name,S_IRUGO |S_IXUGO , parent); -if (pde !=NULL ) - pde->data = net; -return pde; -} -EXPORT_SYMBOL_GPL (proc_net_mkdir); -static __net_init int proc_net_ns_init (struct net *net) {struct proc_dir_entry *netd, *net_statd;
Signature · 1 line
1.5.3.rc5
Re: [PATCH 12/12] netns: assign PDE->data before gluing entry into /proc tree
From: "Denis V. Lunev" <[email protected]>
Date: Tue, 29 Apr 2008 15:13:22 +0400
Quoted 6 lines
In this unfortunate case,
proc_mkdir_modewrapper can't be used anymore and this is no way to reuseproc_create_datadue to nlinks assignment. So, copy the code fromproc_mkdirand assignPDE->dataat the appropriate moment.Signed-off-by Denis V. Lunev <[email protected]>
Oh well, the cost of correctness sometimes :-)
Applied, thanks.