-stable review
Subsystems (15)
cover -stable review
This is the start of the stable review cycle for the 2.6.15.4 release. There are 23 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let us know. If anyone is a maintainer of the proper subsystem, and wants to add a Signed-off-by: line to the patch, please respond with it.
These patches are sent out with a number of different people on the
Cc: line. If you wish to be a reviewer, please email [email protected] to add your name to the list. If you want to be off the reviewer list, also email us.
Responses should be made by Fri Feb 10 06:30 UTC. Anything received after that time, might be too late.
thanks,
the -stable release team
PATCH 1/23 SCSI: turn off ordered flush barriers
-stable review patch. If anyone has any objections, please let us know.
------------------
Turn off ordered flush barriers for SCSI driver, since the SCSI barrier code has a command leak.
---
drivers/scsi/scsi_lib.c | 5 ----- 1 files changed, 5 deletions(-)
drivers/scsi/scsi_lib.c +0 -5
@@ -1534,11 +1534,6 @@ struct request_queue *scsi_alloc_queue(s*/ if (shost->ordered_tag)blk_queue_ordered (q,QUEUE_ORDERED_TAG ); -else if (shost->ordered_flush) { -blk_queue_ordered (q,QUEUE_ORDERED_FLUSH ); - q->prepare_flush_fn = scsi_prepare_flush_fn; - q->end_flush_fn = scsi_end_flush_fn; - }if (!shost->use_clustering)clear_bit (QUEUE_FLAG_CLUSTER , &q->queue_flags);
--
PATCH 2/23 dm-crypt: zero key before freeing it
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Zap the memory before freeing it so we don't leave crypto information around in memory.
---
drivers/md/dm-crypt.c | 5 +++++ 1 files changed, 5 insertions(+)
drivers/md/dm-crypt.c +5 -0
@@ -690,6 +690,8 @@ bad3: bad2:crypto_free_tfm (tfm); bad1: +/* Must zero key material before freeing */ +memset (cc,0 ,sizeof (*cc) + cc->key_size *sizeof (u8));kfree (cc);return -EINVAL ; }
@@ -706,6 +708,9 @@ static void crypt_dtr(struct dm_target * cc->iv_gen_ops->dtr (cc);crypto_free_tfm (cc->tfm);dm_put_device (ti, cc->dev); + +/* Must zero key material before freeing */ +memset (cc,0 ,sizeof (*cc) + cc->key_size *sizeof (u8));kfree (cc); }
--
PATCH 3/23 d_instantiate_unique / NFS inode leakage
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
If we have found aliased dentry that we return, inode reference is not dropped and inode is not attached anywhere, so it seems the reference to inode is leaked in that case.
---
fs/dcache.c | 7 ++++++- 1 files changed, 6 insertions(+), 1 deletion(-)
fs/dcache.c +6 -1
@@ -808,10 +808,14 @@ void d_instantiate(struct dentry *entry,* * Fill in inode information in the entry. On success, it returns NULL. * If an unhashed alias of "entry" already exists, then we return the -* aliased dentry instead. +* aliased dentry instead and drop one reference to inode. * * Note that in order to avoid conflicts with rename() etc, the caller * had better be holding the parent directory semaphore. +* +* This also assumes that the inode count has been incremented +* (or otherwise set) by the caller to indicate that it is now +* in use by the dcache. */ struct dentry *d_instantiate_unique (struct dentry *entry,struct inode *inode) {
@@ -838,6 +842,7 @@ struct dentry *d_instantiate_unique(strudget_locked (alias);spin_unlock (&dcache_lock);BUG_ON (!d_unhashed (alias)); +iput (inode);return alias; }list_add (&entry->d_alias, &inode->i_dentry);
--
PATCH 4/23 seclvl settime fix
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Don't try to "validate" a non-existing timeval.
settime() with a NULL timeval is silly but legal.
Noticed by Dave Jones <[email protected]>
---
security/seclvl.c | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-)
security/seclvl.c +1 -1
@@ -369,7 +369,7 @@ static int seclvl_capable(struct task_ststatic int seclvl_settime (struct timespec *tv,struct timezone *tz) {struct timespec now; -if (seclvl >1 ) { +if (tv && seclvl >1 ) { now =current_kernel_time ();if (tv->tv_sec < now.tv_sec || (tv->tv_sec == now.tv_sec && tv->tv_nsec < now.tv_nsec)) {
--
PATCH 5/23 fix regression in xfs_buf_rele
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Fix regression in xfs_buf_rele dealing with non-hashed buffers, as occur during log replay. Novell bug 145204, Fedora bug 177848.
---
fs/xfs/linux-2.6/xfs_buf.c | 7 +++++++ 1 files changed, 7 insertions(+)
fs/xfs/linux-2.6/xfs_buf.c +7 -0
@@ -830,6 +830,13 @@ pagebuf_rele(PB_TRACE (pb,"rele" , pb->pb_relse); +if (unlikely (!hash)) { +ASSERT (!pb->pb_relse); +if (atomic_dec_and_test (&pb->pb_hold)) +xfs_buf_free (pb); +return ; + } +if (atomic_dec_and_lock (&pb->pb_hold, &hash->bh_lock)) {if (pb->pb_relse) {atomic_inc (&pb->pb_hold);
--
PATCH 6/23 Input: mousedev - fix memory leak
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Apparently, "while true; do cat </dev/null >/dev/input/mice; done" causes an OOM in a short amount of time. Funny that nobody noticed, it actually is very easy to trigger just by switching between VT1 and VT7...
---
drivers/input/mousedev.c | 9 +++------ 1 files changed, 3 insertions(+), 6 deletions(-)
drivers/input/mousedev.c +3 -6
@@ -356,7 +356,7 @@ static void mousedev_free(struct mousedekfree (mousedev); } -static int mixdev_release (void ) +static void mixdev_release (void ) {struct input_handle *handle;
@@ -370,8 +370,6 @@ static int mixdev_release(void)mousedev_free (mousedev); } } - -return 0 ; }static int mousedev_release (struct inode * inode,struct file * file)
@@ -384,9 +382,8 @@ static int mousedev_release(struct inodeif (!--list->mousedev->open) {if (list->mousedev->minor ==MOUSEDEV_MIX ) -return mixdev_release (); - -if (!mousedev_mix.open) { +mixdev_release (); +else if (!mousedev_mix.open) {if (list->mousedev->exist)input_close_device (&list->mousedev->handle);else
--
PATCH 7/23 Input: grip - fix crash when accessing device
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
grip - fix crash when accessing device
---
drivers/input/joystick/grip.c | 3 +++ 1 files changed, 3 insertions(+)
drivers/input/joystick/grip.c +3 -0
@@ -192,6 +192,9 @@ static void grip_poll(struct gameport *gfor (i =0 ; i <2 ; i++) { dev = grip->dev[i]; +if (!dev) +continue ; + grip->reads++;switch (grip->mode[i]) {
--
PATCH 8/23 Input: db9 - fix possible crash with Saturn gamepads
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
db9 - fix possible crash with Saturn gamepads
---
drivers/input/joystick/db9.c | 70 ++++++++++++++++++++++--------------------- 1 files changed, 36 insertions(+), 34 deletions(-)
drivers/input/joystick/db9.c +36 -34
@@ -275,68 +275,70 @@ static unsigned char db9_saturn_read_pac/* * db9_saturn_report() analyzes packet and reports. */ -static int db9_saturn_report (unsigned char id,unsigned char data[60 ],struct input_dev *dev,int n,int max_pads) +static int db9_saturn_report (unsigned char id,unsigned char data[60 ],struct input_dev *devs[],int n,int max_pads) { +struct input_dev *dev;int tmp, i, j; tmp = (id ==0x41 ) ?60 :10 ; -for (j =0 ; (j < tmp) && (n < max_pads); j +=10 , n++) { +for (j =0 ; j < tmp && n < max_pads; j +=10 , n++) { + dev = devs[n];switch (data[j]) {case 0x16 :/* multi controller (analog 4 axis) */ -input_report_abs (dev + n, db9_abs[5 ], data[j +6 ]); +input_report_abs (dev, db9_abs[5 ], data[j +6 ]);case 0x15 :/* mission stick (analog 3 axis) */ -input_report_abs (dev + n, db9_abs[3 ], data[j +4 ]); -input_report_abs (dev + n, db9_abs[4 ], data[j +5 ]); +input_report_abs (dev, db9_abs[3 ], data[j +4 ]); +input_report_abs (dev, db9_abs[4 ], data[j +5 ]);case 0x13 :/* racing controller (analog 1 axis) */ -input_report_abs (dev + n, db9_abs[2 ], data[j +3 ]); +input_report_abs (dev, db9_abs[2 ], data[j +3 ]);case 0x34 :/* saturn keyboard (udlr ZXC ASD QE Esc) */ case 0x02 :/* digital pad (digital 2 axis + buttons) */ -input_report_abs (dev + n, db9_abs[0 ], !(data[j +1 ] &128 ) - !(data[j +1 ] &64 )); -input_report_abs (dev + n, db9_abs[1 ], !(data[j +1 ] &32 ) - !(data[j +1 ] &16 )); +input_report_abs (dev, db9_abs[0 ], !(data[j +1 ] &128 ) - !(data[j +1 ] &64 )); +input_report_abs (dev, db9_abs[1 ], !(data[j +1 ] &32 ) - !(data[j +1 ] &16 ));for (i =0 ; i <9 ; i++) -input_report_key (dev + n, db9_cd32_btn[i], ~data[j + db9_saturn_byte[i]] & db9_saturn_mask[i]); +input_report_key (dev, db9_cd32_btn[i], ~data[j + db9_saturn_byte[i]] & db9_saturn_mask[i]);break ;case 0x19 :/* mission stick x2 (analog 6 axis + buttons) */ -input_report_abs (dev + n, db9_abs[0 ], !(data[j +1 ] &128 ) - !(data[j +1 ] &64 )); -input_report_abs (dev + n, db9_abs[1 ], !(data[j +1 ] &32 ) - !(data[j +1 ] &16 )); +input_report_abs (dev, db9_abs[0 ], !(data[j +1 ] &128 ) - !(data[j +1 ] &64 )); +input_report_abs (dev, db9_abs[1 ], !(data[j +1 ] &32 ) - !(data[j +1 ] &16 ));for (i =0 ; i <9 ; i++) -input_report_key (dev + n, db9_cd32_btn[i], ~data[j + db9_saturn_byte[i]] & db9_saturn_mask[i]); -input_report_abs (dev + n, db9_abs[2 ], data[j +3 ]); -input_report_abs (dev + n, db9_abs[3 ], data[j +4 ]); -input_report_abs (dev + n, db9_abs[4 ], data[j +5 ]); +input_report_key (dev, db9_cd32_btn[i], ~data[j + db9_saturn_byte[i]] & db9_saturn_mask[i]); +input_report_abs (dev, db9_abs[2 ], data[j +3 ]); +input_report_abs (dev, db9_abs[3 ], data[j +4 ]); +input_report_abs (dev, db9_abs[4 ], data[j +5 ]);/* -input_report_abs(dev + n, db9_abs[8], (data[j + 6] & 128 ? 0 : 1) - (data[j + 6] & 64 ? 0 : 1)); -input_report_abs(dev + n, db9_abs[9], (data[j + 6] & 32 ? 0 : 1) - (data[j + 6] & 16 ? 0 : 1)); +input_report_abs(dev, db9_abs[8], (data[j + 6] & 128 ? 0 : 1) - (data[j + 6] & 64 ? 0 : 1)); +input_report_abs(dev, db9_abs[9], (data[j + 6] & 32 ? 0 : 1) - (data[j + 6] & 16 ? 0 : 1)); */ -input_report_abs (dev + n, db9_abs[6 ], data[j +7 ]); -input_report_abs (dev + n, db9_abs[7 ], data[j +8 ]); -input_report_abs (dev + n, db9_abs[5 ], data[j +9 ]); +input_report_abs (dev, db9_abs[6 ], data[j +7 ]); +input_report_abs (dev, db9_abs[7 ], data[j +8 ]); +input_report_abs (dev, db9_abs[5 ], data[j +9 ]);break ;case 0xd3 :/* sankyo ff (analog 1 axis + stop btn) */ -input_report_key (dev + n,BTN_A , data[j +3 ] &0x80 ); -input_report_abs (dev + n, db9_abs[2 ], data[j +3 ] &0x7f ); +input_report_key (dev,BTN_A , data[j +3 ] &0x80 ); +input_report_abs (dev, db9_abs[2 ], data[j +3 ] &0x7f );break ;case 0xe3 :/* shuttle mouse (analog 2 axis + buttons. signed value) */ -input_report_key (dev + n,BTN_START , data[j +1 ] &0x08 ); -input_report_key (dev + n,BTN_A , data[j +1 ] &0x04 ); -input_report_key (dev + n,BTN_C , data[j +1 ] &0x02 ); -input_report_key (dev + n,BTN_B , data[j +1 ] &0x01 ); -input_report_abs (dev + n, db9_abs[2 ], data[j +2 ] ^0x80 ); -input_report_abs (dev + n, db9_abs[3 ], (0xff -(data[j +3 ] ^0x80 ))+1 );/* */ +input_report_key (dev,BTN_START , data[j +1 ] &0x08 ); +input_report_key (dev,BTN_A , data[j +1 ] &0x04 ); +input_report_key (dev,BTN_C , data[j +1 ] &0x02 ); +input_report_key (dev,BTN_B , data[j +1 ] &0x01 ); +input_report_abs (dev, db9_abs[2 ], data[j +2 ] ^0x80 ); +input_report_abs (dev, db9_abs[3 ], (0xff -(data[j +3 ] ^0x80 ))+1 );/* */ break ;case 0xff :default :/* no pad */ -input_report_abs (dev + n, db9_abs[0 ],0 ); -input_report_abs (dev + n, db9_abs[1 ],0 ); +input_report_abs (dev, db9_abs[0 ],0 ); +input_report_abs (dev, db9_abs[1 ],0 );for (i =0 ; i <9 ; i++) -input_report_key (dev + n, db9_cd32_btn[i],0 ); +input_report_key (dev, db9_cd32_btn[i],0 );break ; } }return n; } -static int db9_saturn (int mode,struct parport *port,struct input_dev *dev) +static int db9_saturn (int mode,struct parport *port,struct input_dev *devs[]) {unsigned char id, data[60 ];int type, n, max_pads;
@@ -361,7 +363,7 @@ static int db9_saturn(int mode, struct p max_pads =min (db9_modes[mode].n_pads,DB9_MAX_DEVICES );for (tmp =0 , i =0 ; i < n; i++) { id =db9_saturn_read_packet (port, data, type + i,1 ); - tmp =db9_saturn_report (id, data, dev, tmp, max_pads); + tmp =db9_saturn_report (id, data, devs, tmp, max_pads); }return 0 ; }
@@ -489,7 +491,7 @@ static void db9_timer(unsigned long privcase DB9_SATURN_DPP :case DB9_SATURN_DPP_2 : -db9_saturn (db9->mode, port, dev); +db9_saturn (db9->mode, port, db9->dev);break ;case DB9_CD32_PAD :
--
PATCH 9/23 Input: sidewinder - fix an oops
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Dynalloc conversion strikes again...
---
drivers/input/joystick/sidewinder.c | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-)
drivers/input/joystick/sidewinder.c +1 -1
@@ -736,7 +736,7 @@ static int sw_connect(struct gameport *gsprintf (sw->name,"Microsoft SideWinder %s" , sw_name[sw->type]);sprintf (sw->phys[i],"%s/input%d" , gameport->phys, i); - input_dev =input_allocate_device (); + sw->dev[i] = input_dev =input_allocate_device ();if (!input_dev) { err = -ENOMEM ; goto fail3;
--
PATCH 10/23 Input: iforce - do not return ENOMEM upon successful allocation
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
iforce - do not return ENOMEM upon successful allocation
---
drivers/input/joystick/iforce/iforce-main.c | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-)
drivers/input/joystick/iforce/iforce-main.c +1 -1
@@ -345,7 +345,7 @@ int iforce_init_device(struct iforce *ifint i; input_dev =input_allocate_device (); -if (input_dev) +if (!input_dev)return -ENOMEM ;init_waitqueue_head (&iforce->wait);
--
PATCH 11/23 Input: iforce - fix detection of USB devices
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Recent conversion to wait_event_interruptible_timeout() caused USB detection routine erroneously report timeouts for perfectly working devices.
---
drivers/input/joystick/iforce/iforce-packets.c | 4 ++-- drivers/input/joystick/iforce/iforce-usb.c | 1 - 2 files changed, 2 insertions(+), 3 deletions(-)
drivers/input/joystick/iforce/iforce-packets.c +2 -2
@@ -167,9 +167,9 @@ void iforce_process_packet(struct iforce iforce->expect_packet =0 ; iforce->ecmd = cmd; memcpy(iforce->edata, data, IFORCE_MAX_LENGTH); - wake_up(&iforce->wait); }#endif +wake_up (&iforce->wait);if (!iforce->type) { being_used--;
@@ -264,7 +264,7 @@ int iforce_get_id_packet(struct iforce *wait_event_interruptible_timeout (iforce->wait, iforce->ctrl->status != -EINPROGRESS ,HZ ); -if (iforce->ctrl->status != -EINPROGRESS ) { +if (iforce->ctrl->status) {usb_unlink_urb (iforce->ctrl);return -1 ; }
drivers/input/joystick/iforce/iforce-usb.c +0 -1
@@ -95,7 +95,6 @@ static void iforce_usb_irq(struct urb *u goto exit; } -wake_up (&iforce->wait);iforce_process_packet (iforce, (iforce->data[0 ] <<8 ) | (urb->actual_length -1 ), iforce->data +1 , regs);
--
PATCH 12/23 : Kill compat_sys_clock_settime sign extension stub.
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
It's wrong and totally unneeded.
---
arch/sparc64/kernel/sys32.S | 1 - arch/sparc64/kernel/systbls.S | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-)
arch/sparc64/kernel/sys32.S +0 -1
@@ -84,7 +84,6 @@ SIGN2(sys32_fadvise64_64, compat_sys_fad SIGN2(sys32_bdflush, sys_bdflush, %o0, %o1) SIGN1(sys32_mlockall, sys_mlockall, %o0) SIGN1(sys32_nfsservctl, compat_sys_nfsservctl, %o0) -SIGN1(sys32_clock_settime, compat_sys_clock_settime, %o1) SIGN1(sys32_clock_nanosleep, compat_sys_clock_nanosleep, %o1) SIGN1(sys32_timer_settime, compat_sys_timer_settime, %o1) SIGN1(sys32_io_submit, compat_sys_io_submit, %o1)
arch/sparc64/kernel/systbls.S +1 -1
@@ -71,7 +71,7 @@ sys_call_table32: /*240*/ .word sys_munlockall, sys32_sched_setparam, sys32_sched_getparam, sys32_sched_setscheduler, sys32_sched_getscheduler .word sys_sched_yield, sys32_sched_get_priority_max, sys32_sched_get_priority_min, sys32_sched_rr_get_interval, compat_sys_nanosleep /*250*/ .word sys32_mremap, sys32_sysctl, sys32_getsid, sys_fdatasync, sys32_nfsservctl - .word sys_ni_syscall, sys32_clock_settime, compat_sys_clock_gettime, compat_sys_clock_getres, sys32_clock_nanosleep + .word sys_ni_syscall, compat_sys_clock_settime, compat_sys_clock_gettime, compat_sys_clock_getres, sys32_clock_nanosleep /*260*/ .word compat_sys_sched_getaffinity, compat_sys_sched_setaffinity, sys32_timer_settime, compat_sys_timer_gettime, sys_timer_getoverrun .word sys_timer_delete, sys32_timer_create, sys_ni_syscall, compat_sys_io_setup, sys_io_destroy /*270*/ .word sys32_io_submit, sys_io_cancel, compat_sys_io_getevents, sys32_mq_open, sys_mq_unlink
--
PATCH 13/23 Fix keyctl usage of strnlen_user()
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
In the small window between strnlen_user() and copy_from_user() userspace could alter the terminating `\0' character.
---
security/keys/keyctl.c | 15 ++++++++++----- 1 files changed, 10 insertions(+), 5 deletions(-)
security/keys/keyctl.c +10 -5
@@ -66,9 +66,10 @@ asmlinkage long sys_add_key(const char _ description =kmalloc (dlen +1 ,GFP_KERNEL );if (!description) goto error; + description[dlen] ='\0' ; ret = -EFAULT ; -if (copy_from_user (description, _description, dlen +1 ) !=0 ) +if (copy_from_user (description, _description, dlen) !=0 ) goto error2;/* pull the payload in if one was supplied */
@@ -160,9 +161,10 @@ asmlinkage long sys_request_key(const ch description =kmalloc (dlen +1 ,GFP_KERNEL );if (!description) goto error; + description[dlen] ='\0' ; ret = -EFAULT ; -if (copy_from_user (description, _description, dlen +1 ) !=0 ) +if (copy_from_user (description, _description, dlen) !=0 ) goto error2;/* pull the callout info into kernel space */
@@ -181,9 +183,10 @@ asmlinkage long sys_request_key(const ch callout_info =kmalloc (dlen +1 ,GFP_KERNEL );if (!callout_info) goto error2; + callout_info[dlen] ='\0' ; ret = -EFAULT ; -if (copy_from_user (callout_info, _callout_info, dlen +1 ) !=0 ) +if (copy_from_user (callout_info, _callout_info, dlen) !=0 ) goto error3; }
@@ -278,9 +281,10 @@ long keyctl_join_session_keyring(const c name =kmalloc (nlen +1 ,GFP_KERNEL );if (!name) goto error; + name[nlen] ='\0' ; ret = -EFAULT ; -if (copy_from_user (name, _name, nlen +1 ) !=0 ) +if (copy_from_user (name, _name, nlen) !=0 ) goto error2; }
@@ -582,9 +586,10 @@ long keyctl_keyring_search(key_serial_t description =kmalloc (dlen +1 ,GFP_KERNEL );if (!description) goto error; + description[dlen] ='\0' ; ret = -EFAULT ; -if (copy_from_user (description, _description, dlen +1 ) !=0 ) +if (copy_from_user (description, _description, dlen) !=0 ) goto error2;/* get the keyring at which to begin the search */
--
David Howells
maintainerRe: [PATCH 13/23] Fix keyctl usage of strnlen_user()
Quoted 1 line · Chris Wright
-stable review patch. If anyone has any objections, please let us know.
It looks fine by me.
David
PATCH 14/23 PCMCIA=m, HOSTAP_CS=y is not a legal configuration
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
CONFIG_PCMCIA=m, CONFIG_HOSTAP_CS=y doesn't compile.
Reported by "Gabriel C." <[email protected]>.
This patch was already included in 2.6.16-rc2.
---
drivers/net/wireless/hostap/Kconfig | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/hostap/Kconfig +1 -1
@@ -61,7 +61,7 @@ config HOSTAP_PCI config HOSTAP_CS tristate "Host AP driver for Prism2/2.5/3 PC Cards" - depends on PCMCIA!=n && HOSTAP + depends on PCMCIA && HOSTAP ---help--- Host AP driver's version for Prism2/2.5/3 PC Cards.
--
PATCH 15/23 SELinux: fix size-128 slab leak
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Remove private inode tests from security_inode_alloc and security_inode_free, as we otherwise end up leaking inode security structures for private inodes.
--- include/linux/security.h | 4 ---- 1 files changed, 4 deletions(-)
include/linux/security.h +0 -4
@@ -1437,15 +1437,11 @@ static inline void security_sb_post_pivostatic inline int security_inode_alloc (struct inode *inode) { -if (unlikely (IS_PRIVATE (inode))) -return 0 ;return security_ops->inode_alloc_security (inode); }static inline void security_inode_free (struct inode *inode) { -if (unlikely (IS_PRIVATE (inode))) -return ; security_ops->inode_free_security (inode); }
--
PATCH 16/23 : Fixed hardware RX checksum handling
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
When we pull the PPP protocol off the skb, we forgot to update the hardware RX checksum. This may lead to messages such as
dsl0: hw csum failure.
Similarly, we need to clear the hardware checksum flag when we use the existing packet to store the decompressed result.
---
drivers/net/ppp_generic.c | 3 +++ 1 files changed, 3 insertions(+)
drivers/net/ppp_generic.c +3 -0
@@ -1610,6 +1610,8 @@ ppp_receive_nonmp_frame(struct ppp *ppp, }else if (!pskb_may_pull (skb, skb->len)) goto err; +else + skb->ip_summed =CHECKSUM_NONE ; len =slhc_uncompress (ppp->vj, skb->data +2 , skb->len -2 );if (len <=0 ) {
@@ -1690,6 +1692,7 @@ ppp_receive_nonmp_frame(struct ppp *ppp,kfree_skb (skb); }else {skb_pull (skb,2 );/* chop off protocol */ +skb_postpull_rcsum (skb, skb->data -2 ,2 ); skb->dev = ppp->dev; skb->protocol =htons (npindex_to_ethertype[npi]); skb->mac.raw = skb->data;
--
PATCH 17/23 x86_64: Let impossible CPUs point to reference per cpu data
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Don't mark the reference per cpu data init data (so it stays around after boot) and point all impossible CPUs to it. This way they reference some valid - although shared memory. Usually this is only initialization like INIT_LIST_HEADs and there won't be races because these CPUs never run. Still somewhat hackish.
---
arch/x86_64/kernel/vmlinux.lds.S | 10 ++++++---- 1 files changed, 6 insertions(+), 4 deletions(-)
arch/x86_64/kernel/vmlinux.lds.S +6 -4
@@ -170,13 +170,15 @@ SECTIONS . = ALIGN(4096); __initramfs_start = .; .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) } - __initramfs_end = .; - . = ALIGN(32); + __initramfs_end = .; + /* temporary here to work around NR_CPUS. If you see this comment in 2.6.17+ + complain */ + . = ALIGN(4096); + __init_end = .; + . = ALIGN(128); __per_cpu_start = .; .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) } __per_cpu_end = .; - . = ALIGN(4096); - __init_end = .; . = ALIGN(4096); __nosave_begin = .;
--
PATCH 18/23 x86_64: Clear more state when ignoring empty node in SRAT parsing
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Fix boot failures on systems with bad PXMs.
---
arch/x86_64/mm/srat.c | 28 +++++++++++++++++++++------- 1 files changed, 21 insertions(+), 7 deletions(-)
arch/x86_64/mm/srat.c +21 -7
@@ -25,6 +25,10 @@ static nodemask_t nodes_found __initdatastatic struct node nodes[MAX_NUMNODES ]__initdata ;static __u8 pxm2node[256 ] = { [0 ...255 ] =0xff }; +/* Too small nodes confuse the VM badly. Usually they result +from BIOS bugs. */ +#define NODE_MIN_SIZE (4*1024*1024) +static int node_to_pxm (int n);int pxm_to_node (int pxm)
@@ -168,22 +172,32 @@ acpi_numa_memory_affinity_init(struct ac nd->start, nd->end); } +static void unparse_node (int node) +{ +int i; +node_clear (node, nodes_parsed); +for (i =0 ; i <MAX_LOCAL_APIC ; i++) { +if (apicid_to_node[i] == node) + apicid_to_node[i] =NUMA_NO_NODE ; + } +} +void __init acpi_numa_arch_fixup (void ) {}/* Use the information discovered above to actually set up the nodes. */ int __init acpi_scan_nodes (unsigned long start,unsigned long end) {int i; + +for (i =0 ; i <MAX_NUMNODES ; i++) { +cutoff_node (i, start, end); +if ((nodes[i].end - nodes[i].start) <NODE_MIN_SIZE ) +unparse_node (i); + } +if (acpi_numa <=0 )return -1 ; -/* First clean up the node list */ -for_each_node_mask (i, nodes_parsed) { -cutoff_node (i, start, end); -if (nodes[i].start == nodes[i].end) -node_clear (i, nodes_parsed); - } - memnode_shift =compute_hash_shift (nodes,nodes_weight (nodes_parsed));if (memnode_shift <0 ) { printk(KERN_ERR
--
PATCH 19/23 bridge: netfilter races on device removal
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Fix bridge netfilter to handle case where interface is deleted from bridge while packet is being processed (on other CPU).
---
net/bridge/br_netfilter.c | 55 +++++++++++++++++++++++++++++++--------------- 1 files changed, 38 insertions(+), 17 deletions(-)
net/bridge/br_netfilter.c +38 -17
@@ -47,9 +47,6 @@#define store_orig_dstaddr (skb) (skb_origaddr(skb) = (skb)->nh.iph->daddr)#define dnat_took_place (skb) (skb_origaddr(skb) != (skb)->nh.iph->daddr) -#define has_bridge_parent (device) ((device)->br_port != NULL) -#define bridge_parent (device) ((device)->br_port->br->dev) -#ifdef CONFIG_SYSCTL static struct ctl_table_header *brnf_sysctl_header;static int brnf_call_iptables =1 ;
@@ -94,6 +91,12 @@ static struct rtable __fake_rtable = { .rt_flags = 0, }; +static inline struct net_device *bridge_parent (const struct net_device *dev) +{ +struct net_bridge_port *port =rcu_dereference (dev->br_port); + +return port ? port->br->dev :NULL ; +}/* PF_BRIDGE/PRE_ROUTING *********************************************/ /* Undo the changes made for ip6tables PREROUTING and continue the
@@ -185,11 +188,15 @@ static int br_nf_pre_routing_finish_brid skb->nf_bridge->mask ^=BRNF_NF_BRIDGE_PREROUTING ; skb->dev =bridge_parent (skb->dev); -if (skb->protocol ==__constant_htons (ETH_P_8021Q )) { -skb_pull (skb,VLAN_HLEN ); - skb->nh.raw +=VLAN_HLEN ; +if (!skb->dev) +kfree_skb (skb); +else { +if (skb->protocol ==__constant_htons (ETH_P_8021Q )) { +skb_pull (skb,VLAN_HLEN ); + skb->nh.raw +=VLAN_HLEN ; + } + skb->dst->output (skb); } - skb->dst->output (skb);return 0 ; }
@@ -266,7 +273,7 @@ bridged_dnat: }/* Some common code for IPv4/IPv6 */ -static void setup_pre_routing (struct sk_buff *skb) +static struct net_device *setup_pre_routing (struct sk_buff *skb) {struct nf_bridge_info *nf_bridge = skb->nf_bridge;
@@ -278,6 +285,8 @@ static void setup_pre_routing(struct sk_ nf_bridge->mask |=BRNF_NF_BRIDGE_PREROUTING ; nf_bridge->physindev = skb->dev; skb->dev =bridge_parent (skb->dev); + +return skb->dev; }/* We only check the length. A bridge shouldn't do any hop-by-hop stuff anyway */
@@ -372,7 +381,8 @@ static unsigned int br_nf_pre_routing_ipnf_bridge_put (skb->nf_bridge);if ((nf_bridge =nf_bridge_alloc (skb)) ==NULL )return NF_DROP ; -setup_pre_routing (skb); +if (!setup_pre_routing (skb)) +return NF_DROP ;NF_HOOK (PF_INET6 ,NF_IP6_PRE_ROUTING , skb, skb->dev,NULL , br_nf_pre_routing_finish_ipv6);
@@ -409,7 +419,6 @@ static unsigned int br_nf_pre_routing(unif (skb->protocol ==__constant_htons (ETH_P_8021Q )) {skb_pull (skb,VLAN_HLEN ); - (skb)->nh.raw +=VLAN_HLEN ; }return br_nf_pre_routing_ipv6 (hook, skb, in, out, okfn); }
@@ -426,7 +435,6 @@ static unsigned int br_nf_pre_routing(unif (skb->protocol ==__constant_htons (ETH_P_8021Q )) {skb_pull (skb,VLAN_HLEN ); - (skb)->nh.raw +=VLAN_HLEN ; }if (!pskb_may_pull (skb,sizeof (struct iphdr )))
@@ -456,7 +464,8 @@ static unsigned int br_nf_pre_routing(unnf_bridge_put (skb->nf_bridge);if ((nf_bridge =nf_bridge_alloc (skb)) ==NULL )return NF_DROP ; -setup_pre_routing (skb); +if (!setup_pre_routing (skb)) +return NF_DROP ;store_orig_dstaddr (skb); NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, skb->dev, NULL,
@@ -530,11 +539,16 @@ static unsigned int br_nf_forward_ip(unsstruct sk_buff *skb = *pskb;struct nf_bridge_info *nf_bridge;struct vlan_ethhdr *hdr =vlan_eth_hdr (skb); +struct net_device *parent;int pf;if (!skb->nf_bridge)return NF_ACCEPT ; + parent =bridge_parent (out); +if (!parent) +return NF_DROP ; +if (skb->protocol ==__constant_htons (ETH_P_IP ) ||IS_VLAN_IP ) pf =PF_INET ;else
@@ -555,8 +569,8 @@ static unsigned int br_nf_forward_ip(uns nf_bridge->mask |=BRNF_BRIDGED ; nf_bridge->physoutdev = skb->dev; -NF_HOOK (pf,NF_IP_FORWARD , skb,bridge_parent (in), -bridge_parent (out), br_nf_forward_finish); +NF_HOOK (pf,NF_IP_FORWARD , skb,bridge_parent (in), parent, + br_nf_forward_finish);return NF_STOLEN ; }
@@ -679,6 +693,8 @@ static unsigned int br_nf_local_out(unsi goto out; } realoutdev =bridge_parent (skb->dev); +if (!realoutdev) +return NF_DROP ;#if defined(CONFIG_VLAN_8021Q ) || defined(CONFIG_VLAN_8021Q_MODULE )/* iptables should match -o br0.x */
@@ -692,9 +708,11 @@ static unsigned int br_nf_local_out(unsi/* IP forwarded traffic has a physindev, locally * generated traffic hasn't. */ if (realindev !=NULL ) { -if (!(nf_bridge->mask &BRNF_DONT_TAKE_PARENT ) && -has_bridge_parent (realindev)) - realindev =bridge_parent (realindev); +if (!(nf_bridge->mask &BRNF_DONT_TAKE_PARENT ) ) { +struct net_device *parent =bridge_parent (realindev); +if (parent) + realindev = parent; + } NF_HOOK_THRESH(pf, NF_IP_FORWARD, skb, realindev, realoutdev, br_nf_local_out_finish,
@@ -734,6 +752,9 @@ static unsigned int br_nf_post_routing(uif (!nf_bridge)return NF_ACCEPT ; +if (!realoutdev) +return NF_DROP ; +if (skb->protocol ==__constant_htons (ETH_P_IP ) ||IS_VLAN_IP ) pf =PF_INET ;else
--
PATCH 20/23 bridge: fix RCU race on device removal
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Patch to 2.6.15 stable kernel to fix race conditions on device removal. These are reproducible by doing delif while packets are in flight.
---
net/bridge/br_if.c | 7 ++----- net/bridge/br_input.c | 10 ++++++++-- net/bridge/br_stp_bpdu.c | 8 ++++++-- 3 files changed, 16 insertions(+), 9 deletions(-)
net/bridge/br_if.c +2 -5
@@ -99,7 +99,6 @@ static void del_nbp(struct net_bridge_postruct net_bridge *br = p->br;struct net_device *dev = p->dev; - dev->br_port =NULL ;dev_set_promiscuity (dev,-1 );spin_lock_bh (&br->lock);
@@ -110,9 +109,7 @@ static void del_nbp(struct net_bridge_polist_del_rcu (&p->list); -del_timer_sync (&p->message_age_timer); -del_timer_sync (&p->forward_delay_timer); -del_timer_sync (&p->hold_timer); +rcu_assign_pointer (dev->br_port,NULL );call_rcu (&p->rcu, destroy_nbp_rcu); }
@@ -217,7 +214,6 @@ static struct net_bridge_port *new_nbp(s p->dev = dev; p->path_cost = cost; p->priority =0x8000 >>BR_PORT_BITS ; - dev->br_port = p; p->port_no = index;br_init_port (p); p->state =BR_STATE_DISABLED ;
@@ -360,6 +356,7 @@ int br_add_if(struct net_bridge *br, strelse if ((err =br_sysfs_addif (p)))del_nbp (p);else { +rcu_assign_pointer (dev->br_port, p);dev_set_promiscuity (dev,1 );list_add_rcu (&p->list, &br->port_list);
net/bridge/br_input.c +8 -2
@@ -45,11 +45,17 @@ static void br_pass_frame_up(struct net_int br_handle_frame_finish (struct sk_buff *skb) {const unsigned char *dest =eth_hdr (skb)->h_dest; -struct net_bridge_port *p = skb->dev->br_port; -struct net_bridge *br = p->br; +struct net_bridge_port *p =rcu_dereference (skb->dev->br_port); +struct net_bridge *br;struct net_bridge_fdb_entry *dst;int passedup =0 ; +if (unlikely (!p || p->state ==BR_STATE_DISABLED )) { +kfree_skb (skb); +return 0 ; + } + + br = p->br;/* insert into forwarding database after filtering to avoid spoofing */ br_fdb_update (p->br, p,eth_hdr (skb)->h_source);
net/bridge/br_stp_bpdu.c +6 -2
@@ -136,10 +136,13 @@ static const unsigned char header[6] = {/* NO locks */ int br_stp_handle_bpdu (struct sk_buff *skb) { -struct net_bridge_port *p = skb->dev->br_port; -struct net_bridge *br = p->br; +struct net_bridge_port *p =rcu_dereference (skb->dev->br_port); +struct net_bridge *br;unsigned char *buf; +if (!p) + goto err; +/* insert into forwarding database after filtering to avoid spoofing */ br_fdb_update (p->br, p,eth_hdr (skb)->h_source);
@@ -150,6 +153,7 @@ int br_stp_handle_bpdu(struct sk_buff *s buf =skb_pull (skb,sizeof (header)); + br = p->br;spin_lock_bh (&br->lock);if (p->state ==BR_STATE_DISABLED || !(br->dev->flags & IFF_UP)
--
PATCH 21/23 emu10k1 - Fix the confliction of Front control
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Fix the confliction of 'Front' controls on models with STAC9758 codec.
---
sound/pci/emu10k1/emumixer.c | 2 ++ 1 files changed, 2 insertions(+)
sound/pci/emu10k1/emumixer.c +2 -0
@@ -750,6 +750,8 @@ int __devinit snd_emu10k1_mixer(emu10k1_"Master Mono Playback Volume" ,"PCM Out Path & Mute" ,"Mono Output Select" , +"Front Playback Switch" , +"Front Playback Volume" ,"Surround Playback Switch" ,"Surround Playback Volume" , "Center Playback Switch",
--
PATCH 22/23 x86_64: Dont record local apic ids when they are disabled in MADT
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
Some broken BIOS's had processors disabled, but same apic id as a valid processor. This causes acpi_processor_start() to think this disabled cpu is ok, and croak. So we dont record bad apicid's anymore.
http://bugzilla.kernel.org/show_bug.cgi?id=5930
---
arch/i386/kernel/acpi/boot.c | 13 ++++++++++--- 1 files changed, 10 insertions(+), 3 deletions(-)
arch/i386/kernel/acpi/boot.c +10 -3
@@ -248,10 +248,17 @@ acpi_parse_lapic(acpi_table_entry_headeracpi_table_print_madt_entry (header); -/* Register even disabled CPUs for cpu hotplug */ - - x86_acpiid_to_apicid[processor->acpi_id] = processor->id; +/* Record local apic id only when enabled */ +if (processor->flags.enabled) + x86_acpiid_to_apicid[processor->acpi_id] = processor->id; +/* +* We need to register disabled CPU as well to permit +* counting disabled CPUs. This allows us to size +* cpus_possible_map more accurately, to permit +* to not preallocating memory for all NR_CPUS +* when we use CPU hotplug. +*/ mp_register_lapic (processor->id,/* APIC ID */ processor->flags.enabled);/* Enabled? */
--
PATCH 23/23 __cmpxchg() must really always be inlined
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know.
------------------
With the latest 2.6.15 kernel builds for alpha on Debian, we ran into a problem with undefined references to __cmpxchg_called_with_bad_pointer() in a couple of kernel modules (xfs.ko and drm.ko; see
http://bugs.debian.org/347556).
It looks like people have been trying to out-clever each other wrt the definition of "inline" on this architecture :), with the result that __cmpxchg(), which must be inlined so the compiler can see its argument is const, is not guaranteed to be inlined. Indeed, it was not being inlined when building with -Os.
The attached patch fixes the issue by adding an __attribute__((always_inline)) explicitly to the definition of __cmpxchg() instead of relying on redefines of "inline" elsewhere to make this happen.
---
include/asm-alpha/system.h | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-)
include/asm-alpha/system.h +1 -1
@@ -562,7 +562,7 @@ __cmpxchg_u64(volatile long *m, unsignedif something tries to do an invalid cmpxchg(). */ extern void __cmpxchg_called_with_bad_pointer (void ); -static inline unsigned long +static __always_inline unsigned long __cmpxchg (volatile void *ptr,unsigned long old,unsigned long new,int size) {switch (size) {
--
[PATCH 24/23] md: remove slashes from disk names when creation dev names in sysfs
In reply to Chris Wright-stable review patch. If anyone has any objections, please let us know. (this should've been in the series. i missed this one, mea culpa)
------------------
e.g. The sx8 driver uses names like sx8/0.
This would make a md component dev name like
/sys/block/md0/md/dev-sx8/0
which is not allowed. So we change the '/' to '!' just like fs/partitions/check.c(register_disk) does.
--- drivers/md/md.c | 3 +++ 1 files changed, 3 insertions(+)
drivers/md/md.c +3 -0
@@ -1182,6 +1182,7 @@ static int bind_rdev_to_array(mdk_rdev_tmdk_rdev_t *same_pdev;char b[BDEVNAME_SIZE ], b2[BDEVNAME_SIZE ];struct kobject *ko; +char *s;if (rdev->mddev) {MD_BUG ();
@@ -1213,6 +1214,8 @@ static int bind_rdev_to_array(mdk_rdev_tbdevname (rdev->bdev,b);if (kobject_set_name (&rdev->kobj,"dev-%s" , b) <0 )return -ENOMEM ; +while ( (s=strchr (rdev->kobj.k_name,'/' )) !=NULL ) + *s ='!' ;list_add (&rdev->same_set, &mddev->disks); rdev->mddev = mddev;