- 06 Mar, 2013 40 commits
-
-
David Vrabel authored
[ Upstream commit 35876b5f ] netbk_count_requests() could detect an error, call netbk_fatal_tx_error() but return 0. The vif may then be used afterwards (e.g., in a call to netbk_tx_error(). Since netbk_fatal_tx_error() could set vif->refcnt to 1, the vif may be freed immediately after the call to netbk_fatal_tx_error() (e.g., if the vif is also removed). Netback thread Xenwatch thread ------------------------------------------- netbk_fatal_tx_err() netback_remove() xenvif_disconnect() ... free_netdev() netbk_tx_err() Oops! Signed-off-by:
Wei Liu <wei.liu2@citrix.com> Signed-off-by:
Jan Beulich <JBeulich@suse.com> Signed-off-by:
David Vrabel <david.vrabel@citrix.com> Reported-by:
Christopher S. Aker <caker@theshore.net> Acked-by:
Ian Campbell <ian.campbell@citrix.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Ben Hutchings <ben@decadent.org.uk>
-
Stephen Hemminger authored
[ Upstream commit 547b4e71 ] Spanning Tree Protocol packets should have always been marked as control packets, this causes them to get queued in the high prirority FIFO. As Radia Perlman mentioned in her LCA talk, STP dies if bridge gets overloaded and can't communicate. This is a long-standing bug back to the first versions of Linux bridge. Signed-off-by:
Stephen Hemminger <stephen@networkplumber.org> Signed-off-by:
-
Jan Beulich authored
commit 51ac8893 upstream. ... as being guest triggerable (e.g. by invoking XEN_PCI_OP_enable_msi{,x} on a device not being MSI/MSI-X capable). This is CVE-2013-0231 / XSA-43. Also make the two messages uniform in both their wording and severity. Signed-off-by:
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> [bwh: Backported to 3.2: add #include <linux/ratelimited.h>, needed by printk_ratelimited()] Signed-off-by:
-
Helge Deller authored
commit 4f4ffc3a upstream. automount-support is broken on the parisc architecture, because the existing #if list does not include a check for defined(__hppa__). The HPPA (parisc) architecture is similiar to other 64bit Linux targets where we have to define autofs_wqt_t (which is passed back and forth to user space) as int type which has a size of 32bit across 32 and 64bit kernels. During the discussion on the mailing list, H. Peter Anvin suggested to invert the #if list since only specific platforms (specifically those who do not have a 32bit userspace, like IA64 and Alpha) should have autofs_wqt_t as unsigned long type. This suggestion is probably the best way to go, since Arm64 (and maybe others?) seems to have a non-working automounter. So in the long run even for other new upcoming architectures this inverted check seem to be the best solution, since it will not require them to change this #if again (unless they are 64bit only). Signed-off-by:
Helge Deller <deller@gmx.de> Acked-by:
H. Peter Anvin <hpa@zytor.com> Acked-by:
Ian Kent <raven@themaw.net> Acked-by:
Catalin Marinas <catalin.marinas@arm.com> CC: James Bottomley <James.Bottomley@HansenPartnership.com> CC: Rolf Eike Beer <eike-kernel@sf-tec.de> [bwh: Backported to 3.2: adjust filename] Signed-off-by:
-
Heiko Carstens authored
commit d911e03d upstream. Since ed4f2094 "s390/time: fix sched_clock() overflow" a new helper function is used to avoid overflows when converting TOD format values to nanosecond values. The kvm interrupt code formerly however only worked by accident because of an overflow. It tried to program a timer that would expire in more than ~29 years. Because of the old TOD-to-nanoseconds overflow bug the real expiry value however was much smaller, but now it isn't anymore. This however triggers yet another bug in the function that programs the clock comparator s390_next_ktime(): if the absolute "expires" value is after 2042 this will result in an overflow and the programmed value is lower than the current TOD value which immediatly triggers a clock comparator (= timer) interrupt. Since the timer isn't expired it will be programmed immediately again and so on... the result is a dead system. To fix this simply program the maximum possible value if an overflow is detected. Reported-by:
Christian Borntraeger <borntraeger@de.ibm.com> Tested-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by:
-
Alex Deucher authored
commit ed39fadd upstream. Some chips seem to need a little delay after blacking out the MC before the requests actually stop. May fix: https://bugs.freedesktop.org/show_bug.cgi?id=56139 https://bugs.freedesktop.org/show_bug.cgi?id=57567Signed-off-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Alexander Duyck authored
commit ae1c07a6 upstream. For some reason the reading of the RQDPC register was being artificially limited to 4K. Instead of limiting the value we should read the value and add the full amount. Otherwise this can lead to a misleading number of dropped packets when the actual value is in fact much higher. Signed-off-by:
Alexander Duyck <alexander.h.duyck@intel.com> Tested-by:
Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by:
-
Paolo Bonzini authored
commit 3a2d63f8 upstream. There are two problems with shutdown in the NBD driver. 1: Receiving the NBD_DISCONNECT ioctl does not sync the filesystem. This patch adds the sync operation into __nbd_ioctl()'s NBD_DISCONNECT handler. This is useful because BLKFLSBUF is restricted to processes that have CAP_SYS_ADMIN, and the NBD client may not possess it (fsync of the block device does not sync the filesystem, either). 2: Once we clear the socket we have no guarantee that later reads will come from the same backing storage. The patch adds calls to kill_bdev() in __nbd_ioctl()'s socket clearing code so the page cache is cleaned, lest reads that hit on the page cache will return stale data from the previously-accessible disk. Example: # qemu-nbd -r -c/dev/nbd0 /dev/sr0 # file -s /dev/nbd0 /dev/stdin: # UDF filesystem data (version 1.5) etc. # qemu-nbd -d /dev/nbd0 # qemu-nbd -r -c/dev/nbd0 /dev/sda # file -s /dev/nbd0 /dev/stdin: # UDF filesystem data (version 1.5) etc. While /dev/sda has: # file -s /dev/sda /dev/sda: x86 boot sector; etc. Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com> Acked-by:
Paul Clements <Paul.Clements@steeleye.com> Cc: Alex Bligh <alex@alex.org.uk> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> [bwh: Backported to 3.2: - Adjusted context - s/\bnbd\b/lo/ - Incorporate export of kill_bdev() from commit ff01bb48 ('fs: move code out of buffer.c')] Signed-off-by:
-
Xi Wang authored
commit df1778be upstream. The null check of `strchr() + 1' is broken, which is always non-null, leading to OOB read. Instead, check the result of strchr(). Signed-off-by:
Xi Wang <xi.wang@gmail.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by:
-
Tejun Heo authored
commit 326cf0f0 upstream. Most functions in idr fail to deal with the high bits when the idr tree grows to the maximum height. * idr_get_empty_slot() stops growing idr tree once the depth reaches MAX_IDR_LEVEL - 1, which is one depth shallower than necessary to cover the whole range. The function doesn't even notice that it didn't grow the tree enough and ends up allocating the wrong ID given sufficiently high @starting_id. For example, on 64 bit, if the starting id is 0x7fffff01, idr_get_empty_slot() will grow the tree 5 layer deep, which only covers the 30 bits and then proceed to allocate as if the bit 30 wasn't specified. It ends up allocating 0x3fffff01 without the bit 30 but still returns 0x7fffff01. * __idr_remove_all() will not remove anything if the tree is fully grown. * idr_find() can't find anything if the tree is fully grown. * idr_for_each() and idr_get_next() can't iterate anything if the tree is fully grown. Fix it by introducing idr_max() which returns the maximum possible ID given the depth of tree and replacing the id limit checks in all affected places. As the idr_layer pointer array pa[] needs to be 1 larger than the maximum depth, enlarge pa[] arrays by one. While this plugs the discovered issues, the whole code base is horrible and in desparate need of rewrite. It's fragile like hell, Signed-off-by:
Tejun Heo <tj@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
-
Hugh Dickins authored
commit 9f7de827 upstream. Make one small adjustment to idr_get_next(): take the height from the top layer (stable under RCU) instead of from the root (unprotected by RCU), as idr_find() does: so that it can be used with RCU locking. Copied comment on RCU locking from idr_find(). Signed-off-by:
Hugh Dickins <hughd@google.com> Acked-by:
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by:
Li Zefan <lizf@cn.fujitsu.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Acked-by:
-
Tejun Heo authored
commit 3bec60d5 upstream. fw_device_init() didn't check whether the allocated minor number isn't too large. Fail if it goes overflows MINORBITS. Signed-off-by:
Stefan Richter <stefanr@s5r6.in-berlin.de> Acked-by:
-
Tejun Heo authored
commit ce23bba8 upstream. idr allocation in blk_alloc_devt() wasn't synchronized against lookup and removal, and its limit check was off by one - 1 << MINORBITS is the number of minors allowed, not the maximum allowed minor. Add locking and rename MAX_EXT_DEVT to NR_EXT_DEVT and fix limit checking. Signed-off-by:
Jens Axboe <axboe@kernel.dk> Signed-off-by:
-
Tejun Heo authored
commit 6cdae741 upstream. The iteration logic of idr_get_next() is borrowed mostly verbatim from idr_for_each(). It walks down the tree looking for the slot matching the current ID. If the matching slot is not found, the ID is incremented by the distance of single slot at the given level and repeats. The implementation assumes that during the whole iteration id is aligned to the layer boundaries of the level closest to the leaf, which is true for all iterations starting from zero or an existing element and thus is fine for idr_for_each(). However, idr_get_next() may be given any point and if the starting id hits in the middle of a non-existent layer, increment to the next layer will end up skipping the same offset into it. For example, an IDR with IDs filled between [64, 127] would look like the following. [ 0 64 ... ] /----/ | | | NULL [ 64 ... 127 ] If idr_get_next() is called with 63 as the starting point, it will try to follow down the pointer from 0. As it is NULL, it will then try to proceed to the next slot in the same level by adding the slot distance at that level which is 64 - making the next try 127. It goes around the loop and finds and returns 127 skipping [64, 126]. Note that this bug also triggers in idr_for_each_entry() loop which deletes during iteration as deletions can make layers go away leaving the iteration with unaligned ID into missing layers. Fix it by ensuring proceeding to the next slot doesn't carry over the unaligned offset - ie. use round_up(id + 1, slot_distance) instead of id += slot_distance. Signed-off-by:
David Teigland <teigland@redhat.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by:
-
Tomas Henzl authored
commit 7b74e912 upstream. While adding and removing a lot of disks disks and partitions this sometimes shows up: WARNING: at fs/sysfs/dir.c:512 sysfs_add_one+0xc9/0x130() (Not tainted) Hardware name: sysfs: cannot create duplicate filename '/dev/block/259:751' Modules linked in: raid1 autofs4 bnx2fc cnic uio fcoe libfcoe libfc 8021q scsi_transport_fc scsi_tgt garp stp llc sunrpc cpufreq_ondemand powernow_k8 freq_table mperf ipv6 dm_mirror dm_region_hash dm_log power_meter microcode dcdbas serio_raw amd64_edac_mod edac_core edac_mce_amd i2c_piix4 i2c_core k10temp bnx2 sg ixgbe dca mdio ext4 mbcache jbd2 dm_round_robin sr_mod cdrom sd_mod crc_t10dif ata_generic pata_acpi pata_atiixp ahci mptsas mptscsih mptbase scsi_transport_sas dm_multipath dm_mod [last unloaded: scsi_wait_scan] Pid: 44103, comm: async/16 Not tainted 2.6.32-195.el6.x86_64 #1 Call Trace: warn_slowpath_common+0x87/0xc0 warn_slowpath_fmt+0x46/0x50 sysfs_add_one+0xc9/0x130 sysfs_do_create_link+0x12b/0x170 sysfs_create_link+0x13/0x20 device_add+0x317/0x650 idr_get_new+0x13/0x50 add_partition+0x21c/0x390 rescan_partitions+0x32b/0x470 sd_open+0x81/0x1f0 [sd_mod] __blkdev_get+0x1b6/0x3c0 blkdev_get+0x10/0x20 register_disk+0x155/0x170 add_disk+0xa6/0x160 sd_probe_async+0x13b/0x210 [sd_mod] add_wait_queue+0x46/0x60 async_thread+0x102/0x250 default_wake_function+0x0/0x20 async_thread+0x0/0x250 kthread+0x96/0xa0 child_rip+0xa/0x20 kthread+0x0/0xa0 child_rip+0x0/0x20 This most likely happens because dev_t is freed while the number is still used and idr_get_new() is not protected on every use. The fix adds a mutex where it wasn't before and moves the dev_t free function so it is called after device del. Signed-off-by:
Tomas Henzl <thenzl@redhat.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by:
-
Xiaowei.Hu authored
commit 309a85b6 upstream. ocfs2_block_group_alloc_discontig() disables chain relink by setting ac->ac_allow_chain_relink = 0 because it grabs clusters from multiple cluster groups. It doesn't keep the credits for all chain relink,but ocfs2_claim_suballoc_bits overrides this in this call trace: ocfs2_block_group_claim_bits()->ocfs2_claim_clusters()-> __ocfs2_claim_clusters()->ocfs2_claim_suballoc_bits() ocfs2_claim_suballoc_bits set ac->ac_allow_chain_relink = 1; then call ocfs2_search_chain() one time and disable it again, and then we run out of credits. Fix is to allow relink by default and disable it in ocfs2_block_group_alloc_discontig. Without this patch, End-users will run into a crash due to run out of credits, backtrace like this: RIP: 0010:[<ffffffffa0808b14>] [<ffffffffa0808b14>] jbd2_journal_dirty_metadata+0x164/0x170 [jbd2] RSP: 0018:ffff8801b919b5b8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff88022139ddc0 RCX: ffff880159f652d0 RDX: ffff880178aa3000 RSI: ffff880159f652d0 RDI: ffff880087f09bf8 RBP: ffff8801b919b5e8 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000001e00 R11: 00000000000150b0 R12: ffff880159f652d0 R13: ffff8801a0cae908 R14: ffff880087f09bf8 R15: ffff88018d177800 FS: 00007fc9b0b6b6e0(0000) GS:ffff88022fd40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 000000000040819c CR3: 0000000184017000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process dd (pid: 9945, threadinfo ffff8801b919a000, task ffff880149a264c0) Call Trace: ocfs2_journal_dirty+0x2f/0x70 [ocfs2] ocfs2_relink_block_group+0x111/0x480 [ocfs2] ocfs2_search_chain+0x455/0x9a0 [ocfs2] ... Signed-off-by:
Xiaowei.Hu <xiaowei.hu@oracle.com> Reviewed-by:
Srinivas Eeda <srinivas.eeda@oracle.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Signed-off-by:
-
Jeff Liu authored
commit 32918dd9 upstream. We need to re-initialize the security for a new reflinked inode with its parent dirs if it isn't specified to be preserved for ocfs2_reflink(). However, the code logic is broken at ocfs2_init_security_and_acl() although ocfs2_init_security_get() succeed. As a result, ocfs2_acl_init() does not involked and therefore the default ACL of parent dir was missing on the new inode. Note this was introduced by 9d8f13ba ("security: new security_inode_init_security API adds function callback") To reproduce: set default ACL for the parent dir(ocfs2 in this case): $ setfacl -m default:user:jeff:rwx ../ocfs2/ $ getfacl ../ocfs2/ # file: ../ocfs2/ # owner: jeff # group: jeff user::rwx group::r-x other::r-x default:user::rwx default:user:jeff:rwx default:group::r-x default:mask::rwx default:other::r-x $ touch a $ getfacl a # file: a # owner: jeff # group: jeff user::rw- group::rw- other::r-- Before patching, create reflink file b from a, the user default ACL entry(user:jeff:rwx)was missing: $ ./ocfs2_reflink a b $ getfacl b # file: b # owner: jeff # group: jeff user::rw- group::rw- other::r-- In this case, the end user can also observed an error message at syslog: (ocfs2_reflink,3229,2):ocfs2_init_security_and_acl:7193 ERROR: status = 0 After applying this patch, create reflink file c from a: $ ./ocfs2_reflink a c $ getfacl c # file: c # owner: jeff # group: jeff user::rw- user:jeff:rwx #effective:rw- group::r-x #effective:r-- mask::rw- other::r-- Test program: /* Usage: reflink <source> <dest> */ #include <stdio.h> #include <stdint.h> #include <stdbool.h> #include <string.h> #include <errno.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <sys/ioctl.h> static int reflink_file(char const *src_name, char const *dst_name, bool preserve_attrs) { int fd; #ifndef REFLINK_ATTR_NONE # define REFLINK_ATTR_NONE 0 #endif #ifndef REFLINK_ATTR_PRESERVE # define REFLINK_ATTR_PRESERVE 1 #endif #ifndef OCFS2_IOC_REFLINK struct reflink_arguments { uint64_t old_path; uint64_t new_path; uint64_t preserve; }; # define OCFS2_IOC_REFLINK _IOW ('o', 4, struct reflink_arguments) #endif struct reflink_arguments args = { .old_path = (unsigned long) src_name, .new_path = (unsigned long) dst_name, .preserve = preserve_attrs ? REFLINK_ATTR_PRESERVE : REFLINK_ATTR_NONE, }; fd = open(src_name, O_RDONLY); if (fd < 0) { fprintf(stderr, "Failed to open %s: %s\n", src_name, strerror(errno)); return -1; } if (ioctl(fd, OCFS2_IOC_REFLINK, &args) < 0) { fprintf(stderr, "Failed to reflink %s to %s: %s\n", src_name, dst_name, strerror(errno)); return -1; } } int main(int argc, char *argv[]) { if (argc != 3) { fprintf(stdout, "Usage: %s source dest\n", argv[0]); return 1; } return reflink_file(argv[1], argv[2], 0); } Signed-off-by:
Jie Liu <jeff.liu@oracle.com> Reviewed-by:
Tao Ma <boyu.mt@taobao.com> Cc: Mimi Zohar <zohar@linux.vnet.ibm.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Signed-off-by:
-
H. Peter Anvin authored
commit 7c100936 upstream. On non-BIOS platforms it is possible that the BIOS data area contains garbage instead of being zeroed or something equivalent (firmware people: we are talking of 1.5K here, so please do the sane thing.) We need on the order of 20-30K of low memory in order to boot, which may grow up to < 64K in the future. We probably want to avoid the lowest of the low memory. At the same time, it seems extremely unlikely that a legitimate EBDA would ever reach down to the 128K (which would require it to be over half a megabyte in size.) Thus, pick 128K as the cutoff for "this is insane, ignore." We may still end up reserving a bunch of extra memory on the low megabyte, but that is not really a major issue these days. In the worst case we lose 512K of RAM. This code really should be merged with trim_bios_range() in arch/x86/kernel/setup.c, but that is a bigger patch for a later merge window. Reported-by:
Darren Hart <dvhart@linux.intel.com> Signed-off-by:
H. Peter Anvin <hpa@linux.intel.com> Cc: Matt Fleming <matt.fleming@intel.com> Link: http://lkml.kernel.org/n/tip-oebml055yyfm8yxmria09rja@git.kernel.orgSigned-off-by:
-
Jan Kara authored
commit 9b171e0c upstream. Running AIO is pinning inode in memory using file reference. Once AIO is completed using aio_complete(), file reference is put and inode can be freed from memory. So we have to be sure that calling aio_complete() is the last thing we do with the inode. Signed-off-by:
Jan Kara <jack@suse.cz> Acked-by:
Jeff Moyer <jmoyer@redhat.com> Acked-by:
Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
-
Konrad Rzeszutek Wilk authored
commit a2fd6419 upstream. Both the PowerPC hypervisor and Xen hypervisor can utilize the hvc driver. Cc: Greg KH <gregkh@linuxfoundation.org> Signed-off-by:
-
Konrad Rzeszutek Wilk authored
commit 2482a92e upstream. The earlyprintk for Xen PV guests utilizes a simple hypercall (console_io) to provide output to Xen emergency console. Note that the Xen hypervisor should be booted with 'loglevel=all' to output said information. Reported-by:
-
Shawn Guo authored
commit ef4d0888 upstream. When commit 95a2482a (mmc: sdhci-esdhc-imx: add basic imx6q usdhc support) works around host version issue on imx6q, it gets the register address fixup "reg ^= 2" lost for imx25/35/51/53 esdhc. Thus, the controller version on these SoCs is wrongly identified as v1 while it's actually v2. Add the address fixup back and take a different approach to correct imx6q host version, so that the host version read gets back to work for all SoCs. Signed-off-by:
Shawn Guo <shawn.guo@linaro.org> Signed-off-by:
Chris Ball <cjb@laptop.org> Signed-off-by:
-
Greg Thelen authored
commit 5f00110f upstream. The tmpfs remount logic preserves filesystem mempolicy if the mpol=M option is not specified in the remount request. A new policy can be specified if mpol=M is given. Before this patch remounting an mpol bound tmpfs without specifying mpol= mount option in the remount request would set the filesystem's mempolicy object to a freed mempolicy object. To reproduce the problem boot a DEBUG_PAGEALLOC kernel and run: # mkdir /tmp/x # mount -t tmpfs -o size=100M,mpol=interleave nodev /tmp/x # grep /tmp/x /proc/mounts nodev /tmp/x tmpfs rw,relatime,size=102400k,mpol=interleave:0-3 0 0 # mount -o remount,size=200M nodev /tmp/x # grep /tmp/x /proc/mounts nodev /tmp/x tmpfs rw,relatime,size=204800k,mpol=??? 0 0 # note ? garbage in mpol=... output above # dd if=/dev/zero of=/tmp/x/f count=1 # panic here Panic: BUG: unable to handle kernel NULL pointer dereference at (null) IP: [< (null)>] (null) [...] Oops: 0010 [#1] SMP DEBUG_PAGEALLOC Call Trace: mpol_shared_policy_init+0xa5/0x160 shmem_get_inode+0x209/0x270 shmem_mknod+0x3e/0xf0 shmem_create+0x18/0x20 vfs_create+0xb5/0x130 do_last+0x9a1/0xea0 path_openat+0xb3/0x4d0 do_filp_open+0x42/0xa0 do_sys_open+0xfe/0x1e0 compat_sys_open+0x1b/0x20 cstar_dispatch+0x7/0x1f Non-debug kernels will not crash immediately because referencing the dangling mpol will not cause a fault. Instead the filesystem will reference a freed mempolicy object, which will cause unpredictable behavior. The problem boils down to a dropped mpol reference below if shmem_parse_options() does not allocate a new mpol: config = *sbinfo shmem_parse_options(data, &config, true) mpol_put(sbinfo->mpol) sbinfo->mpol = config.mpol /* BUG: saves unreferenced mpol */ This patch avoids the crash by not releasing the mempolicy if shmem_parse_options() doesn't create a new mpol. How far back does this issue go? I see it in both 2.6.36 and 3.3. I did not look back further. Signed-off-by:
Greg Thelen <gthelen@google.com> Acked-by:
-
Mel Gorman authored
commit 67d46b29 upstream. Rob van der Heij reported the following (paraphrased) on private mail. The scenario is that I want to avoid backups to fill up the page cache and purge stuff that is more likely to be used again (this is with s390x Linux on z/VM, so I don't give it as much memory that we don't care anymore). So I have something with LD_PRELOAD that intercepts the close() call (from tar, in this case) and issues a posix_fadvise() just before closing the file. This mostly works, except for small files (less than 14 pages) that remains in page cache after the face. Unfortunately Rob has not had a chance to test this exact patch but the test program below should be reproducing the problem he described. The issue is the per-cpu pagevecs for LRU additions. If the pages are added by one CPU but fadvise() is called on another then the pages remain resident as the invalidate_mapping_pages() only drains the local pagevecs via its call to pagevec_release(). The user-visible effect is that a program that uses fadvise() properly is not obeyed. A possible fix for this is to put the necessary smarts into invalidate_mapping_pages() to globally drain the LRU pagevecs if a pagevec page could not be discarded. The downside with this is that an inode cache shrink would send a global IPI and memory pressure potentially causing global IPI storms is very undesirable. Instead, this patch adds a check during fadvise(POSIX_FADV_DONTNEED) to check if invalidate_mapping_pages() discarded all the requested pages. If a subset of pages are discarded it drains the LRU pagevecs and tries again. If the second attempt fails, it assumes it is due to the pages being mapped, locked or dirty and does not care. With this patch, an application using fadvise() correctly will be obeyed but there is a downside that a malicious application can force the kernel to send global IPIs and increase overhead. If accepted, I would like this to be considered as a -stable candidate. It's not an urgent issue but it's a system call that is not working as advertised which is weak. The following test program demonstrates the problem. It should never report that pages are still resident but will without this patch. It assumes that CPU 0 and 1 exist. int main() { int fd; int pagesize = getpagesize(); ssize_t written = 0, expected; char *buf; unsigned char *vec; int resident, i; cpu_set_t set; /* Prepare a buffer for writing */ expected = FILESIZE_PAGES * pagesize; buf = malloc(expected + 1); if (buf == NULL) { printf("ENOMEM\n"); exit(EXIT_FAILURE); } buf[expected] = 0; memset(buf, 'a', expected); /* Prepare the mincore vec */ vec = malloc(FILESIZE_PAGES); if (vec == NULL) { printf("ENOMEM\n"); exit(EXIT_FAILURE); } /* Bind ourselves to CPU 0 */ CPU_ZERO(&set); CPU_SET(0, &set); if (sched_setaffinity(getpid(), sizeof(set), &set) == -1) { perror("sched_setaffinity"); exit(EXIT_FAILURE); } /* open file, unlink and write buffer */ fd = open("fadvise-test-file", O_CREAT|O_EXCL|O_RDWR); if (fd == -1) { perror("open"); exit(EXIT_FAILURE); } unlink("fadvise-test-file"); while (written < expected) { ssize_t this_write; this_write = write(fd, buf + written, expected - written); if (this_write == -1) { perror("write"); exit(EXIT_FAILURE); } written += this_write; } free(buf); /* * Force ourselves to another CPU. If fadvise only flushes the local * CPUs pagevecs then the fadvise will fail to discard all file pages */ CPU_ZERO(&set); CPU_SET(1, &set); if (sched_setaffinity(getpid(), sizeof(set), &set) == -1) { perror("sched_setaffinity"); exit(EXIT_FAILURE); } /* sync and fadvise to discard the page cache */ fsync(fd); if (posix_fadvise(fd, 0, expected, POSIX_FADV_DONTNEED) == -1) { perror("posix_fadvise"); exit(EXIT_FAILURE); } /* map the file and use mincore to see which parts of it are resident */ buf = mmap(NULL, expected, PROT_READ, MAP_SHARED, fd, 0); if (buf == NULL) { perror("mmap"); exit(EXIT_FAILURE); } if (mincore(buf, expected, vec) == -1) { perror("mincore"); exit(EXIT_FAILURE); } /* Check residency */ for (i = 0, resident = 0; i < FILESIZE_PAGES; i++) { if (vec[i]) resident++; } if (resident != 0) { printf("Nr unexpected pages resident: %d\n", resident); exit(EXIT_FAILURE); } munmap(buf, expected); close(fd); free(vec); exit(EXIT_SUCCESS); } Signed-off-by:
Mel Gorman <mgorman@suse.de> Reported-by:
Rob van der Heij <rvdheij@gmail.com> Tested-by:
-
Robin Holt authored
commit 751efd86 upstream. There is a race condition between mmu_notifier_unregister() and __mmu_notifier_release(). Assume two tasks, one calling mmu_notifier_unregister() as a result of a filp_close() ->flush() callout (task A), and the other calling mmu_notifier_release() from an mmput() (task B). A B t1 srcu_read_lock() t2 if (!hlist_unhashed()) t3 srcu_read_unlock() t4 srcu_read_lock() t5 hlist_del_init_rcu() t6 synchronize_srcu() t7 srcu_read_unlock() t8 hlist_del_rcu() <--- NULL pointer deref. Additionally, the list traversal in __mmu_notifier_release() is not protected by the by the mmu_notifier_mm->hlist_lock which can result in callouts to the ->release() notifier from both mmu_notifier_unregister() and __mmu_notifier_release(). -stable suggestions: The stable trees prior to 3.7.y need commits 21a92735 and 70400303 cherry-picked in that order prior to cherry-picking this commit. The 3.7.y tree already has those two commits. Signed-off-by:
Robin Holt <holt@sgi.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Cc: Avi Kivity <avi@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Sagi Grimberg <sagig@mellanox.co.il> Cc: Haggai Eran <haggaie@mellanox.com> Signed-off-by:
-
Andrea Arcangeli authored
commit 70400303 upstream. The variable must be static especially given the variable name. s/RCU/SRCU/ over a few comments. Signed-off-by:
Andrea Arcangeli <aarcange@redhat.com> Cc: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Cc: Sagi Grimberg <sagig@mellanox.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Haggai Eran <haggaie@mellanox.com> Signed-off-by:
-
Sagi Grimberg authored
commit 21a92735 upstream. With an RCU based mmu_notifier implementation, any callout to mmu_notifier_invalidate_range_{start,end}() or mmu_notifier_invalidate_page() would not be allowed to call schedule() as that could potentially allow a modification to the mmu_notifier structure while it is currently being used. Since srcu allocs 4 machine words per instance per cpu, we may end up with memory exhaustion if we use srcu per mm. So all mms share a global srcu. Note that during large mmu_notifier activity exit & unregister paths might hang for longer periods, but it is tolerable for current mmu_notifier clients. Signed-off-by:
Sagi Grimberg <sagig@mellanox.co.il> Signed-off-by:
-
Phileas Fogg authored
commit 8520e443 upstream. Disable hard IRQ before kexec a new kernel image. Not doing it can result in corrupted data in the memory segments reserved for the new kernel. Signed-off-by:
Phileas Fogg <phileas-fogg@mail.ru> Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Jan Kara authored
commit 54c807e7 upstream. Running AIO is pinning inode in memory using file reference. Once AIO is completed using aio_complete(), file reference is put and inode can be freed from memory. So we have to be sure that calling aio_complete() is the last thing we do with the inode. CC: Christoph Hellwig <hch@infradead.org> CC: Jens Axboe <axboe@kernel.dk> CC: Jeff Moyer <jmoyer@redhat.com> Acked-by:
-
Lukas Czerner authored
commit 304e220f upstream. ext4_has_free_clusters() should tell us whether there is enough free clusters to allocate, however number of free clusters in the file system is converted to blocks using EXT4_C2B() which is not only wrong use of the macro (we should have used EXT4_NUM_B2C) but it's also completely wrong concept since everything else is in cluster units. Moreover when calculating number of root clusters we should be using macro EXT4_NUM_B2C() instead of EXT4_B2C() otherwise the result might be off by one. However r_blocks_count should always be a multiple of the cluster ratio so doing a plain bit shift should be enough here. We avoid using EXT4_B2C() because it's confusing. As a result of the first problem number of free clusters is much bigger than it should have been and ext4_has_free_clusters() would return 1 even if there is really not enough free clusters available. Fix this by removing the EXT4_C2B() conversion of free clusters and using bit shift when calculating number of root clusters. This bug affects number of xfstests tests covering file system ENOSPC situation handling. With this patch most of the ENOSPC problems with bigalloc file system disappear, especially the errors caused by delayed allocation not having enough space when the actual allocation is finally requested. Signed-off-by:
Lukas Czerner <lczerner@redhat.com> Signed-off-by:
"Theodore Ts'o" <tytso@mit.edu> Signed-off-by:
-
Lars-Peter Clausen authored
commit 5eb02c01 upstream. Clearing the NSTBY bit in the control register also automatically clears the BLEN bit. So we need to make sure to set it again during resume, otherwise the backlight will stay off. Signed-off-by:
Lars-Peter Clausen <lars@metafoo.de> Acked-by:
Michael Hennerich <michael.hennerich@analog.com> Signed-off-by:
-
Junxiao Bi authored
commit 3278bb74 upstream. If lockres refresh failed, the super lock will never be released which will cause some processes on other cluster nodes hung forever. Signed-off-by:
Junxiao Bi <junxiao.bi@oracle.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Signed-off-by:
-
MITSUNARI Shigeo authored
commit 7630b661 upstream. We found that bdev->bd_invalidated was left set once revalidate_disk() is called, which results in page cache flush every time that device is open. Specifically, we found this problem in MD block device. Once we resize a MD device, mdadm --monitor periodically flush all page cache for that device every 60 or 1000 seconds when it opens the device. This bug lies since at least 3.2.0 till the latest kernel(3.6.2). Patch is attached. The following steps will reproduce the problem. 1. prepair a block device (eg /dev/sdb). 2. create two partitions: sudo parted /dev/sdb mklabel gpt mkpart primary 0% 50% mkpart primary 50% 100% 3. create a md device. sudo mdadm -C /dev/md/hoge -l 1 -n 2 -e 1.2 --assume-clean --auto=md --symlink=no /dev/sdb1 /dev/sdb2 4. create file system and mount it sudo mkfs.ext3 /dev/md/hoge sudo mkdir /mnt/test sudo mount /dev/md/hoge /mnt/test 5. try to resize the device sudo mdadm -G /dev/md/hoge --size=max 6. create a file to fill file cache. sudo dd if=/dev/urandom of=/mnt/test/data bs=1M count=10 and verify the current status of file by free command. 7. mdadm monitor will open the md device every 1000 seconds and you will find all file cache on the device are cleared. The timing can be reduced by the following steps. a) kill mdadm and restart it with --delay option /sbin/mdadm --monitor --delay=30 --pid-file /var/run/mdadm/monitor.pid --daemonise --scan --syslog or open the md device directly. sudo dd if=/dev/md/hoge of=/dev/null bs=4096 count=1 Signed-off-by:
MITSUNARI Shigeo <herumi@nifty.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Jeff Moyer <jmoyer@redhat.com> Signed-off-by:
-
Jim Somerville authored
commit 676a0675 upstream. Running the command: inotifywait -e unmount /mnt/disk immediately aborts with a -EINVAL return code. This is however a valid parameter. This abort occurs only if unmount is the sole event parameter. If other event parameters are supplied, then the unmount event wait will work. The problem was introduced by commit 44b350fc ("inotify: Fix mask checks"). In that commit, it states: The mask checks in inotify_update_existing_watch() and inotify_new_watch() are useless because inotify_arg_to_mask() sets FS_IN_IGNORED and FS_EVENT_ON_CHILD bits anyway. But instead of removing the useless checks, it did this: mask = inotify_arg_to_mask(arg); - if (unlikely(!mask)) + if (unlikely(!(mask & IN_ALL_EVENTS))) return -EINVAL; The problem is that IN_ALL_EVENTS doesn't include IN_UNMOUNT, and other parts of the code keep IN_UNMOUNT separate from IN_ALL_EVENTS. So the check should be: if (unlikely(!(mask & (IN_ALL_EVENTS | IN_UNMOUNT)))) But inotify_arg_to_mask(arg) always sets the IN_UNMOUNT bit in the mask anyway, so the check is always going to pass and thus should simply be removed. Also note that inotify_arg_to_mask completely controls what mask bits get set from arg, there's no way for invalid bits to get enabled there. Lets fix it by simply removing the useless broken checks. Signed-off-by:
Jim Somerville <Jim.Somerville@windriver.com> Signed-off-by:
Paul Gortmaker <paul.gortmaker@windriver.com> Cc: Jerome Marchand <jmarchan@redhat.com> Cc: John McCutchan <john@johnmccutchan.com> Cc: Robert Love <rlove@rlove.org> Cc: Eric Paris <eparis@parisplace.org> Signed-off-by:
-
Tejun Heo authored
commit e182bb38 upstream. When idr_find() was fed a negative ID, it used to look up the ID ignoring the sign bit before recent ("idr: remove MAX_IDR_MASK and move left MAX_IDR_* into idr.c") patch. Now a negative ID triggers a WARN_ON_ONCE(). __lock_timer() feeds timer_id from userland directly to idr_find() without sanitizing it which can trigger the above malfunctions. Add a range check on @timer_id before invoking idr_find() in __lock_timer(). While timer_t is defined as int by all archs at the moment, Andrew worries that it may be defined as a larger type later on. Make the test cover larger integers too so that it at least is guaranteed to not return the wrong timer. Note that WARN_ON_ONCE() in idr_find() on id < 0 is transitional precaution while moving away from ignoring MSB. Once it's gone we can remove the guard as long as timer_t isn't larger than int. Signed-off-by: Tejun Heo <tj@kernel.org>nnn Reported-by:
Sasha Levin <sasha.levin@oracle.com> Cc: Andrew Morton <akpm@linux-foundation.org> Link: http://lkml.kernel.org/r/20130220232412.GL3570@htj.dyndns.orgSigned-off-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Pawel Moll authored
commit b531f81b upstream. Commit 99fc8645 "ALSA: usb-mixer: parse descriptors with structs" introduced a set of useful parsers for descriptors. Unfortunately the parses for the Processing Unit Descriptor came with a very subtle bug... Functions uac_processing_unit_iProcessing() and uac_processing_unit_specific() were indexing the baSourceID array forgetting the fields before the iProcessing and process-specific descriptors. The problem was observed with Sound Blaster Extigy mixer, where nNrModes in Up/Down-mix Processing Unit Descriptor was accessed at offset 10 of the descriptor (value 0) instead of offset 15 (value 7). In result the resulting control had interesting limit values: Simple mixer control 'Channel Routing Mode Select',0 Capabilities: volume volume-joined penum Playback channels: Mono Capture channels: Mono Limits: 0 - -1 Mono: -1 [100%] Fixed by starting from the bmControls, which was calculated correctly, instead of baSourceID. Now the mentioned control is fine: Simple mixer control 'Channel Routing Mode Select',0 Capabilities: volume volume-joined penum Playback channels: Mono Capture channels: Mono Limits: 0 - 6 Mono: 0 [0%] Signed-off-by:
Pawel Moll <mail@pawelmoll.com> Signed-off-by:
Takashi Iwai <tiwai@suse.de> [bwh: Backported to 3.2: adjust filename] Signed-off-by:
-
Matt Fleming authored
commit fb834c7a upstream. commit 1de63d60 ("efi: Clear EFI_RUNTIME_SERVICES rather than EFI_BOOT by "noefi" boot parameter") attempted to make "noefi" true to its documentation and disable EFI runtime services to prevent the bricking bug described in commit e0094244 ("samsung-laptop: Disable on EFI hardware"). However, it's not possible to clear EFI_RUNTIME_SERVICES from an early param function because EFI_RUNTIME_SERVICES is set in efi_init() *after* parse_early_param(). This resulted in "noefi" effectively becoming a no-op and no longer providing users with a way to disable EFI, which is bad for those users that have buggy machines. Reported-by:
Walt Nelson Jr <walt0924@gmail.com> Cc: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com> Signed-off-by:
Matt Fleming <matt.fleming@intel.com> Link: http://lkml.kernel.org/r/1361392572-25657-1-git-send-email-matt@console-pimps.orgSigned-off-by:
-
Stefan Bader authored
commit 76eaca03 upstream. There is a loophole between Xen's current implementation of pv-spinlocks and the scheduler. This was triggerable through a testcase until v3.6 changed the TLB flushing code. The problem potentially is still there just not observable in the same way. What could happen was (is): 1. CPU n tries to schedule task x away and goes into a slow wait for the runq lock of CPU n-# (must be one with a lower number). 2. CPU n-#, while processing softirqs, tries to balance domains and goes into a slow wait for its own runq lock (for updating some records). Since this is a spin_lock_irqsave in softirq context, interrupts will be re-enabled for the duration of the poll_irq hypercall used by Xen. 3. Before the runq lock of CPU n-# is unlocked, CPU n-1 receives an interrupt (e.g. endio) and when processing the interrupt, tries to wake up task x. But that is in schedule and still on_cpu, so try_to_wake_up goes into a tight loop. 4. The runq lock of CPU n-# gets unlocked, but the message only gets sent to the first waiter, which is CPU n-# and that is busily stuck. 5. CPU n-# never returns from the nested interruption to take and release the lock because the scheduler uses a busy wait. And CPU n never finishes the task migration because the unlock notification only went to CPU n-#. To avoid this and since the unlocking code has no real sense of which waiter is best suited to grab the lock, just send the IPI to all of them. This causes the waiters to return from the hyper- call (those not interrupted at least) and do active spinlocking. BugLink: http://bugs.launchpad.net/bugs/1011792Acked-by:
Stefan Bader <stefan.bader@canonical.com> Signed-off-by:
-
Wei Liu authored
commit e7e44e44 upstream. Signed-off-by:
-
Daniel Vetter authored
commit 210561ff upstream. We already have the quirk entry for the mobile platform, but also reports on some desktop versions. So be paranoid and set it everywhere. References: http://www.mail-archive.com/dri-devel@lists.freedesktop.org/msg33138.html Cc: David Woodhouse <dwmw2@infradead.org> Cc: "Sankaran, Rajesh" <rajesh.sankaran@intel.com> Reported-and-tested-by:
Mihai Moldovan <ionic@ionic.de> Signed-off-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by:
-
