- 26 Feb, 2008 11 commits
-
-
Paul Moore authored
Upstream commit: e1770d97 As pointed out by Adrian Bunk, commit 45c950e0 ("fix memory leak in netlabel code") caused a double-free when security_netlbl_sid_to_secattr() fails. This patch fixes this by removing the netlbl_secattr_destroy() call from that function since we are already releasing the secattr memory in selinux_netlbl_sock_setsid(). Signed-off-by:
Paul Moore <paul.moore@hp.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stephen Hemminger authored
Upstream commit: 04f217ac If userspace passes a unknown match index into em_meta, then em_meta_change will return an error and the data for the match will not be set. This then causes an null pointer dereference when the cleanup is done in the error path via tcf_em_tree_destroy. Since the tree structure comes kzalloc, it is initialized to NULL. Discovered when testing a new version of tc command against an accidental older kernel. Signed-off-by:
Stephen Hemminger <shemminger@vyatta.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Shan Wei authored
Upstream commit: 16ca3f91 In strategy_allowed_congestion_control of the 2.6.24 kernel, when sysctl_string return 1 on success,it should call tcp_set_allowed_congestion_control to set the allowed congestion control.But, it don't. the sysctl_string return 1 on success, otherwise return negative, never return 0.The patch fix the problem. Signed-off-by:
Shan Wei <shanwei@cn.fujitsu.com> Acked-by:
Stephen Hemminger <shemminger@vyatta.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
James Bottomley authored
patch 366c246d in mainline. Some devices report medium error locations incorrectly. Add guards to make sure the reported bad lba is actually in the request that caused it. Additionally remove the large case statment for sector sizes and replace it with the proper u64 divisions. Tested-by:
Mike Snitzer <snitzer@gmail.com> Cc: Stable Tree <stable@kernel.org> Cc: Tony Battersby <tonyb@cybernetics.com> Signed-off-by:
James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Al Viro authored
patch 9c52fab2 in mainline. The MSSR constants didn't match the reality - bitfield declarations used to be correct (1000BT_FD - bit 11, 1000BT_HD - bit 10), but enum had them the other way round. Went unnoticed until the switch from the bitfields use to the explicit arithmetics and I hadn't caught that one when verifying correctness of change... Signed-off-by:
Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
David Chinner authored
patch 450790a2 in mainline. Several occurrences of oops in xfs_file_readdir() on ia32 have been reported since 2.6.24 was released. This is a regression introduced in 2.6.24 and is relatively easy to hit. The patch below fixes the problem. Signed-off-by:
Dave Chinner <dgc@sgi.com> Signed-off-by:
Lachlan McIlroy <lachlan@sgi.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Nishanth Aravamudan authored
patch a3d0c6aa in mainline. When I replaced hugetlb_dynamic_pool with nr_overcommit_hugepages I used proc_doulongvec_minmax() directly. However, hugetlb.c's locking rules require that all counter modifications occur under the hugetlb_lock. Add a callback into the hugetlb code similar to the one for nr_hugepages. Grab the lock around the manipulation of nr_overcommit_hugepages in proc_doulongvec_minmax(). Signed-off-by:
Nishanth Aravamudan <nacc@us.ibm.com> Acked-by:
Adam Litke <agl@us.ibm.com> Cc: David Gibson <david@gibson.dropbear.id.au> Cc: William Lee Irwin III <wli@holomorphy.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Ulisses Furquim authored
patch ac74c00e in mainline. As the IN_ONESHOT bit is never set when an event is sent we must check it in the watch's mask and not in the event's mask. Signed-off-by:
Ulisses Furquim <ulissesf@gmail.com> Reported-by:
"Clem Taylor" <clem.taylor@gmail.com> Tested-by:
"Clem Taylor" <clem.taylor@gmail.com> Cc: Amy Griffis <amy.griffis@hp.com> Cc: Robert Love <rlove@google.com> Cc: John McCutchan <ttb@tentacle.dhs.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Trond Myklebust authored
patch 5d47a356 in mainline. If the inode is flagged as having an invalid mapping, then we can't rely on the PageUptodate() flag. Ensure that we don't use the "anti-fragmentation" write optimisation in nfs_updatepage(), since that will cause NFS to write out areas of the page that are no longer guaranteed to be up to date. A potential corruption could occur in the following scenario: client 1 client 2 =============== =============== fd=open("f",O_CREAT|O_WRONLY,0644); write(fd,"fubar\n",6); // cache last page close(fd); fd=open("f",O_WRONLY|O_APPEND); write(fd,"foo\n",4); close(fd); fd=open("f",O_WRONLY|O_APPEND); write(fd,"bar\n",4); close(fd); ----- The bug may lead to the file "f" reading 'fubar\n\0\0\0\nbar\n' because client 2 does not update the cached page after re-opening the file for write. Instead it keeps it marked as PageUptodate() until someone calls invalidate_inode_pages2() (typically by calling read()). The bug was introduced by commit 44b11874 "NFS: Separate metadata and page cache revalidation mechanisms" Signed-off-by:
Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Jozsef Kadlecsik authored
[NETFILTER]: nf_conntrack_tcp: conntrack reopening fix [Upstream commits b2155e7f + d0c1fd7a ] TCP connection tracking in netfilter did not handle TCP reopening properly: active close was taken into account for one side only and not for any side, which is fixed now. The patch includes more comments to explain the logic how the different cases are handled. The bug was discovered by Jeff Chua. Signed-off-by:
Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> Signed-off-by:
Patrick McHardy <kaber@trash.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
David Miller authored
[SPARC/SPARC64]: Fix usage of .section .sched.text in assembler code. Upstream commit: c6d64c16 ld will generate an unique named section when assembler do not use "ax" but gcc does. Add the missing annotation. Signed-off-by:
Sam Ravnborg <sam@ravnborg.org> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
- 11 Feb, 2008 2 commits
-
-
Greg Kroah-Hartman authored
-
Bastian Blank authored
patch 712a30e6 in mainline. Commit 8811930d ("splice: missing user pointer access verification") added the proper access_ok() calls to copy_from_user_mmap_sem() which ensures we can copy the struct iovecs from userspace to the kernel. But we also must check whether we can access the actual memory region pointed to by the struct iovec to fix the access checks properly. Signed-off-by:
Bastian Blank <waldi@debian.org> Acked-by:
Oliver Pinter <oliver.pntr@gmail.com> Cc: Jens Axboe <jens.axboe@oracle.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
- 08 Feb, 2008 27 commits
-
-
Greg Kroah-Hartman authored
-
Jens Axboe authored
patch 8811930d in mainline. vmsplice_to_user() must always check the user pointer and length with access_ok() before copying. Likewise, for the slow path of copy_from_user_mmap_sem() we need to check that we may read from the user region. Signed-off-by:
Jens Axboe <jens.axboe@oracle.com> Cc: Wojciech Purczynski <cliph@research.coseinc.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Dave Airlie authored
(submitted upstream as 19a8f59a ) perhaps bonghits could turn on my bus-mastering because the drm certainly never bothered doing it before. Signed-off-by:
Dave Airlie <airlied@linux.ie> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Michael E Brown authored
patch 7d640c4a in mainline. This reverts commit 109f0e93 . The original patch breaks BIOS updates on all Dell machines. The path to the firmware file for the dell_rbu driver changes, which breaks all of the userspace tools which rely on it. Note that this patch re-introduces a problem with i2c name collision that was previously fixed by this patch. Signed-off-by:
Michael E Brown <michael_e_brown@dell.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Nick Piggin authored
patch 124d3b70 in mainline. Frederik Himpe reported an unkillable and un-straceable pan process. Zero length iovecs can go into an infinite loop in writev, because the iovec iterator does not always advance over them. The sequence required to trigger this is not trivial. I think it requires that a zero-length iovec be followed by a non-zero-length iovec which causes a pagefault in the atomic usercopy. This causes the writev code to drop back into single-segment copy mode, which then tries to copy the 0 bytes of the zero-length iovec; a zero length copy looks like a failure though, so it loops. Put a test into iov_iter_advance to catch zero-length iovecs. We could just put the test in the fallback path, but I feel it is more robust to skip over zero-length iovecs throughout the code (iovec iterator may be used in filesystems too, so it should be robust). Signed-off-by:
Nick Piggin <npiggin@suse.de> Signed-off-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Srivatsa Vaddagiri authored
patch 296825cb in mainline. The reason why we are getting better wakeup latencies for !FAIR_USER_SCHED is because of this snippet of code in place_entity(): if (!initial) { /* sleeps upto a single latency don't count. */ if (sched_feat(NEW_FAIR_SLEEPERS) && entity_is_task(se)) ^^^^^^^^^^^^^^^^^^ vruntime -= sysctl_sched_latency; /* ensure we never gain time by being placed backwards. */ vruntime = max_vruntime(se->vruntime, vruntime); } NEW_FAIR_SLEEPERS feature gives credit for sleeping only to tasks and not group-level entities. With the patch attached, I could see that wakeup latencies with FAIR_USER_SCHED are restored to the same level as !FAIR_USER_SCHED. Signed-off-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Peter Zijlstra authored
patch ef9884e6 in mainline. Michel Dänzr has bisected an interactivity problem with plus-reniced tasks back to this commit: 810e95cc is first bad commit commit 810e95cc Author: Peter Zijlstra <a.p.zijlstra@chello.nl> Date: Mon Oct 15 17:00:14 2007 +0200 sched: another wakeup_granularity fix unit mis-match: wakeup_gran was used against a vruntime fix this by assymetrically scaling the vtime of positive reniced tasks. Bisected-by:
Michel Dänzer <michel@tungstengraphics.com> Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Michael Buesch authored
(not in mainline, as it is not applicable.) We must reject new incompatible firmware early to avoid running into strange transmission failures. The current development tree supports newer firmware revisions. These revisions cause strange failures on the stable 2.6.24 kernel. Add a check to avoid confusing users a lot. Signed-off-by:
Michael Buesch <mb@bu3sch.de> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stephen Smalley authored
patch b1aa5301 in mainline. The proc net rewrite had a side effect on selinux, leading it to mislabel the /proc/net inodes, thereby leading to incorrect denials. Fix security_genfs_sid to ignore extra leading / characters in the path supplied by selinux_proc_get_sid since we now get "//net/..." rather than "/net/...". Signed-off-by:
Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by:
James Morris <jmorris@namei.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stefano Brivio authored
patch 8dd0100c in mainline. This fixes four resource leakages. In any error path we must deallocate the DMA frame slots we previously allocated by request_slot(). This is done by storing the ring pointers before doing any ring allocation and restoring the old pointers in case of an error. This patch by Michael Buesch has been ported to b43legacy. Cc: Michael Buesch <mb@bu3sch.de> Signed-off-by:
Stefano Brivio <stefano.brivio@polimi.it> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stefano Brivio authored
patch 9eca9a8e in mainline. We must drop any packets we are not able to encrypt. We must not send them unencrypted or with an all-zero-key (which basically is the same as unencrypted, from a security point of view). This might only trigger shortly after resume before mac80211 reassociated and reconfigured the keys. It is safe to drop these packets, as the association they belong to is not guaranteed anymore anyway. This is a security fix in the sense that it prevents information leakage. This patch by Michael Buesch has been ported to b43legacy. Cc: Michael Buesch <mb@bu3sch.de> Signed-off-by:
Stefano Brivio <stefano.brivio@polimi.it> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stefano Brivio authored
patch ada50731 in mainline. This patch makes suspend/resume work with the b43legacy driver. We must not overwrite the MAC addresses in the init function, as this would also overwrite the MAC on resume. With an all-zero MAC the device firmware is not able to ACK any received packets anymore. Fix this by moving the initializion stuff that must be done on init but not on resume to the start function. Also zero out filter_flags to make sure we don't have some flags from a previous instance for a tiny timeframe until mac80211 reconfigures them. This patch by Michael Buesch has been ported to b43legacy. Cc: Michael Buesch <mb@bu3sch.de> Signed-off-by:
Stefano Brivio <stefano.brivio@polimi.it> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stefano Brivio authored
patch 0cd67d48 in mainline. Fix the crash reported below, which seems to happen on bcm4306 rev. 2 devices only while using PIO: Oops: 0000 [#1] PREEMPT Modules linked in: b43(F) rfkill(F) led_class(F) input_polldev(F) arc4 b43legacy mac80211 cfg80211 i915 drm snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device ohci1394 ieee1394 ssb pcmcia snd_intel8x0m ehci_hcd uhci_hcd evdev Pid: 0, comm: swapper Tainted: GF (2.6.24st3 #2) EIP: 0060:[<f90f667b>] EFLAGS: 00010002 CPU: 0 EIP is at b43legacy_pio_handle_txstatus+0xbb/0x210 [b43legacy] EAX: 0000049b EBX: f11f8044 ECX: 00000001 EDX: 00000000 ESI: f1ff8000 EDI: 00000000 EBP: f11f8040 ESP: c04f4ef4 DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068 Process swapper (pid: 0, ti=c04f4000 task=c0488300 task.ti=c04b8000) Stack: f90f2788 c05009f0 c0500900 000010f7 f1053823 c04f4f24 dfb8e800 00000003 f1368000 00000007 00000296 f90f1975 00001000 010c0800 01000000 00000007 f90f6391 f11f8000 00000082 c04f4f4a 00000000 00004fd0 10f70000 8c061000 Call Trace: [<f90f2788>] b43legacy_debugfs_log_txstat+0x48/0xb0 [b43legacy] [<f90f1975>] b43legacy_handle_hwtxstatus+0x75/0x80 [b43legacy] [<f90f6391>] b43legacy_pio_rx+0x201/0x280 [b43legacy] [<f90e4fa3>] b43legacy_interrupt_tasklet+0x2e3/0x870 [b43legacy] [<c0123567>] tasklet_action+0x27/0x60 [<c01237b4>] __do_softirq+0x54/0xb0 [<c010686b>] do_softirq+0x7b/0xe0 [<c01457c0>] handle_level_irq+0x0/0x110 [<c01457c0>] handle_level_irq+0x0/0x110 [<c0123758>] irq_exit+0x38/0x40 [<c0106953>] do_IRQ+0x83/0xd0 [<c011812f>] __update_rq_clock+0x4f/0x180 [<c0104b4f>] common_interrupt+0x23/0x28 [<c011007b>] wakeup_code+0x7b/0xde [<c02b1039>] acpi_processor_idle+0x24a/0x3c9 [<c01025c7>] cpu_idle+0x47/0x80 [<c04b9ad5>] start_kernel+0x205/0x290 [<c04b9360>] unknown_bootoption+0x0/0x1f0 ======================= Code: 0f 00 00 81 fb ff 00 00 00 0f 87 36 01 00 00 8d 04 db 85 ff 8d 6c c6 40 8d 5d 04 0f 85 ef 00 00 00 fe 4e 0e 0f b7 46 0c 8b 53 04 <8b> 4a 50 29 c8 83 e8 52 66 89 46 0c 8b 54 24 14 80 7a 0b 00 74 EIP: [<f90f667b>] b43legacy_pio_handle_txstatus+0xbb/0x210 [b43legacy] SS:ESP 0068:c04f4ef4 Kernel panic - not syncing: Fatal exception in interrupt Signed-off-by:
Stefano Brivio <stefano.brivio@polimi.it> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Michael Buesch authored
patch 8dd0100c in mainline. This fixes four resource leakages. In any error path we must deallocate the DMA frame slots we previously allocated by request_slot(). This is done by storing the ring pointers before doing any ring allocation and restoring the old pointers in case of an error. Signed-off-by:
Michael Buesch <mb@bu3sch.de> Signed-off-by:
Stefano Brivio <stefano.brivio@polimi.it> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Michael Buesch authored
patch 09552ccd in mainline We must drop any packets we are not able to encrypt. We must not send them unencrypted or with an all-zero-key (which basically is the same as unencrypted, from a security point of view). This might only trigger shortly after resume before mac80211 reassociated and reconfigured the keys. It is safe to drop these packets, as the association they belong to is not guaranteed anymore anyway. This is a security fix in the sense that it prevents information leakage. Signed-off-by:
Michael Buesch <mb@bu3sch.de> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Michael Buesch authored
patch 7be1bb6b in mainline. This patch makes suspend/resume work with the b43 driver. We must not overwrite the MAC addresses in the init function, as this would also overwrite the MAC on resume. With an all-zero MAC the device firmware is not able to ACK any received packets anymore. Fix this by moving the initializion stuff that must be done on init but not on resume to the start function. Also zero out filter_flags to make sure we don't have some flags from a previous instance for a tiny timeframe until mac80211 reconfigures them. Signed-off-by:
Michael Buesch <mb@bu3sch.de> Signed-off-by:
John W. Linville <linville@tuxdriver.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stephen Hemminger authored
patch 82637e80 in mainline This patch disables config mode access after clearing PCI settings. Without this change WOL won't work on some BIOS's Signed-off-by:
Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stephen Hemminger authored
patch a7bffe72 in mainline. If the sky2 deadman timer forces a recovery, the multicast hash list is lost. Move the call to sky2_set_multicast to the end of sky2_up() so all paths that bring device up will restore multicast. Signed-off-by:
Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Sam Ravnborg authored
patch 3a900d89 in mainline The apm module were renamed to apm_32 during the merge of 32 and 64 bit x86 which is unfortunate. As apm is 32 bit specific we like to keep the _32 in the filename but the module should be named apm. Fix this in the Makefile. Reported-by:
"A.E.Lawrence" <lawrence_a_e@ntlworld.com> Signed-off-by:
Sam Ravnborg <sam@ravnborg.org> Cc: Cc: Ingo Molnar <mingo@elte.hu> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: "A.E.Lawrence" <lawrence_a_e@ntlworld.com> Signed-off-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Len Brown authored
These minor changes sync the latest ACPI blacklist into 2.6.24. The main benefit of this patch is to make any future changes easier to apply. The immediate benefit is one less dmesg line on Acer systems. Signed-off-by:
Len Brown <len.brown@intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Ian Abbott authored
This patch works around a problem in the fakephp driver when a process writing "0" to a "power" sysfs file to fake removal of a PCI device ends up deadlocking itself in the sysfs code. The patch was recently accepted into Linus' tree after the 2.6.24 release: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=5c796ae7a7ebe56967ed9b9963d7c16d733635ff Signed-off-by:
Ian Abbott <abbotti@mev.co.uk> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Oleg Nesterov authored
patch 8a459e44 in mainline. Fix ->vm_file accounting, mmap_region() may do do_munmap(). Signed-off-by:
Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by:
Miklos Szeredi <mszeredi@suse.cz> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Peter Zijlstra authored
patch 0ccf831c in mainline. On Sat, 2008-01-05 at 13:35 -0800, Davide Libenzi wrote: > I remember I talked with Arjan about this time ago. Basically, since 1) > you can drop an epoll fd inside another epoll fd 2) callback-based wakeups > are used, you can see a wake_up() from inside another wake_up(), but they > will never refer to the same lock instance. > Think about: > > dfd = socket(...); > efd1 = epoll_create(); > efd2 = epoll_create(); > epoll_ctl(efd1, EPOLL_CTL_ADD, dfd, ...); > epoll_ctl(efd2, EPOLL_CTL_ADD, efd1, ...); > > When a packet arrives to the device underneath "dfd", the net code will > issue a wake_up() on its poll wake list. Epoll (efd1) has installed a > callback wakeup entry on that queue, and the wake_up() performed by the > "dfd" net code will end up in ep_poll_callback(). At this point epoll > (efd1) notices that it may have some event ready, so it needs to wake up > the waiters on its poll wait list (efd2). So it calls ep_poll_safewake() > that ends up in another wake_up(), after having checked about the > recursion constraints. That are, no more than EP_MAX_POLLWAKE_NESTS, to > avoid stack blasting. Never hit the same queue, to avoid loops like: > > epoll_ctl(efd2, EPOLL_CTL_ADD, efd1, ...); > epoll_ctl(efd3, EPOLL_CTL_ADD, efd2, ...); > epoll_ctl(efd4, EPOLL_CTL_ADD, efd3, ...); > epoll_ctl(efd1, EPOLL_CTL_ADD, efd4, ...); > > The code "if (tncur->wq == wq || ..." prevents re-entering the same > queue/lock. Since the epoll code is very careful to not nest same instance locks allow the recursion. Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Tested-by:
Stefan Richter <stefanr@s5r6.in-berlin.de> Acked-by:
Davide Libenzi <davidel@xmailserver.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Ayaz Abdulla authored
patch 2b912130 in mainline. This patch is a critical fix for MCP77 and MCP79 devices. The feature flags were missing the define for correct mac address (DEV_HAS_CORRECT_MACADDR). Signed-off-by:
Ayaz Abdulla <aabdulla@nvidia.com> Signed-off-by:
Jeff Garzik <jeff@garzik.org> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Stefan Bader authored
The device setup did miss to initialize the num_interrupt_out field, thus failing to successfully complete the probe function. Signed-off-by:
Stefan Bader <stefan.bader@canonical.com> Cc: Alan Cox <alan@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Greg Kroah-Hartman authored
Some crazy devices in the wild have a vendor id of 0x0000. If we try to add a module alias with this id, we just can't do it due to a check in the file2alias.c file. Change the test to verify that both the vendor and product ids are 0x0000 to show a real "blank" module alias. Note, the module-init-tools package also needs to be changed to properly generate the depmod tables. Cc: Janusz <janumix@poczta.fm> Cc: Jon Masters <jcm@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
Jan Andersson authored
usbtest did not swap the received status information when checking for a non-zero value and failed to discover halted endpoints on big endian systems. Signed-off-by:
Jan Andersson <jan@gaisler.com> Acked-by:
David Brownell <david-b@pacbell.net> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-