- 14 Jun, 2007 3 commits
-
-
David S. Miller authored
-
Herbert Xu authored
The recent patch that added ipv6_hwtype is broken on tuntap tunnels. Indeed, it's broken on any device that does not pass the ipv6_hwtype test. The reason is that the original test only applies to autoconfiguration, not IPv6 support. IPv6 support is allowed on any device. In fact, even with the ipv6_hwtype patch applied you can still add IPv6 addresses to any interface that doesn't pass thw ipv6_hwtype test provided that they have a sufficiently large MTU. This is a serious problem because come deregistration time these devices won't be cleaned up properly. I've gone back and looked at the rationale for the patch. It appears that the real problem is that we were creating IPv6 devices even if the MTU was too small. So here's a patch which fixes that and reverts the ipv6_hwtype stuff. Thanks to Kanru Chen for reporting this issue. Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Ilpo Järvinen authored
This flaw does not affect any behavior (currently). Signed-off-by:
Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by:
-
- 13 Jun, 2007 7 commits
-
-
Vlad Yasevich authored
Right now, when we receive a mtu estimate smaller then minim threshold in the ICMP message, we disable the path mtu discovery on the transport. This leads to the never increasing sctp fragmentation point even when the real path mtu has increased. Signed-off-by:Vlad Yasevich <vladislav.yasevich@hp.com>
-
Vlad Yasevich authored
Currently, if the socket is owned by the user, we drop the ICMP message. As a result SCTP forgets that path MTU changed and never adjusting it's estimate. This causes all subsequent packets to be fragmented. With this patch, we'll flag the association that it needs to udpate it's estimate based on the already updated routing information. Signed-off-by:
Sridhar Samudrala <sri@us.ibm.com>
-
Vlad Yasevich authored
Introduce new function sctp_transport_update_pmtu that updates the transports and destination caches view of the path mtu. Signed-off-by:
-
Vlad Yasevich authored
If the copy_to_user or copy_user calls fail in sctp_getsockopt_local_addrs(), the function should free locally allocated storage before returning error. Spotted by Coverity. Signed-off-by:
-
Vlad Yasevich authored
Allow sctp_bindx() to accept multiple address with unspecified port. In this case, all addresses inherit the first bound port. We still catch full mis-matches. Signed-off-by:
-
Vlad Yasevich authored
During peeloff of AF_INET6 socket, the inet6_sk(sk)->daddr wasn't set correctly since the code was assuming IPv4 only. Now we use a correct call to set the destination address. Signed-off-by:
-
David S. Miller authored
Because of the current default of 100, Cubic and BIC perform very poorly compared to standard Reno. In the worst case, this change makes Cubic and BIC as aggressive as Reno. So this change should be very safe. Signed-off-by:
-
- 12 Jun, 2007 5 commits
-
-
Ilpo Järvinen authored
Without FRTO, the tcp_try_to_open is never called with lost_out > 0 (see tcp_time_to_recover). However, when FRTO is enabled, the !tp->lost condition is not used until end of FRTO because that way TCP avoids premature entry to fast recovery during FRTO. Signed-off-by:
-
David S. Miller authored
Signed-off-by: -
Konstantin Sharlaimov authored
Prevent mppe_decompress() from generating "osize too small" errors when checking for output buffer size. When receiving a packet of mru size the output buffer for decrypted data is 1 byte too small since mppe_decompress() tries to account for possible PFC, however later in code it is assumed no PFC. Adjusting the check prevented these errors from occurring. Signed-off-by:
Konstantin Sharlaimov <konstantin.sharlaimov@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
David S. Miller authored
Merge branch 'mac80211-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6
-
Mattias Nissler authored
mac80211 stops the tx queues during scans. This is wrong with respect to the master deivce tx queue, since stopping it prevents any probes from being sent during the scan. Instead, they accumulate in the queue and are only sent after the scan is finished, which is obviously wrong. Signed-off-by:
Mattias Nissler <mattias.nissler@gmx.de> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
- 11 Jun, 2007 2 commits
-
-
Johannes Berg authored
This patch fixes a typo in mac80211's debugfs.c. Signed-off-by:
Johannes Berg <johannes@sipsolutions.net> Signed-off-by:
-
David Lamparter authored
Fix signedness mixup making mac addresses show up strangely (like 00:11:22:33:44:ffffffaa) in /sys/class/ieee80211/*/macaddress. Signed-off-by:
David Lamparter <equinox@diac24.net> Acked-by:
-
- 09 Jun, 2007 23 commits
-
-
G. Liakhovetski authored
Jean II was right: you have to re-charge the final timer when resending rejected frames. Otherwise it triggers at a wrong time and can break the currently running communication. Reproducible under rt-preempt. Signed-off-by:
G. Liakhovetski <gl@dsa-ac.de> Signed-off-by:
Samuel Ortiz <samuel@sortiz.org> Signed-off-by:
-
G. Liakhovetski authored
From: G. Liakhovetski <gl@dsa-ac.de> We need to switch to NRM _before_ sending the final packet otherwise we might hit a race condition where we get the first packet from the peer while we're still in LAP_XMIT_P. Signed-off-by:
-
Linus Torvalds authored
* 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: [CIPSO]: Fix several unaligned kernel accesses in the CIPSO engine. [NetLabel]: consolidate the struct socket/sock handling to just struct sock [IPV4]: Do not remove idev when addresses are cleared
-
Linus Torvalds authored
* 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6: [SPARC64]: Handle PCI bridges without 'ranges' property. [SPARC64]: Include <linux/rwsem.h> instead of <asm/rwsem.h>.
-
Linus Torvalds authored
* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/usb-2.6: OHCI: Fix machine check in ohci_hub_status_data USB: Fix up bogus bInterval values in endpoint descriptors USB: cxacru: ignore error trying to start ADSL in atm_start USB: cxacru: create sysfs attributes in atm_start instead of bind USB: cxacru: add Documentation file USB: UNUSUAL_DEV: Sync up some reported devices from Ubuntu USB: usb gadgets avoid le{16,32}_to_cpup() usblp: Don't let suspend to kill ->used USB: set default y for CONFIG_USB_DEVICE_CLASS -
Linus Torvalds authored
* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/driver-2.6: firmware: remove orphaned Email kobject: use the proper printk level for kobject error Driver core: kill unused code Driver core: keep PHYSDEV for old struct class_device update Documentation/driver-model/platform.txt
-
Robert P. J. Day authored
Prevent <linux/console_struct.h> from being included more than once, otherwise you get a redefinition error if you happen to include <linux/vt_kern.h> first. Signed-off-by:
Robert P. J. Day <rpjday@mindspring.com> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Andy Whitcroft authored
This version brings a some new tests, and a host of changes to fix false positives, of particular note: - check for and report #if 0 - extend checking of line lengths and spacing for .pl, .sh etc - extends the pointer type checks to multiple levels - updates printk handling to track newlines - adds a wrapped patch detector - drops the leading component of the filenames - extends switch indent handling to switch statmentes rooted in the context - adds foo * bar single pointer checks This version of checkpatch.pl can be found at the following URL: http://www.shadowen.org/~apw/public/checkpatch/checkpatch.pl-0.04 Full Changelog: Andy Whitcroft (16): allow checking line lengths and spacing on other source files clean up that whitespace sanitise the input line standardising the content of quotes clean up pointer type * and space checks fix up the sanitiser so it maintains the line length apply the printk facility checks only to the first printk in a set switch/case indent checks may anchor in the context add a wrapped patch detector put the #ifdef in C file checks on ice asm volatile is acceptable check for and report #if 0 drop the leading component of the filename as patches are -p1 use the original line when reporting operator errors correct spelling of Joel's name Version: 0.04 add support for struct foo * bar checks Geert Uytterhoeven (1): Fix checkpatch.pl name in usage template Randy Dunlap (1): checkpatch: produce fewer lines of output Signed-off-by:Andy Whitcroft <apw@shadowen.org> Cc: Randy Dunlap <rdunlap@xenotime.net> Cc: Joel Schopp <jschopp@austin.ibm.com> Signed-off-by:
-
Randy Dunlap authored
Add a prefix string parameter. Callers are responsible for any string length/alignment that they want to see in the output. I.e., callers should pad strings to achieve alignment if they want that. Add rowsize parameter. This is the number of raw data bytes to be printed per line. Must be 16 or 32. Add a groupsize parameter. This allows callers to dump values as 1-byte, 2-byte, 4-byte, or 8-byte numbers. Default is 1-byte numbers. If the total length is not an even multiple of groupsize, 1-byte numbers are printed. Add an "ascii" output parameter. This causes ASCII data output following the hex data output. Clean up some doc examples. Align the ASCII output on all lines that are produced by one call. Add a new interface, print_hex_dump_bytes(), that is a shortcut to print_hex_dump(), using default parameter values to print 16 bytes in byte-size chunks of hex + ASCII output, using printk level KERN_DEBUG. Signed-off-by:
Randy Dunlap <randy.dunlap@oracle.com> Cc: Christoph Lameter <clameter@sgi.com> Signed-off-by:
-
Alan Cox authored
The UFS entry was misformatted The NEC V850 links are all broken The Berkshire watchdog links are all broken Signed-off-by:
Alan Cox <alan@redhat.com> Signed-off-by:
-
Randy Dunlap authored
Produce one less line of output per flagged incident. Change this: use tabs not spaces PATCH: /home/rddunlap/arcmsr1200014.patch4:756: FILE: b/drivers/scsi/arcmsr/arcmsr_hba.c:1843: + return PCI_ERS_RESULT_NEED_RESET;$ to this: use tabs not spaces #756: FILE: b/drivers/scsi/arcmsr/arcmsr_hba.c:1843: + return PCI_ERS_RESULT_NEED_RESET;$ Cc: Andy Whitcroft <apw@shadowen.org> Signed-off-by:
-
Jeff Mahoney authored
This patch changes MAINTAINERS to reflect the new location of the reiserfs development mailing list. The old list forwards to the new one. Signed-off-by:
Jeff Mahoney <jeffm@suse.com> Signed-off-by:
-
Andrew Morton authored
We don't want to see this: > BUG: using smp_processor_id() in preemptible [00000001] code: bash/3857 > caller is oops_begin+0xb/0x6f > > Call Trace: > [<ffffffff8020ab4d>] show_trace+0x34/0x4f > [<ffffffff8020ab7a>] dump_stack+0x12/0x17 > [<ffffffff8030d92d>] debug_smp_processor_id+0xad/0xbc > [<ffffffff8042388f>] oops_begin+0xb/0x6f > [<ffffffff8042520b>] do_page_fault+0x66a/0x7c0 > [<ffffffff804234bd>] error_exit+0x0/0x84 > coming out when the kernel is trying to oops. Cc: Andi Kleen <ak@suse.de> Signed-off-by:
-
Bob Picco authored
We aren't sampling for holes in memory. Thus we encounter a section hole with empty section map pointer for SPARSEMEM and OOPs for show_mem. This issue has been seen in 2.6.21, current git and current mm. The patch below is for mainline and mm. It was boot tested for SPARSEMEM, current VMEMMAP of Andy's in mm ml and DISCONTIGMEM. A slightly different patch will be posted to stable for 2.6.21. Previous to commit f0a5a58a memory_present was called for node_start_pfn to node_end_pfn. This would cover the hole(s) with reserved pages and valid sections. Most SPARSEMEM supported arches do a pfn_valid check in show_mem before computing the page structure address. This issue was brought to my attention on IRC by Arnaldo Carvalho de Melo. Thanks to Arnaldo for testing. Signed-off-by:
Bob Picco <bob.picco@hp.com> Cc: Chuck Ebbert <cebbert@redhat.com> Cc: Andi Kleen <ak@suse.de> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Acked-by:
-
Alexey Kuznetsov authored
1. New entries can be added to tsk->pi_state_list after task completed exit_pi_state_list(). The result is memory leakage and deadlocks. 2. handle_mm_fault() is called under spinlock. The result is obvious. 3. results in self-inflicted deadlock inside glibc. Sometimes futex_lock_pi returns -ESRCH, when it is not expected and glibc enters to for(;;) sleep() to simulate deadlock. This problem is quite obvious and I think the patch is right. Though it looks like each "if" in futex_lock_pi() got some stupid special case "else if". :-) 4. sometimes futex_lock_pi() returns -EDEADLK, when nobody has the lock. The reason is also obvious (see comment in the patch), but correct fix is far beyond my comprehension. I guess someone already saw this, the chunk: if (rt_mutex_trylock(&q.pi_state->pi_mutex)) ret = 0; is obviously from the same opera. But it does not work, because the rtmutex is really taken at this point: wake_futex_pi() of previous owner reassigned it to us. My fix works. But it looks very stupid. I would think about removal of shift of ownership in wake_futex_pi() and making all the work in context of process taking lock. From: Thomas Gleixner <tglx@linutronix.de> Fix 1) Avoid the tasklist lock variant of the exit race fix by adding an additional state transition to the exit code. This fixes also the issue, when a task with recursive segfaults is not able to release the futexes. Fix 2) Cleanup the lookup_pi_state() failure path and solve the -ESRCH problem finally. Fix 3) Solve the fixup_pi_state_owner() problem which needs to do the fixup in the lock protected section by using the in_atomic userspace access functions. This removes also the ugly lock drop / unqueue inside of fixup_pi_state() Fix 4) Fix a stale lock in the error path of futex_wake_pi() Added some error checks for verification. The -EDEADLK problem is solved by the rtmutex fixups. Signed-off-by:Thomas Gleixner <tglx@linutronix.de> Acked-by:
Ingo Molnar <mingo@elte.hu> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ulrich Drepper <drepper@redhat.com> Cc: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by:
-
Thomas Gleixner authored
Alexey Kuznetsov found some problems in the pi-futex code. One of the root causes is: When a wakeup happens, we do not to stop the chain walk so we follow a not longer relevant locking chain. Drop out when this happens. Signed-off-by:
-
Thomas Gleixner authored
Alexey Kuznetsov found some problems in the pi-futex code. The major problem is a stale return value in rt_mutex_slowlock(): When the pi chain walk returns -EDEADLK, but the waiter was woken up during the phases where the locks were dropped, the rtmutex could be acquired, but due to the stale return value -EDEADLK returned to the caller. Reset the return value in the retry path. Signed-off-by:
-
Tejun Heo authored
sata_promise uses two different command modes - packet and TF. Packet mode is intelligent low-overhead mode while TF is the same old taskfile interface. As with other advanced interface (ahci/sil24), ATA_TFLAG_POLLING has no effect in packet mode. However, PIO commands are issued using TF interface in polling mode, so pdc_interrupt() considers interrupts spurious if ATA_TFLAG_POLLING is set. This is broken for polling NODATA commands because command is issued using packet mode but the interrupt handler ignores it due to ATA_TFLAG_POLLING. Fix pdc_qc_issue_prot() such that ATA/ATAPI NODATA commands are issued using TF interface if ATA_TFLAG_POLLING is set. This patch fixes detection failure introduced by polling SETXFERMODE. Signed-off-by:
Tejun Heo <htejun@gmail.com> Acked-by:
Mikael Pettersson <mikpe@it.uu.se> Acked-by:
Jeff Garzik <jeff@garzik.org> Signed-off-by:
-
Randy Dunlap authored
__exit function is used by both init and exit routines, so it cannot be marked __init. (from allyesconfig) WARNING: drivers/built-in.o(.text+0x9b83cf): Section mismatch: reference to .exit.text: (between 'divasfunc_exit' and 'didd_callback') Signed-off-by:
Armin Schindler <armin@melware.de> Signed-off-by:
-
Jeff Dike authored
Include linux/kernel.h wherever simple_strtoul is used. This kills a compile warning in stderr_console.c and potential ones in the other files. This also fixes a bunch of style violations in exitcode.c. Signed-off-by:
Jeff Dike <jdike@linux.intel.com> Signed-off-by:
-
Jiri Slaby authored
do not return 0 in one case and return proper values in other 2. Signed-off-by:
Jiri Slaby <jirislaby@gmail.com> Signed-off-by:
-
Jiri Slaby authored
this causes oops, because pci probe function calls tty_register_device for each device found. Thanks to Ingo. Signed-off-by:
-
Jiri Slaby authored
Since it's not neccesary to have MAX_PANELS on the card, don't fail to let users use this card even in this case. Stop the testing for loop instead. Thanks to Ingo. Signed-off-by:
-
