- 06 Jun, 2014 18 commits
-
-
Corey Minyard authored
commit eb6d78ec upstream. The OBF timer in KCS was not reset in one situation when error recovery was started, resulting in an immediate timeout. Reported-by:
Bodo Stroesser <bstroesser@ts.fujitsu.com> Signed-off-by:
Corey Minyard <cminyard@mvista.com> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
Jiri Slaby <jslaby@suse.cz>
-
Jiri Bohac authored
commit 98a01e77 upstream. On architectures with sizeof(int) < sizeof (long), the computation of mask inside apply_slack() can be undefined if the computed bit is > 32. E.g. with: expires = 0xffffe6f5 and slack = 25, we get: expires_limit = 0x20000000e bit = 33 mask = (1 << 33) - 1 /* undefined */ On x86, mask becomes 1 and and the slack is not applied properly. On s390, mask is -1, expires is set to 0 and the timer fires immediately. Use 1UL << bit to solve that issue. Suggested-by:
Deborah Townsend <dstownse@us.ibm.com> Signed-off-by:
Jiri Bohac <jbohac@suse.cz> Link: http://lkml.kernel.org/r/20140418152310.GA13654@midget.suse.czSigned-off-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Stephen Warren authored
commit 22bbd5d9 upstream. BIT_WORD() truncates rather than rounds, so the loops in syncpt_thresh_isr() and _host1x_intr_disable_all_syncpt_intrs() use <= rather than < in an attempt to process the correct number of registers when rounding of the conversion of count of bits to count of words is necessary. However, when rounding isn't necessary because the value is already a multiple of the divisor (as is the case for all values of nb_pts the code actually sees), this causes one too many registers to be processed. Solve this by using and explicit DIV_ROUND_UP() call, rather than BIT_WORD(), and comparing with < rather than <=. Fixes: 7ede0b0b ("gpu: host1x: Add syncpoint wait and interrupts") Signed-off-by:
Stephen Warren <swarren@nvidia.com> Acked-By:
Terje Bergstrom <tbergstrom@nvidia.com> Signed-off-by:
Thierry Reding <treding@nvidia.com> Signed-off-by:
-
Loic Poulain authored
commit b08c9c31 upstream. On transmit-hold-register empty, serial8250_tx_chars should be called only if we don't use DMA. DMA has its own tx cycle. Signed-off-by:
Loic Poulain <loic.poulain@intel.com> Reviewed-by:
Heikki Krogerus <heikki.krogerus@linux.intel.com> Signed-off-by:
-
Loic Poulain authored
commit f8fd1b03 upstream. __dma_tx_complete is not protected against concurrent call of serial8250_tx_dma. it can lead to circular tail index corruption or parallel call of serial_tx_dma on the same data portion. This patch fixes this issue by holding the port lock. Signed-off-by:
-
Linus Torvalds authored
commit 1b17844b upstream. fixup_user_fault() is used by the futex code when the direct user access fails, and the futex code wants it to either map in the page in a usable form or return an error. It relied on handle_mm_fault() to map the page, and correctly checked the error return from that, but while that does map the page, it doesn't actually guarantee that the page will be mapped with sufficient permissions to be then accessed. So do the appropriate tests of the vma access rights by hand. [ Side note: arguably handle_mm_fault() could just do that itself, but we have traditionally done it in the caller, because some callers - notably get_user_pages() - have been able to access pages even when they are mapped with PROT_NONE. Maybe we should re-visit that design decision, but in the meantime this is the minimal patch. ] Found by Dave Jones running his trinity tool. Reported-by:
Dave Jones <davej@redhat.com> Acked-by:
Hugh Dickins <hughd@google.com> Signed-off-by:
-
Eric Dumazet authored
commit 404ca80e upstream. A va_list needs to be copied in case it needs to be used twice. Thanks to Hugh for debugging this issue, leading to various panics. Tested: lpq84:~# echo "|/foobar12345 %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h" >/proc/sys/kernel/core_pattern 'produce_core' is simply : main() { *(int *)0 = 1;} lpq84:~# ./produce_core Segmentation fault (core dumped) lpq84:~# dmesg | tail -1 [ 614.352947] Core dump to |/foobar12345 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 lpq84 (null) pipe failed Notice the last argument was replaced by a NULL (we were lucky enough to not crash, but do not try this on your production machine !) After fix : lpq83:~# echo "|/foobar12345 %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h %h" >/proc/sys/kernel/core_pattern lpq83:~# ./produce_core Segmentation fault lpq83:~# dmesg | tail -1 [ 740.800441] Core dump to |/foobar12345 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 lpq83 pipe failed Fixes: 5fe9d8ca ("coredump: cn_vprintf() has no reason to call vsnprintf() twice") Signed-off-by:
Eric Dumazet <edumazet@google.com> Diagnosed-by:
Oleg Nesterov <oleg@redhat.com> Cc: Neil Horman <nhorman@tuxdriver.com> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Bartlomiej Zolnierkiewicz authored
commit 27aa64b9 upstream. Add missing clk_put() call to ata_host_activate() failure path. Sergei says, "Hm, I have once fixed that (see that *if* (!ret)) but looks like a later commit 477c87e9 (ARM: at91/pata: use gpio_is_valid to check the gpio) broke it again. :-( Would be good if the changelog did mention that..." Cc: Andrew Victor <linux@maxim.org.za> Cc: Nicolas Ferre <nicolas.ferre@atmel.com> Cc: Jean-Christophe Plagniol-Villard <plagnioj@jcrosoft.com> Cc: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by:
Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by:
Tejun Heo <tj@kernel.org> Signed-off-by:
-
Krzysztof Kozlowski authored
commit 8db6e510 upstream. After hotplugging CPU1 the first call of interrupt handler for CPU1 oneshot timer was called on CPU0 because it fired before setting IRQ affinity. Affected are SoCs where Multi Core Timer interrupts are shared (SPI), e.g. Exynos 4210. During setup of the MCT timers the clock event device should be registered after setting the affinity for interrupt. This will prevent starting the timer too early. Signed-off-by:
Krzysztof Kozlowski <k.kozlowski@samsung.com> Signed-off-by:
-
Thomas Gleixner authored
commit 30ccf03b upstream. The starting cpu is not yet in the online mask so irq_set_affinity() fails which results in per cpu timers for this cpu ending up on some other online cpu, ususally cpu 0. Use irq_force_affinity() which disables the online mask check and makes things work. Signed-off-by:
-
Thomas Gleixner authored
commit 01f8fa4f upstream. The current implementation of irq_set_affinity() refuses rightfully to route an interrupt to an offline cpu. But there is a special case, where this is actually desired. Some of the ARM SoCs have per cpu timers which require setting the affinity during cpu startup where the cpu is not yet in the online mask. If we can't do that, then the local timer interrupt for the about to become online cpu is routed to some random online cpu. The developers of the affected machines tried to work around that issue, but that results in a massive mess in that timer code. We have a yet unused argument in the set_affinity callbacks of the irq chips, which I added back then for a similar reason. It was never required so it got not used. But I'm happy that I never removed it. That allows us to implement a sane handling of the above scenario. So the affected SoC drivers can add the required force handling to their interrupt chip, switch the timer code to irq_force_affinity() and things just work. This does not affect any existing user of irq_set_affinity(). Tagged for stable to allow a simple fix of the affected SoC clock event drivers. Reported-and-tested-by:
-
Thomas Gleixner authored
commit ffde1de6 upstream. To support the affinity setting of per cpu timers in the early startup of a not yet online cpu, implement the force logic, which disables the cpu online check. Tagged for stable to allow a simple fix of the affected SoC clock event drivers. Signed-off-by:
-
Steven Rostedt (Red Hat) authored
commit a949ae56 upstream. A race exists between module loading and enabling of function tracer. CPU 1 CPU 2 ----- ----- load_module() module->state = MODULE_STATE_COMING register_ftrace_function() mutex_lock(&ftrace_lock); ftrace_startup() update_ftrace_function(); ftrace_arch_code_modify_prepare() set_all_module_text_rw(); <enables-ftrace> ftrace_arch_code_modify_post_process() set_all_module_text_ro(); [ here all module text is set to RO, including the module that is loading!! ] blocking_notifier_call_chain(MODULE_STATE_COMING); ftrace_init_module() [ tries to modify code, but it's RO, and fails! ftrace_bug() is called] When this race happens, ftrace_bug() will produces a nasty warning and all of the function tracing features will be disabled until reboot. The simple solution is to treate module load the same way the core kernel is treated at boot. To hardcode the ftrace function modification of converting calls to mcount into nops. This is done in init/main.c there's no reason it could not be done in load_module(). This gives a better control of the changes and doesn't tie the state of the module to its notifiers as much. Ftrace is special, it needs to be treated as such. The reason this would work, is that the ftrace_module_init() would be called while the module is in MODULE_STATE_UNFORMED, which is ignored by the set_all_module_text_ro() call. Link: http://lkml.kernel.org/r/1395637826-3312-1-git-send-email-indou.takao@jp.fujitsu.comReported-by:
Takao Indoh <indou.takao@jp.fujitsu.com> Acked-by:
Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
Steven Rostedt <rostedt@goodmis.org> Signed-off-by:
-
Leif Lindholm authored
commit dfc44f80 upstream. A few platforms lack a 'device_type = "memory"' for their memory nodes, relying on an old ppc quirk in order to discover its memory. Add the missing data so that all parsing code can find memory nodes correctly. Signed-off-by:
Leif Lindholm <leif.lindholm@linaro.org> Cc: linux-mips@linux-mips.org Cc: devicetree@vger.kernel.org Cc: Mark Rutland <mark.rutland@arm.com> Acked-by:
John Crispin <blogic@openwrt.org> Signed-off-by:
Grant Likely <grant.likely@linaro.org> Signed-off-by:
-
Thomas Gleixner authored
commit f0d71b3d upstream. We happily allow userspace to declare a random kernel thread to be the owner of a user space PI futex. Found while analysing the fallout of Dave Jones syscall fuzzer. We also should validate the thread group for private futexes and find some fast way to validate whether the "alleged" owner has RW access on the file which backs the SHM, but that's a separate issue. Signed-off-by:
-
Thomas Gleixner authored
commit 866293ee upstream. Dave Jones trinity syscall fuzzer exposed an issue in the deadlock detection code of rtmutex: http://lkml.kernel.org/r/20140429151655.GA14277@redhat.com That underlying issue has been fixed with a patch to the rtmutex code, but the futex code must not call into rtmutex in that case because - it can detect that issue early - it avoids a different and more complex fixup for backing out If the user space variable got manipulated to 0x80000000 which means no lock holder, but the waiters bit set and an active pi_state in the kernel is found we can figure out the recursive locking issue by looking at the pi_state owner. If that is the current task, then we can safely return -EDEADLK. The check should have been added in commit 59fa6245 (futex: Handle futex_pi OWNER_DIED take over correctly) already, but I did not see the above issue caused by user space manipulation back then. Signed-off-by:
-
Aaron Lu authored
commit 9efa5e50 upstream. When testing if the firmware's initial value is valid, we should use the corrected level value instead of the raw value returned from firmware. Signed-off-by:
Aaron Lu <aaron.lu@intel.com> Signed-off-by:
Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by:
-
Peter De Schrijver authored
commit 8e9cc80a upstream. Use pll_ref instead of pll_re_vco as the pll_e parent on Tegra114. Also add a 12Mhz pll_ref table entry for pll_e for Tegra114. This prevents the system from crashing at bootup because of an unsupported pll_re_vco rate. Signed-off-by:
Peter De Schrijver <pdeschrijver@nvidia.com> Signed-off-by:
-
- 04 Jun, 2014 8 commits
-
-
Bjørn Mork authored
commit 4d7c0136 upstream. Dan writes: "The Dell drivers use the same configuration for PIDs: 81A2: Dell Wireless 5806 Gobi(TM) 4G LTE Mobile Broadband Card 81A3: Dell Wireless 5570 HSPA+ (42Mbps) Mobile Broadband Card 81A4: Dell Wireless 5570e HSPA+ (42Mbps) Mobile Broadband Card 81A8: Dell Wireless 5808 Gobi(TM) 4G LTE Mobile Broadband Card 81A9: Dell Wireless 5808e Gobi(TM) 4G LTE Mobile Broadband Card These devices are all clearly Sierra devices, but are also definitely Gobi-based. The A8 might be the MC7700/7710 and A9 is likely a MC7750. >From DellGobi5kSetup.exe from the Dell drivers: usbif0: serial/firmware loader? usbif2: nmea usbif3: modem/ppp usbif8: net/QMI" Cc: <stable@vger.kernel.org> Reported-by:
AceLan Kao <acelan.kao@canonical.com> Reported-by:
Dan Williams <dcbw@redhat.com> Signed-off-by:
Bjørn Mork <bjorn@mork.no> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Sergey Popovich authored
commit a8951d58 upstream. Dst is released one line before we access it again with dst->error. Fixes: 58e35d14 netfilter: ipv6: propagate routing errors from ip6_route_me_harder() Signed-off-by:
Sergey Popovich <popovich_sergei@mail.ru> Signed-off-by:
Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by:
-
Johannes Berg authored
commit b4b177a5 upstream. Jouni reported that if a remain-on-channel was active on the same channel as the current operating channel, then the ROC would start, but any frames transmitted using mgmt-tx on the same channel would get delayed until after the ROC. The reason for this is that the ROC starts, but doesn't have any handling for "remain on the same channel", so it stops the interface queues. The later mgmt-tx then puts the frame on the interface queues (since it's on the current operating channel) and thus they get delayed until after the ROC. To fix this, add some logic to handle remaining on the same channel specially and not stop the queues etc. in this case. This not only fixes the bug but also improves behaviour in this case as data frames etc. can continue to flow. Cc: stable@vger.kernel.org Reported-by:
Jouni Malinen <j@w1.fi> Tested-by:
Johannes Berg <johannes.berg@intel.com> Signed-off-by:
-
David Rientjes authored
commit d49ad935 upstream. When two threads have the same badness score, it's preferable to kill the thread group leader so that the actual process name is printed to the kernel log rather than the thread group name which may be shared amongst several processes. This was the behavior when select_bad_process() used to do for_each_process(), but it now iterates threads instead and leads to ambiguity. Signed-off-by:
David Rientjes <rientjes@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Greg Thelen <gthelen@google.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
-
Oleg Nesterov authored
commit 4d4048be upstream. find_lock_task_mm() expects it is called under rcu or tasklist lock, but it seems that at least oom_unkillable_task()->task_in_mem_cgroup() and mem_cgroup_out_of_memory()->oom_badness() can call it lockless. Perhaps we could fix the callers, but this patch simply adds rcu lock into find_lock_task_mm(). This also allows to simplify a bit one of its callers, oom_kill_process(). Signed-off-by:
Michal Hocko <mhocko@suse.cz> Cc: "Tu, Xiaobing" <xiaobing.tu@intel.com> Acked-by:
-
Oleg Nesterov authored
commit ad962441 upstream. At least out_of_memory() calls has_intersects_mems_allowed() without even rcu_read_lock(), this is obviously buggy. Add the necessary rcu_read_lock(). This means that we can not simply return from the loop, we need "bool ret" and "break". While at it, swap the names of task_struct's (the argument and the local). This cleans up the code a little bit and avoids the unnecessary initialization. Signed-off-by:
Sergey Dyasly <dserrg@gmail.com> Tested-by:
Sameer Nanda <snanda@chromium.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Mandeep Singh Baines <msb@chromium.org> Cc: "Ma, Xindong" <xindong.ma@intel.com> Reviewed-by:
-
Oleg Nesterov authored
commit 1da4db0c upstream. Change oom_kill.c to use for_each_thread() rather than the racy while_each_thread() which can loop forever if we race with exit. Note also that most users were buggy even if while_each_thread() was fine, the task can exit even _before_ rcu_read_lock(). Fortunately the new for_each_thread() only requires the stable task_struct, so this change fixes both problems. Signed-off-by:
-
Oleg Nesterov authored
commit 0c740d0a upstream. while_each_thread() and next_thread() should die, almost every lockless usage is wrong. 1. Unless g == current, the lockless while_each_thread() is not safe. while_each_thread(g, t) can loop forever if g exits, next_thread() can't reach the unhashed thread in this case. Note that this can happen even if g is the group leader, it can exec. 2. Even if while_each_thread() itself was correct, people often use it wrongly. It was never safe to just take rcu_read_lock() and loop unless you verify that pid_alive(g) == T, even the first next_thread() can point to the already freed/reused memory. This patch adds signal_struct->thread_head and task->thread_node to create the normal rcu-safe list with the stable head. The new for_each_thread(g, t) helper is always safe under rcu_read_lock() as long as this task_struct can't go away. Note: of course it is ugly to have both task_struct->thread_node and the old task_struct->thread_group, we will kill it later, after we change the users of while_each_thread() to use for_each_thread(). Perhaps we can kill it even before we convert all users, we can reimplement next_thread(t) using the new thread_head/thread_node. But we can't do this right now because this will lead to subtle behavioural changes. For example, do/while_each_thread() always sees at least one task, while for_each_thread() can do nothing if the whole thread group has died. Or thread_group_empty(), currently its semantics is not clear unless thread_group_leader(p) and we need to audit the callers before we can change it. So this patch adds the new interface which has to coexist with the old one for some time, hopefully the next changes will be more or less straightforward and the old one will go away soon. Signed-off-by:
-
- 29 May, 2014 14 commits
-
-
Jiri Slaby authored
-
Eric Dumazet authored
[ Upstream commit e33d0ba8 ] Recycling skb always had been very tough... This time it appears GRO layer can accumulate skb->truesize adjustments made by drivers when they attach a fragment to skb. skb_gro_receive() can only subtract from skb->truesize the used part of a fragment. I spotted this problem seeing TcpExtPruneCalled and TcpExtTCPRcvCollapsed that were unexpected with a recent kernel, where TCP receive window should be sized properly to accept traffic coming from a driver not overshooting skb->truesize. Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
-
Li RongQing authored
[ Upstream commit fbdc0ad0 ] the value of itag is a random value from stack, and may not be initiated by fib_validate_source, which called fib_combine_itag if CONFIG_IP_ROUTE_CLASSID is not set This will make the cached dst uncertainty Signed-off-by:
Li RongQing <roy.qing.li@gmail.com> Acked-by:
Alexei Starovoitov <ast@plumgrid.com> Signed-off-by:
-
Steffen Klassert authored
[ Upstream commit 78ff4be4 ] We need to initialize the fallback device to have a correct mtu set on this device. Otherwise the mtu is set to null and the device is unusable. Fixes: fd58156e ("IPIP: Use ip-tunneling code.") Cc: Pravin B Shelar <pshelar@nicira.com> Signed-off-by:
Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by:
-
Antonio Quartulli authored
[ Upstream commit 377fe0f9 ] A pointer to the orig_node representing a bat-gateway is stored in the gw_node->orig_node member, but the refcount for such orig_node is never increased. This leads to memory faults when gw_node->orig_node is accessed and the originator has already been freed. Fix this by increasing the refcount on gw_node creation and decreasing it on gw_node free. Signed-off-by:
Antonio Quartulli <antonio@open-mesh.com> Signed-off-by:
Marek Lindner <mareklindner@neomailbox.ch> Signed-off-by:
-
Cong Wang authored
[ Upstream commit 200b916f ] From: Cong Wang <cwang@twopensource.com> commit 50624c93 (net: Delay default_device_exit_batch until no devices are unregistering) introduced rtnl_lock_unregistering() for default_device_exit_batch(). Same race could happen we when rmmod a driver which calls rtnl_link_unregister() as we call dev->destructor without rtnl lock. For long term, I think we should clean up the mess of netdev_run_todo() and net namespce exit code. Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: David S. Miller <davem@davemloft.net> Signed-off-by:
Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by:
Cong Wang <cwang@twopensource.com> Signed-off-by:
-
Susant Sahani authored
[ Upstream commit c8965932 ] The function ip6_tnl_validate assumes that the rtnl attribute IFLA_IPTUN_PROTO always be filled . If this attribute is not filled by the userspace application kernel get crashed with NULL pointer dereference. This patch fixes the potential kernel crash when IFLA_IPTUN_PROTO is missing . Signed-off-by:
Susant Sahani <susant@redhat.com> Acked-by:
Thomas Graf <tgraf@suug.ch> Signed-off-by:
-
Nikolay Aleksandrov authored
[ Upstream commit 1c363900 ] If the sfc driver is in legacy interrupt mode (either explicitly by using interrupt_mode module param or by falling back to it) it will hit a warning at kernel/irq/manage.c because it will try to free an irq which wasn't allocated by it in the first place because the MSI(X) irqs are zero and it'll try to free them unconditionally. So fix it by checking if we're in legacy mode and freeing the appropriate irqs. CC: Zenghui Shi <zshi@redhat.com> CC: Ben Hutchings <ben@decadent.org.uk> CC: <linux-net-drivers@solarflare.com> CC: Shradha Shah <sshah@solarflare.com> CC: David S. Miller <davem@davemloft.net> Fixes: 1899c111 ("sfc: Fix IRQ cleanup in case of a probe failure") Reported-by:
Zenghui Shi <zshi@redhat.com> Signed-off-by:
Nikolay Aleksandrov <nikolay@redhat.com> Acked-by:
Shradha Shah <sshah@solarflare.com> Signed-off-by:
-
Peter Christensen authored
[ Upstream commit bbeb0ead ] Clearing the IFF_ALLMULTI flag on a down interface could cause an allmulti overflow on the underlying interface. Attempting the set IFF_ALLMULTI on the underlying interface would cause an error and the log message: "allmulti touches root, set allmulti failed." Signed-off-by:
Peter Christensen <pch@ordbogen.com> Signed-off-by:
-
Bjørn Mork authored
[ Upstream commit 6b5eeb7f ] This driver maps 802.1q VLANs to MBIM sessions. The mapping is based on a bogus assumption that all tagged frames will use the acceleration API because we enable NETIF_F_HW_VLAN_CTAG_TX. This fails for e.g. frames tagged in userspace using packet sockets. Such frames will erroneously be considered as untagged and silently dropped based on not being IP. Fix by falling back to looking into the ethernet header for a tag if no accelerated tag was found. Fixes: a82c7ce5 ("net: cdc_ncm: map MBIM IPS SessionID to VLAN ID") Cc: Greg Suarez <gsuarez@smithmicro.com> Signed-off-by:
-
Sergey Popovich authored
[ Upstream commit aeefa1ec ] Increment fib_info_cnt in fib_create_info() right after successfuly alllocating fib_info structure, overwise fib_metrics allocation failure leads to fib_info_cnt incorrectly decremented in free_fib_info(), called on error path from fib_create_info(). Signed-off-by:
-
Florian Westphal authored
[ Upstream commit 418a3156 ] If conntrack defragments incoming ipv6 frags it stores largest original frag size in ip6cb and sets ->local_df. We must thus first test the largest original frag size vs. mtu, and not vice versa. Without this patch PKTTOOBIG is still generated in ip6_fragment() later in the stack, but 1) IPSTATS_MIB_INTOOBIGERRORS won't increment 2) packet did (needlessly) traverse netfilter postrouting hook. Fixes: fe6cc55f ("net: ip, ipv6: handle gso skbs in forwarding path") Signed-off-by:
Florian Westphal <fw@strlen.de> Signed-off-by:
-
Florian Westphal authored
[ Upstream commit ca6c5d4a ] local_df means 'ignore DF bit if set', so if its set we're allowed to perform ip fragmentation. This wasn't noticed earlier because the output path also drops such skbs (and emits needed icmp error) and because netfilter ip defrag did not set local_df until couple of days ago. Only difference is that DF-packets-larger-than MTU now discarded earlier (f.e. we avoid pointless netfilter postrouting trip). While at it, drop the repeated test ip_exceeds_mtu, checking it once is enough... Fixes: fe6cc55f ("net: ip, ipv6: handle gso skbs in forwarding path") Signed-off-by:
-
Ying Cai authored
[ Upstream commit e96f2e7c ] In ip_tunnel_rcv(), set skb->network_header to inner IP header before IP_ECN_decapsulate(). Without the fix, IP_ECN_decapsulate() takes outer IP header as inner IP header, possibly causing error messages or packet drops. Note that this skb_reset_network_header() call was in this spot when the original feature for checking consistency of ECN bits through tunnels was added in eccc1bb8 ("tunnel: drop packet if ECN present with not-ECT"). It was only removed from this spot in 3d7b46cd ("ip_tunnel: push generic protocol handling to ip_tunnel module."). Fixes: 3d7b46cd ("ip_tunnel: push generic protocol handling to ip_tunnel module.") Reported-by:
Neal Cardwell <ncardwell@google.com> Signed-off-by:
Ying Cai <ycai@google.com> Acked-by:
-
