- Nov 18, 2021
-
-
Miquel Raynal authored
commit f16b7d2a upstream. Following the introduction of the generic ECC engine infrastructure, it was necessary to reorganize the code and move the ECC configuration in the ->attach_chip() hook. Failing to do that properly lead to a first series of fixes supposed to stabilize the situation. Unfortunately, this only fixed the use of software ECC engines, preventing any other kind of engine to be used, including on-die ones. It is now time to (finally) fix the situation by ensuring that we still provide a default (eg. software ECC) but will still support different ECC engines such as on-die ECC engines if properly described in the device tree. There are no changes needed on the core side in order to do this, but we just need to leverage the logic there which allows: 1- a subsystem default (set to Host engines in the raw NAND world) 2- a driver specific default (here set to software ECC engines) 3- any type of engine requested by the user (ie. described in the DT) As the raw NAND subsystem has not yet been fully converted to the ECC engine infrastructure, in order to provide a default ECC engine for this driver we need to set chip->ecc.engine_type *before* calling nand_scan(). During the initialization step, the core will consider this entry as the default engine for this driver. This value may of course be overloaded by the user if the usual DT properties are provided. Fixes: 8fc6f1f0 ("mtd: rawnand: pasemi: Move the ECC initialization to ->attach_chip()") Cc: stable@vger.kernel.org Signed-off-by:
Miquel Raynal <miquel.raynal@bootlin.com> Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-7-miquel.raynal@bootlin.com Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Miquel Raynal authored
commit b5b5b4dc upstream. Following the introduction of the generic ECC engine infrastructure, it was necessary to reorganize the code and move the ECC configuration in the ->attach_chip() hook. Failing to do that properly lead to a first series of fixes supposed to stabilize the situation. Unfortunately, this only fixed the use of software ECC engines, preventing any other kind of engine to be used, including on-die ones. It is now time to (finally) fix the situation by ensuring that we still provide a default (eg. software ECC) but will still support different ECC engines such as on-die ECC engines if properly described in the device tree. There are no changes needed on the core side in order to do this, but we just need to leverage the logic there which allows: 1- a subsystem default (set to Host engines in the raw NAND world) 2- a driver specific default (here set to software ECC engines) 3- any type of engine requested by the user (ie. described in the DT) As the raw NAND subsystem has not yet been fully converted to the ECC engine infrastructure, in order to provide a default ECC engine for this driver we need to set chip->ecc.engine_type *before* calling nand_scan(). During the initialization step, the core will consider this entry as the default engine for this driver. This value may of course be overloaded by the user if the usual DT properties are provided. Fixes: f6341f64 ("mtd: rawnand: gpio: Move the ECC initialization to ->attach_chip()") Cc: stable@vger.kernel.org Signed-off-by:
-
Miquel Raynal authored
commit f9d8570b upstream. Following the introduction of the generic ECC engine infrastructure, it was necessary to reorganize the code and move the ECC configuration in the ->attach_chip() hook. Failing to do that properly lead to a first series of fixes supposed to stabilize the situation. Unfortunately, this only fixed the use of software ECC engines, preventing any other kind of engine to be used, including on-die ones. It is now time to (finally) fix the situation by ensuring that we still provide a default (eg. software ECC) but will still support different ECC engines such as on-die ECC engines if properly described in the device tree. There are no changes needed on the core side in order to do this, but we just need to leverage the logic there which allows: 1- a subsystem default (set to Host engines in the raw NAND world) 2- a driver specific default (here set to software ECC engines) 3- any type of engine requested by the user (ie. described in the DT) As the raw NAND subsystem has not yet been fully converted to the ECC engine infrastructure, in order to provide a default ECC engine for this driver we need to set chip->ecc.engine_type *before* calling nand_scan(). During the initialization step, the core will consider this entry as the default engine for this driver. This value may of course be overloaded by the user if the usual DT properties are provided. Fixes: 6dd09f77 ("mtd: rawnand: mpc5121: Move the ECC initialization to ->attach_chip()") Cc: stable@vger.kernel.org Signed-off-by:
-
Miquel Raynal authored
commit 6bcd2960 upstream. Following the introduction of the generic ECC engine infrastructure, it was necessary to reorganize the code and move the ECC configuration in the ->attach_chip() hook. Failing to do that properly lead to a first series of fixes supposed to stabilize the situation. Unfortunately, this only fixed the use of software ECC engines, preventing any other kind of engine to be used, including on-die ones. It is now time to (finally) fix the situation by ensuring that we still provide a default (eg. software ECC) but will still support different ECC engines such as on-die ECC engines if properly described in the device tree. There are no changes needed on the core side in order to do this, but we just need to leverage the logic there which allows: 1- a subsystem default (set to Host engines in the raw NAND world) 2- a driver specific default (here set to software ECC engines) 3- any type of engine requested by the user (ie. described in the DT) As the raw NAND subsystem has not yet been fully converted to the ECC engine infrastructure, in order to provide a default ECC engine for this driver we need to set chip->ecc.engine_type *before* calling nand_scan(). During the initialization step, the core will consider this entry as the default engine for this driver. This value may of course be overloaded by the user if the usual DT properties are provided. Fixes: d525914b ("mtd: rawnand: xway: Move the ECC initialization to ->attach_chip()") Cc: stable@vger.kernel.org Cc: Jan Hoffmann <jan@3e8.eu> Cc: Kestrel seventyfour <kestrelseventyfour@gmail.com> Signed-off-by:
Jan Hoffmann <jan@3e8.eu> Link: https://lore.kernel.org/linux-mtd/20210928222258.199726-10-miquel.raynal@bootlin.com Signed-off-by:
-
Miquel Raynal authored
commit d707bb74 upstream. Following the introduction of the generic ECC engine infrastructure, it was necessary to reorganize the code and move the ECC configuration in the ->attach_chip() hook. Failing to do that properly lead to a first series of fixes supposed to stabilize the situation. Unfortunately, this only fixed the use of software ECC engines, preventing any other kind of engine to be used, including on-die ones. It is now time to (finally) fix the situation by ensuring that we still provide a default (eg. software ECC) but will still support different ECC engines such as on-die ECC engines if properly described in the device tree. There are no changes needed on the core side in order to do this, but we just need to leverage the logic there which allows: 1- a subsystem default (set to Host engines in the raw NAND world) 2- a driver specific default (here set to software ECC engines) 3- any type of engine requested by the user (ie. described in the DT) As the raw NAND subsystem has not yet been fully converted to the ECC engine infrastructure, in order to provide a default ECC engine for this driver we need to set chip->ecc.engine_type *before* calling nand_scan(). During the initialization step, the core will consider this entry as the default engine for this driver. This value may of course be overloaded by the user if the usual DT properties are provided. Fixes: 59d93473 ("mtd: rawnand: ams-delta: Move the ECC initialization to ->attach_chip()") Cc: stable@vger.kernel.org Signed-off-by:
-
Halil Pasic authored
commit ad9a1451 upstream. Since commit 48720ba5 ("virtio/s390: use DMA memory for ccw I/O and classic notifiers") we were supposed to make sure that virtio_ccw_release_dev() completes before the ccw device and the attached dma pool are torn down, but unfortunately we did not. Before that commit it used to be OK to delay cleaning up the memory allocated by virtio-ccw indefinitely (which isn't really intuitive for guys used to destruction happens in reverse construction order), but now we trigger a BUG_ON if the genpool is destroyed before all memory allocated from it is deallocated. Which brings down the guest. We can observe this problem, when unregister_virtio_device() does not give up the last reference to the virtio_device (e.g. because a virtio-scsi attached scsi disk got removed without previously unmounting its previously mounted partition). To make sure that the genpool is only destroyed after all the necessary freeing is done let us take a reference on the ccw device on each ccw_device_dma_zalloc() and give it up on each ccw_device_dma_free(). Actually there are multiple approaches to fixing the problem at hand that can work. The upside of this one is that it is the safest one while remaining simple. We don't crash the guest even if the driver does not pair allocations and frees. The downside is the reference counting overhead, that the reference counting for ccw devices becomes more complex, in a sense that we need to pair the calls to the aforementioned functions for it to be correct, and that if we happen to leak, we leak more than necessary (the whole ccw device instead of just the genpool). Some alternatives to this approach are taking a reference in virtio_ccw_online() and giving it up in virtio_ccw_release_dev() or making sure virtio_ccw_release_dev() completes its work before virtio_ccw_remove() returns. The downside of these approaches is that these are less safe against programming errors. Cc: <stable@vger.kernel.org> # v5.3 Signed-off-by:
Halil Pasic <pasic@linux.ibm.com> Fixes: 48720ba5 ("virtio/s390: use DMA memory for ccw I/O and classic notifiers") Reported-by:
<bfu@redhat.com> Reviewed-by:
Vineeth Vijayan <vneethv@linux.ibm.com> Acked-by:
Cornelia Huck <cohuck@redhat.com> Signed-off-by:
Vasily Gorbik <gor@linux.ibm.com> Signed-off-by:
-
Harald Freudenberger authored
commit 3826350e upstream. When a queue is switched to soft offline during heavy load and later switched to soft online again and now used, it may be that the caller is blocked forever in the ioctl call. The failure occurs because there is a pending reply after the queue(s) have been switched to offline. This orphaned reply is received when the queue is switched to online and is accidentally counted for the outstanding replies. So when there was a valid outstanding reply and this orphaned reply is received it counts as the outstanding one thus dropping the outstanding counter to 0. Voila, with this counter the receive function is not called any more and the real outstanding reply is never received (until another request comes in...) and the ioctl blocks. The fix is simple. However, instead of readjusting the counter when an orphaned reply is detected, I check the queue status for not empty and compare this to the outstanding counter. So if the queue is not empty then the counter must not drop to 0 but at least have a value of 1. Signed-off-by:
Harald Freudenberger <freude@linux.ibm.com> Cc: stable@vger.kernel.org Signed-off-by:
-
Sven Schnelle authored
commit 213fca9e upstream. commit 9c6c273a ("timer: Remove init_timer_on_stack() in favor of timer_setup_on_stack()") changed the timer setup from init_timer_on_stack(() to timer_setup(), but missed to change the mod_timer() call. And while at it, use msecs_to_jiffies() instead of the open coded timeout calculation. Cc: stable@vger.kernel.org Fixes: 9c6c273a ("timer: Remove init_timer_on_stack() in favor of timer_setup_on_stack()") Signed-off-by:
Sven Schnelle <svens@linux.ibm.com> Reviewed-by:
-
Vineeth Vijayan authored
commit a4751f15 upstream. Check the validity of subchanel before reading other fields in the schib. Fixes: d3683c05 ("s390/cio: add dev_busid sysfs entry for each subchannel") CC: <stable@vger.kernel.org> Reported-by:
-
Marek Vasut authored
commit 33a5471f upstream. The note in c2adda27 ("video: backlight: Add of_find_backlight helper in backlight.c") says that gpio-backlight uses brightness as power state. This has been fixed since in ec665b75 ("backlight: gpio-backlight: Correct initial power state handling") and other backlight drivers do not require this workaround. Drop the workaround. This fixes the case where e.g. pwm-backlight can perfectly well be set to brightness 0 on boot in DT, which without this patch leads to the display brightness to be max instead of off. Fixes: c2adda27 ("video: backlight: Add of_find_backlight helper in backlight.c") Cc: <stable@vger.kernel.org> # 5.4+ Cc: <stable@vger.kernel.org> # 4.19.x: ec665b75: backlight: gpio-backlight: Correct initial power state handling Signed-off-by:
Marek Vasut <marex@denx.de> Acked-by:
Noralf Trønnes <noralf@tronnes.org> Reviewed-by:
Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by:
Lee Jones <lee.jones@linaro.org> Signed-off-by:
-
Jack Andersen authored
commit 313c84b5 upstream. This patch extends the DLN2 driver; adding cell for adc_dln2 module. The original patch[1] fell through the cracks when the driver was added so ADC has never actually been usable. That patch did not have ACPI support which was added in v5.9, so the oldest supported version this current patch can be backported to is 5.10. [1] https://www.spinics.net/lists/linux-iio/msg33975.html Cc: <stable@vger.kernel.org> # 5.10+ Signed-off-by:
Jack Andersen <jackoalan@gmail.com> Signed-off-by:
-
Michal Hocko authored
commit 60e2793d upstream. Any allocation failure during the #PF path will return with VM_FAULT_OOM which in turn results in pagefault_out_of_memory. This can happen for 2 different reasons. a) Memcg is out of memory and we rely on mem_cgroup_oom_synchronize to perform the memcg OOM handling or b) normal allocation fails. The latter is quite problematic because allocation paths already trigger out_of_memory and the page allocator tries really hard to not fail allocations. Anyway, if the OOM killer has been already invoked there is no reason to invoke it again from the #PF path. Especially when the OOM condition might be gone by that time and we have no way to find out other than allocate. Moreover if the allocation failed and the OOM killer hasn't been invoked then we are unlikely to do the right thing from the #PF context because we have already lost the allocation context and restictions and therefore might oom kill a task from a different NUMA domain. This all suggests that there is no legitimate reason to trigger out_of_memory from pagefault_out_of_memory so drop it. Just to be sure that no #PF path returns with VM_FAULT_OOM without allocation print a warning that this is happening before we restart the #PF. [VvS: #PF allocation can hit into limit of cgroup v1 kmem controller. This is a local problem related to memcg, however, it causes unnecessary global OOM kills that are repeated over and over again and escalate into a real disaster. This has been broken since kmem accounting has been introduced for cgroup v1 (3.8). There was no kmem specific reclaim for the separate limit so the only way to handle kmem hard limit was to return with ENOMEM. In upstream the problem will be fixed by removing the outdated kmem limit, however stable and LTS kernels cannot do it and are still affected. This patch fixes the problem and should be backported into stable/LTS.] Link: https://lkml.kernel.org/r/f5fd8dd8-0ad4-c524-5f65-920b01972a42@virtuozzo.com Signed-off-by:
Michal Hocko <mhocko@suse.com> Signed-off-by:
Vasily Averin <vvs@virtuozzo.com> Acked-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Vasily Averin authored
commit 0b28179a upstream. Patch series "memcg: prohibit unconditional exceeding the limit of dying tasks", v3. Memory cgroup charging allows killed or exiting tasks to exceed the hard limit. It can be misused and allowed to trigger global OOM from inside a memcg-limited container. On the other hand if memcg fails allocation, called from inside #PF handler it triggers global OOM from inside pagefault_out_of_memory(). To prevent these problems this patchset: (a) removes execution of out_of_memory() from pagefault_out_of_memory(), becasue nobody can explain why it is necessary. (b) allow memcg to fail allocation of dying/killed tasks. This patch (of 3): Any allocation failure during the #PF path will return with VM_FAULT_OOM which in turn results in pagefault_out_of_memory which in turn executes out_out_memory() and can kill a random task. An allocation might fail when the current task is the oom victim and there are no memory reserves left. The OOM killer is already handled at the page allocator level for the global OOM and at the charging level for the memcg one. Both have much more information about the scope of allocation/charge request. This means that either the OOM killer has been invoked properly and didn't lead to the allocation success or it has been skipped because it couldn't have been invoked. In both cases triggering it from here is pointless and even harmful. It makes much more sense to let the killed task die rather than to wake up an eternally hungry oom-killer and send him to choose a fatter victim for breakfast. Link: https://lkml.kernel.org/r/0828a149-786e-7c06-b70a-52d086818ea3@virtuozzo.com Signed-off-by:
-
Naveen N. Rao authored
upstream commit b7540d62 Emit similar instruction sequences to commit a048a07d ("powerpc/64s: Add support for a store forwarding barrier at kernel entry/exit") when encountering BPF_NOSPEC. Mitigations are enabled depending on what the firmware advertises. In particular, we do not gate these mitigations based on current settings, just like in x86. Due to this, we don't need to take any action if mitigations are enabled or disabled at runtime. Signed-off-by:
Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/956570cbc191cd41f8274bed48ee757a86dac62a.1633464148.git.naveen.n.rao@linux.vnet.ibm.com [adjust macros to account for commits 1c9debbc and ef909ba9. adjust security feature checks to account for commit 84ed26fd] Signed-off-by:
-
Naveen N. Rao authored
upstream commit 03090592 Add a helper to return the stf_barrier type for the current processor. Signed-off-by:
-
Naveen N. Rao authored
upstream commit 3832ba4e Add checks to ensure that we never emit branch instructions with truncated branch offsets. Suggested-by:
Johan Almbladh <johan.almbladh@anyfinetworks.com> Reviewed-by:
Christophe Leroy <christophe.leroy@csgroup.eu> Acked-by:
Song Liu <songliubraving@fb.com> Acked-by:
-
Naveen N. Rao authored
upstream commit 4549c3ea Add a helper to check if a given offset is within the branch range for a powerpc conditional branch instruction, and update some sites to use the new helper. Signed-off-by:
-
Vasily Averin authored
commit a4ebf1b6 upstream. Memory cgroup charging allows killed or exiting tasks to exceed the hard limit. It is assumed that the amount of the memory charged by those tasks is bound and most of the memory will get released while the task is exiting. This is resembling a heuristic for the global OOM situation when tasks get access to memory reserves. There is no global memory shortage at the memcg level so the memcg heuristic is more relieved. The above assumption is overly optimistic though. E.g. vmalloc can scale to really large requests and the heuristic would allow that. We used to have an early break in the vmalloc allocator for killed tasks but this has been reverted by commit b8c8a338 ("Revert "vmalloc: back off when the current task is killed""). There are likely other similar code paths which do not check for fatal signals in an allocation&charge loop. Also there are some kernel objects charged to a memcg which are not bound to a process life time. It has been observed that it is not really hard to trigger these bypasses and cause global OOM situation. One potential way to address these runaways would be to limit the amount of excess (similar to the global OOM with limited oom reserves). This is certainly possible but it is not really clear how much of an excess is desirable and still protects from global OOMs as that would have to consider the overall memcg configuration. This patch is addressing the problem by removing the heuristic altogether. Bypass is only allowed for requests which either cannot fail or where the failure is not desirable while excess should be still limited (e.g. atomic requests). Implementation wise a killed or dying task fails to charge if it has passed the OOM killer stage. That should give all forms of reclaim chance to restore the limit before the failure (ENOMEM) and tell the caller to back off. In addition, this patch renames should_force_charge() helper to task_is_dying() because now its use is not associated witch forced charging. This patch depends on pagefault_out_of_memory() to not trigger out_of_memory(), because then a memcg failure can unwind to VM_FAULT_OOM and cause a global OOM killer. Link: https://lkml.kernel.org/r/8f5cebbb-06da-4902-91f0-6566fc4b4203@virtuozzo.com Signed-off-by:
-
Dominique Martinet authored
commit 27eb4c31 upstream. Link: https://lkml.kernel.org/r/99338965-d36c-886e-cd0e-1d8fff2b4746@gmail.com Reported-by:
<syzbot+06472778c97ed94af66d@syzkaller.appspotmail.com> Cc: stable@vger.kernel.org Signed-off-by:
Dominique Martinet <asmadeus@codewreck.org> Signed-off-by:
-
Daniel Borkmann authored
[ Upstream commit 3dc20f47 ] Currently, it is not possible to migrate a neighbor entry between NUD_PERMANENT state and NTF_USE flag with a dynamic NUD state from a user space control plane. Similarly, it is not possible to add/remove NTF_EXT_LEARNED flag from an existing neighbor entry in combination with NTF_USE flag. This is due to the latter directly calling into neigh_event_send() without any meta data updates as happening in __neigh_update(). Thus, to enable this use case, extend the latter with a NEIGH_UPDATE_F_USE flag where we break the NUD_PERMANENT state in particular so that a latter neigh_event_send() is able to re-resolve a neighbor entry. Before fix, NUD_PERMANENT -> NUD_* & NTF_USE: # ./ip/ip n replace 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a PERMANENT [...] # ./ip/ip n replace 192.168.178.30 dev enp5s0 use extern_learn # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a PERMANENT [...] As can be seen, despite the admin-triggered replace, the entry remains in the NUD_PERMANENT state. After fix, NUD_PERMANENT -> NUD_* & NTF_USE: # ./ip/ip n replace 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a PERMANENT [...] # ./ip/ip n replace 192.168.178.30 dev enp5s0 use extern_learn # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a extern_learn REACHABLE [...] # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a extern_learn STALE [...] # ./ip/ip n replace 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a PERMANENT [...] After the fix, the admin-triggered replace switches to a dynamic state from the NTF_USE flag which triggered a new neighbor resolution. Likewise, we can transition back from there, if needed, into NUD_PERMANENT. Similar before/after behavior can be observed for below transitions: Before fix, NTF_USE -> NTF_USE | NTF_EXT_LEARNED -> NTF_USE: # ./ip/ip n replace 192.168.178.30 dev enp5s0 use # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a REACHABLE [...] # ./ip/ip n replace 192.168.178.30 dev enp5s0 use extern_learn # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a REACHABLE [...] After fix, NTF_USE -> NTF_USE | NTF_EXT_LEARNED -> NTF_USE: # ./ip/ip n replace 192.168.178.30 dev enp5s0 use # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a REACHABLE [...] # ./ip/ip n replace 192.168.178.30 dev enp5s0 use extern_learn # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a extern_learn REACHABLE [...] # ./ip/ip n replace 192.168.178.30 dev enp5s0 use # ./ip/ip n 192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a REACHABLE [..] Signed-off-by:
Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Roopa Prabhu <roopa@nvidia.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Sasha Levin <sashal@kernel.org>
-
Jaegeuk Kim authored
commit 92d602bc upstream. We use inline_dentry which requires to allocate dentry page when adding a link. If we allow to reclaim memory from filesystem, we do down_read(&sbi->cp_rwsem) twice by f2fs_lock_op(). I think this should be okay, but how about stopping the lockdep complaint [1]? f2fs_create() - f2fs_lock_op() - f2fs_do_add_link() - __f2fs_find_entry - f2fs_get_read_data_page() -> kswapd - shrink_node - f2fs_evict_inode - f2fs_lock_op() [1] fs_reclaim ){+.+.}-{0:0} : kswapd0: lock_acquire+0x114/0x394 kswapd0: __fs_reclaim_acquire+0x40/0x50 kswapd0: prepare_alloc_pages+0x94/0x1ec kswapd0: __alloc_pages_nodemask+0x78/0x1b0 kswapd0: pagecache_get_page+0x2e0/0x57c kswapd0: f2fs_get_read_data_page+0xc0/0x394 kswapd0: f2fs_find_data_page+0xa4/0x23c kswapd0: find_in_level+0x1a8/0x36c kswapd0: __f2fs_find_entry+0x70/0x100 kswapd0: f2fs_do_add_link+0x84/0x1ec kswapd0: f2fs_mkdir+0xe4/0x1e4 kswapd0: vfs_mkdir+0x110/0x1c0 kswapd0: do_mkdirat+0xa4/0x160 kswapd0: __arm64_sys_mkdirat+0x24/0x34 kswapd0: el0_svc_common.llvm.17258447499513131576+0xc4/0x1e8 kswapd0: do_el0_svc+0x28/0xa0 kswapd0: el0_svc+0x24/0x38 kswapd0: el0_sync_handler+0x88/0xec kswapd0: el0_sync+0x1c0/0x200 kswapd0: -> #1 ( &sbi->cp_rwsem ){++++}-{3:3} : kswapd0: lock_acquire+0x114/0x394 kswapd0: down_read+0x7c/0x98 kswapd0: f2fs_do_truncate_blocks+0x78/0x3dc kswapd0: f2fs_truncate+0xc8/0x128 kswapd0: f2fs_evict_inode+0x2b8/0x8b8 kswapd0: evict+0xd4/0x2f8 kswapd0: iput+0x1c0/0x258 kswapd0: do_unlinkat+0x170/0x2a0 kswapd0: __arm64_sys_unlinkat+0x4c/0x68 kswapd0: el0_svc_common.llvm.17258447499513131576+0xc4/0x1e8 kswapd0: do_el0_svc+0x28/0xa0 kswapd0: el0_svc+0x24/0x38 kswapd0: el0_sync_handler+0x88/0xec kswapd0: el0_sync+0x1c0/0x200 Cc: stable@vger.kernel.org Fixes: bdbc90fa ("f2fs: don't put dentry page in pagecache into highmem") Reviewed-by:
Chao Yu <chao@kernel.org> Reviewed-by:
Stanley Chu <stanley.chu@mediatek.com> Reviewed-by:
Light Hsieh <light.hsieh@mediatek.com> Tested-by:
Jaegeuk Kim <jaegeuk@kernel.org> Signed-off-by:
-
Guo Ren authored
commit 69ea4630 upstream. When using "devm_request_threaded_irq(,,,,IRQF_ONESHOT,,)" in a driver, only the first interrupt is handled, and following interrupts are never delivered (initially reported in [1]). That's because the RISC-V PLIC cannot EOI masked interrupts, as explained in the description of Interrupt Completion in the PLIC spec [2]: <quote> The PLIC signals it has completed executing an interrupt handler by writing the interrupt ID it received from the claim to the claim/complete register. The PLIC does not check whether the completion ID is the same as the last claim ID for that target. If the completion ID does not match an interrupt source that *is currently enabled* for the target, the completion is silently ignored. </quote> Re-enable the interrupt before completion if it has been masked during the handling, and remask it afterwards. [1] http://lists.infradead.org/pipermail/linux-riscv/2021-July/007441.html [2] https://github.com/riscv/riscv-plic-spec/blob/8bc15a35d07c9edf7b5d23fec9728302595ffc4d/riscv-plic.adoc Fixes: bb0fed1c ("irqchip/sifive-plic: Switch to fasteoi flow") Reported-by:
Vincent Pelletier <plr.vincent@gmail.com> Tested-by:
Nikita Shubin <nikita.shubin@maquefel.me> Signed-off-by:
Guo Ren <guoren@linux.alibaba.com> Cc: stable@vger.kernel.org Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Atish Patra <atish.patra@wdc.com> Reviewed-by:
Anup Patel <anup@brainfault.org> [maz: amended commit message] Signed-off-by:
Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20211105094748.3894453-1-guoren@kernel.org Signed-off-by:
-
Michael Pratt authored
commit ca7752ca upstream. copy_process currently copies task_struct.posix_cputimers_work as-is. If a timer interrupt arrives while handling clone and before dup_task_struct completes then the child task will have: 1. posix_cputimers_work.scheduled = true 2. posix_cputimers_work.work queued. copy_process clears task_struct.task_works, so (2) will have no effect and posix_cpu_timers_work will never run (not to mention it doesn't make sense for two tasks to share a common linked list). Since posix_cpu_timers_work never runs, posix_cputimers_work.scheduled is never cleared. Since scheduled is set, future timer interrupts will skip scheduling work, with the ultimate result that the task will never receive timer expirations. Together, the complete flow is: 1. Task 1 calls clone(), enters kernel. 2. Timer interrupt fires, schedules task work on Task 1. 2a. task_struct.posix_cputimers_work.scheduled = true 2b. task_struct.posix_cputimers_work.work added to task_struct.task_works. 3. dup_task_struct() copies Task 1 to Task 2. 4. copy_process() clears task_struct.task_works for Task 2. 5. Future timer interrupts on Task 2 see task_struct.posix_cputimers_work.scheduled = true and skip scheduling work. Fix this by explicitly clearing contents of task_struct.posix_cputimers_work in copy_process(). This was never meant to be shared or inherited across tasks in the first place. Fixes: 1fb497dd ("posix-cpu-timers: Provide mechanisms to defer timer handling to task_work") Reported-by:
Rhys Hiltner <rhys@justin.tv> Signed-off-by:
Michael Pratt <mpratt@google.com> Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20211101210615.716522-1-mpratt@google.com Signed-off-by:
-
Dave Jones authored
commit e629fc14 upstream. Errata SKX37 is word-for-word identical to the other errata listed in this workaround. I happened to notice this after investigating a CMCI storm on a Skylake host. While I can't confirm this was the root cause, spurious corrected errors does sound like a likely suspect. Fixes: 2976908e ("x86/mce: Do not log spurious corrected mce errors") Signed-off-by:
Dave Jones <davej@codemonkey.org.uk> Signed-off-by:
Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by:
Tony Luck <tony.luck@intel.com> Cc: <stable@vger.kernel.org> Link: https://lkml.kernel.org/r/20211029205759.GA7385@codemonkey.org.uk Signed-off-by:
-
Maciej W. Rozycki authored
commit a923a267 upstream. Fix assembly errors like: {standard input}: Assembler messages: {standard input}:287: Error: opcode not supported on this processor: mips3 (mips3) `dins $10,$7,32,32' {standard input}:680: Error: opcode not supported on this processor: mips3 (mips3) `dins $10,$7,32,32' {standard input}:1274: Error: opcode not supported on this processor: mips3 (mips3) `dins $12,$9,32,32' {standard input}:2175: Error: opcode not supported on this processor: mips3 (mips3) `dins $10,$7,32,32' make[1]: *** [scripts/Makefile.build:277: mm/highmem.o] Error 1 with code produced from `__cmpxchg64' for MIPS64r2 CPU configurations using CONFIG_32BIT and CONFIG_PHYS_ADDR_T_64BIT. This is due to MIPS_ISA_ARCH_LEVEL downgrading the assembly architecture to `r4000' i.e. MIPS III for MIPS64r2 configurations, while there is a block of code containing a DINS MIPS64r2 instruction conditionalized on MIPS_ISA_REV >= 2 within the scope of the downgrade. The assembly architecture override code pattern has been put there for LL/SC instructions, so that code compiles for configurations that select a processor to build for that does not support these instructions while still providing run-time support for processors that do, dynamically switched by non-constant `cpu_has_llsc'. It went in with linux-mips.org commit aac8aa77 ("Enable a suitable ISA for the assembler around ll/sc so that code builds even for processors that don't support the instructions. Plus minor formatting fixes.") back in 2005. Fix the problem by wrapping these instructions along with the adjacent SYNC instructions only, following the practice established with commit cfd54de3 ("MIPS: Avoid move psuedo-instruction whilst using MIPS_ISA_LEVEL") and commit 378ed6f0 ("MIPS: Avoid using .set mips0 to restore ISA"). Strictly speaking the SYNC instructions do not have to be wrapped as they are only used as a Loongson3 erratum workaround, so they will be enabled in the assembler by default, but do this so as to keep code consistent with other places. Reported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
Maciej W. Rozycki <macro@orcam.me.uk> Fixes: c7e2d71d ("MIPS: Fix set_pte() for Netlogic XLR using cmpxchg64()") Cc: stable@vger.kernel.org # v5.1+ Signed-off-by:
Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by:
-
Helge Deller authored
commit 279917e2 upstream. I noticed that sometimes at kernel startup the backtraces did not included the function names of init functions. Their address were not resolved to function names and instead only the address was printed. Debugging shows that the culprit is is_ksym_addr() which is called by the backtrace functions to check if an address belongs to a function in the kernel. The problem occurs only for CONFIG_KALLSYMS_ALL=y. When looking at is_ksym_addr() one can see that for CONFIG_KALLSYMS_ALL=y the function only tries to resolve the address via is_kernel() function, which checks like this: if (addr >= _stext && addr <= _end) return 1; On parisc the init functions are located before _stext, so this check fails. Other platforms seem to have all functions (including init functions) behind _stext. The following patch moves the _stext symbol at the beginning of the kernel and thus includes the init section. This fixes the check and does not seem to have any negative side effects on where the kernel mapping happens in the map_pages() function in arch/parisc/mm/init.c. Signed-off-by:
Helge Deller <deller@gmx.de> Cc: stable@kernel.org # 5.4+ Signed-off-by:
-
Arnd Bergmann authored
commit 418ace99 upstream. Naresh and Antonio ran into a build failure with latest Debian armhf compilers, with lots of output like tmp/ccY3nOAs.s:2215: Error: selected processor does not support `cpsid i' in ARM mode As it turns out, $(cc-option) fails early here when the FPU is not selected before CPU architecture is selected, as the compiler option check runs before enabling -msoft-float, which causes a problem when testing a target architecture level without an FPU: cc1: error: '-mfloat-abi=hard': selected architecture lacks an FPU Passing e.g. -march=armv6k+fp in place of -march=armv6k would avoid this issue, but the fallback logic is already broken because all supported compilers (gcc-5 and higher) are much more recent than these options, and building with -march=armv5t as a fallback no longer works. The best way forward that I see is to just remove all the checks, which also has the nice side-effect of slightly improving the startup time for 'make'. The -mtune=marvell-f option was apparently never supported by any mainline compiler, and the custom Codesourcery gcc build that did support is now too old to build kernels, so just use -mtune=xscale unconditionally for those. This should be safe to apply on all stable kernels, and will be required in order to keep building them with gcc-11 and higher. Link: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=996419 Reported-by:
Antonio Terceiro <antonio.terceiro@linaro.org> Reported-by:
Naresh Kamboju <naresh.kamboju@linaro.org> Reported-by:
Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Tested-by:
Sebastian Reichel <sebastian.reichel@collabora.com> Tested-by:
Klaus Kudielka <klaus.kudielka@gmail.com> Cc: Matthias Klose <doko@debian.org> Cc: stable@vger.kernel.org Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by:
-
Michał Mirosław authored
commit 0d08e7bf upstream. Currently __set_fixmap() bails out with a warning when called in early boot from early_iounmap(). Fix it, and while at it, make the comment a bit easier to understand. Cc: <stable@vger.kernel.org> Fixes: b089c31c ("ARM: 8667/3: Fix memory attribute inconsistencies when using fixmap") Acked-by:
Ard Biesheuvel <ardb@kernel.org> Signed-off-by:
Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by:
-
Willem de Bruijn authored
[ Upstream commit d336509c ] The below commit added optional support for passing a bind address. It configures the sockaddr bind arguments before parsing options and reconfigures on options -b and -4. This broke support for passing port (-p) on its own. Configure sockaddr after parsing all arguments. Fixes: 3327a9c4 ("selftests: add functionals test for UDP GRO") Reported-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
Willem de Bruijn <willemb@google.com> Signed-off-by:
-
Rahul Lakkireddy authored
[ Upstream commit 4ca110bf ] Ensure diagnostics monitoring support is implemented for the SFF 8472 compliant port module and set the correct length for ethtool port module eeprom read. Fixes: f56ec676 ("cxgb4: Add support for ethtool i2c dump") Signed-off-by:
Manoj Malviya <manojmalviya@chelsio.com> Signed-off-by:
Rahul Lakkireddy <rahul.lakkireddy@chelsio.com> Signed-off-by:
-
Dust Li authored
[ Upstream commit e5d5aadc ] We got the following WARNING when running ab/nginx test with RDMA link flapping (up-down-up). The reason is when smc_sock fallback and at linkdown happens simultaneously, we may got the following situation: __smc_lgr_terminate() --> smc_conn_kill() --> smc_close_active_abort() smc_sock->sk_state = SMC_CLOSED sock_put(smc_sock) smc_sock was set to SMC_CLOSED and sock_put() been called when terminate the link group. But later application call close() on the socket, then we got: __smc_release(): if (smc_sock->fallback) smc_sock->sk_state = SMC_CLOSED sock_put(smc_sock) Again we set the smc_sock to CLOSED through it's already in CLOSED state, and double put the refcnt, so the following warning happens: refcount_t: underflow; use-after-free. WARNING: CPU: 5 PID: 860 at lib/refcount.c:28 refcount_warn_saturate+0x8d/0xf0 Modules linked in: CPU: 5 PID: 860 Comm: nginx Not tainted 5.10.46+ #403 Hardware name: Alibaba Cloud Alibaba Cloud ECS, BIOS 8c24b4c 04/01/2014 RIP: 0010:refcount_warn_saturate+0x8d/0xf0 Code: 05 5c 1e b5 01 01 e8 52 25 bc ff 0f 0b c3 80 3d 4f 1e b5 01 00 75 ad 48 RSP: 0018:ffffc90000527e50 EFLAGS: 00010286 RAX: 0000000000000026 RBX: ffff8881300df2c0 RCX: 0000000000000027 RDX: 0000000000000000 RSI: ffff88813bd58040 RDI: ffff88813bd58048 RBP: 0000000000000000 R08: 0000000000000003 R09: 0000000000000001 R10: ffff8881300df2c0 R11: ffffc90000527c78 R12: ffff8881300df340 R13: ffff8881300df930 R14: ffff88810b3dad80 R15: ffff8881300df4f8 FS: 00007f739de8fb80(0000) GS:ffff88813bd40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000000a01b008 CR3: 0000000111b64003 CR4: 00000000003706e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: smc_release+0x353/0x3f0 __sock_release+0x3d/0xb0 sock_close+0x11/0x20 __fput+0x93/0x230 task_work_run+0x65/0xa0 exit_to_user_mode_prepare+0xf9/0x100 syscall_exit_to_user_mode+0x27/0x190 entry_SYSCALL_64_after_hwframe+0x44/0xa9 This patch adds check in __smc_release() to make sure we won't do an extra sock_put() and set the socket to CLOSED when its already in CLOSED state. Fixes: 51f1de79 (net/smc: replace sock_put worker by socket refcounting) Signed-off-by:
Dust Li <dust.li@linux.alibaba.com> Reviewed-by:
Tony Lu <tonylu@linux.alibaba.com> Signed-off-by:
Karsten Graul <kgraul@linux.ibm.com> Signed-off-by:
-
Eiichi Tsukata authored
[ Upstream commit c7cd82b9 ] Currently vosck_connect() increments sock refcount for nonblocking socket each time it's called, which can lead to memory leak if it's called multiple times because connect timeout function decrements sock refcount only once. Fixes it by making vsock_connect() return -EALREADY immediately when sock state is already SS_CONNECTING. Fixes: d021c344 ("VSOCK: Introduce VM Sockets") Reviewed-by:
Stefano Garzarella <sgarzare@redhat.com> Signed-off-by:
Eiichi Tsukata <eiichi.tsukata@nutanix.com> Signed-off-by:
-
Vladimir Oltean authored
[ Upstream commit f64ab8e4 ] Commit fe28c53e ("net: stmmac: fix taprio configuration when base_time is in the past") allowed some base time values in the past, but apparently not all, the base-time value of 0 (Jan 1st 1970) is still explicitly denied by the driver. Remove the bogus check. Fixes: b60189e0 ("net: stmmac: Integrate EST with TAPRIO scheduler API") Signed-off-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by:
Kurt Kanzenbach <kurt@linutronix.de> Signed-off-by:
-
Guangbin Huang authored
[ Upstream commit 688db0c7 ] Currently, driver only allow configuring ETS bandwidth of TCs according to the max TC number queried from firmware. However, the hardware actually supports 8 TCs and users may need to configure ETS bandwidth of all TCs, so remove the restriction. Fixes: 330baff5 ("net: hns3: add ETS TC weight setting in SSU module") Signed-off-by:
Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by:
-
Yufeng Mo authored
[ Upstream commit e140c798 ] When fully configure VLANs for a VF, then unload the VF while triggering a reset to PF, will cause a kernel crash because the irq is already uninit. [ 293.177579] ------------[ cut here ]------------ [ 293.183502] kernel BUG at drivers/pci/msi.c:352! [ 293.189547] Internal error: Oops - BUG: 0 [#1] SMP ...... [ 293.390124] Workqueue: hclgevf hclgevf_service_task [hclgevf] [ 293.402627] pstate: 80c00009 (Nzcv daif +PAN +UAO) [ 293.414324] pc : free_msi_irqs+0x19c/0x1b8 [ 293.425429] lr : free_msi_irqs+0x18c/0x1b8 [ 293.436545] sp : ffff00002716fbb0 [ 293.446950] x29: ffff00002716fbb0 x28: 0000000000000000 [ 293.459519] x27: 0000000000000000 x26: ffff45b91ea16b00 [ 293.472183] x25: 0000000000000000 x24: ffffa587b08f4700 [ 293.484717] x23: ffffc591ac30e000 x22: ffffa587b08f8428 [ 293.497190] x21: ffffc591ac30e300 x20: 0000000000000000 [ 293.509594] x19: ffffa58a062a8300 x18: 0000000000000000 [ 293.521949] x17: 0000000000000000 x16: ffff45b91dcc3f48 [ 293.534013] x15: 0000000000000000 x14: 0000000000000000 [ 293.545883] x13: 0000000000000040 x12: 0000000000000228 [ 293.557508] x11: 0000000000000020 x10: 0000000000000040 [ 293.568889] x9 : ffff45b91ea1e190 x8 : ffffc591802d0000 [ 293.580123] x7 : ffffc591802d0148 x6 : 0000000000000120 [ 293.591190] x5 : ffffc591802d0000 x4 : 0000000000000000 [ 293.602015] x3 : 0000000000000000 x2 : 0000000000000000 [ 293.612624] x1 : 00000000000004a4 x0 : ffffa58a1e0c6b80 [ 293.623028] Call trace: [ 293.630340] free_msi_irqs+0x19c/0x1b8 [ 293.638849] pci_disable_msix+0x118/0x140 [ 293.647452] pci_free_irq_vectors+0x20/0x38 [ 293.656081] hclgevf_uninit_msi+0x44/0x58 [hclgevf] [ 293.665309] hclgevf_reset_rebuild+0x1ac/0x2e0 [hclgevf] [ 293.674866] hclgevf_reset+0x358/0x400 [hclgevf] [ 293.683545] hclgevf_reset_service_task+0xd0/0x1b0 [hclgevf] [ 293.693325] hclgevf_service_task+0x4c/0x2e8 [hclgevf] [ 293.702307] process_one_work+0x1b0/0x448 [ 293.710034] worker_thread+0x54/0x468 [ 293.717331] kthread+0x134/0x138 [ 293.724114] ret_from_fork+0x10/0x18 [ 293.731324] Code: f940b000 b4ffff00 a903e7b8 f90017b6 (d4210000) This patch fixes the problem by waiting for the VF reset done while unloading the VF. Fixes: e2cb1dec ("net: hns3: Add HNS3 VF HCL(Hardware Compatibility Layer) Support") Signed-off-by:
Yufeng Mo <moyufeng@huawei.com> Signed-off-by:
-
Eric Dumazet authored
[ Upstream commit 6dc25401 ] 1) if q->tk_offset == TK_OFFS_MAX, then get_tcp_tstamp() calls ktime_mono_to_any() with out-of-bound value. 2) if q->tk_offset is changed in taprio_parse_clockid(), taprio_get_time() might also call ktime_mono_to_any() with out-of-bound value as sysbot found: UBSAN: array-index-out-of-bounds in kernel/time/timekeeping.c:908:27 index 3 is out of range for type 'ktime_t *[3]' CPU: 1 PID: 25668 Comm: kworker/u4:0 Not tainted 5.15.0-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Workqueue: bat_events batadv_iv_send_outstanding_bat_ogm_packet Call Trace: <TASK> __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106 ubsan_epilogue+0xb/0x5a lib/ubsan.c:151 __ubsan_handle_out_of_bounds.cold+0x62/0x6c lib/ubsan.c:291 ktime_mono_to_any+0x1d4/0x1e0 kernel/time/timekeeping.c:908 get_tcp_tstamp net/sched/sch_taprio.c:322 [inline] get_packet_txtime net/sched/sch_taprio.c:353 [inline] taprio_enqueue_one+0x5b0/0x1460 net/sched/sch_taprio.c:420 taprio_enqueue+0x3b1/0x730 net/sched/sch_taprio.c:485 dev_qdisc_enqueue+0x40/0x300 net/core/dev.c:3785 __dev_xmit_skb net/core/dev.c:3869 [inline] __dev_queue_xmit+0x1f6e/0x3630 net/core/dev.c:4194 batadv_send_skb_packet+0x4a9/0x5f0 net/batman-adv/send.c:108 batadv_iv_ogm_send_to_if net/batman-adv/bat_iv_ogm.c:393 [inline] batadv_iv_ogm_emit net/batman-adv/bat_iv_ogm.c:421 [inline] batadv_iv_send_outstanding_bat_ogm_packet+0x6d7/0x8e0 net/batman-adv/bat_iv_ogm.c:1701 process_one_work+0x9b2/0x1690 kernel/workqueue.c:2298 worker_thread+0x658/0x11f0 kernel/workqueue.c:2445 kthread+0x405/0x4f0 kernel/kthread.c:327 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295 Fixes: 7ede7b03 ("taprio: make clock reference conversions easier") Fixes: 54002066 ("taprio: Adjust timestamps for TCP packets") Signed-off-by:
syzbot <syzkaller@googlegroups.com> Reviewed-by:
Vinicius Costa Gomes <vinicius.gomes@intel.com> Link: https://lore.kernel.org/r/20211108180815.1822479-1-eric.dumazet@gmail.com Signed-off-by:
Jakub Kicinski <kuba@kernel.org> Signed-off-by:
-
Muchun Song authored
[ Upstream commit 10a6de19 ] DEFINE_PROC_SHOW_ATTRIBUTE() is supposed to be used to define a series of functions and variables to register proc file easily. And the users can use proc_create_data() to pass their own private data and get it via seq->private in the callback. Unfortunately, the proc file system use PDE_DATA() to get private data instead of inode->i_private. So fix it. Fortunately, there only one user of it which does not pass any private data, so this bug does not break any in-tree codes. Link: https://lkml.kernel.org/r/20211029032638.84884-1-songmuchun@bytedance.com Fixes: 97a32539 ("proc: convert everything to "struct proc_ops"") Signed-off-by:
Muchun Song <songmuchun@bytedance.com> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Stephen Rothwell <sfr@canb.auug.org.au> Cc: Florent Revest <revest@chromium.org> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by:
-
Dan Carpenter authored
[ Upstream commit 1c360cc1 ] The priv->ntfy_blocks[] has "priv->num_ntfy_blks" elements so this > needs to be >= to prevent an off by one bug. The priv->ntfy_blocks[] array is allocated in gve_alloc_notify_blocks(). Fixes: 87a7f321 ("gve: Recover from queue stall due to missed IRQ") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
John Fastabend authored
[ Upstream commit e0dc3b93 ] Strparser is reusing the qdisc_skb_cb struct to stash the skb message handling progress, e.g. offset and length of the skb. First this is poorly named and inherits a struct from qdisc that doesn't reflect the actual usage of cb[] at this layer. But, more importantly strparser is using the following to access its metadata. (struct _strp_msg *)((void *)skb->cb + offsetof(struct qdisc_skb_cb, data)) Where _strp_msg is defined as: struct _strp_msg { struct strp_msg strp; /* 0 8 */ int accum_len; /* 8 4 */ /* size: 12, cachelines: 1, members: 2 */ /* last cacheline: 12 bytes */ }; So we use 12 bytes of ->data[] in struct. However in BPF code running parser and verdict the user has read capabilities into the data[] array as well. Its not too problematic, but we should not be exposing internal state to BPF program. If its really needed then we can use the probe_read() APIs which allow reading kernel memory. And I don't believe cb[] layer poses any API breakage by moving this around because programs can't depend on cb[] across layers. In order to fix another issue with a ctx rewrite we need to stash a temp variable somewhere. To make this work cleanly this patch builds a cb struct for sk_skb types called sk_skb_cb struct. Then we can use this consistently in the strparser, sockmap space. Additionally we can start allowing ->cb[] write access after this. Fixes: 604326b4 ("bpf, sockmap: convert to generic sk_msg interface") Signed-off-by:
John Fastabend <john.fastabend@gmail.com> Signed-off-by:
Jussi Maki <joamaki@gmail.com> Reviewed-by:
Jakub Sitnicki <jakub@cloudflare.com> Link: https://lore.kernel.org/bpf/20211103204736.248403-5-john.fastabend@gmail.com Signed-off-by:
-
John Fastabend authored
[ Upstream commit b8b8315e ] We do not need to handle unhash from BPF side we can simply wait for the close to happen. The original concern was a socket could transition from ESTABLISHED state to a new state while the BPF hook was still attached. But, we convinced ourself this is no longer possible and we also improved BPF sockmap to handle listen sockets so this is no longer a problem. More importantly though there are cases where unhash is called when data is in the receive queue. The BPF unhash logic will flush this data which is wrong. To be correct it should keep the data in the receive queue and allow a receiving application to continue reading the data. This may happen when tcp_abort() is received for example. Instead of complicating the logic in unhash simply moving all this to tcp_close() hook solves this. Fixes: 51199405 ("bpf: skb_verdict, support SK_PASS on RX BPF path") Signed-off-by:
-
