- 07 Aug, 2017 40 commits
-
-
Zheng Li authored
ipv6: Should use consistent conditional judgement for ip6 fragment between __ip6_append_data and ip6_finish_output [ Upstream commit e4c5e13a ] There is an inconsistent conditional judgement between __ip6_append_data and ip6_finish_output functions, the variable length in __ip6_append_data just include the length of application's payload and udp6 header, don't include the length of ipv6 header, but in ip6_finish_output use (skb->len > ip6_skb_dst_mtu(skb)) as judgement, and skb->len include the length of ipv6 header. That causes some particular application's udp6 payloads whose length are between (MTU - IPv6 Header) and MTU were fragmented by ip6_fragment even though the rst->dev support UFO feature. Add the length of ipv6 header to length in __ip6_append_data to keep consistent conditional judgement as ip6_finish_output for ip6 fragment. Signed-off-by:
Zheng Li <james.z.li@ericsson.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Signed-off-by:
Sasha Levin <alexander.levin@verizon.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Or Gerlitz authored
[ Upstream commit 9da34cd3 ] Under the switchdev/offloads mode, packets that don't match any e-switch steering rule are sent towards the e-switch management port. We use a NIC HW steering rule set per vport (uplink and VFs) to make them be received into the host OS through the respective vport representor netdevice. Currnetly such missed RoCE packets will not get to this NIC steering rule, and hence VF RoCE will not work over the slow path of the offloads mode. This is b/c these packets will be matched by a steering rule added by the firmware that serves RoCE traffic set on the PF NIC vport which is also the e-switch management port under SRIOV. Disabling RoCE on the e-switch management vport when we are in the offloads mode, will signal to the firmware to remove their RoCE rule, and then the missed RoCE packets will be matched by the representor NIC steering rule as any other missed packets. To achieve that, we disable RoCE on the PF vport. We do that by removing (hot-unplugging) the IB device instance associated with the PF. This is also required by our current model where the PF serves as the uplink representor and hence only SW switching (TC, bridge, OVS) applications and slow path vport mlx5e net-device should be running over that vport. Fixes: c930a3ad ('net/mlx5e: Add devlink based SRIOV mode changes') Signed-off-by:
Or Gerlitz <ogerlitz@mellanox.com> Reviewed-by:
Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by:
Saeed Mahameed <saeedm@mellanox.com> Signed-off-by:
-
Pali Rohár authored
[ Upstream commit 4cf48f1d ] Trying to initialize eMMC slot as SDIO or SD cause failure in n900 port of qemu. eMMC itself is not detected and is not working. Real Nokia N900 harware does not have this problem. As eMMC is really not SDIO or SD based such change is harmless and will fix support for qemu. Signed-off-by:
Pali Rohár <pali.rohar@gmail.com> Acked-by:
Pavel Machek <pavel@ucw.cz> Signed-off-by:
Tony Lindgren <tony@atomide.com> Signed-off-by:
-
Roger Quadros authored
[ Upstream commit 5acd016c ] USB2 port can be operated in dual-role mode but till we have dual-role support in dwc3 driver let's limit this port to peripheral mode. If we don't do so it defaults to host mode. USB1 port is meant for host only operation and we don't want both ports in host only mode. Signed-off-by:
Roger Quadros <rogerq@ti.com> Signed-off-by:
-
Milo Kim authored
[ Upstream commit 82038157 ] Specify the power button interrupt number which is from the datasheet. Signed-off-by:
Milo Kim <woogyom.kim@gmail.com> Acked-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Milo Kim authored
[ Upstream commit 81d7358d ] Add interrupt specifiers for USB and AC charger input. Interrupt numbers are from the datasheet. Fix wrong property for compatible string. Signed-off-by:
-
Maninder Singh authored
[ Upstream commit 5066d529 ] Issue caught with static analysis tool: "Dangerous usage of 'name' (strncpy doesn't always 0-terminate it)" Use strlcpy _includes_ the NUL terminator, and strlcat() which ensures that it won't overflow the buffer. Reported-by:
Maninder Singh <maninder1.s@samsung.com> Signed-off-by:
Vaneet Narang <v.narang@samsung.com> CC: Russell King <linux@armlinux.org.uk> Signed-off-by:
-
Chun-Hao Lin authored
[ Upstream commit 610c9087 ] This chip is the same as RTL8168, but its device id is 0x8161. Signed-off-by:
Chun-Hao Lin <hau@realtek.com> Signed-off-by:
-
Thomas Gleixner authored
[ Upstream commit 0dad3a30 ] If mce_device_init() fails then the mce device pointer is NULL and the AMD mce code happily dereferences it. Add a sanity check. Reported-by:
Markus Trippelsdorf <markus@trippelsdorf.de> Reported-by:
Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by:
-
Dan Williams authored
commit bbb3be17 upstream. Fix warnings of the form... WARNING: CPU: 10 PID: 4983 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x62/0x80 sysfs: cannot create duplicate filename '/class/dax/dax12.0' Call Trace: dump_stack+0x63/0x86 __warn+0xcb/0xf0 warn_slowpath_fmt+0x5a/0x80 ? kernfs_path_from_node+0x4f/0x60 sysfs_warn_dup+0x62/0x80 sysfs_do_create_link_sd.isra.2+0x97/0xb0 sysfs_create_link+0x25/0x40 device_add+0x266/0x630 devm_create_dax_dev+0x2cf/0x340 [dax] dax_pmem_probe+0x1f5/0x26e [dax_pmem] nvdimm_bus_probe+0x71/0x120 ...by reusing the namespace id for the device-dax instance name. Now that we have decided that there will never by more than one device-dax instance per libnvdimm-namespace parent device [1], we can directly reuse the namepace ids. There are some possible follow-on cleanups, but those are saved for a later patch to simplify the -stable backport. [1]: https://lists.01.org/pipermail/linux-nvdimm/2016-December/008266.html Fixes: 98a29c39 ("libnvdimm, namespace: allow creation of multiple pmem...") Cc: Jeff Moyer <jmoyer@redhat.com> Reported-by:
Dariusz Dokupil <dariusz.dokupil@intel.com> Signed-off-by:
Dan Williams <dan.j.williams@intel.com> Signed-off-by:
-
Eric Dumazet authored
commit 6e7bc478 upstream. My recent change missed fact that UFO would perform a complete UDP checksum before segmenting in frags. In this case skb->ip_summed is set to CHECKSUM_NONE. We need to add this valid case to skb_needs_check() Fixes: b2504a5d ("net: reduce skb_warn_bad_offload() noise") Signed-off-by:
Eric Dumazet <edumazet@google.com> Cc: Willem de Bruijn <willemb@google.com> Signed-off-by:
-
Kees Cook authored
commit e9a330c4 upstream. The per-prz spinlock should be using the dynamic initializer so that lockdep can correctly track it. Without this, under lockdep, we get a warning at boot that the lock is in non-static memory. Fixes: 10970449 ("pstore: Make spinlock per zone instead of global") Fixes: 76d5692a ("pstore: Correctly initialize spinlock and flags") Signed-off-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
-
Kees Cook authored
commit 76d5692a upstream. The ram backend wasn't always initializing its spinlock correctly. Since it was coming from kzalloc memory, though, it was harmless on architectures that initialize unlocked spinlocks to 0 (at least x86 and ARM). This also fixes a possibly ignored flag setting too. When running under CONFIG_DEBUG_SPINLOCK, the following Oops was visible: [ 0.760836] persistent_ram: found existing buffer, size 29988, start 29988 [ 0.765112] persistent_ram: found existing buffer, size 30105, start 30105 [ 0.769435] persistent_ram: found existing buffer, size 118542, start 118542 [ 0.785960] persistent_ram: found existing buffer, size 0, start 0 [ 0.786098] persistent_ram: found existing buffer, size 0, start 0 [ 0.786131] pstore: using zlib compression [ 0.790716] BUG: spinlock bad magic on CPU#0, swapper/0/1 [ 0.790729] lock: 0xffffffc0d1ca9bb0, .magic: 00000000, .owner: <none>/-1, .owner_cpu: 0 [ 0.790742] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.10.0-rc2+ #913 [ 0.790747] Hardware name: Google Kevin (DT) [ 0.790750] Call trace: [ 0.790768] [<ffffff900808ae88>] dump_backtrace+0x0/0x2bc [ 0.790780] [<ffffff900808b164>] show_stack+0x20/0x28 [ 0.790794] [<ffffff9008460ee0>] dump_stack+0xa4/0xcc [ 0.790809] [<ffffff9008113cfc>] spin_dump+0xe0/0xf0 [ 0.790821] [<ffffff9008113d3c>] spin_bug+0x30/0x3c [ 0.790834] [<ffffff9008113e28>] do_raw_spin_lock+0x50/0x1b8 [ 0.790846] [<ffffff9008a2d2ec>] _raw_spin_lock_irqsave+0x54/0x6c [ 0.790862] [<ffffff90083ac3b4>] buffer_size_add+0x48/0xcc [ 0.790875] [<ffffff90083acb34>] persistent_ram_write+0x60/0x11c [ 0.790888] [<ffffff90083aab1c>] ramoops_pstore_write_buf+0xd4/0x2a4 [ 0.790900] [<ffffff90083a9d3c>] pstore_console_write+0xf0/0x134 [ 0.790912] [<ffffff900811c304>] console_unlock+0x48c/0x5e8 [ 0.790923] [<ffffff900811da18>] register_console+0x3b0/0x4d4 [ 0.790935] [<ffffff90083aa7d0>] pstore_register+0x1a8/0x234 [ 0.790947] [<ffffff90083ac250>] ramoops_probe+0x6b8/0x7d4 [ 0.790961] [<ffffff90085ca548>] platform_drv_probe+0x7c/0xd0 [ 0.790972] [<ffffff90085c76ac>] driver_probe_device+0x1b4/0x3bc [ 0.790982] [<ffffff90085c7ac8>] __device_attach_driver+0xc8/0xf4 [ 0.790996] [<ffffff90085c4bfc>] bus_for_each_drv+0xb4/0xe4 [ 0.791006] [<ffffff90085c7414>] __device_attach+0xd0/0x158 [ 0.791016] [<ffffff90085c7b18>] device_initial_probe+0x24/0x30 [ 0.791026] [<ffffff90085c648c>] bus_probe_device+0x50/0xe4 [ 0.791038] [<ffffff90085c35b8>] device_add+0x3a4/0x76c [ 0.791051] [<ffffff90087d0e84>] of_device_add+0x74/0x84 [ 0.791062] [<ffffff90087d19b8>] of_platform_device_create_pdata+0xc0/0x100 [ 0.791073] [<ffffff90087d1a2c>] of_platform_device_create+0x34/0x40 [ 0.791086] [<ffffff900903c910>] of_platform_default_populate_init+0x58/0x78 [ 0.791097] [<ffffff90080831fc>] do_one_initcall+0x88/0x160 [ 0.791109] [<ffffff90090010ac>] kernel_init_freeable+0x264/0x31c [ 0.791123] [<ffffff9008a25bd0>] kernel_init+0x18/0x11c [ 0.791133] [<ffffff9008082ec0>] ret_from_fork+0x10/0x50 [ 0.793717] console [pstore-1] enabled [ 0.797845] pstore: Registered ramoops as persistent store backend [ 0.804647] ramoops: attached 0x100000@0xf7edc000, ecc: 0/0 Fixes: 663deb47 ("pstore: Allow prz to control need for locking") Fixes: 10970449 ("pstore: Make spinlock per zone instead of global") Reported-by:
Brian Norris <briannorris@chromium.org> Signed-off-by:
-
Joel Fernandes authored
commit 663deb47 upstream. In preparation of not locking at all for certain buffers depending on if there's contention, make locking optional depending on the initialization of the prz. Signed-off-by:
Joel Fernandes <joelaf@google.com> [kees: moved locking flag into prz instead of via caller arguments] Signed-off-by:
-
Andrzej Hajda authored
commit a2370ba2 upstream. Bool values should be negated using logical operators. Using bitwise operators results in unexpected and possibly incorrect results. Reported-by:
David Binderman <dcb314@hotmail.com> Signed-off-by:
Andrzej Hajda <a.hajda@samsung.com> Signed-off-by:
Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by:
Mauro Carvalho Chehab <mchehab@s-opensource.com> Signed-off-by:
-
Al Viro authored
commit 49d31c2f upstream. take_dentry_name_snapshot() takes a safe snapshot of dentry name; if the name is a short one, it gets copied into caller-supplied structure, otherwise an extra reference to external name is grabbed (those are never modified). In either case the pointer to stable string is stored into the same structure. dentry must be held by the caller of take_dentry_name_snapshot(), but may be freely dropped afterwards - the snapshot will stay until destroyed by release_dentry_name_snapshot(). Intended use: struct name_snapshot s; take_dentry_name_snapshot(&s, dentry); ... access s.name ... release_dentry_name_snapshot(&s); Replaces fsnotify_oldname_...(), gets used in fsnotify to obtain the name to pass down with event. Signed-off-by:
Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
-
Valentin Vidic authored
commit 860f01e9 upstream. systemd by default starts watchdog on reboot and sets the timer to ShutdownWatchdogSec=10min. Reboot handler in ipmi_watchdog than reduces the timer to 120s which is not enough time to boot a Xen machine with a lot of RAM. As a result the machine is rebooted the second time during the long run of (XEN) Scrubbing Free RAM..... Fix this by setting the timer to 120s only if it was previously set to a low value. Signed-off-by:
Valentin Vidic <Valentin.Vidic@CARNet.hr> Signed-off-by:
Corey Minyard <cminyard@mvista.com> Signed-off-by:
-
Ismail, Mustafa authored
commit 5a7a88f1 upstream. The port number is only valid if IB_QP_PORT is set in the mask. So only check port number if it is valid to prevent modify_qp from failing due to an invalid port number. Fixes: 5ecce4c9("Check port number supplied by user verbs cmds") Reviewed-by:
Steve Wise <swise@opengridcomputing.com> Signed-off-by:
Mustafa Ismail <mustafa.ismail@intel.com> Tested-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by:
Doug Ledford <dledford@redhat.com> Signed-off-by:
-
Konstantin Khlebnikov authored
commit 96b77745 upstream. Commit: 2f5177f0 ("sched/cgroup: Fix/cleanup cgroup teardown/init") .. moved sched_online_group() from css_online() to css_alloc(). It exposes half-baked task group into global lists before initializing generic cgroup stuff. LTP testcase (third in cgroup_regression_test) written for testing similar race in kernels 2.6.26-2.6.28 easily triggers this oops: BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: kernfs_path_from_node_locked+0x260/0x320 CPU: 1 PID: 30346 Comm: cat Not tainted 4.10.0-rc5-test #4 Call Trace: ? kernfs_path_from_node+0x4f/0x60 kernfs_path_from_node+0x3e/0x60 print_rt_rq+0x44/0x2b0 print_rt_stats+0x7a/0xd0 print_cpu+0x2fc/0xe80 ? __might_sleep+0x4a/0x80 sched_debug_show+0x17/0x30 seq_read+0xf2/0x3b0 proc_reg_read+0x42/0x70 __vfs_read+0x28/0x130 ? security_file_permission+0x9b/0xc0 ? rw_verify_area+0x4e/0xb0 vfs_read+0xa5/0x170 SyS_read+0x46/0xa0 entry_SYSCALL_64_fastpath+0x1e/0xad Here the task group is already linked into the global RCU-protected 'task_groups' list, but the css->cgroup pointer is still NULL. This patch reverts this chunk and moves online back to css_online(). Signed-off-by:
Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Tejun Heo <tj@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Fixes: 2f5177f0 ("sched/cgroup: Fix/cleanup cgroup teardown/init") Link: http://lkml.kernel.org/r/148655324740.424917.5302984537258726349.stgit@buzzSigned-off-by:
Ingo Molnar <mingo@kernel.org> Signed-off-by:
Matt Fleming <matt@codeblueprint.co.uk> Signed-off-by:
-
Sudeep Holla authored
commit cb710ab1 upstream. We already check if the message is empty before calling the client tx_done callback. Calling completion on a wait event is also invalid if the message is empty. This patch moves the existing empty message check earlier. Fixes: 2b6d83e2 ("mailbox: Introduce framework for mailbox") Signed-off-by:
Sudeep Holla <sudeep.holla@arm.com> Signed-off-by:
Jassi Brar <jaswinder.singh@linaro.org> Signed-off-by:
Amit Pundir <amit.pundir@linaro.org> Signed-off-by:
-
Sudeep Holla authored
commit cc6eeaa3 upstream. If a wait_for_completion_timeout() call returns due to a timeout, complete() can get called after returning from the wait which is incorrect and can cause subsequent transmissions on a channel to fail. Since the wait_for_completion_timeout() sees the completion variable is non-zero caused by the erroneous/spurious complete() call, and it immediately returns without waiting for the time as expected by the client. This patch fixes the issue by skipping complete() call for the timer expiry. Fixes: 2b6d83e2 ("mailbox: Introduce framework for mailbox") Reported-by:
Alexey Klimov <alexey.klimov@arm.com> Signed-off-by:
-
Sudeep Holla authored
commit c61b781e upstream. There exists a race when msg_submit return immediately as there was an active request being processed which may have completed just before it's checked again in mbox_send_message. This will result in return to the caller without waiting in mbox_send_message even when it's blocking Tx. This patch fixes the issue by waiting for the completion always if Tx is in blocking mode. Fixes: 2b6d83e2 ("mailbox: Introduce framework for mailbox") Reported-by:
-
Lior David authored
commit dfb5b098 upstream. When FW crashes with no_fw_recovery option, driver waits for manual recovery with wil->mutex held, this can easily create deadlocks. Fix the problem by moving the wait outside the lock. Signed-off-by:
Lior David <qca_liord@qca.qualcomm.com> Signed-off-by:
Maya Erez <qca_merez@qca.qualcomm.com> Signed-off-by:
Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by:
-
Michal Kazior authored
commit 18ae68ff upstream. WMI ops wrappers did not properly check for null function pointers for spectral scan. This caused null dereference crash with WMI-TLV based firmware which doesn't implement spectral scan. The crash could be triggered with: ip link set dev wlan0 up echo background > /sys/kernel/debug/ieee80211/phy0/ath10k/spectral_scan_ctl The crash looked like this: [ 168.031989] BUG: unable to handle kernel NULL pointer dereference at (null) [ 168.037406] IP: [< (null)>] (null) [ 168.040395] PGD cdd4067 PUD fa0f067 PMD 0 [ 168.043303] Oops: 0010 [#1] SMP [ 168.045377] Modules linked in: ath10k_pci(O) ath10k_core(O) ath mac80211 cfg80211 [last unloaded: cfg80211] [ 168.051560] CPU: 1 PID: 1380 Comm: bash Tainted: G W O 4.8.0 #78 [ 168.054336] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 168.059183] task: ffff88000c460c00 task.stack: ffff88000d4bc000 [ 168.061736] RIP: 0010:[<0000000000000000>] [< (null)>] (null) ... [ 168.100620] Call Trace: [ 168.101910] [<ffffffffa03b9566>] ? ath10k_spectral_scan_config+0x96/0x200 [ath10k_core] [ 168.104871] [<ffffffff811386e2>] ? filemap_fault+0xb2/0x4a0 [ 168.106696] [<ffffffffa03b97e6>] write_file_spec_scan_ctl+0x116/0x280 [ath10k_core] [ 168.109618] [<ffffffff812da3a1>] full_proxy_write+0x51/0x80 [ 168.111443] [<ffffffff811957b8>] __vfs_write+0x28/0x120 [ 168.113090] [<ffffffff812f1a2d>] ? security_file_permission+0x3d/0xc0 [ 168.114932] [<ffffffff8109b912>] ? percpu_down_read+0x12/0x60 [ 168.116680] [<ffffffff811965f8>] vfs_write+0xb8/0x1a0 [ 168.118293] [<ffffffff81197966>] SyS_write+0x46/0xa0 [ 168.119912] [<ffffffff818f2972>] entry_SYSCALL_64_fastpath+0x1a/0xa4 [ 168.121737] Code: Bad RIP value. [ 168.123318] RIP [< (null)>] (null) Signed-off-by:
Michal Kazior <michal.kazior@tieto.com> Signed-off-by:
-
Annie Cherkaev authored
commit 9f5af546 upstream. This fixes a potential buffer overflow in isdn_net.c caused by an unbounded strcpy. [ ISDN seems to be effectively unmaintained, and the I4L driver in particular is long deprecated, but in case somebody uses this.. - Linus ] Signed-off-by:
Jiten Thakkar <jitenmt@gmail.com> Signed-off-by:
Annie Cherkaev <annie.cherk@gmail.com> Cc: Karsten Keil <isdn@linux-pingi.de> Cc: Kees Cook <keescook@chromium.org> Signed-off-by:
-
Jia-Ju Bai authored
commit e8f4ae85 upstream. The driver may sleep under a spin lock, the function call path is: isdn_ppp_mp_receive (acquire the lock) isdn_ppp_mp_reassembly isdn_ppp_push_higher isdn_ppp_decompress isdn_ppp_ccp_reset_trans isdn_ppp_ccp_reset_alloc_state kzalloc(GFP_KERNEL) --> may sleep To fixed it, the "GFP_KERNEL" is replaced with "GFP_ATOMIC". Signed-off-by:
Jia-Ju Bai <baijiaju1990@163.com> Signed-off-by:
-
Florian Fainelli authored
commit 0878fff1 upstream. The Generic PHY driver is a catch-all PHY driver and it should preserve whatever prior initialization has been done by boot loader or firmware agents. For specific PHY device configuration it is expected that a specialized PHY driver would take over that role. Resetting the generic PHY was a bad idea that has lead to several complaints and downstream workarounds e.g: in OpenWrt/LEDE so restore the behavior prior to 87aa9f9c ("net: phy: consolidate PHY reset in phy_init_hw()"). Reported-by:
Felix Fietkau <nbd@nbd.name> Fixes: 87aa9f9c ("net: phy: consolidate PHY reset in phy_init_hw()") Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Sudip Mukherjee authored
commit b6355fb3 upstream. We are checking phy after dereferencing it. We can print the debug information after checking it. If phy is NULL then we will get a good stack trace to tell us that we are in this irq handler. Signed-off-by:
Sudip Mukherjee <sudip.mukherjee@codethink.co.uk> Signed-off-by:
Samuel Ortiz <sameo@linux.intel.com> Signed-off-by:
-
OGAWA Hirofumi authored
commit 24971281 upstream. If port100_send_ack() was called twice or more, it has race to hangup. port100_send_ack() port100_send_ack() init_completion() [...] dev->cmd_cancel = true /* this removes previous from completion */ init_completion() [...] dev->cmd_cancel = true wait_for_completion() /* never be waked up */ wait_for_completion() Like above race, this code is not assuming port100_send_ack() is called twice or more. To fix, this checks dev->cmd_cancel to know if prior cancel is in-flight or not. And never be remove prior task from completion by using reinit_completion(), so this guarantees to be waked up properly soon or later. Signed-off-by:
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by:
-
Thomas Gleixner authored
commit dea1d0f5 upstream. The move of the unpark functions to the control thread moved the BUG_ON() there as well. While it made some sense in the idle thread of the upcoming CPU, it's bogus to crash the control thread on the already online CPU, especially as the function has a return value and the callsite is prepared to handle an error return. Replace it with a WARN_ON_ONCE() and return a proper error code. Fixes: 9cd4f1a4 ("smp/hotplug: Move unparking of percpu threads to the control CPU") Rightfully-ranted-at-by:
-
Thomas Gleixner authored
commit 9cd4f1a4 upstream. Vikram reported the following backtrace: BUG: scheduling while atomic: swapper/7/0/0x00000002 CPU: 7 PID: 0 Comm: swapper/7 Not tainted 4.9.32-perf+ #680 schedule schedule_hrtimeout_range_clock schedule_hrtimeout wait_task_inactive __kthread_bind_mask __kthread_bind __kthread_unpark kthread_unpark cpuhp_online_idle cpu_startup_entry secondary_start_kernel He analyzed correctly that a parked cpu hotplug thread of an offlined CPU was still on the runqueue when the CPU came back online and tried to unpark it. This causes the thread which invoked kthread_unpark() to call wait_task_inactive() and subsequently schedule() with preemption disabled. His proposed workaround was to "make sure" that a parked thread has scheduled out when the CPU goes offline, so the situation cannot happen. But that's still wrong because the root cause is not the fact that the percpu thread is still on the runqueue and neither that preemption is disabled, which could be simply solved by enabling preemption before calling kthread_unpark(). The real issue is that the calling thread is the idle task of the upcoming CPU, which is not supposed to call anything which might sleep. The moron, who wrote that code, missed completely that kthread_unpark() might end up in schedule(). The solution is simpler than expected. The thread which controls the hotplug operation is waiting for the CPU to call complete() on the hotplug state completion. So the idle task of the upcoming CPU can set its state to CPUHP_AP_ONLINE_IDLE and invoke complete(). This in turn wakes the control task on a different CPU, which then can safely do the unpark and kick the now unparked hotplug thread of the upcoming CPU to complete the bringup to the final target state. Control CPU AP bringup_cpu(); __cpu_up() ------------> bringup_ap(); bringup_wait_for_ap() wait_for_completion(); cpuhp_online_idle(); <------------ complete(); unpark(AP->stopper); unpark(AP->hotplugthread); while(1) do_idle(); kick(AP->hotplugthread); wait_for_completion(); hotplug_thread() run_online_callbacks(); complete(); Fixes: 8df3e07e ("cpu/hotplug: Let upcoming cpu bring itself fully up") Reported-by:
Vikram Mulukutla <markivx@codeaurora.org> Signed-off-by:
-
Laurent Pinchart authored
commit 4f7b0d26 upstream. It isn't safe to call drm_dev_unregister() without first initializing mode setting with drm_mode_config_init(). This leads to a crash if either IO memory can't be remapped or vblank initialization fails. Fix this by reordering the initialization sequence. Move vblank initialization after the drm_mode_config_init() call, and move IO remapping before drm_dev_alloc() to avoid the need to perform clean up in case of failure. While at it remove the explicit drm_vblank_cleanup() call from rcar_du_remove() as the drm_dev_unregister() function already cleans up vblank. Signed-off-by:
Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Signed-off-by:
thongsyho <thong.ho.px@rvc.renesas.com> Signed-off-by:
Nhan Nguyen <nhan.nguyen.yb@renesas.com> Signed-off-by:
-
Cheah Kok Cheong authored
commit bf279ece upstream. Move comedi_proc_init to the end to avoid orphaned proc entry if module loading failed. Signed-off-by:
Cheah Kok Cheong <thrust73@gmail.com> Reviewed-by:
Ian Abbott <abbotti@mev.co.uk> Signed-off-by:
-
Greg Kroah-Hartman authored
This reverts commit b4624ff9 which is commit ba4a648f upstream. Michal Hocko writes: JFYI. We have encountered a regression after applying this patch on a large ppc machine. While the patch is the right thing to do it doesn't work well with the current vmalloc area size on ppc and large machines where NUMA nodes are very far from each other. Just for the reference the boot fails on such a machine with bunch of warning preceeding it. See http://lkml.kernel.org/r/20170724134240.GL25221@dhcp22.suse.cz It seems the right thing to do is to enlarge the vmalloc space on ppc but this is not the case in the upstream kernel yet AFAIK. It is also questionable whether that is a stable material but I will decision on you here. We have reverted this patch from our 4.4 based kernel. Newer kernels do not have enlarged vmalloc space yet AFAIK so they won't work properly eiter. This bug is quite rare though because you need a specific HW configuration to trigger the issue - namely NUMA nodes have to be far away from each other in the physical memory space. Cc: Michal Hocko <mhocko@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Nicholas Piggin <npiggin@gmail.com> Signed-off-by:
-
Paul Mackerras authored
commit 7ceaa6dc upstream. At present, HV KVM on POWER8 and POWER9 machines loses any instruction or data breakpoint set in the host whenever a guest is run. Instruction breakpoints are currently only used by xmon, but ptrace and the perf_event subsystem can set data breakpoints as well as xmon. To fix this, we save the host values of the debug registers (CIABR, DAWR and DAWRX) before entering the guest and restore them on exit. To provide space to save them in the stack frame, we expand the stack frame allocated by kvmppc_hv_entry() from 112 to 144 bytes. [paulus@ozlabs.org - Adjusted stack offsets since we aren't saving POWER9-specific registers.] Fixes: b005255e ("KVM: PPC: Book3S HV: Context-switch new POWER8 SPRs", 2014-01-08) Signed-off-by:
Paul Mackerras <paulus@ozlabs.org> Signed-off-by:
-
Paul Mackerras authored
commit 4c3bb4cc upstream. This restores several special-purpose registers (SPRs) to sane values on guest exit that were missed before. TAR and VRSAVE are readable and writable by userspace, and we need to save and restore them to prevent the guest from potentially affecting userspace execution (not that TAR or VRSAVE are used by any known program that run uses the KVM_RUN ioctl). We save/restore these in kvmppc_vcpu_run_hv() rather than on every guest entry/exit. FSCR affects userspace execution in that it can prohibit access to certain facilities by userspace. We restore it to the normal value for the task on exit from the KVM_RUN ioctl. IAMR is normally 0, and is restored to 0 on guest exit. However, with a radix host on POWER9, it is set to a value that prevents the kernel from executing user-accessible memory. On POWER9, we save IAMR on guest entry and restore it on guest exit to the saved value rather than 0. On POWER8 we continue to set it to 0 on guest exit. PSPB is normally 0. We restore it to 0 on guest exit to prevent userspace taking advantage of the guest having set it non-zero (which would allow userspace to set its SMT priority to high). UAMOR is normally 0. We restore it to 0 on guest exit to prevent the AMR from being used as a covert channel between userspace processes, since the AMR is not context-switched at present. [paulus@ozlabs.org - removed IAMR bits that are only needed on POWER9] Fixes: b005255e ("KVM: PPC: Book3S HV: Context-switch new POWER8 SPRs", 2014-01-08) Signed-off-by:
-
Ben Skeggs authored
commit 38bcb208 upstream. Bit 30 being set causes the upper half of BAR2 to stay in physical mode, mapped over the end of VRAM, even when the rest of the BAR has been set to virtual mode. We inherited our initial value from RM, but I'm not aware of any reason we need to keep it that way. This fixes severe GPU hang/lockup issues revealed by Wayland on F26. Shout-out to NVIDIA for the quick response with the potential cause! Signed-off-by:
Ben Skeggs <bskeggs@redhat.com> Signed-off-by:
-
Ilia Mirkin authored
commit a90e049c upstream. GP102's cursors go from chan 17..20. Increase the array size to hold their data properly. Fixes: e50fcff1 ("drm/nouveau/disp/gp102: fix cursor/overlay immediate channel indices") Signed-off-by:
Ilia Mirkin <imirkin@alum.mit.edu> Signed-off-by:
-
Sinclair Yeh authored
commit fcfffdd8 upstream. The current code does not look correct, and the reason for it is probably lost. Since this now generates a compiler warning, fix it to what makes sense. Reported-by:
Arnd Bergmann <arnd@arndb.de> Reported-by:
Sinclair Yeh <syeh@vmware.com> Reviewed-by:
Brian Paul <brianp@vmware.com> Signed-off-by:
-
Ofer Heifetz authored
commit 7e96d559 upstream. Since thread_group worker and raid5d kthread are not in sync, if worker writes stripe before raid5d then requests will be waiting for issue_pendig. Issue observed when building raid5 with ext4, in some build runs jbd2 would get hung and requests were waiting in the HW engine waiting to be issued. Fix this by adding a call to async_tx_issue_pending_all in the raid5_do_work. Signed-off-by:
Ofer Heifetz <oferh@marvell.com> Signed-off-by:
Shaohua Li <shli@fb.com> Signed-off-by:
-
