kernel/linux.git/include, branch v5.4.7

cpufreq: Avoid leaving stale IRQ work items during CPU offline

2019-12-31T15:46:06+00:00

commit 85572c2c4a45a541e880e087b5b17a48198b2416 upstream. The scheduler code calling cpufreq_update_util() may run during CPU offline on the target CPU after the IRQ work lists have been flushed for it, so the target CPU should be prevented from running code that may queue up an IRQ work item on it at that point. Unfortunately, that may not be the case if dvfs_possible_from_any_cpu is set for at least one cpufreq policy in the system, because that allows the CPU going offline to run the utilization update callback of the cpufreq governor on behalf of another (online) CPU in some cases. If that happens, the cpufreq governor callback may queue up an IRQ work on the CPU running it, which is going offline, and the IRQ work may not be flushed after that point. Moreover, that IRQ work cannot be flushed until the "offlining" CPU goes back online, so if any other CPU calls irq_work_sync() to wait for the completion of that IRQ work, it will have to wait until the "offlining" CPU is back online and that may not happen forever. In particular, a system-wide deadlock may occur during CPU online as a result of that. The failing scenario is as follows. CPU0 is the boot CPU, so it creates a cpufreq policy and becomes the "leader" of it (policy->cpu). It cannot go offline, because it is the boot CPU. Next, other CPUs join the cpufreq policy as they go online and they leave it when they go offline. The last CPU to go offline, say CPU3, may queue up an IRQ work while running the governor callback on behalf of CPU0 after leaving the cpufreq policy because of the dvfs_possible_from_any_cpu effect described above. Then, CPU0 is the only online CPU in the system and the stale IRQ work is still queued on CPU3. When, say, CPU1 goes back online, it will run irq_work_sync() to wait for that IRQ work to complete and so it will wait for CPU3 to go back online (which may never happen even in principle), but (worse yet) CPU0 is waiting for CPU1 at that point too and a system-wide deadlock occurs. To address this problem notice that CPUs which cannot run cpufreq utilization update code for themselves (for example, because they have left the cpufreq policies that they belonged to), should also be prevented from running that code on behalf of the other CPUs that belong to a cpufreq policy with dvfs_possible_from_any_cpu set and so in that case the cpufreq_update_util_data pointer of the CPU running the code must not be NULL as well as for the CPU which is the target of the cpufreq utilization update in progress. Accordingly, change cpufreq_this_cpu_can_update() into a regular function in kernel/sched/cpufreq.c (instead of a static inline in a header file) and make it check the cpufreq_update_util_data pointer of the local CPU if dvfs_possible_from_any_cpu is set for the target cpufreq policy. Also update the schedutil governor to do the cpufreq_this_cpu_can_update() check in the non-fast-switch case too to avoid the stale IRQ work issues. Fixes: 99d14d0e16fa ("cpufreq: Process remote callbacks from any CPU if the platform permits") Link: https://lore.kernel.org/linux-pm/20191121093557.bycvdo4xyinbc5cb@vireshk-i7/ Reported-by: Anson Huang Tested-by: Anson Huang Cc: 4.14+ # 4.14+ Signed-off-by: Rafael J. Wysocki Acked-by: Viresh Kumar Tested-by: Peng Fan (i.MX8QXP-MEK) Signed-off-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman

net: avoid potential false sharing in neighbor related code

2019-12-31T15:45:03+00:00

[ Upstream commit 25c7a6d1f90e208ec27ca854b1381ed39842ec57 ] There are common instances of the following construct : if (n->confirmed != now) n->confirmed = now; A C compiler could legally remove the conditional. Use READ_ONCE()/WRITE_ONCE() to avoid this problem. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Sasha Levin

nvmem: core: fix nvmem_cell_write inline function

2019-12-31T15:44:43+00:00

[ Upstream commit 9b8303fc6efa724bd6a90656434fbde2cc6ceb2c ] nvmem_cell_write's buf argument uses different types based on the configuration of CONFIG_NVMEM. The function prototype for enabled NVMEM uses 'void *' type, but the static dummy function for disabled NVMEM uses 'const char *' instead. Fix the different behaviour by always expecting a 'void *' typed buf argument. Fixes: 7a78a7f7695b ("power: reset: nvmem-reboot-mode: use NVMEM as reboot mode write interface") Reported-by: kbuild test robot Cc: Han Nandor Cc: Srinivas Kandagatla Cc: linux-kernel@vger.kernel.org Signed-off-by: Sebastian Reichel Reviewed-By: Han Nandor Signed-off-by: Srinivas Kandagatla Link: https://lore.kernel.org/r/20191029114240.14905-2-srinivas.kandagatla@linaro.org Signed-off-by: Greg Kroah-Hartman Signed-off-by: Sasha Levin

nvme: introduce "Command Aborted By host" status code

2019-12-31T15:44:40+00:00

[ Upstream commit 2dc3947b53f573e8a75ea9cbec5588df88ca502e ] Fix the status code of canceled requests initiated by the host according to TP4028 (Status Code 0x371): "Command Aborted By host: The command was aborted as a result of host action (e.g., the host disconnected the Fabric connection)." Also in a multipath environment, unless otherwise specified, errors of this type (path related) should be retried using a different path, if one is available. Signed-off-by: Max Gurtovoy Reviewed-by: Christoph Hellwig Signed-off-by: Keith Busch Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

ipmi: Don't allow device module unload when in use

2019-12-31T15:44:14+00:00

[ Upstream commit cbb79863fc3175ed5ac506465948b02a893a8235 ] If something has the IPMI driver open, don't allow the device module to be unloaded. Before it would unload and the user would get errors on use. This change is made on user request, and it makes it consistent with the I2C driver, which has the same behavior. It does change things a little bit with respect to kernel users. If the ACPI or IPMI watchdog (or any other kernel user) has created a user, then the device module cannot be unloaded. Before it could be unloaded, This does not affect hot-plug. If the device goes away (it's on something removable that is removed or is hot-removed via sysfs) then it still behaves as it did before. Reported-by: tony camuso Signed-off-by: Corey Minyard Tested-by: tony camuso Signed-off-by: Sasha Levin

block: Fix writeback throttling W=1 compiler warnings

2019-12-31T15:43:47+00:00

[ Upstream commit 1d200e9d6f635ae894993a7d0f1b9e0b6e522e3b ] Fix the following compiler warnings: In file included from ./include/linux/bitmap.h:9, from ./include/linux/cpumask.h:12, from ./arch/x86/include/asm/cpumask.h:5, from ./arch/x86/include/asm/msr.h:11, from ./arch/x86/include/asm/processor.h:21, from ./arch/x86/include/asm/cpufeature.h:5, from ./arch/x86/include/asm/thread_info.h:53, from ./include/linux/thread_info.h:38, from ./arch/x86/include/asm/preempt.h:7, from ./include/linux/preempt.h:78, from ./include/linux/spinlock.h:51, from ./include/linux/mmzone.h:8, from ./include/linux/gfp.h:6, from ./include/linux/mm.h:10, from ./include/linux/bvec.h:13, from ./include/linux/blk_types.h:10, from block/blk-wbt.c:23: In function 'strncpy', inlined from 'perf_trace_wbt_stat' at ./include/trace/events/wbt.h:15:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'strncpy', inlined from 'perf_trace_wbt_lat' at ./include/trace/events/wbt.h:58:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'strncpy', inlined from 'perf_trace_wbt_step' at ./include/trace/events/wbt.h:87:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'strncpy', inlined from 'perf_trace_wbt_timer' at ./include/trace/events/wbt.h:126:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'strncpy', inlined from 'trace_event_raw_event_wbt_stat' at ./include/trace/events/wbt.h:15:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'strncpy', inlined from 'trace_event_raw_event_wbt_lat' at ./include/trace/events/wbt.h:58:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'strncpy', inlined from 'trace_event_raw_event_wbt_timer' at ./include/trace/events/wbt.h:126:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'strncpy', inlined from 'trace_event_raw_event_wbt_step' at ./include/trace/events/wbt.h:87:1: ./include/linux/string.h:260:9: warning: '__builtin_strncpy' specified bound 32 equals destination size [-Wstringop-truncation] return __builtin_strncpy(p, q, size); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Cc: Christoph Hellwig Cc: Ming Lei Cc: Hannes Reinecke Cc: Johannes Thumshirn Fixes: e34cbd307477 ("blk-wbt: add general throttling mechanism"; v4.10). Signed-off-by: Bart Van Assche Signed-off-by: Jens Axboe Signed-off-by: Sasha Levin

media: cec-funcs.h: add status_req checks

2019-12-31T15:43:43+00:00

[ Upstream commit 9b211f9c5a0b67afc435b86f75d78273b97db1c5 ] The CEC_MSG_GIVE_DECK_STATUS and CEC_MSG_GIVE_TUNER_DEVICE_STATUS commands both have a status_req argument: ON, OFF, ONCE. If ON or ONCE, then the follower will reply with a STATUS message. Either once or whenever the status changes (status_req == ON). If status_req == OFF, then it will stop sending continuous status updates, but the follower will *not* send a STATUS message in that case. This means that if status_req == OFF, then msg->reply should be 0 as well since no reply is expected in that case. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Sasha Levin

drm: mst: Fix query_payload ack reply struct

2019-12-31T15:42:19+00:00

[ Upstream commit 268de6530aa18fe5773062367fd119f0045f6e88 ] Spec says[1] Allocated_PBN is 16 bits [1]- DisplayPort 1.2 Spec, Section 2.11.9.8, Table 2-98 Fixes: ad7f8a1f9ced ("drm/helper: add Displayport multi-stream helper (v0.6)") Cc: Lyude Paul Cc: Todd Previte Cc: Dave Airlie Cc: Maarten Lankhorst Cc: Maxime Ripard Cc: Sean Paul Cc: David Airlie Cc: Daniel Vetter Cc: dri-devel@lists.freedesktop.org Reviewed-by: Lyude Paul Signed-off-by: Sean Paul Link: https://patchwork.freedesktop.org/patch/msgid/20190829165223.129662-1-sean@poorly.run Signed-off-by: Sasha Levin

neighbour: remove neigh_cleanup() method

2019-12-31T15:41:37+00:00

[ Upstream commit f394722fb0d0f701119368959d7cd0ecbc46363a ] neigh_cleanup() has not been used for seven years, and was a wrong design. Messing with shared pointer in bond_neigh_init() without proper memory barriers would at least trigger syzbot complains eventually. It is time to remove this stuff. Fixes: b63b70d87741 ("IPoIB: Use a private hash table for path lookup in xmit path") Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

net: phy: ensure that phy IDs are correctly typed

2019-12-31T15:41:26+00:00

[ Upstream commit 7d49a32a66d2215c5b3bf9bc67c9036ea9904111 ] PHY IDs are 32-bit unsigned quantities. Ensure that they are always treated as such, and not passed around as "int"s. Fixes: 13d0ab6750b2 ("net: phy: check return code when requesting PHY driver module") Signed-off-by: Russell King Reviewed-by: Florian Fainelli Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman