It was reported that 'perf stat' crashed when using with armv8_pmu (CPU)
events with the task mode. As 'perf stat' uses an empty cpu map for
task mode but armv8_pmu has its own cpu mask, it has confused which map
it should use when accessing file descriptors and this causes segfaults:
(gdb) bt
#0 0x0000000000603fc8 in perf_evsel__close_fd_cpu (evsel=<optimized out>,
cpu=<optimized out>) at evsel.c:122
#1 perf_evsel__close_cpu (evsel=evsel@entry=0x716e950, cpu=7) at evsel.c:156
#2 0x00000000004d4718 in evlist__close (evlist=0x70a7cb0) at util/evlist.c:1242
#3 0x0000000000453404 in __run_perf_stat (argc=3, argc@entry=1, argv=0x30,
argv@entry=0xfffffaea2f90, run_idx=119, run_idx@entry=1701998435)
at builtin-stat.c:929
#4 0x0000000000455058 in run_perf_stat (run_idx=1701998435, argv=0xfffffaea2f90,
argc=1) at builtin-stat.c:947
#5 cmd_stat (argc=1, argv=0xfffffaea2f90) at builtin-stat.c:2357
#6 0x00000000004bb888 in run_builtin (p=p@entry=0x9764b8 <commands+288>,
argc=argc@entry=4, argv=argv@entry=0xfffffaea2f90) at perf.c:312
#7 0x00000000004bbb54 in handle_internal_command (argc=argc@entry=4,
argv=argv@entry=0xfffffaea2f90) at perf.c:364
#8 0x0000000000435378 in run_argv (argcp=<synthetic pointer>,
argv=<synthetic pointer>) at perf.c:408
#9 main (argc=4, argv=0xfffffaea2f90) at perf.c:538
To fix this, I simply used the given cpu map unless the evsel actually
is not a system-wide event (like uncore events).
Fixes: 7736627b865d ("perf stat: Use affinity for closing file descriptors") Reported-by: Wei Li <liwei391@huawei.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org> Tested-by: Barry Song <song.bao.hua@hisilicon.com> Acked-by: Jiri Olsa <jolsa@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lore.kernel.org/lkml/20201007081311.1831003-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 9e9f60108423f ("powerpc/perf/{hv-gpci, hv-common}: generate
requests with counters annotated") adds a framework for defining
gpci counters.
In this patch, they adds starting_index value as '0xffffffffffffffff'.
which is wrong as starting_index is of size 32 bits.
Because of this, incase we try to run hv-gpci event we get error.
In power9 machine:
command#: perf stat -e hv_gpci/system_tlbie_count_and_time_tlbie_instructions_issued/
-C 0 -I 1000
event syntax error: '..bie_count_and_time_tlbie_instructions_issued/'
\___ value too big for format, maximum is 4294967295
This patch fix this issue and changes starting_index value to '0xffffffff'
PMU counter support functions enforces event constraints for group of
events to check if all events in a group can be monitored. Incase of
event codes using PMC5 and PMC6 ( 500fa and 600f4 respectively ), not
all constraints are applicable, say the threshold or sample bits. But
current code includes pmc5 and pmc6 in some group constraints (like
IC_DC Qualifier bits) which is actually not applicable and hence
results in those events not getting counted when scheduled along with
group of other events. Patch fixes this by excluding PMC5/6 from
constraints which are not relevant for it.
Replayed interrupts get an "artificial" struct pt_regs constructed to
pass to interrupt handler functions. This did not get the softe field
set correctly, it's as though the interrupt has hit while irqs are
disabled. It should be IRQS_ENABLED.
This is possibly harmless, asynchronous handlers should not be testing
if irqs were disabled, but it might be possible for example some code
is shared with synchronous or NMI handlers, and it makes more sense if
debug output looks at this.
Fixes: 3282a3da25bd ("powerpc/64: Implement soft interrupt replay in C") Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200915114650.3980244-2-npiggin@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Prior to commit 3282a3da25bd ("powerpc/64: Implement soft interrupt
replay in C"), replayed interrupts returned by the regular interrupt
exit code, which performs preemption in case an interrupt had set
need_resched.
This logic was missed by the conversion. Adding preempt_disable/enable
around the interrupt replay and final irq enable will reschedule if
needed.
Fixes: 3282a3da25bd ("powerpc/64: Implement soft interrupt replay in C") Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200915114650.3980244-1-npiggin@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The various array_size functions use SIZE_MAX define, but missed limits.h
causes to failure to compile code that needs overflow.h.
In file included from drivers/infiniband/core/uverbs_std_types_device.c:6:
./include/linux/overflow.h: In function 'array_size':
./include/linux/overflow.h:258:10: error: 'SIZE_MAX' undeclared (first use in this function)
258 | return SIZE_MAX;
| ^~~~~~~~
Currently using forward search doesn't handle multi-line strings correctly.
The search routine replaces line breaks with \0 during the search and, for
regular searches ("help | grep Common\n"), there is code after the line
has been discarded or printed to replace the break character.
However during a pager search ("help\n" followed by "/Common\n") when the
string is matched we will immediately return to normal output and the code
that should restore the \n becomes unreachable. Fix this by restoring the
replaced character when we disable the search mode and update the comment
accordingly.
Fixes: fb6daa7520f9d ("kdb: Provide forward search at more prompt") Link: https://lore.kernel.org/r/20200909141708.338273-1-daniel.thompson@linaro.org Reviewed-by: Douglas Anderson <dianders@chromium.org> Signed-off-by: Daniel Thompson <daniel.thompson@linaro.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Commit 7c2f66a960fc ("mtd: rawnand: ams-delta: Add module device
tables") introduced an OF module device table but wrapped a reference
to it with of_match_ptr() which resolves to NULL in non-OF configs.
That resulted in a clang compiler warning on unused variable in non-OF
builds. Fix it.
The following GigaDevice chips have the QE BIT in the feature flags, I
checked the datasheets, but did not try this.
* GD5F1GQ4xExxG
* GD5F1GQ4xFxxG
* GD5F1GQ4UAYIG
* GD5F4GQ4UAYIG
The Quad operations like 0xEB mention that the QE bit has to be set.
The hardware will add AckReq flag in BTH header according to the value of
ack_req_freq to request ACK from responder for the packets with this flag.
It should be greater than or equal to lp_pktn_ini instead of using a fixed
value.
The rnr_retry returned to the user is not correct, it should be got from
another fields in QPC.
Fixes: bfe860351e31 ("RDMA/hns: Fix cast from or to restricted __le32 for driver") Link: https://lore.kernel.org/r/1600509802-44382-7-git-send-email-liweihang@huawei.com Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
calc_pg_sz() may gets a data calculation overflow if the PAGE_SIZE is 64 KB
and hop_num is 2. It is because that all variables involved in calculation
are defined in type of int. So change the type of bt_chunk_size,
buf_chunk_size and obj_per_chunk_default to u64.
Fixes: ba6bb7e97421 ("RDMA/hns: Add interfaces to get pf capabilities from firmware") Link: https://lore.kernel.org/r/1600509802-44382-6-git-send-email-liweihang@huawei.com Signed-off-by: Jiaran Zhang <zhangjiaran@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
According to the RoCE v1 specification, the sl (service level) 0-7 are
mapped directly to priorities 0-7 respectively, sl 8-15 are reserved. The
driver should verify whether the the value of sl is larger than 7, if so,
an exception should be returned.
Fixes: 926a01dc000d ("RDMA/hns: Add QP operations support for hip08 SoC") Link: https://lore.kernel.org/r/1600509802-44382-5-git-send-email-liweihang@huawei.com Signed-off-by: Jiaran Zhang <zhangjiaran@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Some metrics (such as DRAM_BW_Use) consists of uncore events and
duration_time. For uncore events, counter->core.system_wide is true. But
for duration_time, counter->core.system_wide is false so
target.system_wide is set to false.
Then 'enable_on_exec' is set in perf_event_attr of uncore event. Kernel
will return error when trying to open the uncore event.
This patch skips the duration_time in setup_system_wide then
target.system_wide will be set to true for the evlist of uncore events +
duration_time.
Before (tested on skylake desktop):
# perf stat -M DRAM_BW_Use -- sleep 1
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (arb/event=0x84,umask=0x1/).
/bin/dmesg | grep -i perf may provide additional information.
Fixes: e3ba76deef23064f ("perf tools: Force uncore events to system wide monitoring") Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jin Yao <yao.jin@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lore.kernel.org/lkml/20200922015004.30114-1-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Occasionally ib_write_bw crash is seen due to access of a pd object in
i40iw_sc_qp_destroy after it is freed. Destroy qp is not synchronous in
i40iw and thus the iwqp object could be referencing a pd object that is
freed by ib core as a result of successful return from i40iw_destroy_qp.
Wait in i40iw_destroy_qp till all QP references are released and destroy
the QP and its associated resources before returning. Switch to use the
refcount API vs atomic API for lifetime management of the qp.
set_reg_wr() always fails if !umr_modify_entity_size_disabled because
mlx5_ib_can_use_umr() always fails. Without set_reg_wr() IB_WR_REG_MR
doesn't work and that means the device should not advertise
IB_DEVICE_MEM_MGT_EXTENSIONS.
Fixes: 841b07f99a47 ("IB/mlx5: Block MR WR if UMR is not possible") Link: https://lore.kernel.org/r/20200914112653.345244-5-leon@kernel.org Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Any mkey that is not enabled and assigned to userspace should have the PD
set to a kernel owned PD.
When cache entries are created for the first time the PDN is set to 0,
which is probably a kernel PD, but be explicit.
When a MR is registered using the hybrid reg_create with UMR xlt & enable
the disabled mkey is pointing at the user PD, keep it pointing at the
kernel until a UMR enables it and sets the user PD.
Fixes: 9ec4483a3f0f ("IB/mlx5: Move MRs to a kernel PD when freeing them to the MR cache") Link: https://lore.kernel.org/r/20200914112653.345244-4-leon@kernel.org Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
A context_switch event can have no tid because pids can be detached from
a task while the task is still running (in do_exit()). Note this won't
happen with per-task contexts because then tracing stops at
perf_event_exit_task()
If a task with no tid gets preempted, or a dying task gets preempted and
its parent releases it, when it subsequently gets switched back in,
Intel PT will not be able to determine what task is running and prints
an error "context_switch event has no tid". However, it is not really an
error because the task is in kernel space and the decoder can continue
to decode successfully. Fix by changing the error to be only a logged
message, and make allowance for tid == -1.
Example:
Using 5.9-rc4 with Preemptible Kernel (Low-Latency Desktop) e.g.
$ uname -r
5.9.0-rc4
$ grep PREEMPT .config
# CONFIG_PREEMPT_NONE is not set
# CONFIG_PREEMPT_VOLUNTARY is not set
CONFIG_PREEMPT=y
CONFIG_PREEMPT_COUNT=y
CONFIG_PREEMPTION=y
CONFIG_PREEMPT_RCU=y
CONFIG_PREEMPT_NOTIFIERS=y
CONFIG_DRM_I915_PREEMPT_TIMEOUT=640
CONFIG_DEBUG_PREEMPT=y
# CONFIG_PREEMPT_TRACER is not set
# CONFIG_PREEMPTIRQ_DELAY_TEST is not set
The roce path triggers a work queue that continues to touch the id_priv
but doesn't hold any reference on it. Futher, unlike in the IB case, the
work queue is not fenced during rdma_destroy_id().
This can trigger a use after free if a destroy is triggered in the
incredibly narrow window after the queue_work and the work starting and
obtaining the handler_mutex.
The only purpose of this work queue is to run the ULP event callback from
the standard context, so switch the design to use the existing
cma_work_handler() scheme. This simplifies quite a lot of the flow:
- Use the cma_work_handler() callback to launch the work for roce. This
requires generating the event synchronously inside the
rdma_join_multicast(), which in turn means the dummy struct
ib_sa_multicast can become a simple stack variable.
- cm_work_handler() used the id_priv kref, so we can entirely eliminate
the kref inside struct cma_multicast. Since the cma_multicast never
leaks into an unprotected work queue the kfree can be done at the same
time as for IB.
- Eliminating the general multicast.ib requires using cma_set_mgid() in a
few places to recompute the mgid.
Fixes: 3c86aa70bf67 ("RDMA/cm: Add RDMA CM support for IBoE devices") Link: https://lore.kernel.org/r/20200902081122.745412-9-leon@kernel.org Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
These are the same thing, except that cma_ndev_work doesn't have a state
transition. Signal no state transition by setting old_state and new_state
== 0.
In all cases the handler function should not be called once
rdma_destroy_id() has progressed passed setting the state.
Commit 0cef77c7798a7 ("powerpc/64s/radix: flush remote CPUs out of
single-threaded mm_cpumask") added a mechanism to trim the mm_cpumask of
a process under certain conditions. One of the assumptions is that
mm_users would not be incremented via a reference outside the process
context with mmget_not_zero() then go on to kthread_use_mm() via that
reference.
That invariant was broken by io_uring code (see previous sparc64 fix),
but I'll point Fixes: to the original powerpc commit because we are
changing that assumption going forward, so this will make backports
match up.
Fix this by no longer relying on that assumption, but by having each CPU
check the mm is not being used, and clearing their own bit from the mask
only if it hasn't been switched-to by the time the IPI is processed.
This relies on commit 38cf307c1f20 ("mm: fix kthread_use_mm() vs TLB
invalidate") and ARCH_WANT_IRQS_OFF_ACTIVATE_MM to disable irqs over mm
switch sequences.
Fixes: 0cef77c7798a7 ("powerpc/64s/radix: flush remote CPUs out of single-threaded mm_cpumask") Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Reviewed-by: Michael Ellerman <mpe@ellerman.id.au>
Depends-on: 38cf307c1f20 ("mm: fix kthread_use_mm() vs TLB invalidate") Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200914045219.3736466-5-npiggin@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Before the commit identified below, pages tables allocation was
performed after the allocation of final shadow area for linear memory.
But that commit switched the order, leading to page tables being
already allocated at the time 8xx kasan_init_shadow_8M() is called.
Due to this, kasan_init_shadow_8M() doesn't map the needed
shadow entries because there are already page tables.
kasan_init_shadow_8M() installs huge PMD entries instead of page
tables. We could at that time free the page tables, but there is no
point in creating page tables that get freed before being used.
Only book3s/32 hash needs early allocation of page tables. For other
variants, we can keep the initial order and create remaining page
tables after the allocation of final shadow memory for linear mem.
Move back the allocation of shadow page tables for
CONFIG_KASAN_VMALLOC into kasan_init() after the loop which creates
final shadow memory for linear mem.
Resolve this problem by disabling each THRMn comparator when handling
the associated THRMn interrupt and by disabling the TAU entirely when
updating THRMn thresholds.
According to Freescale's documentation, MPC74XX processors have an
erratum that prevents the TAU interrupt from working, so don't try to
use it when running on those processors.
The commentary at the call site seems to disagree with the code. The
conditional prevents calling set_thresholds() via the exception handler,
which appears to crash. Perhaps that's because it immediately triggers
another TAU exception. Anyway, calling set_thresholds() from TAUupdate()
is redundant because tau_timeout() does so.
Since commit 19dbdcb8039cf ("smp: Warn on function calls from softirq
context") the Thermal Assist Unit driver causes a warning like the
following when CONFIG_SMP is enabled.
According to the MPC750 Users Manual, the SITV value in Thermal
Management Register 3 is 13 bits long. The present code calculates the
SITV value as 60 * 500 cycles. This would overflow to give 10 us on
a 500 MHz CPU rather than the intended 60 us. (But according to the
Microprocessor Datasheet, there is also a factor of 266 that has to be
applied to this value on certain parts i.e. speed sort above 266 MHz.)
Always use the maximum cycle count, as recommended by the Datasheet.
With commit: 0034d395f89d ("powerpc/mm/hash64: Map all the kernel
regions in the same 0xc range"), we now split the 64TB address range
into 4 contexts each of 16TB. That implies we can do only 16TB linear
mapping.
On some systems, eg. Power9, memory attached to nodes > 0 will appear
above 16TB in the linear mapping. This resulted in kernel crash when
we boot such systems in hash translation mode with 4K PAGE_SIZE.
This patch updates the kernel mapping such that we now start supporting upto
61TB of memory with 4K. The kernel mapping now looks like below 4K PAGE_SIZE
and hash translation.
Our MAX_PHYSMEM_BITS for 4K is still 64TB even though we can only map 61TB.
We prevent bolt mapping anything outside 61TB range by checking against
H_VMALLOC_START.
Fixes: 0034d395f89d ("powerpc/mm/hash64: Map all the kernel regions in the same 0xc range") Reported-by: Cameron Berkenpas <cam@neo-zeon.de> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200608070904.387440-3-aneesh.kumar@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
There are couple of places where we set len but not hw_len. For
ptrace/perf watchpoints, when CONFIG_HAVE_HW_BREAKPOINT=Y, hw_len
will be calculated and set internally while parsing watchpoint.
But when CONFIG_HAVE_HW_BREAKPOINT=N, we need to manually set
'hw_len'. Similarly for xmon as well, hw_len needs to be set
directly.
Fixes: b57aeab811db ("powerpc/watchpoint: Fix length calculation for unaligned target") Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200902042945.129369-7-ravi.bangoria@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Vector load/store instructions are special because they are always
aligned. Thus unaligned EA needs to be aligned down before comparing
it with watch ranges. Otherwise we might consider valid event as
invalid.
Fixes: 74c6881019b7 ("powerpc/watchpoint: Prepare handler to handle more than one watchpoint") Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200902042945.129369-3-ravi.bangoria@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
On p10 predecessors, watchpoint with quadword access is compared at
quadword length. If the watch range is doubleword or less than that
in a first half of quadword aligned 16 bytes, and if there is any
unaligned quadword access which will access only the 2nd half, the
handler should consider it as extraneous and emulate/single-step it
before continuing.
Fixes: 74c6881019b7 ("powerpc/watchpoint: Prepare handler to handle more than one watchpoint") Reported-by: Pedro Miraglia Franco de Carvalho <pedromfc@linux.ibm.com> Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200902042945.129369-2-ravi.bangoria@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
POWER secure guests (i.e., guests which use the Protected Execution
Facility) need to use SWIOTLB to be able to do I/O with the
hypervisor, but they don't need the SWIOTLB memory to be in low
addresses since the hypervisor doesn't have any addressing limitation.
This solves a SWIOTLB initialization problem we are seeing in secure
guests with 128 GB of RAM: they are configured with 4 GB of
crashkernel reserved memory, which leaves no space for SWIOTLB in low
addresses.
To do this, we use mostly the same code as swiotlb_init(), but
allocate the buffer using memblock_alloc() instead of
memblock_alloc_low().
Fixes: 2efbc58f157a ("powerpc/pseries/svm: Force SWIOTLB for secure guests") Signed-off-by: Thiago Jung Bauermann <bauerman@linux.ibm.com> Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200818221126.391073-1-bauerman@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 59e8970b3798 ("RDMA/qedr: Return max inline data in QP query
result") changed query_qp max_inline size to return the max roce inline
size. When iwarp was introduced, this should have been modified to return
the max inline size based on protocol. This size is cached in the device
attributes
Fixes: 69ad0e7fe845 ("RDMA/qedr: Add support for iWARP in user space") Link: https://lore.kernel.org/r/20200902165741.8355-8-michal.kalderon@marvell.com Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Change the doorbell setting so that the maximum value between the last and
current value is set. This is to avoid doorbells being lost.
Fixes: a7efd7773e31 ("qedr: Add support for PD,PKEY and CQ verbs") Link: https://lore.kernel.org/r/20200902165741.8355-3-michal.kalderon@marvell.com Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
rdma_for_each_block() makes assumptions about how the SGL is constructed
that don't work if the block size is below the page size used to to build
the SGL.
The rules for umem SGL construction require that the SG's all be PAGE_SIZE
aligned and we don't encode the actual byte offset of the VA range inside
the SGL using offset and length. So rdma_for_each_block() has no idea
where the actual starting/ending point is to compute the first/last block
boundary if the starting address should be within a SGL.
Fixing the SGL construction turns out to be really hard, and will be the
subject of other patches. For now block smaller pages.
Fixes: 4a35339958f1 ("RDMA/umem: Add API to find best driver supported page size in an MR") Link: https://lore.kernel.org/r/2-v2-270386b7e60b+28f4-umem_1_jgg@nvidia.com Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
It is possible for a single SGL to span an aligned boundary, eg if the SGL
is
61440 -> 90112
Then the length is 28672, which currently limits the block size to
32k. With a 32k page size the two covering blocks will be:
32768->65536 and 65536->98304
However, the correct answer is a 128K block size which will span the whole
28672 bytes in a single block.
Instead of limiting based on length figure out which high IOVA bits don't
change between the start and end addresses. That is the highest useful
page size.
Fixes: 4a35339958f1 ("RDMA/umem: Add API to find best driver supported page size in an MR") Link: https://lore.kernel.org/r/1-v2-270386b7e60b+28f4-umem_1_jgg@nvidia.com Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Like any other verbs objects, CQ shouldn't fail during destroy, but
mlx5_ib didn't follow this contract with mixed IB verbs objects with
DEVX. Such mix causes to the situation where FW and kernel are fully
interdependent on the reference counting of each side.
Kernel verbs and drivers that don't have DEVX flows shouldn't fail.
Fixes: e39afe3d6dbd ("RDMA: Convert CQ allocations to be under core responsibility") Link: https://lore.kernel.org/r/20200907120921.476363-7-leon@kernel.org Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The ib_alloc_cq*() and ib_free_cq*() are solely kernel verbs to manage CQs
and doesn't need extra indirection just to call same functions with
constant parameter NULL as udata.
The rnbd_server module's communication manager (cm) initialization depends
on the registration of the "network namespace subsystem" of the RDMA CM
agent module. As such, when the kernel is configured to load the
rnbd_server and the RDMA cma module during initialization; and if the
rnbd_server module is initialized before RDMA cma module, a null ptr
dereference occurs during the RDMA bind operation.
All this happens cause the cm init is in the call chain of the module
init, which is not a preferred practice.
So remove the call to rdma_create_id() from the module init call chain.
Instead register rtrs-srv as an ib client, which makes sure that the
rdma_create_id() is called only when an ib device is added.
Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality") Link: https://lore.kernel.org/r/20200907103106.104530-1-haris.iqbal@cloud.ionos.com Reported-by: kernel test robot <rong.a.chen@intel.com> Signed-off-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Reviewed-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix some off-by-one errors in xfs_rtalloc_query_range. The highest key
in the realtime bitmap is always one less than the number of rt extents,
which means that the key clamp at the start of the function is wrong.
The 4th argument to xfs_rtfind_forw is the highest rt extent that we
want to probe, which means that passing 1 less than the high key is
wrong. Finally, drop the rem variable that controls the loop because we
can compare the iteration point (rtstart) against the high key directly.
The sordid history of this function is that the original commit (fb3c3)
incorrectly passed (high_rec->ar_startblock - 1) as the 'limit' parameter
to xfs_rtfind_forw. This was wrong because the "high key" is supposed
to be the largest key for which the caller wants result rows, not the
key for the first row that could possibly be outside the range that the
caller wants to see.
A subsequent attempt (8ad56) to strengthen the parameter checking added
incorrect clamping of the parameters to the number of rt blocks in the
system (despite the bitmap functions all taking units of rt extents) to
avoid querying ranges past the end of rt bitmap file but failed to fix
the incorrect _rtfind_forw parameter. The original _rtfind_forw
parameter error then survived the conversion of the startblock and
blockcount fields to rt extents (a0e5c), and the most recent off-by-one
fix (a3a37) thought it was patching a problem when the end of the rt
volume is not in use, but none of these fixes actually solved the
original problem that the author was confused about the "limit" argument
to xfs_rtfind_forw.
Sadly, all four of these patches were written by this author and even
his own usage of this function and rt testing were inadequate to get
this fixed quickly.
Original-problem: fb3c3de2f65c ("xfs: add a couple of queries to iterate free extents in the rtbitmap") Not-fixed-by: 8ad560d2565e ("xfs: strengthen rtalloc query range checks") Not-fixed-by: a0e5c435babd ("xfs: fix xfs_rtalloc_rec units") Fixes: a3a374bf1889 ("xfs: fix off-by-one error in xfs_rtalloc_query_range") Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Chandan Babu R <chandanrlinux@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
"mount -o local_lock=posix..." was broken by the mount API conversion
due to the missing constant.
Fixes: e38bb238ed8c ("NFS: Convert mount option parsing to use functionality from fs_parser.h") Signed-off-by: Scott Mayhew <smayhew@redhat.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Refactor xfs_getfsmap to improve its performance: instead of indirectly
calling a function that copies one record to userspace at a time, create
a shadow buffer in the kernel and copy the whole array once at the end.
On the author's computer, this reduces the runtime on his /home by ~20%.
This also eliminates a deadlock when running GETFSMAP against the
realtime device. The current code locks the rtbitmap to create
fsmappings and copies them into userspace, having not released the
rtbitmap lock. If the userspace buffer is an mmap of a sparse file that
itself resides on the realtime device, the write page fault will recurse
into the fs for allocation, which will deadlock on the rtbitmap lock.
Fixes: 4c934c7dd60c ("xfs: report realtime space information via the rtbitmap") Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Chandan Babu R <chandanrlinux@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If userspace asked fsmap to count the number of entries, we cannot
return more than UINT_MAX entries because fmh_entries is u32.
Therefore, stop counting if we hit this limit or else we will waste time
to return truncated results.
Fixes: e89c041338ed ("xfs: implement the GETFSMAP ioctl") Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Chandan Babu R <chandanrlinux@gmail.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
If a bitmap needs to be allocated, and then by the time the thread
is scheduled to be run again all the indices which would satisfy the
allocation have been allocated then we would leak the allocation. Almost
impossible to hit in practice, but a trivial fix. Found by Coverity.
When ARC_SOC_HSDK is enabled and RESET_CONTROLLER is disabled, it results
in the following Kbuild warning:
WARNING: unmet direct dependencies detected for RESET_HSDK
Depends on [n]: RESET_CONTROLLER [=n] && HAS_IOMEM [=y] && (ARC_SOC_HSDK [=y] || COMPILE_TEST [=n])
Selected by [y]:
- ARC_SOC_HSDK [=y] && ISA_ARCV2 [=y]
The reason is that ARC_SOC_HSDK selects RESET_HSDK without depending on or
selecting RESET_CONTROLLER while RESET_HSDK is subordinate to
RESET_CONTROLLER.
Honor the kconfig menu hierarchy to remove kconfig dependency warnings.
The mere fact that the kernel has the MMC subsystem enabled (CONFIG_MMC
enabled) does not mean that the underlying hardware platform has the
SDHC hardware present. Within the ColdFire hardware defines that is
signified by MCFSDHC_BASE being defined with an address.
The platform data for the ColdFire parts is including the SDHC hardware
if CONFIG_MMC is enabled, instead of MCFSDHC_BASE. This means that if
you are compiling for a ColdFire target that does not support SDHC but
enable CONFIG_MMC you will fail to compile with errors like this:
arch/m68k/coldfire/device.c:565:12: error: ‘MCFSDHC_BASE’ undeclared here (not in a function)
.start = MCFSDHC_BASE,
^
arch/m68k/coldfire/device.c:566:25: error: ‘MCFSDHC_SIZE’ undeclared here (not in a function)
.end = MCFSDHC_BASE + MCFSDHC_SIZE - 1,
^
arch/m68k/coldfire/device.c:569:12: error: ‘MCF_IRQ_SDHC’ undeclared here (not in a function)
.start = MCF_IRQ_SDHC,
^
Make the SDHC platform support depend on MCFSDHC_BASE, that is only
include it if the specific ColdFire SoC has that hardware module.
Nathan popped up on #xfs and pointed out that we fail to handle
finobt btree blocks in xlog_recover_get_buf_lsn(). This means they
always fall through the entire magic number matching code to "recover
immediately". Whilst most of the time this is the correct behaviour,
occasionally it will be incorrect and could potentially overwrite
more recent metadata because we don't check the LSN in the on disk
metadata at all.
This bug has been present since the finobt was first introduced, and
is a potential cause of the occasional xfs_iget_check_free_state()
failures we see that indicate that the inode btree state does not
match the on disk inode state.
Fixes: aafc3c246529 ("xfs: support the XFS_BTNUM_FINOBT free inode btree type") Reported-by: Nathan Scott <nathans@redhat.com> Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The L310_PREFETCH_CTRL register bits 28 and 29 to enable data and
instruction prefetch respectively can also be accessed via the
L2X0_AUX_CTRL register. They appear to be actually wired together in
hardware between the registers. Changing them in the prefetch
register only will get undone when restoring the aux control register
later on. For this reason, set these bits in both registers during
initialisation according to the devicetree property values.
We were failing that due to GTK2+ and then for the ZSTD test, which made
test-all.c, the fast path feature detection file to fail and thus
trigger building all of the feature tests, slowing down the test.
Eventually the ZSTD test would be built and would succeed, since it had
the needed -lzstd, avoiding:
$ cat /tmp/build/perf/feature/test-all.make.output
/usr/bin/ld: /tmp/ccRRJQ4u.o: in function `main_test_libzstd':
/home/acme/git/perf/tools/build/feature/test-libzstd.c:8: undefined reference to `ZSTD_createCStream'
/usr/bin/ld: /home/acme/git/perf/tools/build/feature/test-libzstd.c:9: undefined reference to `ZSTD_freeCStream'
collect2: error: ld returned 1 exit status
$
Fix it by adding -lzstd to the test-all target.
Now I need an entry to 'perf test' to make sure that
/tmp/build/perf/feature/test-all.make.output is empty...
Fixes: 3b1c5d9659718263 ("tools build: Implement libzstd feature check, LIBZSTD_DIR and NO_LIBZSTD defines") Reviewed-by: Alexei Budankov <alexey.budankov@linux.intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Link: http://lore.kernel.org/lkml/20200904202611.GJ3753976@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This is bitrotting, nobody is stepping up to work on it, and since we
treat warnings as errors, feature detection is failing in its main,
faster test (tools/build/feature/test-all.c) because of the GTK+2
infobar check.
So make this opt-in, at some point ditch this if nobody volunteers to
take care of this.
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
At memory hot-remove time we can retrieve an LMB's nid from its
corresponding memory_block. There is no need to store the nid
in multiple locations.
Note that lmb_to_memblock() uses find_memory_block() to get the
corresponding memory_block. As find_memory_block() runs in sub-linear
time this approach is negligibly slower than what we do at present.
In exchange for this lookup at hot-remove time we no longer need to
call memory_add_physaddr_to_nid() during drmem_init() for each LMB.
On powerpc, memory_add_physaddr_to_nid() is a linear search, so this
spares us an O(n^2) initialization during boot.
On systems with many LMBs that initialization overhead is palpable and
disruptive. For example, on a box with 249854 LMBs we're seeing
drmem_init() take upwards of 30 seconds to complete:
With a patched kernel on the same machine we're no longer seeing the
soft lockup. drmem_init() now completes in negligible time, even when
the LMB count is large.
Fixes: b2d3b5ee66f2 ("powerpc/pseries: Track LMB nid instead of using device tree") Signed-off-by: Scott Cheloha <cheloha@linux.ibm.com> Reviewed-by: Nathan Lynch <nathanl@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200811015115.63677-1-cheloha@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The drmem lmb list can have hundreds of thousands of entries, and
unfortunately lookups take the form of linear searches. As long as
this is the case, traversals have the potential to monopolize the CPU
and provoke lockup reports, workqueue stalls, and the like unless
they explicitly yield.
Rather than placing cond_resched() calls within various
for_each_drmem_lmb() loop blocks in the code, put it in the iteration
expression of the loop macro itself so users can't omit it.
Introduce a drmem_lmb_next() iteration helper function which calls
cond_resched() at a regular interval during array traversal. Each
iteration of the loop in DLPAR code paths can involve around ten RTAS
calls which can each take up to 250us, so this ensures the check is
performed at worst every few milliseconds.
Fixes: 6c6ea53725b3 ("powerpc/mm: Separate ibm, dynamic-memory data from DT format") Signed-off-by: Nathan Lynch <nathanl@linux.ibm.com> Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/20200813151131.2070161-1-nathanl@linux.ibm.com Signed-off-by: Sasha Levin <sashal@kernel.org>
The original function returns unsigned long and 0 on failure.
Fixes: 4a35339958f1 ("RDMA/umem: Add API to find best driver supported page size in an MR") Link: https://lore.kernel.org/r/0-v1-982a13cc5c6d+501ae-fix_best_pgsz_stub_jgg@nvidia.com Reviewed-by: Gal Pressman <galpress@amazon.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
It should be considered an illegal operation if the ULP attempts to modify
a QP from another state to the current hardware state. Otherwise, the ULP
can modify some fields of QPC at any time. For example, for a QP in state
of RTS, modify it from RTR to RTS can change the PSN, which is always not
as expected.
Fixes: 9a4435375cd1 ("IB/hns: Add driver files for hns RoCE driver") Link: https://lore.kernel.org/r/1598353674-24270-1-git-send-email-liweihang@huawei.com Signed-off-by: Lang Cheng <chenglang@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Building lpddr2_nvm with clang can result in a giant stack usage
in one function:
drivers/mtd/lpddr/lpddr2_nvm.c:399:12: error: stack frame size of 1144 bytes in function 'lpddr2_nvm_probe' [-Werror,-Wframe-larger-than=]
The problem is that clang decides to build a copy of the mtd_info
structure on the stack and then do a memcpy() into the actual version. It
shouldn't really do it that way, but it's not strictly a bug either.
As a workaround, use a static const version of the structure to assign
most of the members upfront and then only set the few members that
require runtime knowledge at probe time.
All entry points to the rdma_cm from a ULP must be single threaded,
even this error unwinds. Add the missing locking.
Fixes: 7c11910783a1 ("RDMA/ucma: Put a lock around every call to the rdma_cm layer") Link: https://lore.kernel.org/r/20200818120526.702120-11-leon@kernel.org Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
This value is locked under the file->mut, ensure it is held whenever
touching it.
The case in ucma_migrate_id() is a race, while in ucma_free_uctx() it is
already not possible for the write side to run, the movement is just for
clarity.
Fixes: 88314e4dda1e ("RDMA/cma: add support for rdma_migrate_id()") Link: https://lore.kernel.org/r/20200818120526.702120-10-leon@kernel.org Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The conversion of rcu_fwds to dynamic allocation failed to actually
allocate the required structure. This commit therefore allocates it,
frees it, and updates rcu_fwds accordingly. While in the area, it
abstracts the cleanup actions into rcu_torture_fwd_prog_cleanup().
Fixes: 5155be9994e5 ("rcutorture: Dynamically allocate rcu_fwds structure") Reported-by: kernel test robot <rong.a.chen@intel.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
On callback overload, it is necessary to quickly detect idle CPUs,
and rcu_gp_fqs_check_wake() checks for this condition. Unfortunately,
the code following the call to this function does not repeat this check,
which means that in reality no actual quiescent-state forcing, instead
only a couple of quick and pointless wakeups at the beginning of the
grace period.
This commit therefore adds a check for the RCU_GP_FLAG_OVLD flag in
the post-wakeup "if" statement in rcu_gp_fqs_loop().
Fixes: 1fca4d12f4637 ("rcu: Expedite first two FQS scans under callback-overload conditions") Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Neeraj Upadhyay <neeraju@codeaurora.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Both of_find_compatible_node() and of_find_node_by_type() will return
a refcounted node on success - thus for the success path the node must
be explicitly released with a of_node_put().
The call to of_find_compatible_node() returns a node pointer with
refcount incremented thus it must be explicitly decremented here
before returning.
Fixes: a489043f4626 ("powerpc/pseries: Implement arch_get_random_long() based on H_RANDOM") Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Link: https://lore.kernel.org/r/1530522496-14816-1-git-send-email-hofrat@osadl.org Signed-off-by: Sasha Levin <sashal@kernel.org>
The mlx4 driver will proxy MAD packets through the PF driver. A VM or an
instantiated VF will send its MAD packets to the PF driver using
loop-back. The PF driver will be informed by an interrupt, but defer the
handling and polling of CQEs to a worker thread running on an ordered
work-queue.
Consider the following scenario: the VMs will in short proximity in time,
for example due to a network event, send many MAD packets to the PF
driver. Lets say there are K VMs, each sending N packets.
The interrupt from the first VM will start the worker thread, which will
poll N CQEs. A common case here is where the PF driver will multiplex the
packets received from the VMs out on the wire QP.
But before the wire QP has returned a send CQE and associated interrupt,
the other K - 1 VMs have sent their N packets as well.
The PF driver has to multiplex K * N packets out on the wire QP. But the
send-queue on the wire QP has a finite capacity.
So, in this scenario, if K * N is larger than the send-queue capacity of
the wire QP, we will get MAD packets dropped on the floor with this
dynamic debug message:
mlx4_ib_multiplex_mad: failed sending GSI to wire on behalf of slave 2 (-11)
and this despite the fact that the wire send-queue could have capacity,
but the PF driver isn't aware, because the wire send CQEs have not yet
been polled.
We can also have a similar scenario inbound, with a wire recv-queue larger
than the tunnel QP's send-queue. If many remote peers send MAD packets to
the very same VM, the tunnel send-queue destined to the VM could allegedly
be construed to be full by the PF driver.
This starvation is fixed by introducing separate work queues for the wire
QPs vs. the tunnel QPs.
With this fix, using a dual ported HCA, 8 VFs instantiated, we could run
cmtime on each of the 18 interfaces towards a similar configured peer,
each cmtime instance with 800 QPs (all in all 14400 QPs) without a single
CM packet getting lost.
Boardinfo was lost if I3C object for devices with boardinfo
available are not created or not added to the I3C device list
because of some failure e.g. SETDASA failed, retrieve info failed etc
This patch adds i3c_master_attach_boardinfo which scan boardinfo list
in the master object and 'attach' it to the I3C device object.
This test uses waking+wakeup_latency as an event name, which doesn't
make sense since it includes an operator. Illegal names are now
detected by the synthetic event command parsing, which causes this
test to fail. Change the name to 'waking_plus_wakeup_latency' to
prevent this.
Fix data race in prepend_path() with re-reading mnt->mnt_ns twice
without holding the lock.
is_mounted() does check for NULL, but is_anon_ns(mnt->mnt_ns) might
re-read the pointer again which could be NULL already, if in between
reads one of kern_unmount()/kern_unmount_array()/umount_tree() sets
mnt->mnt_ns to NULL.
This is seen in production with the following stack trace:
Currently __set_oom_adj loops through all processes in the system to keep
oom_score_adj and oom_score_adj_min in sync between processes sharing
their mm. This is done for any task with more that one mm_users, which
includes processes with multiple threads (sharing mm and signals).
However for such processes the loop is unnecessary because their signal
structure is shared as well.
Android updates oom_score_adj whenever a tasks changes its role
(background/foreground/...) or binds to/unbinds from a service, making it
more/less important. Such operation can happen frequently. We noticed
that updates to oom_score_adj became more expensive and after further
investigation found out that the patch mentioned in "Fixes" introduced a
regression. Using Pixel 4 with a typical Android workload, write time to
oom_score_adj increased from ~3.57us to ~362us. Moreover this regression
linearly depends on the number of multi-threaded processes running on the
system.
Mark the mm with a new MMF_MULTIPROCESS flag bit when task is created with
(CLONE_VM && !CLONE_THREAD && !CLONE_VFORK). Change __set_oom_adj to use
MMF_MULTIPROCESS instead of mm_users to decide whether oom_score_adj
update should be synchronized between multiple processes. To prevent
races between clone() and __set_oom_adj(), when oom_score_adj of the
process being cloned might be modified from userspace, we use
oom_adj_mutex. Its scope is changed to global.
The combination of (CLONE_VM && !CLONE_THREAD) is rarely used except for
the case of vfork(). To prevent performance regressions of vfork(), we
skip taking oom_adj_mutex and setting MMF_MULTIPROCESS when CLONE_VFORK is
specified. Clearing the MMF_MULTIPROCESS flag (when the last process
sharing the mm exits) is left out of this patch to keep it simple and
because it is believed that this threading model is rare. Should there
ever be a need for optimizing that case as well, it can be done by hooking
into the exit path, likely following the mm_update_next_owner pattern.
With the combination of (CLONE_VM && !CLONE_THREAD && !CLONE_VFORK) being
quite rare, the regression is gone after the change is applied.
[surenb@google.com: v3] Link: https://lkml.kernel.org/r/20200902012558.2335613-1-surenb@google.com Fixes: 44a70adec910 ("mm, oom_adj: make sure processes sharing mm have same view of oom_score_adj") Reported-by: Tim Murray <timmurray@google.com> Suggested-by: Michal Hocko <mhocko@kernel.org> Signed-off-by: Suren Baghdasaryan <surenb@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Christian Brauner <christian.brauner@ubuntu.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Oleg Nesterov <oleg@redhat.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Eugene Syromiatnikov <esyr@redhat.com> Cc: Christian Kellner <christian@kellner.me> Cc: Adrian Reber <areber@redhat.com> Cc: Shakeel Butt <shakeelb@google.com> Cc: Aleksa Sarai <cyphar@cyphar.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Alexey Gladkov <gladkov.alexey@gmail.com> Cc: Michel Lespinasse <walken@google.com> Cc: Daniel Jordan <daniel.m.jordan@oracle.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Bernd Edlinger <bernd.edlinger@hotmail.de> Cc: John Johansen <john.johansen@canonical.com> Cc: Yafang Shao <laoar.shao@gmail.com> Link: https://lkml.kernel.org/r/20200824153036.3201505-1-surenb@google.com Debugged-by: Minchan Kim <minchan@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The code in mc_handle_swap_pte() checks for non_swap_entry() and returns
NULL before checking is_device_private_entry() so device private pages are
never handled. Fix this by checking for non_swap_entry() after handling
device private swap PTEs.
I assume the memory cgroup accounting would be off somehow when moving
a process to another memory cgroup. Currently, the device private page
is charged like a normal anonymous page when allocated and is uncharged
when the page is freed so I think that path is OK.
Signed-off-by: Ralph Campbell <rcampbell@nvidia.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Vladimir Davydov <vdavydov.dev@gmail.com> Cc: Jerome Glisse <jglisse@redhat.com> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Ira Weiny <ira.weiny@intel.com> Link: https://lkml.kernel.org/r/20201009215952.2726-1-rcampbell@nvidia.com
xFixes: c733a82874a7 ("mm/memcontrol: support MEMORY_DEVICE_PRIVATE") Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Dump vlan tag and proto for the usual vlan offload case if the
NF_LOG_MACDECODE flag is set on. Without this information the logging is
misleading as there is no reference to the VLAN header.
Fixes: 83e96d443b37 ("netfilter: log: split family specific code to nf_log_{ip,ip6,common}.c files") Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Calling skb_orphan() is unnecessary in the strp rcv handler because the skb
is from a skb_clone() in __strp_recv. So it never has a destructor or a
sk assigned. Plus its confusing to read because it might hint to the reader
that the skb could have an sk assigned which is not true. Even if we did
have an sk assigned it would be cleaner to simply wait for the upcoming
kfree_skb().
Additionally, move the comment about strparser clone up so its closer to
the logic it is describing and add to it so that it is more complete.
fq qdisc requires tstamp to be cleared in forwarding path
Reported-by: Evgeny B <abt-admin@mail.ru> Link: https://bugzilla.kernel.org/show_bug.cgi?id=209427 Suggested-by: Eric Dumazet <eric.dumazet@gmail.com> Fixes: 8203e2d844d3 ("net: clear skb->tstamp in forwarding paths") Fixes: fb420d5d91c1 ("tcp/fq: move back to CLOCK_MONOTONIC") Fixes: 80b14dee2bea ("net: Add a new socket option for a future transmit time.") Signed-off-by: Julian Anastasov <ja@ssi.bg> Reviewed-by: Simon Horman <horms@verge.net.au> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Amlogic SoC devices report the following errors frequently causing excessive
dmesg log spam and early log rotataion, although the errors appear to be
harmless as everything works fine:
[ 7.202702] panfrost ffe40000.gpu: error powering up gpu L2
[ 7.203760] panfrost ffe40000.gpu: error powering up gpu shader
ARM staff have advised increasing the timeout values to eliminate the errors
in most normal scenarios, and testing with several different G31/G52 devices
shows 20000 to be a reliable value.
Fixes: f3ba91228e8e ("drm/panfrost: Add initial panfrost driver") Suggested-by: Steven Price <steven.price@arm.com> Signed-off-by: Christian Hewitt <christianshewitt@gmail.com> Reviewed-by: Steven Price <steven.price@arm.com> Signed-off-by: Steven Price <steven.price@arm.com> Link: https://patchwork.freedesktop.org/patch/msgid/20201008141738.13560-1-christianshewitt@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org>
sdio.c:2403:3: warning: Attempt to free released memory
kfree(card->mpa_rx.buf);
^~~~~~~~~~~~~~~~~~~~~~~
When mwifiex_init_sdio() fails in its first call to
mwifiex_alloc_sdio_mpa_buffer, it falls back to calling it
again. If the second alloc of mpa_tx.buf fails, the error
handler will try to free the old, previously freed mpa_rx.buf.
Reviewing the code, it looks like a second double free would
happen with mwifiex_cleanup_sdio().
So set both pointers to NULL when they are freed.
Fixes: 5e6e3a92b9a4 ("wireless: mwifiex: initial commit for Marvell mwifiex driver") Signed-off-by: Tom Rix <trix@redhat.com> Reviewed-by: Brian Norris <briannorris@chromium.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org> Link: https://lore.kernel.org/r/20201004131931.29782-1-trix@redhat.com Signed-off-by: Sasha Levin <sashal@kernel.org>
Remove PSU EEPROM configuration for systems class equipped with
Mellanox chip Spectrume-2. Till now all the systems from this class
used few types of power units, all equipped with EEPROM device with
address space two bytes. Thus, all these devices have been handled by
EEPROM driver "24c32".
There is a new requirement is to support power unit replacement by "off
the shelf" device, matching electrical required parameters. Such device
could be equipped with different EEPROM type, which could be one byte
address space addressing or even could be not equipped with EEPROM.
In such case "24c32" will not work.
Fixes: 1bd42d94ccab ("platform/x86: mlx-platform: Add support for new 200G IB and Ethernet systems") Signed-off-by: Vadim Pasternak <vadimp@nvidia.com> Reviewed-by: Hans de Goede <hdegoede@redhat.com> Link: https://lore.kernel.org/r/20200923172053.26296-2-vadimp@nvidia.com Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
synth_field_size() returns either a positive size or an error (zero or
a negative value). However, the existing code assumes the only error
value is 0. It doesn't handle negative error codes, as it assigns
directly to field->size (a size_t; unsigned), thereby interpreting the
error code as a valid size instead.
Do the test before assignment to field->size.
[ axelrasmussen@google.com: changelog addition, first paragraph above ]
Link: https://lkml.kernel.org/r/9b6946d9776b2eeb43227678158196de1c3c6e1d.1601848695.git.zanussi@kernel.org Fixes: 4b147936fa50 (tracing: Add support for 'synthetic' events) Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org> Tested-by: Axel Rasmussen <axelrasmussen@google.com> Signed-off-by: Tom Zanussi <zanussi@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
dma_alloc_coherent() is called with a fixed SZ_2M size, but frees happen
with IOAT_CHUNK_SIZE. Recently, IOAT_CHUNK_SIZE was reduced to 512M but
the allocation did not change. To fix, change to using the
IOAT_CHUNK_SIZE define.
This was caught with the upcoming patchset for converting Intel platforms to the
dma-iommu implementation. It has a warning when the unmapped size differs from
the mapped size.
The be_fill_queue() function can only fail when "eq_vaddress" is NULL and
since it's non-NULL here that means the function call can't fail. But
imagine if it could, then in that situation we would want to store the
"paddr" so that dma memory can be released.
Link: https://lore.kernel.org/r/20200928091300.GD377727@mwanda Fixes: bfead3b2cb46 ("[SCSI] be2iscsi: Adding msix and mcc_rings V3") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Corrects drivers/target/target_core_user.c:688:6: warning: 'page' may be
used uninitialized.
Link: https://lore.kernel.org/r/20200924001920.43594-1-john.p.donnelly@oracle.com Fixes: 3c58f737231e ("scsi: target: tcmu: Optimize use of flush_dcache_page") Cc: Mike Christie <michael.christie@oracle.com> Acked-by: Mike Christie <michael.christie@oracle.com> Signed-off-by: John Donnelly <john.p.donnelly@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In DDMA mode if INTR OUT transfers mps not multiple of 4 then single packet
corresponds to single descriptor.
Descriptor limit set to mps and desc chain limit set to mps *
MAX_DMA_DESC_NUM_GENERIC. On that descriptors complete, to calculate
transfer size should be considered correction value for each descriptor.
In start request function, if "continue" is true then dma buffer address
should be incremmented by offset for all type of transfers, not only for
Control DATA_OUT transfers.
Fixes: cf77b5fb9b394 ("usb: dwc2: gadget: Transfer length limit checking for DDMA") Fixes: e02f9aa6119e0 ("usb: dwc2: gadget: EP 0 specific DDMA programming") Fixes: aa3e8bc81311e ("usb: dwc2: gadget: DDMA transfer start and complete") Signed-off-by: Minas Harutyunyan <hminas@synopsys.com> Signed-off-by: Felipe Balbi <balbi@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
When dumping wiphy information, we try to split the data into
many submessages, but for old userspace we still support the
old mode where this doesn't happen.
However, in this case we were not resetting our state correctly
and dumping multiple messages for each wiphy, which would have
broken such older userspace.
This was broken pretty much immediately afterwards because it
only worked in the original commit where non-split dumps didn't
have any more data than split dumps...
projects / thirdparty / kernel / stable.git / log
