| Age | Commit message (Collapse) | Author |
|---|
|
git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm into master
Pull device mapper fix from Mike Snitzer:
"A stable fix for DM integrity target's integrity recalculation that
gets skipped when resuming a device. This is a fix for a previous
stable@ fix"
* tag 'for-5.8/dm-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm integrity: fix integrity recalculation that is improperly skipped
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux into master
Pull i2c fixes from Wolfram Sang:
"Again some driver bugfixes and some documentation fixes"
* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: i2c-qcom-geni: Fix DMA transfer race
i2c: rcar: always clear ICSAR to avoid side effects
MAINTAINERS: i2c: at91: handover maintenance to Codrin Ciubotariu
i2c: drop duplicated word in the header file
i2c: cadence: Clear HOLD bit at correct time in Rx path
Revert "i2c: cadence: Fix the hold bit setting"
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc into master
Pull MMC fix from Ulf Hansson:
"Fix clock divider calculation in the ASPEED SDHCI controller"
* tag 'mmc-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:
mmc: sdhci-of-aspeed: Fix clock divider calculation
|
|
into master
Pull drm fixes from Dave Airlie:
"Quiet fixes, I may have a single regression fix follow up to this for
nouveau, but it might be next week, Ben was testing it a bit more .
Otherwise two amdgpu fixes, one lima and one sun4i:
amdgpu:
- Fix crash when overclocking VegaM
- Fix possible crash when editing dpm levels
sun4i:
- Fix inverted HPD result; fixes an earlier fix
lima:
- fix timeout during reset"
* tag 'drm-fixes-2020-07-24' of git://anongit.freedesktop.org/drm/drm:
drm/amdgpu: Fix NULL dereference in dpm sysfs handlers
drm/amd/powerplay: fix a crash when overclocking Vega M
drm/lima: fix wait pp reset timeout
drm: sun4i: hdmi: Fix inverted HPD result
|
|
Commit ed66f991bb19 ("module: Refactor section attr into bin attribute")
removed the 'name' field from 'struct module_sect_attr' triggering the
following error when invoking lx-symbols:
(gdb) lx-symbols
loading vmlinux
scanning for modules in linux/build
loading @0xffffffffc014f000: linux/build/drivers/net/tun.ko
Python Exception <class 'gdb.error'> There is no member named name.:
Error occurred in Python: There is no member named name.
This patch fixes the issue taking the module name from the 'struct
attribute'.
Fixes: ed66f991bb19 ("module: Refactor section attr into bin attribute")
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Jan Kiszka <jan.kiszka@siemens.com>
Reviewed-by: Kieran Bingham <kbingham@kernel.org>
Link: http://lkml.kernel.org/r/20200722102239.313231-1-sgarzare@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
To link KCOV to the kasan-dev@ mailing list.
Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Marco Elver <elver@google.com>
Link: http://lkml.kernel.org/r/5fa344db7ac4af2213049e5656c0f43d6ecaa379.1595331682.git.andreyknvl@google.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
The !ATOMIC_IOMAP version of io_maping_init_wc will always return
success, even when the ioremap fails.
Since the ATOMIC_IOMAP version returns NULL when the init fails, and
callers check for a NULL return on error this is unexpected.
During a device probe, where the ioremap failed, a crash can look like
this:
BUG: unable to handle page fault for address: 0000000000210000
#PF: supervisor write access in kernel mode
#PF: error_code(0x0002) - not-present page
Oops: 0002 [#1] PREEMPT SMP
CPU: 0 PID: 177 Comm:
RIP: 0010:fill_page_dma [i915]
gen8_ppgtt_create [i915]
i915_ppgtt_create [i915]
intel_gt_init [i915]
i915_gem_init [i915]
i915_driver_probe [i915]
pci_device_probe
really_probe
driver_probe_device
The remap failure occurred much earlier in the probe. If it had been
propagated, the driver would have exited with an error.
Return NULL on ioremap failure.
[akpm@linux-foundation.org: detect ioremap_wc() errors earlier]
Fixes: cafaf14a5d8f ("io-mapping: Always create a struct to hold metadata about the io-mapping")
Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200721171936.81563-1-michael.j.ruhl@intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Currently the basepath is removed only from the beginning of the string.
When the symbol is inlined and there's multiple line outputs of
addr2line, only the first line would have basepath removed.
Change to remove the basepath prefix from all lines.
Fixes: 31013836a71e ("scripts/decode_stacktrace: match basepath using shell prefix operator, not regex")
Co-developed-by: Shik Chen <shik@chromium.org>
Signed-off-by: Pi-Hsun Shih <pihsun@chromium.org>
Signed-off-by: Shik Chen <shik@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Cc: Sasha Levin <sashal@kernel.org>
Cc: Nicolas Boichat <drinkcat@chromium.org>
Cc: Jiri Slaby <jslaby@suse.cz>
Link: http://lkml.kernel.org/r/20200720082709.252805-1-pihsun@chromium.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
This is a regression introduced by the "migrate from ll_rw_block usage
to BIO" patch.
Squashfs packs structures on byte boundaries, and due to that the length
field (of the metadata block) may not be fully in the current block.
The new code rewrote and introduced a faulty check for that edge case.
Fixes: 93e72b3c612adcaca1 ("squashfs: migrate from ll_rw_block usage to BIO")
Reported-by: Bernd Amend <bernd.amend@gmail.com>
Signed-off-by: Phillip Lougher <phillip@squashfs.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Adrien Schildknecht <adrien+dev@schischi.me>
Cc: Guenter Roeck <groeck@chromium.org>
Cc: Daniel Rosenberg <drosen@google.com>
Link: http://lkml.kernel.org/r/20200717195536.16069-1-phillip@squashfs.org.uk
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Add an entry to correct my email addresses.
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20200708095414.12275-1-rppt@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
khugepaged has to drop mmap lock several times while collapsing a page.
The situation can change while the lock is dropped and we need to
re-validate that the VMA is still in place and the PMD is still subject
for collapse.
But we miss one corner case: while collapsing an anonymous pages the VMA
could be replaced with file VMA. If the file VMA doesn't have any
private pages we get NULL pointer dereference:
general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
anon_vma_lock_write include/linux/rmap.h:120 [inline]
collapse_huge_page mm/khugepaged.c:1110 [inline]
khugepaged_scan_pmd mm/khugepaged.c:1349 [inline]
khugepaged_scan_mm_slot mm/khugepaged.c:2110 [inline]
khugepaged_do_scan mm/khugepaged.c:2193 [inline]
khugepaged+0x3bba/0x5a10 mm/khugepaged.c:2238
The fix is to make sure that the VMA is anonymous in
hugepage_vma_revalidate(). The helper is only used for collapsing
anonymous pages.
Fixes: 99cb0dbd47a1 ("mm,thp: add read-only THP support for (non-shmem) FS")
Reported-by: syzbot+ed318e8b790ca72c5ad0@syzkaller.appspotmail.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Acked-by: Yang Shi <yang.shi@linux.alibaba.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200722121439.44328-1-kirill.shutemov@linux.intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
hugetlb_cma[0] can be NULL due to various reasons, for example, node0
has no memory. so NULL hugetlb_cma[0] doesn't necessarily mean cma is
not enabled. gigantic pages might have been reserved on other nodes.
This patch fixes possible double reservation and CMA leak.
[akpm@linux-foundation.org: fix CONFIG_CMA=n warning]
[sfr@canb.auug.org.au: better checks before using hugetlb_cma]
Link: http://lkml.kernel.org/r/20200721205716.6dbaa56b@canb.auug.org.au
Fixes: cf11e85fc08c ("mm: hugetlb: optionally allocate gigantic hugepages using cma")
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Jonathan Cameron <jonathan.cameron@huawei.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200710005726.36068-1-song.bao.hua@hisilicon.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
If the kmem_cache refcount is greater than one, we should not mark the
root kmem_cache as dying. If we mark the root kmem_cache dying
incorrectly, the non-root kmem_cache can never be destroyed. It
resulted in memory leak when memcg was destroyed. We can use the
following steps to reproduce.
1) Use kmem_cache_create() to create a new kmem_cache named A.
2) Coincidentally, the kmem_cache A is an alias for kmem_cache B,
so the refcount of B is just increased.
3) Use kmem_cache_destroy() to destroy the kmem_cache A, just
decrease the B's refcount but mark the B as dying.
4) Create a new memory cgroup and alloc memory from the kmem_cache
B. It leads to create a non-root kmem_cache for allocating memory.
5) When destroy the memory cgroup created in the step 4), the
non-root kmem_cache can never be destroyed.
If we repeat steps 4) and 5), this will cause a lot of memory leak. So
only when refcount reach zero, we mark the root kmem_cache as dying.
Fixes: 92ee383f6daa ("mm: fix race between kmem_cache destroy, create and deactivate")
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200716165103.83462-1-songmuchun@bytedance.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
It was hard to keep a test running, moving tasks between memcgs with
move_charge_at_immigrate, while swapping: mem_cgroup_id_get_many()'s
refcount is discovered to be 0 (supposedly impossible), so it is then
forced to REFCOUNT_SATURATED, and after thousands of warnings in quick
succession, the test is at last put out of misery by being OOM killed.
This is because of the way moved_swap accounting was saved up until the
task move gets completed in __mem_cgroup_clear_mc(), deferred from when
mem_cgroup_move_swap_account() actually exchanged old and new ids.
Concurrent activity can free up swap quicker than the task is scanned,
bringing id refcount down 0 (which should only be possible when
offlining).
Just skip that optimization: do that part of the accounting immediately.
Fixes: 615d66c37c75 ("mm: memcontrol: fix memcg id ref counter on swap charge move")
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/alpine.LSU.2.11.2007071431050.4726@eggly.anvils
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Prabhakar reported an OOPS inside mem_cgroup_get_nr_swap_pages()
function in a corner case seen on some arm64 boards when kdump kernel
runs with "cgroup_disable=memory" passed to the kdump kernel via
bootargs.
The root-cause behind the same is that currently mem_cgroup_swap_init()
function is implemented as a subsys_initcall() call instead of a
core_initcall(), this means 'cgroup_memory_noswap' still remains set to
the default value (false) even when memcg is disabled via
"cgroup_disable=memory" boot parameter.
This may result in premature OOPS inside mem_cgroup_get_nr_swap_pages()
function in corner cases:
Unable to handle kernel NULL pointer dereference at virtual address 0000000000000188
Mem abort info:
ESR = 0x96000006
EC = 0x25: DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
Data abort info:
ISV = 0, ISS = 0x00000006
CM = 0, WnR = 0
[0000000000000188] user address but active_mm is swapper
Internal error: Oops: 96000006 [#1] SMP
Modules linked in:
<..snip..>
Call trace:
mem_cgroup_get_nr_swap_pages+0x9c/0xf4
shrink_lruvec+0x404/0x4f8
shrink_node+0x1a8/0x688
do_try_to_free_pages+0xe8/0x448
try_to_free_pages+0x110/0x230
__alloc_pages_slowpath.constprop.106+0x2b8/0xb48
__alloc_pages_nodemask+0x2ac/0x2f8
alloc_page_interleave+0x20/0x90
alloc_pages_current+0xdc/0xf8
atomic_pool_expand+0x60/0x210
__dma_atomic_pool_init+0x50/0xa4
dma_atomic_pool_init+0xac/0x158
do_one_initcall+0x50/0x218
kernel_init_freeable+0x22c/0x2d0
kernel_init+0x18/0x110
ret_from_fork+0x10/0x18
Code: aa1403e3 91106000 97f82a27 14000011 (f940c663)
---[ end trace 9795948475817de4 ]---
Kernel panic - not syncing: Fatal exception
Rebooting in 10 seconds..
Fixes: eccb52e78809 ("mm: memcontrol: prepare swap controller setup for integration")
Reported-by: Prabhakar Kushwaha <pkushwaha@marvell.com>
Signed-off-by: Bhupesh Sharma <bhsharma@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: James Morse <james.morse@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Link: http://lkml.kernel.org/r/1593641660-13254-2-git-send-email-bhsharma@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
clang static analysis reports a garbage return
In file included from mm/memory.c:84:
mm/memory.c:1612:2: warning: Undefined or garbage value returned to caller [core.uninitialized.UndefReturn]
return err;
^~~~~~~~~~
The setting of err depends on a loop executing. So initialize err.
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20200703155354.29132-1-trix@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
After commit fdc85222d58e ("kernfs: kvmalloc xattr value instead of
kmalloc"), simple xattr entry is allocated with kvmalloc() instead of
kmalloc(), so we should release it with kvfree() instead of kfree().
Fixes: fdc85222d58e ("kernfs: kvmalloc xattr value instead of kmalloc")
Signed-off-by: Chengguang Xu <cgxu519@mykernel.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Daniel Xu <dxu@dxuuu.xyz>
Cc: Chris Down <chris@chrisdown.name>
Cc: Andreas Dilger <adilger@dilger.ca>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org> [5.7]
Link: http://lkml.kernel.org/r/20200704051608.15043-1-cgxu519@mykernel.net
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
VMA with VM_GROWSDOWN or VM_GROWSUP flag set can change their size under
mmap_read_lock(). It can lead to race with __do_munmap():
Thread A Thread B
__do_munmap()
detach_vmas_to_be_unmapped()
mmap_write_downgrade()
expand_downwards()
vma->vm_start = address;
// The VMA now overlaps with
// VMAs detached by the Thread A
// page fault populates expanded part
// of the VMA
unmap_region()
// Zaps pagetables partly
// populated by Thread B
Similar race exists for expand_upwards().
The fix is to avoid downgrading mmap_lock in __do_munmap() if detached
VMAs are next to VM_GROWSDOWN or VM_GROWSUP VMA.
[akpm@linux-foundation.org: s/mmap_sem/mmap_lock/ in comment]
Fixes: dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Yang Shi <yang.shi@linux.alibaba.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: <stable@vger.kernel.org> [4.20+]
Link: http://lkml.kernel.org/r/20200709105309.42495-1-kirill.shutemov@linux.intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
git://people.freedesktop.org/~gabbayo/linux into char-misc-next
Oded writes:
This tag contains the following changes for kernel 5.9-rc1:
- Remove rate limiters from GAUDI configuration (no longer needed).
- Set maximum amount of in-flight CS per ASIC type and increase
the maximum amount for GAUDI.
- Refactor signal/wait command submissions code
- Calculate trace frequency from PLLs to show accurate profiling data
- Rephrase error messages to make them more clear to the common user
- Add statistics of dropped CS (counter per possible reason for drop)
- Get ECC information from firmware
- Remove support for partial SoC reset in Gaudi
- Halt device CPU only when reset is certain to happen. Sometimes we abort
the reset procedure and in that case we can't leave device CPU in halt
mode.
- set each CQ to its own work queue to prevent a race between completions
on different CQs.
- Use queue pi/ci in order to determine queue occupancy. This is done to
make the code reusable between current and future ASICs.
- Add more validations for user inputs.
- Refactor PCIe controller configuration to make the code reusable between
current and future ASICs.
- Update firmware interface headers to latest version
- Move all common code to a dedicated common sub-folder
* tag 'misc-habanalabs-next-2020-07-24' of git://people.freedesktop.org/~gabbayo/linux: (28 commits)
habanalabs: Fix memory leak in error flow of context initialization
habanalabs: use no flags on MMU cache invalidation
habanalabs: enable device before hw_init()
habanalabs: create internal CB pool
habanalabs: update hl_boot_if.h from firmware
habanalabs: create common folder
habanalabs: check for DMA errors when clearing memory
habanalabs: verify queue can contain all cs jobs
habanalabs: Assign each CQ with its own work queue
habanalabs: halt device CPU only upon certain reset
habanalabs: remove unused hash
habanalabs: use queue pi/ci in order to determine queue occupancy
habanalabs: configure maximum queues per asic
habanalabs: remove soft-reset support from GAUDI
habanalabs: PCIe iATU refactoring
habanalabs: Extract ECC information from FW
habanalabs: Add dropped cs statistics info struct
habanalabs: extract cpu boot status lookup
habanalabs: rephrase error messages
habanalabs: Increase queues depth
...
|
|
Add a missing free of the cs_pending array in the error flow of context
initialization.
Fixes: c16d45f42b64 ("habanalabs: Use pending CS amount per ASIC")
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
gaudi_mmu_invalidate_cache() doesn't use the flags parameter, and thus
it can be set to 0 when the function is called in the gaudi only files.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Device is now enabled before the hw_init() because part of the
initialization requires communication with the device firmware to get
information that is required for the initialization itself
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
|
|
Create a device MMU-mapped internal command buffer pool, in order to allow
the driver to allocate CBs for the signal/wait operations
that are fetched by the queues when they are configured with the user's
address space ID.
We must pre-map this internal pool due to performance issues.
This pool is needed for future ASIC support and it is currently unused in
GOYA and GAUDI.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Update the boot interface file from the latest version from firmware.
Defines for secure boot were added.
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
|
|
For internal needs of our CI we need to move all the common code into a
common folder instead of putting them in the root folder of the driver.
Same applies to the common header files under include/
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
|
|
In GAUDI we use QMAN0 DMA for clearing the MMU memory region
at initialization. if this operation fails it places the DMA in an error
state and then when trying to initialize QMAN0 we fail and erroneously
assume its the QMAN that failed.
This commit adds a check and clear of such DMA errors at initialization so
we will have a better understanding of what went wrong.
Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
In order for the user to be aware of wrong inputs, we must return
error in case the amount of jobs per cs exceeds the corresponding
queue size.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
We identified a possible race during job completion when working
with a single multi-threaded work queue. In order to overcome this
race we suggest using a single threaded work queue per completion
queue, hence we guarantee jobs completion in order.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Currently the driver halts the device CPU in the halt engines function,
which halts all the engines of the ASIC. The problem is that if later on we
stop the reset process (due to inability to clean memory mappings in time),
the CPU will remain in halt mode. This creates many issues, such as
thermal/power control and FLR handling.
Therefore, move the halting of the device CPU to the very end of the reset
process, just before writing to the registers to initiate the reset. In
addition, the driver now needs to send a message to the device F/W to
disable it from sending interrupts to the host machine because during halt
engines function the driver disables the MSI/MSI-X interrupts.
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
|
|
Remove an old hash that is not in use anymore.
Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Instead of using the free slots amount on the compute CQ to determine
whether we can submit work to queues, use the queues pi/ci.
This is needed in future ASICs where we don't have CQ per queue.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Currently the amount of maximum queues is statically configured.
Using a static value is causing redundunt cycles when traversing
all queues and consumes more memory than actually needed.
In this patch we configure each asic with the exact number of
queues needed.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Soft-reset isn't supported in GAUDI. Remove the code that performs it and
print error in case the user wants to do it via sysfs.
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
|
|
Divide iATU initialization into inbound/outbound methods.
We must separate it in order to enable different match mode
per PCIe region.
In addition, added support for PCI address match mode.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
ECC (Error Correcting Code) interrupts are going to be handled
by the FW. Hence, we define an interface in which the driver can
obtain the relevant ECC information.
This information is needed for monitoring and can also lead
to a hard reset if ECC error is not correctable.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Add command submission statistics structure which can be obtained
through the info ioctl. Each drop counter describes the reason for
which the command submission was dropped.
This information is needed for the user to be aware of the specific
reason for which the submitted work was dropped. The user can then
utilize the driver more efficiently.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Extract detection of the cpu boot status to a function
to allow code reuse
Signed-off-by: Christine Gharzuzi <cgharzuzi@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
rephrase some error/warning/notice messages to make them more accessible to
ordinary users.
There is no need to print context ASID as the driver currently doesn't
support multiple contexts.
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
|
|
After recent concurrent cs amount increase, we must also
increase queues depth since much more concurrent work can be done.
All external queue depths were increased to 4096 as gaudi's
internal queue depths were also increased to 1024.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Rephrase F/W error message to make it more understandable to ordinary
users.
Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
The profiler needs to know the PLL values for correctly showing the
profiling data. Because our firmware can use different PLL configurations,
we need to read the PLL values from the ASIC to pass them to the profiler.
Signed-off-by: Adam Aharon <aaharon@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Once there is a 64-bit field in a structure, GCC compiler for ARM aligns
the structure to 8 bytes. In order to avoid confusion when these
structures are being passed between CPUs from different architectures, we
explicitly align the structure to 8 bytes.
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
MAP/UNMAP are done also for device memory.
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Use proper bitfield masks instead of shifting values when configuring
packets sent to device.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Currently sync stream is limited only for external queues. We want to
remove this constraint by adding a new queue property dedicated for sync
stream. In addition we move the initialization and reset methods to the
common code since we can re-use them with slight changes.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
Training schemes requires much more concurrent command submissions than
inference does. In addition, training command submissions can be completed
in a non serialized manner. Hence, we add support in which each ASIC will
be able to configure the amount of concurrent pending command submissions,
rather than use a predefined amount. This change will enhance performance
by allowing the user to add more concurrent work without waiting for the
previous work to be completed.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
We no longer need to initialize the rate limiters in GAUDI A1.
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
|
|
into char-misc-next
Georgi writes:
interconnect changes for 5.9
Here are the interconnect changes for the 5.9-rc1 merge window
consisting mostly of changes that give the core more flexibility
in order to support some new provider drivers.
Core changes:
- Export of_icc_get_from_provider()
- Relax requirement in of_icc_get_from_provider()
- Allow inter-provider pairs to be configured
- Mark all dummy functions as static inline
Signed-off-by: Georgi Djakov <georgi.djakov@linaro.org>
* tag 'icc-5.9-rc1' of https://git.linaro.org/people/georgi.djakov/linux:
interconnect: Mark all dummy functions as static inline
interconnect: Allow inter-provider pairs to be configured
interconnect: Relax requirement in of_icc_get_from_provider()
interconnect: Export of_icc_get_from_provider()
|
|
Export ddebug_exec_queries() for use by modules.
This will allow module authors to control all their *pr_debug*s
dynamically. And since ddebug_exec_queries() is what implements
"echo $query >control", it gives the same per-callsite control.
Virtues of this:
- simplicity. just an export.
- full control over any/all subsets of callsites.
- same "query/command-string" in code and console
- full callsite selectivity with module file line format
Format in particular deserves special attention; it is where
low-hanging fruit will be found.
Consider: drivers/gpu/drm/amd/display/include/logger_types.h:
#define DC_LOG_SURFACE(...) pr_debug("[SURFACE]:"__VA_ARGS__)
#define DC_LOG_HW_LINK_TRAINING(...) pr_debug("[HW_LINK_TRAINING]:"__VA_ARGS__)
.. 9 more ..
Thats 11 string prefixes, used in 804 places in drivers/gpu/**
Clearly this is a systematized classification of those callsites.
And one I'd expect to see repeated often.
Using ddebug_exec_queries(), authors can select on those prefixes
as a unitary set, equivalent to:
echo "module=MODULE_NAME format=^[SURFACE]: +p" >control
Trivially, those sets can be subsected with the other query terms too,
say file=foo, should the author see fit.
Perhaps as important, users can modify the set of enabled callsites,
presumably to aid debugging by enabling helpful debug callsites, and
disabling those that just clutter the info.
Authors could even alter [fmlt] flags, though I dont see a good reason
why they would. Perhaps harnessed by bug-logging automation to get
fuller, or more minimal bug-reports.
DRM
drm has both drm.debug, which defines 32 categories of drm_printk
logging, and entirely separate uses of pr_debug, which are dynamic on
this i915 laptop, running mainline. So I can observe and report on
both.
The i915 driver has 118 dyndbg callsites, with following
"classifications" defined in drivers/gpu/drm/i915/gvt/**
$ grep 915 /proc/dynamic_debug/control | cut -d= -f2 | cut -d: -f1,2 | sort -u
_ "gvt: cmd
_ "gvt: core
_ "gvt: dpy
_ "gvt: el
_ "gvt: irq
_ "gvt: mm
_ "gvt: mmio
_ "gvt: render
_ "gvt: sched
_ "%s for root hub!\012"
_ "Vendor defined info completion code %u\012"
This classification is entirely out-of-band for control by drm.debug,
and is only available to root user at the console. But module authors
can activate them with ddebug_exec_queries(sprintf("format=^%s +p")),
and then decide how to expose the groups to the user for max utility.
drm.debug
drm.debug has 32 bit-flags, and matching enum drm_debug_category
values to classify the ~2943 DRM_DEBUG*() callsites in drivers/gpu
The drm.debug callback could invoke ddebug_exec_queries() with 32
different hardcoded query strings, needing only (bit) ? " +p" : " -p"
added.
I briefly enabled drm.debug=0xff on my i915 laptop, which yielded
these unique prefixes: (dmesg | cut -c17- | cut -d\] -f1 | sort -u)
[drm:drm_atomic_check_only [drm
[drm:drm_atomic_get_crtc_state [drm
[drm:drm_atomic_get_plane_state [drm
[drm:drm_atomic_nonblocking_commit [drm
[drm:drm_atomic_set_fb_for_plane [drm
[drm:drm_atomic_state_default_clear [drm
[drm:__drm_atomic_state_free [drm
[drm:drm_atomic_state_init [drm
[drm:drm_crtc_vblank_helper_get_vblank_timestamp_internal [drm
[drm:drm_handle_vblank [drm
[drm:drm_ioctl [drm
[drm:drm_mode_addfb2 [drm
[drm:drm_mode_object_get [drm
[drm:drm_mode_object_put.part.0 [drm
[drm:drm_update_vblank_count [drm
[drm:drm_vblank_enable [drm
[drm:drm_vblank_restore [drm
[drm:vblank_disable_fn [drm
i915 0000:00:02.0: [drm:gen9_set_dc_state [i915
i915 0000:00:02.0: [drm:intel_atomic_get_global_obj_state [i915
i915 0000:00:02.0: [drm:__intel_display_power_get_domain.part.0 [i915
i915 0000:00:02.0: [drm:__intel_display_power_put_domain [i915
i915 0000:00:02.0: [drm:intel_plane_atomic_calc_changes [i915
i915 0000:00:02.0: [drm:skl_enable_dc6 [i915
Several good format=^prefixes are apparent there, and some misses.
^[drm:drm_atomic_ # misses: [drm:__drm_atomic_state_free [drm
^[drm:drm_ioctl
^[drm:drm_mode
^[drm:drm_vblank_ # misses: [drm:drm_update_vblank_count & [drm:vblank_disable_fn
Its not a perfect 1:1 single format-match per class, but the misses
above can be covered with 1 & 2 additional queries, which can be
concatenated together with ";" separators and submitted with 1 call.
Benefits:
For drm, adapting DRM_DEBUG to use dynamic-debug inside could
replicate (and thereby obsolete) lots of bit-checking in current
DRM_DEBUG callsites, at least with JUMP_LABEL optimized code.
ddebug_exec_queries() and a handful of fixed query-strings can select
and thereby control the already classified callsites.
With the classes mapped to queries, the enum type and parameter can be
eliminated (folded away with macro magic), at least for DYNAMIC_DEBUG
& JUMP_LABEL builds.
Is it safe ?
ddebug_exec_queries() is currently exposed to user space in
several limited ways;
1 it is called from module-load callback, where it implements the
$modname.dyndbg=+p "fake" parameter provided to all modules.
2 it handles query input via >control directly
IOW, it is "fully" exposed to local root user; exposing the same
functionality to other kernel modules is no additional risk.
The other standard issue to check is locking:
dyndbg has a single mutex, taken by ddebug_change to handle >control,
and by ddebug_proc_(start|stop) to span `cat control`. Queries
submitted via export will typically have module specified, which
dramatically cuts the scan by ddebug_change vs "module=* +p".
ISTM this proposed export presents no locking problems.
TLDR;
It would be interesting to see how drm.dyndbg=$QUERY and
drm.debug=$HEXY would interact; it might be order dependent, as
if given as modprobe args or in /etc/modprobe.d/
Acked-by: <jbaron@akamai.com>
Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
Link: https://lore.kernel.org/r/20200719231058.1586423-19-jim.cromie@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
For log-message output, reduce column space consumed by current
pr_fmt by dropping __func__ and shortening "dynamic_debug" to
"dyndbg". This improves readability on narrow consoles, and better
matches other kernel boot info messages.
Acked-by: <jbaron@akamai.com>
Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
Link: https://lore.kernel.org/r/20200719231058.1586423-18-jim.cromie@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
