Lucas De Marchi [Tue, 3 May 2022 06:15:56 +0000 (23:15 -0700)]
drm/i915/gem: Make drop_pages() return bool
Commit e4e806253003 ("drm/i915: Change shrink ordering to use locking
around unbinding.") changed the return type to int without changing the
return values or their meaning to "0 is success". Move it back to
boolean.
Colin Ian King [Mon, 9 May 2022 11:57:32 +0000 (14:57 +0300)]
ath11k: remove redundant assignment to variables vht_mcs and he_mcs
The variables vht_mcs and he_mcs are being initialized in the
start of for-loops however they are re-assigned new values in
the loop and not used outside the loop. The initializations
are redundant and can be removed.
Cleans up clang scan warnings:
warning: Although the value stored to 'vht_mcs' is used in the
enclosing expression, the value is never actually read from
'vht_mcs' [deadcode.DeadStores]
warning: Although the value stored to 'he_mcs' is used in the
enclosing expression, the value is never actually read from
'he_mcs' [deadcode.DeadStores]
Anilkumar Kolli [Mon, 9 May 2022 11:57:32 +0000 (14:57 +0300)]
ath11k: Reuse the available memory after firmware reload
Ath11k allocates memory when firmware requests memory in QMI.
Coldboot calibration and firmware recovery uses firmware reload.
On firmware reload, firmware sends memory request again. If Ath11k
allocates memory on first firmware boot, reuse the available
memory. Also check if the segment type and size is same
on the next firmware boot. Reuse if segment type/size is
same as previous firmware boot else free the segment and
allocate the segment with size/type.
Baochen Qiang [Mon, 9 May 2022 11:57:32 +0000 (14:57 +0300)]
ath11k: Designating channel frequency when sending management frames
In case of Passpoint, the WLAN interface may be requested to
remain on a specific channel and then to send some management
frames on that channel. Now chanfreq of wmi_mgmt_send_cmd is set
as 0, as a result firmware may choose a default but wrong channel.
Fix it by assigning chanfreq field with the designated channel.
This change only applies to WCN6855 and QCA6390, other chips are
not affected.
Baochen Qiang [Mon, 9 May 2022 11:57:31 +0000 (14:57 +0300)]
ath11k: Don't check arvif->is_started before sending management frames
Commit 66307ca04057 ("ath11k: fix mgmt_tx_wmi cmd sent to FW for
deleted vdev") wants both of below two conditions are true before
sending management frames:
Baochen Qiang [Mon, 9 May 2022 11:57:31 +0000 (14:57 +0300)]
ath11k: Handle keepalive during WoWLAN suspend and resume
With WoWLAN enabled and after sleeping for a rather long time,
we are seeing that with some APs, it is not able to wake up
the STA though the correct wake up pattern has been configured.
This is because the host doesn't send keepalive command to
firmware, thus firmware will not send any packet to the AP and
after a specific time the AP kicks out the STA.
Fix this issue by enabling keepalive before going to suspend
and disabling it after resume back.
Mathieu Poirier [Mon, 9 May 2022 16:35:10 +0000 (10:35 -0600)]
MAINTAINERS: Add James and Mike as Arm64 performance events reviewers
James, Mike and Leo have been doing all the reviews and development work
for the Coresight perf tools for a couple of years now. As such remove
my name and add James and Mike as official reviewers (Leo is already
listed as such).
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: James Clark <james.clark@arm.com> Cc: Leo Yan <leo.yan@linaro.org> Cc: Mike Leach <mike.leach@linaro.org> Cc: linux-arm-kernel@lists.infradead.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Takashi Iwai [Tue, 10 May 2022 10:36:26 +0000 (12:36 +0200)]
ALSA: wavefront: Proper check of get_user() error
The antient ISA wavefront driver reads its sample patch data (uploaded
over an ioctl) via __get_user() with no good reason; likely just for
some performance optimizations in the past. Let's change this to the
standard get_user() and the error check for handling the fault case
properly.
Dan Carpenter [Fri, 6 May 2022 14:53:52 +0000 (17:53 +0300)]
hte: Fix off by one in hte_push_ts_ns()
The &chip->gdev->ei[] array has chip->nlines elements so this >
comparison needs to be >= to prevent an out of bounds access. The
gdev->ei[] array is allocated in hte_register_chip().
Yang Yingliang [Fri, 6 May 2022 08:48:51 +0000 (16:48 +0800)]
hte: Fix possible use-after-free in tegra_hte_test_remove()
del_timer() does not wait until the timer handler finishing.
This means that the timer handler may still be running after
the driver's remove function has finished, which would result
in a use-after-free.
Fix it by calling del_timer_sync(), which makes sure the timer
handler has finished.
Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Reviewed-by: Dipen Patel <dipenp@nvidia.com> Acked-by: Dipen Patel <dipenp@nvidia.com> Signed-off-by: Thierry Reding <treding@nvidia.com>
Shengjiu Wang [Mon, 9 May 2022 09:14:22 +0000 (17:14 +0800)]
ASoC: fsl_micfil: Add support for i.MX8MPlus
On i.MX8Plus there are two updates for micfil module.
One is that the output format is S32_LE, only the 24 more
significative bits have information, the other bits are always
zero. Add 'formats' variable in soc data to distinguish the
format on different platform.
Another is that the fifo depth is 32 entries.
YC Hung [Mon, 9 May 2022 17:04:25 +0000 (12:04 -0500)]
ASoC: SOF: ipc: introduce cont_update_posn in sof_ipc_stream_params struct
The host stream position is updated when no_stream_position is set as 0.
However current implementation updates host stream position only when
report data is larger than or equal to host period size which is decided
by the period size of host side. It maybe cause host stream position
update not in time. Therefore this patch introduces another field
"cont_update_posn", a boolean value aimed to update host stream position
continuously and based on period size of pipeline. It can get better
precise when need to update host stream position from firmware.
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com> Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com> Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com> Signed-off-by: YC Hung <yc.hung@mediatek.com> Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Link: https://lore.kernel.org/r/20220509170425.54640-1-pierre-louis.bossart@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org>
Brent Lu [Mon, 9 May 2022 17:09:21 +0000 (12:09 -0500)]
ASoC: Intel: sof_ssp_amp: fix no DMIC BE Link on Chromebooks
The SOF topology supports 2 BE Links(dmic01 and dmic16k) and each
link supports up to four DMICs. However, Chromebook does not implement
ACPI NHLT table so the mach->mach_params.dmic_num is always zero. We
add a quirk so machine driver knows it's running on a Chromebook and
need to create BE Links for DMIC.
Expose mlx5_sriov_blocking_notifier_register / unregister APIs to let a
VF register to be notified for its enablement / disablement by the PF.
Upon VF probe it will call mlx5_sriov_blocking_notifier_register() with
its notifier block and upon VF remove it will call
mlx5_sriov_blocking_notifier_unregister() to drop its registration.
This can give a VF the ability to clean some resources upon disable
before that the command interface goes down and on the other hand sets
some stuff before that it's enabled.
This may be used by a VF which is migration capable in few cases.(e.g.
PF load/unload upon an health recovery).
"Redistribution of this file is permitted under the GNU General Public
License." and was added by Ted in 1993, at which point GPLv2 only
was the default Linux license.
Jan Kara [Tue, 10 May 2022 10:36:04 +0000 (12:36 +0200)]
udf: Avoid using stale lengthOfImpUse
udf_write_fi() uses lengthOfImpUse of the entry it is writing to.
However this field has not yet been initialized so it either contains
completely bogus value or value from last directory entry at that place.
In either case this is wrong and can lead to filesystem corruption or
kernel crashes.
Reported-by: butt3rflyh4ck <butterflyhuangxx@gmail.com> CC: stable@vger.kernel.org Fixes: 979a6e28dd96 ("udf: Get rid of 0-length arrays in struct fileIdentDesc") Signed-off-by: Jan Kara <jack@suse.cz>
Tanmay Jagdale [Tue, 10 May 2022 10:26:57 +0000 (15:56 +0530)]
perf/marvell_cn10k: Fix tad_pmu_event_init() to check pmu type first
Make sure to check the pmu type first and then check event->attr.disabled.
Doing so would avoid reading the disabled attribute of an event that is
not handled by TAD PMU.
Mark Brown [Tue, 10 May 2022 11:12:50 +0000 (12:12 +0100)]
Clean up usage of the endianness flag
Merge series from Charles Keepax <ckeepax@opensource.cirrus.com>:
Before componentisation any part registered as a CODEC would have
automatically supported both little and big endian, ie. the core
would duplicate any supported LE or BE PCM format to support the other
endian as well. As componentisation removed the distinction between
CODEC drivers and platform drivers, a flag was added to specify
if this behaviour is required for a particular component. However,
as most systems tend to use little endian the absence of the flag
is rarely noticed. Also the naming of the flag "endianness" is a
little unobvious as to if it should be applied to a particular
component.
This series adds a comment to better explain the meaning of the
flag and then tidys up the usage of the flag. A couple of uses
of the flag are removed where is has been used inappropriately
on the CPU side of the DAI link, this is clearly not valid in the
cases it has been used, and I suspect never would be valid. Then
some redundant formats are removed, since they would be covered by
existing endianness flags. And finally a bunch of devices that are
missing the flag have it added.
It is worth noting that since componenisation there are now a couple
of cases where it is not entire clear to me that the flag should
be applied to all CODECs as it was before. In those cases I haven't
updated the driver to add the flag and they are outlined here:
1) Build into the AP CODECs, these are actual silicon inside the main
processor and they typically receive audio directly from an internal
bus. It is not obvious to me that these can happily ignore endian. On
the CODEC side these include: jz4725b.c, jz4760.c, jz4770.c,
rk3328_codec.c, lpass-va-macro.c, lpass-rx-macro.c, lpass-tx-macro.c,
lpass-wsa-macro.c. There are also some examples of this scattered
around the various platform support directories in sound/soc.
2) Devices behind non-audio buses, SPI just moves bits and doesn't
really define an endian for audio data on the bus. Thus it seems the
CODEC probably can care about the endian. The only devices that fall
into this group (mostly for AoV) are: rt5514-spi.c, rt5677-spi.c,
cros_ec_codec.c (only the AoV).
3) CODECs with no DAIs, these could specify the flag and plenty of
them do; CODECs from the initial conversion to componentisation. But
the flag makes no difference here since there is nothing for it to
apply to. This includes purely analogue CODECs: aw8738.c, ssm2305.c,
tpa6130a2.c, tda7419.c, max9759.c, max9768.c, max9877.c, lm4857.c,
simple-mux.c, simple-amplifier.c. And devices that only do jack
detection: ts3a227e.c, mt6359-accdet.c.
If there are any opinions on adding the flag to any of those three
groups they would be greatfully received. But I am leaning towards
leaving 1,2 without endianness flags since it feels inappropriate,
and removing the endian flag from devices in catagory 3 that already
have it. Assuming no one objects to that I will do a follow up
series for that.
Joey Gouly [Tue, 10 May 2022 10:27:21 +0000 (11:27 +0100)]
arm64: vdso: fix makefile dependency on vdso.so
There is currently no dependency for vdso*-wrap.S on vdso*.so, which means that
you can get a build that uses a stale vdso*-wrap.o.
In commit a5b8ca97fbf8, the file that includes the vdso.so was moved and renamed
from arch/arm64/kernel/vdso/vdso.S to arch/arm64/kernel/vdso-wrap.S, when this
happened the Makefile was not updated to force the dependcy on vdso.so.
Fixes: a5b8ca97fbf8 ("arm64: do not descend to vdso directories twice") Signed-off-by: Joey Gouly <joey.gouly@arm.com> Cc: Masahiro Yamada <masahiroy@kernel.org> Cc: Vincenzo Frascino <vincenzo.frascino@arm.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will@kernel.org> Link: https://lore.kernel.org/r/20220510102721.50811-1-joey.gouly@arm.com Signed-off-by: Will Deacon <will@kernel.org>
Robin Murphy [Tue, 10 May 2022 08:38:58 +0000 (09:38 +0100)]
iommu/arm-smmu: Force identity domains for legacy binding
When using the legacy "mmu-masters" DT binding, we reject DMA domains
since we have no guarantee of driver probe order and thus can't rely on
client drivers getting the correct DMA ops. However, we can do better
than fall back to the old no-default-domain behaviour now, by forcing an
identity default domain instead. This also means that detaching from a
VFIO domain can actually work - that looks to have been broken for over
6 years, so clearly isn't something that legacy binding users care
about, but we may as well make the driver code make sense anyway.
If an eMMC card supports TRIM and indicates that it erases to zeros, we can
use it to support hardware offloading of REQ_OP_WRITE_ZEROES, so let's add
support for this.
Jing Xia [Tue, 10 May 2022 02:35:14 +0000 (10:35 +0800)]
writeback: Avoid skipping inode writeback
We have run into an issue that a task gets stuck in
balance_dirty_pages_ratelimited() when perform I/O stress testing.
The reason we observed is that an I_DIRTY_PAGES inode with lots
of dirty pages is in b_dirty_time list and standard background
writeback cannot writeback the inode.
After studing the relevant code, the following scenario may lead
to the issue:
This patch moves the dirty inode to b_dirty list when the inode
currently is not queued in b_io or b_more_io list at the end of
writeback_single_inode.
Reviewed-by: Jan Kara <jack@suse.cz> Reviewed-by: Christoph Hellwig <hch@lst.de> CC: stable@vger.kernel.org Fixes: 0ae45f63d4ef ("vfs: add support for a lazytime mount option") Signed-off-by: Jing Xia <jing.xia@unisoc.com> Signed-off-by: Jan Kara <jack@suse.cz> Link: https://lore.kernel.org/r/20220510023514.27399-1-jing.xia@unisoc.com
Colin Ian King [Sun, 8 May 2022 21:45:00 +0000 (22:45 +0100)]
x25: remove redundant pointer dev
Pointer dev is being assigned a value that is never used, the assignment
and the variable are redundant and can be removed. Also replace null check
with the preferred !ptr idiom.
Cleans up clang scan warning:
net/x25/x25_proc.c:94:26: warning: Although the value stored to 'dev' is
used in the enclosing expression, the value is never actually read
from 'dev' [deadcode.DeadStores]
====================
This is a patch series for Ethernet driver of Sunplus SP7021 SoC.
Sunplus SP7021 is an ARM Cortex A7 (4 cores) based SoC. It integrates
many peripherals (ex: UART, I2C, SPI, SDIO, eMMC, USB, SD card and
etc.) into a single chip. It is designed for industrial control
applications.
Refer to:
https://sunplus.atlassian.net/wiki/spaces/doc/overview
https://tibbo.com/store/plus1.html
====================
dma-buf: call dma_buf_stats_setup after dmabuf is in valid list
When dma_buf_stats_setup() fails, it closes the dmabuf file which
results into the calling of dma_buf_file_release() where it does
list_del(&dmabuf->list_node) with out first adding it to the proper
list. This is resulting into panic in the below path:
__list_del_entry_valid+0x38/0xac
dma_buf_file_release+0x74/0x158
__fput+0xf4/0x428
____fput+0x14/0x24
task_work_run+0x178/0x24c
do_notify_resume+0x194/0x264
work_pending+0xc/0x5f0
Fix it by moving the dma_buf_stats_setup() after dmabuf is added to the
list.
Fixes: bdb8d06dfefd ("dmabuf: Add the capability to expose DMA-BUF stats in sysfs") Signed-off-by: Charan Teja Reddy <quic_charante@quicinc.com> Tested-by: T.J. Mercier <tjmercier@google.com> Acked-by: T.J. Mercier <tjmercier@google.com> Cc: <stable@vger.kernel.org> # 5.15.x+ Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/1652125797-2043-1-git-send-email-quic_charante@quicinc.com
Wan Jiabing [Tue, 10 May 2022 09:05:31 +0000 (17:05 +0800)]
cpufreq: mediatek: Fix potential deadlock problem in mtk_cpufreq_set_target
Fix following coccichek error:
./drivers/cpufreq/mediatek-cpufreq.c:199:2-8: preceding lock on line
./drivers/cpufreq/mediatek-cpufreq.c:208:2-8: preceding lock on line
mutex_lock is acquired but not released before return.
Use 'goto out' to help releasing the mutex_lock.
Fixes: c210063b40ac ("cpufreq: mediatek: Add opp notification support") Signed-off-by: Wan Jiabing <wanjiabing@vivo.com> Reviewed-by: Rex-BC Chen <rex-bc.chen@mediatek.com> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Ravi Bangoria [Fri, 29 Apr 2022 05:14:41 +0000 (10:44 +0530)]
perf/amd/ibs: Use interrupt regs ip for stack unwinding
IbsOpRip is recorded when IBS interrupt is triggered. But there is
a skid from the time IBS interrupt gets triggered to the time the
interrupt is presented to the core. Meanwhile processor would have
moved ahead and thus IbsOpRip will be inconsistent with rsp and rbp
recorded as part of the interrupt regs. This causes issues while
unwinding stack using the ORC unwinder as it needs consistent rip,
rsp and rbp. Fix this by using rip from interrupt regs instead of
IbsOpRip for stack unwinding.
Fixes: ee9f8fce99640 ("x86/unwind: Add the ORC unwinder") Reported-by: Dmitry Monakhov <dmtrmonakhov@yandex-team.ru> Suggested-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20220429051441.14251-1-ravi.bangoria@amd.com
Manuel Ullmann [Wed, 4 May 2022 19:30:44 +0000 (21:30 +0200)]
net: atlantic: always deep reset on pm op, fixing up my null deref regression
The impact of this regression is the same for resume that I saw on
thaw: the kernel hangs and nothing except SysRq rebooting can be done.
Fixes regression in commit cbe6c3a8f8f4 ("net: atlantic: invert deep
par in pm functions, preventing null derefs"), where I disabled deep
pm resets in suspend and resume, trying to make sense of the
atl_resume_common() deep parameter in the first place.
It turns out, that atlantic always has to deep reset on pm
operations. Even though I expected that and tested resume, I screwed
up by kexec-rebooting into an unpatched kernel, thus missing the
breakage.
This fixup obsoletes the deep parameter of atl_resume_common, but I
leave the cleanup for the maintainers to post to mainline.
Suspend and hibernation were successfully tested by the reporters.
Hyeonggon Yoo [Sat, 7 May 2022 07:35:06 +0000 (16:35 +0900)]
MAINTAINERS: add myself as reviewer for slab
Recently I was involved in slab subsystem (reviewing struct slab,
SLUB debugfs and etc). I would like to help maintainers and people
working on slab allocators by reviewing and testing their work.
Matthew Gerlach [Thu, 5 May 2022 10:06:17 +0000 (06:06 -0400)]
fpga: dfl: Allow Port to be linked to FME's DFL
Currently we use PORTn_OFFSET to locate PORT DFLs, and PORT DFLs are not
connected FME DFL. But for some cases (e.g. Intel Open FPGA Stack device),
PORT DFLs are connected to FME DFL directly, so we don't need to search
PORT DFLs via PORTn_OFFSET again. If BAR value of PORTn_OFFSET is 0x7
(FME_PORT_OFST_BAR_SKIP) then driver will skip searching the DFL for that
port. If BAR value is invalid, return -EINVAL.
fpga: dfl: check feature type before parse irq info
Previously the feature IDs defined are unique, no matter
which feature type. But currently we want to extend its
usage to have a per-type feature ID space, so this patch
adds feature type checking as well just before look into
feature ID for different features which have irq info.
To fix below kernel-doc warnings this patch does the following
->Replaced Return\Returns with 'Return:' keyword.
->Added 'Return' description For __init of_fpga_region_init()' API.
->Added description for 'child_regions_with_firmware()' API.
warning: No description found for return value of
'of_fpga_region_find'.
warning: No description found for return value of
'of_fpga_region_get_bridges'.
warning: missing initial short description on line:
* child_regions_with_firmware
warning: No description found for return value of
'child_regions_with_firmware'.
warning: No description found for return value of
'of_fpga_region_notify_pre_apply'.
warning: No description found for return value of
'of_fpga_region_notify'.
warning: No description found for return value of
'of_fpga_region_init'.
Xiubo Li [Thu, 5 May 2022 10:53:09 +0000 (18:53 +0800)]
ceph: check folio PG_private bit instead of folio->private
The pages in the file mapping maybe reclaimed and reused by other
subsystems and the page->private maybe used as flags field or
something else, if later that pages are used by page caches again
the page->private maybe not cleared as expected.
Here will check the PG_private bit instead of the folio->private.
Cc: stable@vger.kernel.org
URL: https://tracker.ceph.com/issues/55421 Signed-off-by: Xiubo Li <xiubli@redhat.com> Reviewed-by: Luis Henriques <lhenriques@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Jeff Layton [Mon, 25 Apr 2022 19:54:27 +0000 (15:54 -0400)]
ceph: fix setting of xattrs on async created inodes
Currently when we create a file, we spin up an xattr buffer to send
along with the create request. If we end up doing an async create
however, then we currently pass down a zero-length xattr buffer.
Fix the code to send down the xattr buffer in req->r_pagelist. If the
xattrs span more than a page, however give up and don't try to do an
async create.
Cc: stable@vger.kernel.org
URL: https://bugzilla.redhat.com/show_bug.cgi?id=2063929 Fixes: 9a8d03ca2e2c ("ceph: attempt to do async create when possible") Reported-by: John Fortin <fortinj66@gmail.com> Reported-by: Sri Ramanujam <sri@ramanujam.io> Signed-off-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: Xiubo Li <xiubli@redhat.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
====================
ptp: Support hardware clocks with additional free running cycle counter
ptp vclocks require a clock with free running time for the timecounter.
Currently only a physical clock forced to free running is supported.
If vclocks are used, then the physical clock cannot be synchronized
anymore. The synchronized time is not available in hardware in this
case. As a result, timed transmission with TAPRIO hardware support
is not possible anymore.
If hardware would support a free running time additionally to the
physical clock, then the physical clock does not need to be forced to
free running. Thus, the physical clocks can still be synchronized while
vclocks are in use.
The physical clock could be used to synchronize the time domain of the
TSN network and trigger TAPRIO. In parallel vclocks can be used to
synchronize other time domains.
One year ago I thought for two time domains within a TSN network also
two physical clocks are required. This would lead to new kernel
interfaces for asking for the second clock, ... . But actually for a
time triggered system like TSN there can be only one time domain that
controls the system itself. All other time domains belong to other
layers, but not to the time triggered system itself. So other time
domains can be based on a free running counter if similar mechanisms
like 2 step synchroisation are used.
Synchronisation was tested with two time domains between two directly
connected hosts. Each host run two ptp4l instances, the first used the
physical clock and the second used the virtual clock. I used my FPGA
based network controller as network device. ptp4l was used in
combination with the virtual clock support patches from Miroslav
Lichvar.
v4:
- if_index of 0 is invalid (Jonathan Lemon)
- set if_index to 0 in the SOF_TIMESTAMPING_RAW_HARDWARE block (Jonathan
Lemon)
- add helper function for netdev_get_tstamp() call (Jonathan Lemon)
- update SKBTX_ANY_TSTAMP (Paolo Abeni)
- use separate bits for new tx_flags (Richard Cochran)
v3:
- optimize ptp_convert_timestamp (Richard Cochran)
- call dev_get_by_napi_id() only if needed (Richard Cochran)
- use non-negated logical test (Richard Cochran)
- add comment for skipped output (Richard Cochran)
- add comment for SKBTX_HW_TSTAMP_USE_CYCLES masking (Richard Cochran)
The TSN endpoint Ethernet MAC supports a free running counter
additionally to its clock. This free running counter can be read and
hardware timestamps are supported. As the name implies, this counter
cannot be set and its frequency cannot be adjusted.
Add free running cycle counter support based on this free running
counter to physical clock. This also requires hardware time stamps
based on that free running counter.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
ptp_convert_timestamp() is called in the RX path of network messages.
The current implementation takes ~5000ns on 1.2GHz A53. This is too much
for the hot path of packet processing.
Introduce hash table for fast vclock lookup in ptp_convert_timestamp().
The execution time of ptp_convert_timestamp() is reduced to ~700ns on
1.2GHz A53.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Acked-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
If a physical clock supports a free running cycle counter, then
timestamps shall be based on this time too. For TX it is known in
advance before the transmission if a timestamp based on the free running
cycle counter is needed. For RX it is impossible to know which timestamp
is needed before the packet is received and assigned to a socket.
Support late timestamp determination by a network device. Therefore, an
address/cookie is stored within the new netdev_data field of struct
skb_shared_hwtstamps. This address/cookie is provided to a new network
device function called ndo_get_tstamp(), which returns a timestamp based
on the normal/adjustable time or based on the free running cycle
counter. If function is not supported, then timestamp handling is not
changed.
This mechanism is intended for RX, but TX use is also possible.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com> Acked-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
ptp_convert_timestamp() converts only the timestamp hwtstamp, which is
a field of the argument with the type struct skb_shared_hwtstamps *. So
a pointer to the hwtstamp field of this structure is sufficient.
Rework ptp_convert_timestamp() to use an argument of type ktime_t *.
This allows to add additional timestamp manipulation stages before the
call of ptp_convert_timestamp().
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Acked-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The free running cycle counter of physical clocks called cycles shall be
used for hardware timestamps to enable synchronisation.
Introduce new flag SKBTX_HW_TSTAMP_USE_CYCLES, which signals driver to
provide a TX timestamp based on cycles if cycles are supported.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Acked-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
ptp vclocks require a free running time for their timecounter.
Currently only a physical clock forced to free running is supported.
If vclocks are used, then the physical clock cannot be synchronized
anymore. The synchronized time is not available in hardware in this
case. As a result, timed transmission with TAPRIO hardware support
is not possible anymore.
If hardware would support a free running time additionally to the
physical clock, then the physical clock does not need to be forced to
free running. Thus, the physical clocks can still be synchronized
while vclocks are in use.
The physical clock could be used to synchronize the time domain of the
TSN network and trigger TAPRIO. In parallel vclocks can be used to
synchronize other time domains.
Introduce support for a free running cycle counter called cycles to
physical clocks. Rework ptp vclocks to use this free running cycle
counter. Default implementation is based on time of physical clock.
Thus, behavior of ptp vclocks based on physical clocks without free
running cycle counter is identical to previous behavior.
Signed-off-by: Gerhard Engleder <gerhard@engleder-embedded.com> Acked-by: Richard Cochran <richardcochran@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Jakub Kicinski [Fri, 6 May 2022 20:00:29 +0000 (13:00 -0700)]
eth: dpaa2-mac: remove a dead-code NULL check on fwnode parent
Since commit 4e30e98c4b4c ("dpaa2-mac: return -EPROBE_DEFER from dpaa2_mac_open in case the fwnode is not set")
@parent can't be NULL after the if. It's either the address
of the ->fwnode of @dpmacs or @fwnode in case of ACPI.
Mark Bloch [Tue, 15 Mar 2022 16:56:50 +0000 (16:56 +0000)]
net/mlx5: Lag, add debugfs to query hardware lag state
Lag state has become very complicated with many modes, flags, types and
port selections methods and future work will add additional features.
Add a debugfs to query the current lag state. A new directory named "lag"
will be created under the mlx5 debugfs directory. As the driver has
debugfs per pci function the location will be: <debugfs>/mlx5/<BDF>/lag
For example:
/sys/kernel/debug/mlx5/0000:08:00.0/lag
The following files are exposed:
- state: Returns "active" or "disabled". If "active" it means hardware
lag is active.
- members: Returns the BDFs of all the members of lag object.
- type: Returns the type of the lag currently configured. Valid only
if hardware lag is active.
* "roce" - Members are bare metal PFs.
* "switchdev" - Members are in switchdev mode.
* "multipath" - ECMP offloads.
- port_sel_mode: Returns the egress port selection method, valid
only if hardware lag is active.
* "queue_affinity" - Egress port is selected by
the QP/SQ affinity.
* "hash" - Egress port is selected by hash done on
each packet. Controlled by: xmit_hash_policy of the
bond device.
- flags: Returns flags that are specific per lag @type. Valid only if
hardware lag is active.
* "shared_fdb" - "on" or "off", if "on" single FDB is used.
- mapping: Returns the mapping which is used to select egress port.
Valid only if hardware lag is active.
If @port_sel_mode is "hash" returns the active egress ports.
The hash result will select only active ports.
if @port_sel_mode is "queue_affinity" returns the mapping
between the configured port affinity of the QP/SQ and actual
egress port. For example:
* 1:1 - Mapping means if the configured affinity is port 1
traffic will egress via port 1.
* 1:2 - Mapping means if the configured affinity is port 1
traffic will egress via port 2. This can happen
if port 1 is down or in active/backup mode and port 1
is backup.
Signed-off-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Mark Bloch [Wed, 2 Mar 2022 15:38:50 +0000 (15:38 +0000)]
net/mlx5: Lag, use buckets in hash mode
When in hardware lag and the NIC has more than 2 ports when one port
goes down need to distribute the traffic between the remaining
active ports.
For better spread in such cases instead of using 1-to-1 mapping and only
4 slots in the hash, use many.
Each port will have many slots that point to it. When a port goes down
go over all the slots that pointed to that port and spread them between
the remaining active ports. Once the port comes back restore the default
mapping.
We will have number_of_ports * MLX5_LAG_MAX_HASH_BUCKETS slots.
Each MLX5_LAG_MAX_HASH_BUCKETS belong to a different port.
The native mapping is such that:
port 1: The first MLX5_LAG_MAX_HASH_BUCKETS slots are: [1, 1, .., 1]
which means if a packet is hased into one of this slots it will hit the
wire via port 1.
port 2: The second MLX5_LAG_MAX_HASH_BUCKETS slots are: [2, 2, .., 2]
which means if a packet is hased into one of this slots it will hit the
wire via port2.
and this mapping is the same of the rest of the ports.
On a failover, lets say port 2 goes down (port 1, 3, 4 are still up).
the new mapping for port 2 will be:
port 2: The second MLX5_LAG_MAX_HASH_BUCKETS are: [1, 3, 1, 4, .., 4]
which means the mapping was changed from the native mapping to a mapping
that consists of only the active ports.
With this if a port goes down the traffic will be split between the
active ports randomly
Signed-off-by: Mark Bloch <mbloch@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Mark Bloch [Tue, 1 Mar 2022 17:34:31 +0000 (17:34 +0000)]
net/mlx5: Lag, use hash when in roce lag on 4 ports
Downstream patches will add support for lag over 4 ports.
In that mode we will only use hash as the uplink selection method.
Using hash instead of queue affinity (before this patch) offers key
advantages like:
- Align ports selection method with the method used by the bond device
- Better packets distribution where a single queue can transmit from
multiple ports (with queue affinity a queue is bound to a single port
regardless of the packet being sent).
- In case of failover we traffic is split between multiple ports and not
a single one like in queue affinity.
Going forward it was decided that queue affinity will be deprecated
as using hash provides a better user experience which means on 4 ports
HCAs hash will always be used.
Future work will add hash support for 2 ports HCAs as well.
Signed-off-by: Mark Bloch <mbloch@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Mark Bloch [Tue, 1 Mar 2022 17:24:40 +0000 (17:24 +0000)]
net/mlx5: Lag, support single FDB only on 2 ports
E-Switch currently doesn't support more than 2 E-Switch managers
being aggregated under a single hardware lag. Have specific checks
to disallow creating lag when the code doesn't support it.
Signed-off-by: Mark Bloch <mbloch@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Mark Bloch [Sun, 27 Feb 2022 13:45:59 +0000 (13:45 +0000)]
net/mlx5: Lag, store number of ports inside lag object
Store the number of lag ports inside the lag object. Lag object is a single
shared object managing the lag state of multiple mlx5 devices on the same
physical HCA.
Downstream patches will allow hardware lag to be created over devices with
more than 2 ports.
Signed-off-by: Mark Bloch <mbloch@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Mark Bloch [Sun, 27 Feb 2022 12:40:39 +0000 (12:40 +0000)]
net/mlx5: Lag, filter non compatible devices
When search for a peer lag device we can filter based on that
device's capabilities.
Downstream patch will be less strict when filtering compatible devices
and remove the limitation where we require exact MLX5_MAX_PORTS and
change it to a range.
Signed-off-by: Mark Bloch <mbloch@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Gavin Li [Sun, 27 Mar 2022 14:36:44 +0000 (17:36 +0300)]
net/mlx5: Increase FW pre-init timeout for health recovery
Currently, health recovery will reload driver to recover it from fatal
errors. During the driver's load process, it would wait for FW to set the
pre-init bit for up to 120 seconds, beyond this threshold it would abort
the load process. In some cases, such as a FW upgrade on the DPU, this
timeout period is insufficient, and the user has no way to recover the
host device.
To solve this issue, introduce a new FW pre-init timeout for health
recovery, which is set to 2 hours.
The timeout for devlink reload and probe will use the original one because
they are user triggered flows, and therefore should not have a
significantly long timeout, during which the user command would hang.
Gavin Li [Sun, 27 Mar 2022 14:45:32 +0000 (17:45 +0300)]
net/mlx5: Add exit route when waiting for FW
Currently, removing a device needs to get the driver interface lock before
doing any cleanup. If the driver is waiting in a loop for FW init, there
is no way to cancel the wait, instead the device cleanup waits for the
loop to conclude and release the lock.
To allow immediate response to remove device commands, check the TEARDOWN
flag while waiting for FW init, and exit the loop if it has been set.
Slark Xiao [Tue, 3 May 2022 02:43:49 +0000 (10:43 +0800)]
bus: mhi: host: Add support for Foxconn T99W373 and T99W368
Product's enumeration align with previous Foxconn
SDX55, so T99W373(SDX62)/T99W368(SDX65) would use
the same config as Foxconn SDX55.
Remove fw and edl for this new commit.
The SMB2 Write packet contains data that is to be written
to a file or to a pipe. Depending on the client, there may
be padding between the header and the data field.
Currently, the length is validated only in the case padding
is present.
Since the DataOffset field always points to the beginning
of the data, there is no need to have a special case for
padding. By removing this, the length is validated in both
cases.
Signed-off-by: Marios Makassikis <mmakassikis@freebox.fr> Acked-by: Namjae Jeon <linkinjeon@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
ksmbd: fix reference count leak in smb_check_perm_dacl()
The issue happens in a specific path in smb_check_perm_dacl(). When
"id" and "uid" have the same value, the function simply jumps out of
the loop without decrementing the reference count of the object
"posix_acls", which is increased by get_acl() earlier. This may
result in memory leaks.
Fix it by decreasing the reference count of "posix_acls" before
jumping to label "check_access_bits".
Fixes: 777cad1604d6 ("ksmbd: remove select FS_POSIX_ACL in Kconfig") Signed-off-by: Xin Xiong <xiongx18@fudan.edu.cn> Signed-off-by: Xin Tan <tanxin.ctf@gmail.com> Acked-by: Namjae Jeon <linkinjeon@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
Masahiro Yamada [Sat, 7 May 2022 05:22:43 +0000 (14:22 +0900)]
crypto: vmx - Fix build error
When I refactored this Makefile, I accidentally changed the CONFIG
option.
Fixes: b52455a73db9 ("crypto: vmx - Align the short log with Makefile cleanups") Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add a wrapper that converts back from the folio to the page. This
entire file needs to be converted to use folios, but that's a
task for a different set of patches.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
All but two of the callers already have a folio; pass a folio into
try_to_free_buffers(). This removes the last user of cancel_dirty_page()
so remove that wrapper function too.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Reviewed-by: Jeff Layton <jlayton@kernel.org>
projects / people / ms / linux.git / log
