Jens Axboe [Mon, 13 Jun 2022 12:52:52 +0000 (06:52 -0600)]
Merge branch 'io_uring/io_uring-5.19' of https://github.com/isilence/linux into io_uring-5.19
Pull io_uring fixes from Pavel.
* 'io_uring/io_uring-5.19' of https://github.com/isilence/linux:
io_uring: fix double unlock for pbuf select
io_uring: kbuf: fix bug of not consuming ring buffer in partial io case
io_uring: openclose: fix bug of closing wrong fixed file
io_uring: fix not locked access to fixed buf table
io_uring: fix races with buffer table unregister
io_uring: fix races with file table unregister
Eric Dumazet [Sat, 11 Jun 2022 03:30:16 +0000 (20:30 -0700)]
tcp: sk_forced_mem_schedule() optimization
sk_memory_allocated_add() has three callers, and returns
to them @memory_allocated.
sk_forced_mem_schedule() is one of them, and ignores
the returned value.
Change sk_memory_allocated_add() to return void.
Change sock_reserve_memory() and __sk_mem_raise_allocated()
to call sk_memory_allocated().
This removes one cache line miss [1] for RPC workloads,
as first skbs in TCP write queue and receive queue go through
sk_forced_mem_schedule().
[1] Cache line holding tcp_memory_allocated.
Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: smsc95xx: add support for Microchip EVB-LAN8670-USB
This patch adds support for Microchip's EVB-LAN8670-USB 10BASE-T1S
ethernet device to the existing smsc95xx driver by adding the new
USB VID/PID pairs.
Signed-off-by: Parthiban Veerasooran <Parthiban.Veerasooran@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Yinjun Zhang [Mon, 13 Jun 2022 09:58:31 +0000 (11:58 +0200)]
nfp: support 48-bit DMA addressing for NFP3800
48-bit DMA addressing is supported in NFP3800 HW and implemented
in NFDK firmware, so enable this feature in driver now. Note that
with this change, NFD3 firmware, which doesn't implement 48-bit
DMA, cannot be used for NFP3800 any more.
RX free list descriptor, used by both NFD3 and NFDK, is also modified
to support 48-bit DMA. That's OK because the top bits is always get
set to 0 when assigned with 40-bit address.
Based on initial work of Jakub Kicinski <jakub.kicinski@netronome.com>.
Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Move peripheral properties for Tegra QSPI controller to
nvidia,tegra210-quad-peripheral-props.yaml and add reference
to spi-peripheral-props.yaml file.
Stephen Kitt [Tue, 7 Jun 2022 18:53:04 +0000 (20:53 +0200)]
regulator: rpi-panel-attiny: Use backlight helper
backlight_properties.fb_blank is deprecated. The states it represents
are handled by other properties; but instead of accessing those
properties directly, drivers should use the helpers provided by
backlight.h.
Instead of retrieving the backlight brightness in struct
backlight_properties manually, and then checking whether the backlight
should be on at all, use backlight_get_brightness() which does all
this and insulates this from future changes.
Mark Brown [Fri, 3 Jun 2022 12:46:09 +0000 (14:46 +0200)]
ASoC: wcd9335: Fix spurious event generation
The slimbus mux put operation unconditionally reports a change in value
which means that spurious events are generated. Fix this by exiting early
in that case.
Yassine Oudjana [Mon, 6 Jun 2022 15:22:26 +0000 (19:22 +0400)]
ASoC: wcd9335: Remove RX channel from old list before adding it to a new one
Currently in slim_rx_mux_put, an RX channel gets added to a new list
even if it is already in one. This can mess up links and make either
it, the new list head, or both, get linked to the wrong entries.
This can cause an entry to link to itself which in turn ends up
making list_for_each_entry in other functions loop infinitely.
To avoid issues, always remove the RX channel from any list it's in
before adding it to a new list.
We ran into multiple DMA TX errors while writing files over a network
block device running on top of a DMA-connected AXI Ethernet device on
64-bit RISC-V machines. The errors indicated that the DMA had fetched a
null descriptor and we found that the reason for this is that AXI DMA had
unexpectedly processed a partially updated tail descriptor pointer. To
fix it, we suggest that the driver should use one 64-bit write instead
of two 32-bit writes to perform such update if possible. For those
archectures where double-word load/stores are unavailable, e.g. 32-bit
archectures, force a driver probe failure if the driver finds 64-bit
capability on DMA.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Andy Chiu [Mon, 13 Jun 2022 03:42:02 +0000 (11:42 +0800)]
net: axienet: Use iowrite64 to write all 64b descriptor pointers
According to commit f735c40ed93c ("net: axienet: Autodetect 64-bit DMA
capability") and AXI-DMA spec (pg021), on 64-bit capable dma, only
writing MSB part of tail descriptor pointer causes DMA engine to start
fetching descriptors. However, we found that it is true only if dma is in
idle state. In other words, dma would use a tailp even if it only has LSB
updated, when the dma is running.
The non-atomicity of this behavior could be problematic if enough
delay were introduced in between the 2 writes. For example, if an
interrupt comes right after the LSB write and the cpu spends long
enough time in the handler for the dma to get back into idle state by
completing descriptors, then the seconcd write to MSB would treat dma
to start fetching descriptors again. Since the descriptor next to the
one pointed by current tail pointer is not filled by the kernel yet,
fetching a null descriptor here causes a dma internal error and halt
the dma engine down.
We suggest that the dma engine should start process a 64-bit MMIO write
to the descriptor pointer only if ONE 32-bit part of it is written on all
states. Or we should restrict the use of 64-bit addressable dma on 32-bit
platforms, since those devices have no instruction to guarantee the write
to LSB and MSB part of tail pointer occurs atomically to the dma.
Signed-off-by: Andy Chiu <andy.chiu@sifive.com> Reported-by: Max Hsu <max.hsu@sifive.com> Reviewed-by: Greentime Hu <greentime.hu@sifive.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andy Chiu [Mon, 13 Jun 2022 03:42:01 +0000 (11:42 +0800)]
net: axienet: make the 64b addresable DMA depends on 64b archectures
Currently it is not safe to config the IP as 64-bit addressable on 32-bit
archectures, which cannot perform a double-word store on its descriptor
pointers. The pointer is 64-bit wide if the IP is configured as 64-bit,
and the device would process the partially updated pointer on some
states if the pointer was updated via two store-words. To prevent such
condition, we force a probe fail if we discover that the IP has 64-bit
capability but it is not running on a 64-Bit kernel.
This is a series of patch (1/2). The next patch must be applied in order
to make 64b DMA safe on 64b archectures.
Signed-off-by: Andy Chiu <andy.chiu@sifive.com> Reported-by: Max Hsu <max.hsu@sifive.com> Reviewed-by: Greentime Hu <greentime.hu@sifive.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dylan Yudaken [Mon, 13 Jun 2022 10:11:57 +0000 (03:11 -0700)]
io_uring: limit size of provided buffer ring
The type of head and tail do not allow more than 2^15 entries in a
provided buffer ring, so do not allow this.
At 2^16 while each entry can be indexed, there is no way to
disambiguate full vs empty.
David S. Miller [Mon, 13 Jun 2022 11:01:58 +0000 (12:01 +0100)]
Merge branch 'ipa-refactoring'
Alex Elder says:
====================
net: ipa: simple refactoring
This series contains some minor code improvements.
The first patch verifies that the configuration is compatible with a
recently-defined limit. The second and third rename two fields so
they better reflect their use in the code. The next gets rid of an
empty function by reworking its only caller.
The last two begin to remove the assumption that an event ring is
associated with a single channel. Eventually we'll support having
multiple channels share an event ring but some more needs to be done
before that can happen.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Elder [Fri, 10 Jun 2022 15:46:15 +0000 (10:46 -0500)]
net: ipa: derive channel from transaction
In gsi_channel_tx_queued(), we report when a transaction gets passed
to hardware. Change that function so it takes transaction rather
than a channel as its argument, and derive the channel from the
transaction. Rename the function accordingly.
Delete the header comments above the function definition; the ones
above the declaration in "gsi_private.h" should suffice. In
addition, the comments above gsi_channel_tx_update() do a fine job
of explaining what's going on.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Elder [Fri, 10 Jun 2022 15:46:14 +0000 (10:46 -0500)]
net: ipa: determine channel from event
Each event in an event ring describes the TRE whose completion
caused the event. Currently, every event ring is dedicated to a
single channel, so the channel is easily derived from the event
ring.
An event ring can actually be shared by more than one channel
though, and to distinguish events for one channel from another, the
event structure contains a field indicating which channel the event
is associated with.
In gsi_event_trans(), use the channel ID in an event to determine
which channel the event is for. This makes the channel pointer now
passed to that function irrelevant; pass the GSI pointer to that
function instead.
And although it shouldn't happen, warn if an event arrives that
records a channel ID that's not in use, or if the event does not
have a transaction associated with it.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
When a GSI transaction completes, ipa_endpoint_trans_complete() is
eventually called. That handles TX and RX completions separately,
but ipa_endpoint_tx_complete() is a no-op.
Instead, have ipa_endpoint_trans_complete() return immediately for a
TX transaction, and incorporate code from ipa_endpoint_rx_complete()
to handle RX transactions.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Elder [Fri, 10 Jun 2022 15:46:12 +0000 (10:46 -0500)]
net: ipa: rename endpoint->trans_tre_max
The trans_tre_max field of the IPA endpoint structure is only used
to limit the number of fragments allowed for an SKB being prepared
for transmission. Recognizing that, rename the field skb_frag_max,
and reduce its value by 1.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Elder [Fri, 10 Jun 2022 15:46:11 +0000 (10:46 -0500)]
net: ipa: rename channel->tlv_count
Each GSI channel has a TLV FIFO of a certain size, specified in the
configuration data for an AP channel. That size dictates the
maximum number of TREs that are allowed in a single transaction.
The only way that value is used after initialization is as a limit
on the number of TREs in a transaction; calling it "tlv_count"
isn't helpful, and in fact gsi_channel_trans_tre_max() exists to
sort of abstract it.
Instead, rename the channel->tlv_count field trans_tre_max, and get
rid of the helper function. Update a couple of comments as well.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Elder [Fri, 10 Jun 2022 15:46:10 +0000 (10:46 -0500)]
net: ipa: verify command channel TLV count
In commit 8797972afff3d ("net: ipa: remove command info pool"), the
maximum number of IPA commands that would be sent in a single
transaction was defined. That number can't exceed the size of the
TLV FIFO on the command channel, and we can check that at runtime.
To add this check, pass a new flag to gsi_channel_data_valid() to
indicate the channel being checked is being used for IPA commands.
Knowing that we can also verify the channel direction is correct.
Use a new local variable that refers to the command-specific portion
of the data being checked.
Signed-off-by: Alex Elder <elder@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Guangbin Huang [Sat, 11 Jun 2022 12:25:29 +0000 (20:25 +0800)]
net: hns3: fix tm port shapping of fibre port is incorrect after driver initialization
Currently in driver initialization process, driver will set shapping
parameters of tm port to default speed read from firmware. However, the
speed of SFP module may not be default speed, so shapping parameters of
tm port may be incorrect.
To fix this problem, driver sets new shapping parameters for tm port
after getting exact speed of SFP module in this case.
Fixes: 88d10bd6f730 ("net: hns3: add support for multiple media type") Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jie Wang [Sat, 11 Jun 2022 12:25:28 +0000 (20:25 +0800)]
net: hns3: fix PF rss size initialization bug
Currently hns3 driver misuses the VF rss size to initialize the PF rss size
in hclge_tm_vport_tc_info_update. So this patch fix it by checking the
vport id before initialization.
Fixes: 7347255ea389 ("net: hns3: refactor PF rss get APIs with new common rss get APIs") Signed-off-by: Jie Wang <wangjie125@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Guangbin Huang [Sat, 11 Jun 2022 12:25:27 +0000 (20:25 +0800)]
net: hns3: restore tm priority/qset to default settings when tc disabled
Currently, settings parameters of schedule mode, dwrr, shaper of tm
priority or qset of one tc are only be set when tc is enabled, they are
not restored to the default settings when tc is disabled. It confuses
users when they cat tm_priority or tm_qset files of debugfs. So this
patch fixes it.
Fixes: 848440544b41 ("net: hns3: Add support of TX Scheduler & Shaper to HNS3 driver") Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jie Wang [Sat, 11 Jun 2022 12:25:26 +0000 (20:25 +0800)]
net: hns3: modify the ring param print info
Currently tx push is also a ring param. So the original ring param print
info in hns3_is_ringparam_changed should be adjusted.
Fixes: 07fdc163ac88 ("net: hns3: refactor hns3_set_ringparam()") Signed-off-by: Jie Wang <wangjie125@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jian Shen [Sat, 11 Jun 2022 12:25:25 +0000 (20:25 +0800)]
net: hns3: don't push link state to VF if unalive
It's unnecessary to push link state to unalive VF, and the VF will
query link state from PF when it being start works.
Fixes: 18b6e31f8bf4 ("net: hns3: PF add support for pushing link status to VFs") Signed-off-by: Jian Shen <shenjian15@huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Guangbin Huang [Sat, 11 Jun 2022 12:25:24 +0000 (20:25 +0800)]
net: hns3: set port base vlan tbl_sta to false before removing old vlan
When modify port base vlan, the port base vlan tbl_sta needs to set to
false before removing old vlan, to indicate this operation is not finish.
Fixes: c0f46de30c96 ("net: hns3: fix port base vlan add fail when concurrent with reset") Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Pavel Begunkov [Sun, 12 Jun 2022 13:31:38 +0000 (14:31 +0100)]
io_uring: fix double unlock for pbuf select
io_buffer_select(), which is the only caller of io_ring_buffer_select(),
fully handles locking, mutex unlock in io_ring_buffer_select() will lead
to double unlock.
Fixes: c7fb19428d67d ("io_uring: add support for ring mapped supplied buffers") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Hao Xu [Sat, 11 Jun 2022 12:29:52 +0000 (20:29 +0800)]
io_uring: kbuf: fix bug of not consuming ring buffer in partial io case
When we use ring-mapped provided buffer, we should consume it before
arm poll if partial io has been done. Otherwise the buffer may be used
by other requests and thus we lost the data.
Fixes: c7fb19428d67 ("io_uring: add support for ring mapped supplied buffers") Signed-off-by: Hao Xu <howeyxu@tencent.com>
[pavel: 5.19 rebase] Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Ever since {suspend,resume}_device_irqs() were introduced in 2009
by commit 0a0c5168df27 ("PM: Introduce functions for suspending and
resuming device interrupts"), they've been exported even though there
are no module users and never will be: The functions are solely called
by the PM core, which is always built-in. Unexport them.
Nirmoy Das [Wed, 25 May 2022 09:59:55 +0000 (11:59 +0200)]
drm/i915: Individualize fences before adding to dma_resv obj
_i915_vma_move_to_active() can receive > 1 fences for
multiple batch buffers submission. Because dma_resv_add_fence()
can only accept one fence at a time, change _i915_vma_move_to_active()
to be aware of multiple fences so that it can add individual
fences to the dma resv object.
v6: fix multi-line comment.
v5: remove double fence reservation for batch VMAs.
v4: Reserve fences for composite_fence on multi-batch contexts and
also reserve fence slots to composite_fence for each VMAs.
v3: dma_resv_reserve_fences is not cumulative so pass num_fences.
v2: make sure to reserve enough fence slots before adding.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5614 Fixes: 544460c33821 ("drm/i915: Multi-BB execbuf") Cc: <stable@vger.kernel.org> # v5.16+ Signed-off-by: Nirmoy Das <nirmoy.das@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com> Signed-off-by: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20220525095955.15371-1-nirmoy.das@intel.com
(cherry picked from commit 420a07b841d03f6a436d8c06571c69aa5c783897) Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Ashutosh Dixit [Wed, 25 May 2022 13:19:20 +0000 (06:19 -0700)]
drm/i915/gt: Fix memory leaks in per-gt sysfs
All kmalloc'd kobjects need a kobject_put() to free memory. For example in
previous code, kobj_gt_release() never gets called. The requirement of
kobject_put() now results in a slightly different code organization.
Alan Previn [Fri, 11 Mar 2022 00:43:11 +0000 (16:43 -0800)]
drm/i915/reset: Fix error_state_read ptr + offset use
Fix our pointer offset usage in error_state_read
when there is no i915_gpu_coredump but buf offset
is non-zero.
This fixes a kernel page fault can happen when
multiple tests are running concurrently in a loop
and one is producing engine resets and consuming
the i915 error_state dump while the other is
forcing full GT resets. (takes a while to trigger).
The bindings already uses <dt-bindings/clock/r9a0*-cpg.h> to refer to
the header files with DT binding definitions for core clocks.
Use more wildcards to simplify more references to these files.
The register block base pointer as stored in the reg member of the
rz_cpg structure is only used during initialization. Hence move
it to a local variable, and pass it as a parameter to
rz_cpg_register_clock().
After this, the data member is the only remaining member of the rz_cpg
structure, so the whole structure can be replaced by the data member.
The register block base pointer as stored in the reg member of the
sh73a0_cpg structure is only used during initialization. Hence move
it to a local variable, and pass it as a parameter to
sh73a0_cpg_register_clock().
The register block base pointer as stored in the reg member of the
r8a7740_cpg structure is only used during initialization. Hence move
it to a local variable, and pass it as a parameter to
r8a7740_cpg_register_clock().
The register block base pointer as stored in the reg member of the
r8a73a4_cpg structure is only used during initialization. Hence move
it to a local variable, and pass it as a parameter to
r8a73a4_cpg_register_clock().
Reduce the sizes of the rcar_sysc_area and rcar_gen4_sysc_area
structures:
- parent is in the range [-1..32] or [-1..64], so s8 is sufficient,
- flags needs just 3 bits (for now), so u8 is plenty.
This reduces kernel size by:
- 288 bytes for an R-Car Gen1+Gen2 kernel (arm32),
- 1772 bytes for an R-Car Gen3+Gen4 kernel (arm64).
Jani Nikula [Wed, 8 Jun 2022 07:50:35 +0000 (10:50 +0300)]
drm/edid: add new interfaces around struct drm_edid
Add new functions drm_edid_read(), drm_edid_read_ddc(), and
drm_edid_read_custom() to replace drm_get_edid() and drm_do_get_edid()
for reading the EDID. The transition is expected to happen over a fairly
long time.
Note that the new drm_edid_read*() functions do not do any of the
connector updates anymore. The reading and parsing will be completely
separated from each other.
Add new functions drm_edid_alloc(), drm_edid_dup(), and drm_edid_free()
for allocating and freeing drm_edid containers.
Jani Nikula [Wed, 8 Jun 2022 07:50:31 +0000 (10:50 +0300)]
drm/edid: fix CTA data block collection size for CTA version 3
The CTA Data Block Collection is valid only for CTA extension version
3. In versions 1 and 2, it is a reserved block, which we ignore.
The DTD start offset (byte 2, or d in CTA-861 spec), which determines
the CTA Data Block Collection size, is specified slightly differently
for different versions:
Version 1:
d = offset for the byte following the reserved data block. If no
data is provided in the reserved data block, then d=4. If no DTDs
are provided, then d=0
Version 2:
d = offset for the byte following the reserved data block. If no
data is provided in the reserved data block, then d=4. If d=0, then
no detailed timing descriptors are provided, and no data is provided
in the reserved data block.
Version 3:
d = offset for the byte following the data block collection. If no
data is provided in the data block collection, then d=4. If d=0,
then no detailed timing descriptors are provided, and no data is
provided in the data block collection.
Ever since commit 9e50b9d55e9c ("drm: edid: Add some bounds checking"),
we've interpreted 0 to mean there are no DTDs but it's all Data
Blocks. Per the spec, Data Blocks are only valid for version 3, where we
should interpret 0 to mean there are no data blocks.
Follow the spec (and hope the EDIDs follow it too).
Rob Herring [Fri, 10 Jun 2022 21:33:07 +0000 (15:33 -0600)]
arm64: dts: arm/juno: Drop erroneous 'mbox-name' property
The 'mbox-name' property in the Juno mailbox node is undocumented and
unused. It's the consumer side of the mailbox binding that have
'mbox-names' properties.
The rtsm-display panel timing node was removed in commit 928faf5e3e8d
("arm64: dts: fvp: Remove panel timings"). Without the node, 'panel-dpi'
is not needed either.
Refresh the defconfig for Renesas ARM systems:
- Move CONFIG_SLAB=y (moved in commit 7b42f1041c98f5d7 ("mm: Kconfig:
move swap and slab config options to the MM section")),
- Enable CONFIG_RENESAS_RZN1WDT (RZN1D-DB),
- Enable CONFIG_RTC_DRV_RZN1 (RZN1-DB).
Pavel Begunkov [Thu, 9 Jun 2022 07:34:35 +0000 (08:34 +0100)]
io_uring: fix not locked access to fixed buf table
We can look inside the fixed buffer table only while holding
->uring_lock, however in some cases we don't do the right async prep for
IORING_OP_{WRITE,READ}_FIXED ending up with NULL req->imu forcing making
an io-wq worker to try to resolve the fixed buffer without proper
locking.
Move req->imu setup into early req init paths, i.e. io_prep_rw(), which
is called unconditionally for rw requests and under uring_lock.
Pavel Begunkov [Mon, 13 Jun 2022 05:30:06 +0000 (06:30 +0100)]
io_uring: fix races with buffer table unregister
Fixed buffer table quiesce might unlock ->uring_lock, potentially
letting new requests to be submitted, don't allow those requests to
use the table as they will race with unregistration.
Reported-and-tested-by: van fantasy <g1042620637@gmail.com> Fixes: bd54b6fe3316ec ("io_uring: implement fixed buffers registration similar to fixed files") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Pavel Begunkov [Mon, 13 Jun 2022 05:32:44 +0000 (06:32 +0100)]
io_uring: fix races with file table unregister
Fixed file table quiesce might unlock ->uring_lock, potentially letting
new requests to be submitted, don't allow those requests to use the
table as they will race with unregistration.
Reported-and-tested-by: van fantasy <g1042620637@gmail.com> Fixes: 05f3fb3c53975 ("io_uring: avoid ring quiesce for fixed file set unregister and update") Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Tianchen Ding [Wed, 8 Jun 2022 23:34:12 +0000 (07:34 +0800)]
sched: Remove the limitation of WF_ON_CPU on wakelist if wakee cpu is idle
Wakelist can help avoid cache bouncing and offload the overhead of waker
cpu. So far, using wakelist within the same llc only happens on
WF_ON_CPU, and this limitation could be removed to further improve
wakeup performance.
The commit 518cd6234178 ("sched: Only queue remote wakeups when
crossing cache boundaries") disabled queuing tasks on wakelist when
the cpus share llc. This is because, at that time, the scheduler must
send IPIs to do ttwu_queue_wakelist. Nowadays, ttwu_queue_wakelist also
supports TIF_POLLING, so this is not a problem now when the wakee cpu is
in idle polling.
Benefits:
Queuing the task on idle cpu can help improving performance on waker cpu
and utilization on wakee cpu, and further improve locality because
the wakee cpu can handle its own rq. This patch helps improving rt on
our real java workloads where wakeup happens frequently.
Consider the normal condition (CPU0 and CPU1 share same llc)
Before this patch:
We see CPU0 can finish its work earlier. It only needs to put task to
wakelist and return.
While CPU1 is idle, so let itself handle its own runqueue data.
This patch brings no difference about IPI.
This patch only takes effect when the wakee cpu is:
1) idle polling
2) idle not polling
For 1), there will be no IPI with or without this patch.
For 2), there will always be an IPI before or after this patch.
Before this patch: waker cpu will enqueue task and check preempt. Since
"idle" will be sure to be preempted, waker cpu must send a resched IPI.
After this patch: waker cpu will put the task to the wakelist of wakee
cpu, and send an IPI.
Benchmark:
We've tested schbench, unixbench, and hachbench on both x86 and arm64.
On x86 (Intel Xeon Platinum 8269CY):
schbench -m 2 -t 8
hackbench -g 1 -l 100000
before after
Time 4.217 2.916
Our patch has improvement on schbench, hackbench
and Pipe-based Context Switching of unixbench
when there exists idle cpus,
and no obvious regression on other tests of unixbench.
This can help improve rt in scenes where wakeup happens frequently.
Tianchen Ding [Wed, 8 Jun 2022 23:34:11 +0000 (07:34 +0800)]
sched: Fix the check of nr_running at queue wakelist
The commit 2ebb17717550 ("sched/core: Offload wakee task activation if it
the wakee is descheduling") checked rq->nr_running <= 1 to avoid task
stacking when WF_ON_CPU.
Per the ordering of writes to p->on_rq and p->on_cpu, observing p->on_cpu
(WF_ON_CPU) in ttwu_queue_cond() implies !p->on_rq, IOW p has gone through
the deactivate_task() in __schedule(), thus p has been accounted out of
rq->nr_running. As such, the task being the only runnable task on the rq
implies reading rq->nr_running == 0 at that point.
Josh Don [Thu, 9 Jun 2022 02:55:15 +0000 (19:55 -0700)]
sched: Allow newidle balancing to bail out of load_balance
While doing newidle load balancing, it is possible for new tasks to
arrive, such as with pending wakeups. newidle_balance() already accounts
for this by exiting the sched_domain load_balance() iteration if it
detects these cases. This is very important for minimizing wakeup
latency.
However, if we are already in load_balance(), we may stay there for a
while before returning back to newidle_balance(). This is most
exacerbated if we enter a 'goto redo' loop in the LBF_ALL_PINNED case. A
very straightforward workaround to this is to adjust should_we_balance()
to bail out if we're doing a CPU_NEWLY_IDLE balance and new tasks are
detected.
This was tested with the following reproduction:
- two threads that take turns sleeping and waking each other up are
affined to two cores
- a large number of threads with 100% utilization are pinned to all
other cores
Without this patch, wakeup latency was ~120us for the pair of threads,
almost entirely spent in load_balance(). With this patch, wakeup latency
is ~6us.
Yajun Deng [Tue, 7 Jun 2022 10:18:07 +0000 (18:18 +0800)]
sched/deadline: Use proc_douintvec_minmax() limit minimum value
sysctl_sched_dl_period_max and sysctl_sched_dl_period_min are unsigned
integer, but proc_dointvec() wouldn't return error even if we set a
negative number.
Use proc_douintvec_minmax() instead of proc_dointvec(). Add extra1 for
sysctl_sched_dl_period_max and extra2 for sysctl_sched_dl_period_min.
It's just an optimization for match data and proc_handler in struct
ctl_table. The 'if (period < min || period > max)' in __checkparam_dl()
will work fine even if there hasn't this patch.
Chengming Zhou [Wed, 1 Jun 2022 02:18:48 +0000 (10:18 +0800)]
sched/fair: Optimize and simplify rq leaf_cfs_rq_list
We notice the rq leaf_cfs_rq_list has two problems when do bugfix
backports and some test profiling.
1. cfs_rqs under throttled subtree could be added to the list, and
make their fully decayed ancestors on the list, even though not needed.
2. #1 also make the leaf_cfs_rq_list management complex and error prone,
this is the list of related bugfix so far:
commit 31bc6aeaab1d ("sched/fair: Optimize update_blocked_averages()")
commit fe61468b2cbc ("sched/fair: Fix enqueue_task_fair warning")
commit b34cb07dde7c ("sched/fair: Fix enqueue_task_fair() warning some more")
commit 39f23ce07b93 ("sched/fair: Fix unthrottle_cfs_rq() for leaf_cfs_rq list")
commit 0258bdfaff5b ("sched/fair: Fix unfairness caused by missing load decay")
commit a7b359fc6a37 ("sched/fair: Correctly insert cfs_rq's to list on unthrottle")
commit fdaba61ef8a2 ("sched/fair: Ensure that the CFS parent is added after unthrottling")
commit 2630cde26711 ("sched/fair: Add ancestors of unthrottled undecayed cfs_rq")
commit 31bc6aeaab1d ("sched/fair: Optimize update_blocked_averages()")
delete every cfs_rq under throttled subtree from rq->leaf_cfs_rq_list,
and delete the throttled_hierarchy() test in update_blocked_averages(),
which optimized update_blocked_averages().
But those later bugfix add cfs_rqs under throttled subtree back to
rq->leaf_cfs_rq_list again, with their fully decayed ancestors, for
the integrity of rq->leaf_cfs_rq_list.
This patch takes another method, skip all cfs_rqs under throttled
hierarchy when list_add_leaf_cfs_rq(), to completely make cfs_rqs
under throttled subtree off the leaf_cfs_rq_list.
So we don't need to consider throttled related things in
enqueue_entity(), unthrottle_cfs_rq() and enqueue_task_fair(),
which simplify the code a lot. Also optimize update_blocked_averages()
since cfs_rqs under throttled hierarchy and their ancestors
won't be on the leaf_cfs_rq_list.
sched/fair: Consider CPU affinity when allowing NUMA imbalance in find_idlest_group()
In the case of systems containing multiple LLCs per socket, like
AMD Zen systems, users want to spread bandwidth hungry applications
across multiple LLCs. Stream is one such representative workload where
the best performance is obtained by limiting one stream thread per LLC.
To ensure this, users are known to pin the tasks to a specify a subset
of the CPUs consisting of one CPU per LLC while running such bandwidth
hungry tasks.
Suppose we kickstart a multi-threaded task like stream with 8 threads
using taskset or numactl to run on a subset of CPUs on a 2 socket Zen3
server where each socket contains 128 CPUs
(0-63,128-191 in one socket, 64-127,192-255 in another socket)
Eg: numactl -C 0,16,32,48,64,80,96,112 ./stream8
Here each CPU in the list is from a different LLC and 4 of those LLCs
are on one socket, while the other 4 are on another socket.
Ideally we would prefer that each stream thread runs on a different
CPU from the allowed list of CPUs. However, the current heuristics in
find_idlest_group() do not allow this during the initial placement.
Suppose the first socket (0-63,128-191) is our local group from which
we are kickstarting the stream tasks. The first four stream threads
will be placed in this socket. When it comes to placing the 5th
thread, all the allowed CPUs are from the local group (0,16,32,48)
would have been taken.
However, the current scheduler code simply checks if the number of
tasks in the local group is fewer than the allowed numa-imbalance
threshold. This threshold was previously 25% of the NUMA domain span
(in this case threshold = 32) but after the v6 of Mel's patchset
"Adjust NUMA imbalance for multiple LLCs", got merged in sched-tip,
Commit: e496132ebedd ("sched/fair: Adjust the allowed NUMA imbalance
when SD_NUMA spans multiple LLCs") it is now equal to number of LLCs
in the NUMA domain, for processors with multiple LLCs.
(in this case threshold = 8).
For this example, the number of tasks will always be within threshold
and thus all the 8 stream threads will be woken up on the first socket
thereby resulting in sub-optimal performance.
The following sched_wakeup_new tracepoint output shows the initial
placement of tasks in the current tip/sched/core on the Zen3 machine:
Once the first four threads are distributed among the allowed CPUs of
socket one, the rest of the treads start piling on these same CPUs
when clearly there are CPUs on the second socket that can be used.
Following the initial pile up on a small number of CPUs, though the
load-balancer eventually kicks in, it takes a while to get to {4}{4}
and even {4}{4} isn't stable as we observe a bunch of ping ponging
between {4}{4} to {5}{3} and back before a stable state is reached
much later (1 Stream thread per allowed CPU) and no more migration is
required.
We can detect this piling and avoid it by checking if the number of
allowed CPUs in the local group are fewer than the number of tasks
running in the local group and use this information to spread the
5th task out into the next socket (after all, the goal in this
slowpath is to find the idlest group and the idlest CPU during the
initial placement!).
The following sched_wakeup_new tracepoint output shows the initial
placement of tasks after adding this fix on the Zen3 machine:
We see that threads are using all of the allowed CPUs and there is
no pileup.
No output is generated for tracepoint sched_migrate_task with this
patch due to a perfect initial placement which removes the need
for balancing later on - both across NUMA boundaries and within
NUMA boundaries for stream.
Following are the results from running 8 Stream threads with and
without pinning on a dual socket Zen3 Machine (2 x 64C/128T):
During the testing of this patch, the tip sched/core was at
commit: 089c02ae2771 "ftrace: Use preemption model accessors for trace
header printout"
Pinning is done using: numactl -C 0,16,32,48,64,80,96,112 ./stream8
5.18.0-rc1 5.18.0-rc1 5.18.0-rc1
tip sched/core tip sched/core tip sched/core
(no pinning) + pinning + this-patch
+ pinning
Pinning currently hurts the performance compared to unbound case on
tip/sched/core. With the addition of this patch, we are able to
outperform tip/sched/core by a good margin with pinning.
Following are the results from running 16 Stream threads with and
without pinning on a dual socket IceLake Machine (2 x 32C/64T):
NUMA Topology of Intel Skylake machine:
Node 1: 0,2,4,6 ... 126 (Even numbers)
Node 2: 1,3,5,7 ... 127 (Odd numbers)
Pinning is done using: numactl -C 0-15 ./stream16
5.18.0-rc1 5.18.0-rc1 5.18.0-rc1
tip sched/core tip sched/core tip sched/core
(no pinning) +pinning + this-patch
+ pinning
In case of Icelake machine, with single LLC per socket, pinning across
the two sockets reduces cache contention, thus showing great
improvement in pinned case which is further benefited by this patch.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Acked-by: Mel Gorman <mgorman@techsingularity.net> Link: https://lkml.kernel.org/r/20220407111222.22649-1-kprateek.nayak@amd.com
Mel Gorman [Fri, 20 May 2022 10:35:19 +0000 (11:35 +0100)]
sched/numa: Adjust imb_numa_nr to a better approximation of memory channels
For a single LLC per node, a NUMA imbalance is allowed up until 25%
of CPUs sharing a node could be active. One intent of the cut-off is
to avoid an imbalance of memory channels but there is no topological
information based on active memory channels. Furthermore, there can
be differences between nodes depending on the number of populated
DIMMs.
A cut-off of 25% was arbitrary but generally worked. It does have a severe
corner cases though when an parallel workload is using 25% of all available
CPUs over-saturates memory channels. This can happen due to the initial
forking of tasks that get pulled more to one node after early wakeups
(e.g. a barrier synchronisation) that is not quickly corrected by the
load balancer. The LB may fail to act quickly as the parallel tasks are
considered to be poor migrate candidates due to locality or cache hotness.
On a range of modern Intel CPUs, 12.5% appears to be a better cut-off
assuming all memory channels are populated and is used as the new cut-off
point. A minimum of 1 is specified to allow a communicating pair to
remain local even for CPUs with low numbers of cores. For modern AMDs,
there are multiple LLCs and are not affected.
The imbalance limitations are applied inconsistently at fork time
and at runtime. At fork, a new task can remain local until there are
too many running tasks even if the degree of imbalance is larger than
NUMA_IMBALANCE_MIN which is different to runtime. Secondly, the imbalance
figure used during load balancing is different to the one used at NUMA
placement. Load balancing uses the number of tasks that must move to
restore imbalance where as NUMA balancing uses the total imbalance.
In combination, it is possible for a parallel workload that uses a small
number of CPUs without applying scheduler policies to have very variable
run-to-run performance.
[lkp@intel.com: Fix build breakage for arc-allyesconfig]
Mel Gorman [Fri, 20 May 2022 10:35:17 +0000 (11:35 +0100)]
sched/numa: Do not swap tasks between nodes when spare capacity is available
If a destination node has spare capacity but there is an imbalance then
two tasks are selected for swapping. If the tasks have no numa group
or are within the same NUMA group, it's simply shuffling tasks around
without having any impact on the compute imbalance. Instead, it's just
punishing one task to help another.
Mel Gorman [Fri, 20 May 2022 10:35:16 +0000 (11:35 +0100)]
sched/numa: Initialise numa_migrate_retry
On clone, numa_migrate_retry is inherited from the parent which means
that the first NUMA placement of a task is non-deterministic. This
affects when load balancing recognises numa tasks and whether to
migrate "regular", "remote" or "all" tasks between NUMA scheduler
domains.
locking/lockdep: Use sched_clock() for random numbers
Since the rewrote of prandom_u32(), in the commit mentioned below, the
function uses sleeping locks which extracing random numbers and filling
the batch.
This breaks lockdep on PREEMPT_RT because lock_pin_lock() disables
interrupts while calling __lock_pin_lock(). This can't be moved earlier
because the main user of the function (rq_pin_lock()) invokes that
function after disabling interrupts in order to acquire the lock.
The cookie does not require random numbers as its goal is to provide a
random value in order to notice unexpected "unlock + lock" sites.
Use sched_clock() to provide random numbers.
Fixes: a0103f4d86f88 ("random32: use real rng for non-deterministic randomness") Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/YoNn3pTkm5+QzE5k@linutronix.de
Furkan Kardame [Sat, 11 Jun 2022 17:08:52 +0000 (20:08 +0300)]
arm64: dts: meson: add gpio-fan control to GS-King-X
GS-King-X has a single speed GPIO fan which is always-on by default.
If we add controls for the fan and a trip point,
the fan stays off most of the time, reducing background
noise from the unit.
The current RDPMC assignment scheme maps four DF PMCs and
six L3 PMCs from index 6 to 15.
If AMD Performance Monitoring Version 2 (PerfMonV2) is
supported, there may be additional DF counters available
which are mapped starting from index 16 i.e. just after
the L3 counters. Update the RDPMC assignments accordingly.
Sandipan Das [Thu, 19 May 2022 10:03:33 +0000 (15:33 +0530)]
perf/x86/amd/uncore: Add PerfMonV2 DF event format
If AMD Performance Monitoring Version 2 (PerfMonV2) is
supported, use bits 0-7, 32-37 as EventSelect and bits
8-15, 24-27 as UnitMask for Data Fabric (DF) events.
Sandipan Das [Thu, 19 May 2022 10:03:32 +0000 (15:33 +0530)]
perf/x86/amd/uncore: Detect available DF counters
If AMD Performance Monitoring Version 2 (PerfMonV2) is
supported, use CPUID leaf 0x80000022 EBX to detect the
number of Data Fabric (DF) PMCs. This offers more
flexibility if the counts change in later processor
families.
Sandipan Das [Thu, 19 May 2022 10:03:31 +0000 (15:33 +0530)]
perf/x86/amd/uncore: Use attr_update for format attributes
Use the update_attrs attribute group introduced by commit f3a3a8257e5a ("perf/core: Add attr_groups_update into struct
pmu") and the is_visible() callback to populate the family
specifc attributes for uncore events.
The changes apply to attributes that are unique to families
such as slicemask for Family 17h and coreid for Family 19h.
The addition of common attributes such as event and umask,
whose formats change across families, remain unchanged.
Sandipan Das [Thu, 19 May 2022 10:03:30 +0000 (15:33 +0530)]
perf/x86/amd/uncore: Use dynamic events array
If AMD Performance Monitoring Version 2 (PerfMonV2) is
supported, the number of available counters for a given
uncore PMU may not be fixed across families and models
and has to be determined at runtime.
The per-cpu uncore PMU data currently uses a fixed-sized
array for event information. Make it dynamic based on the
number of available counters.
Like Xu [Thu, 19 May 2022 15:19:13 +0000 (23:19 +0800)]
x86/events/intel/ds: Enable large PEBS for PERF_SAMPLE_WEIGHT_TYPE
All the information required by the PERF_SAMPLE_WEIGHT is
available in the pebs record. Thus large PEBS could be enabled
for PERF_SAMPLE_WEIGHT sample type to save PMIs overhead until
other non-compatible flags such as PERF_SAMPLE_DATA_PAGE_SIZE
(due to lack of munmap tracking) stop it.
To cover new weight extension, add PERF_SAMPLE_WEIGHT_TYPE
to the guardian LARGE_PEBS_FLAGS.
Tested it with:
$ perf mem record -c 1000 workload
Before: Captured and wrote 0.126 MB perf.data (958 samples) [958 PMIs]
After: Captured and wrote 0.313 MB perf.data (4859 samples) [3 PMIs]
Peter Zijlstra [Tue, 7 Jun 2022 20:41:55 +0000 (22:41 +0200)]
sched: Fix balance_push() vs __sched_setscheduler()
The purpose of balance_push() is to act as a filter on task selection
in the case of CPU hotplug, specifically when taking the CPU out.
It does this by (ab)using the balance callback infrastructure, with
the express purpose of keeping all the unlikely/odd cases in a single
place.
In order to serve its purpose, the balance_push_callback needs to be
(exclusively) on the callback list at all times (noting that the
callback always places itself back on the list the moment it runs,
also noting that when the CPU goes down, regular balancing concerns
are moot, so ignoring them is fine).
And here-in lies the problem, __sched_setscheduler()'s use of
splice_balance_callbacks() takes the callbacks off the list across a
lock-break, making it possible for, an interleaving, __schedule() to
see an empty list and not get filtered.
Josh Poimboeuf [Thu, 9 Jun 2022 07:17:32 +0000 (00:17 -0700)]
x86/mm: Fix RESERVE_BRK() for older binutils
With binutils 2.26, RESERVE_BRK() causes a build failure:
/tmp/ccnGOKZ5.s: Assembler messages:
/tmp/ccnGOKZ5.s:98: Error: missing ')'
/tmp/ccnGOKZ5.s:98: Error: missing ')'
/tmp/ccnGOKZ5.s:98: Error: missing ')'
/tmp/ccnGOKZ5.s:98: Error: junk at end of line, first unrecognized
character is `U'
The problem is this line:
RESERVE_BRK(early_pgt_alloc, INIT_PGT_BUF_SIZE)
Specifically, the INIT_PGT_BUF_SIZE macro which (via PAGE_SIZE's use
_AC()) has a "1UL", which makes older versions of the assembler unhappy.
Unfortunately the _AC() macro doesn't work for inline asm.
Inline asm was only needed here to convince the toolchain to add the
STT_NOBITS flag. However, if a C variable is placed in a section whose
name is prefixed with ".bss", GCC and Clang automatically set
STT_NOBITS. In fact, ".bss..page_aligned" already relies on this trick.
So fix the build failure (and simplify the macro) by allocating the
variable in C.
Also, add NOLOAD to the ".brk" output section clause in the linker
script. This is a failsafe in case the ".bss" prefix magic trick ever
stops working somehow. If there's a section type mismatch, the GNU
linker will force the ".brk" output section to be STT_NOBITS. The LLVM
linker will fail with a "section type mismatch" error.
Note this also changes the name of the variable from .brk.##name to
__brk_##name. The variable names aren't actually used anywhere, so it's
harmless.
Fixes: a1e2c031ec39 ("x86/mm: Simplify RESERVE_BRK()") Reported-by: Joe Damato <jdamato@fastly.com> Reported-by: Byungchul Park <byungchul.park@lge.com> Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: Joe Damato <jdamato@fastly.com> Link: https://lore.kernel.org/r/22d07a44c80d8e8e1e82b9a806ddc8c6bbb2606e.1654759036.git.jpoimboe@kernel.org
Yang Yingliang [Mon, 13 Jun 2022 06:34:54 +0000 (14:34 +0800)]
drm/syncobj: add missing error return code in drm_syncobj_transfer_to_timeline()
If dma_fence_unwrap_merge() fails, it should return error code
in drm_syncobj_transfer_to_timeline()
Fixes: ec8d985ff26f ("drm: use dma_fence_unwrap_merge() in drm_syncobj") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20220613063454.2609364-1-yangyingliang@huawei.com
Sergey Shtylyov [Fri, 13 May 2022 20:50:14 +0000 (23:50 +0300)]
ata: libata-core: fix sloppy typing in ata_id_n_sectors()
The code multiplying the # of cylinders/heads/sectors in ata_id_n_sectors()
to get a disk capacity implicitly uses the *int* type for that calculation
and casting the result to 'u64' before returning ensues a sign extension.
Explicitly casting the 'u16' typed multipliers to 'u32' results in avoiding
a sign extension instruction and so in a more compact code...
Found by Linux Verification Center (linuxtesting.org) with the SVACE static
analysis tool.
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Sergey Shtylyov [Sun, 15 May 2022 20:41:26 +0000 (23:41 +0300)]
ata: pata_hpt3x2n: pass base DPLL frequency to hpt3x2n_pci_clock()
Currently, the base DPLL frequency is hardcoded in hpt3x2n_pci_clock().
Align with the updated 'pata_hpt37x' driver, where this frequency is a
parameter to hpt37x_pci_clock().
While at it, also do the following to align with the 'pata_hpt37x' driver:
- fix the 'freq' local variable's type;
- remove the 'iobase' local variable;
- extend the comment to the inl() call;
- move the 'total' local variable's declaration.
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Sergey Shtylyov [Sun, 15 May 2022 20:41:25 +0000 (23:41 +0300)]
ata: pata_hpt37x: merge hpt374_read_freq() to hpt37x_pci_clock()
With hpt374_read_freq() implemented as a separate function, there's
some code duplication going on, not to mention that this function is
named incorrectly: it returns the f_CNT register value saved by BIOS,
not the PCI clock frequency.
Folding hpt374_read_freq() into hpt37x_pci_clock() saves 20 bytes of
object code with x86_64 gcc 10.3.1...
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Sergey Shtylyov [Sun, 15 May 2022 20:41:24 +0000 (23:41 +0300)]
ata: pata_hpt37x: factor out hpt37x_pci_clock()
Factor out the PCI clock frequency detection code into hpt37x_pci_clock(),
so that this driver becomes more like 'pata_hpt3x2n'. Note that I decided
to change the way HPT374 is identified to using the PCI device ID...
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Sergey Shtylyov [Sun, 15 May 2022 20:41:23 +0000 (23:41 +0300)]
ata: pata_hpt37x: move claculating PCI clock from hpt37x_clock_slot()
hpt37x_init_one() incorrectly calls an averaged f_CNT register value 'freq'
and hpt37x_clock_slot() takes that value as the 'freq' parameter -- rename
the former variable to 'fcnt' and move the actual code calculating the PCI
clock from hpt37x_clock_slot() to hpt37x_init_one(), along with adding the
frequency clamping code, in preparation for the factoring out the PCI clock
detection, so that this driver would become more like the 'pata_hpt3x2n'
driver...
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru> Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Igor Torrente [Mon, 4 Apr 2022 20:45:07 +0000 (17:45 -0300)]
drm: vkms: Alloc the compose frame using vzalloc
Currently, the memory to the composition frame is being allocated using
the kzmalloc. This comes with the limitation of maximum size of one
page size(which in the x86_64 is 4Kb and 4MB for default and hugepage
respectively).
Somes test of igt (e.g. kms_plane@pixel-format) uses more than 4MB when
testing some pixel formats like ARGB16161616 and the following error were
showing up when running kms_plane@plane-panning-bottom-right*:
[drm:vkms_composer_worker [vkms]] *ERROR* Cannot allocate memory for
output frame.
This problem is addessed by allocating the memory using kvzalloc that
circunvents this limitation.
V5: Improve the commit message and drop the debugging issues in VKMS
TO-DO(Melissa Wen).
drm/vkms: check plane_composer->map[0] before using it
Fix a copypasta error. The caller of compose_plane() already checks
primary_composer->map. In contrast, plane_composer->map is never
verified here before handling.