| Age | Commit message (Collapse) | Author |
|---|
|
The MMU mapping flow is optimized for improve the performance, the
unmapping flow could also be optimized to follow same flow.
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
ipu6_mmu_map() operated on a per-page basis, it leads frequent
spin_lock/unlock() and clflush_cache_range() for each page, it
will cause inefficiencies especially when handling dma-bufs
with large number of pages. However, the pages are likely concentrated
pages by IOMMU DMA driver, IPU MMU driver can map the concentrated
pages into less entries in l1 table.
This change enhances ipu6_mmu_map() with batching process multiple
contiguous pages. It significantly reduces calls for spin_lock/unlock
and clflush_cache_range() and improve the performance.
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Jianhui Dai <jianhui.j.dai@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
l2_map() and l2_unmap() are better to be grouped together.
l2_unmap() will soon be called from l2_map() for mapping
optimization.
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Jianhui Dai <jianhui.j.dai@intel.com>
[Sakari Ailus: Rebase on debug print fixes on 32-bit.]
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Rename the pad field in __media_pipeline_start() to both better describe
what it is and avoid masking it during the loop.
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
The buttress ipc ish structure is not effectively used on IPU6 - data
is nullified on init. Remove the ish structure and handing of related
interrupts to cleanup the code.
Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
When using buffers from DRM, DMA-API gives a warning about: "mapping sg
segment longer than device claims to support [len=307200] [max=65536]"
Add a call to vb2_dma_contig_set_max_seg_size() to tell the DMA-API
about the supported segment size (which is UINT_MAX).
Fixes: 6edb685abb2a ("media: raspberrypi: Add support for RP1-CFE")
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Some IPU6 devices have shared interrupts. We need to handle properly
case when interrupt is triggered from other device on shared irq line
and IPU6 itself disabled. In such case we get 0xffffffff from
ISR_STATUS register and handle all irq's cases, for what we are not
not prepared and usually hang the whole system.
To avoid the issue use pm_runtime_get_if_active() to check if
the device is enabled and prevent suspending it when we handle irq
until the end of irq. Additionally use synchronize_irq() in suspend
Fixes: ab29a2478e70 ("media: intel/ipu6: add IPU6 buttress interface driver")
Cc: stable@vger.kernel.org
Signed-off-by: Stanislaw Gruszka <stanislaw.gruszka@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Hans de Goede <hdegoede@redhat.com> # ThinkPad X1 Yoga Gen 8, ov2740
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
2X.. mbus formats are related to parallel interface and
as such should not be acceptable on the sink pad of the bridge.
Only keep their csi counterpart in 1X..
Signed-off-by: Alain Volmat <alain.volmat@foss.st.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Add handling of pm_runtime in order to control the supplies
and clocks of the bridge.
Signed-off-by: Alain Volmat <alain.volmat@foss.st.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Add pad enable_streams and disable_streams ops in addition to
v4l2_subdev_s_stream_helper. Moreover, use functions
v4l2_subdev_enable_streams and v4l2_subdev_disable_streams to
control the source subdev.
Signed-off-by: Alain Volmat <alain.volmat@foss.st.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Following addition of cascaded s_stream call within
mipid02_stream_enable and mipid02_stream_disable,
the mipid02_stream_enable error handling must be fixed to simply
disable the clocks & data lanes without also calling
again the s_stream of the source subdev.
Moreover, main control registers are reset to 0 for all cases
of error within the mipid02_stream_enable even if they haven't yet
been written. Correct this behavior by simply returning an
error when register clear are not necessary.
Signed-off-by: Alain Volmat <alain.volmat@foss.st.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Introduce enable_streams and disable_streams pad ops and
replace s_stream with the v4l2_subdev_s_stream_helper
function.
Signed-off-by: Alain Volmat <alain.volmat@foss.st.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Add new structure 'mt9p031_model_info' to encapsulate format codes for
the mt9p031 camera sensor family. This approach enhances code clarity
and maintainability.
Signed-off-by: Tarang Raval <tarang.raval@siliconsignals.io>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
The ACPI version of this driver "just works" on dts based systems with a
few extensions to facilitate.
- Add support for DT based probing
- Add support for taking the part out of reset via a GPIO reset pin
- Add in regulator bulk on/off logic for the power rails.
Once done this sensor works nicely on a Qualcomm X1E80100 CRD.
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
The data-sheet and documentation for this part uses the name xvclk not
ext_clk for the input reference clock. Rename the variables and defines in
this driver to align with the data-sheet name.
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Add bindings for the already upstream OV08X40 to enable usage of this
sensor on DTS based systems.
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
[Sakari Ailus: Add the file to MAINTAINERS.]
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
It is necessary to account for I2C quirks in the burst mode path of this
driver. Not all I2C controllers can accept arbitrarily long writes and this
is represented in the quirks field of the adapter structure.
Prior to this patch the following error message is seen on a Qualcomm
X1E80100 CRD.
[ 38.773524] i2c i2c-2: adapter quirk: msg too long (addr 0x0036, size 290, write)
[ 38.781454] ov08x40 2-0036: Failed regs transferred: -95
[ 38.787076] ov08x40 2-0036: ov08x40_start_streaming failed to set regs
Fix the error by breaking up the write sequence into the advertised maximum
write size of the quirks field if the quirks field is populated.
Fixes: 8f667d202384 ("media: ov08x40: Reduce start streaming time")
Cc: stable@vger.kernel.org # v6.9+
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
In set_frame_rate(), select a rate in rate_0 or rate_1 by checking
sd->frame_rate >= r->fps in a loop, but the loop condition terminates when
the index reaches zero, which fails to check the last elememt in rate_0 or
rate_1.
Check for >= 0 so that the last one in rate_0 or rate_1 is also checked.
Fixes: 189d92af707e ("V4L/DVB (13422): gspca - ov534: ov772x changes from Richard Kaswy.")
Cc: stable@vger.kernel.org
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
It is not valid to call pm_runtime_set_suspended() for devices
with runtime PM enabled because it returns -EAGAIN if it is enabled
already and working. So, call pm_runtime_disable() before to fix it.
Cc: stable@vger.kernel.org
Fixes: af2c3834c8ca ("[media] media: venus: adding core part and helper functions")
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Acked-by: Stanimir Varbanov <stanimir.k.varbanov@gmail.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
It is not valid to call pm_runtime_set_suspended() for devices
with runtime PM enabled because it returns -EAGAIN if it is enabled
already and working. So, call pm_runtime_disable() before to fix it.
Cc: stable@vger.kernel.org
Fixes: b50a64fc54af ("media: amphion: add amphion vpu device driver")
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
It is not valid to call pm_runtime_set_suspended() and
pm_runtime_set_active() for devices with runtime PM enabled because it
returns -EAGAIN if it is enabled already and working. So, adjust the
order to fix it.
Cc: stable@vger.kernel.org
Fixes: 5f9a089b6de3 ("dw9768: Enable low-power probe on ACPI")
Suggested-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
v4l2_subdev_init_finalize() already sets the HAS_EVENTS flag if a
control handler is set. Let's drop the HAS_EVENTS flag.
Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Tommaso Merciai <tomm.merciai@gmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Michael Riesch <michael.riesch@wolfvision.net>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
v4l2_subdev_init_finalize() already sets the HAS_EVENTS flag if a
control handler is set, and subdev_do_ioctl() uses
v4l2_ctrl_subdev_subscribe_event() and v4l2_event_subdev_unsubscribe()
as defaults if the subdev doesn't have .(un)subscribe.
Let's drop the HAS_EVENTS flag and event handlers.
Signed-off-by: Tommaso Merciai <tomm.merciai@gmail.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Remove extraneous assigned-clock* from media/i2c/* schemas.
assigned-clock and assigned-clock-parent assume that there is a clock that
can be assigned and/or a clock parent that similarly can be assigned by
firmware which may not always be the case.
The sensor properties themselves should not mandate assigned-clock* as a
result.
Link: https://lore.kernel.org/linux-media/j7kgz2lyxnler5qwd7yiazdq6fmsv77kyozdrxf33h54ydakjz@uqjhwhoyv6re
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Remove the direct dependency on OF by using the fwnode API. This makes
the driver usable on non-OF systems, such as ACPI-based platforms.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
The mt9p031 driver doesn't support platform data anymore, and requires
the I2C device to be instantiated through a method that supports device
properties. As the driver lacks an ACPI device ID table, the device
matching will always go through the OF device ID table, either for OF
devices, or for ACPI devices using PRP0001. The I2C device ID table is
therefore not needed. Drop it.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
No user of the mt9p031_platform_data legacy platform data for board
files exist in the kernel anymore. Drop support from the driver.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
If devm_clk_get() fails then we need to free "cfe" before returning.
Fixes: 6edb685abb2a ("media: raspberrypi: Add support for RP1-CFE")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Controls can be exposed to userspace via a v4l-subdevX device, and
userspace has to be able to subscribe to control events so that it is
notified when the control changes value. If a control handler is set for
the subdev then set the HAS_EVENTS flag automatically into
v4l2_subdev_init_finalize() and use v4l2_ctrl_subdev_subscribe_event() and
v4l2_event_subdev_unsubscribe() as default if subdev don't have
.(un)subscribe control operations. This simplifies subdev drivers by
avoiding the need to set the V4L2_SUBDEV_FL_HAS_EVENTS flag and plug the
event handlers, and ensures consistency of the API exposed to userspace.
Signed-off-by: Tommaso Merciai <tomm.merciai@gmail.com>
Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
platform_get_irq() already provides a error message.
This fixes the following cocci error:
drivers/media/platform/raspberrypi/rp1-cfe/cfe.c:2326:2-9: line 2326 is redundant because platform_get_irq() already prints an error
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Use PFN_UP() and sg_virt() can be used to simplify the code.
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
dma_ops override is not in code anymore, so remove it.
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
IPU6 driver doesn't override the dma_ops of device now, it doesn't
depends on the ARCH_HAS_DMA_OPS, so remove the dependency in Kconfig.
Fixes: de6c85bf918e ("dma-mapping: clearly mark DMA ops as an architecture feature")
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
dma_ops is removed from the IPU6 auxiliary device, ISYS driver
should use the IPU6 DMA mapping APIs directly instead of depending
on the device callbacks.
ISYS driver switch from the videobuf2 DMA contig memory allocator to
scatter/gather memory allocator.
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
[Sakari Ailus: Rebased on recent videobuf2 wait changes.]
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
DMA ops are a helper for architectures and not for drivers to override the
DMA implementation. Driver should not override the DMA implementation.
This patch removes the dma_ops override from auxiliary device and adds
driver-internal helpers that use the actual DMA mapping APIs.
Fixes: 9163d83573e4 ("media: intel/ipu6: add IPU6 DMA mapping API and MMU table")
Signed-off-by: Bingbu Cao <bingbu.cao@intel.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
[Sakari Ailus: Fix the commit message a little.]
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
Fix printing DMA and physical address printing on 32-bit platforms, by
using correct types. Also cast DMA_BIT_MASK() result to dma_addr_t to make
Clang happy.
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
All ov2740_check_hwcfg() error-exit paths already log a detailed reason,
logging a second generic "failed to check HW configuration" error is
not useful, drop this.
The one exception is the -EPROBE_DEFER exit on
fwnode_graph_get_next_endpoint() returning NULL.
Call dev_err_probe() there to register the reason for deferring the probe,
this is used if the endpoint is still not there after 30 seconds, e.g. :
i2c-INT3474:00: deferred probe pending: waiting for fwnode graph endpoint
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Bingbu Cao <bingbu.cao@intel.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
|
|
The function workingset_activation() is called from folio_mark_accessed()
with the guarantee that the given folio can not be freed under us in
workingset_activation(). In addition, the association of the folio and
its memcg can not be broken here because charge migration is no more.
There is no need to use folio_memcg_rcu. Simply use folio_memcg_charged()
because that is what this function cares about.
[akpm@linux-foundation.org: provide folio_memcg_charged stub for CONFIG_MEMCG=n]
Link: https://lkml.kernel.org/r/20241026163707.2479526-1-shakeel.butt@linux.dev
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Suggested-by: Yu Zhao <yuzhao@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
By this point can_vma_merge_right() must have returned true, which implies
can_vma_merge_before() also returned true, which already asserts that the
pgoff is as expected for a merge with the following VMA, thus this
assignment is redundant.
Below is a more detail explanation.
Current definition of can_vma_merge_right() is:
static bool can_vma_merge_right(struct vma_merge_struct *vmg,
bool can_merge_left)
{
if (!vmg->next || vmg->end != vmg->next->vm_start ||
!can_vma_merge_before(vmg))
return false;
...
}
And:
static bool can_vma_merge_before(struct vma_merge_struct *vmg)
{
pgoff_t pglen = PHYS_PFN(vmg->end - vmg->start);
...
if (vmg->next->vm_pgoff == vmg->pgoff + pglen)
return true;
...
}
Which implies vmg->pgoff == vmg->next->vm_pgoff - pglen.
None of these values are changed between the check and prior assignment,
so this was an entirely redundant assignment.
[akpm@linux-foundation.org: remove now-unused local]
[lorenzo.stoakes@oracle.com: rephrase the changelog]
Link: https://lkml.kernel.org/r/20241024093347.18057-1-richard.weiyang@gmail.com
Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Jann Horn <jannh@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Rather than trying to merge again when ostensibly allocating a new VMA,
instead defer until the VMA is added and attempt to merge the existing
range.
This way we have no complicated unwinding logic midway through the process
of mapping the VMA.
In addition this removes limitations on the VMA not being able to be the
first in the virtual memory address space which was previously implicitly
required.
In theory, for this very same reason, we should unconditionally attempt
merge here, however this is likely to have a performance impact so it is
better to avoid this given the unlikely outcome of a merge.
[lorenzo.stoakes@oracle.com: remove unnecessary indirection]
Link: https://lkml.kernel.org/r/5106696d-e625-4d8a-8545-9d1430301730@lucifer.local
Link: https://lkml.kernel.org/r/d4f84502605d7651ac114587f507395c0fc76004.1729858176.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Liam R. Howlett <Liam.Howlett@Oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The only place where this was used was in mmap_region(), which we have now
adjusted to not require this to be performed (we reset ourselves in
effect).
It also created a dangerous assumption that VMG state could be safely
reused after a merge, at which point it may have been mutated in
unexpected ways, leading to subtle bugs.
Note that it was discovered by Wei Yang that there was also an error in
this code - we are comparing vmg->vma with prev after setting it to NULL.
This however had no impact, as we previously reset VMA iterator state
before attempting merge again, but it was useless effort.
In any case, this patch removes all of the logic so also eliminates this
wasted effort.
Link: https://lkml.kernel.org/r/5d9a59eee6498ae017cc87d89aa723de7179f75d.1729858176.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Liam R. Howlett <Liam.Howlett@Oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
We have seen bugs and resource leaks arise from the complexity of the
__mmap_region() function. This, and the generally deeply fragile error
handling logic and complexity which makes understanding the function
difficult make it highly desirable to refactor it into something readable.
Achieve this by separating the function into smaller logical parts which
are easier to understand and follow, and which importantly very
significantly simplify the error handling.
Note that we now call vms_abort_munmap_vmas() in more error paths than we
used to, however in cases where no abort need occur, vms->nr_pages will be
equal to zero and we simply exit this function without doing more than we
would have done previously.
Importantly, the invocation of the driver mmap hook via mmap_file() now
has very simple and obvious handling (this was previously the most
problematic part of the mmap() operation).
Use a generalised stack-based 'mmap state' to thread through values and
also retrieve state as needed.
Also avoid ever relying on vma merge (vmg) state after a merge is
attempted, instead maintain meaningful state in the mmap state and
establish vmg state as and when required.
This avoids any subtle bugs arising from merge logic mutating this state
and mmap_region() logic later relying upon it.
Link: https://lkml.kernel.org/r/25bd2edc3275450f448cbfe0756ce2a7cd06810f.1729858176.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Liam R. Howlett <Liam.Howlett@Oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
In previous commits we effected improvements to the mmap() logic in
mmap_region() and its newly introduced internal implementation function
__mmap_region().
However as these changes are intended to be backported, we kept the delta
as small as is possible and made as few changes as possible to the newly
introduced mm/vma.* files.
Take the opportunity to move this logic to mm/vma.c which not only
isolates it, but also makes it available for later userland testing which
can help us catch such logic errors far earlier.
Link: https://lkml.kernel.org/r/93fc2c3aa37dd30590b7e4ee067dfd832007bf7e.1729858176.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Jann Horn <jannh@google.com>
Cc: Liam R. Howlett <Liam.Howlett@Oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "fix error handling in mmap_region() and refactor", v3.
The mmap_region() function is somewhat terrifying, with spaghetti-like
control flow and numerous means by which issues can arise and incomplete
state, memory leaks and other unpleasantness can occur.
This series goes to great lengths to simplify how mmap_region() works and
to avoid unwinding errors late on in the process of setting up the VMA for
the new mapping, and equally avoids such operations occurring while the
VMA is in an inconsistent state.
This series builds on the previously submitted hotfix patches (see link to
v2 below) which addresses the most critical issues around mmap_region(),
and further works to improve mmap_region() complexity, stability, and
testability.
This series moves the code to mm/vma.c to render it userland testable,
refactors and simplifies it into smaller functions that are significantly
more readable.
It additionally avoids performing an attempt at a second merge mid-way
through allocating a new VMA, a dubious proposition at best and one that
is highly subject to subtle bugs.
Rather than do this, we simply note that we ought to retry the merge and
do this as a final step.
This patch (of 3):
Add some additional vma_internal.h stubs in preparation for
__mmap_region() being moved to mm/vma.c. Without these the move would
result in the tests no longer compiling.
Link: https://lkml.kernel.org/r/cover.1729858176.git.lorenzo.stoakes@oracle.com
Link: https://lkml.kernel.org/r/74b27e159e261d2ac1fe66a130edad1d61fdc176.1729858176.git.lorenzo.stoakes@oracle.com
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Jann Horn <jannh@google.com>
Cc: Liam R. Howlett <Liam.Howlett@Oracle.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Xu <peterx@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The memcg v1's charge move feature has been deprecated. All the places
using the memcg move lock, have stopped using it as they don't need the
protection any more. Let's proceed to remove all the locking code related
to charge moving.
Link: https://lkml.kernel.org/r/20241025012304.2473312-7-shakeel.butt@linux.dev
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
While updating the generation of the folios, MGLRU requires that the
folio's memcg association remains stable. With the charge migration
deprecated, there is no need for MGLRU to acquire locks to keep the folio
and memcg association stable.
[yuzhao@google.com: remove !rcu_read_lock_held() assertion]
Link: https://lkml.kernel.org/r/ZykEtcHrQRq-KrBC@google.com
Link: https://syzkaller.appspot.com/bug?extid=24f45b8beab9788e467e
Link: https://lore.kernel.org/lkml/67294349.050a0220.701a.0010.GAE@google.com/
[akpm@linux-foundation.org: remove now-unused local]
[shakeel.butt@linux.dev: folio_rcu() fixup, per Yu Zhao]
Link: https://lkml.kernel.org/r/iwmabnye3nl4merealrawt3bdvfii2pwavwrddrqpraoveet7h@ezrsdhjwwej7
Link: https://lkml.kernel.org/r/20241025012304.2473312-6-shakeel.butt@linux.dev
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
During the era of memcg charge migration, the kernel has to be make
sure that the writeback stat updates do not race with the charge
migration. Otherwise it might update the writeback stats of the wrong
memcg. Now with the memcg charge migration gone, there is no more race
for writeback stat updates and the previous locking can be removed.
Link: https://lkml.kernel.org/r/20241025012304.2473312-5-shakeel.butt@linux.dev
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
During the era of memcg charge migration, the kernel has to be make
sure that the dirty stat updates do not race with the charge migration.
Otherwise it might update the dirty stats of the wrong memcg. Now
with the memcg charge migration gone, there is no more race for dirty
stat updates and the previous locking can be removed.
Link: https://lkml.kernel.org/r/20241025012304.2473312-4-shakeel.butt@linux.dev
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The memcg-v1 charge move feature has been deprecated completely and let's
remove the relevant code as well.
Link: https://lkml.kernel.org/r/20241025012304.2473312-3-shakeel.butt@linux.dev
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "memcg-v1: fully deprecate charge moving".
The memcg v1's charge moving feature has been deprecated for almost 2
years and the kernel warns if someone try to use it. This warning has
been backported to all stable kernel and there have not been any report of
the warning or the request to support this feature anymore. Let's proceed
to fully deprecate this feature.
This patch (of 6):
Proceed with the complete deprecation of memcg v1's charge moving feature.
The deprecation warning has been in the kernel for almost two years and
has been ported to all stable kernel since. Now is the time to fully
deprecate this feature.
Link: https://lkml.kernel.org/r/20241025012304.2473312-1-shakeel.butt@linux.dev
Link: https://lkml.kernel.org/r/20241025012304.2473312-2-shakeel.butt@linux.dev
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Yosry Ahmed <yosryahmed@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
