The PTP hardware semaphore (PFTSYN_SEM) is used to synchronize
operations that program the PTP timers. The operations involve issuing
commands to the sideband queue. The E810 does not have a hardware
sideband queue, so the admin queue is used. The admin queue is slow.
I have observed delays in hundreds of milliseconds waiting for
ice_sq_done.
When phc2sys reads the time from the ice PTP clock and PFTSYN_SEM is
held by a task performing one of the slow operations, ice_ptp_lock can
easily time out. phc2sys gets -EBUSY and the kernel prints:
ice 0000:XX:YY.0: PTP failed to get time
These messages appear once every few seconds, causing log spam.
The E810 datasheet recommends an algorithm for reading the upper 64 bits
of the GLTSYN_TIME register. It matches what's implemented in
ice_ptp_read_src_clk_reg. It is robust against wrap-around, but not
necessarily against the concurrent setting of the register (with
GLTSYN_CMD_{INIT,ADJ}_TIME commands). Perhaps that's why
ice_ptp_gettimex64 also takes PFTSYN_SEM.
The race with time setters can be prevented without relying on the PTP
hardware semaphore. Using the "ice_adapter" from the previous patch,
we can have a common spinlock for the PFs that share the clock hardware.
It will protect the reading and writing to the GLTSYN_TIME register.
The writing is performed indirectly, by the hardware, as a result of
the driver writing GLTSYN_CMD_SYNC in ice_ptp_exec_tmr_cmd. I wasn't
sure if the ice_flush there is enough to make sure GLTSYN_TIME has been
updated, but it works well in my testing.
My test code can be seen here:
https://gitlab.com/mschmidt2/linux/-/commits/ice-ptp-host-side-lock-10
It consists of:
- kernel threads reading the time in a busy loop and looking at the
deltas between consecutive values, reporting new maxima.
- a shell script that sets the time repeatedly;
- a bpftrace probe to produce a histogram of the measured deltas.
Without the spinlock ptp_gltsyn_time_lock, it is easy to see tearing.
Deltas in the [2G, 4G) range appear in the histograms.
With the spinlock added, there is no tearing and the biggest delta I saw
was in the range [1M, 2M), that is under 2 ms.
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
There is a need for synchronization between ice PFs on the same physical
adapter.
Add a "struct ice_adapter" for holding data shared between PFs of the
same multifunction PCI device. The struct is refcounted - each ice_pf
holds a reference to it.
Its first use will be for PTP. I expect it will be useful also to
improve the ugliness that is ice_prot_id_tbl.
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Rename functions of mtk_ddp_comp:
- To align the naming rule
- To reduce the code size
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-15-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename files mtk_drm_plane.c to mtk_plane.c and
modify the Makefile accordingly.
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-12-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename files mtk_drm_plane.h to mtk_plane.h.
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-11-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename files mtk_drm_ddp_comp.c to mtk_ddp_comp.c and
modify the Makefile accordingly.
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-10-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename files mtk_drm_ddp_comp.h to mtk_ddp_comp.h.
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-9-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename files mtk_drm_crtc.c to mtk_crtc.c and
modify the Makefile accordingly.
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-8-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename files mtk_drm_crtc.h to mtk_crtc.h.
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-7-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename all "mtk_drm_hdmi" to "mtk_hdmi":
- To align the naming rule
- To reduce the code size
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-6-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename all "mtk_drm_gem" to "mtk_gem":
- To align the naming rule
- To reduce the code size
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-5-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename all "mtk_drm_plane" to "mtk_plane":
- To align the naming rule
- To reduce the code size
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-4-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename all "mtk_drm_ddp_comp" to "mtk_ddp_comp":
- To align the naming rule
- To reduce the code size
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-3-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Rename all "mtk_drm_crtc" to "mtk_crtc" due to the following benefits:
- Lower the matches when searching the native drm_crtc* codes
- Reduce the code size
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240322091232.26387-2-shawn.sung@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
In the case where `conn_routes` is true we allocate an extra slot in
the `ddp_comp` array but mtk_drm_crtc_create() never seemed to
initialize it in the test case I ran. For me, this caused a later
crash when we looped through the array in mtk_drm_crtc_mode_valid().
This showed up for me when I booted with `slub_debug=FZPUA` which
poisons the memory initially. Without `slub_debug` I couldn't
reproduce, presumably because the later code handles the value being
NULL and in most cases (not guaranteed in all cases) the memory the
allocator returned started out as 0.
It really doesn't hurt to initialize the array with devm_kcalloc()
since the array is small and the overhead of initting a handful of
elements to 0 is small. In general initting memory to zero is a safer
practice and usually it's suggested to only use the non-initting alloc
functions if you really need to.
Let's switch the function to use an allocation function that zeros the
memory. For me, this avoids the crash.
Fixes: 01389b324c ("drm/mediatek: Add connector dynamic selection capability")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240328092248.1.I2e73c38c0f264ee2fa4a09cdd83994e37ba9f541@changeid/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Add a check to mtk_drm_gem_init if we attempt to allocate a GEM object
of 0 bytes. Currently, no such check exists and the kernel will panic if
a userspace application attempts to allocate a 0x0 GBM buffer.
Tested by attempting to allocate a 0x0 GBM buffer on an MT8188 and
verifying that we now return EINVAL.
Fixes: 119f517362 ("drm/mediatek: Add DRM Driver for Mediatek SoC MT8173.")
Signed-off-by: Justin Green <greenjustin@chromium.org>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240307180051.4104425-1-greenjustin@chromium.org/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Since MT8195 compatible is in the single enum group, we have to add its
compatible into mediatek-drm component binding table to ensure that
it can be bound as a ddp_comp.
Signed-off-by: Jason-JH.Lin <jason-jh.lin@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogiaocchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240229144844.1688-4-jason-jh.lin@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
The gamma LUT setting of MT8188 and MT8195 are the same, so we create
a one of items for MT8188 to reuse the driver data settings of MT8195.
Signed-off-by: Jason-JH.Lin <jason-jh.lin@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240229144844.1688-3-jason-jh.lin@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Since MT8195 gamma has multiple bank for 12 bits LUT and it is
different from any other SoC LUT setting.
So we add MT8195 compatible to the single enum group to make its
driver data settings can be reused by other SoC.
Signed-off-by: Jason-JH.Lin <jason-jh.lin@mediatek.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20240229144844.1688-2-jason-jh.lin@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
A wrapper function is available since the commit 890cc39a87
("drivers: provide devm_platform_get_and_ioremap_resource()").
Thus reuse existing functionality instead of keeping duplicate
source code.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Link: https://patchwork.kernel.org/project/dri-devel/patch/2a4ecf78-20e7-4678-a67d-0d66956b07cc@web.de/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
The old state is helpful for debugging, as the current state is always
IB_CM_IDLE when timeout happens.
Fixes: 96d9cbe2f2 ("RDMA/cm: add timeout to cm_destroy_id wait")
Signed-off-by: Mark Zhang <markzhang@nvidia.com>
Link: https://lore.kernel.org/r/20240322112049.2022994-1-markzhang@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
struct rdma_restrack_entry's kern_name was set to KBUILD_MODNAME
in ib_create_cq(), while if the module exited but forgot del this
rdma_restrack_entry, it would cause a invalid address access in
rdma_restrack_clean() when print the owner of this rdma_restrack_entry.
These code is used to help find one forgotten PD release in one of the
ULPs. But it is not needed anymore, so delete them.
Signed-off-by: Wenchao Hao <haowenchao2@huawei.com>
Link: https://lore.kernel.org/r/20240318092320.1215235-1-haowenchao2@huawei.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
As described in the added code comment, a reference to .exit.text is ok
for drivers registered via module_platform_driver_probe(). Make this
explicit to prevent the following section mismatch warning
WARNING: modpost: sound/oss/dmasound/dmasound_paula: section mismatch in reference: amiga_audio_driver+0x8 (section: .data) -> amiga_audio_remove (section: .exit.text)
that triggers on an allmodconfig W=1 build.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Message-ID: <c216a129aa88f3af5c56fe6612a472f7a882f048.1711748999.git.u.kleine-koenig@pengutronix.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The dma_alloc_coherent() interface automatically zero the memory returned.
Thus, we do not need to specify the __GFP_ZERO flag explicitly when we call
dma_alloc_coherent().
Reviewed-by: Cheng Xu <chengyou@linux.alibaba.com>
Signed-off-by: Boshi Yu <boshiyu@linux.alibaba.com>
Link: https://lore.kernel.org/r/20240311113821.22482-4-boshiyu@alibaba-inc.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
There exist two different names for the doorbell records: db_info and
db_record. We use dbrec for cpu address of the doorbell record and
dbrec_dma for dma address of the doorbell recordi uniformly.
Reviewed-by: Cheng Xu <chengyou@linux.alibaba.com>
Signed-off-by: Boshi Yu <boshiyu@linux.alibaba.com>
Link: https://lore.kernel.org/r/20240311113821.22482-3-boshiyu@alibaba-inc.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Currently, the 8 byte doorbell record is allocated along with the queue
buffer, which may result in waste of dma space when the queue buffer is
page aligned. To address this issue, we introduce a dma pool named
db_pool and allocate doorbell record from it.
Reviewed-by: Cheng Xu <chengyou@linux.alibaba.com>
Signed-off-by: Boshi Yu <boshiyu@linux.alibaba.com>
Link: https://lore.kernel.org/r/20240311113821.22482-2-boshiyu@alibaba-inc.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When a mutex lock is not used any more, the function mutex_destroy
should be called to mark the mutex lock uninitialized.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Signed-off-by: Yanjun.Zhu <yanjun.zhu@linux.dev>
Link: https://lore.kernel.org/r/20240314065140.27468-1-yanjun.zhu@linux.dev
Reviewed-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Currently, ath11k fails to resume from system suspend/hibernation on some
the x86 host machines with below error message:
```
ath11k_pci 0000:06:00.0: timeout while waiting for restart complete
```
This happens because, ath11k powers down the MHI stack during suspend and
that leads to destruction of the struct device associated with the MHI
channels. And during resume, ath11k calls calling mhi_sync_power_up() to
power up the MHI subsystem and that eventually calls the driver framework's
device_add() API from mhi_create_devices(). But the PM framework blocks the
struct device creation during device_add() and this leads to probe deferral
as below:
```
mhi mhi0_IPCR: Driver qcom_mhi_qrtr force probe deferral
```
The reason for deferring device creation during resume is explained in
dpm_prepare():
/*
* It is unsafe if probing of devices will happen during suspend or
* hibernation and system behavior will be unpredictable in this
* case. So, let's prohibit device's probing here and defer their
* probes instead. The normal behavior will be restored in
* dpm_complete().
*/
Due to the device probe deferral, qcom_mhi_qrtr_probe() API is not getting
called during resume and thus MHI channels are not prepared. So this blocks
the QMI messages from being transferred between ath11k and firmware,
resulting in a firmware initialization failure.
After consulting with Rafael, it was decided to not destroy the struct
device for the MHI channels during system suspend/hibernation because the
device is bound to appear again during resume.
So to achieve this, a new API called mhi_power_down_keep_dev() is
introduced for MHI controllers to keep the struct device when required.
This API is similar to the existing mhi_power_down() API, except that it
keeps the struct device associated with MHI channels instead of destroying
them.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Reviewed-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://lore.kernel.org/r/20240305021320.3367-2-quic_bqiang@quicinc.com
[mani: reworded the commit message and subject]
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Most of the functions that read values return a status and put the value
itself in an a function parameter. Update health reading to match.
As health is not checked for changes as part of the update loop, remove
the read of this from the periodic update loop. This saves I2C/1W
bandwidth. It also means we do not have to cache it, fresh values are
read when requested.
Signed-off-by: Andrew Davis <afd@ti.com>
Link: https://lore.kernel.org/r/20240325203129.150030-6-afd@ti.com
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Most of the functions that read values return a status and put the value
itself in an a function parameter. Update cycle count reading to match.
As cycle count is not checked for changes as part of the update loop,
remove the read of this from the periodic update loop. This saves I2C/1W
bandwidth. It also means we do not have to cache it, fresh values are
read when requested.
Signed-off-by: Andrew Davis <afd@ti.com>
Link: https://lore.kernel.org/r/20240325203129.150030-5-afd@ti.com
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Most of the functions that read values return a status and put the value
itself in an a function parameter. Update energy reading to match.
As energy is not checked for changes as part of the update loop,
remove the read of this from the periodic update loop. This saves
I2C/1W bandwidth. It also means we do not have to cache it, fresh
values are read when requested.
Signed-off-by: Andrew Davis <afd@ti.com>
Link: https://lore.kernel.org/r/20240325203129.150030-4-afd@ti.com
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Most of the functions that read values return a status and put the value
itself in an a function parameter. Update charge reading to match.
As charge state is not checked for changes as part of the update loop,
remove the read of this from the periodic update loop. This saves
I2C/1W bandwidth. It also means we do not have to cache it, fresh
values are read when requested.
Signed-off-by: Andrew Davis <afd@ti.com>
Link: https://lore.kernel.org/r/20240325203129.150030-3-afd@ti.com
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Most of the functions that read values return a status and put the value
itself in an a function parameter. Update time reading to match.
As time is not checked for changes as part of the update loop, remove
the read of the this from the periodic update loop. This saves
I2C/1W bandwidth. It also means we do not have to cache it, fresh
values are read when requested.
Signed-off-by: Andrew Davis <afd@ti.com>
Link: https://lore.kernel.org/r/20240325203129.150030-2-afd@ti.com
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Most of the functions that read values return a status and put the value
itself in an a function parameter. Update temperature reading to match.
As temp is not checked for changes as part of the update loop, remove
the read of the temperature from the periodic update loop. This saves
I2C/1W bandwidth. It also means we do not have to cache it, fresh
values are read when requested.
Signed-off-by: Andrew Davis <afd@ti.com>
Link: https://lore.kernel.org/r/20240325203129.150030-1-afd@ti.com
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Eric Dumazet says:
====================
net: rps: misc changes
Make RPS/RFS a bit more efficient with better cache locality
and heuristics.
Aso shrink include/linux/netdevice.h a bit.
v2: fixed a build issue in patch 6/8 with CONFIG_RPS=n
(Jakub and kernel build bots)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 14d898f3c1 ("dev: Move received_rps counter next
to RPS members in softnet data") was unfortunate:
received_rps is dirtied by a cpu and never read by other
cpus in fast path.
Its presence in the hot RPS cache line (shared by many cpus)
is hurting RPS/RFS performance.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
process_backlog() can batch increments of sd->input_queue_head,
saving some memory bandwidth.
Also add READ_ONCE()/WRITE_ONCE() annotations around
sd->input_queue_head accesses.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
input_queue_tail_incr_save() is incrementing the sd queue_tail
and save it in the flow last_qtail.
Two issues here :
- no lock protects the write on last_qtail, we should use appropriate
annotations.
- We can perform this write after releasing the per-cpu backlog lock,
to decrease this lock hold duration (move away the cache line miss)
Also move input_queue_head_incr() and rps helpers to include/net/rps.h,
while adding rps_ prefix to better reflect their role.
v2: Fixed a build issue (Jakub and kernel build bots)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can remove a goto and a label by reversing a condition.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If under extreme cpu backlog pressure enqueue_to_backlog() has
to drop a packet, it could do this without dirtying a cache line
and potentially slowing down the target cpu.
Move sd->dropped into a separate cache line, and make it atomic.
In non pressure mode, this field is not touched, no need to consume
valuable space in a hot cache line.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the device attached to the packet given to enqueue_to_backlog()
is not running, we drop the packet.
But we accidentally increase sd->dropped, giving false signals
to admins: sd->dropped should be reserved to cpu backlog pressure,
not to temporary glitches at device dismantles.
While we are at it, perform the netif_running() test before
we get the rps lock, and use REASON_DEV_READY
drop reason instead of NOT_SPECIFIED.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move dev_xmit_recursion() and friends to net/core/dev.h
They are only used from net/core/dev.c and net/core/filter.c.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
kick_defer_list_purge() is defined in net/core/dev.c
and used from net/core/skubff.c
Because we need softnet_data, include <linux/netdevice.h>
from net/core/dev.h
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alexander Lobakin says:
====================
ice: add PFCP filter support
Add support for creating PFCP filters in switchdev mode. Add pfcp module
that allows to create a PFCP-type netdev. The netdev then can be passed to
tc when creating a filter to indicate that PFCP filter should be created.
To add a PFCP filter, a special netdev must be created and passed to tc
command:
ip link add pfcp0 type pfcp
tc filter add dev eth0 ingress prio 1 flower pfcp_opts \
1:12ab/ff:fffffffffffffff0 skip_hw action mirred egress redirect \
dev pfcp0
Changes in iproute2 [1] are required to use pfcp_opts in tc.
ICE COMMS package is required as it contains PFCP profiles.
Part of this patchset modifies IP_TUNNEL_*_OPTs, which were previously
stored in a __be16. All possible values have already been used, making
it impossible to add new ones.
* 1-3: add new bitmap_{read,write}(), which is used later in the IP
tunnel flags code (from Alexander's ARM64 MTE series[2]);
* 4-14: some bitmap code preparations also used later in IP tunnels;
* 15-17: convert IP tunnel flags from __be16 to a bitmap;
* 18-21: add PFCP module and support for it in ice.
[1] https://lore.kernel.org/netdev/20230614091758.11180-1-marcin.szycik@linux.intel.com
[2] https://lore.kernel.org/linux-kernel/20231218124033.551770-1-glider@google.com
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for creating PFCP filters in switchdev mode. Add support
for parsing PFCP-specific tc options: S flag and SEID.
To create a PFCP filter, a special netdev must be created and passed
to tc command:
ip link add pfcp0 type pfcp
tc filter add dev eth0 ingress prio 1 flower pfcp_opts \
1:123/ff:fffffffffffffff0 skip_hw action mirred egress redirect \
dev pfcp0
Changes in iproute2 [1] are required to be able to use pfcp_opts in tc.
ICE COMMS package is required to create a filter as it contains PFCP
profiles.
Link: https://lore.kernel.org/netdev/20230614091758.11180-1-marcin.szycik@linux.intel.com [1]
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
FLOW_DISSECTOR_KEY_ENC_OPTS can be used for multiple headers, but currently
it is treated as GTP-exclusive in ice. Rename ICE_TC_FLWR_FIELD_ENC_OPTS to
ICE_TC_FLWR_FIELD_GTP_OPTS and check for tunnel type earlier. After this
refactor, it is easier to add new headers using FLOW_DISSECTOR_KEY_ENC_OPTS
- instead of checking tunnel type in ice_tc_count_lkups() and
ice_tc_fill_tunnel_outer(), it needs to be checked only once, in
ice_parse_tunnel_attr().
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>