| Age | Commit message (Collapse) | Author |
|---|
|
Maciej Fijalkowski says:
====================
Although this work started as an effort to add multi-buffer XDP support
to ice driver, as usual it turned out that some other side stuff needed
to be addressed, so let me give you an overview.
First patch adjusts legacy-rx in a way that it will be possible to refer
to skb_shared_info being at the end of the buffer when gathering up
frame fragments within xdp_buff.
Then, patches 2-9 prepare ice driver in a way that actual multi-buffer
patches will be easier to swallow.
10 and 11 are the meat. What is worth mentioning is that this set
actually *fixes* things as patch 11 removes the logic based on
next_dd/rs and we previously stepped away from this for ice_xmit_zc().
Currently, AF_XDP ZC XDP_TX workload is off as there are two cleaning
sides that can be triggered and two of them work on different internal
logic. This set unifies that and allows us to improve the performance by
2x with a trick on the last (13) patch.
12th is a simple cleanup of no longer fields from Tx ring.
I might be wrong but I have not seen anyone reporting performance impact
among patches that add XDP multi-buffer support to a particular driver.
Numbers below were gathered via xdp_rxq_info and xdp_redirect_map on
1500 MTU:
XDP_DROP +1%
XDP_PASS -1,2%
XDP_TX -0,5%
XDP_REDIRECT -3,3%
Cherry on top, which is not directly related to mbuf support (last
patch):
XDP_TX ZC +126%
Target the we agreed on was to not degrade performance for any action by
anything that would be over 5%, so our goal was met. Basically this set
keeps the performance where it was. Redirect is slower due to more
frequent tail bumps.
====================
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
|
|
Let us store pointer to xdp_buff that came from xsk_buff_pool on tx_buf
so that it will be possible to recycle it via xsk_buff_free() on Tx
cleaning side. This way it is not necessary to do expensive copy to
another xdp_buff backed by a newly allocated page.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-14-maciej.fijalkowski@intel.com
|
|
Now that both ZC and standard XDP data paths stopped using Tx logic
based on next_dd and next_rs fields, we can safely remove these fields
and shrink Tx ring structure.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-13-maciej.fijalkowski@intel.com
|
|
Similarly as for Rx side in previous patch, logic on XDP Tx in ice
driver needs to be adjusted for multi-buffer support. Specifically, the
way how HW Tx descriptors are produced and cleaned.
Currently, XDP_TX works on strict ring boundaries, meaning it sets RS
bit (on producer side) / looks up DD bit (on consumer/cleaning side)
every quarter of the ring. It means that if for example multi buffer
frame would span across the ring quarter boundary (say that frame
consists of 4 frames and we start from 62 descriptor where ring is sized
to 256 entries), RS bit would be produced in the middle of multi buffer
frame, which would be a broken behavior as it needs to be set on the
last descriptor of the frame.
To make it work, set RS bit at the last descriptor from the batch of
frames that XDP_TX action was used on and make the first entry remember
the index of last descriptor with RS bit set. This way, cleaning side
can take the index of descriptor with RS bit, look up DD bit's presence
and clean from first entry to last.
In order to clean up the code base introduce the common ice_set_rs_bit()
which will return index of descriptor that got RS bit produced on so
that standard driver can store this within proper ice_tx_buf and ZC
driver can simply ignore return value.
Co-developed-by: Martyna Szapar-Mudlaw <martyna.szapar-mudlaw@linux.intel.com>
Signed-off-by: Martyna Szapar-Mudlaw <martyna.szapar-mudlaw@linux.intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-12-maciej.fijalkowski@intel.com
|
|
Ice driver needs to be a bit reworked on Rx data path in order to
support multi-buffer XDP. For skb path, it currently works in a way that
Rx ring carries pointer to skb so if driver didn't manage to combine
fragmented frame at current NAPI instance, it can restore the state on
next instance and keep looking for last fragment (so descriptor with EOP
bit set). What needs to be achieved is that xdp_buff needs to be
combined in such way (linear + frags part) in the first place. Then skb
will be ready to go in case of XDP_PASS or BPF program being not present
on interface. If BPF program is there, it would work on multi-buffer
XDP. At this point xdp_buff resides directly on Rx ring, so given the
fact that skb will be built straight from xdp_buff, there will be no
further need to carry skb on Rx ring.
Besides removing skb pointer from Rx ring, lots of members have been
moved around within ice_rx_ring. First and foremost reason was to place
rx_buf with xdp_buff on the same cacheline. This means that once we
touch rx_buf (which is a preceding step before touching xdp_buff),
xdp_buff will already be hot in cache. Second thing was that xdp_rxq is
used rather rarely and it occupies a separate cacheline, so maybe it is
better to have it at the end of ice_rx_ring.
Other change that affects ice_rx_ring is the introduction of
ice_rx_ring::first_desc. Its purpose is twofold - first is to propagate
rx_buf->act to all the parts of current xdp_buff after running XDP
program, so that ice_put_rx_buf() that got moved out of the main Rx
processing loop will be able to tak an appriopriate action on each
buffer. Second is for ice_construct_skb().
ice_construct_skb() has a copybreak mechanism which had an explicit
impact on xdp_buff->skb conversion in the new approach when legacy Rx
flag is toggled. It works in a way that linear part is 256 bytes long,
if frame is bigger than that, remaining bytes are going as a frag to
skb_shared_info.
This means while memcpying frags from xdp_buff to newly allocated skb,
care needs to be taken when picking the destination frag array entry.
Upon the time ice_construct_skb() is called, when dealing with
fragmented frame, current rx_buf points to the *last* fragment, but
copybreak needs to be done against the first one. That's where
ice_rx_ring::first_desc helps.
When frame building spans across NAPI polls (DD bit is not set on
current descriptor and xdp->data is not NULL) with current Rx buffer
handling state there might be some problems.
Since calls to ice_put_rx_buf() were pulled out of the main Rx
processing loop and were scoped from cached_ntc to current ntc, remember
that now mentioned function relies on rx_buf->act, which is set within
ice_run_xdp(). ice_run_xdp() is called when EOP bit was found, so
currently we could put Rx buffer with rx_buf->act being *uninitialized*.
To address this, change scoping to rely on first_desc on both boundaries
instead.
This also implies that cleaned_count which is used as an input to
ice_alloc_rx_buffers() and tells how many new buffers should be refilled
has to be adjusted. If it stayed as is, what could happen is a case
where ntc would go over ntu.
Therefore, remove cleaned_count altogether and use against allocing
routine newly introduced ICE_RX_DESC_UNUSED() macro which is an
equivalent of ICE_DESC_UNUSED() dedicated for Rx side and based on
struct ice_rx_ring::first_desc instead of next_to_clean.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-11-maciej.fijalkowski@intel.com
|
|
SKB path calculates truesize on three different functions, which could
be avoided as xdp_buff carries the already calculated truesize under
xdp_buff::frame_sz. If ice_add_rx_frag() is adjusted to take the
xdp_buff as an input just like functions responsible for creating
sk_buff initially, codebase could be simplified by removing these
redundant recalculations and rely on xdp_buff::frame_sz instead.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-10-maciej.fijalkowski@intel.com
|
|
Currently ice_finalize_xdp_rx() is called only when xdp_prog is present
on VSI, which is a good thing. However, this optimization can be
enhanced and check only if any of the XDP_TX/XDP_REDIRECT took place in
current Rx processing. Non-zero value of @xdp_xmit indicates that
xdp_prog is present on VSI. This way XDP_DROP-based workloads will not
suffer from unnecessary calls to external function.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-9-maciej.fijalkowski@intel.com
|
|
This should have been used in there from day 1, let us address that
before introducing XDP multi-buffer support for Rx side.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-8-maciej.fijalkowski@intel.com
|
|
Currently calls to ice_put_rx_buf() are sprinkled through
ice_clean_rx_irq() - first place is for explicit flow director's
descriptor handling, second is after running XDP prog and the last one
is after taking care of skb.
1st callsite was actually only for ntc bump purpose, as Rx buffer to be
recycled is not even passed to a function.
It is possible to walk through Rx buffers processed in particular NAPI
cycle by caching ntc from beginning of the ice_clean_rx_irq().
To do so, let us store XDP verdict inside ice_rx_buf, so action we need
to take on will be known. For XDP prog absence, just store ICE_XDP_PASS
as a verdict.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-7-maciej.fijalkowski@intel.com
|
|
This might be in future used by ZC driver and might potentially yield a
minor performance boost. While at it, constify arguments that
ice_is_non_eop() takes, since they are pointers and this will help compiler
while generating asm.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-6-maciej.fijalkowski@intel.com
|
|
Plan is to move ice_put_rx_buf() to the end of ice_clean_rx_irq() so
in order to keep the ability of walking through HW Rx descriptors, pull
out next_to_clean handling out of ice_put_rx_buf().
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-5-maciej.fijalkowski@intel.com
|
|
This will allow us to avoid carrying additional auxiliary array of page
counts when dealing with XDP multi buffer support. Previously combining
fragmented frame to skb was not affected in the same way as XDP would be
as whole frame is needed to be in place before executing XDP prog.
Therefore, when going through HW Rx descriptors one-by-one, calls to
ice_put_rx_buf() need to be taken *after* running XDP prog on a
potentially multi buffered frame, so some additional storage of
page count is needed.
By adding page count to rx buf, it will make it easier to walk through
processed entries at the end of rx cleaning routine and decide whether
or not buffers should be recycled.
While at it, bump ice_rx_buf::pagecnt_bias from u16 up to u32. It was
proven many times that calculations on variables smaller than standard
register size are harmful. This was also the case during experiments
with embedding page count to ice_rx_buf - when this was added as u16 it
had a performance impact.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-4-maciej.fijalkowski@intel.com
|
|
In preparation for XDP multi-buffer support, let's store xdp_buff on
Rx ring struct. This will allow us to combine fragmented frames across
separate NAPI cycles in the same way as currently skb fragments are
handled. This means that skb pointer on Rx ring will become redundant
and will be removed. For now it is kept and layout of Rx ring struct was
not inspected, some member movement will be needed later on so that will
be the time to take care of it.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-3-maciej.fijalkowski@intel.com
|
|
Rx path is going to be modified in a way that fragmented frame will be
gathered within xdp_buff in the first place. This approach implies that
underlying buffer has to provide tailroom for skb_shared_info. This is
currently the case when ring uses build_skb but not when legacy-rx knob
is turned on. This case configures 2k Rx buffers and has no way to
provide either headroom or tailroom - FWIW it currently has
XDP_PACKET_HEADROOM which is broken and in here it is removed. 2k Rx
buffers were used so driver in this setting was able to support 9k MTU
as it can chain up to 5 Rx buffers. With offset configuring HW writing
2k of a data was passing the half of the page which broke the assumption
of our internal page recycling tricks.
Now if above got fixed and legacy-rx path would be left as is, when
referring to skb_shared_info via xdp_get_shared_info_from_buff(),
packet's content would be corrupted again. Hence size of Rx buffer needs
to be lowered and therefore supported MTU. This operation will allow us
to keep the unified data path and with 8k MTU users (if any of
legacy-rx) would still be good to go. However, tendency is to drop the
support for this code path at some point.
Add ICE_RXBUF_1664 as vsi::rx_buf_len and ICE_MAX_FRAME_LEGACY_RX (8320)
as vsi::max_frame for legacy-rx. For bigger page sizes configure 3k Rx
buffers, not 2k.
Since headroom support is removed, disable data_meta support on legacy-rx.
When preparing XDP buff, rely on ice_rx_ring::rx_offset setting when
deciding whether to support data_meta or not.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/bpf/20230131204506.219292-2-maciej.fijalkowski@intel.com
|
|
Wire up the missing ptrace requests PTRACE_GETREGS, PTRACE_SETREGS,
PTRACE_GETFPREGS and PTRACE_SETFPREGS when running 32-bit applications
on 64-bit kernels.
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: stable@vger.kernel.org # 4.7+
|
|
Prefer usage of the PRIV_USER constant over the hard-coded value to set
the lowest 2 bits for the userspace privilege.
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: stable@vger.kernel.org # 5.16+
|
|
Jiri Pirko says:
====================
devlink: trivial names cleanup
This is a follow-up to Jakub's devlink code split and dump iteration
helper patchset. No functional changes, just couple of renames to makes
things consistent and perhaps easier to follow.
====================
Link: https://lore.kernel.org/r/20230131090613.2131740-1-jiri@resnulli.us
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In order to maintain naming consistency, rename and reorder all usages
of struct struct devlink_cmd in the following way:
1) Remove "gen" and replace it with "cmd" to match the struct name
2) Order devl_cmds[] and the header file to match the order
of enum devlink_command
3) Move devl_cmd_rate_get among the peers
4) Remove "inst" for DEVLINK_CMD_GET
5) Add "_get" suffix to all to match DEVLINK_CMD_*_GET (only rate had it
done correctly)
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
No need to have "gen" inside name of the structure for devlink commands.
Remove it.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
To have the name of the function consistent with the struct cb name,
rename devlink_nl_instance_iter_dump() to
devlink_nl_instance_iter_dumpit().
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Pull virtio fixes from Michael Tsirkin:
"Just small bugfixes all over the place"
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
vdpa: ifcvf: Do proper cleanup if IFCVF init fails
vhost-scsi: unbreak any layout for response
tools/virtio: fix the vringh test for virtio ring changes
vhost/net: Clear the pending messages when the backend is removed
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"A bit higher volume of changes than wished, but each change is
relatively small and the fix targets are mostly device-specific, so
those should be safe as a late stage merge.
The most significant LoC is about the memalloc helper fix, which is
applied only to Xen PV. The other major parts are ASoC Intel SOF and
AVS fixes that are scattered as various small code changes. The rest
are device-specific fixes and quirks for HD- and USB-audio, FireWire
and ASoC AMD / HDMI"
* tag 'sound-6.2-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (30 commits)
ALSA: firewire-motu: fix unreleased lock warning in hwdep device
ALSA: memalloc: Workaround for Xen PV
ASoC: cs42l56: fix DT probe
ASoC: codecs: wsa883x: correct playback min/max rates
ALSA: hda/realtek: Add Acer Predator PH315-54
ASoC: amd: yc: Add Xiaomi Redmi Book Pro 15 2022 into DMI table
ALSA: hda: Do not unset preset when cleaning up codec
ASoC: SOF: sof-audio: prepare_widgets: Check swidget for NULL on sink failure
ASoC: hdmi-codec: zero clear HDMI pdata
ASoC: SOF: ipc4-mtrace: prevent underflow in sof_ipc4_priority_mask_dfs_write()
ASoC: Intel: sof_ssp_amp: always set dpcm_capture for amplifiers
ASoC: Intel: sof_nau8825: always set dpcm_capture for amplifiers
ASoC: Intel: sof_cs42l42: always set dpcm_capture for amplifiers
ASoC: Intel: sof_rt5682: always set dpcm_capture for amplifiers
ALSA: hda/via: Avoid potential array out-of-bound in add_secret_dac_path()
ALSA: usb-audio: Add FIXED_RATE quirk for JBL Quantum610 Wireless
ALSA: hda/realtek: fix mute/micmute LEDs, speaker don't work for a HP platform
ASoC: SOF: keep prepare/unprepare widgets in sink path
ASoC: SOF: sof-audio: skip prepare/unprepare if swidget is NULL
ASoC: SOF: sof-audio: unprepare when swidget->use_count > 0
...
|
|
The Flash Interface Unit (FIU) should have a reference to the Shared
Memory controller (SHM) so that flash access from the host (x86 computer
managed by the WPCM450 BMC) can be blocked during flash access by the
FIU driver.
Fixes: 38abcb0d68767 ("ARM: dts: wpcm450: Add FIU SPI controller node")
Signed-off-by: Jonathan Neuschäfer <j.neuschaefer@gmx.net>
Link: https://lore.kernel.org/r/20230129112611.1176517-1-j.neuschaefer@gmx.net
Signed-off-by: Joel Stanley <joel@jms.id.au>
Link: https://lore.kernel.org/r/20230201044158.962417-1-joel@jms.id.au
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
The linux-mediatek IRC channel has moved to liber.chat for quite some
time. Apart from that, not all patches are also send to LKML, so add
this ML explicitly.
And last but not least:
Angelo does a wunderfull job in reviewing patches for all kind of
devices from MediaTek.
Cc: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com>
Link: https://lore.kernel.org/r/20230201152256.19514-1-matthias.bgg@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
NVMe In-Band authentication uses two kinds of works: chap->auth_work and
ctrl->dhchap_auth_work. The latter work flushes or cancels the former
work. However, the both works are queued to the same workqueue nvme-wq.
It results in the lockdep WARNING as follows:
WARNING: possible recursive locking detected
6.2.0-rc4+ #1 Not tainted
--------------------------------------------
kworker/u16:7/69 is trying to acquire lock:
ffff902d52e65548 ((wq_completion)nvme-wq){+.+.}-{0:0}, at: start_flush_work+0x2c5/0x380
but task is already holding lock:
ffff902d52e65548 ((wq_completion)nvme-wq){+.+.}-{0:0}, at: process_one_work+0x210/0x410
To avoid the WARNING, introduce a new workqueue nvme-auth-wq dedicated
to chap->auth_work.
Reported-by: Daniel Wagner <dwagner@suse.de>
Link: https://lore.kernel.org/linux-nvme/20230130110802.paafkiipmitwtnwr@carbon.lan/
Fixes: f50fff73d620 ("nvme: implement In-Band authentication")
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Tested-by: Daniel Wagner <dwagner@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
In nvme_alloc_io_tag_set(), the connect_q pointer should be set to NULL
in case of error to avoid potential invalid pointer dereferences.
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
If nvme_alloc_admin_tag_set() fails, the admin_q and fabrics_q pointers
are left with an invalid, non-NULL value. Other functions may then check
the pointers and dereference them, e.g. in
nvme_probe() -> out_disable: -> nvme_dev_remove_admin().
Fix the bug by setting admin_q and fabrics_q to NULL in case of error.
Also use the set variable to free the tag_set as ctrl->admin_tagset isn't
initialized yet.
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
As part of nvmet_fc_ls_create_association there is a case where
nvmet_fc_alloc_target_queue fails right after a new association with an
admin queue is created. In this case, no one releases the get taken in
nvmet_fc_alloc_target_assoc. This fix is adding the missing put.
Signed-off-by: Amit Engel <Amit.Engel@dell.com>
Reviewed-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
The unprepare sequence has started to fail after moving to panel bridge
code in the msm drm driver (commit 007ac0262b0d ("drm/msm/dsi: switch to
DRM_PANEL_BRIDGE")). You'll see messages like this in the kernel logs:
panel-boe-tv101wum-nl6 ae94000.dsi.0: failed to set panel off: -22
This is because boe_panel_enter_sleep_mode() needs an operating DSI link
to set the panel into sleep mode. Performing those writes in the
unprepare phase of bridge ops is too late, because the link has already
been torn down by the DSI controller in post_disable, i.e. the PHY has
been disabled, etc. See dsi_mgr_bridge_post_disable() for more details
on the DSI .
Split the unprepare function into a disable part and an unprepare part.
For now, just the DSI writes to enter sleep mode are put in the disable
function. This fixes the panel off routine and keeps the panel happy.
My Wormdingler has an integrated touchscreen that stops responding to
touch if the panel is only half disabled too. This patch fixes it. And
finally, this saves power when the screen is off because without this
fix the regulators for the panel are left enabled when nothing is being
displayed on the screen.
Fixes: 007ac0262b0d ("drm/msm/dsi: switch to DRM_PANEL_BRIDGE")
Fixes: a869b9db7adf ("drm/panel: support for boe tv101wum-nl6 wuxga dsi video mode panel")
Cc: yangcong <yangcong5@huaqin.corp-partner.google.com>
Cc: Douglas Anderson <dianders@chromium.org>
Cc: Jitao Shi <jitao.shi@mediatek.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Rob Clark <robdclark@chromium.org>
Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230106030108.2542081-1-swboyd@chromium.org
(cherry picked from commit c913cd5489930abbb557ef144a333846286754c3)
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
|
|
This is a single fix, but it conflicts with some recent features. I'm
merging it on top of the commit it fixes to ease backporting.
* b4-shazam-merge:
riscv: Fix build with CONFIG_CC_OPTIMIZE_FOR_SIZE=y
Link: https://lore.kernel.org/r/20220922060958.44203-1-samuel@sholland.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
|
|
commit 8eb060e10185 ("arch/riscv: add Zihintpause support") broke
building with CONFIG_CC_OPTIMIZE_FOR_SIZE enabled (gcc 11.1.0):
CC arch/riscv/kernel/vdso/vgettimeofday.o
In file included from <command-line>:
./arch/riscv/include/asm/jump_label.h: In function 'cpu_relax':
././include/linux/compiler_types.h:285:33: warning: 'asm' operand 0 probably does not match constraints
285 | #define asm_volatile_goto(x...) asm goto(x)
| ^~~
./arch/riscv/include/asm/jump_label.h:41:9: note: in expansion of macro 'asm_volatile_goto'
41 | asm_volatile_goto(
| ^~~~~~~~~~~~~~~~~
././include/linux/compiler_types.h:285:33: error: impossible constraint in 'asm'
285 | #define asm_volatile_goto(x...) asm goto(x)
| ^~~
./arch/riscv/include/asm/jump_label.h:41:9: note: in expansion of macro 'asm_volatile_goto'
41 | asm_volatile_goto(
| ^~~~~~~~~~~~~~~~~
make[1]: *** [scripts/Makefile.build:249: arch/riscv/kernel/vdso/vgettimeofday.o] Error 1
make: *** [arch/riscv/Makefile:128: vdso_prepare] Error 2
Having a static branch in cpu_relax() is problematic because that
function is widely inlined, including in some quite complex functions
like in the VDSO. A quick measurement shows this static branch is
responsible by itself for around 40% of the jump table.
Drop the static branch, which ends up being the same number of
instructions anyway. If Zihintpause is supported, we trade the nop from
the static branch for a div. If Zihintpause is unsupported, we trade the
jump from the static branch for (what gets interpreted as) a nop.
Fixes: 8eb060e10185 ("arch/riscv: add Zihintpause support")
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Cc: stable@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
|
|
Alex Elder says:
====================
net: ipa: remaining IPA v5.0 support
This series includes almost all remaining IPA code changes required
to support IPA v5.0. IPA register definitions and configuration
data for IPA v5.0 will be sent later (soon). Note that the GSI
register definitions still require work. GSI for IPA v5.0 supports
up to 256 (rather than 32) channels, and this changes the way GSI
register offsets are calculated. A few GSI register fields also
change.
The first patch in this series increases the number of IPA endpoints
supported by the driver, from 32 to 36. The next updates the width
of the destination field for the IP_PACKET_INIT immediate command so
it can represent up to 256 endpoints rather than just 32. The next
adds a few definitions of some IPA registers and fields that are
first available in IPA v5.0.
The next two patches update the code that handles router and filter
table caches. Previously these were referred to as "hashed" tables,
and the IPv4 and IPv6 tables are now combined into one "unified"
table. The sixth and seventh patches add support for a new pulse
generator, which allows time periods to be specified with a wider
range of clock resolution. And the last patch just defines two new
memory regions that were not previously used.
====================
Link: https://lore.kernel.org/r/20230130210158.4126129-1-elder@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
IPA v5.0 uses two memory regions not previously used. Define them
and treat them as valid only for IPA v5.0.
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The AP has third pulse generator available starting with IPA v5.0.
Redefine ipa_qtime_val() to support that possibility. Pass the IPA
pointer as an argument so the version can be determined. And stop
using the sign of the returned tick count to indicate which of two
pulse generators to use.
Instead, have the caller provide the address of a variable that will
hold the selected pulse generator for the Qtime value. And for
version 5.0, check whether the third pulse generator best represents
the time period.
Add code in ipa_qtime_config() to configure the fourth pulse
generator for IPA v5.0+; in that case configure both the third and
fourth pulse generators to use 10 msec granularity.
Consistently use "ticks" for local variables that represent a tick
count.
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Starting with IPA v5.0, the head-of-line blocking timer has more
than two pulse generators available to define timer granularity.
To prepare for that, change the way the field value is encoded
to use ipa_reg_encode() rather than ipa_reg_bit().
The aggregation granularity selection could (in principle) also use
an additional pulse generator starting with IPA v5.0. Encode the
AGGR_GRAN_SEL field differently to allow that as well.
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
IPA v5.0+ separates the configuration of entries in the cached
(previously "hashed") routing and filtering tables into distinct
registers. Previously a single "filter and router" register updated
entries in both tables at once; now the routing and filter table
caches have separate registers that define their content.
This patch updates the code that zeroes entries in the cached filter
and router tables to support IPA versions including v5.0+.
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Update the code that causes filter and router table caches to be
flushed so that it supports IPA versions 5.0+. It adds a comment in
ipa_hardware_config_hashing() that explains that cacheing does not
need to be enabled, just as before, because it's enabled by default.
(For the record, the FILT_ROUT_CACHE_CFG register would have been
used if we wanted to explicitly enable these.)
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Define some new registers that appear starting with IPA v5.0, along
with enumerated types identifying their fields. Code that uses
these will be added by upcoming patches.
Most of the new registers are related to filter and routing tables,
and in particular, their "hashed" variant. These tables are better
described as "cached", where a hash value determines which entries
are cached. From now on, naming related to this functionality will
use "cache" instead of "hash", and that is reflected in these new
register names. Some registers for managing these caches and their
contents have changed as well.
A few other new field definitions for registers (unrelated to table
caches) are also defined.
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The IP_PACKET_INIT immediate command defines the destination
endpoint to which a packet should be sent. Prior to IPA v5.0, a
5 bit field in that command represents the endpoint, but starting
with IPA v5.0, the field is extended to 8 bits to support more than
32 endpoints.
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Increase the number of endpoints supported by the driver to 36,
which IPA v5.0 supports. This makes it impossible to check at build
time whether the supported number is too big to fit within the
(5-bit) PACKET_INIT destination endpoint field. Instead, convert
the build time check to compare against what fits in 8 bits.
Add a check in ipa_endpoint_config() to also ensure the hardware
reports an endpoint count that's in the expected range. Just
open-code 32 as the limit (the PACKET_INIT field mask is not
available where we'd want to use it).
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2023-01-30
Add fast update encryption key
Jianbo Liu Says:
================
Data encryption keys (DEKs) are the keys used for data encryption and
decryption operations. Starting from version 22.33.0783, firmware is
optimized to accelerate the update of user keys into DEK object in
hardware. The support for bulk allocation and destruction of DEK
objects is added, and the bulk allocated DEKs are uninitialized, as
the bulk creation requires no input key. When offload
encryption/decryption, user gets one object from a bulk, and updates
key by a new "modify DEK" command. This command is the same as create
DEK object, but requires no heavy context memory allocation in
firmware, which consumes most cpu cycles of the create DEK command.
DEKs are cached internally by the NIC, so invalidating internal NIC
caches is required before reusing DEKs. The SYNC_CRYPTO command is
added to support it. DEK object can be reused, the keys in it can be
updated after this command is executed.
This patchset enhances the key creation and destruction flow, to get
use of this new feature. Any user, for example, ktls, ipsec and
macsec, can use it to offload keys. But, only ktls uses it, as others
don't need many keys, and caching two many DEKs in pool is wasteful.
There are two new data struts added:
a. DEK pool. One pool is created for each key type. The bulks by
the type, are placed in the pool's different bulk lists, according to
the number of available and in_used DEKs in the bulk.
b. DEK bulk. All DEKs in one bulk allocation are store here. There
are two bitmaps to indicate the state of each DEK.
New APIs are then added. When user need a DEK object,
a. Fetch one bulk with avail DEKs, from the partial_list or
avail_list, otherwise create new one.
b. Pick one DEK, and set its need_sync and in_used bits to 1.
Move the bulk to full_list if no more available keys, or put it to
partial_list if the bulk is newly created.
c. Update DEK object's key with user key, by the "modify DEK"
command.
d. Return DEK struct to user, then it gets the object id and fills
it into the offload commands.
When user free a DEK,
a. Set in_use bit to 0. If all need_sync bits are 1 and all in_use
bits of this bulk are 0, move it to sync_list.
b. If the number of DEKs, which are freed by users, is over the
threshold (128), schedule a workqueue to do the sync process.
For the sync process, the SYNC_CRYPTO command is executed first. Then,
for each bulks in partial_list, full_list and sync_list, reset
need_sync bits of the freed DEK objects. If all need_sync bits in one
bulk are zero, move it to avail_list.
We already supported TIS pool to recycle the TISes. With this series
and TIS pool, TLS CPS performance is improved greatly.
And we tested https on the system:
CPU: dual AMD EPYC 7763 64-Core processors
RAM: 512G
DEV: ConnectX-6 DX, with FW ver 22.33.0838 and TLS_OPTIMISE=true
TLS CPS performance numbers are:
Before: 11k connections/sec
After: 101 connections/sec
================
* tag 'mlx5-updates-2023-01-30' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
net/mlx5e: kTLS, Improve connection rate by using fast update encryption key
net/mlx5: Keep only one bulk of full available DEKs
net/mlx5: Add async garbage collector for DEK bulk
net/mlx5: Reuse DEKs after executing SYNC_CRYPTO command
net/mlx5: Use bulk allocation for fast update encryption key
net/mlx5: Add bulk allocation and modify_dek operation
net/mlx5: Add support SYNC_CRYPTO command
net/mlx5: Add new APIs for fast update encryption key
net/mlx5: Refactor the encryption key creation
net/mlx5: Add const to the key pointer of encryption key creation
net/mlx5: Prepare for fast crypto key update if hardware supports it
net/mlx5: Change key type to key purpose
net/mlx5: Add IFC bits and enums for crypto key
net/mlx5: Add IFC bits for general obj create param
net/mlx5: Header file for crypto
====================
Link: https://lore.kernel.org/r/20230131031201.35336-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
1) Release bridge info once packet escapes the br_netfilter path,
from Florian Westphal.
2) Revert incorrect fix for the SCTP connection tracking chunk
iterator, also from Florian.
First path fixes a long standing issue, the second path addresses
a mistake in the previous pull request for net.
* git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
Revert "netfilter: conntrack: fix bug in for_each_sctp_chunk"
netfilter: br_netfilter: disable sabotage_in hook after first suppression
====================
Link: https://lore.kernel.org/r/20230131133158.4052-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The Meson G12A Internal PHY does not support standard IEEE MMD extended
register access, therefore add generic dummy stubs to fail the read and
write MMD calls. This is necessary to prevent the core PHY code from
erroneously believing that EEE is supported by this PHY even though this
PHY does not support EEE, as MMD register access returns all FFFFs.
Fixes: 5c3407abb338 ("net: phy: meson-gxl: add g12a support")
Reviewed-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Chris Healy <healych@amazon.com>
Reviewed-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20230130231402.471493-1-cphealy@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Commit 3a1296a38d0c ("net: Support GRO/GSO fraglist chaining.")
introduced UDP listifyed GRO. The segmentation relies on frag_list being
untouched when passing through the network stack. This assumption can be
broken sometimes, where frag_list itself gets pulled into linear area,
leaving frag_list being NULL. When this happens it can trigger
following NULL pointer dereference, and panic the kernel. Reverse the
test condition should fix it.
[19185.577801][ C1] BUG: kernel NULL pointer dereference, address:
...
[19185.663775][ C1] RIP: 0010:skb_segment_list+0x1cc/0x390
...
[19185.834644][ C1] Call Trace:
[19185.841730][ C1] <TASK>
[19185.848563][ C1] __udp_gso_segment+0x33e/0x510
[19185.857370][ C1] inet_gso_segment+0x15b/0x3e0
[19185.866059][ C1] skb_mac_gso_segment+0x97/0x110
[19185.874939][ C1] __skb_gso_segment+0xb2/0x160
[19185.883646][ C1] udp_queue_rcv_skb+0xc3/0x1d0
[19185.892319][ C1] udp_unicast_rcv_skb+0x75/0x90
[19185.900979][ C1] ip_protocol_deliver_rcu+0xd2/0x200
[19185.910003][ C1] ip_local_deliver_finish+0x44/0x60
[19185.918757][ C1] __netif_receive_skb_one_core+0x8b/0xa0
[19185.927834][ C1] process_backlog+0x88/0x130
[19185.935840][ C1] __napi_poll+0x27/0x150
[19185.943447][ C1] net_rx_action+0x27e/0x5f0
[19185.951331][ C1] ? mlx5_cq_tasklet_cb+0x70/0x160 [mlx5_core]
[19185.960848][ C1] __do_softirq+0xbc/0x25d
[19185.968607][ C1] irq_exit_rcu+0x83/0xb0
[19185.976247][ C1] common_interrupt+0x43/0xa0
[19185.984235][ C1] asm_common_interrupt+0x22/0x40
...
[19186.094106][ C1] </TASK>
Fixes: 3a1296a38d0c ("net: Support GRO/GSO fraglist chaining.")
Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/r/Y9gt5EUizK1UImEP@debian
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When memory allocation fails in lynx_pcs_create() and it returns NULL,
there remains a dangling reference to the mdiodev returned by
of_mdio_find_device() which is leaked as soon as memac_pcs_create()
returns empty-handed.
Fixes: a7c2a32e7f22 ("net: fman: memac: Use lynx pcs driver")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Sean Anderson <sean.anderson@seco.com>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Link: https://lore.kernel.org/r/20230130193051.563315-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue
Tony Nguyen says:
====================
Intel Wired LAN: Remove redundant Device Control Error Reporting Enable
Bjorn Helgaas says:
Since f26e58bf6f54 ("PCI/AER: Enable error reporting when AER is native"),
the PCI core sets the Device Control bits that enable error reporting for
PCIe devices.
This series removes redundant calls to pci_enable_pcie_error_reporting()
that do the same thing from several NIC drivers.
There are several more drivers where this should be removed; I started with
just the Intel drivers here.
* '10GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
ixgbe: Remove redundant pci_enable_pcie_error_reporting()
igc: Remove redundant pci_enable_pcie_error_reporting()
igb: Remove redundant pci_enable_pcie_error_reporting()
ice: Remove redundant pci_enable_pcie_error_reporting()
iavf: Remove redundant pci_enable_pcie_error_reporting()
i40e: Remove redundant pci_enable_pcie_error_reporting()
fm10k: Remove redundant pci_enable_pcie_error_reporting()
e1000e: Remove redundant pci_enable_pcie_error_reporting()
====================
Link: https://lore.kernel.org/r/20230130192519.686446-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Petr Machata says:
====================
selftests: mlxsw: Convert to iproute2 dcb
There is a dedicated tool for configuration of DCB in iproute2. Use it
in the selftests instead of lldpad.
Patches #1-#3 convert three tests. Patch #4 drops the now-unnecessary
lldpad helpers.
====================
Link: https://lore.kernel.org/r/cover.1675096231.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The existing users of these helpers have been converted to iproute2 dcb.
Drop the helpers.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Set up default port priority through the iproute2 dcb tool, which is easier
to understand and manage.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Set up DSCP prioritization through the iproute2 dcb tool, which is easier
to understand and manage.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
