| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
FW sends a notification indicating whether activating EMLSR mode is
recommended or not.
Support the notification and enter EMLSR only if recommended.
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240505091420.2fd3387882eb.I7a8a5b24658744ed732bfc03b1872c9298483d62@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
Adjust EMLSR activation to account for traffic levels. By
tracking the number of RX/TX MPDUs, EMLSR will be activated only when
traffic volume meets the required threshold.
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240505091420.9480f99ac8fc.If9eb946e929a39e10fe5f4638bc8bc3f8976edf1@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
When an event occurs to unblock EMLSR, the code attempts to re-enable
EMLSR. However, the current implementation always tries to activate
EMLSR, regardless of whether the blocker was set before the unblocking
event or not. If EMLSR was already unblocked, there is no need to
re-activate it.
Fixes: 6cf7df9f013f ("wifi: iwlwifi: mvm: Add helper functions to update EMLSR status")
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240505091420.eb861402dac9.I6a1d9f774f5551cfab60ea37b71a62640496af9b@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
EMLSR can't be activated from mac80211. Except for the debugfs, which is
intended for testing purposes. Currently we don't allow entering EMLSR
from debugfs if EMLSR is blocked, i.e. if mvmvif::esr_disable_reason is
not 0. But we need a way to activate EMLSR regardless of the vif being
blocked, for testing. Remove the check of esr_disable_reason
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240505091420.bc3c24d9e0e6.Iad60e22a0d7e2b2b989051e1140b6dc98bef7bcc@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
This is needed for testing purposes.
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240505091420.eba2b6f0664c.I5f058e02abda11bf2eccfd2bcb59ca26bae87a3a@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
If the reason for exiting EMLSR was a blocking reason, wait for the
corresponding unblocking event:
- if there is an ongoing scan - do nothing. Link selection will be
triggered at the end of it.
- If more than 30 seconds passed since the exit, trigger MLO scan, which
will trigger link selection
- If less then 30 seconds passed since exit, reuse the latest link
selection result
If the reason for exiting EMLSR was an exit reason (IWL_MVM_EXIT_*),
schedule MLO scan in 30 seconds.
Signed-off-by: Miri Korenblit <[email protected]>
Reviewed-by: Ilan Peer <[email protected]>
Link: https://msgid.link/20240505091420.6a808c4ae8f5.Ia79605838eb6deee9358bec633ef537f2653db92@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
BT Coex disables EMLSR only for a 2.4 GHz link, but doesn't block the
vif from using EMLSR with a different link pair. In addition, storing it
in mvmvif:disable_esr_reason requires extracting the BT Coex bit before
checking if EMLSR is blocked or not for a specific vif.
Therefore, change the BT Coex bit to be an exit reason and not a
blocker. On link selection, EMLSR mode will be re-calculated for the 2.4
GHz link instead of checking that bit.
While at it, move the relevant function declarations to the EMLSR
functions area in mvm.h
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240505091420.a2e93b67c895.I183a0039ef076613144648cc46fbe9ab3d47c574@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
Given how late we are in the cycle, merge the two fixes from
wireless into wireless-next as they don't see that urgent.
This way, the wireless tree won't need rebasing later.
Signed-off-by: Johannes Berg <[email protected]>
|
|
The PTP_CMD_CTL is a self clearing register which controls the PTP clock
values. In the current implementation driver waits for a duration of 20
sec in case of HW failure to clear the PTP_CMD_CTL register bit. This
timeout of 20 sec is very long to recognize a HW failure, as it is
typically cleared in one clock(<16ns). Hence reducing the timeout to 1 sec
would be sufficient to conclude if there is any HW failure observed. The
usleep_range will sleep somewhere between 1 msec to 20 msec for each
iteration. By setting the PTP_CMD_CTL_TIMEOUT_CNT to 50 the max timeout
is extended to 1 sec.
Signed-off-by: Rengarajan S <[email protected]>
Reviewed-by: Simon Horman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Paolo Abeni <[email protected]>
|
|
rtw-next patches for v6.10
Major changes are listed as below
rtl8xxxu:
- remove rtl8xxxu_ prefix from filename
- cleanup includes of header files
rtlwifi:
- adjust code to share with coming support of rtl8192du
rtw89:
- complete features of new WiFi 7 chip 8922AE including BT-coexistence
and WoWLAN
- use BIOS ACPI settings to set TX power and channels
|
|
Every tx and rx ring has its own queue-page-list (QPL) that serves as
the bounce buffer. Previously we were allocating QPLs for all queues
before the queues themselves were allocated and later associating a QPL
with a queue. This is avoidable complexity: it is much more natural for
each queue to allocate and free its own QPL.
Moreover, the advent of new queue-manipulating ndo hooks make it hard to
keep things as is: we would need to transfer a QPL from an old queue to
a new queue, and that is unpleasant.
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
We now account for the fact that the NIC might send us stats for a
subset of queues. Without this change, gve_get_ethtool_stats might make
an invalid access on the priv->stats_report->stats array.
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This does not fix any existing bug. In anticipation of the ndo queue api
hooks that alloc/free/start/stop a single Rx queue, the already existing
per-queue stop functions are being made more robust. Specifically for
this use case: rx_queue_n.stop() + rx_queue_n.start()
Note that this is not the use case being used in devmem tcp (the first
place these new ndo hooks would be used). There the usecase is:
new_queue.alloc() + old_queue.stop() + new_queue.start() + old_queue.free()
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In order to make possible the implementation of per-queue ndo hooks,
gve_turnup was changed in a previous patch to account for queues already
having some unprocessed descriptors: it does a one-off napi_schdule to
handle them. If conditions of consistent high traffic persist in the
immediate aftermath of this, the poll routine for a queue can be "stuck"
on the cpu on which the ndo hooks ran, instead of the cpu its irq has
affinity with.
This situation is exacerbated by the fact that the ndo hooks for all the
queues are invoked on the same cpu, potentially causing all the napi
poll routines to be residing on the same cpu.
A self correcting mechanism in the poll method itself solves this
problem.
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
gVNIC has a requirement that all queues have to be quiesced before any
queue is operated on (created or destroyed). To enable the
implementation of future ndo hooks that work on a single queue, we need
to evolve gve_turnup to account for queues already having some
unprocessed descriptors in the ring.
Say rxq 4 is being stopped and started via the queue api. Due to gve's
requirement of quiescence, queues 0 through 3 are not processing their
rings while queue 4 is being toggled. Once they are made live, these
queues need to be poked to cause them to check their rings for
descriptors that were written during their brief period of quiescence.
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Currently the queues are either all live or all dead, toggling from one
state to the other via the ndo open and stop hooks. The future addition
of single-queue ndo hooks changes this, and thus gve_turnup and
gve_turndown should evolve to account for a state where some queues are
live and some aren't.
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
This allows for implementing future ndo hooks that act on a single
queue.
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Although this is not fixing any existing double free bug, making these
functions idempotent allows for a simpler implementation of future ndo
hooks that act on a single queue.
Tested-by: Mina Almasry <[email protected]>
Reviewed-by: Praveen Kaligineedi <[email protected]>
Reviewed-by: Harshitha Ramamurthy <[email protected]>
Signed-off-by: Shailend Chand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
le32p_replace_bits() only updates partial bits of rate_mask, and GCC warns
below. Set initial value to avoid warnings, and prevent random value of
missed bits (bit 6 of rate_mask.macid_and_short_gi).
In file included from ./include/linux/fortify-string.h:5,
from ./include/linux/string.h:369,
from ./include/linux/bitmap.h:13,
from ./include/linux/cpumask.h:13,
from ./include/linux/sched.h:16,
from drivers/net/wireless/realtek/rtlwifi/rtl8192d/../wifi.h:9,
from drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c:4:
In function 'le32p_replace_bits',
inlined from 'rtl92de_update_hal_rate_mask.isra' at drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c:986:2:
./include/linux/bitfield.h:189:15: warning: 'rate_mask' is used uninitialized [-Wuninitialized]
189 | *p = (*p & ~to(field)) | type##_encode_bits(val, field); \
| ^~
./include/linux/bitfield.h:196:9: note: in expansion of macro '____MAKE_OP'
196 | ____MAKE_OP(le##size,u##size,cpu_to_le##size,le##size##_to_cpu) \
| ^~~~~~~~~~~
./include/linux/bitfield.h:201:1: note: in expansion of macro '__MAKE_OP'
201 | __MAKE_OP(32)
| ^~~~~~~~~
drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c: In function 'rtl92de_update_hal_rate_mask.isra':
drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c:863:37: note: 'rate_mask' declared here
863 | struct rtl92d_rate_mask_h2c rate_mask;
| ^~~~~~~~~
Compile tested only.
Fixes: 014bba73b525 ("wifi: rtlwifi: Adjust rtl8192d-common for USB")
Cc: Bitterblue Smith <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Wflex-array-member-not-at-end is coming in GCC-14, and we are getting
ready to enable it globally.
So, remove unused structs and fix the following
-Wflex-array-member-not-at-end warnings:
drivers/net/wireless/realtek/rtlwifi/btcoexist/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8188ee/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8192c/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8192ce/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8192cu/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8192de/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8192ee/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8192se/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8723ae/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8723be/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8723com/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/rtl8821ae/../wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/net/wireless/realtek/rtlwifi/wifi.h:1063:30: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
Signed-off-by: Gustavo A. R. Silva <[email protected]>
Reviewed-by: Kees Cook <[email protected]>
Acked-by: Ping-Ke Shih <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/ZjLFIa31BGPVCGh1@neat
|
|
aSIFSTime is 10us for 2GHz band and 16us for 5GHz and 6GHz bands.
Originally, it doesn't consider 6GHz band and use wrong value, so correct
it accordingly.
Cc: [email protected]
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Add H2C command and offload template packet to allow firmware send ARP
response in WoWLAN mode. Then, firmware in WoWLAN mode can interactive
with peer that issue ARP request to query MAC address.
Signed-off-by: Chin-Yen Lee <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
When using the WEP cipher, we need to add the address cam type as all
unicast mode to let firmware to work. Although WEP only set GTK in
mac80211, but we need to set both PTK and GTK information to firmware.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Once we connect to AP with 802.11w enabled, IGTK rekey happen during GTK
happen. We get IGTK IPN from mac80211 and set to firmware, and get latest
IGTK IPN from AOAC report then update to mac80211 after resume. When rekey
happen, also update new IGTK key info to mac80211. Furthermore, We
construct SA query reply packet to firmware. If firmware received SA query
request from AP can transmit reply back when suspend.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
When resume we parse AOAC report from firmware using H2C and C2H
registers before enable interrupt, then update PTK RX PN and GTK RX PN.
After enable interrupt, we parse AOAC report using H2C and C2H commands,
then update PTK TX PN and update new GTK key info if GTK rekey during
suspend. Furthermore, We update pattern match index if wakeup by pattern.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Add PTK TRX IV, GTK RX IV, key encryption algorithm to H2C command to
enable GTK rekey feature.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
We construct EAPoL packet with various encryption method, and download
to firmware. Also we add Key Encryption Key (KEK) and Key Confirmation Key
(KCK) to H2C command. Once firmware received EAPoL group rekey packet(1/2)
can TX EAPoL group rekey packet(2/2) when suspend.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
This H2C command is used to set WoWLAN global config, and we correct
the H2C format by enlarging the H2C size to fill GTK and PTK info.
This fix is compatible with old firmware.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
This H2C command set key information into security CAM including key
index, entry index and valid map. No logic is changed.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Get the PTK and PTK TRX PN value and transfer to IV value, these
values will used by firmware to generate packets with correct IV value.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Need Auth Key Management(AKM) to let firmware to generate appropriate
EAPoL packet for GTK rekey. The AKM is present in the association request
RSN IE to indicate which cipher that station selected.
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
After enabling packet offload, the TX will be stuck after resume from
WoWLAN mode. And the 8852c gets error messages like
rtw89_8852ce 0000:04:00.0: No busy txwd pages available
rtw89_8852ce 0000:04:00.0: queue 0 txwd 100 is not idle
rtw89_8852ce 0000:04:00.0: queue 0 txwd 101 is not idle
rtw89_8852ce 0000:04:00.0: queue 0 txwd 102 is not idle
rtw89_8852ce 0000:04:00.0: queue 0 txwd 103 is not idle
If suspend/resume many times that firmware will download failed and
disconnection.
To fix these issues, We removed the rtw89_hci_disable_intr() and
rtw89_hci_enable_intr() during rtw89_wow_swap_fw() to prevent add packet
offload can't receive c2h back due to interrupt disable. Only 8852C and
8922A needs to disable interrupt before downloading fw.
Furthermore, we avoid using low power HCI mode on WoWLAN mode, to prevent
interrupt enabled, then get interrupt and calculate RXBD mismatched due to
software RXBD index already reset but hardware RXBD index not yet.
Fixes: 5c12bb66b79d ("wifi: rtw89: refine packet offload flow")
Signed-off-by: Chih-Kang Chang <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
802.11be WiFi chips need a RFK (RF calibration) notify H2C command after
downloading WoWLAN firmware to make sure RF TX/RX work fine when leaving
power save mode, so add it to correct RF TX/RX in WoWLAN mode.
Signed-off-by: Chin-Yen Lee <[email protected]>
Signed-off-by: Ping-Ke Shih <[email protected]>
Link: https://msgid.link/[email protected]
|
|
This reverts commit a580ea994fd37f4105028f5a85c38ff6508a2b25.
This revert is to resolve Dragos's report of page_pool leak here:
https://lore.kernel.org/lkml/[email protected]/
The reverted patch interacts very badly with commit 2cc3aeb5eccc ("skbuff:
Fix a potential race while recycling page_pool packets"). The reverted
commit hopes that the pp_recycle + is_pp_page variables do not change
between the skb_frag_ref and skb_frag_unref operation. If such a change
occurs, the skb_frag_ref/unref will not operate on the same reference type.
In the case of Dragos's report, the grabbed ref was a pp ref, but the unref
was a page ref, because the pp_recycle setting on the skb was changed.
Attempting to fix this issue on the fly is risky. Lets revert and I hope
to reland this with better understanding and testing to ensure we don't
regress some edge case while streamlining skb reffing.
Fixes: a580ea994fd3 ("net: mirror skb frag ref/unref helpers")
Reported-by: Dragos Tatulea <[email protected]>
Signed-off-by: Mina Almasry <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
|
|
Current net-next/main does not boot for older chipsets e.g. Stratus.
Sample dmesg:
[ 11.368315] bnxt_en 0000:02:00.0 (unnamed net_device) (uninitialized): Able to reserve only 0 out of 9 requested RX rings
[ 11.390181] bnxt_en 0000:02:00.0 (unnamed net_device) (uninitialized): Unable to reserve tx rings
[ 11.438780] bnxt_en 0000:02:00.0 (unnamed net_device) (uninitialized): 2nd rings reservation failed.
[ 11.487559] bnxt_en 0000:02:00.0 (unnamed net_device) (uninitialized): Not enough rings available.
[ 11.506012] bnxt_en 0000:02:00.0: probe with driver bnxt_en failed with error -12
This is caused by bnxt_get_avail_msix() returning a negative value for
these chipsets not using the new resource manager i.e. !BNXT_NEW_RM.
This in turn causes hwr.cp in __bnxt_reserve_rings() to be set to 0.
In the current call stack, __bnxt_reserve_rings() is called from
bnxt_set_dflt_rings() before bnxt_init_int_mode(). Therefore,
bp->total_irqs is always 0 and for !BNXT_NEW_RM bnxt_get_avail_msix()
always returns a negative number.
Historically, MSIX vectors were requested by the RoCE driver during
run-time and bnxt_get_avail_msix() was used for this purpose. Today,
RoCE MSIX vectors are statically allocated. bnxt_get_avail_msix() should
only be called for the BNXT_NEW_RM() case to reserve the MSIX ahead of
time for RoCE use.
bnxt_get_avail_msix() is also be simplified to handle the BNXT_NEW_RM()
case only.
Fixes: d630624ebd70 ("bnxt_en: Utilize ulp client resources if RoCE is not registered")
Signed-off-by: David Wei <[email protected]>
Reviewed-by: Michael Chan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
|
|
Currently the driver uses local_bh_disable()/local_bh_enable() in its
IRQ handler to avoid triggering net_rx_action() softirq on exit from
netif_rx(). The net_rx_action() could trigger this driver .start_xmit
callback, which is protected by the same lock as the IRQ handler, so
calling the .start_xmit from netif_rx() from the IRQ handler critical
section protected by the lock could lead to an attempt to claim the
already claimed lock, and a hang.
The local_bh_disable()/local_bh_enable() approach works only in case
the IRQ handler is protected by a spinlock, but does not work if the
IRQ handler is protected by mutex, i.e. this works for KS8851 with
Parallel bus interface, but not for KS8851 with SPI bus interface.
Remove the BH manipulation and instead of calling netif_rx() inside
the IRQ handler code protected by the lock, queue all the received
SKBs in the IRQ handler into a queue first, and once the IRQ handler
exits the critical section protected by the lock, dequeue all the
queued SKBs and push them all into netif_rx(). At this point, it is
safe to trigger the net_rx_action() softirq, since the netif_rx()
call is outside of the lock that protects the IRQ handler.
Fixes: be0384bf599c ("net: ks8851: Handle softirqs at the end of IRQ thread to fix hang")
Tested-by: Ronald Wahl <[email protected]> # KS8851 SPI
Signed-off-by: Marek Vasut <[email protected]>
Reviewed-by: Eric Dumazet <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
|
|
In mesh node traffic, internal firmware-transmitted failures are
reported as transmitted failures in mesh metric calculation, leading
to the breakage of the mesh link.
Fix the issue by dropping the internal firmware-transmitted failures
before updating the TX completion status to mac80211, in order to
prevent false failure averaging in mesh metric calculation.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Karthikeyan Kathirvel <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
When a station idles for a long time, hostapd will try to send
a QoS Null frame to the station as "poll". NL80211_CMD_PROBE_CLIENT
is used for this purpose.
And the skb will be added to ack_status_frame - waiting for a
completion via ieee80211_report_ack_skb().
But when the peer was already removed before the tx_complete arrives,
the peer will be missing. And when using dev_kfree_skb_any (instead
of going through mac80211), the entry will stay inside
ack_status_frames thus not clean up related information in its
internal data structures. This IDR will therefore run full after
8K request were generated for such clients.
At this point, the access point will then just stall and not allow
any new clients because idr_alloc() for ack_status_frame will fail.
ieee80211_free_txskb() on the other hand will (when required) call
ieee80211_report_ack_skb() and make sure that (when required) remove
the entry from the ack_status_frame and clean up related
information in its internal data structures.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Sven Eckelmann <[email protected]>
Signed-off-by: Sarika Sharma <[email protected]>
Signed-off-by: Karthikeyan Kathirvel <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Currently, inter device MLO is not supported. Therefore, the WBM idle ring
buffers choose the implicit return buffer manager (DEV0_IDLE_DESC_LIST).
However, this implicit return buffer manager design is not meeting the
requirements to support inter device MLO. In inter device MLO, multiple
devices participate. The device specific WBM idle ring buffers transmit
to multiple device REO rings. To distinguish between device specific WBM
idle buffers, the setup configuration need to choose a different return
buffer manager based on the unique identifier (device index).
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Karthikeyan Periyasamy <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Introduce a device index parameter in the ath12k_base structure. This
value is used as unique identifier for devices involved in Inter-device
MLO. It is also used to configure the MLO parameter within the QMI host
capability send request. While currently set to zero, future logic will
be added for inter device support to populate this parameter with
non-zero values.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Karthikeyan Periyasamy <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Currently, the WBM idle ring descriptor setup uses implicit value
HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST for the return buffer manager
parameter. To support inter-device MLO (Multi-link operation), this
parameter needs to be configure dynamically based on the device identifier
within the MLO group. Therefore, introduce a new argument to the helper
function ath12k_hal_set_link_desc_addr().
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Karthikeyan Periyasamy <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Prefer the term "device" over "chip". Therefore, modify the relevant terms
in hal Rx return buffer manager to use "device" instead.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Karthikeyan Periyasamy <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Currently, in ath11k_ahb_probe(), if a failure occurs after
ath11k_ahb_ce_remap() is called, and if hw_params->ce_remap is
enabled, the CE register memory is not unmapped. So add a call to
ath11k_ahb_ce_unmap() in the error path.
This issue was identified by smatch/smatch_scripts/kchecker:
drivers/net/wireless/ath/ath11k/ahb.c:1218 ath11k_ahb_probe() warn: 'ab->mem_ce' from ioremap() not released on lines: 1218.
Since this is a rare error path with no simple way to test, and since
the change is trivial to review, compile tested only.
Signed-off-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Currently the logic that handles hw_params->ce_remap is inline code,
both for doing the remap and the unmap. An upcoming change needs to do
the unmap in a second place, so refactor the unmap logic into a
separate function. And although it is only called from one place,
refactor the remap logic as well to have functional symmetry.
No functional changes, compile tested only.
Signed-off-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Currently, hardware state is not protected across the reconfigure
operations. However, in single wiphy models, multiple radio/links is
exposed as a MAC hardware (ieee80211_hw) through the driver hardware
abstraction (ath12k_hw) layer. In such scenario, we need to protect
hardware state across the multiple radio/link at the driver hardware
abstraction (ath12k_hw) layer. Therefore, introduce a new mutex in
the ath12k_hw layer.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Karthikeyan Periyasamy <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Currently, in multi wiphy models, the mac80211 hardware state is maintained
within the radio/link structure. However, in single wiphy models, the
mac80211 hardware state is needed at the hardware abstraction layer
(ath12k_hw). Therefore, move the hardware state from the radio/link
structure to the hardware abstraction layer (ath12k_hw). Additionally,
update the naming convention of the state enums to enhance clarity and
consistency.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Karthikeyan Periyasamy <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Currently, in multi-wiphy models, the recovery handler access mac80211
HW from the radio/link structure. This will be incorrect for single wiphy
model, as they will hold multiple link/radio structures. To fix this,
access mac80211 HW based on the number of hardware in the SoC/chip. This
approach makes the recovery handler compatible with both multi wiphy and
single wiphy models.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Karthikeyan Periyasamy <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Link: https://msgid.link/[email protected]
|
|
Simplify the loop by using list_for_each_entry(). No functional changes.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.37
Signed-off-by: Kalle Valo <[email protected]>
Acked-by: Jeff Johnson <[email protected]>
Link: https://msgid.link/[email protected]
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath
ath.git patches for v6.10
ath12k
* debugfs support
* dfs_simulate_radar debugfs file
* disable Wireless Extensions
* suspend and hibernation support
* ACPI support
* refactoring in preparation of multi-link support
ath11k
* support hibernation (required changes in qrtr and MHI subsystems)
* ieee80211-freq-limit Device Tree property support
ath10k
* firmware-name Device Tree property support
|
|
mt76 patches for 6.10
- fixes
- mt7603 stability improvements
- mt7921 LED control
- mt7925 EHT radiotap support
|