This patch changes mac80211 rate control for the ath9k driver. The rate lookup
per packet is changed from legacy usage of ieee80211_get_tx_rates() to the new
rate table based lookup in struct ieee80211_sta->rates.
The most recent rate control API, introduced with commit 0d528d85c5
("mac80211: improve the rate control API"), allows drivers to directly get
rates from ieee80211_sta->rates. This is not used by every driver yet, the
translation/merge is currently performed in ieee80211_get_tx_rates. This patch
changes the behaviour and avoids the call to ieee80211_get_tx_rates and
subsequent calls. ath9k now directly reads rates from sta->rates into its rate
table. Cause ath9k does not expect rate selection in SKB->CB, the table merge
does not consider rate array in SKB->CB except for the first entry (used for
probing).
Tested with a 8devices Rambutan with QCA9558 SoC by performing two runs, one
without the patch and one with. Generated traffic between AP and multiple STAs
in each run, measured throughput and captured rc_stats. Comparison of both
runs resulted in same rate selection and no performance loss or other negative
effects.
Co-developed-by: Thomas Huehn <thomas.huehn@hs-nordhausen.de>
Signed-off-by: Thomas Huehn <thomas.huehn@hs-nordhausen.de>
Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20211128090753.958-1-jelonek.jonas@gmail.com
On an imx6dl-pico-pi board with a QCA9377 SDIO chip, simply trying to
connect via ssh to another machine causes:
[ 55.824159] ath10k_sdio mmc1:0001:1: failed to transmit packet, dropping: -12
[ 55.832169] ath10k_sdio mmc1:0001:1: failed to submit frame: -12
[ 55.838529] ath10k_sdio mmc1:0001:1: failed to push frame: -12
[ 55.905863] ath10k_sdio mmc1:0001:1: failed to transmit packet, dropping: -12
[ 55.913650] ath10k_sdio mmc1:0001:1: failed to submit frame: -12
[ 55.919887] ath10k_sdio mmc1:0001:1: failed to push frame: -12
, leading to an ssh connection failure.
One user inspected the size of frames on Wireshark and reported
the followig:
"I was able to narrow the issue down to the mtu. If I set the mtu for
the wlan0 device to 1486 instead of 1500, the issue does not happen.
The size of frames that I see on Wireshark is exactly 1500 after
setting it to 1486."
Clearing the HI_ACS_FLAGS_ALT_DATA_CREDIT_SIZE avoids the problem and
the ssh command works successfully after that.
Introduce a 'credit_size_workaround' field to ath10k_hw_params for
the QCA9377 SDIO, so that the HI_ACS_FLAGS_ALT_DATA_CREDIT_SIZE
is not set in this case.
Tested with QCA9377 SDIO with firmware WLAN.TF.1.1.1-00061-QCATFSWPZ-1.
Fixes: 2f918ea986 ("ath10k: enable alt data of TX path for sdio")
Signed-off-by: Fabio Estevam <festevam@denx.de>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20211124131047.713756-1-festevam@denx.de
Currently 'ar' reference is not added in skb_cb during
WMI mgmt tx. Though this is generally not used during tx completion
callbacks, on interface removal the remaining idr cleanup callback
uses the ar ptr from skb_cb from mgmt txmgmt_idr. Hence
fill them during tx call for proper usage.
Also free the skb which is missing currently in these
callbacks.
Crash_info:
[19282.489476] Unable to handle kernel NULL pointer dereference at virtual address 00000000
[19282.489515] pgd = 91eb8000
[19282.496702] [00000000] *pgd=00000000
[19282.502524] Internal error: Oops: 5 [#1] PREEMPT SMP ARM
[19282.783728] PC is at ath11k_mac_vif_txmgmt_idr_remove+0x28/0xd8 [ath11k]
[19282.789170] LR is at idr_for_each+0xa0/0xc8
Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.5.0.1-00729-QCAHKSWPL_SILICONZ-3 v2
Signed-off-by: Sriram R <quic_srirrama@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1637832614-13831-1-git-send-email-quic_srirrama@quicinc.com
Since firmware uses its own sequence number counters, we need to
use firmware number as well when mac80211 generates the ADD_BA
request packet. Indeed the firmware sequence counters tend to
slightly drift from the mac80211 ones because of firmware offload
features like ARP responses. This causes the starting sequence
number field of the ADD_BA request to be unaligned, and can possibly
cause issues with strict/picky APs.
To fix this, we retrieve the current firmware sequence number for
a given TID through the smd_trigger_ba API, and use that number as
replacement of the mac80211 starting sequence number.
This change also ensures that any issue in the smd *ba procedures
will cause the ba action to properly fail, and remove useless call
to smd_trigger_ba() from IEEE80211_AMPDU_RX_START.
Signed-off-by: Loic Poulain <loic.poulain@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1637604251-11763-1-git-send-email-loic.poulain@linaro.org
CE interrupt configuration uses host ce parameters to assign/free
interrupts. Use host ce parameters to enable/disable interrupts.
This patch fixes below BUG,
BUG: KASAN: global-out-of-bounds in 0xffffffbffdfb035c at addr
ffffffbffde6eeac
Read of size 4 by task kworker/u8:2/132
Address belongs to variable ath11k_core_qmi_firmware_ready+0x1b0/0x5bc [ath11k]
OOB is due to ath11k_ahb_ce_irqs_enable() iterates ce_count(which is 12)
times and accessing 12th element in target_ce_config
(which has only 11 elements) from ath11k_ahb_ce_irq_enable().
With this change host ce configs are used to enable/disable interrupts.
Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.5.0.1-00471-QCAHKSWPL_SILICONZ-1
Fixes: 967c1d1131 ("ath11k: move target ce configs to hw_params")
Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1637249558-12793-1-git-send-email-akolli@codeaurora.org
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memset(), avoid intentionally writing across
neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing
beyond the destination member that is intended to be the starting point
of zeroing through the end of the struct. Additionally split up a later
field-spanning memset() so that memset() can reason about the size.
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211118202416.1286046-1-keescook@chromium.org
This reverts commit 46e46db313. Mark reported
that it breaks QCA6390 hw2.0 on Dell XPS 13 9310:
[ 5.537034] ath11k_pci 0000:72:00.0: chip_id 0x0 chip_family 0xb board_id 0xff soc_id 0xffffffff
[ 5.537038] ath11k_pci 0000:72:00.0: fw_version 0x101c06cc fw_build_timestamp 2020-06-24 19:50 fw_build_id
[ 5.537236] ath11k_pci 0000:72:00.0: failed to fetch board data for bus=pci,qmi-chip-id=0,qmi-board-id=255,variant=DE_1901 from ath11k/QCA6390/hw2.0/board-2.bin
[ 5.537255] ath11k_pci 0000:72:00.0: failed to fetch board-2.bin or board.bin from QCA6390/hw2.0
[ 5.537257] ath11k_pci 0000:72:00.0: qmi failed to fetch board file: -2
[ 5.537258] ath11k_pci 0000:72:00.0: failed to load board data file: -2
So we need to back to the drawing board and implement it so that backwards
compatiblity is not broken.
Reported-by: Mark Herbert <mark.herbert42@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211124094316.9096-1-kvalo@codeaurora.org
In HE monitor capture, HAL_TLV_STATUS_PPDU_DONE is received
on processing multiple skb. Do not clear the ppdu_info
till the HAL_TLV_STATUS_PPDU_DONE is received.
This fixes below warning and packet drops in monitor mode.
"Rate marked as an HE rate but data is invalid: MCS: 6, NSS: 0"
WARNING: at
PC is at ieee80211_rx_napi+0x624/0x840 [mac80211]
Tested-on: IPQ8074 hw2.0 AHB WLAN.HK.2.4.0.1-01693-QCAHKSWPL_SILICONZ-1
Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1637249433-10316-1-git-send-email-akolli@codeaurora.org
This is to read variant from SMBIOS such as read from DT, the variant
string will be used to one part of string which used to search board
data from board-2.bin.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-01720.1-QCAHSPSWPL_V1_V2_SILICONZ_LITE-1
Signed-off-by: Wen Gong <quic_wgong@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211118100033.8384-1-quic_wgong@quicinc.com
The ath11k driver currently sends vdev down to the firmware before
updating the channel context, which is followed by a vdev restart
command.
Sending vdev down is not required before sending a vdev restart,
because the firmware internally does vdev down when ath11k sends
a vdev restart command.
Firmware will happen crash while channel switch without this change.
Hence skip the vdev down command sending when updating the channel
context and then fix the firmware crash issue.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Wen Gong <quic_wgong@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211118095901.8271-1-quic_wgong@quicinc.com
The pdev id is set to 0 for single pdev configured hardware, the real
pdev id is not 0 in firmware, for example, its pdev id is 1 for 5G/6G
phy and 2 for 2G band phy. For HTT_H2T_MSG_TYPE_EXT_STATS_CFG message,
firmware parse the pdev_mask to its pdev id, ath11k set it to 0 for
single pdev, it is not correct, need set it with the real pdev id of
firmware.
Save the real pdev id report by firmware and set it correctly.
Below commands run success with this patch:
cat /sys/kernel/debug/ieee80211/phy0/ath11k/htt_stats
cat /sys/kernel/debug/ieee80211/phy0/netdev\:wls1/stations/00\:03\:7f\:75\:59\:85/htt_peer_stats
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Wen Gong <quic_wgong@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211118095700.8149-1-quic_wgong@quicinc.com
When connected to 6G mode AP, it does not have VHT/HT capabilities,
so the NSS is not set, then it is 1 by default.
This patch is to calculate the NSS with supported HE-MCS and NSS set
of HE capabilities.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-01280-QCAHSPSWPL_V1_V2_SILICONZ_LITE-1
Signed-off-by: Wen Gong <quic_wgong@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211118095453.8030-1-quic_wgong@quicinc.com
Some firmware versions for WCN6855 report the default regdomain with
alpha code "na" by default when load as a world wide regdomain, ath11k
should treat it as a world wide alpha code.
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-01720.1-QCAHSPSWPL_V1_V2_SILICONZ_LITE-1
Signed-off-by: Wen Gong <quic_wgong@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211118094848.7776-1-quic_wgong@quicinc.com
With VT-d disabled on Intel platform, ath11k gets only one MSI
vector. In that case, ath11k does not free IRQ when doing suspend,
hence the kernel has to migrate it to CPU0 (if it was affine to
other CPUs) and allocates a new MSI vector. However, ath11k has
no chance to reconfig it to HW srngs during this phase, thus
ath11k fails to resume.
This issue can be fixed by setting IRQ affinity to CPU0 before
request_irq is called. With such affinity, migration will not
happen and thus the vector keeps unchanged during suspend/resume.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Baochen Qiang <bqiang@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211026041732.5323-1-bqiang@codeaurora.org
Current code enables ASPM by default, it allows MHI to enter M2 state.
In case of one MSI vector, system hang is observed if ath11k does MHI
register reading in this state. The issue was reported on Dell XPS 13
9310 but is seen also on XPS 15 and XPS 17 laptops.
The workaround here is to prevent MHI from entering M2 state, this can
be done by disabling ASPM if only one MSI vector is used. When using 32
vectors ASPM is enabled as before.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Carl Huang <cjhuang@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Baochen Qiang <bqiang@codeaurora.org>
Link: https://lore.kernel.org/r/20211026041722.5271-1-bqiang@codeaurora.org
On some platforms it's not possible to allocate 32 MSI vectors for various
reasons, be it kernel configuration, VT-d disabled, buggy BIOS etc. So
ath11k was not able to use QCA6390 PCI devices on those platforms. Add
support for one MSI vector to solve that.
In case of one MSI vector, interrupt migration needs to be disabled. This
is because when interrupt migration happens, the msi_data may change.
However, msi_data is already programmed to rings during initial phase and
ath11k has no way to know that msi_data is changed during run time and
reprogram again.
In case of one MSI vector, MHI subsystem should not use IRQF_NO_SUSPEND
as QCA6390 doesn't set this flag too. Ath11k doesn't need to leave
IRQ enabled in suspend state.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Carl Huang <cjhuang@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Baochen Qiang <bqiang@codeaurora.org>
Link: https://lore.kernel.org/r/20211026041714.5219-1-bqiang@codeaurora.org
This is to prepare for one MSI vector support. IRQ enable and disable
of CE and DP are done only in case of multiple MSI vectors.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Carl Huang <cjhuang@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Baochen Qiang <bqiang@codeaurora.org>
Link: https://lore.kernel.org/r/20211026041705.5167-1-bqiang@codeaurora.org
Like ATH11K_PCI_IRQ_CE0_OFFSET, define ATH11K_PCI_IRQ_DP_OFFSET for
DP to save the IRQ instead of base_vector from MSI config.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Carl Huang <cjhuang@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Baochen Qiang <bqiang@codeaurora.org>
Link: https://lore.kernel.org/r/20211026041655.5112-1-bqiang@codeaurora.org
This change adds two flags to indicate whether IRQ handler for CE
and DP can be called. This is because in one MSI vector case,
interrupt is not disabled in hif_stop and hif_irq_disable. Otherwise,
MHI interrupt is disabled too.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Carl Huang <cjhuang@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Signed-off-by: Baochen Qiang <bqiang@codeaurora.org>
Link: https://lore.kernel.org/r/20211026041646.5060-1-bqiang@codeaurora.org
The reservation mode of interrupts in kernel assigns a dummy vector
when the interrupt is allocated and assigns a real vector when the
request_irq is called. The reservation mode helps to ease vector
pressure when devices with a large amount of queues/interrupts
are initialized, but only a minimal subset of those queues/interrupts
is actually used.
So on reservation mode, the msi_data may change after request_irq
is called, so ath11k reads msi_data again after request_irq is called,
and then the correct msi_data is programmed into QCA6390 hardware
components. Without this change, spurious interrupt occurs in case of
one MSI vector. When VT-d in BIOS is enabled and ath11k can get 32 MSI
vectors, ath11k always get the same msi_data before and after request_irq,
that's why this change is only required when one MSI vector is to be
supported.
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
Signed-off-by: Carl Huang <cjhuang@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211026041636.5008-1-bqiang@codeaurora.org
ath.git patches for v5.17. Major changes:
ath10k
* fetch (pre-)calibration data via nvmem subsystem
ath11k
* enable 802.11 power save mode in station mode for qca6390 and wcn6855
* trace log support
* proper board file detection for WCN6855 based on PCI ids
* BSS color change support
The struct ath_hw member intr_txqs is never reset/assigned outside
of ath9k_hw_init_queues() and with the used bitwise-or in the interrupt
handling ar9002_hw_get_isr() accumulates all ever set interrupt flags.
Fix this by using a pure assign instead of bitwise-or for the
first line (note: intr_txqs is only evaluated in case ATH9K_INT_TX bit
is set).
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20211116220720.30145-1-ps.report@gmx.net
Wakeup mhi is needed before pci_read/write only for QCA6390 and WCN6855. Since
wakeup & release mhi is enabled for all hardwares, below mhi assert is seen in
QCN9074 when doing 'rmmod ath11k_pci':
Kernel panic - not syncing: dev_wake != 0
CPU: 2 PID: 13535 Comm: procd Not tainted 4.4.60 #1
Hardware name: Generic DT based system
[<80316dac>] (unwind_backtrace) from [<80313700>] (show_stack+0x10/0x14)
[<80313700>] (show_stack) from [<805135dc>] (dump_stack+0x7c/0x9c)
[<805135dc>] (dump_stack) from [<8032136c>] (panic+0x84/0x1f8)
[<8032136c>] (panic) from [<80549b24>] (mhi_pm_disable_transition+0x3b8/0x5b8)
[<80549b24>] (mhi_pm_disable_transition) from [<80549ddc>] (mhi_power_down+0xb8/0x100)
[<80549ddc>] (mhi_power_down) from [<7f5242b0>] (ath11k_mhi_op_status_cb+0x284/0x3ac [ath11k_pci])
[E][__mhi_device_get_sync] Did not enter M0 state, cur_state:RESET pm_state:SHUTDOWN Process
[E][__mhi_device_get_sync] Did not enter M0 state, cur_state:RESET pm_state:SHUTDOWN Process
[E][__mhi_device_get_sync] Did not enter M0 state, cur_state:RESET pm_state:SHUTDOWN Process
[<7f5242b0>] (ath11k_mhi_op_status_cb [ath11k_pci]) from [<7f524878>] (ath11k_mhi_stop+0x10/0x20 [ath11k_pci])
[<7f524878>] (ath11k_mhi_stop [ath11k_pci]) from [<7f525b94>] (ath11k_pci_power_down+0x54/0x90 [ath11k_pci])
[<7f525b94>] (ath11k_pci_power_down [ath11k_pci]) from [<8056b2a8>] (pci_device_shutdown+0x30/0x44)
[<8056b2a8>] (pci_device_shutdown) from [<805cfa0c>] (device_shutdown+0x124/0x174)
[<805cfa0c>] (device_shutdown) from [<8033aaa4>] (kernel_restart+0xc/0x50)
[<8033aaa4>] (kernel_restart) from [<8033ada8>] (SyS_reboot+0x178/0x1ec)
[<8033ada8>] (SyS_reboot) from [<80301b80>] (ret_fast_syscall+0x0/0x34)
Hence, disable wakeup/release mhi using hw_param for other hardwares.
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.5.0.1-01060-QCAHKSWPL_SILICONZ-1
Fixes: a05bd85133 ("ath11k: read and write registers below unwindowed address")
Signed-off-by: Seevalamuthu Mariappan <quic_seevalam@quicinc.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1636702019-26142-1-git-send-email-quic_seevalam@quicinc.com
This patch introduces a NAPI-style approach for processing completed
Rx URBs that contributes to improving driver throughput and reducing
CPU load.
Packets in completed URBs are copied to NAPI SKBs and passed to the
network stack for processing. Each frame passed to the stack is one
work item in the NAPI budget.
If the NAPI budget is consumed and frames remain, they are added to
an overflow queue that is processed at the start of the next NAPI
polling cycle.
The NAPI handler is also responsible for copying pending Tx data to
Tx URBs and submitting them to the USB host controller for
transmission.
Signed-off-by: John Efstathiades <john.efstathiades@pebblebay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove hardware-specific header length adjustment as it is no longer
required. It also breaks generic receive offload (GRO) processing of
received TCP frames that results in a TCP ACK being sent for each
received frame.
Signed-off-by: John Efstathiades <john.efstathiades@pebblebay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move position of rx_submit() to remove forward declaration of
rx_complete() which is now no longer required.
Signed-off-by: John Efstathiades <john.efstathiades@pebblebay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces a new approach to allocating and managing
Rx URBs that contributes to improving driver throughput and reducing
CPU load.
A pool of Rx URBs is created during driver instantiation. All the
URBs are initially submitted to the USB host controller for
processing.
The default URB buffer size is different for each USB bus speed.
The chosen sizes provide good USB utilisation with little impact on
overall packet latency.
Completed URBs are processed in the driver bottom half. The URB
buffer contents are copied to a dynamically allocated SKB, which is
then passed to the network stack. The URB is then re-submitted to
the USB host controller.
NOTE: the call to skb_copy() in rx_process() that copies the URB
contents to a new SKB is a temporary change to make this patch work
in its own right. This call will be removed when the NAPI processing
is introduced by patch 6 in this patch set.
Signed-off-by: John Efstathiades <john.efstathiades@pebblebay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces a new approach to allocating and managing
Tx URBs that contributes to improving driver throughput and reducing
CPU load.
A pool of Tx URBs is created during driver instantiation. A URB is
allocated from the pool when there is data to transmit. The URB is
released back to the pool when the data has been transmitted by the
device.
The default URB buffer size is different for each USB bus speed.
The chosen sizes provide good USB utilisation with little impact on
overall packet latency.
SKBs to be transmitted are added to a pending queue for processing.
The driver tracks the available Tx URB buffer space and copies as
much pending data as possible into each free URB. Each full URB
is then submitted to the USB host controller for transmission.
Signed-off-by: John Efstathiades <john.efstathiades@pebblebay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix memory allocation that fails to check for NULL return.
Signed-off-by: John Efstathiades <john.efstathiades@pebblebay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
PSFP rules take effect on the streams from any port of VSC9959 switch.
This patch use ingress port to limit the rule only active on this port.
Each stream can only match two ingress source ports in VSC9959. Streams
from lowest port gets the configuration of SFID pointed by MAC Table
lookup and streams from highest port gets the configuration of (SFID+1)
pointed by MAC Table lookup. This patch defines the PSFP rule on highest
port as dummy rule, which means that it does not modify the MAC table.
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch add police action to set flow meter table which is defined
in IEEE802.1Qci. Flow metering is two rates two buckets and three color
marker to policing the frames, we only enable one rate one bucket in
this patch.
Flow metering shares a same policer pool with VCAP policers, so the PSFP
policer calls ocelot_vcap_policer_add() and ocelot_vcap_policer_del() to
set flow meter police.
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Policer was previously automatically assigned from the highest index to
the lowest index from policer pool. But police action of tc flower now
uses index to set an police entry. This patch uses the police index to
set vcap policers, so that one policer can be shared by multiple rules.
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds stream gate settings for PSFP. Use SGI table to store
stream gate entries. Disable the gate entry when it is not used by any
stream.
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
VSC9959 supports Per-Stream Filtering and Policing(PSFP) that complies
with the IEEE 802.1Qci standard. The stream is identified by Null stream
identification(DMAC and VLAN ID) defined in IEEE802.1CB.
For PSFP, four tables need to be set up: stream table, stream filter
table, stream gate table, and flow meter table. Identify the stream by
parsing the tc flower keys and add it to the stream table. The stream
filter table is automatically maintained, and its index is determined by
SGID(flow gate index) and FMID(flow meter index).
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
PSFP support gate and police action. This patch add the gate and police
action to flower parse action, check chain ID to determine which block
to offload. Adding psfp callback functions to add, delete and update gate
and police in PSFP table if hardware supports it.
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some chips in the ocelot series such as VSC9959 support Per-Stream
Filtering and Policing(PSFP), which is processing after VCAP blocks.
We set this block on chain 30000 and set vcap IS2 chain to goto PSFP
chain if hardware support.
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ocelot_mact_learn_streamdata() can be used in VSC9959 to overwrite an
FDB entry with stream data. The stream data includes SFID and SSID which
can be used for PSFP and FRER set.
ocelot_mact_lookup() can be used to check if the given {DMAC, VID} FDB
entry is exist, and also can retrieve the DEST_IDX and entry type for
the FDB entry.
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The definition of macro MOTO_SROM_BUG is:
#define MOTO_SROM_BUG (lp->active == 8 && (get_unaligned_le32(
dev->dev_addr) & 0x00ffffff) == 0x3e0008)
and the if statement
if (MOTO_SROM_BUG) lp->active = 0;
using this macro indicates lp->active could be 8. If lp->active is 8 and
the second comparison of this macro is false. lp->active will remain 8 in:
lp->phy[lp->active].gep = (*p ? p : NULL); p += (2 * (*p) + 1);
lp->phy[lp->active].rst = (*p ? p : NULL); p += (2 * (*p) + 1);
lp->phy[lp->active].mc = get_unaligned_le16(p); p += 2;
lp->phy[lp->active].ana = get_unaligned_le16(p); p += 2;
lp->phy[lp->active].fdx = get_unaligned_le16(p); p += 2;
lp->phy[lp->active].ttm = get_unaligned_le16(p); p += 2;
lp->phy[lp->active].mci = *p;
However, the length of array lp->phy is 8, so array overflows can occur.
To fix these possible array overflows, we first check lp->active and then
return -EINVAL if it is greater or equal to ARRAY_SIZE(lp->phy) (i.e. 8).
Reported-by: TOTE Robot <oslab@tsinghua.edu.cn>
Signed-off-by: Teng Qi <starmiku1207184332@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
In line 5001, if all id in the array 'lp->phy[8]' is not 0, when the
'for' end, the 'k' is 8.
At this time, the array 'lp->phy[8]' may be out of bound.
Signed-off-by: zhangyue <zhangyue1@kylinos.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tony Nguyen says:
====================
10GbE Intel Wired LAN Driver Updates 2021-11-17
Radoslaw Tyl says:
The change is a consequence of errors reported by the ixgbevf driver
while starting several virtual guests at the same time on ESX host.
During this, VF was not able to communicate correctly with the PF,
as a result reported "PF still in reset state. Is the PF interface up?"
and then goes to locked state. The only thing left was to reload
the VF driver on the guest OS.
The background of the problem is that the current PFU and VFU
semaphore locking mechanism between sender and receiver may cause
overriding Mailbox memory (VFMBMEM), in such scenario receiver of
the original message will read the invalid, corrupted or one (or more)
message may be lost.
This change is actually as a support for communication with PF ESX
driver and does not contains changes and support for ixgbe driver.
For maintain backward compatibility, previous communication method
has been preserved in the form of LEGACY functions.
In the future there is a plan to add a support for a 1.5 mailbox API
communication also to ixgbe driver.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2021-11-17
This series contains updates to i40e driver only.
Eryk adds accounting for VLAN header in packet size when VF port VLAN is
configured. He also fixes TC queue distribution when the user has changed
queue counts as well as for configuration of VF ADQ which caused dropped
packets.
Michal adds tracking for when a VSI is being released to prevent null
pointer dereference when managing filters.
Karen ensures PF successfully initiates VF requested reset which could
cause a call trace otherwise.
Jedrzej moves validation of channel queue value earlier to prevent
partial configuration when the value is invalid.
Grzegorz corrects the reported error when adding filter fails.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Killing the kernel because a certain MDIO bus object is not in the
desired state at various points in the registration or unregistration
paths is excessive and is not helping in troubleshooting or fixing
issues. Replace the BUG_ON() with WARN() and print out the MDIO bus name
to facilitate debugging.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
As reported in [1], e100 was no longer working for suspend/resume
cycles. The previous commit mentioned in the fixes appears to have
broken things and this attempts to practice best known methods for
device power management and keep wake-up working while allowing
suspend/resume to work. To do this, I reorder a little bit of code
and fix the resume path to make sure the device is enabled.
[1] https://bugzilla.kernel.org/show_bug.cgi?id=214933
Fixes: 69a74aef8a ("e100: use generic power management")
Cc: Vaibhav Gupta <vaibhavgupta40@gmail.com>
Reported-by: Alexey Kuznetsov <axet@me.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Alexey Kuznetsov <axet@me.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
DPAA2 has no special behaviour in its validation implementation, so can
be switched to phylink_generic_validate().
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
As phylink checks the interface mode against the supported_interfaces
bitmap, we no longer need to validate the interface mode, nor handle
PHY_INTERFACE_MODE_NA in the validation function. Remove these to
simplify the implementation.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Populate the phy interface mode bitmap for the Freescale DPAA2 driver
with interfaces modes supported by the MAC.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
ag71xx apparently only supports MII port type, which makes it different
from other implementations. However, Oleksij says there is no special
reason for this.
Convert the driver to use phylink_generic_validate(), which will allow
all ethtool port linkmodes instead of only MII, giving the driver
consistent behaviour with other drivers.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
As phylink checks the interface mode against the supported_interfaces
bitmap, we no longer need to validate the interface mode, nor handle
PHY_INTERFACE_MODE_NA in the validation function. Remove these to
simplify the implementation.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>