Some printks in here don't have newlines at the end, meaning the log
will be sort of hard to read. Add newlines.
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
The management packets, send to firmware via WMI, are
mapped using the direction DMA_TO_DEVICE. Currently in
case of wmi cleanup, these buffers are being unmapped
using an incorrect DMA direction. This can cause unwanted
behavior when the host driver is handling a restart
of the wlan firmware.
We might see a trace like below
[<ffffff8008098b18>] __dma_inv_area+0x28/0x58
[<ffffff8001176734>] ath10k_wmi_mgmt_tx_clean_up_pending+0x60/0xb0 [ath10k_core]
[<ffffff80088c7c50>] idr_for_each+0x78/0xe4
[<ffffff80011766a4>] ath10k_wmi_detach+0x4c/0x7c [ath10k_core]
[<ffffff8001163d7c>] ath10k_core_stop+0x58/0x68 [ath10k_core]
[<ffffff800114fb74>] ath10k_halt+0xec/0x13c [ath10k_core]
[<ffffff8001165110>] ath10k_core_restart+0x11c/0x1a8 [ath10k_core]
[<ffffff80080c36bc>] process_one_work+0x16c/0x31c
Fix the incorrect DMA direction during the wmi
management tx buffer cleanup.
Tested HW: WCN3990
Tested FW: WLAN.HL.3.1-00784-QCAHLSWMTPLZ-1
Fixes: dc405152bb ("ath10k: handle mgmt tx completion event")
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
There are two error paths where "ret" wasn't set. Also one error path
we set the error code to -EINVAL but we should just preserve the error
code from ath11k_hal_srng_get_entrysize(). That function only returns
-EINVAL so this doesn't change anything.
I removed the "ret = 0;" initializers so that hopefully GCC will be able
to detect these sorts of bugs in the future.
Fixes: d5c65159f2 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Add IEEE80211_RADIOTAP_HE_DATA1_BW_RU_ALLOC_KNOWN to the list of known
fields. Not doing so will result in wireshark not calculating the
bitrate correctly.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
In the unlikely event that cap->supported_bands has neither
WMI_HOST_WLAN_2G_CAP set or WMI_HOST_WLAN_5G_CAP set then pointer
band is null and a null dereference occurs when assigning
band->n_iftype_data. Move the assignment to the if blocks to
avoid this. Cleans up static analysis warnings.
Addresses-Coverity: ("Explicit null dereference")
Fixes: 9f056ed8ee ("ath11k: add HE support")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
This adds missing rx rate info stats like pilot evm,
per chain rssi, per user ul ppdu and mpdu counts and
ul ofdma rate info etc.
And add null checks for memory alloc failures.
Signed-off-by: Bhagavathi Perumal S <bperumal@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Currently the structure ts is not inititalized and ts.flags contains
garbage values from the stack. This is being passed into function
ath11k_dp_tx_status_parse that bit-wise or'ing in settings into the
ts.flags field. To avoid flags (and other fields) from containing
garbage, initialize the structure to zero before use.
Addresses-Coverity: ("Uninitialized scalar variable)"
Fixes: d5c65159f2 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
The channel info parameters are required by the firmware to process TxBf
parameters. Currently TxBf is passed prior to the channel info. This patch
moves TxBf setup after the channel setup.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Bhagavathi Perumal S <bperumal@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
This patch enables HE tlvs in ppdu stats for pktlog lite mode.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
When CONFIG_ATH11K_DEBUGFS is disabled, but CONFIG_MAC80211_DEBUGFS
is turned on, the driver fails to build:
drivers/net/wireless/ath/ath11k/debugfs_sta.c: In function 'ath11k_dbg_sta_open_htt_peer_stats':
drivers/net/wireless/ath/ath11k/debugfs_sta.c:416:4: error: 'struct ath11k' has no member named 'debug'
ar->debug.htt_stats.stats_req = stats_req;
^~
It appears that just using the former symbol is sufficient here,
adding a Kconfig dependency takes care of the corner cases.
Fixes: d5c65159f2 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
For some targets ex: QCS404, SCM permissions for MSA region is
statically configured in TrustZone fw. Add SCM call disable option
for such targets to avoid duplicate permissions.
Testing: Tested on WCN3990 HW
Tested FW: WLAN.HL.3.1-01040-QCAHLSWMTPLZ-1
Signed-off-by: Govind Singh <govinds@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Add boolean context flag to disable SCM call for statically
mapped msa region.
Signed-off-by: Govind Singh <govinds@codeaurora.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
This reverts commit 76d164f582.
PCIe hung issue was observed on multiple platforms. The issue was reproduced
when DUT was configured as AP and associated with 50+ STAs.
For QCA9984/QCA9888, the DMA_BURST_SIZE register controls the AXI burst size
of the RD/WR access to the HOST MEM.
0 - No split , RAW read/write transfer size from MAC is put out on bus
as burst length
1 - Split at 256 byte boundary
2,3 - Reserved
With PCIe protocol analyzer, we can see DMA Read crossing 4KB boundary when
issue happened. It broke PCIe spec and caused PCIe stuck. So revert
the default value from 0 to 1.
Tested: IPQ8064 + QCA9984 with firmware 10.4-3.10-00047
QCS404 + QCA9984 with firmware 10.4-3.9.0.2--00044
Synaptics AS370 + QCA9888 with firmware 10.4-3.9.0.2--00040
Signed-off-by: Zhi Chen <zhichen@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
When simulate random transfer fail for sdio write and read, it crash
sometimes.
Test steps:
1. Add config and update kernel:
CONFIG_FAIL_MMC_REQUEST=y
CONFIG_FAULT_INJECTION=y
CONFIG_FAULT_INJECTION_DEBUG_FS=y
2. run simulate fail:
cd /sys/kernel/debug/mmc1/fail_mmc_request
echo 10 > probability
echo 10 > times # repeat until hitting issues
3. it crash, the act len of ath10k_htc_hdr is higher than allocate len, it cause panic:
[ 99.723482] skbuff: skb_over_panic: text:00000000caa0f780 len:57013 put:57013 head:000000004116f24a data:0000000019ecb4dc tail:0xdef5 end:0x640 dev:<NULL>
[ 99.737697] ------------[ cut here ]------------
[ 99.742327] kernel BUG at /mnt/host/source/src/third_party/kernel/v4.19/net/core/skbuff.c:104!
[ 99.750937] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP
[ 99.831154] Process kworker/0:2 (pid: 151, stack limit = 0x00000000728010bf)
[ 99.838200] CPU: 0 PID: 151 Comm: kworker/0:2 Tainted: G W 4.19.85 #48
[ 99.846022] Hardware name: MediaTek krane sku0 board (DT)
[ 99.851429] Workqueue: events sdio_irq_work
[ 99.855614] pstate: 60000005 (nZCv daif -PAN -UAO)
[ 99.860402] pc : skb_panic+0x64/0x68
[ 99.863974] lr : skb_panic+0x64/0x68
[ 99.867542] sp : ffffff8008833a90
[ 99.870850] x29: ffffff8008833ac0 x28: ffffffe52e337370
[ 99.876159] x27: ffffffe52e328a90 x26: 000000000000e0d0
[ 99.881469] x25: ffffffe52e336b60 x24: 000000000000deb5
[ 99.886779] x23: ffffffe52e340680 x22: ffffffe4efd47e00
[ 99.892088] x21: 000000000000deb5 x20: ffffffa516d85b4c
[ 99.897397] x19: ffffffa526928037 x18: 0000000000000000
[ 99.902706] x17: 000000000000003c x16: ffffffa5265b6c80
[ 99.908015] x15: 0000000000000006 x14: 3a76656420303436
[ 99.913325] x13: 0000000000029bf0 x12: 0000000000000000
[ 99.918634] x11: 0000000000000000 x10: 0000000000000000
[ 99.923943] x9 : a3b907e4b2783000 x8 : a3b907e4b2783000
[ 99.929253] x7 : 0000000000000000 x6 : ffffffa526f66d76
[ 99.934563] x5 : 0000000000000000 x4 : 0000000000000000
[ 99.939872] x3 : 000000000002a5ab x2 : ffffffe53feed918
[ 99.945182] x1 : ffffffe53fee4a08 x0 : 000000000000008e
[ 99.950491] Call trace:
[ 99.952937] skb_panic+0x64/0x68
[ 99.956165] skb_put+0x7c/0x84
[ 99.959224] ath10k_sdio_irq_handler+0x740/0xbb8 [ath10k_sdio]
[ 99.965055] process_sdio_pending_irqs+0x58/0x1a4
[ 99.969758] sdio_run_irqs+0x34/0x60
[ 99.973329] sdio_irq_work+0x1c/0x28
[ 99.974930] cros-ec-spi spi2.0: SPI transfer timed out
[ 99.976904] process_one_work+0x210/0x410
[ 99.976911] worker_thread+0x234/0x3dc
[ 99.976923] kthread+0x120/0x130
[ 99.982090] cros-ec-spi spi2.0: spi transfer failed: -110
[ 99.986054] ret_from_fork+0x10/0x18
[ 99.986063] Code: aa1403e2 2a1503e4 a90023e9 97e37d1a (d4210000)
[ 99.986068] ---[ end trace cb6d948c5a0fd6c7 ]---
[ 100.017250] Kernel panic - not syncing: Fatal exception
[ 100.018879] cros-ec-spi spi2.0: Command xfer error (err:-110)
[ 100.023659] SMP: stopping secondary CPUs
[ 100.023703] Kernel Offset: 0x251dc00000 from 0xffffff8008000000
[ 100.023707] CPU features: 0x0,2188200c
[ 100.023709] Memory Limit: none
The simulate fail of sdio is not a real sdio transter fail, it only
set an error status in mmc_should_fail_request after the transfer end,
actually the transfer is success, then sdio_io_rw_ext_helper will
return error status and stop transfer the left data. For example,
the really RX len is 286 bytes, then it will split to 2 blocks in
sdio_io_rw_ext_helper, one is 256 bytes, left is 30 bytes, if the
first 256 bytes get an error status by mmc_should_fail_request,then
the left 30 bytes will not read in this RX operation. Then when the
next RX arrive, the left 30 bytes will be considered as the header
of the read, the top 8 bytes will be considered as ath10k_htc_hdr,
but actually the 8 bytes is not the ath10k_htc_hdr, so the act_len
from this ath10k_htc_hdr is not correct, if it is a big value, such
as 57013, it will trigger skb_panic.
Drop the skb with invalid length will be reasonable.
This patch only effect sdio chips.
Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00029.
Signed-off-by: Wen Gong <wgong@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Fixes coccicheck warning:
drivers/net/wireless/ath/ath10k/htt_rx.c:2143:2-31: WARNING: Assignment of 0/1 to bool variable
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
The WiFi firmware found on sm8150 requires that the QDSS clock is
ticking in order to operate, so add an optional clock to the binding to
allow this to be specified in the sm8150 dts and add the clock to the
list of clocks in the driver.
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
The description of ath10k_pci_dump_memory_sram() is inaccurate, an error
can never be returned, it is always the length. Update the comment to
reflect.
Fixes: 219cc084c6 ("ath10k: add memory dump support QCA9984")
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
ath10k_pci_dump_memory_reg() will try to access memory of type
ATH10K_MEM_REGION_TYPE_IOREG however, if a hardware restart is in progress
this can crash a system.
Individual ioread32() time has been observed to jump from 15-20 ticks to >
80k ticks followed by a secure-watchdog bite and a system reset.
Work around this corner case by only issuing the read transaction when the
driver state is ATH10K_STATE_ON.
Tested-on: QCA9988 PCI 10.4-3.9.0.2-00044
Fixes: 219cc084c6 ("ath10k: add memory dump support QCA9984")
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Fixes coccicheck warning:
drivers/net/wireless/ath/wil6210/main.c:765:1-14: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/txrx.c:1143:1-19: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1516:4-23: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1523:4-23: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1538:4-30: WARNING: Assignment of 0/1 to bool variable
drivers/net/wireless/ath/wil6210/wmi.c:1545:4-30: WARNING: Assignment of 0/1 to bool variable
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Fixes coccicheck warning:
drivers/net/wireless/ath/ath9k/ar9003_aic.c:409:2-12: WARNING: Assignment of 0/1 to bool variable
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Use the print_hex_dump_debug() helper, instead of open-coding the same
operations.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Remove unused fields, copied from the Sun LANCE driver eons ago.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Linus Walleij says:
====================
IXP4xx networking cleanups
This is a patch series which jams together Arnds and mine
cleanups for the IXP4xx networking.
I also have patches for device tree support but that
requires more elaborate work, this series is some of
mine and some of Arnds patches that is a good foundation
for his multiplatform work and my device tree work.
These are for application to the networking tree so
that can be taken in one separate sweep.
I have tested the patches for a bit using zeroday builds
and some boots on misc IXP4xx devices and haven't run
into any major problems. We might find some new stuff
as a result from the new compiler coverage.
I had to depromote enabling compiler coverage at one
point in the v2 set because it depended on other patches
making the code more generic.
The change in v3 was simply dropping one offending
patch hardcoding base addresses into the driver.
The change in v4 drops a stable@ tag that was
unnecessary.
This v5 is a rebase of the v4 patch set on top of
net-next.
====================
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use the netdevice struct device .parent field when calling
dma_pool_create(): the .dma_coherent_mask and .dma_mask
pertains to the bus device on the hardware (platform)
bus in this case, not the struct device inside the network
device. This makes the pool allocation work.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In order to probe this ethernet interface from the device tree
all physical MMIO regions must be passed as resources. Begin
this rewrite by first passing the port base address as a
resource for all platforms using this driver, remap it in
the driver and avoid using any reference of the statically
mapped virtual address in the driver.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Simplify and correct a bunch of messages using printk
directly to use the netdev_* macros. I have not changed
all of them, just the low-hanging fruit.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use "ndev" for the struct net_device and "dev" for the
struct device in probe() and remove(). Add the local
"dev" pointer for later use in refactoring.
Take this opportunity to fix inverse christmas tree
coding style.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The IXP4xx driver was initializing the MDIO bus before even
probing, in the callbacks supposed to be used for setting up
the module itself, and with the side effect of trying to
register the MDIO bus as soon as this module was loaded or
compiled into the kernel whether the device was discovered
or not.
This does not work with multiplatform environments.
To get rid of this: set up the MDIO bus from the probe()
callback and remove it in the remove() callback. Rename
the probe() and remove() calls to reflect the most common
conventions.
Since there is a bit of checking for the ethernet feature
to be present in the MDIO registering function, making the
whole module not even be registered if we can't find an
MDIO bus, we need something similar: register the MDIO
bus when the corresponding ethernet is probed, and
return -EPROBE_DEFER on the other interfaces until this
happens. If no MDIO bus is present on any of the
registered interfaces we will eventually bail out.
None of the platforms I've seen has e.g. MDIO on EthB
and only uses EthC, there is always a Ethernet hardware
on the NPE (B, C) that has the MDIO bus, we just might
have to wait for it.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The platform data is needed to compile the driver as standalone,
so move it to a global location along with similar files.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The ixp46x ptp driver has a somewhat unusual setup, where the ptp
driver and the ethernet driver are in different directories but
access the same registers that are defined a platform specific
header file.
Moving everything into drivers/net/ makes it look more like most
other ptp drivers and allows compile-testing this driver on
other targets.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The ixp4xx_hss driver needs the platform data definition and the
system clock rate to be compiled. Move both into a new platform_data
header file.
This is a prerequisite for compile testing, but turning on compile
testing requires further patches to isolate the SoC headers.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Change the driver to use portable integer types to avoid
warnings during compile testing:
drivers/net/wan/ixp4xx_hss.c:863:21: error: cast to 'u32 *' (aka 'unsigned int *') from smaller integer type 'int' [-Werror,-Wint-to-pointer-cast]
memcpy_swab32(mem, (u32 *)((int)skb->data & ~3), bytes / 4);
^
drivers/net/wan/ixp4xx_hss.c:979:12: error: incompatible pointer types passing 'u32 *' (aka 'unsigned int *') to parameter of type 'dma_addr_t *' (aka 'unsigned long long *') [-Werror,-Wincompatible-pointer-types]
&port->desc_tab_phys)))
^~~~~~~~~~~~~~~~~~~~
include/linux/dmapool.h:27:20: note: passing argument to parameter 'handle' here
dma_addr_t *handle);
^
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
On modern hardware with a large number of cpus and using XDP,
the current MSIX limit is insufficient. Bump the limit in
order to allow more queues.
Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Reviewed-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Huazhong Tan says:
====================
net: hns3: add some misc update about reset issue
This series includes some misc update relating to reset issue.
[patch 1/7] & [patch 2/7] splits hclge_reset()/hclgevf_reset()
into two parts: preparing and rebuilding. Since the procedure
of FLR should be separated out from the reset task([patch 3/7 &
patch 3/7]), then the FLR's processing can reuse these codes.
pci_error_handlers.reset_prepare() is void type function, so
[patch 6/7] & [patch 7/7] factor some codes related to PF
function reset to make the preparing done before .reset_prepare()
return.
BTW, [patch 5/7] enlarges the waiting time of reset for matching
the hardware's.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
hclge_reset_prepare_down() is only used to inform VF that PF is
going to do function reset, then using hclge_func_reset_sync_vf()
in hclge_reset_prepare_wait() to query whether VF is ready before
asserting PF function reset. To make the code more readable,
this patch uses a new function hclge_function_reset_notify_vf()
to do this job.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When synchronizes with VFs fail before PF function reset,
PF driver should go on its function reset, otherwise it
can not run normally anymore. So, hclge_func_reset_sync_vf()
should not affect the processing of PF reset, this patch
modifies its return type to void.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the load of firmware is high, its reset task may takes
more time(which will be as long as 35 seconds). So this
patch modifies HCLGE_RESET_WAIT_CNT to match the firmware's.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the actual work of VF FLR is handled in the reset task,
which is asynchronous. So in some case, if the preparing and
rebuilding are not done, then the VF FLR will trigger some problems,
for example, makes hardware go into chaos.
So this patch separates the process of VF FLR from reset task, and
adds a semaphore to serialize this reset and others.
When FLR's preparing fails, if there has other higher level reset
pending or failing times less than the HCLGE_FLR_RETRY_CNT, this
preparing should be retried, otherwise it will get into a wrong state.
BTW, while the hardware reports misc interrupt during pcie_flr(),
the driver can not receive this interrupt anymore, so disable it
when hclgevf_flr_prepare() return, and re-enable it when enter
hclgevf_flr_done().
Avoid declaring internal function hclgevf_enable_vector(), this patch
also moves its definition forward, and removes unused enum
hnae3_flr_state.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the actual work of PF FLR is handled in the reset task,
which is asynchronous. So in some case, if the preparing and
rebuilding are not done, then the PF FLR will trigger some problems,
for example, makes hardware go into chaos.
So this patch separates the process of PF FLR from reset task, and
adds a semaphore to serialize this reset and others.
When FLR's preparing fails, if there has other higher level reset
pending or failing times less than the HCLGE_FLR_RETRY_CNT, this
preparing should be retried, otherwise PF and its VF may get into
wrong state.
BTW, while the hardware reports misc interrupt during pcie_flr(),
the driver can not receive this interrupt anymore, so disable it
when hclge_flr_prepare() return, and re-enable it when enter
hclge_flr_done().
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
hclgevf_reset() is a little bloated, and the process of VF FLR will
be separated from the reset task later. So this patch splits
hclgevf_reset() into hclgevf_reset_prepare() and hclge_reset_rebuild(),
then FLR can also reuse these two functions. Also moves HNAE3_UP_CLIENT
into hclgevf_reset_stack().
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
hclge_reset() is a little bloated, and the process of PF FLR will
be separated from the reset task later. So this patch splits
hclge_reset() into hclge_reset_prepare() and hclge_reset_rebuild(),
then FLR can also reuse these two functions.
BTW, since hclge_clear_reset_cause() and hclge_reset_prepare_up()
will not affect the device, so in hclge_reset_rebuild(), these
functions are called without rtnl_lock.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/net/ethernet/sfc/mcdi_functions.c: In function 'efx_mcdi_ev_init':
drivers/net/ethernet/sfc/mcdi_functions.c:79:28: warning:
variable 'nic_data' set but not used [-Wunused-but-set-variable]
commit 4438b587fe ("sfc: move MCDI event queue management code")
introduces this unused variable.
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove duplicated include.
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When running in XDP mode, pages come from the page pool, and should
be freed back to the same pool or specifically detached. Currently,
when the driver re-initializes, the page pool destruction is delayed
forever since it thinks there are oustanding pages.
Fixes: 322b87ca55 ("bnxt_en: add page_pool support")
Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Reviewed-by: Andy Gospodarek <gospo@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jacob Keller says:
====================
devlink documentation refactor
This series updates the devlink documentation, with a few primary goals
* move all of the devlink documentation into a dedicated subfolder
* convert that documentation to the reStructuredText format
* merge driver-specific documentations into a single file per driver
* add missing documentation, including per-driver and devlink generally
For each driver, I took the time to review the code and add further
documentation on the various features it currently supports. Additionally, I
added new documentation files for some of the features such as
devlink-dpipe, devlink-resource, and devlink-regions.
Note for the region snapshot triggering, I kept that as a separate patch as
that is based on work that has not yet been merged to net-next, and may
change.
I also improved the existing documentation for devlink-info and
devlink-param by adding a bit more of an introduction when converting it to
the rst format.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that devlink regions can be triggered via
DEVLINK_CMD_REGION_TRIGGER, document this in the devlink-region.rst
file.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Primarily based on the DPIPE netdev conference paper, introduce a new
file to document the dpipe interface.
This likely needs further improvement, but is at least a good overall
start.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Take the little bit of documentation for resources from various commit
messages and combine it into a new devlink-resource.rst file.
This could probably be expanded on even further by someone with more
knowledge of how the devlink resources work.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename the trap-specific netdevimsim.rst file, and expand it to include
documentation of all the devlink features currently implemented by the
netdevsim driver code.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The IONIC device driver allocates a devlink and reports versions. Add
documentation for this driver.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Cc: Shannon Nelson <snelson@pensando.io>
Acked-by: Shannon Nelson <snelson@pensandi.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
