IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
If we have a per CE (Copy Engine) IRQ then we have no summary
register. Right now the code generates a summary register by
iterating over all copy engines and seeing if they have an interrupt
pending.
This has a problem. Specifically if _none_ if the Copy Engines have
an interrupt pending then they might go into low power mode and
reading from their address space will cause a full system crash. This
was seen to happen when two interrupts went off at nearly the same
time. Both were handled by a single call of ath10k_snoc_napi_poll()
but, because there were two interrupts handled and thus two calls to
napi_schedule() there was still a second call to
ath10k_snoc_napi_poll() which ran with no interrupts pending.
Instead of iterating over all the copy engines, let's just keep track
of the IRQs that fire. Then we can effectively generate our own
summary without ever needing to read the Copy Engines.
Tested-on: WCN3990 SNOC WLAN.HL.3.2.2-00490-QCAHLSWMTPL-1
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Rakesh Pillai <pillair@codeaurora.org>
Reviewed-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20200709082024.v2.1.I4d2f85ffa06f38532631e864a3125691ef5ffe06@changeid
Currently the NAPI processing loops through all
the copy engines and processes a particular copy
engine is the copy completion is set for that copy
engine. The host driver is not supposed to access
any copy engine register after clearing the interrupt
status register.
This might result in kernel crash like the one below
[ 1159.220143] Call trace:
[ 1159.220170] ath10k_snoc_read32+0x20/0x40 [ath10k_snoc]
[ 1159.220193] ath10k_ce_per_engine_service_any+0x78/0x130 [ath10k_core]
[ 1159.220203] ath10k_snoc_napi_poll+0x38/0x8c [ath10k_snoc]
[ 1159.220270] net_rx_action+0x100/0x3b0
[ 1159.220312] __do_softirq+0x164/0x30c
[ 1159.220345] run_ksoftirqd+0x2c/0x64
[ 1159.220380] smpboot_thread_fn+0x1b0/0x288
[ 1159.220405] kthread+0x11c/0x12c
[ 1159.220423] ret_from_fork+0x10/0x18
To avoid such a scenario, we generate an interrupt
summary by reading the copy completion for all the
copy engine before actually processing any of them.
This will avoid reading the interrupt status register
for any CE after the interrupt status is cleared.
Tested-on: WCN3990 hw1.0 SNOC WLAN.HL.3.1-01040-QCAHLSWMTPLZ-1
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1593193967-29897-1-git-send-email-pillair@codeaurora.org
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
sizeof(flexible-array-member) triggers a warning because flexible array
members have incomplete type[1]. There are some instances of code in
which the sizeof operator is being incorrectly/erroneously applied to
zero-length arrays and the result is zero. Such instances may be hiding
some bugs. So, this work (flexible-array member conversions) will also
help to get completely rid of those sorts of issues.
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20200507041127.GA31587@embeddedor
As the exisiting CE configurations are defined in global, there
are the chances of QCA99X0 family chipsets CE configurations
are getting changed by the ath10k_pci_override_ce_config()
function.
The override will be hit and CE5 configurations will be changed,
when the user bring up the QCA99X0 chipsets along with QCA6174
or QCA9377 chipset. (Bring up QCA99X0 family chipsets after
QCA6174 or QCA9377).
Hence, fixing this issue by moving the global CE configuration
to radio specific CE configuration.
Tested hardware: QCA9888 & QCA6174
Tested firmware: 10.4-3.10-00047 & WLAN.RM.4.4.1.c3-00058
Signed-off-by: Maharaja Kennadyrajan <mkenna@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1587649759-14381-1-git-send-email-mkenna@codeaurora.org
When the driver receives the tx completion of the
descriptor over ce, it clears the nbytes configured
for that particular descriptor. WCN3990 uses ce
descriptors with 64-bit address.
Currently during handling the tx completion of the
descriptors, the nbytes are accessed from the descriptors
using ce_desc for 32-bit targets. This will lead to clearing
of memory at incorrect offset if DMA MASK is set to greater
than 32 bits.
Attach different ce tx copy completed handler for targets
using address above 32-bit address.
Tested HW: WCN3990
Tested FW: WLAN.HL.2.0-01387-QCAHLSWMTPLZ-1
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Use SPDX identifiers everywhere in ath10k.
Makefile was incorrectly marked in commit b24413180f56 ("License cleanup: add
SPDX GPL-2.0 license identifier to files with no license"), fix that as well.
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
WCN3990 supports shadow registers write operation support
for copy engine for regular operation in powersave mode.
Since WCN3990 is a 64-bit target, the shadow register
implementation needs to be done in the copy engine handlers
for 64-bit target. Currently the shadow register implementation
is present in the 32-bit target handlers of copy engine.
Fix the shadow register copy engine write operation
implementation for 64-bit target(WCN3990).
Tested HW: WCN3990
Tested FW: WLAN.HL.2.0-01188-QCAHLSWMTPLZ-1
Fixes: b7ba83f7c414 ("ath10k: add support for shadow register for WNC3990")
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
WCN3990 is a 37-bit target but can address memory range
only upto 35 bits. The 36th bit is used to control the
smmu/iommu translation and the 37th bit is used by the
internal bus masters to access the wifi subsystem internal
SRAM. With the DMA mask set to 37i-bit, the host driver
can get 37-bit dma address, which leads to incorrect
address access in the target.
Hence the host driver can used addresses upto 35-bit
for WCN3990. Fix the dma mask for wcn3990 to 35-bit,
instead of 37-bit.
Tested HW: WCN3990
Tested FW: WLAN.HL.2.0-01188-QCAHLSWMTPLZ-1
Tested-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Existing copy engine interrupt enable logic assumes that last
CE is using polling mode and due to this interrupt for last copy engine
are always disabled. WCN3990 uses last CE for pktlog and
interrupt remains disabled with existing logic.
To mitigate this issue, introduce CE_ATTR_POLL flag and control
the interrupt based on the flag which can be set in ce_attr.
Testing:
Tested on WCN3990 and QCA6174 HW.
Tested FW: WLAN.HL.2.0-01192-QCAHLSWMTPLZ-1,
WLAN.RM.4.4.1-00109-QCARMSWPZ-1
Signed-off-by: Govind Singh <govinds@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
It's easier to violate abstractions and introduce bugs when snoc.h is
including pci.h. Let's not do that.
I'm not extremely familiar with this driver yet, but several of the
shared PCI/SNOC bits seem to be related to the Copy Engine, so move them
to ce.h.
Signed-off-by: Brian Norris <briannorris@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
SRRI/DRRI are not mapped in the HW Shadow block and can lead
to un-clocked access if common subsystem in the target is
powered down due to idle mode.
To mitigate this problem SRRI/DRRI can be read from
DDR instead of doing an actual hardware read.
Host allocates non cached memory on ddr and configures
the physical address of this memory to the CE hardware.
The hardware updates the RRI on this particular location.
Read SRRI/DRRI from DDR location instead of
direct target read.
Enable retention restore on ddr using hw params to enable
in specific targets.
Signed-off-by: Govind Singh <govinds@codeaurora.org>
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
WCN3990 needs shadow register write operation support
for copy engine for regular operation in powersave mode.
Add support for copy engine shadow register write in
datapath tx for WCN3990
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
WCN3990 has interrupts per CE and the interrupt summary
is not retained after the interrupt handler has finished
execution. We need to check if we received any
ce in rx and tx completion path.
Generate a interrupt summary with all CE interrupts if
the target does not retain interrupt summary after the
execution of interrupt handler.
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Signed-off-by: Govind Singh <govinds@codeaurora.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Checkpatch found these issues:
drivers/net/wireless/ath/ath10k/ce.h:324: Please use a blank line after function/struct/union/enum declarations
drivers/net/wireless/ath/ath10k/core.c:1321: Please don't use multiple blank lines
drivers/net/wireless/ath/ath10k/htt.h:1859: Please use a blank line after function/struct/union/enum declarations
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
WCN3990 CE descriptor uses 64bit address for
src/dst ring buffer. It has extended field for toeplitz
hash result, which is being used for HW assisted
hash results.
To accommodate WCN3990 descriptor, define new CE
descriptor for extended addressing mode and related
methods to handle the descriptor data.
Signed-off-by: Govind Singh <govinds@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
CE send and receive API's are using u32 ring address, which
truncates the address for target with 64bit addressing range.
Use dma_addr_t for ce buffers to support target with extended
addressing range.
Signed-off-by: Govind Singh <govinds@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Remove bus specific dependencies from CE layer
to have common CE layer across multiple targets.
This is required for adding support for WCN3990
chipset support as WCN3990 chipset uses SNOC
bus interface with Copy Engine endpoint.
Signed-off-by: Govind Singh <govinds@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Define structures for the copy engine ctrl/misc registers,
that includes CE CMD halt, watermark source, watermark destination,
host IE ring, source, destination and dmax ring.
This adds support to avoid the conditional compilation,
code optimization and dynamic configuration of the copy engine
register map for respective hardware bus interface.
Signed-off-by: Sarada Prasanna Garnayak <c_sgarna@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Fix checkpatch warnings about use of spaces with operators:
spaces preferred around that '*' (ctx:VxV)
This has been recently added to checkpatch.
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Whenever htt rx indication i.e target to host messages are received
on rx copy engine (CE5), the message will be freed after processing
the response. Then CE 5 will be refilled with new descriptors at
post rx processing. This memory alloc and free operations can be avoided
by reusing the same descriptors.
During CE pipe allocation, full ring is not initialized i.e n-1 entries
are filled up. So for CE 5 full ring should be filled up to reuse
descriptors. Moreover CE 5 write index will be updated in single shot
instead of incremental access. This could avoid multiple pci_write and
ce_ring access. From experiments, It improves CPU usage by ~3% in IPQ4019
platform.
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
The physical address necessary to unmap DMA ('bufferp') is stored
in ath10k_skb_cb as 'paddr'. For diag register read and write
operations, 'paddr' is stored in transfer context. ath10k doesn't rely
on the meta/transfer_id. So the unused output arguments {bufferp, nbytesp
and transfer_idp} are removed from CE recv_next completion.
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
QCA4019 can queue up to 2500 frames at a time.
This means it requires roughly 5000 entires on the
ring to work properly. Otherwise random tx failure
may occur.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
For the messages from host to target, shadow copy of CE descriptors
are maintained in source ring. Before writing actual CE descriptor,
first shadow copy is filled and then it is copied to CE address space.
To optimize in download path and to reduce d-cache pressure, removing
shadow copy of CE descriptors. This will also reduce driver memory
consumption by 33KB during on device probing.
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
The physical address necessary to unmap DMA ('bufferp') is stored
in ath10k_skb_cb as 'paddr'. ath10k doesn't rely on the meta/transfer_id
when handling send completion (htc ep id is stored in sk_buff control
buffer). So the unused output arguments {bufferp, nbytesp and transfer_idp}
are removed from CE send completion. This change is needed before removing
the shadow copy of copy engine (CE) descriptors in follow up patch.
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Register receive callbacks for every copy engines (CE) separately
instead of having common receive handler. Some of the copy engines
receives different type of messages (i.e HTT/HTC/pktlog) from target.
Hence to service them accordingly, register per copy engine receive
callbacks.
Reviewed-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Register send completion callbacks for every copy engines (CE) separately
instead of having common completion handler. Since some of the copy
engines delivers different type of messages, per-CE callbacks help to
service them differently.
Reviewed-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Rajkumar Manoharan <rmanohar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
QCA99X0 uses two new copy engine src desc flags for interrupt
indication. Bit_2 is to mark if host interrupt is disabled after
processing the current desc and bit_3 is to mark if target interrupt
is diabled after the processing of current descriptor.
CE_DESC_FLAGS_META_DATA_MASK and CE_DESC_FLAGS_META_DATA_LSB are based
on the target type.
Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
QCA99X0 supports upto 12 Copy engines. Host and target
CE configuration table is updated to support new copy engine
pipes. This also fixes the assumption of diagnostic CE by making
CE_7 as the one instead of CE_COUNT - 1.
Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Add more register address and mask which can be different
for newer chip to hw_reg table.
Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
The value was off by one. The error probably has no negative
affect on any upstream firmware, but should be fixed anyway
in case it comes into use in the future.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
The QCA6174 in combination with new wmi-tlv firmware is capable of
multi-channel, beamforming, tdls and other features.
This patch just makes it possible to boot these devices and do some basic stuff
like connect to an AP without encryption. Some things may not work or may be
unreliable. New features will be implemented later. This will be addressed
eventually with future patches.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Calling init to reinit ce pipe state would also
re-set all static structure links and setting
(which don't change over driver lifecycle).
Make it so alloc links structures and initializes
static data and init part to setup state
variables and clear stuff.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
diag_read() is used for reading from firmware memory via the diagnose window.
First user will be cal_data debugfs file.
To serialise diagnostic window access and make it safe to use while firmware is
running take ce_lock both in ath10k_pci_diag_write_mem() and
ath10k_pci_diag_read_mem(). Because of that all the CE calls had to be changed
to _nolock variants.
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
It was possible on a host system running low on
memory to end up with no rx buffers on pci pipes.
This makes the driver more robust as it won't fail
to start if it can't allocate all rx buffers right
away. If it is fatal then upper layers will notice
trouble anyway.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
It doesn't make much sense to overwrite send_cb
and recv_cb callbacks over and over again whenever
transport starts. Just make sure to unmask copy
engine interrupts when starting.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
This prevents leaving incomplete scatter-gather
transfer on CE rings which can lead firmware to
crash.
Reported-By: Avery Pennarun <apenwarr@gmail.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Definitions by which copy engine structure are
allocated do not change so it doesn't make much
sense to re-create those structures each time
device is booted (e.g. due to firmware recovery).
This should decrease chance of memory allocation
failures.
While at it remove per_transfer_context pointer
indirection. The array has been trailing the copy
engine ringbuffer structure anyway. This also
saves pointer size worth of bytes for each copy
engine ringbuffer.
Reported-By: Avery Pennarun <apenwarr@gmail.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Going through full htc tx path for htt tx is a
waste of resources. By skipping it it's possible
to easily submit scatter-gather to the pci hif for
reduced host cpu load and improved performance.
The new approach uses dma pool to store the
following metadata for each tx request:
* msdu fragment list
* htc header
* htt tx command
The htt tx command contains a msdu prefetch.
Instead of copying it original mapped msdu address
is used to submit a second scatter-gather item to
hif to make a complete htt tx command.
The htt tx command itself hands over dma mapped
pointers to msdus and completion of the command
itself doesn't mean the frame has been sent and
can be unmapped/freed. This is why htc tx
completion is skipped for htt tx as all tx related
resources are freed upon htt tx completion
indication event (which also implicitly means htt
tx command itself was completed).
Since now each htt tx request effectively consists
of 2 copy engine items CE_HTT_H2T_MSG_SRC_NENTRIES
is updated to allow maximum of
TARGET_10X_NUM_MSDU_DESC msdus being queued. This
keeps the tx path resource management simple.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
PCI is capable of handling scatter-gather lists.
This can be used to avoid copying memory.
Change the name of the callback while at to
reflect its purpose.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
It's impossible to rely on disable_irq() and/or CE
interrupt masking with legacy shared interrupts.
Other devices sharing the same irq line may assert
it while ath10k is doing something that requires
no interrupts.
Irq handlers are now registered after all
preparations are complete so spurious/foreign
interrupts won't do any harm. The handlers are
unregistered when no interrupts are required (i.e.
during driver teardown).
This also removes the ability to receive FW early
indication (since interrupts are not registered
until early boot is complete). This is not mission
critical (it's more of a hint that early boot
failed due to unexpected FW crash) and will be
re-added in a follow up patch.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
This patch moves irq registering after necessary
structures have been allocated and initialized.
This should prevent interrupts from causing
tasklet access invalid memory pointers.
Reported-By: Ben Greear <greearb@candelatech.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
This shouldn't be silenced. This will be necessary
for PCI init code reordering.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
The value provided by num_sends_allowed is now
derived from CE source ringbuffer state.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
It is completely pointless to keep this function
around. It doesn't do anything different than
ce_send except it introduces more overhead.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
struct ce_sendlist is useless as we always add just one buffer onto it.
And most importantly, it's ugly as it doesn't use skb properly.
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
The double while() could end up running forever.
Inner while() would complete very fast. However
the completion processing could take enough time
for more completions to flow in. In that case the
outer while() would not terminate and run again,
and again. This could happen especially on a slow
host system.
This could lead to a system freeze during heavy
traffic. Note: this doesn't solve all known
starvation issues yet.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
