Commit graph

1,058,063 commits

Author SHA1 Message Date
Mike Marciniszyn
b4b90a50cb IB/hfi1: Remove atomic completion count
The atomic is not needed.

Fixes: d99dc602e2 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132847.131370.54250.stgit@awfm-01.cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27 20:06:42 -03:00
Mike Marciniszyn
f5dc70a0e1 IB/hfi1: Tune netdev xmit cachelines
This patch moves fields in the ring and creates a line for the producer
and the consumer.

The adds a consumer side variable that tracks the ring avail so that the
code doesn't have the read the other cacheline to get a count for every
packet. A read now only occurs when the avail is at 0.

Fixes: d99dc602e2 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132842.131370.15636.stgit@awfm-01.cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27 20:06:42 -03:00
Mike Marciniszyn
a7125869b2 IB/hfi1: Get rid of tx priv backpointer
The txq has the backpointer, so this is a micro optimization for the tx
path.

Fixes: d99dc602e2 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132836.131370.89704.stgit@awfm-01.cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27 20:06:41 -03:00
Mike Marciniszyn
4bf0ca0c9f IB/hfi1: Get rid of hot path divide
The pointer math in this statemet does a divide;
	struct hfi1_ipoib_txq *txq = &priv->txqs[napi - priv->tx_napis];

Elminate the divide by embedding the struct napi_strut in the txq and
getting the txq with a container_of() using the newly embedded napi.

Fixes: d99dc602e2 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132831.131370.3993.stgit@awfm-01.cornelisnetworks.com
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27 20:06:41 -03:00
Mike Marciniszyn
d47dfc2b00 IB/hfi1: Remove cache and embed txreq in ring
This patch removes kmem cache allocation and deallocation in favor of
having the ipoib_txreq in the ring.

The consumer is now the packet sending side allocating tx descriptors from
ring and the producer is the napi interrupt handling freeing tx
descriptors.

The locks are now eliminated because the napi tx lock insures a single
consumer and the napi handling insures a single producer.

The napi poll is converted to memory poll looking for items that have been
marked completed.

Fixes: d99dc602e2 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20210913132826.131370.4397.stgit@awfm-01.cornelisnetworks.com
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27 20:06:41 -03:00
Sibi Sankar
0025fac17b arm64: dts: qcom: sc7280: Update Q6V5 MSS node
Update MSS node to support MSA based modem boot on SC7280 SoCs.

Signed-off-by: Sibi Sankar <sibis@codeaurora.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1631886935-14691-11-git-send-email-sibis@codeaurora.org
2021-09-27 17:48:59 -05:00
Sibi Sankar
4882cafb99 arm64: dts: qcom: sc7280: Add Q6V5 MSS node
This patch adds Q6V5 MSS PAS remoteproc node for SC7280 SoCs.

Signed-off-by: Sibi Sankar <sibis@codeaurora.org>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1631886935-14691-10-git-send-email-sibis@codeaurora.org
2021-09-27 17:48:59 -05:00
Sibi Sankar
dddf4b0621 arm64: dts: qcom: sc7280: Add nodes to boot modem
Add miscellaneous nodes to boot the modem and support post-mortem debug
on SC7280 SoCs.

Signed-off-by: Sibi Sankar <sibis@codeaurora.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1631886935-14691-9-git-send-email-sibis@codeaurora.org
2021-09-27 17:48:44 -05:00
Sibi Sankar
f831468901 arm64: dts: qcom: sc7280: Add/Delete/Update reserved memory nodes
Add, delete and update platform specific reserved memory nodes.

Signed-off-by: Sibi Sankar <sibis@codeaurora.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1631886935-14691-8-git-send-email-sibis@codeaurora.org
2021-09-27 17:48:44 -05:00
Sibi Sankar
eca7d3a366 arm64: dts: qcom: sc7280: Update reserved memory map
Add missing reserved regions as described in v1 of SC7280 memory map.

Signed-off-by: Sibi Sankar <sibis@codeaurora.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1631886935-14691-7-git-send-email-sibis@codeaurora.org
2021-09-27 17:48:33 -05:00
AngeloGioacchino Del Regno
cea8351135 arm64: dts: qcom: msm8998-fxtec-pro1: Add tlmm keyboard keys
This device has a physical matrix keyboard, connected to a GPIO
expander, for which there's still no support yet.
Though, some of the keys are connected to the MSM8998 GPIOs and not
as a matrix, so these can be added.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210909123823.368199-4-angelogioacchino.delregno@somainline.org
2021-09-27 17:47:57 -05:00
AngeloGioacchino Del Regno
f66ea51f0e arm64: dts: qcom: msm8998-fxtec-pro1: Add Goodix GT9286 touchscreen
This smartphone has a Goodix GT8296 touch IC, reachable at address
0x14 on blsp2 i2c-1.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210909123823.368199-3-angelogioacchino.delregno@somainline.org
2021-09-27 17:47:57 -05:00
AngeloGioacchino Del Regno
946c9a2cf8 arm64: dts: qcom: msm8998-fxtec-pro1: Add physical keyboard leds
Add configuration for the physical keyboard LEDs, including the
caps lock indicator and keyboard backlight.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210909123823.368199-2-angelogioacchino.delregno@somainline.org
2021-09-27 17:47:57 -05:00
AngeloGioacchino Del Regno
122d2c5f31 arm64: dts: qcom: Add support for MSM8998 F(x)tec Pro1 QX1000
Add device tree support for the F(x)tec Pro 1 (QX1000) smartphone;
this is a minimal configuration to boot to serial console.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210909123823.368199-1-angelogioacchino.delregno@somainline.org
2021-09-27 17:47:57 -05:00
Stephan Gerhold
8199a0b31e arm64: dts: qcom: msm8916: Fix Secondary MI2S bit clock
At the moment, playing audio on Secondary MI2S will just end up getting
stuck, without actually playing any audio. This happens because the wrong
bit clock is configured when playing audio on Secondary MI2S.

The PRI_I2S_CLK (better name: SPKR_I2S_CLK) is used by the SPKR audio mux
block that provides both Primary and Secondary MI2S.

The SEC_I2S_CLK (better name: MIC_I2S_CLK) is used by the MIC audio mux
block that provides Tertiary MI2S. Quaternary MI2S is also part of the
MIC audio mux but has its own clock (AUX_I2S_CLK).

This means that (quite confusingly) the SEC_I2S_CLK is not actually
used for Secondary MI2S as the name would suggest. Secondary MI2S
needs to have the same clock as Primary MI2S configured.

Fix the clock list for the lpass node in the device tree and add
a comment to clarify this confusing naming. With these changes,
audio can be played correctly on Secondary MI2S.

Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Fixes: 3761a3618f ("arm64: dts: qcom: add lpass node")
Tested-by: Vincent Knecht <vincent.knecht@mailoo.org>
Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210816181810.2242-1-stephan@gerhold.net
2021-09-27 17:32:38 -05:00
Stephan Gerhold
51c7786f5d arm64: dts: qcom: msm8916-longcheer-l8150: Add missing sensor interrupts
So far there were no interrupts set up for the BMC150 accelerometer
+ magnetometer combo because they were broken for some reason.
It turns out Longcheer L8150 actually has a BMC156 which is very similar
to BMC150, but only has an INT2 pin for the accelerometer part.

This requires some minor changes in the bmc150-accel driver which is now
supported by using the more correct bosch,bmc156_accel compatible.
Unfortunately it looks like even INT2 is not functional on most boards
because the interrupt line is not actually connected to the BMC156.
However, there are two pads next to the chip that can be shorted
to make it work if needed.

While at it, add the missing interrupts for the magnetometer part
and extra BMG160 gyroscope, those seem to work without any problems.
Also correct the magnetometer compatible to bosch,bmc156_magn for clarity
(no functional difference for the magnetometer part).

Tested-by: Nikita Travkin <nikita@trvn.ru>
Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210816123544.14027-1-stephan@gerhold.net
2021-09-27 17:31:55 -05:00
Sai Prakash Ranjan
ede638c42c arm64: dts: qcom: sc7180: Add IMEM and pil info regions
Add IMEM and pil info DT nodes for SC7180 SoC which will help in the
post-mortem debug.

Signed-off-by: Sai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>
[bjorn: Dropped dload-mode subnode, as no agreement was reached on this binding]
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/39064a2db95ccc2cb5eef003569bef2de651c8ed.1628757036.git.saiprakash.ranjan@codeaurora.org
2021-09-27 17:27:50 -05:00
Konrad Dybcio
a9a5ca5c8c arm64: dts: qcom: pm6150l: Add missing include
Add missing include to make it compile.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-17-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
ed1648d52a arm64: dts: qcom: sm6350: Add device tree for Sony Xperia 10 III
Add initial SM6350 SoC and Sony Xperia 10 III (PDX213, Lena platform) device
trees. There is no sign of another Lena devices on the horizon, so a common
DTSI is not created for now. 10 III features a Full HD OLED display and 5G
support, among other nice things like USB3.

The bootloader is VERY unpleasant, to get a bootable setup you have to run:

mkbootimg --kernel arch/arm64/boot/Image.gz --ramdisk [some initrd] \
--dtb arch/arm64/boot/dts/qcom/sm6350-sony-xperia-lena-pdx213.dtb \
--cmdline "[some cmdline]" --base 0 --kernel_offset 0x8000 \
--ramdisk_offset 0x1000000 --dtb_offset 0x1f00000 --os_version 11 \
--os_patch_level "2021-08" --tags_offset 0x100 --pagesize 4096 \
--header_version 2 -o mainline.img

adb reboot bootloader

// You have to either pull vbmeta{"","_system"} from
// /dev/block/bootdevice/by-name/ or build one as a part of AOSP build process
fastboot --disable-verity --disable-verification flash vbmeta vbmeta.img
fastboot --disable-verity --disable-verification flash vbmeta_system \
vbmeta_system.img

fastboot flash boot mainline.img
fastboot erase dtbo // This will take approx 70s...
fastboot reboot

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-16-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
4ef13f7fe4 arm64: dts: qcom: sm6350: Add apps_smmu and assign iommus prop to USB1
Add a node for the APPS SMMU to allow for managing memory access to peripherals
such as the USB controller.

While at it, add iommus property to the USB1 node to make sure its registers can
be accessed, as they seem to be gated by default.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-15-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
1797e1c9a9 arm64: dts: qcom: sm6350: Add SDHCI1/2 nodes
Add SDHCI1/2 nodes for eMMC and uSD card respectively.
Do note that most SM6350 devices seem to come with UFS.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
[bjorn: Replaced SM6350_CX with its constant value]
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-14-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
9264d3c8ee arm64: dts: qcom: sm6350: Add RPMHPD and BCM voter
Add RPMHPD node, its OPP table and BCM voter to prepare for performance level
voting.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-13-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
574af54562 arm64: dts: qcom: sm6350: Add PRNG node
Add a node for the PRNG to enable hw-accelerated pseudo-random number
generation.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-12-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
001eaf9514 arm64: dts: qcom: sm6350: Add SPMI bus
Add a node for SPMI to allow for communication with on-board PMICs.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-11-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
8fe2e0d9db arm64: dts: qcom: sm6350: Add AOSS_QMP
Add a node for AOSS_QMP in preparation for remote processor enablement.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-10-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
25e0ae6848 arm64: dts: qcom: sm6350: Add TSENS nodes
Add nodes required for TSENS block using the common qcom,tsens-v2 binding.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-9-konrad.dybcio@somainline.org
2021-09-27 17:21:29 -05:00
Konrad Dybcio
3cc415413f arm64: dts: qcom: sm6350: Add cpufreq-hw support
Add cpufreq-hw node and assign qcom,freq-domain properties to CPUs to enable
CPU clock scaling.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-8-konrad.dybcio@somainline.org
2021-09-27 17:21:28 -05:00
Konrad Dybcio
23737b9557 arm64: dts: qcom: sm6350: Add USB1 nodes
Add nodes required for USB1 to function. SM6350 (thankfully) resuses SDM845 and
SC7180 IP, so no additional code porting is required.

Acked-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
[bjorn: Renamed dwc3 node "usb"]
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-7-konrad.dybcio@somainline.org
2021-09-27 17:21:15 -05:00
Daniel Borkmann
4c9f093720 Merge branch 'bpf-xsk-rx-batch'
Magnus Karlsson says:

====================
This patch set introduces a batched interface for Rx buffer allocation
in AF_XDP buffer pool. Instead of using xsk_buff_alloc(*pool), drivers
can now use xsk_buff_alloc_batch(*pool, **xdp_buff_array,
max). Instead of returning a pointer to an xdp_buff, it returns the
number of xdp_buffs it managed to allocate up to the maximum value of
the max parameter in the function call. Pointers to the allocated
xdp_buff:s are put in the xdp_buff_array supplied in the call. This
could be a SW ring that already exists in the driver or a new
structure that the driver has allocated.

  u32 xsk_buff_alloc_batch(struct xsk_buff_pool *pool,
                           struct xdp_buff **xdp,
                           u32 max);

When using this interface, the driver should also use the new
interface below to set the relevant fields in the struct xdp_buff. The
reason for this is that xsk_buff_alloc_batch() does not fill in the
data and data_meta fields for you as is the case with
xsk_buff_alloc(). So it is not sufficient to just set data_end
(effectively the size) anymore in the driver. The reason for this is
performance as explained in detail in the commit message.

  void xsk_buff_set_size(struct xdp_buff *xdp, u32 size);

Patch 6 also optimizes the buffer allocation in the aligned case. In
this case, we can skip the reinitialization of most fields in the
xdp_buff_xsk struct at allocation time. As the number of elements in
the heads array is equal to the number of possible buffers in the
umem, we can initialize them once and for all at bind time and then
just point to the correct one in the xdp_buff_array that is returned
to the driver. No reason to have a stack of free head entries. In the
unaligned case, the buffers can reside anywhere in the umem, so this
optimization is not possible as we still have to fill in the right
information in the xdp_buff every single time one is allocated.

I have updated i40e and ice to use this new batched interface.

These are the throughput results on my 2.1 GHz Cascade Lake system:

Aligned mode:
ice: +11% / -9 cycles/pkt
i40e: +12% / -9 cycles/pkt

Unaligned mode:
ice: +1.5% / -1 cycle/pkt
i40e: +1% / -1 cycle/pkt

For the aligned case, batching provides around 40% of the performance
improvement and the aligned optimization the rest, around 60%. Would
have expected a ~4% boost for unaligned with this data, but I only get
around 1%. Do not know why. Note that memory consumption in aligned
mode is also reduced by this patch set.
====================

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2021-09-28 00:19:13 +02:00
Magnus Karlsson
e34087fc00 selftests: xsk: Add frame_headroom test
Add a test for the frame_headroom feature that can be set on the
umem. The logic added validates that all offsets in all tests and
packets are valid, not just the ones that have a specifically
configured frame_headroom.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-14-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
e4e9baf06a selftests: xsk: Change interleaving of packets in unaligned mode
Change the interleaving of packets in unaligned mode. With the current
buffer addresses in the packet stream, the last buffer in the umem
could not be used as a large packet could potentially write over the
end of the umem. The kernel correctly threw this buffer address away
and refused to use it. This is perfectly fine for all regular packet
streams, but the ones used for unaligned mode have every other packet
being at some different offset. As we will add checks for correct
offsets in the next patch, this needs to be fixed. Just start these
page-boundary straddling buffers one page earlier so that the last
one is not on the last page of the umem, making all buffers valid.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-13-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
96a40678ce selftests: xsk: Add single packet test
Add a test where a single packet is sent and received. This might
sound like a silly test, but since many of the interfaces in xsk are
batched, it is important to be able to validate that we did not break
something as fundamental as just receiving single packets, instead of
batches of packets at high speed.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-12-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
1bf3649688 selftests: xsk: Introduce pacing of traffic
Introduce pacing of traffic so that the Tx thread can never send more
packets than the receiver has processed plus the number of packets it
can have in its umem. So at any point in time, the number of in flight
packets (not processed by the Rx thread) are less than or equal to the
number of packets that can be held in the Rx thread's umem.

The batch size is also increased to improve running time.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-11-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
89013b8a29 selftests: xsk: Fix socket creation retry
The socket creation retry unnecessarily registered the umem once for
every retry. No reason to do this. It wastes memory and it might lead
to too many pages being locked at some point and the failure of a
test.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-10-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
872a1184db selftests: xsk: Put the same buffer only once in the fill ring
Fix a problem where the fill ring was populated with too many
entries. If number of buffers in the umem was smaller than the fill
ring size, the code used to loop over from the beginning of the umem
and start putting the same buffers in again. This is racy indeed as a
later packet can be received overwriting an earlier one before the Rx
thread manages to validate it.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-9-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
5b13205612 selftests: xsk: Fix missing initialization
Fix missing initialization of the member rx_pkt_nb in the packet
stream. This leads to some tests declaring success too early as the
test thought all packets had already been received.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-8-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
94033cd8e7 xsk: Optimize for aligned case
Optimize for the aligned case by precomputing the parameter values of
the xdp_buff_xsk and xdp_buff structures in the heads array. We can do
this as the heads array size is equal to the number of chunks in the
umem for the aligned case. Then every entry in this array will reflect
a certain chunk/frame and can therefore be prepopulated with the
correct values and we can drop the use of the free_heads stack. Note
that it is not possible to allocate more buffers than what has been
allocated in the aligned case since each chunk can only contain a
single buffer.

We can unfortunately not do this in the unaligned case as one chunk
might contain multiple buffers. In this case, we keep the old scheme
of populating a heads entry every time it is used and using
the free_heads stack.

Also move xp_release() and xp_get_handle() to xsk_buff_pool.h. They
were for some reason in xsk.c even though they are buffer pool
operations.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-7-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
6aab0bb0c5 i40e: Use the xsk batched rx allocation interface
Use the new xsk batched rx allocation interface for the zero-copy data
path. As the array of struct xdp_buff pointers kept by the driver is
really a ring that wraps, the allocation routine is modified to detect
a wrap and in that case call the allocation function twice. The
allocation function cannot deal with wrapped rings, only arrays. As we
now know exactly how many buffers we get and that there is no
wrapping, the allocation function can be simplified even more as all
if-statements in the allocation loop can be removed, improving
performance.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-6-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
db804cfc21 ice: Use the xsk batched rx allocation interface
Use the new xsk batched rx allocation interface for the zero-copy data
path. As the array of struct xdp_buff pointers kept by the driver is
really a ring that wraps, the allocation routine is modified to detect
a wrap and in that case call the allocation function twice. The
allocation function cannot deal with wrapped rings, only arrays. As we
now know exactly how many buffers we get and that there is no
wrapping, the allocation function can be simplified even more as all
if-statements in the allocation loop can be removed, improving
performance.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-5-magnus.karlsson@gmail.com
2021-09-28 00:18:35 +02:00
Magnus Karlsson
57f7f8b6bc ice: Use xdp_buf instead of rx_buf for xsk zero-copy
In order to use the new xsk batched buffer allocation interface, a
pointer to an array of struct xsk_buff pointers need to be provided so
that the function can put the result of the allocation there. In the
ice driver, we already have a ring that stores pointers to
xdp_buffs. This is only used for the xsk zero-copy driver and is a
union with the structure that is used for the regular non zero-copy
path. Unfortunately, that structure is larger than the xdp_buffs
pointers which mean that there will be a stride (of 20 bytes) between
each xdp_buff pointer. And feeding this into the xsk_buff_alloc_batch
interface will not work since it assumes a regular array of xdp_buff
pointers (each 8 bytes with 0 bytes in-between them on a 64-bit
system).

To fix this, remove the xdp_buff pointer from the rx_buf union and
move it one step higher to the union above which only has pointers to
arrays in it. This solves the problem and we can directly feed the SW
ring of xdp_buff pointers straight into the allocation function in the
next patch when that interface is used. This will improve performance.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-4-magnus.karlsson@gmail.com
2021-09-28 00:18:34 +02:00
Magnus Karlsson
47e4075df3 xsk: Batched buffer allocation for the pool
Add a new driver interface xsk_buff_alloc_batch() offering batched
buffer allocations to improve performance. The new interface takes
three arguments: the buffer pool to allocated from, a pointer to an
array of struct xdp_buff pointers which will contain pointers to the
allocated xdp_buffs, and an unsigned integer specifying the max number
of buffers to allocate. The return value is the actual number of
buffers that the allocator managed to allocate and it will be in the
range 0 <= N <= max, where max is the third parameter to the function.

u32 xsk_buff_alloc_batch(struct xsk_buff_pool *pool, struct xdp_buff **xdp,
                         u32 max);

A second driver interface is also introduced that need to be used in
conjunction with xsk_buff_alloc_batch(). It is a helper that sets the
size of struct xdp_buff and is used by the NIC Rx irq routine when
receiving a packet. This helper sets the three struct members data,
data_meta, and data_end. The two first ones is in the xsk_buff_alloc()
case set in the allocation routine and data_end is set when a packet
is received in the receive irq function. This unfortunately leads to
worse performance since the xdp_buff is touched twice with a long time
period in between leading to an extra cache miss. Instead, we fill out
the xdp_buff with all 3 fields at one single point in time in the
driver, when the size of the packet is known. Hence this helper. Note
that the driver has to use this helper (or set all three fields
itself) when using xsk_buff_alloc_batch(). xsk_buff_alloc() works as
before and does not require this.

void xsk_buff_set_size(struct xdp_buff *xdp, u32 size);

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-3-magnus.karlsson@gmail.com
2021-09-28 00:18:34 +02:00
Magnus Karlsson
10a5e009b9 xsk: Get rid of unused entry in struct xdp_buff_xsk
Get rid of the unused entry "unaligned" in struct xdp_buff_xsk.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210922075613.12186-2-magnus.karlsson@gmail.com
2021-09-28 00:18:34 +02:00
Konrad Dybcio
538f4bcd51 arm64: dts: qcom: sm6350: Add TLMM block node
Add TLMM pinctrl node to enable referencing the SoC pins in other nodes.

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-6-konrad.dybcio@somainline.org
2021-09-27 17:11:13 -05:00
Konrad Dybcio
30de1108df arm64: dts: qcom: sm6350: Add GCC node
Add and configure GCC node to allow for referencing GCC-controlled clocks
in other nodes.

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-5-konrad.dybcio@somainline.org
2021-09-27 17:11:13 -05:00
Konrad Dybcio
985e02e7c0 arm64: dts: qcom: sm6350: Add RPMHCC node
Add RPMHCC node to allow for referencing RPMH-controlled clocks in other
nodes.

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-4-konrad.dybcio@somainline.org
2021-09-27 17:11:13 -05:00
Konrad Dybcio
ced2f0d75e arm64: dts: qcom: sm6350: Add LLCC node
Add a node for LLCC with SM6350-specific compatible.

Acked-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-3-konrad.dybcio@somainline.org
2021-09-27 17:11:13 -05:00
Konrad Dybcio
5f82b9cda6 arm64: dts: qcom: Add SM6350 device tree
Add a base DT for SM6350 SoC

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-2-konrad.dybcio@somainline.org
2021-09-27 17:11:13 -05:00
Konrad Dybcio
55d0feb3ab dt-bindings: arm: cpus: Add Kryo 560 CPUs
Document Kryo 560 CPUs found in Qualcomm Snapdragon 690 (SM6350).

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@somainline.org>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20210923162204.21752-1-konrad.dybcio@somainline.org
2021-09-27 17:11:13 -05:00
Liang Chen
98419a39d1 arm64: dts: rockchip: add pwm nodes for rk3568
Add the pwm controller nodes to the core rk3568 dtsi.

Signed-off-by: Liang Chen <cl@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20210726090355.1548483-2-heiko@sntech.de
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
2021-09-27 23:45:49 +02:00
Linus Torvalds
0513e464f9 perf tools fixes for v5.15: 2nd batch
- Fix 'perf test' DWARF unwind for optimized builds.
 
 - Fix 'perf test' 'Object code reading' when dealing with samples in @plt
   symbols.
 
 - Fix off-by-one directory paths in the ARM support code.
 
 - Fix error message to eliminate confusion in 'perf config' when first creating
   a config file.
 
 - 'perf iostat' fix for system wide operation.
 
 - Fix printing of metrics when 'perf iostat' is used with one or more
   iio_root_ports and unconnected cpus (using -C).
 
 - Fix several typos in the documentation files.
 
 - Fix spelling mistake "icach" -> "icache" in the power8 JSON vendor files.
 
 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYIAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCYVIO3wAKCRCyPKLppCJ+
 J9piAP4jmxYEnimD6qvVHjOLio2LvwGI0u7MakZCHWVKQZKHbgEArb8l3+D2+YXw
 U7RxDmXoSE+0EjTV8o13sQlerRTU3wM=
 =oVI7
 -----END PGP SIGNATURE-----

Merge tag 'perf-tools-fixes-for-v5.15-2021-09-27' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull more perf tools fixes from Arnaldo Carvalho de Melo:

 - Fix 'perf test' DWARF unwind for optimized builds.

 - Fix 'perf test' 'Object code reading' when dealing with samples in
   @plt symbols.

 - Fix off-by-one directory paths in the ARM support code.

 - Fix error message to eliminate confusion in 'perf config' when first
   creating a config file.

 - 'perf iostat' fix for system wide operation.

 - Fix printing of metrics when 'perf iostat' is used with one or more
   iio_root_ports and unconnected cpus (using -C).

 - Fix several typos in the documentation files.

 - Fix spelling mistake "icach" -> "icache" in the power8 JSON vendor
   files.

* tag 'perf-tools-fixes-for-v5.15-2021-09-27' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  perf iostat: Fix Segmentation fault from NULL 'struct perf_counts_values *'
  perf iostat: Use system-wide mode if the target cpu_list is unspecified
  perf config: Refine error message to eliminate confusion
  perf doc: Fix typos all over the place
  perf arm: Fix off-by-one directory paths.
  perf vendor events powerpc: Fix spelling mistake "icach" -> "icache"
  perf tests: Fix flaky test 'Object code reading'
  perf test: Fix DWARF unwind for optimized builds.
2021-09-27 14:06:42 -07:00