By giving each register its own liveness chain, we elide the skip_callee()
logic. Instead, each register's parent is the state it inherits from;
both check_func_call() and prepare_func_exit() automatically connect
reg states to the correct chain since when they copy the reg state across
(r1-r5 into the callee as args, and r0 out as the return value) they also
copy the parent pointer.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This tag contains a handful of patches that filtered their way in during
the merge window but just didn't make the deadline. It includes:
* Additional documentation in the riscv,cpu-intc device tree binding
that resulted from some feedback I missed in the original patch set.
* A build fix that provides the definition of tlb_flush() before
including tlb.h, which fixes a RISC-V build regression introduced
during this merge window.
* A cosmetic cleanup to sys_riscv_flush_icache().
-----BEGIN PGP SIGNATURE-----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=5UCG
-----END PGP SIGNATURE-----
Merge tag 'riscv-for-linus-4.19-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux
Pull RISC-V fixes from Palmer Dabbelt:
"RISC-V Fixes and Cleanups for 4.19-rc2
This contains a handful of patches that filtered their way in during
the merge window but just didn't make the deadline. It includes:
- Additional documentation in the riscv,cpu-intc device tree binding
that resulted from some feedback I missed in the original patch
set.
- A build fix that provides the definition of tlb_flush() before
including tlb.h, which fixes a RISC-V build regression introduced
during this merge window.
- A cosmetic cleanup to sys_riscv_flush_icache()"
* tag 'riscv-for-linus-4.19-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux:
RISC-V: Use a less ugly workaround for unused variable warnings
riscv: tlb: Provide definition of tlb_flush() before including tlb.h
dt-bindings: riscv,cpu-intc: Cleanups from a missed review
Fixes for 4.19:
- SR-IOV fixes
- Kasan and page fault fix on device removal
- S3 stability fix for CZ/ST
- VCE regression fixes for CIK parts
- Avoid holding the mn_lock when allocating memory
- DC memory leak fix
- BO eviction fix
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180829202555.2653-1-alexander.deucher@amd.com
Add support for the R7S9210 which is part of the RZ/A2 series.
Signed-off-by: Chris Brandt <chris.brandt@renesas.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Peng Li says:
====================
net: hns: fix some bugs about speed and duplex change
If there are packets in hardware when changing the spped
or duplex, it may cause hardware hang up.
This patchset adds the code for waiting chip to clean the all
pkts(TX & RX) in chip when the driver uses the function named
"adjust link".
This patchset cleans the pkts as follows:
1) close rx of chip, close tx of protocol stack.
2) wait rcb, ppe, mac to clean.
3) adjust link
4) open rx of chip, open tx of protocol stack.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
If there are packets in hardware when changing the speed
or duplex, it may cause hardware hang up.
This patch adds netif_carrier_off before change speed and
duplex in ethtool_ops.set_link_ksettings, and adds
netif_carrier_on after complete the change.
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If there are packets in hardware when changing the speed
or duplex, it may cause hardware hang up.
This patch adds the code for waiting chip to clean the all
pkts(TX & RX) in chip when the driver uses the function named
"adjust link".
This patch cleans the pkts as follows:
1) close rx of chip, close tx of protocol stack.
2) wait rcb, ppe, mac to clean.
3) adjust link
4) open rx of chip, open tx of protocol stack.
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I have two Ethernet adapters:
r8169 0000:03:01.0 eth0: RTL8169sb/8110sb, 00:14:d1:14:2d:49, XID 10000000, IRQ 18
r8169 0000:01:00.0 eth0: RTL8168e/8111e, 64:66:b3:11:14:5d, XID 2c200000, IRQ 30
And after upgrading from linux 4.15 [1] to linux 4.18+ [2] RTL8169sb failed to
receive any packets. tcpdump shows a lot of checksum mismatch.
[1]: a0f79386a4
[2]: 0519359784 (4.19 merge window opened)
I started bisecting and the found that [3] breaks it. According to [4]:
"For 8110S, 8110SB, and 8110SC series, the initial value of RxConfig
needs to be set after the tx/rx is enabled."
So I moved rtl_init_rxcfg() after enabling tx/rs and now my adapter works
(RTL8168e works too).
[3]: 3559d81e76
[4]: e542a2269f ("r8169: adjust the RxConfig
settings.")
Also drop "rx" from rtl_set_rx_tx_config_registers(), since it does nothing
with it already.
Fixes: 3559d81e76 ("r8169: simplify
rtl_hw_start_8169")
Cc: Heiner Kallweit <hkallweit1@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: Realtek linux nic maintainers <nic_swsd@realtek.com>
Signed-off-by: Azat Khuzhin <a3at.mail@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
syzbot reported a use-after-free in tipc_group_fill_sock_diag(),
where tipc_group_fill_sock_diag() still reads tsk->group meanwhile
tipc_group_delete() just deletes it in tipc_release().
tipc_nl_sk_walk() aims to lock this sock when walking each sock
in the hash table to close race conditions with sock changes like
this one, by acquiring tsk->sk.sk_lock.slock spinlock, unfortunately
this doesn't work at all. All non-BH call path should take
lock_sock() instead to make it work.
tipc_nl_sk_walk() brutally iterates with raw rht_for_each_entry_rcu()
where RCU read lock is required, this is the reason why lock_sock()
can't be taken on this path. This could be resolved by switching to
rhashtable iterator API's, where taking a sleepable lock is possible.
Also, the iterator API's are friendly for restartable calls like
diag dump, the last position is remembered behind the scence,
all we need to do here is saving the iterator into cb->args[].
I tested this with parallel tipc diag dump and thousands of tipc
socket creation and release, no crash or memory leak.
Reported-by: syzbot+b9c8f3ab2994b7cd1625@syzkaller.appspotmail.com
Cc: Jon Maloy <jon.maloy@ericsson.com>
Cc: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This reverts commit 4ae0169fd1.
This change in the handling of the coalesce timer is causing regression on
(at least) amlogic platforms.
Network will break down very quickly (a few seconds) after starting
a download. This can easily be reproduced using iperf3 for example.
The problem has been reported on the S805, S905, S912 and A113 SoCs
(Realtek and Micrel PHYs) and it is likely impacting all Amlogics
platforms using Gbit ethernet
No problem was seen with the platform using 10/100 only PHYs (GXL internal)
Reverting change brings things back to normal and allows to use network
again until we better understand the problem with the coalesce timer.
Cc: Jose Abreu <joabreu@synopsys.com>
Cc: Joao Pinto <jpinto@synopsys.com>
Cc: Vitor Soares <soares@synopsys.com>
Cc: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Cc: Corentin Labbe <clabbe@baylibre.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rhashtable_walk_exit() must be paired with rhashtable_walk_enter().
Fixes: 40f9f43970 ("tipc: Fix tipc_sk_reinit race conditions")
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Before the commit d6990976af ("vti6: fix PMTU caching and reporting
on xmit") '!skb->ignore_df' check was always true because the function
skb_scrub_packet() was called before it, resetting ignore_df to zero.
In the commit, skb_scrub_packet() was moved below, and now this check
can be false for the packet, e.g. when sending it in the two fragments,
this prevents successful PMTU updates in such case. The next attempts
to send the packet lead to the same tx error. Moreover, vti6 initial
MTU value relies on PMTU adjustments.
This issue can be reproduced with the following LTP test script:
udp_ipsec_vti.sh -6 -p ah -m tunnel -s 2000
Fixes: ccd740cbc6 ("vti6: Add pmtu handling to vti6_xmit.")
Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
Acked-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann says:
====================
pull-request: bpf 2018-08-29
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) Fix a build error in sk_reuseport_convert_ctx_access() when
compiling with clang which cannot resolve hweight_long() at
build time inside the BUILD_BUG_ON() assertion, from Stefan.
2) Several fixes for BPF sockmap, four of them in getting the
bpf_msg_pull_data() helper to work, one use after free case
in bpf_tcp_close() and one refcount leak in bpf_tcp_recvmsg(),
from Daniel.
3) Another fix for BPF sockmap where we misaccount sk_mem_uncharge()
in the socket redirect error case from unwinding scatterlist
twice, from John.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The newly added code that emits ksymtab entries as pairs of 32-bit
relative references interacts poorly with the way powerpc lays out its
address space: when a module exports a per-CPU variable, the primary
module region covering the ksymtab entry -and thus the 32-bit relative
reference- is too far away from the actual per-CPU variable's base
address (to which the per-CPU offsets are applied to obtain the
respective address of each CPU's copy), resulting in corruption when the
module loader attempts to resolve symbol references of modules that are
loaded on top and link to the exported per-CPU symbol.
So let's disable this feature on powerpc. Even though it implements
CONFIG_RELOCATABLE, it does not implement CONFIG_RANDOMIZE_BASE and so
KASLR kernels (which are the main target of the feature) do not exist on
powerpc anyway.
Reported-by: Andreas Schwab <schwab@linux-m68k.org>
Suggested-by: Nicholas Piggin <nicholas.piggin@gmail.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Replace GPL v2.0 and v2.0+ license statements with SPDX license
identifiers.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
simple_strtoul is obsolete, and use kstrtoint instead
Signed-off-by: Ding Xiang <dingxiang@cmss.chinamobile.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEEq1nRK9aeMoq1VSgcnJ2qBz9kQNkFAluGfbAACgkQnJ2qBz9k
QNmkVQgAjv/tCHStwoQ4Hhr6q5wLU9apAW5WC7Or2MyNfqoJsKgyujpikwFMa0xY
wkVqlITYavvUKl/nRlQ6hpZtAY7hi1Y7GvIYs2ci6QO4YAUuBoH7qPLKqyZYzXx+
mBaC68885nMMDqIHvsCcLurwTiDer6XQXXPDmpMoO9g4kyVuhm2e0/M7CPaA4Ue9
WEfBPBROSNdRH7Wtww/MUvtfd+9mezdlpUQZVVO5cdhAVQ8pzW2g2piYtwIpaY7E
DiJ1zcgaqCnni+b69x4wz8TwB0q8wZJjrPfJgf4X7mIGBZrw80yNRlevfe5SxwdJ
mkMDvjaD0TEItvpjgTZReXaAjFOWTA==
=E61M
-----END PGP SIGNATURE-----
Merge tag 'for_v4.19-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Pull misc fs fixes from Jan Kara:
- make UDF to properly mount media created by Win7
- make isofs to properly refuse devices with large physical block size
- fix a Spectre gadget in quotactl(2)
- fix a warning in fsnotify code hit by syzkaller
* tag 'for_v4.19-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
udf: Fix mounting of Win7 created UDF filesystems
udf: Remove dead code from udf_find_fileset()
fs/quota: Fix spectre gadget in do_quotactl
fs/quota: Replace XQM_MAXQUOTAS usage with MAXQUOTAS
isofs: reject hardware sector size > 2048 bytes
fsnotify: fix false positive warning on inode delete
Pointer 'dev' is being assigned but is never used hence it is
redundant and can be removed.
Cleans up clang warning:
variable 'dev' set but not used [-Wunused-but-set-variable]
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
If the display has been disabled by modparam, we still want to connect
together the HW bits and bobs with the associated drivers so that we can
continue to manage their runtime power gating.
Fixes: 108109444f ("drm/i915: Check num_pipes before initializing audio component")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Elaine Wang <elaine.wang@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180817100241.4628-1-chris@chris-wilson.co.uk
(cherry picked from commit 35a5fd9ebf)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
100 ms is not enough time for the LSPCON adapter on Intel NUC devices to
settle. This causes dropped display modes at boot or screen reconfiguration.
Empirical testing can reproduce the error up to a timeout of 190 ms. Basic
boot and stress testing at 200 ms has not (yet) failed.
Increase timeout to 400 ms to get some margin of error.
Changes from v1:
The initial suggestion of 1000 ms was lowered due to concerns about delaying
valid timeout cases.
Update patch metadata.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107503
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1570392
Fixes: 357c0ae919 ("drm/i915/lspcon: Wait for expected LSPCON mode to settle")
Cc: Shashank Sharma <shashank.sharma@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: <stable@vger.kernel.org> # v4.11+
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Fredrik Schön <fredrik.schon@gmail.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180817200728.8154-1-fredrik.schon@gmail.com
(cherry picked from commit 59f1c8ab30)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
The context owns both the ppgtt and the vma within it, and our activity
tracking on the context ensures that we do not release active ppgtt. As
the context fulfils our obligations for active memory tracking, we can
relinquish the reference from the vma.
This fixes a silly transient refleak from closed vma being kept alive
until the entire system was idle, keeping all vm alive as well.
Reported-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Testcase: igt/gem_ctx_create/files
Fixes: 3365e2268b ("drm/i915: Lazily unbind vma on close")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Tested-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180816073448.19396-1-chris@chris-wilson.co.uk
(cherry picked from commit a4417b7b41)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Pull crypto fixes from Herbert Xu:
- Check for the right CPU feature bit in sm4-ce on arm64.
- Fix scatterwalk WARN_ON in aes-gcm-ce on arm64.
- Fix unaligned fault in aesni on x86.
- Fix potential NULL pointer dereference on exit in chtls.
- Fix DMA mapping direction for RSA in caam.
- Fix error path return value for xts setkey in caam.
- Fix address endianness when DMA unmapping in caam.
- Fix sleep-in-atomic in vmx.
- Fix command corruption when queue is full in cavium/nitrox.
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: cavium/nitrox - fix for command corruption in queue full case with backlog submissions.
crypto: vmx - Fix sleep-in-atomic bugs
crypto: arm64/aes-gcm-ce - fix scatterwalk API violation
crypto: aesni - Use unaligned loads from gcm_context_data
crypto: chtls - fix null dereference chtls_free_uld()
crypto: arm64/sm4-ce - check for the right CPU feature bit
crypto: caam - fix DMA mapping direction for RSA forms 2 & 3
crypto: caam/qi - fix error path in xts setkey
crypto: caam/jr - fix descriptor DMA unmapping
This patch adds missing properties to the CODEC and sound nodes, so the
audio will work also on Snow rev5 Chromebook. This patch is an extension
to the commit e9eefc3f8c ("ARM: dts: exynos: Add missing clock and
DAI properties to the max98095 node in Snow Chromebook")
and commit 6ab569936d ("ARM: dts: exynos: Enable HDMI audio on Snow
Chromebook"). It has been reported that such changes work fine on the
rev5 board too.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
[krzk: Fixed typo in phandle to &max98090]
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
The LDO28 is used only on Odroid XU3 for Display Port. Define it so DTS
will describe entire hardware.
Depending on bootloader behavior this might affect the Display Port
because none of drivers are enabling it. By default it is off in S2MPS11
PMIC reset values. However it could be enabled by bootloader so in such
case kernel will later disable it as unused regulator.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Disable unused PMIC regulators on Exynos5422 Odroid boards to reduce
energy used. According to schematics:
1. LDO12, LDO16 and LDO24 are not connected,
2. LDO26 is not used on Odroid HC1.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Define LDO2, LDO23 and LDO27 critical board regulators to describe the
hardware.
Suggested-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
1 GHz CPU OPP is the default boot value for the Exynos5250 SOC, so mark it
as suspend OPP. This fixes suspend/resume on Samsung Exynos5250 Snow
Chomebook, which was broken since switching to generic cpufreq-dt driver
in v4.3.
Cc: <stable@vger.kernel.org> # 4.3.x: cd6f55457e: ARM: dts: exynos: Remove "cooling-{min|max}-level" for CPU nodes
Cc: <stable@vger.kernel.org> # 4.3.x: 672f33198b: arm: dts: exynos: Add missing cooling device properties for CPUs
Cc: <stable@vger.kernel.org> # 4.3.x
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Chanwoo Choi <cw00.choi@samsung.com>
Acked-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Convert Exynos5250 to OPP-v2 bindings. This is a preparation to add proper
support for suspend operation point, which cannot be marked in opp-v1.
Cc: <stable@vger.kernel.org> # 4.3.x: cd6f55457e: ARM: dts: exynos: Remove "cooling-{min|max}-level" for CPU nodes
Cc: <stable@vger.kernel.org> # 4.3.x: 672f33198b: arm: dts: exynos: Add missing cooling device properties for CPUs
Cc: <stable@vger.kernel.org> # 4.3.x
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Chanwoo Choi <cw00.choi@samsung.com>
Acked-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Björn Töpel says:
====================
This patch set introduces zero-copy AF_XDP support for Intel's i40e
driver. In the first preparatory patch we also add support for
XDP_REDIRECT for zero-copy allocated frames so that XDP programs can
redirect them. This was a ToDo from the first AF_XDP zero-copy patch
set from early June. Special thanks to Alex Duyck and Jesper Dangaard
Brouer for reviewing earlier versions of this patch set.
The i40e zero-copy code is located in its own file i40e_xsk.[ch]. Note
that in the interest of time, to get an AF_XDP zero-copy implementation
out there for people to try, some code paths have been copied from the
XDP path to the zero-copy path. It is out goal to merge the two paths
in later patch sets.
In contrast to the implementation from beginning of June, this patch
set does not require any extra HW queues for AF_XDP zero-copy
TX. Instead, the XDP TX HW queue is used for both XDP_REDIRECT and
AF_XDP zero-copy TX.
Jeff, given that most of changes are in i40e, it is up to you how you
would like to route these patches. The set is tagged bpf-next, but
if taking it via the Intel driver tree is easier, let us know.
We have run some benchmarks on a dual socket system with two Broadwell
E5 2660 @ 2.0 GHz with hyperthreading turned off. Each socket has 14
cores which gives a total of 28, but only two cores are used in these
experiments. One for TR/RX and one for the user space application. The
memory is DDR4 @ 2133 MT/s (1067 MHz) and the size of each DIMM is
8192MB and with 8 of those DIMMs in the system we have 64 GB of total
memory. The compiler used is gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0. The
NIC is Intel I40E 40Gbit/s using the i40e driver.
Below are the results in Mpps of the I40E NIC benchmark runs for 64
and 1500 byte packets, generated by a commercial packet generator HW
outputing packets at full 40 Gbit/s line rate. The results are with
retpoline and all other spectre and meltdown fixes, so these results
are not comparable to the ones from the zero-copy patch set in June.
AF_XDP performance 64 byte packets.
Benchmark XDP_SKB XDP_DRV XDP_DRV with zerocopy
rxdrop 2.6 8.2 15.0
txpush 2.2 - 21.9
l2fwd 1.7 2.3 11.3
AF_XDP performance 1500 byte packets:
Benchmark XDP_SKB XDP_DRV XDP_DRV with zerocopy
rxdrop 2.0 3.3 3.3
l2fwd 1.3 1.7 3.1
XDP performance on our system as a base line:
64 byte packets:
XDP stats CPU pps issue-pps
XDP-RX CPU 16 18.4M 0
1500 byte packets:
XDP stats CPU pps issue-pps
XDP-RX CPU 16 3.3M 0
The structure of the patch set is as follows:
Patch 1: Add support for XDP_REDIRECT of zero-copy allocated frames
Patches 2-4: Preparatory patches to common xsk and net code
Patches 5-7: Preparatory patches to i40e driver code for RX
Patch 8: i40e zero-copy support for RX
Patch 9: Preparatory patch to i40e driver code for TX
Patch 10: i40e zero-copy support for TX
Patch 11: Add flags to sample application to force zero-copy/copy mode
====================
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The -c/--copy -z/--zero-copy flags enforces either copy or zero-copy
mode.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch adds zero-copy Tx support for AF_XDP sockets. It implements
the ndo_xsk_async_xmit netdev ndo and performs all the Tx logic from a
NAPI context. This means pulling egress packets from the Tx ring,
placing the frames on the NIC HW descriptor ring and completing sent
frames back to the application via the completion ring.
The regular XDP Tx ring is used for AF_XDP as well. This rationale for
this is as follows: XDP_REDIRECT guarantees mutual exclusion between
different NAPI contexts based on CPU id. In other words, a netdev can
XDP_REDIRECT to another netdev with a different NAPI context, since
the operation is bound to a specific core and each core has its own
hardware ring.
As the AF_XDP Tx action is running in the same NAPI context and using
the same ring, it will also be protected from XDP_REDIRECT actions
with the exact same mechanism.
As with AF_XDP Rx, all AF_XDP Tx specific functions are added to
i40e_xsk.c.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch prepares for the upcoming zero-copy Tx functionality, by
moving common functions and refactor chunks of code into re-usable
functions, used both by the regular path and zero-copy path.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch adds zero-copy Rx support for AF_XDP sockets. Instead of
allocating buffers of type MEM_TYPE_PAGE_SHARED, the Rx frames are
allocated as MEM_TYPE_ZERO_COPY when AF_XDP is enabled for a certain
queue.
All AF_XDP specific functions are added to a new file, i40e_xsk.c.
Note that when AF_XDP zero-copy is enabled, the XDP action XDP_PASS
will allocate a new buffer and copy the zero-copy frame prior passing
it to the kernel stack.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch prepares for the upcoming zero-copy Rx functionality, by
moving/changing linkage of common functions, used both by the regular
path and zero-copy path.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
In this commit, the Rx path is refactored some, as a step torwards the
introduction AF_XDP Rx zero-copy.
The page re-use counter is moved into the i40e_reuse_rx_page, instead
of bumping the counter in many places. The Rx buffer page clearing is
moved for better readability. Lastely, functions to update statistics
and bump the XDP Tx ring are introduced.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add functions for queue pair enable/disable. Instead of resetting the
whole device, only the affected queue pair is disabled or enabled.
This plumbing is used in a later commit, when zero-copy AF_XDP support
is introduced.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The function napi_if_scheduled_mark_missed is used to check if the
NAPI context is scheduled, if so set NAPIF_STATE_MISSED and return
true. Used by the AF_XDP zero-copy i40e Tx code implementation in
order to make sure that irq affinity is honored by the napi context.
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Move the xdp_umem_get_{data,dma} functions to include/net/xdp_sock.h,
so that the upcoming zero-copy implementation in the Ethernet drivers
can utilize them.
Also, supply some dummy function implementations for
CONFIG_XDP_SOCKETS=n configs.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Export __xdp_rxq_info_unreg_mem_model as xdp_rxq_info_unreg_mem_model,
so it can be used from netdev drivers. Also, add additional checks for
the memory type.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This commit adds proper MEM_TYPE_ZERO_COPY support for
convert_to_xdp_frame. Converting a MEM_TYPE_ZERO_COPY xdp_buff to an
xdp_frame is done by transforming the MEM_TYPE_ZERO_COPY buffer into a
MEM_TYPE_PAGE_ORDER0 frame. This is costly, and in the future it might
make sense to implement a more sophisticated thread-safe alloc/free
scheme for MEM_TYPE_ZERO_COPY, so that no allocation and copy is
required in the fast-path.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The pin controller block of S5Pv210 for handling external wakeup
interrupts is different than in newer designs (Exynos).
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Cc: Tomasz Figa <tomasz.figa@gmail.com>
Cc: Sylwester Nawrocki <snawrocki@kernel.org>
Acked-by: Tomasz Figa <tomasz.figa@gmail.com>
HDMI-HPD was set active low, moreover by default pincontrol chip sets
pull-down on the pin. As a result HDMI driver assumes TV is always
connected regardless of actual state. The patch fixes it.
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
HDMI-DDC for unknown reasons doesn't work with Exynos I2C controllers.
Fortunately i2c-gpio comes to the rescue.
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Add to Arndale board: bridge and panel nodes, DSI properties and
regulators required by the bridge.
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Maciej Purski <m.purski@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
