While working on BPF ringbuf implementation, testing, and benchmarking, I've
developed a pretty generic and modular benchmark runner, which seems to be
generically useful, as I've already used it for one more purpose (testing
fastest way to trigger BPF program, to minimize overhead of in-kernel cod
Add generic benchmark runner framework which simplifies writing various
performance benchmarks in a consistent fashion. This framework will be used
in follow up patches to test performance of perf buffer and ring buffer as
well.
Patch #1 adds generic runner implementation and atomic counter bench
It is sometimes desirable to be able to trigger BPF program from user-space
with minimal overhead. sys_enter would seem to be a good candidate, yet in
a lot of cases there will be a lot of noise from syscalls triggered by other
processes on the system. So while searching for low-overhead alternativ
Add fmod_ret BPF program to existing test_overhead selftest. Also re-implement
user-space benchmarking part into benchmark runner to compare results. Results
with ./bench are consistently somewhat lower than test_overhead's, but relative
performance of various types of BPF programs stay consisten
On Wed, May 06, 2020 at 02:54:58PM +0200, Jakub Sitnicki wrote:
> Add a new program type BPF_PROG_TYPE_SK_LOOKUP and a dedicated attach type
> called BPF_SK_LOOKUP. The new program kind is to be invoked by the
> transport layer when looking up a socket for a received packet.
>
> When called, SK_LO
Hi David,
On Thu, May 7, 2020 at 10:17 PM David Miller wrote:
> From: Geert Uytterhoeven
> Date: Thu, 7 May 2020 13:42:05 +0200
>
> > The VIA Rhine Ethernet interface is only present on PCI devices or
> > VIA/WonderMedia VT8500/WM85xx SoCs. Add platform dependencies to the
> > VIA_RHINE config
On Thu, May 7, 2020 at 11:40 PM Ian Rogers wrote:
>
> If bits is 0, the case when the map is empty, then the >> is the size of
> the register which is undefined behavior - on x86 it is the same as a
> shift by 0. Fix by handling the 0 case explicitly.
>
> Signed-off-by: Ian Rogers
> ---
No need.
On Thu, May 07, 2020 at 12:12:13PM -0700, Stanislav Fomichev wrote:
> 1. Move pkt_v4 and pkt_v6 into network_helpers and adjust the users.
> 2. Copy-paste spin_lock_thread into two tests that use it.
>
> Cc: Martin KaFai Lau
> Signed-off-by: Stanislav Fomichev
Acked-by: Martin KaFai Lau
On Thu, May 07, 2020 at 12:12:12PM -0700, Stanislav Fomichev wrote:
> Move the following routines that let us start a background listener
> thread and connect to a server by fd to the test_prog:
> * start_server - socket+bind+listen
> * connect_to_fd - connect to the server identified by fd
>
> Th
On Fri, 8 May 2020 10:05:46 +0800
Jason Wang wrote:
> On 2020/5/7 上午4:34, Michael S. Tsirkin wrote:
> > On Thu, Apr 30, 2020 at 01:22:08PM +0200, Jesper Dangaard Brouer wrote:
> >> The virtio_net driver is running inside the guest-OS. There are two
> >> XDP receive code-paths in virtio_net, nam
Fix to return negative error code -ENOMEM from the kzalloc() error
handling case instead of 0, as done elsewhere in this function.
Fixes: 174ab544e3bc ("nfp: abm: add cls_u32 offload for simple band
classification")
Reported-by: Hulk Robot
Signed-off-by: Wei Yongjun
---
drivers/net/ethernet/ne
Introduce a ingress frame gate control flow action.
Tc gate action does the work like this:
Assume there is a gate allow specified ingress frames can pass at
specific time slot, and also drop at specific time slot. Tc filter
chooses the ingress frames, and tc gate action would specify what slot
doe
This patch is to add the man page for the tc gate action.
Signed-off-by: Po Liu
---
changes before v4:
No changes.
changes from v4:
- Update the examples with input/output time and size routine support
man/man8/tc-gate.8 | 123 +
1 file changed, 123
On Fri, May 8, 2020 at 11:00 AM Kevin Hao wrote:
>
> On Fri, May 08, 2020 at 10:18:27AM +0530, Sunil Kovvuri wrote:
> > On Fri, May 8, 2020 at 9:43 AM Kevin Hao wrote:
> > >
> > > In the current codes, the octeontx2 uses its own method to allocate
> > > the pool buffers, but there are some issues
On Fri, May 8, 2020 at 9:43 AM Kevin Hao wrote:
>
> In the current codes, the octeontx2 uses its own method to allocate
> the pool buffers, but there are some issues in this implementation.
> 1. We have to run the otx2_get_page() for each allocation cycle and
>this is pretty error prone. As I
Fix the following coccicheck warning:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c:1785:5-8:
WARNING: Comparison to bool
Signed-off-by: Jason Yan
Reviewed-by: Chi-hsien Lin
---
v2: Rebased on top of wireless-drivers-next and drop one already fixed line.
drivers/net/wireless/broadco
> On Wed, 6 May 2020 14:33:08 +0300 Igor Russkikh wrote:
>> To implement debug dump retrieval on a live system we add callbacks to
>> collect the same data which is collected now during manual `ethtool -d`
>> call.
>>
>> But we instead collect the dump immediately at the moment bad thing
>> happe
On Mon, May 04, 2020 at 10:06:02AM +0200, Florian Westphal wrote:
> v2: rebase on top of ipsec-next, no other changes.
>
> This patch series removes three more indirect calls from the state_afinfo
> struct.
>
> These are:
> - extract_input (no dependencies on other modules)
> - output_finish (sam
Hi David, Rob,
On 2020/05/02 7:21, David Miller wrote:
From: Kunihiko Hayashi
Date: Tue, 28 Apr 2020 15:31:22 +0900
Convert the UniPhier AVE4 controller binding to DT schema format.
This changes phy-handle property to required.
Signed-off-by: Kunihiko Hayashi
DT folks, is it ok if I take
On 08/05/2020 05:10, Wei Yongjun wrote:
Fix to return negative error code -ENOMEM from the cpsw_ale_create()
error handling case instead of 0, as done elsewhere in this function.
Fixes: 93a76530316a ("net: ethernet: ti: introduce am65x/j721e gigabit eth subsystem
driver")
Reported-by: Hulk R
On Wed, May 06, 2020 at 12:00:08PM +0200, Toke Høiland-Jørgensen wrote:
> > No, I haven't test the performance. Do you have any suggestions about how
> > to test it? I'd like to try forwarding pkts to 10+ ports. But I don't know
> > how to test the throughput. I don't think netperf or iperf support
On Thu, May 07, 2020 at 10:55 PM CEST, Martin KaFai Lau wrote:
> On Wed, May 06, 2020 at 03:53:35PM +0200, Jakub Sitnicki wrote:
>> On Wed, May 06, 2020 at 03:16 PM CEST, Lorenz Bauer wrote:
>> > On Wed, 6 May 2020 at 13:55, Jakub Sitnicki wrote:
>>
>> [...]
>>
>> >> @@ -4012,4 +4051,18 @@ struct
On Thu, 7 May 2020 at 16:48, Toke Høiland-Jørgensen wrote:
>
[]
> Well, my immediate thought would be that the added complexity would not
> be worth it, because:
>
> - A new action would mean either you'd need to patch all drivers or
> (more likely) we'd end up with yet another difference betwee
On Thu, 7 May 2020 at 20:08, John Fastabend wrote:
>
[]
>
> I'm wondering if we can teach the verifier to recognize tail calls,
>
> int xdp_prog1(struct xdp_md *ctx)
> {
> return xdp_do_redirect(ctx, &xsks_map, 0);
> }
>
> This would be useful for normal calls as well. I guess the question
Hi Jakub,
>> This patch moves MAC generation into a separate function, which is
> called
>> from both places to reduce the amount of copy/paste.
>
> Right, but why do you have your own mac generation rather than using
> eth_hw_addr_random(). You need to set NET_ADDR_RANDOM for example,
> just us
From: Clay McClure
My recent commit b6d49cab44b5 ("net: Make PTP-specific drivers depend on
PTP_1588_CLOCK") exposes a missing dependency in defconfigs that select
TI_CPTS without selecting PTP_1588_CLOCK, leading to linker errors of the
form:
drivers/net/ethernet/ti/cpsw.o: in function `cpsw_nd
cpsw_ale_create() can return both NULL and PTR_ERR(), but all of
the caller only check NULL for error handling. This patch convert
it to only return PTR_ERR() in all error cases, all the caller using
IS_ERR() install of NULL test.
Also fix a return negative error code from the cpsw_ale_create()
er
Use the helper function that wraps the calls to platform_get_resource()
and devm_ioremap_resource() together.
Signed-off-by: Wei Yongjun
---
drivers/net/dsa/vitesse-vsc73xx-platform.c | 10 +-
1 file changed, 1 insertion(+), 9 deletions(-)
diff --git a/drivers/net/dsa/vitesse-vsc73xx-pl
On Fri, May 8, 2020 at 11:59 AM Grygorii Strashko
wrote:
>
> From: Clay McClure
>
> My recent commit b6d49cab44b5 ("net: Make PTP-specific drivers depend on
> PTP_1588_CLOCK") exposes a missing dependency in defconfigs that select
> TI_CPTS without selecting PTP_1588_CLOCK, leading to linker erro
On Fri, May 08, 2020 at 09:06 AM CEST, Martin KaFai Lau wrote:
> On Wed, May 06, 2020 at 02:54:58PM +0200, Jakub Sitnicki wrote:
>> Add a new program type BPF_PROG_TYPE_SK_LOOKUP and a dedicated attach type
>> called BPF_SK_LOOKUP. The new program kind is to be invoked by the
>> transport layer whe
On Fri, 1 May 2020 14:32:32 +0200
Jesper Dangaard Brouer wrote:
> On Thu, 30 Apr 2020 20:12:11 +0300
> Tariq Toukan wrote:
>
> > >> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h
> > >> b/drivers/net/ethernet/mellanox/mlx5/core/en.h
> > >> index 23701c0e36ec..ba6a0ee297c6 100644
> >
This driver uses RX page-split when possible. It was recently fixed
in commit 86e85bf6981c ("sfc: fix XDP-redirect in this driver") to
add needed tailroom for XDP-redirect.
After the fix efx->rx_page_buf_step is the frame size, with enough
head and tail-room for XDP-redirect.
Signed-off-by: Jespe
(Send for net-next due to all the driver updates)
V3:
- Fix issue on virtio_net patch spotted by Jason Wang
- Adjust name for variable in mlx5 patch
- Collected more ACKs
V2:
- Fix bug in mlx5 for XDP_PASS case
- Collected nitpicks and ACKs from mailing list
V1:
- Fix bug in dpaa2
XDP have evol
XDP have evolved to support several frame sizes, but xdp_buff was not
updated with this information. The frame size (frame_sz) member of
xdp_buff is introduced to know the real size of the memory the frame is
delivered in.
When introducing this also make it clear that some tailroom is
reserved/req
This driver uses full PAGE_SIZE pages when XDP is enabled.
In case of XDP uses driver uses __bnxt_alloc_rx_page which does full
page DMA-map. Thus, xdp_adjust_tail grow is DMA compliant for XDP_TX
action that does DMA-sync.
Cc: Michael Chan
Cc: Andy Gospodarek
Signed-off-by: Jesper Dangaard Bro
This marvell driver mvneta uses PAGE_SIZE frames, which makes it
really easy to convert. Driver updates rxq and now frame_sz
once per NAPI call.
This driver takes advantage of page_pool PP_FLAG_DMA_SYNC_DEV that
can help reduce the number of cache-lines that need to be flushed
when doing DMA sync
When native XDP redirect into a veth device, the frame arrives in the
xdp_frame structure. It is then processed in veth_xdp_rcv_one(),
which can run a new XDP bpf_prog on the packet. Doing so requires
converting xdp_frame to xdp_buff, but the tricky part is that
xdp_frame memory area is located in
From: Ilias Apalodimas
This driver takes advantage of page_pool PP_FLAG_DMA_SYNC_DEV that
can help reduce the number of cache-lines that need to be flushed
when doing DMA sync for_device. Due to xdp_adjust_tail can grow the
area accessible to the by the CPU (can possibly write into), then max
syn
Use hole in struct xdp_frame, when adding member frame_sz, which keeps
same sizeof struct (32 bytes)
Drivers ixgbe and sfc had bug cases where the necessary/expected
tailroom was not reserved. This can lead to some hard to catch memory
corruption issues. Having the drivers frame_sz this can be det
Knowing the memory size backing the packet/xdp_frame data area, and
knowing it already have reserved room for skb_shared_info, simplifies
using build_skb significantly.
With this change we no-longer lie about the SKB truesize, but more
importantly a significant larger skb_tailroom is now provided,
The veth driver can run XDP in "native" mode in it's own NAPI
handler, and since commit 9fc8d518d9d5 ("veth: Handle xdp_frames in
xdp napi ring") packets can come in two forms either xdp_frame or
skb, calling respectively veth_xdp_rcv_one() or veth_xdp_rcv_skb().
For packets to arrive in xdp_frame
The hyperv NIC driver does memory allocation and copy even without XDP.
In XDP mode it will allocate a new page for each packet and copy over
the payload, before invoking the XDP BPF-prog.
The positive thing it that its easy to determine the xdp.frame_sz.
The XDP implementation for hv_netvsc tran
The driver code cpsw.c and cpsw_new.c both use page_pool
with default order-0 pages or their RX-pages.
Cc: Grygorii Strashko
Cc: Ilias Apalodimas
Signed-off-by: Jesper Dangaard Brouer
Reviewed-by: Grygorii Strashko
---
drivers/net/ethernet/ti/cpsw.c |1 +
drivers/net/ethernet/ti/cpsw_
Frame size ENA_PAGE_SIZE is limited to 16K on systems with larger
PAGE_SIZE than 16K. Change ENA_XDP_MAX_MTU to also take into account
the reserved tailroom.
Cc: Arthur Kiyanovski
Acked-by: Sameeh Jubran
Signed-off-by: Jesper Dangaard Brouer
---
drivers/net/ethernet/amazon/ena/ena_netdev.c |
The SKB "head" pointer points to the data area that contains
skb_shared_info, that can be found via skb_end_pointer(). Given
xdp->data_hard_start have been established (basically pointing to
skb->head), frame size is between skb_end_pointer() and data_hard_start,
plus the size reserved to skb_share
The dpaa2-eth driver reserve some headroom used for hardware and
software annotation area in RX/TX buffers. Thus, xdp.data_hard_start
doesn't start at page boundary.
When XDP is configured the area reserved via dpaa2_fd_get_offset(fd) is
448 bytes of which XDP have reserved 256 bytes. As frame_sz
This patch mirrors the changes to ixgbe in previous patch.
This VF driver doesn't support XDP_REDIRECT, but correct tailroom is
still necessary for BPF-helper xdp_adjust_tail. In legacy-mode +
larger PAGE_SIZE, due to lacking tailroom, we accept that
xdp_adjust_tail shrink doesn't work.
Cc: inte
The ixgbe driver have another memory model when compiled on archs with
PAGE_SIZE above 4096 bytes. In this mode it doesn't split the page in
two halves, but instead increment rx_buffer->page_offset by truesize of
packet (which include headroom and tailroom for skb_shared_info).
This is done correc
The driver qede uses a full page, when XDP is enabled. The drivers value
in rx_buf_seg_size (struct qede_rx_queue) will be PAGE_SIZE when an
XDP bpf_prog is attached.
Cc: Ariel Elior
Cc: gr-everest-linux...@marvell.com
Signed-off-by: Jesper Dangaard Brouer
---
drivers/net/ethernet/qlogic/qede/q
This driver uses different memory models depending on PAGE_SIZE at
compile time. For PAGE_SIZE 4K it uses page splitting, meaning for
normal MTU frame size is 2048 bytes (and headroom 192 bytes). For
larger MTUs the driver still use page splitting, by allocating
order-1 pages (8192 bytes) for RX fr
The virtio_net driver is running inside the guest-OS. There are two
XDP receive code-paths in virtio_net, namely receive_small() and
receive_mergeable(). The receive_big() function does not support XDP.
In receive_small() the frame size is available in buflen. The buffer
backing these frames are a
The netronome nfp driver use PAGE_SIZE when xdp_prog is set, but
xdp.data_hard_start begins at offset NFP_NET_RX_BUF_HEADROOM.
Thus, adjust for this when setting xdp.frame_sz, as it counts
from data_hard_start.
When doing XDP_TX this driver is smart and instead of a full DMA-map
does a DMA-sync on
This driver uses different memory models depending on PAGE_SIZE at
compile time. For PAGE_SIZE 4K it uses page splitting, meaning for
normal MTU frame size is 2048 bytes (and headroom 192 bytes). For
larger MTUs the driver still use page splitting, by allocating
order-1 pages (8192 bytes) for RX fr
The mlx4 drivers size of memory backing the RX packet is stored in
frag_stride. For XDP mode this will be PAGE_SIZE (normally 4096).
For normal mode frag_stride is 2048.
Also adjust MLX4_EN_MAX_XDP_MTU to take tailroom into account.
Cc: Tariq Toukan
Cc: Saeed Mahameed
Signed-off-by: Jesper Dang
Intel drivers implement native AF_XDP zerocopy in separate C-files,
that have its own invocation of bpf_prog_run_xdp(). The setup of
xdp_buff is also handled in separately from normal code path.
This patch update XDP frame_sz for AF_XDP zerocopy drivers i40e, ice
and ixgbe, as the code changes nee
To help reviewers these are the defines related to RCV_FRAG_LEN
#define DMA_BUFFER_LEN 1536 /* In multiples of 128bytes */
#define RCV_FRAG_LEN (SKB_DATA_ALIGN(DMA_BUFFER_LEN + NET_SKB_PAD) + \
SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
Cc: Sunil Goutham
Cc: Rob
In vhost_net_build_xdp() the 'buf' that gets queued via an xdp_buff
have embedded a struct tun_xdp_hdr (located at xdp->data_hard_start)
which contains the buffer length 'buflen' (with tailroom for
skb_shared_info). Also storing this buflen in xdp->frame_sz, does not
obsolete struct tun_xdp_hdr, as
This driver uses different memory models depending on PAGE_SIZE at
compile time. For PAGE_SIZE 4K it uses page splitting, meaning for
normal MTU frame size is 2048 bytes (and headroom 192 bytes). For
larger MTUs the driver still use page splitting, by allocating
order-1 pages (8192 bytes) for RX fr
The tun driver have two code paths for running XDP (bpf_prog_run_xdp).
In both cases 'buflen' contains enough tailroom for skb_shared_info.
Cc: Jason Wang
Signed-off-by: Jesper Dangaard Brouer
Acked-by: Michael S. Tsirkin
Acked-by: Jason Wang
---
drivers/net/tun.c |2 ++
1 file changed, 2
The mlx5 driver have multiple memory models, which are also changed
according to whether a XDP bpf_prog is attached.
The 'rx_striding_rq' setting is adjusted via ethtool priv-flags e.g.:
# ethtool --set-priv-flags mlx5p2 rx_striding_rq off
On the general case with 4K page_size and regular MTU pa
Finally, after all drivers have a frame size, allow BPF-helper
bpf_xdp_adjust_tail() to grow or extend packet size at frame tail.
Remember that helper/macro xdp_data_hard_end have reserved some
tailroom. Thus, this helper makes sure that the BPF-prog don't have
access to this tailroom area.
V2:
Current selftest for BPF-helper xdp_adjust_tail only shrink tail.
Make it more clear that this is a shrink test case.
Signed-off-by: Jesper Dangaard Brouer
---
.../selftests/bpf/prog_tests/xdp_adjust_tail.c |9 +-
.../testing/selftests/bpf/progs/test_adjust_tail.c | 30
Extend BPF selftest xdp_adjust_tail with grow tail tests, which is added
as subtest's. The first grow test stays in same form as original shrink
test. The second grow test use the newer bpf_prog_test_run_xattr() calls,
and does extra checking of data contents.
Signed-off-by: Jesper Dangaard Brouer
Clearing memory of tail when grow happens, because it is too easy
to write a XDP_PASS program that extend the tail, which expose
this memory to users that can run tcpdump.
Signed-off-by: Jesper Dangaard Brouer
Acked-by: Toke Høiland-Jørgensen
---
net/core/filter.c |4
1 file changed, 4
Update the memory requirements, when adding xdp.frame_sz in BPF test_run
function bpf_prog_test_run_xdp() which e.g. is used by XDP selftests.
Specifically add the expected reserved tailroom, but also allocated a
larger memory area to reflect that XDP frames usually comes in this
format. Limit the
On 08/05/2020 13:10, Arnd Bergmann wrote:
On Fri, May 8, 2020 at 11:59 AM Grygorii Strashko
wrote:
From: Clay McClure
My recent commit b6d49cab44b5 ("net: Make PTP-specific drivers depend on
PTP_1588_CLOCK") exposes a missing dependency in defconfigs that select
TI_CPTS without selecting
On Fri, May 8, 2020 at 1:14 PM Grygorii Strashko
wrote:
> On 08/05/2020 13:10, Arnd Bergmann wrote:
> > On Fri, May 8, 2020 at 11:59 AM Grygorii Strashko
> >> That's because TI_CPTS_MOD (which is the symbol gating the _compilation_ of
> >> cpts.c) now depends on PTP_1588_CLOCK, and so is not enab
On 08/05/2020 13:06, Wei Yongjun wrote:
cpsw_ale_create() can return both NULL and PTR_ERR(), but all of
the caller only check NULL for error handling. This patch convert
it to only return PTR_ERR() in all error cases, all the caller using
IS_ERR() install of NULL test.
Also fix a return nega
On Fri, May 08, 2020 at 01:10:00PM +0530, Sunil Kovvuri wrote:
> On Fri, May 8, 2020 at 9:43 AM Kevin Hao wrote:
> >
> > In the current codes, the octeontx2 uses its own method to allocate
> > the pool buffers, but there are some issues in this implementation.
> > 1. We have to run the otx2_get_pa
On Thu, May 07, 2020 at 02:48:07PM -0700, Luke Nelson wrote:
> Thanks for the comments! Responses below:
>
> > It's a bit grotty spreading the checks out now. How about we tweak things
> > slightly along the lines of:
> >
> >
> > diff --git a/arch/arm64/kernel/insn.c b/arch/arm64/kernel/insn.c
> >
In the current codes, the octeontx2 uses its own method to allocate
the pool buffers, but there are some issues in this implementation.
1. We have to run the otx2_get_page() for each allocation cycle and
this is pretty error prone. As I can see there is no invocation
of the otx2_get_page() in
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/net/dsa/sja1105/sja1105_vl.c:468:6: warning: variable ‘prev_time’ set
but not used [-Wunused-but-set-variable]
u32 prev_time = 0;
^
Reported-by: Hulk Robot
Signed-off-by: Samuel Zou
---
drivers/net/dsa/sja1105/sja1105_vl.c
On 2020-05-07 13:42, Björn Töpel wrote:
From: Björn Töpel
Use the new MEM_TYPE_XSK_BUFF_POOL API in lieu of MEM_TYPE_ZERO_COPY in
mlx5e. It allows to drop a lot of code from the driver (which is now
common in AF_XDP core and was related to XSK RX frame allocation, DMA
mapping, etc.) and slightl
On 2020-05-07 13:42, Björn Töpel wrote:
From: Björn Töpel
In order to simplify AF_XDP zero-copy enablement for NIC driver
developers, a new AF_XDP buffer allocation API is added. The
implementation is based on a single core (single producer/consumer)
buffer pool for the AF_XDP UMEM.
A buffer i
On Fri, May 08, 2020 at 08:21:19AM +0200, Johan Hovold wrote:
> On Fri, May 08, 2020 at 03:35:02AM +0530, Naresh Kamboju wrote:
> > On Thu, 7 May 2020 at 16:43, Greg Kroah-Hartman
> > wrote:
> > >
> >
> > > > >
> > > > > Greg, 3f65047c853a ("of_mdio: add helper to deregister fixed-link
> > > > >
On Tue, 28 Apr 2020 14:07:46 -0700 (PDT), David Miller
wrote:
> From: Stephen Kitt
> Date: Tue, 28 Apr 2020 09:52:31 +0200
>
> > This patch changes INET_ADDR_COOKIE to declare a dummy typedef (so it
> > makes checkpatch.pl complain, sorry...)
>
> This is trading one problem for another, so in
On 2020-05-08 13:55, Maxim Mikityanskiy wrote:
On 2020-05-07 13:42, Björn Töpel wrote:
From: Björn Töpel
Use the new MEM_TYPE_XSK_BUFF_POOL API in lieu of MEM_TYPE_ZERO_COPY in
mlx5e. It allows to drop a lot of code from the driver (which is now
common in AF_XDP core and was related to XSK RX
On 2020-05-08 13:55, Maxim Mikityanskiy wrote:
On 2020-05-07 13:42, Björn Töpel wrote:
From: Björn Töpel
In order to simplify AF_XDP zero-copy enablement for NIC driver
developers, a new AF_XDP buffer allocation API is added. The
implementation is based on a single core (single producer/consum
On Fri, May 08, 2020 at 01:18:51PM +1000, Stephen Rothwell wrote:
> Hi all,
>
> Today's linux-next merge of the net-next tree got a conflict in:
>
> drivers/net/bonding/bond_main.c
>
> between commits:
>
> ed7d4f023b1a ("bonding: Rename slave_arr to usable_slaves")
> c071d91d2a89 ("bondin
It is more efficient to use kmemdup_nul() if the size is known exactly
.
According to doc:
"Note: Use kmemdup_nul() instead if the size is known exactly."
Signed-off-by: Chen Zhou
---
net/sunrpc/auth_gss/gss_rpc_upcall.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/su
Hi Florian,
Thank you so much for the review!
On Fri, 8 May 2020 at 06:16, Florian Fainelli wrote:
>
>
>
> On 5/3/2020 3:12 PM, Vladimir Oltean wrote:
> > From: Vladimir Oltean
> >
> > One way of utilizing DSA is by cascading switches which do not all have
> > compatible taggers. Consider the f
Hi Samuel,
On Fri, 8 May 2020 at 14:55, Samuel Zou wrote:
>
> Fixes gcc '-Wunused-but-set-variable' warning:
>
> drivers/net/dsa/sja1105/sja1105_vl.c:468:6: warning: variable ‘prev_time’ set
> but not used [-Wunused-but-set-variable]
> u32 prev_time = 0;
> ^
>
> Reported-by: Hulk
On 2020-05-08 15:27, Björn Töpel wrote:
On 2020-05-08 13:55, Maxim Mikityanskiy wrote:
On 2020-05-07 13:42, Björn Töpel wrote:
From: Björn Töpel
Use the new MEM_TYPE_XSK_BUFF_POOL API in lieu of MEM_TYPE_ZERO_COPY in
mlx5e. It allows to drop a lot of code from the driver (which is now
common
On Fri, May 08, 2020 at 01:08:13PM +0530, Sunil Kovvuri wrote:
> On Fri, May 8, 2020 at 11:00 AM Kevin Hao wrote:
> >
> > On Fri, May 08, 2020 at 10:18:27AM +0530, Sunil Kovvuri wrote:
> > > On Fri, May 8, 2020 at 9:43 AM Kevin Hao wrote:
> > > >
> > > > In the current codes, the octeontx2 uses i
The following commit has been merged into the perf/core branch of tip:
Commit-ID: 5ef86146de941f273d669a8e018036f549bf058c
Gitweb:
https://git.kernel.org/tip/5ef86146de941f273d669a8e018036f549bf058c
Author:Stephane Eranian
AuthorDate:Wed, 29 Apr 2020 16:14:41 -07:00
Commit
On Fri, 8 May 2020 at 15:01, Maxim Mikityanskiy wrote:
>
> On 2020-05-08 15:27, Björn Töpel wrote:
> > On 2020-05-08 13:55, Maxim Mikityanskiy wrote:
> >> On 2020-05-07 13:42, Björn Töpel wrote:
> >>> From: Björn Töpel
> >>>
> >>> Use the new MEM_TYPE_XSK_BUFF_POOL API in lieu of MEM_TYPE_ZERO_CO
The following commit has been merged into the perf/core branch of tip:
Commit-ID: 3a50dc76058d7cd8315f9c712b793d81a7ff4541
Gitweb:
https://git.kernel.org/tip/3a50dc76058d7cd8315f9c712b793d81a7ff4541
Author:Stephane Eranian
AuthorDate:Wed, 29 Apr 2020 16:14:42 -07:00
Commit
The following commit has been merged into the perf/core branch of tip:
Commit-ID: 4b1984491e65c9592f9f2bfbe8fb5bf4d4e51587
Gitweb:
https://git.kernel.org/tip/4b1984491e65c9592f9f2bfbe8fb5bf4d4e51587
Author:Ian Rogers
AuthorDate:Wed, 29 Apr 2020 16:14:40 -07:00
Committer:
On Fri, May 08, 2020 at 12:14:51PM +0300, Igor Russkikh wrote:
> Hi Jakub,
>
>
> >> This patch moves MAC generation into a separate function, which is
> > called
> >> from both places to reduce the amount of copy/paste.
> >
> > Right, but why do you have your own mac generation rather than using
On Fri, 8 May 2020 at 15:08, Björn Töpel wrote:
>
> On Fri, 8 May 2020 at 15:01, Maxim Mikityanskiy wrote:
> >
[]
>
> All zeros hints that you're probably putting in the wrong DMA address
> somewhere.
>
Hmm, I can't see that you're using xsk_buff_xdp_get_dma() anywhere in
the code. Probably it?
>>> Right, but why do you have your own mac generation rather than using
>>> eth_hw_addr_random(). You need to set NET_ADDR_RANDOM for example,
>>> just use standard helpers, please.
>>
>> We want this still be an Aquantia vendor id MAC, not a fully random mac.
>> Thats why the logic below random
Chen Zhou 於 5/8/2020 9:32 AM 寫道:
Fix sparse warning:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c:2206:5:
warning: symbol 'brcmf_p2p_get_conn_idx' was not declared. Should it be
static?
Reported-by: Hulk Robot
Signed-off-by: Chen Zhou
Reviewed-by: Wright Feng
---
dri
On Fri, May 08, 2020 at 04:22:40PM +0300, Igor Russkikh wrote:
>
> >>> Right, but why do you have your own mac generation rather than using
> >>> eth_hw_addr_random(). You need to set NET_ADDR_RANDOM for example,
> >>> just use standard helpers, please.
> >>
> >> We want this still be an Aquantia
On Thu, May 7, 2020 at 9:18 PM Kelly Littlepage wrote:
>
> The stated intent of the original commit is to is to "return the timestamp
> corresponding to the highest sequence number data returned." The current
> implementation returns the timestamp for the last byte of the last fully
> read skb, wh
On Thu, May 7, 2020 at 8:52 PM Cong Wang wrote:
>
> On Tue, May 5, 2020 at 1:46 AM Václav Zindulka
> wrote:
> >
> > On Mon, May 4, 2020 at 7:46 PM Cong Wang wrote:
> > >
> > > Sorry for the delay. I lost connection to my dev machine, I am trying
> > > to setup this on my own laptop.
> >
> > Sorr
Björn Töpel writes:
> On Thu, 7 May 2020 at 20:08, John Fastabend wrote:
>>
> []
>>
>> I'm wondering if we can teach the verifier to recognize tail calls,
>>
>> int xdp_prog1(struct xdp_md *ctx)
>> {
>> return xdp_do_redirect(ctx, &xsks_map, 0);
>> }
>>
>> This would be useful for normal
Hi,
Here is the 2nd version of the series to add nested-kprobes support
to x86, arm64 and arm. This makes kprobes to accept 1-level nesting
instead of incrementing missed count.
In this version, I fixed a mistake for the kprobes on ftrace on x86
and dump nested probes when we detect an unrecovera
Make kprobes to accept 1-level nesting instead of missing it
on arm.
Any kprobes hits in kprobes pre/post handler context can be
nested at once. If the other kprobes hits in the nested pre/post
handler context, that will be missed.
We can test this feature on the kernel with
CONFIG_KPROBE_EVENTS_
The stated intent of the original commit is to is to "return the timestamp
corresponding to the highest sequence number data returned." The current
implementation returns the timestamp for the last byte of the last fully
read skb, which is not necessarily the last byte in the recv buffer. This
patc
Make kprobes to accept 1-level nesting instead of missing it.
Any kprobes hits in kprobes pre/post handler context
can be nested at once. If the other kprobes hits in
the nested pre/post handler context, that will be missed.
This is useful if user wants to trace/debug the code
which can be invoke
1 - 100 of 328 matches
Mail list logo
