it, remove the socket/umem in xdpsock's error path.
Signed-off-by: Maciej Fijalkowski
---
samples/bpf/xdpsock_user.c | 33 ++---
tools/lib/bpf/xsk.c| 32
tools/lib/bpf/xsk.h| 1 +
3 files changed, 43 insertions(+
ue per CPU core.
For xsk counter we could have for example additional entry in qidconf_map, but
that map is removed in Jonathan's patches, so that's not an option. The
resolution I gave a shot is to have a pinned map with a single entry. Further
reasoning is included in commit messag
for having this preparation done after bind().
Signed-off-by: Maciej Fijalkowski
Signed-off-by: Krzysztof Kazimierczak
---
samples/bpf/xdpsock_user.c | 15 ---
tools/lib/bpf/xsk.c| 19 ++-
2 files changed, 18 insertions(+), 16 deletions(-)
diff --git a/s
ap in nr_map_ids field of struct bpf_prog_info. The
described behaviour brings the need to have this map pinned, so in
case when socket is being created and the libbpf detects the presence of
bpf resources, it will be able to access that map.
Signed-off-by: Maciej Fijalkowski
---
tools/lib/bpf/xsk.c
or channels.max_{t,r}x as well in
xsk_get_max_queues.
Signed-off-by: Maciej Fijalkowski
---
tools/lib/bpf/xsk.c | 18 ++
1 file changed, 10 insertions(+), 8 deletions(-)
diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c
index 57dda1389870..514ab3fb06f4 100644
--- a/tools/l
On Tue, 4 Jun 2019 10:06:36 +0200
Björn Töpel wrote:
> On 2019-06-03 15:19, Maciej Fijalkowski wrote:
> > Let's get into the driver via ndo_bpf with command set to XDP_SETUP_UMEM
> > with fill queue that already contains some available entries that can be
> > used
On Tue, 4 Jun 2019 10:06:57 +0200
Björn Töpel wrote:
> On 2019-06-03 15:19, Maciej Fijalkowski wrote:
> > When it comes down to ethtool's get channels API, various drivers are
> > reporting the queue count in two ways - they are setting max_combined or
> > max_tx/max_rx
On Tue, 4 Jun 2019 10:07:25 +0200
Björn Töpel wrote:
>
> On 2019-06-03 15:19, Maciej Fijalkowski wrote:
> > Since xsk support in libbpf loads the xdp program interface, make it
> > also responsible for its removal. Store the prog id in xsk_socket_config
> > so when rem
On Tue, 4 Jun 2019 10:08:03 +0200
Björn Töpel wrote:
> On 2019-06-03 15:19, Maciej Fijalkowski wrote:
> > In case where multiple xsk sockets are attached to a single interface
> > and one of them gets detached, the eBPF maps and program are removed.
> > This should not
On Wed, 7 Aug 2019 10:02:04 -0700
Y Song wrote:
> On Tue, Aug 6, 2019 at 7:25 PM Daniel T. Lee wrote:
> >
> > By this commit, using `bpftool net detach`, the attached XDP prog can
> > be detached. Detaching the BPF prog will be done through libbpf
> > 'bpf_set_link_xdp_fd' with the progfd set to
On Wed, 7 Aug 2019 13:12:17 -0700
Y Song wrote:
> On Wed, Aug 7, 2019 at 11:30 AM Maciej Fijalkowski
> wrote:
> >
> > On Wed, 7 Aug 2019 10:02:04 -0700
> > Y Song wrote:
> >
> > > On Tue, Aug 6, 2019 at 7:25 PM Daniel T. Lee
> > > wrote:
&g
On Thu, Dec 31, 2020 at 12:16:41AM +0100, Daniel Borkmann wrote:
> On 12/29/20 7:09 PM, Lorenzo Bianconi wrote:
> > > > + hard_start = page_address(rx_buffer->page) +
> > > > + rx_buffer->page_offset - offset;
> > > > + xdp_pr
allocation loop
Maciej Fijalkowski (10):
i40e: drop redundant check when setting xdp prog
i40e: drop misleading function comments
i40e: adjust i40e_is_non_eop
ice: simplify ice_run_xdp
ice: move skb pointer from rx_buf to rx_ring
ice: remove redundant checks in ice_change_mtu
ice: skip NULL
Net core handles the case where netdev has no xdp prog attached and
current prog is NULL. Therefore, remove such check within
i40e_xdp_setup.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 3 ---
1 file changed, 3 deletions(-)
diff
There's no need for 'result' variable, we can directly return the
internal status based on action returned by xdp prog.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/ice_txrx.c | 15 +--
1 file changed, 5 insertions(+
n is limited to what the name actually
states.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/i40e/i40e_txrx.c | 23 ++---
1 file changed, 6 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
b/drive
From: Björn Töpel
Fold the count decrement into the while-statement.
Reviewed-by: Maciej Fijalkowski
Signed-off-by: Björn Töpel
---
drivers/net/ethernet/intel/i40e/i40e_xsk.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
b
;rx_offset' field to ice_ring that is meant to hold the
ice_rx_offset() result and use it within ice_clean_rx_irq().
Furthermore, use it within ice_alloc_mapped_page().
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/
ives in .c file.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/i40e/i40e_txrx.c | 35 +++--
drivers/net/ethernet/intel/i40e/i40e_txrx.h | 1 +
2 files changed, 19 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e
ut the inlining.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 1 +
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 15 ---
2 files changed, 9 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/intel/i
dev_validate_mtu checks that mtu value specified by user is not less
than min mtu and not greater than max allowed mtu. It is being done
before calling the ndo_change_mtu exposed by driver, so remove these
redundant checks in ice_change_mtu.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/ice_xsk.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c
b/drivers/net/ethernet/intel/ice/ice_xsk.c
index 1782146db644..b926ca9c4f4f 100644
--- a/drivers/net
i40e_cleanup_headers has a statement about check against skb being
linear or not which is not relevant anymore, so let's remove it.
Same case for i40e_can_reuse_rx_page, it references things that are not
present there anymore.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalk
Similar thing has been done in i40e, as there is no real need for having
the sk_buff pointer in each rx_buf. Non-eop frames can be simply handled
on that pointer moved upwards to rx_ring.
Reviewed-by: Björn Töpel
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/ice_txrx.c
On Fri, Jan 15, 2021 at 05:36:23PM +0100, Eelco Chaudron wrote:
>
>
> On 16 Dec 2020, at 15:08, Eelco Chaudron wrote:
>
> > On 15 Dec 2020, at 19:06, Maciej Fijalkowski wrote:
> >
> > > On Tue, Dec 15, 2020 at 02:28:39PM +0100, Eelco Chaudron wrote:
> >
On Wed, Jan 20, 2021 at 04:57:08PM +0100, Jesper Dangaard Brouer wrote:
> On Wed, 20 Jan 2021 15:15:22 +0200
> Maxim Mikityanskiy wrote:
>
> > On 2021-01-19 17:50, Björn Töpel wrote:
> > > This series extends bind() for XDP sockets, so that the bound socket
> > > is added to the netdev_rx_queue _
On Wed, Jan 20, 2021 at 10:25:09AM +0800, Hangbin Liu wrote:
> From: Jesper Dangaard Brouer
>
> This changes the devmap XDP program support to run the program when the
> bulk queue is flushed instead of before the frame is enqueued. This has
> a couple of benefits:
>
> - It "sorts" the packets b
On Thu, Jan 21, 2021 at 11:54:24AM +0800, Hangbin Liu wrote:
> Hi Maciej,
> On Wed, Jan 20, 2021 at 11:42:38PM +0100, Maciej Fijalkowski wrote:
> > > +static int dev_map_bpf_prog_run(struct bpf_prog *xdp_prog,
> > > + struct x
On Thu, Jan 21, 2021 at 06:11:30PM +0100, Jesper Dangaard Brouer wrote:
> On Wed, 20 Jan 2021 13:27:59 -0800
> Ivan Babrou wrote:
>
> > Without this change the driver tries to allocate too many queues,
> > breaching the number of available msi-x interrupts on machines
> > with many logical cpus a
On Fri, Jan 22, 2021 at 09:59:53AM +0100, Magnus Karlsson wrote:
> On Thu, Jan 21, 2021 at 6:07 PM Jesper Dangaard Brouer
> wrote:
> >
> > On Wed, 20 Jan 2021 17:19:31 +0100
> > Maciej Fijalkowski wrote:
> >
> > > On Wed, Jan 20, 2021 at 04:57:08PM +0100, J
+ }
> + return nframes; /* sent frames count */
> +}
> +
> static void bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
> {
> struct net_device *dev = bq->dev;
> - int sent = 0, drops = 0, err = 0;
> + unsigned int cnt = bq->count;
> +
On Thu, Dec 10, 2020 at 04:50:42PM +0100, Lorenzo Bianconi wrote:
> Introduce xdp_init_buff utility routine to initialize xdp_buff data
> structure. Rely on xdp_init_buff in all XDP capable drivers.
Hm, Jesper was suggesting two helpers, one that you implemented for things
that are set once per NA
leaned frames (16).
You also have a typo in subject (alloction).
Is spinning a v2 worth it?
>
> This change boosts the xdpsock rxdrop scenario with 15% more
> packets-per-second.
>
> Reviewed-by: Maciej Fijalkowski
> Signed-off-by: Björn Töpel
> ---
> dri
On Thu, Dec 10, 2020 at 05:32:41PM +0100, Lorenzo Bianconi wrote:
> > On Thu, Dec 10, 2020 at 04:50:42PM +0100, Lorenzo Bianconi wrote:
> > > Introduce xdp_init_buff utility routine to initialize xdp_buff data
> > > structure. Rely on xdp_init_buff in all XDP capable drivers.
> >
> > Hm, Jesper wa
On Thu, Dec 10, 2020 at 06:59:45PM +0100, Lorenzo Bianconi wrote:
> On Dec 10, Maciej Fijalkowski wrote:
> > On Thu, Dec 10, 2020 at 05:32:41PM +0100, Lorenzo Bianconi wrote:
> > > > On Thu, Dec 10, 2020 at 04:50:42PM +0100, Lorenzo Bianconi wrote:
> > > > > Int
Hi,
This series is mostly about the cleanups on Rx (ZC/normal) paths both in
ice and i40e drivers. Things that stand out are the simplifactions of
ice_change_mtu and i40e_xdp_setup.
Thanks!
Björn Töpel (1):
i40e, xsk: Simplify the do-while allocation loop
Maciej Fijalkowski (7):
i40e: drop
Net core handles the case where netdev has no xdp prog attached and
current prog is NULL. Therefore, remove such check within
i40e_xdp_setup.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/net
i40e_cleanup_headers has a statement about check against skb being
linear or not which is not relevant anymore, so let's remove it.
Same case for i40e_can_reuse_rx_page, it references things that are not
present there anymore.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/
Whole zero-copy variant of clean Rx irq is executed when xsk_pool is
attached to rx_ring and it can happen only when XDP program is present
on interface. Therefore it is safe to assume that program is always
!NULL and there is no need for checking it in ice_run_xdp_zc.
Signed-off-by: Maciej
n is limited to what the name actually
states.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/i40e/i40e_txrx.c | 23 ++---
1 file changed, 6 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
b/drivers/net/ethernet/intel/i
Similar thing had been done in i40e, as there is no real need for having
the sk_buff pointer in each rx_buf. Non-eop frames can be simply handled
on that pointer moved upwards to rx_ring.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/ice_txrx.c | 28
From: Björn Töpel
Fold the count decrement into the while-statement.
Signed-off-by: Björn Töpel
---
drivers/net/ethernet/intel/i40e/i40e_xsk.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
b/drivers/net/ethernet/intel/i40e/i4
dev_validate_mtu checks that mtu value specified by user is not less
than min mtu and not greater than max allowed mtu. It is being done
before calling the ndo_change_mtu exposed by driver, so remove these
redundant checks in ice_change_mtu.
Signed-off-by: Maciej Fijalkowski
---
drivers/net
There's no need for 'result' variable, we can directly return the
internal status based on action returned by xdp prog.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/ice_txrx.c | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff
On Mon, Dec 14, 2020 at 09:51:27AM +0100, Magnus Karlsson wrote:
> From: Magnus Karlsson
>
> Fix a possible memory leak when a bind of an AF_XDP socket fails. When
> the fill and completion rings are created, they are tied to the
> socket. But when the buffer pool is later created at bind time, t
On Fri, Dec 11, 2020 at 08:23:15PM -0800, Jakub Kicinski wrote:
> On Fri, 11 Dec 2020 17:49:53 +0100 Maciej Fijalkowski wrote:
> > @@ -864,14 +865,12 @@ ice_reuse_rx_page(struct ice_ring *rx_ring, struct
> > ice_rx_buf *old_buf)
> > * for use by the CPU.
> > */
&g
Maciej Fijalkowski (7):
i40e: drop redundant check when setting xdp prog
i40e: drop misleading function comments
i40e: adjust i40e_is_non_eop
ice: simplify ice_run_xdp
ice: move skb pointer from rx_buf to rx_ring
ice: remove redundant checks in ice_change_mtu
ice: skip NULL check against
Net core handles the case where netdev has no xdp prog attached and
current prog is NULL. Therefore, remove such check within
i40e_xdp_setup.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/net
i40e_cleanup_headers has a statement about check against skb being
linear or not which is not relevant anymore, so let's remove it.
Same case for i40e_can_reuse_rx_page, it references things that are not
present there anymore.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/
n is limited to what the name actually
states.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/i40e/i40e_txrx.c | 23 ++---
1 file changed, 6 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
b/drivers/net/ethernet/intel/i
Similar thing has been done in i40e, as there is no real need for having
the sk_buff pointer in each rx_buf. Non-eop frames can be simply handled
on that pointer moved upwards to rx_ring.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/ice_txrx.c | 29
There's no need for 'result' variable, we can directly return the
internal status based on action returned by xdp prog.
Signed-off-by: Maciej Fijalkowski
---
drivers/net/ethernet/intel/ice/ice_txrx.c | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff
Whole zero-copy variant of clean Rx irq is executed when xsk_pool is
attached to rx_ring and it can happen only when XDP program is present
on interface. Therefore it is safe to assume that program is always
!NULL and there is no need for checking it in ice_run_xdp_zc.
Signed-off-by: Maciej
From: Björn Töpel
Fold the count decrement into the while-statement.
Signed-off-by: Björn Töpel
---
drivers/net/ethernet/intel/i40e/i40e_xsk.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
b/drivers/net/ethernet/intel/i40e/i4
dev_validate_mtu checks that mtu value specified by user is not less
than min mtu and not greater than max allowed mtu. It is being done
before calling the ndo_change_mtu exposed by driver, so remove these
redundant checks in ice_change_mtu.
Signed-off-by: Maciej Fijalkowski
---
drivers/net
On Sat, Dec 12, 2020 at 06:41:49PM +0100, Lorenzo Bianconi wrote:
> Introduce xdp_prepare_buff utility routine to initialize per-descriptor
> xdp_buff fields (e.g. xdp_buff pointers). Rely on xdp_prepare_buff() in
> all XDP capable drivers.
>
> Signed-off-by: Lorenzo Bianconi
> ---
> drivers/net
On Tue, Dec 15, 2020 at 04:06:20PM +0100, Lorenzo Bianconi wrote:
> > On 12/15/20 2:47 PM, Lorenzo Bianconi wrote:
> > [...]
> > > > > diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
> > > > > index 329397c60d84..61d3f5f8b7f3 100644
> > > > > --- a/drivers/net/xen-netfront.c
>
On Tue, Dec 15, 2020 at 02:28:39PM +0100, Eelco Chaudron wrote:
>
>
> On 9 Dec 2020, at 13:07, Eelco Chaudron wrote:
>
> > On 9 Dec 2020, at 12:10, Maciej Fijalkowski wrote:
>
>
>
> > > > > > +
> > > > > > + ctx_re
On Thu, Dec 17, 2020 at 10:16:06AM -0800, Saeed Mahameed wrote:
> On Wed, 2020-12-16 at 16:01 +0100, Lorenzo Bianconi wrote:
> > > On Tue, 15 Dec 2020 14:47:10 +0100
> > > Lorenzo Bianconi wrote:
> > >
> > > > [...]
> > > > > > xdp_act = bpf_prog_run_xdp(xdp_prog, &xdp);
> > > > > > diff --gi
space. This provides around 100k extra packets in throughput on my
> server when running l2fwd in xdpsock.
>
> Thanks: Magnus
For series:
Acked-by: Maciej Fijalkowski
You only ate 'e' from i40e subject line.
>
> Magnus Karlsson (3):
> i40: optimize for XDP_REDI
On Fri, Dec 04, 2020 at 01:18:31PM +0100, Toke Høiland-Jørgensen wrote:
> alar...@gmail.com writes:
>
> > From: Marek Majtyka
> >
> > Implement support for checking what kind of xdp functionality a netdev
> > supports. Previously, there was no way to do this other than to try
> > to create an AF_
On Fri, Dec 04, 2020 at 11:28:57AM +0100, alar...@gmail.com wrote:
> From: Marek Majtyka
>
> Implement support for checking what kind of xdp functionality a netdev
> supports. Previously, there was no way to do this other than to try
> to create an AF_XDP socket on the interface or load an XDP pr
On Fri, Dec 04, 2020 at 02:11:12PM +0200, akiy...@amazon.com wrote:
> From: Arthur Kiyanovski
>
> Rename the ena_xdp_xmit_buff() function to ena_xdp_xmit_frame() and pass
> it an xdp_frame struct instead of xdp_buff.
> This change lays the ground for XDP redirect implementation which uses
> xdp_f
On Fri, Dec 04, 2020 at 02:11:15PM +0200, akiy...@amazon.com wrote:
> From: Arthur Kiyanovski
>
> This patch implements the ndo_xdp_xmit() net_device function which is
> called when a packet is redirected to this driver using an
> XDP_REDIRECT directive.
>
> The function receives an array of xdp
On Mon, Dec 07, 2020 at 01:15:00PM -0800, Alexander Duyck wrote:
> On Mon, Dec 7, 2020 at 8:36 AM Lorenzo Bianconi wrote:
> >
> > Initialize multi-buffer bit (mb) to 0 in all XDP-capable drivers.
> > This is a preliminary patch to enable xdp multi-buffer support.
> >
> > Signed-off-by: Lorenzo Bia
On Mon, Dec 07, 2020 at 12:52:22PM -0800, John Fastabend wrote:
> Jesper Dangaard Brouer wrote:
> > On Fri, 4 Dec 2020 16:21:08 +0100
> > Daniel Borkmann wrote:
> >
> > > On 12/4/20 1:46 PM, Maciej Fijalkowski wrote:
> > > > On Fri, Dec 04, 2020 at 01:1
On Mon, Dec 07, 2020 at 05:32:42PM +0100, Lorenzo Bianconi wrote:
> From: Eelco Chaudron
>
> This patch adds a new field to the XDP context called frame_length,
> which will hold the full length of the packet, including fragments
> if existing.
The approach you took for ctx access conversion is
On Tue, Dec 08, 2020 at 10:03:51PM -0800, John Fastabend wrote:
> > On Mon, Dec 07, 2020 at 12:52:22PM -0800, John Fastabend wrote:
> > > Jesper Dangaard Brouer wrote:
> > > > On Fri, 4 Dec 2020 16:21:08 +0100
> > > > Daniel Borkmann wrote:
>
> [...] pruning the thread to answer Jesper.
I think
On Wed, Dec 09, 2020 at 11:35:13AM +0100, Eelco Chaudron wrote:
>
>
> On 8 Dec 2020, at 23:17, Maciej Fijalkowski wrote:
>
> > On Mon, Dec 07, 2020 at 05:32:42PM +0100, Lorenzo Bianconi wrote:
> > > From: Eelco Chaudron
> > >
> > > This pat
On Thu, Nov 19, 2020 at 10:34:48PM +, Ramsay, Lincoln wrote:
> When performing IPv6 forwarding, there is an expectation that SKBs
> will have some headroom. When forwarding a packet from the aquantia
> driver, this does not always happen, triggering a kernel warning.
>
> The build_skb path fai
On Thu, Nov 19, 2020 at 06:29:33PM +0200, Camelia Groza wrote:
> Use an xdp_frame structure for managing the frame. Store a backpointer to
> the structure at the start of the buffer before enqueueing. Use the XDP
> API for freeing the buffer when it returns to the driver on the TX
> confirmation pa
On Thu, Nov 19, 2020 at 06:29:31PM +0200, Camelia Groza wrote:
> Implement the XDP_DROP and XDP_PASS actions.
>
> Avoid draining and reconfiguring the buffer pool at each XDP
> setup/teardown by increasing the frame headroom and reserving
> XDP_PACKET_HEADROOM bytes from the start. Since we always
On Thu, Nov 19, 2020 at 03:22:42PM -0800, Andrii Nakryiko wrote:
> Teach libbpf to search for candidate types for CO-RE relocations across kernel
> modules BTFs, in addition to vmlinux BTF. If at least one candidate type is
> found in vmlinux BTF, kernel module BTFs are not iterated. If vmlinux BTF
On Thu, Nov 19, 2020 at 04:22:24PM -0800, David Awogbemila wrote:
> On Thu, Nov 19, 2020 at 12:21 PM Saeed Mahameed wrote:
> >
> > On Wed, 2020-11-18 at 15:20 -0800, David Awogbemila wrote:
> > > From: Catherine Sullivan
> > >
> > > Add support to describe device for parsing device options. As
>
On Thu, Nov 19, 2020 at 05:24:43PM -0800, Andrii Nakryiko wrote:
> On Thu, Nov 19, 2020 at 4:55 PM Maciej Fijalkowski
> wrote:
> >
> > On Thu, Nov 19, 2020 at 03:22:42PM -0800, Andrii Nakryiko wrote:
> > > Teach libbpf to search for candidate types for CO-RE reloc
On Fri, Nov 20, 2020 at 11:18:34AM +0300, Igor Russkikh wrote:
>
>
> On 20/11/2020 1:49 am, Maciej Fijalkowski wrote:
> > External Email
> >
> > --
> > On Thu, Nov 19, 2020 at 10:34:48PM +,
On Fri, Nov 20, 2020 at 06:50:28PM +, Camelia Alexandra Groza wrote:
> > -Original Message-
> > From: Maciej Fijalkowski
> > Sent: Friday, November 20, 2020 02:19
> > To: Camelia Alexandra Groza
> > Cc: k...@kernel.org; bro...@redhat.com; sa...@kerne
On Fri, Nov 20, 2020 at 06:54:42PM +, Camelia Alexandra Groza wrote:
> > -Original Message-
> > From: Maciej Fijalkowski
> > Sent: Friday, November 20, 2020 01:51
> > To: Camelia Alexandra Groza
> > Cc: k...@kernel.org; bro...@redhat.com; sa...@kerne
On Mon, Nov 23, 2020 at 07:36:20PM +0200, Camelia Groza wrote:
> Implement the XDP_DROP and XDP_PASS actions.
>
> Avoid draining and reconfiguring the buffer pool at each XDP
> setup/teardown by increasing the frame headroom and reserving
> XDP_PACKET_HEADROOM bytes from the start. Since we always
tu + VLAN_ETH_HLEN + ETH_FCS_LEN > max_contig_data) {
Do you support VLAN double tagging? We normally take into acount to two vlan
headers in these checks.
Other than that:
Reviewed-by: Maciej Fijalkowski
> + dev_warn(priv->net_dev->dev.parent,
> +
n a word?) in here?
One question below and:
Reviewed-by: Maciej Fijalkowski
>
> This approach will be reused for XDP REDIRECT.
>
> Acked-by: Madalin Bucur
> Signed-off-by: Camelia Groza
> ---
> Changes in v4:
> - call xdp_rxq_info_is_reg() before unregistering
>
is implemented at this time.
>
> Acked-by: Madalin Bucur
> Signed-off-by: Camelia Groza
> ---
> drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 48
> +-
> drivers/net/ethernet/freescale/dpaa/dpaa_eth.h | 1 +
> 2 files changed, 48 insertions(+)
On Mon, Nov 23, 2020 at 07:36:25PM +0200, Camelia Groza wrote:
> For XDP TX, even tough we start out with correctly aligned buffers, the
> XDP program might change the data's alignment. For REDIRECT, we have no
> control over the alignment either.
>
> Create a new workaround for xdp_frame structur
On Wed, Nov 25, 2020 at 03:49:45PM +, Camelia Alexandra Groza wrote:
> > -Original Message-
> > From: Maciej Fijalkowski
> > Sent: Tuesday, November 24, 2020 21:52
> > To: Camelia Alexandra Groza
> > Cc: k...@kernel.org; bro...@redhat.com; sa...@kerne
On Wed, Nov 25, 2020 at 03:52:33PM +, Camelia Alexandra Groza wrote:
> > -Original Message-
> > From: Maciej Fijalkowski
> > Sent: Tuesday, November 24, 2020 22:51
> > To: Camelia Alexandra Groza
> > Cc: k...@kernel.org; bro...@redhat.com; sa...@kerne
t would lead to a double free of the fq resources in 4/7
>
> Changes in v5:
> - report errors on XDP setup with extack in 2/7
> - checkpath fix in 4/7
> - add more details in the commit message in 4/7
> - reduce the impact of the A050385 erratum workaround code on non-erratum
>
On Fri, Nov 27, 2020 at 05:35:00PM +, Camelia Alexandra Groza wrote:
> > -Original Message-
> > From: Maciej Fijalkowski
> > Sent: Friday, November 27, 2020 16:44
> > To: Camelia Alexandra Groza
> > Cc: k...@kernel.org; bro...@redhat.com; sa...@kerne
On Tue, Nov 10, 2020 at 03:25:10PM +0100, Jesper Dangaard Brouer wrote:
> On Tue, 10 Nov 2020 20:46:39 +0800
> Hangbin Liu wrote:
>
> > This patch add running xdp program on egress interface support for
> > xdp_redirect_map sample. The new prog will change the IP ttl based
> > on egress ifindex.
On Sat, Oct 17, 2020 at 09:12:38AM +0200, sven.auha...@voleatech.de wrote:
> From: Sven Auhagen
>
> Since we share the transmit queue with the slow path,
> it is possible that we run into a transmit queue timeout.
> This will reset the queue.
> This happens under high load when the fast path is u
eue timeout
> that can occur when using xdp.
>
> Signed-off-by: Sven Auhagen
Sorry for not getting back at v1 discussion, I took some time off.
For the series:
Acked-by: Maciej Fijalkowski
Couple nits:
- you don't need SOB line within cover letter, I suppose
- next time please
On Wed, Oct 21, 2020 at 01:07:12AM -0500, Lijun Pan wrote:
> Jakub Kicinski brought up a concern in ibmvnic_set_mac().
> ibmvnic_set_mac() does this:
>
> ether_addr_copy(adapter->mac_addr, addr->sa_data);
> if (adapter->state != VNIC_PROBED)
> rc = __ibmvnic_set_mac(netde
On Wed, Nov 04, 2020 at 03:09:02PM +0100, Magnus Karlsson wrote:
> From: Magnus Karlsson
>
> Use the new batched xsk interfaces for the Tx path in the i40e driver
> to improve performance. On my machine, this yields a throughput
> increase of 4% for the l2fwd sample app in xdpsock. If we instead
On Thu, Nov 09, 2023 at 07:11:46PM +0800, Xuan Zhuo wrote:
> On Thu, 9 Nov 2023 03:12:27 -0500, "Michael S. Tsirkin"
> wrote:
> > On Tue, Nov 07, 2023 at 11:12:22AM +0800, Xuan Zhuo wrote:
> > > Implement the logic of filling rq with XSK buffers.
> > >
> > > Signed-off-by: Xuan Zhuo
> > > ---
>
On Fri, Nov 10, 2023 at 10:38:04AM +0800, Xuan Zhuo wrote:
> On Thu, 9 Nov 2023 17:26:33 +0100, Maciej Fijalkowski
> wrote:
> > On Thu, Nov 09, 2023 at 07:11:46PM +0800, Xuan Zhuo wrote:
> > > On Thu, 9 Nov 2023 03:12:27 -0500, "Michael S. Tsirkin"
> > >
On Tue, Nov 07, 2023 at 11:12:24AM +0800, Xuan Zhuo wrote:
> The virtnet_xdp_handler() is re-used. But
>
> 1. We need to copy data to create skb for XDP_PASS.
> 2. We need to call xsk_buff_free() to release the buffer.
> 3. The handle for xdp_buff is difference.
>
> If we pushed this logic into e
...@intel.com/
In the meantime bpf_link was introduced and it seems that it can address
the issue of refcounting the XDP prog on interface. More info on commit
messages.
Thanks.
Maciej Fijalkowski (3):
libbpf: xsk: use bpf_link
libbpf: clear map_info before each bpf_obj_get_info_by_fd
samples: bpf
e, bail out or update the bpf_link's prog given the presence of
XDP_FLAGS_UPDATE_IF_NOEXIST.
Signed-off-by: Maciej Fijalkowski
---
tools/lib/bpf/xsk.c | 143 +---
1 file changed, 122 insertions(+), 21 deletions(-)
diff --git a/tools/lib/bpf/xsk.c b/too
no more
active users.
While at it, don't dump stats on error path.
Signed-off-by: Maciej Fijalkowski
---
samples/bpf/xdpsock_user.c | 55 ++
1 file changed, 14 insertions(+), 41 deletions(-)
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_
;s convert it to strncmp and provide the
length of the array name that we're looking for.
Signed-off-by: Maciej Fijalkowski
---
tools/lib/bpf/xsk.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c
index 5911868efa43..fb259c0bba
On Mon, Feb 15, 2021 at 08:35:29PM +0100, Toke Høiland-Jørgensen wrote:
> Björn Töpel writes:
>
> > On 2021-02-15 18:07, Toke Høiland-Jørgensen wrote:
> >> Maciej Fijalkowski writes:
> >>
> >>> Currently, if there are multiple xdpsock instances runni
1 - 100 of 384 matches
Mail list logo
