On Mon, Jun 9, 2025 at 7:51 AM Christoph Hellwig wrote:
>
> Before you add new features to vduse, please remove the broken abuse of
> the DMA API first. Without that no new feature work should go into
> this code.
>
Hi Christoph,
This code does not touch the DMA API at all. Actually, we could e
On Tue, Jun 17, 2025 at 2:18 AM wrote:
>
> From: "Dr. David Alan Gilbert"
>
> Hi,
> The following pair of patches remove a bunch of small functions
> that have been unused for a long time.
>
Acked-by: Eugenio Pérez
Thanks!
On Mon, Jun 16, 2025 at 10:25 AM Jason Wang wrote:
>
> Hello all:
>
> This sereis tries to implement the VIRTIO_F_IN_ORDER to
> virtio_ring. This is done by introducing virtqueue ops so we can
> implement separate helpers for different virtqueue layout/features
> then the in-order were implemented
On Thu, Jun 12, 2025 at 2:33 AM Jason Wang wrote:
>
> On Fri, Jun 6, 2025 at 7:50 PM Eugenio Pérez wrote:
> >
> > Is the expected behavior with vdpa_sim and mlx.
> >
> > Signed-off-by: Eugenio Pérez
> > ---
> > drivers/vdpa/vdpa_user/vduse_dev.c | 6 ++
> > 1 file changed, 6 insertions(+)
>
On Thu, Jun 12, 2025 at 2:30 AM Jason Wang wrote:
>
> On Fri, Jun 6, 2025 at 7:50 PM Eugenio Pérez wrote:
> >
> > Make one IOTLB domain per address space, and allow the driver to assign
> > each ASID to a vq group. Each address space via an dedicated identifier
> > (ASID).
> >
> > During vDPA de
On Mon, Jun 9, 2025 at 3:50 AM Jason Wang wrote:
>
> On Mon, Jun 9, 2025 at 9:41 AM Jason Wang wrote:
> >
> > On Fri, Jun 6, 2025 at 7:50 PM Eugenio Pérez wrote:
> > >
> > > This allows to define all functions checking the API version set by the
> > > userland device.
> > >
> > > Signed-off-by:
On Mon, Jun 9, 2025 at 3:55 AM Jason Wang wrote:
>
> On Fri, Jun 6, 2025 at 7:50 PM Eugenio Pérez wrote:
> >
> > The virtqueue group is the minimal set of virtqueues that must share an
> > address space. And the address space identifier could only be attached
> > to a specific virtqueue group.
On Thu, May 29, 2025 at 9:30 AM Viresh Kumar wrote:
>
> The virtio vdpa implementation creates a list of virtqueues, while the
> same is already available in the struct virtio_device.
>
> This list is never traversed though, and only the pointer to the struct
> virtio_vdpa_vq_info is used in the c
On Wed, May 28, 2025 at 8:42 AM Jason Wang wrote:
>
> Hello all:
>
> This sereis tries to implement the VIRTIO_F_IN_ORDER to
> virtio_ring. This is done by introducing virtqueue ops so we can
> implement separate helpers for different virtqueue layout/features
> then the in-order were implemented
On 5/19/25 3:04 AM, Matthieu Baerts wrote:
+SEC("cgroup/getsockopt")
+int iters_subflow(struct bpf_sockopt *ctx)
+{
+ struct mptcp_subflow_context *subflow;
+ struct bpf_sock *sk = ctx->sk;
+ struct sock *ssk = NULL;
+ struct mptcp_sock *msk;
+ int local_ids = 0;
+
+ if (ctx->le
() can now return NULL.
- v3:
- Use bpf_core_cast to get the msk instead of bpf_skc_to_mptcp_sock.
- Drop bpf_mptcp_sock_acquire and bpf_mptcp_sock_release (Martin).
- Adapt the commit message accordingly.
- Remove no longer needed export to the mptcp_bpf.h file and adapt
bpf_iter_mpt
this:
mptcp_for_each_subflow(msk, subflow)
kfunc(subflow);
But in the MPTCP BPF program, this has not yet been implemented. As
Martin suggested recently, this conn_list walking + modify-by-kfunc
usage fits the bpf_iter use case.
So this patch adds a new bpf_iter type named "mptcp_subflow&q
On Mon, May 12, 2025 at 5:21 PM Jon Kohler wrote:
>
>
>
> > On Apr 30, 2025, at 9:21 PM, Jon Kohler wrote:
> >
> >
> >
> >> On Apr 16, 2025, at 6:15 AM, Eugenio Perez Martin
> >> wrote:
> >>
> >> !
On 4/22/25 7:21 PM, Jiayuan Chen wrote:
April 23, 2025 at 08:13, "Martin KaFai Lau" wrote:
On 4/16/25 9:40 PM, Jiayuan Chen wrote:
when we attach a prog without cgroup_storage map being used,
cgroup_storage in struct bpf_prog_array_item is empty. Then, if we use
BPF_LINK
On 4/16/25 9:40 PM, Jiayuan Chen wrote:
when we attach a prog without cgroup_storage map being used,
cgroup_storage in struct bpf_prog_array_item is empty. Then, if we use
BPF_LINK_UPDATE to replace old prog with a new one that uses the
cgroup_storage map, we miss cgroup_storage being initiated.
On 4/22/25 11:23 AM, Prabhav Kumar Vaish wrote:
Ensure that 'tcp' is checked for NULL before dereferencing. This resolves
a potential null pointer dereference warning reported by static analysis.
Signed-off-by: Prabhav Kumar Vaish
---
tools/testing/selftests/bpf/progs/skb_pkt_end.c | 4 ++--
On Tue, Apr 8, 2025 at 8:28 AM Jason Wang wrote:
>
> On Tue, Apr 8, 2025 at 9:18 AM Jon Kohler wrote:
> >
> >
> >
> > > On Apr 6, 2025, at 7:14 PM, Jason Wang wrote:
> > >
> > > !---|
> > > CAUTION: External Email
> > >
> > > |
On Mon, Mar 24, 2025 at 7:01 AM Jason Wang wrote:
>
> Factor out core logic of buffer detaching and leave the id population
> to the caller so in order can just call the core logic.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c | 17 +--
On Mon, Mar 24, 2025 at 7:01 AM Jason Wang wrote:
>
> This patch introduces virtqueue ops which is a set of the callbacks
> that will be called for different queue layout or features. This would
> help to avoid branches for split/packed and will ease the future
> implementation like in order.
>
A
On Mon, Mar 24, 2025 at 7:02 AM Jason Wang wrote:
>
> Factor out the split indirect descriptor detaching logic in order to
> make it be reused by the in order support.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c | 63 -
On Mon, Mar 24, 2025 at 7:01 AM Jason Wang wrote:
>
> This variable is used by packed virtqueue now, moving it to
> vring_virtqueue to make it possible to be reused by split virtqueue
> in-order implementation.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virti
On Mon, Mar 24, 2025 at 7:01 AM Jason Wang wrote:
>
> Factor out the core logic for updating last_used_idx to be reused by
> the packed in order implementation.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c | 43 +---
On Mon, Mar 24, 2025 at 7:01 AM Jason Wang wrote:
>
> Let's determine the last descriptor by counting the number of sg. This
> would be consistent with packed virtqueue implementation and ease the
> future in-order implementation.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
>
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Use u16 for last_used_idx in virtqueue_poll_split() to align with the
> spec.
>
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/virtio/virtio_ri
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
> Signed-off-by: Jason Wang
Acked-by: Eugenio Pérez
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
Acked-by: Eugenio Pérez
> Signed-off-by: Jason Wang
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
> Signed-off-by: Jason Wang
Acked-by: Eugenio Pérez
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
> Signed-off-by: Jason Wang
Acked-by: Eugenio Pérez
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:44 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
> Signed-off-by: Jason Wang
Acked-by: Eugenio Pérez
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:43 AM Jason Wang wrote:
>
> Those variants are used internally so let's switch to use
> vring_virtqueue as parameter to be consistent with other internal
> virtqueue helpers.
>
> Signed-off-by: Jason Wang
Acked-by: Eugenio Pérez
> ---
> drivers/virtio/virtio_ring.c |
On Mon, Mar 24, 2025 at 6:43 AM Jason Wang wrote:
>
> To be consistent with virtqueue_reset().
>
> Signed-off-by: Jason Wang
Acked-by: Eugenio Pérez
> ---
> drivers/virtio/virtio_ring.c | 12 ++--
> 1 file changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/virtio/virtio
On Mon, Mar 24, 2025 at 3:44 PM Lei Yang wrote:
>
> QE tested this series of patches with virtio-net regression tests,
> everything works fine.
>
Hi Lei,
Is it possible to test this series also with virtio-net-pci,...,in_order=on?
Thanks!
> Tested-by: Lei Yang
>
> On Mon, Mar 24, 2025 at 1:45
On 3/18/25 3:25 AM, Bastien Curutchet wrote:
Hi all,
On 2/24/25 5:10 PM, Stanislav Fomichev wrote:
On 02/21, Bastien Curutchet (eBPF Foundation) wrote:
Hi all,
This patch series continues the work to migrate the script tests into
prog_tests.
test_xdp_vlan.sh tests the ability of an XDP progr
On 3/4/25 1:24 AM, Bastien Curutchet (eBPF Foundation) wrote:
+int remove_routes_to_gredev(const char *ns1, const char *ns2, const char *vrf)
+{
+ SYS(fail, "ip -n %s route del %s dev veth5 %s", ns1, IP4_ADDR_GRE, vrf);
+ SYS(fail, "ip -n %s route del %s dev veth7 %s", ns2, IP4_ADDR_G
On 3/3/25 8:13 AM, Marcus Wichelmann wrote:
Am 28.02.25 um 20:49 schrieb Martin KaFai Lau:
On 2/27/25 6:23 AM, Marcus Wichelmann wrote:
When the XDP metadata area was used, it is expected that the same
metadata can also be accessed from TC, as can be read in the description
of the
On Mon, Mar 3, 2025 at 9:52 AM Stefano Garzarella wrote:
>
> VHOST_OWNER_SET and VHOST_OWNER_RESET are used in the documentation
> instead of VHOST_SET_OWNER and VHOST_RESET_OWNER respectively.
>
> To avoid confusion, let's use the right names in the documentation.
> No change to the API, only the
On 2/27/25 6:23 AM, Marcus Wichelmann wrote:
When the XDP metadata area was used, it is expected that the same
metadata can also be accessed from TC, as can be read in the description
of the bpf_xdp_adjust_meta helper function. In the tun driver, this was
not yet implemented.
To make this work,
On 2/26/25 5:22 AM, Jiayuan Chen wrote:
Use RCU lock to protect sk_socket, preventing concurrent close and release
by another thread.
Because TCP/UDP are already within a relatively large critical section:
'''
ip_local_deliver_finish
rcu_read_lock
ip_protocol_deliver_rcu
tcp_rcv/udp
On Mon, Feb 24, 2025 at 10:51 PM Michael S. Tsirkin wrote:
>
> On Tue, Jan 21, 2025 at 11:33:46AM +0100, Eugenio Pérez wrote:
> > A VDUSE device that implements virtiofs device works fine just by
> > adding the device id to the whitelist.
> >
> > Signed-off-by: Eugenio Pérez
>
>
> OK, but the com
On Mon, Feb 24, 2025 at 7:44 AM Eugenio Perez Martin
wrote:
>
> On Mon, Feb 24, 2025 at 3:01 AM Jason Wang wrote:
> >
> > On Sat, Feb 22, 2025 at 1:07 AM Eugenio Pérez wrote:
> > >
> > > Disable notifications right before scheduling, instead of wai
On Mon, Feb 24, 2025 at 3:01 AM Jason Wang wrote:
>
> On Sat, Feb 22, 2025 at 1:07 AM Eugenio Pérez wrote:
> >
> > Disable notifications right before scheduling, instead of waiting until
> > the work is running.
> >
> > After applying this patch, fio read test goes from 1191MiB/s to
> > 1263.14Mi
On Mon, Feb 24, 2025 at 2:57 AM Jason Wang wrote:
>
> On Sat, Feb 22, 2025 at 1:06 AM Eugenio Pérez wrote:
> >
> > A VDUSE device that implements virtiofs device works fine just by
> > adding the device id to the whitelist.
> >
> > Signed-off-by: Eugenio Pérez
> > Acked-by: Jason Wang
> > ---
>
On Mon, Feb 17, 2025 at 8:45 PM Dragos Tatulea wrote:
>
> From: Si-Wei Liu
>
> create_user_mr() has correct code to count the number of null keys
> used to fill in a hole for the memory map. However, fill_indir()
> does not follow the same to cap the range up to the 1GB limit
> correspondinly.
s
On Thu, Feb 13, 2025 at 5:18 PM Stefano Garzarella wrote:
>
> As defined in the specification, the `controls` field in the configuration
> space is only valid/present if VIRTIO_SND_F_CTLS is negotiated.
>
> From https://docs.oasis-open.org/virtio/virtio/v1.3/virtio-v1.3.html:
>
> 5.14.4 Device
On Mon, 2025-02-10 at 16:06 -0800, Andrii Nakryiko wrote:
> > Tracking associated maps for a program is not necessary. As long as
> > the last BPF program using the BPF map is unloaded, the kernel will
> > automatically free not-anymore-referenced BPF map. Note that
> > bpf_object itself will keep
On Wed, 2025-02-05 at 14:33 -0800, Andrii Nakryiko wrote:
> > > >
> > > > I see two ways forward for you. Either you can break apart your
> > > > BPF
> > > > object of ~100 BPF programs into more independent BPF objects >
> > > > > (seeing
> > > > that programs can be independently loaded/unloaded
On 1/30/25 11:21 PM, Bastien Curutchet (eBPF Foundation) wrote:
+#define BROADCAST_REDIRECT_SKEL_NB 2
+static void xdp_veth_broadcast_redirect(u32 attach_flags, u64 redirect_flags)
+{
+ struct prog_configuration prog_cfg[VETH_PAIRS_COUNT] = {
+ {
+ .l
On Fri, 2025-01-24 at 10:31 -0800, Andrii Nakryiko wrote:
> > On Wed, Jan 22, 2025 at 1:53 PM Slava Imameev
> > wrote:
> > > >
> > > > BPF programs designated as dynamically loaded can be loaded and
> > > > attached independently after the initial bpf_object loading and
> > > > attaching.
> > > >
On 1/28/25 1:57 AM, Bastien Curutchet (eBPF Foundation) wrote:
Set the tests to run serially to avoid conflicts with
test_xdp_veth_redirect
I think this has been fixed in v3?
Others lgtm also. Thanks for working on this.
+void test_xdp_veth_broadcast_redirect(void)
+{
+ if (test__start
On 1/28/25 1:57 AM, Bastien Curutchet (eBPF Foundation) wrote:
+#define NO_IP "NO_IP"
#define PROG_NAME_MAX_LEN 128
+#define NS_NAME_MAX_LEN32
struct veth_configuration {
char local_veth[VETH_NAME_MAX_LEN]; /* Interface in main namespac
On 1/28/25 1:57 AM, Bastien Curutchet (eBPF Foundation) wrote:
Some tests can't be run in parallel because they use same namespace
names or veth names.
Create an helper that appends the thread ID to a given string. 8
characters are used for it (7 digits + '\0')
Signed-off-by: Bastien Curutchet
On 12/19/24 7:46 AM, Matthieu Baerts (NGI0) wrote:
+SEC("cgroup/getsockopt")
+int iters_subflow(struct bpf_sockopt *ctx)
+{
+ struct mptcp_subflow_context *subflow;
+ struct bpf_sock *sk = ctx->sk;
+ struct sock *ssk = NULL;
+ struct mptcp_sock *msk;
+ int local_ids
On 12/19/24 7:46 AM, Matthieu Baerts (NGI0) wrote:
From: Geliang Tang
The KF_TRUSTED_ARGS flag is used for bpf_iter_mptcp_subflow_new, it
indicates that the all pointer arguments are valid. It's necessary to
add a KF_ACQUIRE helper to get valid "msk".
This feels wrong. It forces an unnecessar
this:
mptcp_for_each_subflow(msk, subflow)
kfunc(subflow);
But in the MPTCP BPF program, this has not yet been implemented. As
Martin suggested recently, this conn_list walking + modify-by-kfunc
usage fits the bpf_iter use case.
So this patch adds a new bpf_iter type named "mptcp_subflow&q
On Thu, Jan 23, 2025 at 2:54 AM Jason Wang wrote:
>
> On Thu, Jan 23, 2025 at 12:32 AM Eugenio Pérez wrote:
> >
> > We need to move the map and unmap out of virtiofs queue lock. We need to
> > store the unmap sgs to call virtqueue_unmap_sgs, so use the request
> > itself.
> >
> > TODO: We need t
On Thu, Jan 23, 2025 at 2:51 AM Jason Wang wrote:
>
> On Thu, Jan 23, 2025 at 12:32 AM Eugenio Pérez wrote:
> >
> > From: Jason Wang
> >
> > Introduce new virtqueue DMA operations which allows the drivers that
> > want to make use of the premapping API but operate at the sg level.
> >
> > Note t
On Thu, Jan 23, 2025 at 2:50 AM Jason Wang wrote:
>
> On Wed, Jan 22, 2025 at 11:49 PM Stefan Hajnoczi wrote:
> >
> > On Tue, Jan 21, 2025 at 11:33:46AM +0100, Eugenio Pérez wrote:
> > > A VDUSE device that implements virtiofs device works fine just by
> > > adding the device id to the whitelist.
.
- Patch 3: remove two kfunc, more restrictions. (Martin)
- Patch 4: add BUILD_BUG_ON(), more restrictions. (Martin)
- Patch 7: adaptations due to modifications in patches 1-4.
- Link to v1:
https://lore.kernel.org/r/20241108-bpf-next-net-mptcp-bpf_iter-subflows-v1-0-cf1695303...@kernel.org
The v2
On 1/10/25 1:21 AM, Bastien Curutchet (eBPF Foundation) wrote:
+static void ping_test(struct test_data *data)
+{
+ struct test_xdp_redirect *skel = NULL;
+ struct xdp_dummy *skel_dummy = NULL;
+ struct nstoken *nstoken = NULL;
+ int i, ret;
+
+ skel_dummy = xdp_dummy
On 1/3/25 2:10 AM, Bastien Curutchet (eBPF Foundation) wrote:
+static int ping_setup(struct test_data *data)
+{
+ struct nstoken *nstoken = NULL;
+ int i;
+
+ data->ns[0] = netns_new(NS0, false);
+ if (!ASSERT_OK_PTR(data->ns[0], "create ns"))
+ return -1;
+
On 12/6/24 12:12 AM, Bastien Curutchet wrote:
+void test_xdp_context_functional(void)
+{
+ LIBBPF_OPTS(bpf_tc_hook, tc_hook, .attach_point = BPF_TC_INGRESS);
+ LIBBPF_OPTS(bpf_tc_opts, tc_opts, .handle = 1, .priority = 1);
+ struct bpf_program *tc_prog, *xdp_prog;
+ struct
ecause the issue of deadlock is inevitable.
Acked-by: Martin KaFai Lau
Jakub, please help to land it to the net tree. Thanks!
this:
mptcp_for_each_subflow(msk, subflow)
kfunc(subflow);
But in the MPTCP BPF program, this has not yet been implemented. As
Martin suggested recently, this conn_list walking + modify-by-kfunc
usage fits the bpf_iter use case.
So this patch adds a new bpf_iter type named "mptcp_subflow&q
On 11/8/24 7:52 AM, Matthieu Baerts (NGI0) wrote:
From: Geliang Tang
MPTCP helper mptcp_sk() is used to convert struct sock to mptcp_sock.
Helpers mptcp_subflow_ctx() and mptcp_subflow_tcp_sock() are used to
convert between struct mptcp_subflow_context and sock. They all will
be used in MPTCP B
On Thu, Nov 7, 2024 at 2:21 PM Colin Ian King wrote:
>
> The boolean variable has_affinity is true when desc is non-null and
> ops->set_vq_affinity is non-null. Hence the call to create_affinity_masks
> does not need to check for desc being non-null is redundant when
> has_affinity is true, so it
On Tue, Nov 5, 2024 at 7:52 PM Dragos Tatulea wrote:
>
> In the error recovery path of mlx5_vdpa_dev_add(), the cleanup is
> executed and at the end put_device() is called which ends up calling
> mlx5_vdpa_free(). This function will execute the same cleanup all over
> again. Most resources support
On 10/30/24 9:31 AM, Matthieu Baerts wrote:
Hi Tao, BPF maintainers,
On 30/10/2024 12:12, Tao Chen wrote:
在 2024/10/30 18:49, Matthieu Baerts 写道:
Hi Tao Chen,
Thank you for having shared this patch.
On 30/10/2024 11:01, Tao Chen wrote:
Fix compile error when MPTCP feature not support, thoug
On 10/16/24 11:35 AM, Alexis Lothoré (eBPF Foundation) wrote:
.../bpf/progs/test_tcp_check_syncookie_kern.c | 167
.../selftests/bpf/test_tcp_check_syncookie.sh | 85
.../selftests/bpf/test_tcp_check_syncookie_user.c | 213 -
Nice.
Le
On 10/16/24 11:35 AM, Alexis Lothoré (eBPF Foundation) wrote:
btf_skc_cls_ingress test currently checks that syncookie and
bpf_sk_assign/release helpers behave correctly in multiple scenarios,
but only with ipv4 socket.
Increase those helpers coverage by adding testing support for IPv6-only
sock
On 10/16/24 11:35 AM, Alexis Lothoré (eBPF Foundation) wrote:
btf_skc_cls_ingress.c currently runs two subtests, and create a
dedicated network namespace for each, but never cleans up the created
namespace once the test has ended.
Add missing namespace cleanup after each namespace to avoid accum
On 10/9/24 3:12 AM, Alexis Lothoré (eBPF Foundation) wrote:
Current test only checks attach/detach on cpu map type program, and so
does not check that it can be properly executed, neither that it
redirects correctly.
Update the existing test to extend its coverage:
- keep the redirected program
On 9/24/24 6:37 PM, Martin KaFai Lau wrote:
There are other .sh tests that could better use the test_progs migration. In
particular the ones without existing test coverage. For non XDP related,
test_tcp_check_syncookie.sh, test_flow_dissector.sh, and test_tc_edt.sh should
be the good ones.
I
On 9/11/24 1:41 AM, Florian Kauer wrote:
@@ -25,14 +28,11 @@ static void test_xdp_with_devmap_helpers(void)
if (!ASSERT_OK_PTR(skel, "test_xdp_with_devmap_helpers__open_and_load"))
return;
- dm_fd = bpf_program__fd(skel->progs.xdp_redir_prog);
- err = bpf_xdp_att
On 9/14/24 3:32 AM, Feng zhou wrote:
From: Feng Zhou
This patch adds a test for TCP over IPv4 via INET6 API.
Signed-off-by: Feng Zhou
---
.../selftests/bpf/prog_tests/setget_sockopt.c | 33 +++
.../selftests/bpf/progs/setget_sockopt.c | 13 ++--
2 files changed, 4
On 9/14/24 3:32 AM, Feng zhou wrote:
From: Feng Zhou
when TCP over IPv4 via INET6 API, bpf_get/setsockopt with ipv4 will
I think you meant bpf_get/setsockopt with SOL_IP will fail. so s/ipv4/SOL_IP/?
fail, because sk->sk_family is AF_INET6. With ipv6 will success, not
take effect, because i
On 9/18/24 12:45 AM, Feng zhou wrote:
From: Feng Zhou
This patch adds a test for cgroup skb to get classid.
Signed-off-by: Feng Zhou
---
.../bpf/prog_tests/cg_skb_get_classid.c | 87 +++
.../selftests/bpf/progs/cg_skb_get_classid.c | 19
2 files changed, 106 in
On 9/25/24 3:37 AM, Martin KaFai Lau wrote:
I am not sure which case in xdp_features.c does not have existing coverage in
test_progs. From a quick look, it seems only BPF_MAP_TYPE_CPUMAP is missing
(please check)?
Re: CPUMAP, I noticed there is a xdp_(cpu)map_attach.c test but it only does
On 9/19/24 11:12 AM, Paolo Abeni wrote:
On 9/11/24 10:41, Florian Kauer wrote:
rxq contains a pointer to the device from where
the redirect happened. Currently, the BPF program
that was executed after a redirect via BPF_MAP_TYPE_DEVMAP*
does not have it set.
Add bugfix and related selftest.
Si
On 9/22/24 12:04 PM, Alexis Lothoré wrote:
Hello all, sorry for the slow feedback, I have been off last week.
On 9/14/24 15:38, Jakub Kicinski wrote:
On Sat, 14 Sep 2024 11:25:47 +0200 Lorenzo Bianconi wrote:
On Sep 13, Martin KaFai Lau wrote:
test a physical network device that supports a
On 9/17/24 6:15 PM, Tiago Lam wrote:
On Fri, Sep 13, 2024 at 11:24:09AM -0700, Martin KaFai Lau wrote:
On 9/13/24 2:39 AM, Tiago Lam wrote:
This follows the same rationale provided for the ipv4 counterpart, where
it now runs a reverse socket lookup when source addresses and/or ports
are
On 9/14/24 11:24 AM, Lin Yikai wrote:
diff --git a/tools/testing/selftests/bpf/Makefile
b/tools/testing/selftests/bpf/Makefile
index 04716a5e43f1..369c5ad8fc4a 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -183,7 +183,7 @@ NON_CHECK_FEAT_TARGETS
On 9/11/24 8:16 AM, Matthieu Baerts (NGI0) wrote:
+static void test_subflow(void)
+{
+ int cgroup_fd, prog_fd, err;
+ struct mptcp_subflow *skel;
+ struct nstoken *nstoken;
+ struct bpf_link *link;
+
+ cgroup_fd = test__join_cgroup("/mptcp_subflow");
+ if (!ASS
On 9/11/24 8:16 AM, Matthieu Baerts (NGI0) wrote:
diff --git a/tools/testing/selftests/bpf/progs/mptcp_bpf.h
b/tools/testing/selftests/bpf/progs/mptcp_bpf.h
new file mode 100644
index ..179b74c1205f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/mptcp_bpf.h
@@ -0,0 +1,42 @@
+/
On 9/10/24 11:10 AM, Alexis Lothoré (eBPF Foundation) wrote:
test_xdp_features.sh is a shell script allowing to test that xdp features
advertised by an interface are indeed delivered. The test works by starting
two instance of the same program, both attaching specific xdp programs to
each side of
On 9/13/24 2:39 AM, Tiago Lam wrote:
This follows the same rationale provided for the ipv4 counterpart, where
it now runs a reverse socket lookup when source addresses and/or ports
are changed, on sendmsg, to check whether egress traffic should be
allowed to go through or not.
As with ipv4, the
On Mon, Sep 9, 2024 at 11:30 AM Dragos Tatulea wrote:
>
>
>
> On 30.08.24 12:58, Dragos Tatulea wrote:
> > This series improves the time of .set_map() operations by parallelizing
> > the MKEY creation and deletion for direct MKEYs. Looking at the top
> > level MKEY creation/deletion functions, the
On Mon, Sep 9, 2024 at 11:33 AM Dragos Tatulea wrote:
>
>
>
> On 16.08.24 11:01, Dragos Tatulea wrote:
> > Currently, commands that qualify as throttled can't be used via the
> > async API. That's due to the fact that the throttle semaphore can sleep
> > but the async API can't.
> >
> > This patch
On Tue, Sep 3, 2024 at 9:48 AM Dragos Tatulea wrote:
>
>
>
> On 03.09.24 09:40, Lei Yang wrote:
> > On Mon, Sep 2, 2024 at 7:05 PM Dragos Tatulea wrote:
> >>
> >> Hi Lei,
> >>
> >> On 02.09.24 12:03, Lei Yang wrote:
> >>> Hi Dragos
> >>>
> >>> QE tested this series with mellanox nic, it failed wi
On Thu, Aug 29, 2024 at 5:23 PM Dragos Tatulea wrote:
>
>
>
> On 29.08.24 17:07, Eugenio Perez Martin wrote:
> > On Wed, Aug 21, 2024 at 1:42 PM Dragos Tatulea wrote:
> >>
> >> Currently, when a new MR is set up, the old MR is deleted. MR deletion
> >>
On Thu, Aug 29, 2024 at 3:54 PM Dragos Tatulea wrote:
>
>
>
> On 29.08.24 15:10, Eugenio Perez Martin wrote:
> > On Wed, Aug 21, 2024 at 1:41 PM Dragos Tatulea wrote:
> >>
> >> Use the async interface to issue MTT MKEY creation.
> >> Extra care is
On Wed, Aug 21, 2024 at 1:42 PM Dragos Tatulea wrote:
>
> Currently, when a new MR is set up, the old MR is deleted. MR deletion
> is about 30-40% the time of MR creation. As deleting the old MR is not
> important for the process of setting up the new MR, this operation
> can be postponed.
>
> Thi
On Wed, Aug 21, 2024 at 1:42 PM Dragos Tatulea wrote:
>
> There's currently not a lot of action happening during
> the init/destroy of MR resources. But more will be added
> in the upcoming patches.
If the series doesn't receive new patches, it is just the next patch :).
>
> Signed-off-by: Drago
On Wed, Aug 21, 2024 at 1:42 PM Dragos Tatulea wrote:
>
> Now that the mr resources have their own namespace in the
> struct, give the lock a clearer name.
>
> Signed-off-by: Dragos Tatulea
> Reviewed-by: Cosmin Ratiu
Acked-by: Eugenio Pérez
> ---
> drivers/vdpa/mlx5/core/mlx5_vdpa.h | 2 +-
On Wed, Aug 21, 2024 at 1:42 PM Dragos Tatulea wrote:
>
> Group all mapping related resources into their own structure.
>
> Upcoming patches will add more members in this new structure.
>
> Signed-off-by: Dragos Tatulea
> Reviewed-by: Cosmin Ratiu
Acked-by: Eugenio Pérez
> ---
> drivers/vdpa
On Wed, Aug 21, 2024 at 1:42 PM Dragos Tatulea wrote:
>
> Use the async interface to issue MTT MKEY deletion.
>
> This makes destroy_user_mr() on average 8x times faster.
> This number is also dependent on the size of the MR being
> deleted.
>
> Signed-off-by: Dragos Tatulea
> Reviewed-by: Cosmin
On Wed, Aug 21, 2024 at 1:41 PM Dragos Tatulea wrote:
>
> Use the async interface to issue MTT MKEY creation.
> Extra care is taken at the allocation of FW input commands
> due to the MTT tables having variable sizes depending on
> MR.
>
> The indirect MKEY is still created synchronously at the
>
On Fri, Aug 16, 2024 at 11:03 AM Dragos Tatulea wrote:
>
> change_num_qps() is still suspending/resuming VQs one by one.
> This change switches to parallel suspend/resume.
>
> When increasing the number of queues the flow has changed a bit for
> simplicity: the setup_vq() function will always be c
1 - 100 of 5419 matches
Mail list logo
