From: "Daniel P. Berrangé" <berrange@redhat.com>
To: Andrew Melnychenko <andrew@daynix.com>
Cc: jasowang@redhat.com, mst@redhat.com, pbonzini@redhat.com,
marcandre.lureau@redhat.com, thuth@redhat.com, philmd@linaro.org,
armbru@redhat.com, eblake@redhat.com, qemu-devel@nongnu.org,
toke@redhat.com, mprivozn@redhat.com,
yuri.benditovich@daynix.com, yan@daynix.com
Subject: Re: [RFC PATCH 2/5] virtio-net: Added property to load eBPF RSS with fds.
Date: Thu, 30 Mar 2023 09:52:49 +0100 [thread overview]
Message-ID: <ZCVN4TWwDsXJGfwi@redhat.com> (raw)
In-Reply-To: <20230330001522.120774-3-andrew@daynix.com>
On Thu, Mar 30, 2023 at 03:15:19AM +0300, Andrew Melnychenko wrote:
> eBPF RSS program and maps may now be passed during initialization.
> Initially was implemented for libvirt to launch qemu without permissions,
> and initialized eBPF program through the helper.
>
> Signed-off-by: Andrew Melnychenko <andrew@daynix.com>
> ---
> hw/net/virtio-net.c | 77 ++++++++++++++++++++++++++++++++--
> include/hw/virtio/virtio-net.h | 1 +
> 2 files changed, 74 insertions(+), 4 deletions(-)
>
> diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
> index 53e1c32643..45d448a83d 100644
> --- a/hw/net/virtio-net.c
> +++ b/hw/net/virtio-net.c
> @@ -42,6 +42,7 @@
> #include "sysemu/sysemu.h"
> #include "trace.h"
> #include "monitor/qdev.h"
> +#include "monitor/monitor.h"
> #include "hw/pci/pci_device.h"
> #include "net_rx_pkt.h"
> #include "hw/virtio/vhost.h"
> @@ -1305,14 +1306,81 @@ static void virtio_net_detach_epbf_rss(VirtIONet *n)
> virtio_net_attach_ebpf_to_backend(n->nic, -1);
> }
>
> -static bool virtio_net_load_ebpf(VirtIONet *n)
> +static int virtio_net_get_ebpf_rss_fds(char *str, char *fds[], int nfds)
Needs to have an 'Error **errp' parameter
> +{
> + char *ptr = str;
> + char *cur = NULL;
> + size_t len = strlen(str);
> + int i = 0;
> +
> + for (; i < nfds && ptr < str + len;) {
> + cur = strchr(ptr, ':');
> +
> + if (cur == NULL) {
> + fds[i] = g_strdup(ptr);
> + } else {
> + fds[i] = g_strndup(ptr, cur - ptr);
> + }
> +
> + i++;
> + if (cur == NULL) {
> + break;
> + } else {
> + ptr = cur + 1;
> + }
> + }
...and here needs to error_setg(...) if the string being parsed has
more FDs present than 'nfds', and return error.
It should also report error if less than 'nfds' were present.
> +
> + return i;
> +}
> +
> +static bool virtio_net_load_ebpf_fds(VirtIONet *n)
> {
> - if (!virtio_net_attach_ebpf_to_backend(n->nic, -1)) {
> - /* backend does't support steering ebpf */
> + char *fds_strs[EBPF_RSS_MAX_FDS];
> + int fds[EBPF_RSS_MAX_FDS];
> + int nfds;
> + int ret = false;
> + Error *errp;
This method should be having 'Error **errp' passed into it
instead
> + int i = 0;
> +
> + if (n == NULL || !n->ebpf_rss_fds) {
> return false;
> }
The only caller has already de-referenced 'n' so n == NULL is
impossible. The caller has also chck ebpf_rss_fds != NULL.
IOW, this is redundant
> - return ebpf_rss_load(&n->ebpf_rss);
> + nfds = virtio_net_get_ebpf_rss_fds(n->ebpf_rss_fds,
> + fds_strs, EBPF_RSS_MAX_FDS);
Needs to pass in 'errp' and check for the error condition on return
if (nfds != EBPF_RSS_MAX_FDS)
return false;
> + for (i = 0; i < nfds; i++) {
> + fds[i] = monitor_fd_param(monitor_cur(), fds_strs[i], &errp);
if (fds[i] < 0)
....close all FDs so far and return error...
> + }
> +
> + if (nfds == EBPF_RSS_MAX_FDS) {
> + ret = ebpf_rss_load_fds(&n->ebpf_rss, fds[0], fds[1], fds[2], fds[3]);
> + }
This condition is not required, since we should have check
it immediately after the virtio_net_get_ebpf_rss_fds() call
> +
> + if (!ret) {
> + for (i = 0; i < nfds; i++) {
> + close(fds[i]);
> + }
> + }
> +
> + for (i = 0; i < nfds; i++) {
> + g_free(fds_strs[i]);
> + }
> +
> + return ret;
> +}
> +
> +static bool virtio_net_load_ebpf(VirtIONet *n)
This needs to have "Error **errp' added to it and return an
error on failure. The caller already has an 'errp' available
to pass in.
> +{
> + bool ret = true;
> +
> + if (virtio_net_attach_ebpf_to_backend(n->nic, -1)) {
> + if (!(n->ebpf_rss_fds
> + && virtio_net_load_ebpf_fds(n))) {
> + ret = ebpf_rss_load(&n->ebpf_rss);
> + }
> + }
> +
> + return ret;
> }
>
> static void virtio_net_unload_ebpf(VirtIONet *n)
> @@ -3900,6 +3968,7 @@ static Property virtio_net_properties[] = {
> VIRTIO_NET_F_RSS, false),
> DEFINE_PROP_BIT64("hash", VirtIONet, host_features,
> VIRTIO_NET_F_HASH_REPORT, false),
> + DEFINE_PROP_STRING("ebpf_rss_fds", VirtIONet, ebpf_rss_fds),
> DEFINE_PROP_BIT64("guest_rsc_ext", VirtIONet, host_features,
> VIRTIO_NET_F_RSC_EXT, false),
> DEFINE_PROP_UINT32("rsc_interval", VirtIONet, rsc_timeout,
> diff --git a/include/hw/virtio/virtio-net.h b/include/hw/virtio/virtio-net.h
> index ef234ffe7e..e10ce88f91 100644
> --- a/include/hw/virtio/virtio-net.h
> +++ b/include/hw/virtio/virtio-net.h
> @@ -219,6 +219,7 @@ struct VirtIONet {
> VirtioNetRssData rss_data;
> struct NetRxPkt *rx_pkt;
> struct EBPFRSSContext ebpf_rss;
> + char *ebpf_rss_fds;
> };
>
> size_t virtio_net_handle_ctrl_iov(VirtIODevice *vdev,
> --
> 2.39.1
>
With regards,
Daniel
--
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o- https://fstop138.berrange.com :|
|: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
next prev parent reply other threads:[~2023-03-30 8:53 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-30 0:15 [RFC PATCH 0/4] eBPF RSS through QMP support Andrew Melnychenko
2023-03-30 0:15 ` [RFC PATCH 1/5] ebpf: Added eBPF initialization by fds and map update Andrew Melnychenko
2023-03-30 6:53 ` Jason Wang
2023-03-30 6:56 ` Jason Wang
2023-03-30 7:10 ` Daniel P. Berrangé
2023-03-30 11:13 ` Andrew Melnichenko
2023-03-30 0:15 ` [RFC PATCH 2/5] virtio-net: Added property to load eBPF RSS with fds Andrew Melnychenko
2023-03-30 8:52 ` Daniel P. Berrangé [this message]
2023-03-30 0:15 ` [RFC PATCH 3/5] ebpf: Added declaration/initialization routines Andrew Melnychenko
2023-03-30 6:54 ` Jason Wang
2023-03-30 8:34 ` Daniel P. Berrangé
2023-03-31 7:48 ` Jason Wang
2023-03-31 7:59 ` Daniel P. Berrangé
2023-03-31 8:03 ` Jason Wang
2023-03-31 8:13 ` Daniel P. Berrangé
2023-03-31 8:21 ` Jason Wang
2023-03-30 8:33 ` Daniel P. Berrangé
2023-03-30 11:02 ` Andrew Melnichenko
2023-03-30 0:15 ` [RFC PATCH 4/5] qmp: Added new command to retrieve eBPF blob Andrew Melnychenko
2023-03-30 8:39 ` Daniel P. Berrangé
2023-03-30 10:42 ` Andrew Melnichenko
2023-03-30 0:15 ` [RFC PATCH 5/5] ebpf: Updated eBPF program and skeleton Andrew Melnychenko
2023-03-30 6:57 ` [RFC PATCH 0/4] eBPF RSS through QMP support Jason Wang
2023-03-30 10:48 ` Andrew Melnichenko
-- strict thread matches above, loose matches on Subject: below --
2021-06-09 10:04 [RFC PATCH 0/5] ebpf: Added ebpf helper for libvirtd Andrew Melnychenko
2021-06-09 10:04 ` [RFC PATCH 2/5] virtio-net: Added property to load eBPF RSS with fds Andrew Melnychenko
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZCVN4TWwDsXJGfwi@redhat.com \
--to=berrange@redhat.com \
--cc=andrew@daynix.com \
--cc=armbru@redhat.com \
--cc=eblake@redhat.com \
--cc=jasowang@redhat.com \
--cc=marcandre.lureau@redhat.com \
--cc=mprivozn@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=thuth@redhat.com \
--cc=toke@redhat.com \
--cc=yan@daynix.com \
--cc=yuri.benditovich@daynix.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).