From: Jesper Dangaard Brouer <brouer@redhat.com>
To: "Björn Töpel" <bjorn.topel@gmail.com>
Cc: brouer@redhat.com, ast@kernel.org, daniel@iogearbox.net,
netdev@vger.kernel.org, bpf@vger.kernel.org,
"Björn Töpel" <bjorn.topel@intel.com>,
magnus.karlsson@intel.com, maciej.fijalkowski@intel.com,
kuba@kernel.org, jonathan.lemon@gmail.com, maximmi@nvidia.com,
davem@davemloft.net, hawk@kernel.org, john.fastabend@gmail.com,
ciara.loftus@intel.com, weqaar.a.janjua@intel.com,
andrii@kernel.org, "Marek Majtyka" <alardam@gmail.com>
Subject: Re: [PATCH bpf-next 3/3] libbpf, xsk: select AF_XDP BPF program based on kernel version
Date: Tue, 26 Jan 2021 09:38:23 +0100 [thread overview]
Message-ID: <20210126093823.633448e0@carbon> (raw)
In-Reply-To: <20210122105351.11751-4-bjorn.topel@gmail.com>
On Fri, 22 Jan 2021 11:53:51 +0100
Björn Töpel <bjorn.topel@gmail.com> wrote:
> From: Björn Töpel <bjorn.topel@intel.com>
>
> Add detection for kernel version, and adapt the BPF program based on
> kernel support. This way, users will get the best possible performance
> from the BPF program.
You say "detection for kernel version", but doesn't the code detect the
feature rather than kernel version ?
If so, please update the description.
> Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
> Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
> Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
> Signed-off-by: Marek Majtyka <alardam@gmail.com>
> ---
> tools/lib/bpf/xsk.c | 82 +++++++++++++++++++++++++++++++++++++++++++--
> 1 file changed, 79 insertions(+), 3 deletions(-)
>
> diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c
> index e3e41ceeb1bc..1df8c133a5bc 100644
> --- a/tools/lib/bpf/xsk.c
> +++ b/tools/lib/bpf/xsk.c
> @@ -46,6 +46,11 @@
> #define PF_XDP AF_XDP
> #endif
>
> +enum xsk_prog {
> + XSK_PROG_FALLBACK,
> + XSK_PROG_REDIRECT_FLAGS,
> +};
> +
> struct xsk_umem {
> struct xsk_ring_prod *fill_save;
> struct xsk_ring_cons *comp_save;
> @@ -351,6 +356,55 @@ int xsk_umem__create_v0_0_2(struct xsk_umem **umem_ptr, void *umem_area,
> COMPAT_VERSION(xsk_umem__create_v0_0_2, xsk_umem__create, LIBBPF_0.0.2)
> DEFAULT_VERSION(xsk_umem__create_v0_0_4, xsk_umem__create, LIBBPF_0.0.4)
>
> +
> +static enum xsk_prog get_xsk_prog(void)
> +{
> + enum xsk_prog detected = XSK_PROG_FALLBACK;
> + struct bpf_load_program_attr prog_attr;
> + struct bpf_create_map_attr map_attr;
> + __u32 size_out, retval, duration;
> + char data_in = 0, data_out;
> + struct bpf_insn insns[] = {
> + BPF_LD_MAP_FD(BPF_REG_1, 0),
> + BPF_MOV64_IMM(BPF_REG_2, 0),
> + BPF_MOV64_IMM(BPF_REG_3, XDP_PASS),
> + BPF_EMIT_CALL(BPF_FUNC_redirect_map),
> + BPF_EXIT_INSN(),
> + };
> + int prog_fd, map_fd, ret;
> +
> + memset(&map_attr, 0, sizeof(map_attr));
> + map_attr.map_type = BPF_MAP_TYPE_XSKMAP;
> + map_attr.key_size = sizeof(int);
> + map_attr.value_size = sizeof(int);
> + map_attr.max_entries = 1;
> +
> + map_fd = bpf_create_map_xattr(&map_attr);
> + if (map_fd < 0)
> + return detected;
> +
> + insns[0].imm = map_fd;
> +
> + memset(&prog_attr, 0, sizeof(prog_attr));
> + prog_attr.prog_type = BPF_PROG_TYPE_XDP;
> + prog_attr.insns = insns;
> + prog_attr.insns_cnt = ARRAY_SIZE(insns);
> + prog_attr.license = "GPL";
> +
> + prog_fd = bpf_load_program_xattr(&prog_attr, NULL, 0);
> + if (prog_fd < 0) {
> + close(map_fd);
> + return detected;
> + }
> +
> + ret = bpf_prog_test_run(prog_fd, 0, &data_in, 1, &data_out, &size_out, &retval, &duration);
> + if (!ret && retval == XDP_PASS)
> + detected = XSK_PROG_REDIRECT_FLAGS;
> + close(prog_fd);
> + close(map_fd);
> + return detected;
> +}
> +
> static int xsk_load_xdp_prog(struct xsk_socket *xsk)
> {
> static const int log_buf_size = 16 * 1024;
> @@ -358,7 +412,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
> char log_buf[log_buf_size];
> int err, prog_fd;
>
> - /* This is the C-program:
> + /* This is the fallback C-program:
> * SEC("xdp_sock") int xdp_sock_prog(struct xdp_md *ctx)
> * {
> * int ret, index = ctx->rx_queue_index;
> @@ -414,9 +468,31 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
> /* The jumps are to this instruction */
> BPF_EXIT_INSN(),
> };
> - size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
>
> - prog_fd = bpf_load_program(BPF_PROG_TYPE_XDP, prog, insns_cnt,
> + /* This is the post-5.3 kernel C-program:
> + * SEC("xdp_sock") int xdp_sock_prog(struct xdp_md *ctx)
> + * {
> + * return bpf_redirect_map(&xsks_map, ctx->rx_queue_index, XDP_PASS);
> + * }
> + */
> + struct bpf_insn prog_redirect_flags[] = {
> + /* r2 = *(u32 *)(r1 + 16) */
> + BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 16),
> + /* r1 = xskmap[] */
> + BPF_LD_MAP_FD(BPF_REG_1, ctx->xsks_map_fd),
> + /* r3 = XDP_PASS */
> + BPF_MOV64_IMM(BPF_REG_3, 2),
> + /* call bpf_redirect_map */
> + BPF_EMIT_CALL(BPF_FUNC_redirect_map),
> + BPF_EXIT_INSN(),
> + };
> + size_t insns_cnt[] = {sizeof(prog) / sizeof(struct bpf_insn),
> + sizeof(prog_redirect_flags) / sizeof(struct bpf_insn),
> + };
> + struct bpf_insn *progs[] = {prog, prog_redirect_flags};
> + enum xsk_prog option = get_xsk_prog();
> +
> + prog_fd = bpf_load_program(BPF_PROG_TYPE_XDP, progs[option], insns_cnt[option],
> "LGPL-2.1 or BSD-2-Clause", 0, log_buf,
> log_buf_size);
> if (prog_fd < 0) {
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
next prev parent reply other threads:[~2021-01-26 8:41 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-22 10:53 [PATCH bpf-next 0/3] AF_XDP clean up/perf improvements Björn Töpel
2021-01-22 10:53 ` [PATCH bpf-next 1/3] xsk: remove explicit_free parameter from __xsk_rcv() Björn Töpel
2021-01-22 10:53 ` [PATCH bpf-next 2/3] xsk: fold xp_assign_dev and __xp_assign_dev Björn Töpel
2021-01-22 10:53 ` [PATCH bpf-next 3/3] libbpf, xsk: select AF_XDP BPF program based on kernel version Björn Töpel
2021-01-25 23:12 ` Daniel Borkmann
2021-01-26 8:38 ` Jesper Dangaard Brouer [this message]
2021-01-22 13:19 ` [PATCH bpf-next 0/3] AF_XDP clean up/perf improvements Toke Høiland-Jørgensen
2021-01-22 13:37 ` Björn Töpel
2021-01-22 13:40 ` Toke Høiland-Jørgensen
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=20210126093823.633448e0@carbon \
--to=brouer@redhat.com \
--cc=alardam@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bjorn.topel@gmail.com \
--cc=bjorn.topel@intel.com \
--cc=bpf@vger.kernel.org \
--cc=ciara.loftus@intel.com \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=hawk@kernel.org \
--cc=john.fastabend@gmail.com \
--cc=jonathan.lemon@gmail.com \
--cc=kuba@kernel.org \
--cc=maciej.fijalkowski@intel.com \
--cc=magnus.karlsson@intel.com \
--cc=maximmi@nvidia.com \
--cc=netdev@vger.kernel.org \
--cc=weqaar.a.janjua@intel.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).