xdp generic default option
Subsystems (3)
Hi Andrii,
working on xdp multi-buff I figured out now xdp generic is the default choice if not specified by userspace. In particular after commit 7f0a838254bd ("bpf, xdp: Maintain info on attached XDP BPF programs in net_device"), running the command below, XDP will run in generic mode even if the underlay driver support XDP in native mode:
$ip link set dev eth0 xdp obj prog.o
$ip link show dev eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 xdpgeneric qdisc mq state UP mode DEFAULT
group default qlen 1024
link/ether f0:ad:4e:09:6b:57 brd ff:ff:ff:ff:ff:ff
prog/xdp id 1 tag 3b185187f1855c4c jited
Is it better to use xdpdrv as default choice if not specified by userspace? doing something like:
net/core/dev.c +3 -3
@@ -8747,9 +8747,9 @@ static enum bpf_xdp_mode dev_xdp_mode(u32 flags) {if (flags &XDP_FLAGS_HW_MODE )return XDP_MODE_HW ; -if (flags &XDP_FLAGS_DRV_MODE ) -return XDP_MODE_DRV ; -return XDP_MODE_SKB ; +if (flags &XDP_FLAGS_SKB_MODE ) +return XDP_MODE_SKB ; +return XDP_MODE_DRV ; }static bpf_op_t dev_xdp_bpf_op (struct net_device *dev,enum bpf_xdp_mode mode)
Regards,
Lorenzo
Quoted 8 lines · Lorenzo Bianconi
Hi Andrii,
working on xdp multi-buff I figured out now xdp generic is the default choice if not specified by userspace. In particular after commit
7f0a838254bd("bpf, xdp: Maintain info on attached XDP BPF programs innet_device"), running the command below, XDP will run in generic mode even if the underlay driver support XDP in native mode:
Quoted 6 lines
$ip link set dev eth0 xdp obj prog.o
$ip link show dev eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 xdpgeneric qdisc mq state UP mode DEFAULT
group default qlen 1024
link/ether f0:ad:4e:09:6b:57 brd ff:ff:ff:ff:ff:ff
prog/xdp id 1 tag3b185187f1855c4cjited
Yeah, defaulting to xdpgeneric is not a good idea (and a change in behaviour; I get native mode on the same command on a 5.8 kernel)...
-Toke
Jakub Kicinski
maintainerQuoted 7 lines · On Wed, 19 Aug 2020 11:28:11 +0200 Lorenzo Bianconi wrote:
Hi Andrii,
working on xdp multi-buff I figured out now xdp generic is the default choice if not specified by userspace. In particular after commit
7f0a838254bd("bpf, xdp: Maintain info on attached XDP BPF programs innet_device"), running the command below, XDP will run in generic mode even if the underlay driver support XDP in native mode:
Make me wonder if bpf/test_offload.py was ever run on those changes :/
Quoted 25 lines · Lorenzo Bianconi
Hi Andrii,
working on xdp multi-buff I figured out now xdp generic is the default choice if not specified by userspace. In particular after commit
7f0a838254bd("bpf, xdp: Maintain info on attached XDP BPF programs innet_device"), running the command below, XDP will run in generic mode even if the underlay driver support XDP in native mode:$ip link set dev eth0 xdp obj prog.o
$ip link show dev eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 xdpgeneric qdisc mq state UP mode DEFAULT
group default qlen 1024
link/ether f0:ad:4e:09:6b:57 brd ff:ff:ff:ff:ff:ff
prog/xdp id 1 tag3b185187f1855c4cjitedIs it better to use xdpdrv as default choice if not specified by userspace? doing something like:
net/core/dev.c +0
-0@@ -8747,9 +8747,9 @@ static enum bpf_xdp_mode dev_xdp_mode(u32 flags) {
net/core/dev.c static enum bpf_xdp_mode dev_xdp_mode(u32 flags) · line 8754
Quoted 10 lines
net/core/dev.c +3
-3if (flags &XDP_FLAGS_HW_MODE )return XDP_MODE_HW ; -if (flags &XDP_FLAGS_DRV_MODE ) -return XDP_MODE_DRV ; -return XDP_MODE_SKB ; +if (flags &XDP_FLAGS_SKB_MODE ) +return XDP_MODE_SKB ; +return XDP_MODE_DRV ; }
I think the better way would be to choose XDP_MODE_DRV if ndo_bpf != NULL and XDP_MODE_SKB otherwise. That seems to be matching original behavior, no?
It was not my intent to change the behavior, sorry about that. I'll post patch a bit later today.
Quoted 4 lines
net/core/dev.c +0
-0static bpf_op_t dev_xdp_bpf_op (struct net_device *dev,enum bpf_xdp_mode mode)Regards,
Lorenzo
Quoted 32 lines · On Wed, 19 Aug 2020 13:57:51 -0700 Andrii Nakryiko <[email protected]>…
Quoted 31 lines · Lorenzo Bianconi
Hi Andrii,
working on xdp multi-buff I figured out now xdp generic is the default choice if not specified by userspace. In particular after commit
7f0a838254bd("bpf, xdp: Maintain info on attached XDP BPF programs innet_device"), running the command below, XDP will run in generic mode even if the underlay driver support XDP in native mode:$ip link set dev eth0 xdp obj prog.o
$ip link show dev eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 xdpgeneric qdisc mq state UP mode DEFAULT
group default qlen 1024
link/ether f0:ad:4e:09:6b:57 brd ff:ff:ff:ff:ff:ff
prog/xdp id 1 tag3b185187f1855c4cjitedIs it better to use xdpdrv as default choice if not specified by userspace? doing something like:
net/core/dev.c +1
-3@@ -8747,9 +8747,9 @@ static enum bpf_xdp_mode dev_xdp_mode(u32 flags) {if (flags &XDP_FLAGS_HW_MODE )return XDP_MODE_HW ; -if (flags &XDP_FLAGS_DRV_MODE ) -return XDP_MODE_DRV ; -return XDP_MODE_SKB ; +if (flags &XDP_FLAGS_SKB_MODE )
Quoted 8 lines
net/core/dev.c static enum bpf_xdp_mode dev_xdp_mode(u32 flags) · line 8754
Quoted 4 lines
net/core/dev.c +2
-0+return XDP_MODE_SKB ; +return XDP_MODE_DRV ; }I think the better way would be to choose
XDP_MODE_DRVifndo_bpf!= NULL andXDP_MODE_SKBotherwise. That seems to be matching original behavior, no?
Yes, but this silent fallback to XDP_MODE_SKB (generic-XDP) have cause a lot of support issues in the past. I wish we could change it. We already changed all the samples/bpf/ to ask for XDP_FLAGS_DRV_MODE, so they behave this way.
d50ecc46d18f ("samples/bpf: Attach XDP programs in driver mode by default")
https://git.kernel.org/torvalds/c/d50ecc46d18fa
Quoted 2 lines
It was not my intent to change the behavior, sorry about that. I'll post patch a bit later today.
Signature · 4 lines
Best regards, Jesper Dangaard Brouer MSc.CS, Principal Kernel Engineer at Red Hat LinkedIn: http://www.linkedin.com/in/brouer
Quoted 21 lines · Jesper Dangaard Brouer
Quoted 12 lines
Quoted 12 lines
net/core/dev.c +1
-3@@ -8747,9 +8747,9 @@ static enum bpf_xdp_mode dev_xdp_mode(u32 flags) {if (flags &XDP_FLAGS_HW_MODE )return XDP_MODE_HW ; -if (flags &XDP_FLAGS_DRV_MODE ) -return XDP_MODE_DRV ; -return XDP_MODE_SKB ; +if (flags &XDP_FLAGS_SKB_MODE )Quoted 8 lines
net/core/dev.c static enum bpf_xdp_mode dev_xdp_mode(u32 flags) · line 8754
Quoted 4 lines
net/core/dev.c +2
-0+return XDP_MODE_SKB ; +return XDP_MODE_DRV ; }I think the better way would be to choose
XDP_MODE_DRVifndo_bpf!= NULL andXDP_MODE_SKBotherwise. That seems to be matching original behavior, no?
Quoted 5 lines
Yes, but this silent fallback to
XDP_MODE_SKB(generic-XDP) have cause a lot of support issues in the past. I wish we could change it. We already changed all thesamples/bpf/to ask forXDP_FLAGS_DRV_MODE, so they behave this way.
I would prefer the flags check in Lorenzo's proposed patch which is an explicit opt in to the SKB mode.