* [PATCH net-next 1/4] ynl: expose xdp-zc-max-segs
2023-07-25 23:35 [PATCH net-next 0/4] ynl: couple of unrelated fixes Stanislav Fomichev
@ 2023-07-25 23:35 ` Stanislav Fomichev
2023-07-25 23:35 ` [PATCH net-next 2/4] ynl: mark max/mask as private for kdoc Stanislav Fomichev
` (2 subsequent siblings)
3 siblings, 0 replies; 10+ messages in thread
From: Stanislav Fomichev @ 2023-07-25 23:35 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, kuba, pabeni, Stanislav Fomichev
Also rename it to dashes, to match the rest. And fix unrelated
spelling error while we're at it.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
---
Documentation/netlink/specs/netdev.yaml | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml
index e41015310a6e..1c7284fd535b 100644
--- a/Documentation/netlink/specs/netdev.yaml
+++ b/Documentation/netlink/specs/netdev.yaml
@@ -14,7 +14,7 @@ name: netdev
-
name: basic
doc:
- XDP feautues set supported by all drivers
+ XDP features set supported by all drivers
(XDP_ABORTED, XDP_DROP, XDP_PASS, XDP_TX)
-
name: redirect
@@ -63,7 +63,7 @@ name: netdev
enum: xdp-act
enum-as-flags: true
-
- name: xdp_zc_max_segs
+ name: xdp-zc-max-segs
doc: max fragment count supported by ZC driver
type: u32
checks:
@@ -83,6 +83,7 @@ name: netdev
attributes:
- ifindex
- xdp-features
+ - xdp-zc-max-segs
dump:
reply: *dev-all
-
--
2.41.0.487.g6d72f3e995-goog
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH net-next 2/4] ynl: mark max/mask as private for kdoc
2023-07-25 23:35 [PATCH net-next 0/4] ynl: couple of unrelated fixes Stanislav Fomichev
2023-07-25 23:35 ` [PATCH net-next 1/4] ynl: expose xdp-zc-max-segs Stanislav Fomichev
@ 2023-07-25 23:35 ` Stanislav Fomichev
2023-07-26 23:33 ` Jakub Kicinski
2023-07-25 23:35 ` [PATCH net-next 3/4] ynl: regenerate all headers Stanislav Fomichev
2023-07-25 23:35 ` [PATCH net-next 4/4] ynl: print xdp-zc-max-segs in the sample Stanislav Fomichev
3 siblings, 1 reply; 10+ messages in thread
From: Stanislav Fomichev @ 2023-07-25 23:35 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, kuba, pabeni, Stanislav Fomichev
Simon mentioned in another thread that it makes kdoc happy
and Jakub confirms that commit e27cb89a22ad ("scripts: kernel-doc: support
private / public marking for enums") actually added the needed
support.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
---
tools/net/ynl/ynl-gen-c.py | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index 71c5e79e877f..0112722fcc37 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -2124,6 +2124,8 @@ _C_KW = {
cw.p(entry.c_name + suffix)
if const.get('render-max', False):
+ cw.nl()
+ cw.p('/* private: */')
cw.nl()
if const['type'] == 'flags':
max_name = c_upper(name_pfx + 'mask')
--
2.41.0.487.g6d72f3e995-goog
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH net-next 2/4] ynl: mark max/mask as private for kdoc
2023-07-25 23:35 ` [PATCH net-next 2/4] ynl: mark max/mask as private for kdoc Stanislav Fomichev
@ 2023-07-26 23:33 ` Jakub Kicinski
2023-07-26 23:35 ` Stanislav Fomichev
0 siblings, 1 reply; 10+ messages in thread
From: Jakub Kicinski @ 2023-07-26 23:33 UTC (permalink / raw)
To: Stanislav Fomichev; +Cc: netdev, davem, edumazet, pabeni
On Tue, 25 Jul 2023 16:35:15 -0700 Stanislav Fomichev wrote:
> + cw.p('/* private: */')
> cw.nl()
I was about to apply but I keep having doubts whether there should
be a new line after the private marker. I know - a very important
question :D
Quick grep of
$ git grep -A1 '\* private:'
shows that all(?) current uses have something immediately following
the comment.
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH net-next 2/4] ynl: mark max/mask as private for kdoc
2023-07-26 23:33 ` Jakub Kicinski
@ 2023-07-26 23:35 ` Stanislav Fomichev
0 siblings, 0 replies; 10+ messages in thread
From: Stanislav Fomichev @ 2023-07-26 23:35 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: netdev, davem, edumazet, pabeni
On Wed, Jul 26, 2023 at 4:33 PM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Tue, 25 Jul 2023 16:35:15 -0700 Stanislav Fomichev wrote:
> > + cw.p('/* private: */')
> > cw.nl()
>
> I was about to apply but I keep having doubts whether there should
> be a new line after the private marker. I know - a very important
> question :D
>
> Quick grep of
>
> $ git grep -A1 '\* private:'
>
> shows that all(?) current uses have something immediately following
> the comment.
:-D I was too lazy to grep. Will drop \n and respin!
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH net-next 3/4] ynl: regenerate all headers
2023-07-25 23:35 [PATCH net-next 0/4] ynl: couple of unrelated fixes Stanislav Fomichev
2023-07-25 23:35 ` [PATCH net-next 1/4] ynl: expose xdp-zc-max-segs Stanislav Fomichev
2023-07-25 23:35 ` [PATCH net-next 2/4] ynl: mark max/mask as private for kdoc Stanislav Fomichev
@ 2023-07-25 23:35 ` Stanislav Fomichev
2023-07-26 23:37 ` Jakub Kicinski
2023-07-25 23:35 ` [PATCH net-next 4/4] ynl: print xdp-zc-max-segs in the sample Stanislav Fomichev
3 siblings, 1 reply; 10+ messages in thread
From: Stanislav Fomichev @ 2023-07-25 23:35 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, kuba, pabeni, Stanislav Fomichev
Also add small (and simple - no dependencies) makefile rule do update
the UAPI ones.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
---
include/uapi/linux/netdev.h | 4 +++-
tools/net/ynl/Makefile | 10 ++++++++++
tools/net/ynl/generated/netdev-user.c | 6 ++++++
tools/net/ynl/generated/netdev-user.h | 2 ++
4 files changed, 21 insertions(+), 1 deletion(-)
diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h
index bf71698a1e82..1a2f6e320f1c 100644
--- a/include/uapi/linux/netdev.h
+++ b/include/uapi/linux/netdev.h
@@ -11,7 +11,7 @@
/**
* enum netdev_xdp_act
- * @NETDEV_XDP_ACT_BASIC: XDP feautues set supported by all drivers
+ * @NETDEV_XDP_ACT_BASIC: XDP features set supported by all drivers
* (XDP_ABORTED, XDP_DROP, XDP_PASS, XDP_TX)
* @NETDEV_XDP_ACT_REDIRECT: The netdev supports XDP_REDIRECT
* @NETDEV_XDP_ACT_NDO_XMIT: This feature informs if netdev implements
@@ -34,6 +34,8 @@ enum netdev_xdp_act {
NETDEV_XDP_ACT_RX_SG = 32,
NETDEV_XDP_ACT_NDO_XMIT_SG = 64,
+ /* private: */
+
NETDEV_XDP_ACT_MASK = 127,
};
diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile
index d664b36deb5b..f4eacafa9665 100644
--- a/tools/net/ynl/Makefile
+++ b/tools/net/ynl/Makefile
@@ -1,5 +1,10 @@
# SPDX-License-Identifier: GPL-2.0
+DIR_SPEC:=../../../Documentation/netlink/specs
+DIR_UAPI:=../../../include/uapi/linux
+SPECS:=netdev fou handshake
+TOOL:=./ynl-gen-c.py
+
SUBDIRS = lib generated samples
all: $(SUBDIRS)
@@ -16,4 +21,9 @@ all: $(SUBDIRS)
fi \
done
+generate:
+ for spec in $(SPECS); do \
+ $(TOOL) --spec $(DIR_SPEC)/$$spec.yaml --header --mode uapi > $(DIR_UAPI)/$$spec.h; \
+ done
+
.PHONY: clean all $(SUBDIRS)
diff --git a/tools/net/ynl/generated/netdev-user.c b/tools/net/ynl/generated/netdev-user.c
index 4eb8aefef0cd..68b408ca0f7f 100644
--- a/tools/net/ynl/generated/netdev-user.c
+++ b/tools/net/ynl/generated/netdev-user.c
@@ -50,6 +50,7 @@ struct ynl_policy_attr netdev_dev_policy[NETDEV_A_DEV_MAX + 1] = {
[NETDEV_A_DEV_IFINDEX] = { .name = "ifindex", .type = YNL_PT_U32, },
[NETDEV_A_DEV_PAD] = { .name = "pad", .type = YNL_PT_IGNORE, },
[NETDEV_A_DEV_XDP_FEATURES] = { .name = "xdp-features", .type = YNL_PT_U64, },
+ [NETDEV_A_DEV_XDP_ZC_MAX_SEGS] = { .name = "xdp-zc-max-segs", .type = YNL_PT_U32, },
};
struct ynl_policy_nest netdev_dev_nest = {
@@ -91,6 +92,11 @@ int netdev_dev_get_rsp_parse(const struct nlmsghdr *nlh, void *data)
return MNL_CB_ERROR;
dst->_present.xdp_features = 1;
dst->xdp_features = mnl_attr_get_u64(attr);
+ } else if (type == NETDEV_A_DEV_XDP_ZC_MAX_SEGS) {
+ if (ynl_attr_validate(yarg, attr))
+ return MNL_CB_ERROR;
+ dst->_present.xdp_zc_max_segs = 1;
+ dst->xdp_zc_max_segs = mnl_attr_get_u32(attr);
}
}
diff --git a/tools/net/ynl/generated/netdev-user.h b/tools/net/ynl/generated/netdev-user.h
index 5554dc69bb9c..0952d3261f4d 100644
--- a/tools/net/ynl/generated/netdev-user.h
+++ b/tools/net/ynl/generated/netdev-user.h
@@ -47,10 +47,12 @@ struct netdev_dev_get_rsp {
struct {
__u32 ifindex:1;
__u32 xdp_features:1;
+ __u32 xdp_zc_max_segs:1;
} _present;
__u32 ifindex;
__u64 xdp_features;
+ __u32 xdp_zc_max_segs;
};
void netdev_dev_get_rsp_free(struct netdev_dev_get_rsp *rsp);
--
2.41.0.487.g6d72f3e995-goog
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH net-next 3/4] ynl: regenerate all headers
2023-07-25 23:35 ` [PATCH net-next 3/4] ynl: regenerate all headers Stanislav Fomichev
@ 2023-07-26 23:37 ` Jakub Kicinski
2023-07-26 23:55 ` Stanislav Fomichev
0 siblings, 1 reply; 10+ messages in thread
From: Jakub Kicinski @ 2023-07-26 23:37 UTC (permalink / raw)
To: Stanislav Fomichev; +Cc: netdev, davem, edumazet, pabeni
On Tue, 25 Jul 2023 16:35:16 -0700 Stanislav Fomichev wrote:
> Also add small (and simple - no dependencies) makefile rule do update
> the UAPI ones.
I was thinking people would use:
./tools/net/ynl/ynl-regen.sh -f
for this. It is slightly more capable. Can we perhaps hook the new make
target to just run that script?
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 3/4] ynl: regenerate all headers
2023-07-26 23:37 ` Jakub Kicinski
@ 2023-07-26 23:55 ` Stanislav Fomichev
2023-07-27 0:22 ` Jakub Kicinski
0 siblings, 1 reply; 10+ messages in thread
From: Stanislav Fomichev @ 2023-07-26 23:55 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: netdev, davem, edumazet, pabeni
On Wed, Jul 26, 2023 at 4:37 PM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Tue, 25 Jul 2023 16:35:16 -0700 Stanislav Fomichev wrote:
> > Also add small (and simple - no dependencies) makefile rule do update
> > the UAPI ones.
>
> I was thinking people would use:
>
> ./tools/net/ynl/ynl-regen.sh -f
>
> for this. It is slightly more capable. Can we perhaps hook the new make
> target to just run that script?
Oh, didn't know about this. Something like this maybe? Ugly?
diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile
index d664b36deb5b..c36380bf1536 100644
--- a/tools/net/ynl/Makefile
+++ b/tools/net/ynl/Makefile
@@ -3,6 +3,7 @@
SUBDIRS = lib generated samples
all: $(SUBDIRS)
+ (cd ../../../ && ./tools/net/ynl/ynl-regen.sh -f)
$(SUBDIRS):
@if [ -f "$@/Makefile" ] ; then \
Or, now that I know about the script I can actually run it manually.
But with a makefile imo a bit easier to discover..
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH net-next 3/4] ynl: regenerate all headers
2023-07-26 23:55 ` Stanislav Fomichev
@ 2023-07-27 0:22 ` Jakub Kicinski
0 siblings, 0 replies; 10+ messages in thread
From: Jakub Kicinski @ 2023-07-27 0:22 UTC (permalink / raw)
To: Stanislav Fomichev; +Cc: netdev, davem, edumazet, pabeni
On Wed, 26 Jul 2023 16:55:07 -0700 Stanislav Fomichev wrote:
> Oh, didn't know about this. Something like this maybe? Ugly?
>
> diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile
> index d664b36deb5b..c36380bf1536 100644
> --- a/tools/net/ynl/Makefile
> +++ b/tools/net/ynl/Makefile
> @@ -3,6 +3,7 @@
> SUBDIRS = lib generated samples
>
> all: $(SUBDIRS)
> + (cd ../../../ && ./tools/net/ynl/ynl-regen.sh -f)
Hm, I thought the script itself would handle this. I guess I did a
stupid. How about we extend the script to default to the full tree:
diff --git a/tools/net/ynl/ynl-regen.sh b/tools/net/ynl/ynl-regen.sh
index 8d4ca6a50582..bdba24066cf1 100755
--- a/tools/net/ynl/ynl-regen.sh
+++ b/tools/net/ynl/ynl-regen.sh
@@ -4,15 +4,18 @@
TOOL=$(dirname $(realpath $0))/ynl-gen-c.py
force=
+search=
while [ ! -z "$1" ]; do
case "$1" in
-f ) force=yes; shift ;;
+ -p ) search=$2; shift 2 ;;
* ) echo "Unrecognized option '$1'"; exit 1 ;;
esac
done
KDIR=$(dirname $(dirname $(dirname $(dirname $(realpath $0)))))
+pushd ${search:-$KDIR} >>/dev/null
files=$(git grep --files-with-matches '^/\* YNL-GEN \(kernel\|uapi\|user\)')
for f in $files; do
@@ -30,3 +33,5 @@ for f in $files; do
$TOOL --mode ${params[2]} --${params[3]} --spec $KDIR/${params[0]} \
$args -o $f
done
+
+popd >>/dev/null
> $(SUBDIRS):
> @if [ -f "$@/Makefile" ] ; then \
>
> Or, now that I know about the script I can actually run it manually.
> But with a makefile imo a bit easier to discover..
Yea, agreed.
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 4/4] ynl: print xdp-zc-max-segs in the sample
2023-07-25 23:35 [PATCH net-next 0/4] ynl: couple of unrelated fixes Stanislav Fomichev
` (2 preceding siblings ...)
2023-07-25 23:35 ` [PATCH net-next 3/4] ynl: regenerate all headers Stanislav Fomichev
@ 2023-07-25 23:35 ` Stanislav Fomichev
3 siblings, 0 replies; 10+ messages in thread
From: Stanislav Fomichev @ 2023-07-25 23:35 UTC (permalink / raw)
To: netdev; +Cc: davem, edumazet, kuba, pabeni, Stanislav Fomichev
Technically we don't have to keep extending the sample, but it
feels useful to run these tools locally to confirm everything
is working.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
---
tools/net/ynl/samples/netdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tools/net/ynl/samples/netdev.c b/tools/net/ynl/samples/netdev.c
index d31268aa47c5..06433400dddd 100644
--- a/tools/net/ynl/samples/netdev.c
+++ b/tools/net/ynl/samples/netdev.c
@@ -38,6 +38,8 @@ static void netdev_print_device(struct netdev_dev_get_rsp *d, unsigned int op)
printf(" %s", netdev_xdp_act_str(1 << i));
}
+ printf(" xdp-zc-max-segs=%u", d->xdp_zc_max_segs);
+
name = netdev_op_str(op);
if (name)
printf(" (ntf: %s)", name);
--
2.41.0.487.g6d72f3e995-goog
^ permalink raw reply related [flat|nested] 10+ messages in thread