max_addr is the max number of addresses, not the highest possible address,
therefore check phydev->mdio.addr > max_addr isn't correct.
To fix this change the semantics of max_addr, so that it represents
the highest possible address. IMO this is also a little bit more intuitive
wrt name max_addr.
Fixes: 4a107a0e83 ("net: stmmac: mdio: use phy_find_first to simplify stmmac_mdio_register")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Reported-by: Simon Horman <horms@kernel.org>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/e869999b-2d4b-4dc1-9890-c2d3d1e8d0f8@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Wei Fang says:
====================
net: enetc: Add i.MX94 ENETC support
i.MX94 NETC has two kinds of ENETCs, one is the same as i.MX95, which
can be used as a standalone network port. The other one is an internal
ENETC, it connects to the CPU port of NETC switch through the pseudo
MAC. Also, i.MX94 have multiple PTP Timers, which is different from
i.MX95. Any PTP Timer can be bound to a specified standalone ENETC by
the IERB ETBCR registers. Currently, this patch only add ENETC support
and Timer support for i.MX94. The switch will be added by a separate
patch set.
In addition, note that i.MX94 SoC is launched after i.MX95, its NETC
has a higher version, so the driver support is added after i.MX95.
====================
Link: https://patch.msgid.link/20251029013900.407583-1-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The ENETC with pseudo MAC is an internal port which connects to the CPU
port of the switch. The switch CPU/host ENETC is fully integrated with
the switch and does not require a back-to-back MAC, instead a light
weight "pseudo MAC" provides the delineation between switch and ENETC.
This translates to lower power (less logic and memory) and lower delay
(as there is no serialization delay across this link).
Different from the standalone ENETC which is used as the external port,
the internal ENETC has a different PCIe device ID, and it does not have
Ethernet MAC port registers, instead, it has a small number of pseudo
MAC port registers, so some features are not supported by pseudo MAC,
such as loopback, half duplex, one-step timestamping and so on.
Therefore, the configuration of this internal ENETC is also somewhat
different from that of the standalone ENETC. So add the basic support
for ENETC with pseudo MAC. More supports will be added in the future.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Link: https://patch.msgid.link/20251029013900.407583-6-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The i.MX94 has three PTP timers, and all standalone ENETCs can select
one of them to bind to as their PHC. The 'ptp-timer' property is used
to represent the PTP device of the Ethernet controller. So users can
add 'ptp-timer' to the ENETC node to specify the PTP timer. The driver
parses this property to bind the two hardware devices.
If the "ptp-timer" property is not present, the first timer of the PCIe
bus where the ENETC is located is used as the default bound PTP timer.
Signed-off-by: Clark Wang <xiaoning.wang@nxp.com>
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20251029013900.407583-5-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
NETC blocks control is used for warm reset and pre-boot initialization.
Different versions of NETC blocks control are not exactly the same. We
need to add corresponding netc_devinfo data for each version. i.MX94
series are launched after i.MX95, so its NETC version (v4.3) is higher
than i.MX95 NETC (v4.1). Currently, the patch adds the following
configurations for ENETCs.
1. Set the link's MII protocol.
2. ENETC 0 (MAC 3) and the switch port 2 (MAC 2) share the same parallel
interface, but due to SoC constraint, they cannot be used simultaneously.
Since the switch is not supported yet, so the interface is assigned to
ENETC 0 by default.
The switch configuration will be added separately in a subsequent patch.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251029013900.407583-4-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Ido Schimmel says:
====================
icmp: Add RFC 5837 support
tl;dr
=====
This patchset extends certain ICMP error messages (e.g., "Time
Exceeded") with incoming interface information in accordance with RFC
5837 [1]. This is required for more meaningful traceroute results in
unnumbered networks. Like other ICMP settings, the feature is controlled
via a per-{netns, address family} sysctl. The interface and the
implementation are designed to support more ICMP extensions.
Motivation
==========
Over the years, the kernel was extended with the ability to derive the
source IP of ICMP error messages from the interface that received the
datagram which elicited the ICMP error [2][3][4]. This is especially
important for "Time Exceeded" messages as it allows traceroute users to
trace the actual packet path along the network.
The above scheme does not work in unnumbered networks. In these
networks, only the loopback / VRF interface is assigned a global IP
address while router interfaces are assigned IPv6 link-local addresses.
As such, ICMP error messages are generated with a source IP derived from
the loopback / VRF interface, making it impossible to trace the actual
packet path when parallel links exist between routers.
The problem can be solved by implementing the solution proposed by RFC
4884 [5] and RFC 5837. The former defines an ICMP extension structure
that can be appended to selected ICMP messages and carry extension
objects. The latter defines an extension object called the "Interface
Information Object" (IIO) that can carry interface information (e.g.,
name, index, MTU) about interfaces with certain roles such as the
interface that received the datagram which elicited the ICMP error.
The payload of the datagram that elicited the error (potentially padded
/ trimmed) along with the ICMP extension structure will be queued to the
error queue of the originating socket, thereby allowing traceroute
applications to parse and display the information encoded in the ICMP
extension structure. Example:
# traceroute6 -e 2001:db8:1::3
traceroute to 2001:db8:1::3 (2001:db8:1::3), 30 hops max, 80 byte packets
1 2001:db8:1::2 (2001:db8:1::2) <INC:11,"eth1",mtu=1500> 0.214 ms 0.171 ms 0.162 ms
2 2001:db8:1::3 (2001:db8:1::3) <INC:12,"eth2",mtu=1500> 0.154 ms 0.135 ms 0.127 ms
# traceroute -e 192.0.2.3
traceroute to 192.0.2.3 (192.0.2.3), 30 hops max, 60 byte packets
1 192.0.2.2 (192.0.2.2) <INC:11,"eth1",mtu=1500> 0.191 ms 0.148 ms 0.144 ms
2 192.0.2.3 (192.0.2.3) <INC:12,"eth2",mtu=1500> 0.137 ms 0.122 ms 0.114 ms
Implementation
==============
As previously stated, the feature is controlled via a per-{netns,
address} sysctl. Specifically, a bit mask where each bit controls the
addition of a different ICMP extension to ICMP error messages.
Currently, only a single value is supported, to append the incoming
interface information.
Key points:
1. Global knob vs finer control. I am not aware of users who require
finer control, but it is possible that some users will want to avoid
appending ICMP extensions when the packet is sent out of a specific
interface (e.g., the management interface) or to a specific subnet. This
can be accomplished via a tc-bpf program that trims the ICMP extension
structure. An example program can be found here [6].
2. Split implementation between IPv4 / IPv6. While the implementation is
currently similar, there are some differences between both address
families. In addition, some extensions (e.g., RFC 8883 [7]) are
IPv6-specific. Given the above and given that the implementation is not
very complex, it makes sense to keep both implementations separate.
3. Compatibility with legacy applications. RFC 4884 from 2007 extended
certain ICMP messages with a length field that encodes the length of the
"original datagram" field, so that applications will be able to tell
where the "original datagram" ends and where the ICMP extension
structure starts.
Before the introduction of the IP{,6}_RECVERR_RFC4884 socket options
[8][9] in 2020 it was impossible for applications to know where the ICMP
extension structure starts and to this day some applications assume that
it starts at offset 128, which is the minimum length of the "original
datagram" field as specified by RFC 4884.
Therefore, in order to be compatible with both legacy and modern
applications, the datagram that elicited the ICMP error is trimmed /
padded to 128 bytes before appending the ICMP extension structure.
This behavior is specifically called out by RFC 4884: "Those wishing to
be backward compatible with non-compliant TRACEROUTE implementations
will include exactly 128 octets" [10].
Note that in 128 bytes we should be able to include enough headers for
the originating node to match the ICMP error message with the relevant
socket. For example, the following headers will be present in the
"original datagram" field when a VXLAN encapsulated IPv6 packet elicits
an ICMP error in an IPv6 underlay: IPv6 (40) | UDP (8) | VXLAN (8) | Eth
(14) | IPv6 (40) | UDP (8). Overall, 118 bytes.
If the 128 bytes limit proves to be insufficient for some use case, we
can consider dedicating a new bit in the previously mentioned sysctl to
allow for more bytes to be included in the "original datagram" field.
4. Extensibility. This patchset adds partial support for a single ICMP
extension. However, the interface and the implementation should be able
to support more extensions, if needed. Examples:
* More interface information objects as part of RFC 5837. We should be
able to derive the outgoing interface information and nexthop IP from
the dst entry attached to the packet that elicited the error.
* Node identification object (e.g., hostname / loopback IP) [11].
* Extended Information object which encodes aggregate header limits as
part of RFC 8883.
A previous proposal from Ishaan Gandhi and Ron Bonica is available here
[12].
Testing
=======
The existing traceroute selftest is extended to test that ICMP
extensions are reported correctly when enabled. Both address families
are tested and with different packet sizes in order to make sure that
trimming / padding works correctly. Tested that packets are parsed
correctly by the IP{,6}_RECVERR_RFC4884 socket options using Willem's
selftest [13].
Changelog
=========
Changes since v1 [14]:
* Patches #1-#2: Added a comment about field ordering and review tags.
* Patch #3: Converted "sysctl" to "echo" when testing the return value.
Added a check to skip the test if traceroute version is older
than 2.1.5.
[1] https://datatracker.ietf.org/doc/html/rfc5837
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=1c2fb7f93cb20621772bf304f3dba0849942e5db
[3] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=fac6fce9bdb59837bb89930c3a92f5e0d1482f0b
[4] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=4a8c416602d97a4e2073ed563d4d4c7627de19cf
[5] https://datatracker.ietf.org/doc/html/rfc4884
[6] https://gist.github.com/idosch/5013448cdb5e9e060e6bfdc8b433577c
[7] https://datatracker.ietf.org/doc/html/rfc8883
[8] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=eba75c587e811d3249c8bd50d22bb2266ccd3c0f
[9] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=01370434df85eb76ecb1527a4466013c4aca2436
[10] https://datatracker.ietf.org/doc/html/rfc4884#section-5.3
[11] https://datatracker.ietf.org/doc/html/draft-ietf-intarea-extended-icmp-nodeid-04
[12] https://lore.kernel.org/netdev/20210317221959.4410-1-ishaangandhi@gmail.com/
[13] https://lore.kernel.org/netdev/aPpMItF35gwpgzZx@shredder/
[14] https://lore.kernel.org/netdev/20251022065349.434123-1-idosch@nvidia.com/
====================
Link: https://patch.msgid.link/20251027082232.232571-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Test that ICMP extensions are reported correctly when enabled and not
reported when disabled. Test both IPv4 and IPv6 and using different
packet sizes, to make sure trimming / padding works correctly.
Disable ICMP rate limiting (defaults to 1 per-second per-target) so that
the kernel will always generate ICMP errors when needed.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20251027082232.232571-4-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add the ability to append the incoming IP interface information to
ICMPv6 error messages in accordance with RFC 5837 and RFC 4884. This is
required for more meaningful traceroute results in unnumbered networks.
The feature is disabled by default and controlled via a new sysctl
("net.ipv6.icmp.errors_extension_mask") which accepts a bitmask of ICMP
extensions to append to ICMP error messages. Currently, only a single
value is supported, but the interface and the implementation should be
able to support more extensions, if needed.
Clone the skb and copy the relevant data portions before modifying the
skb as the caller of icmp6_send() still owns the skb after the function
returns. This should be fine since by default ICMP error messages are
rate limited to 1000 per second and no more than 1 per second per
specific host.
Trim or pad the packet to 128 bytes before appending the ICMP extension
structure in order to be compatible with legacy applications that assume
that the ICMP extension structure always starts at this offset (the
minimum length specified by RFC 4884).
Since commit 20e1954fe2 ("ipv6: RFC 4884 partial support for SIT/GRE
tunnels") it is possible for icmp6_send() to be called with an skb that
already contains ICMP extensions. This can happen when we receive an
ICMPv4 message with extensions from a tunnel and translate it to an
ICMPv6 message towards an IPv6 host in the overlay network. I could not
find an RFC that supports this behavior, but it makes sense to not
overwrite the original extensions that were appended to the packet.
Therefore, avoid appending extensions if the length field in the
provided ICMPv6 header is already filled.
Export netdev_copy_name() using EXPORT_IPV6_MOD_GPL() to make it
available to IPv6 when it is built as a module.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251027082232.232571-3-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add the ability to append the incoming IP interface information to
ICMPv4 error messages in accordance with RFC 5837 and RFC 4884. This is
required for more meaningful traceroute results in unnumbered networks.
The feature is disabled by default and controlled via a new sysctl
("net.ipv4.icmp_errors_extension_mask") which accepts a bitmask of ICMP
extensions to append to ICMP error messages. Currently, only a single
value is supported, but the interface and the implementation should be
able to support more extensions, if needed.
Clone the skb and copy the relevant data portions before modifying the
skb as the caller of __icmp_send() still owns the skb after the function
returns. This should be fine since by default ICMP error messages are
rate limited to 1000 per second and no more than 1 per second per
specific host.
Trim or pad the packet to 128 bytes before appending the ICMP extension
structure in order to be compatible with legacy applications that assume
that the ICMP extension structure always starts at this offset (the
minimum length specified by RFC 4884).
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20251027082232.232571-2-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Socket APIs like recvfrom(), accept(), and getsockname() expect socklen_t*
arg, but tests were using int variables. This causes -Wpointer-sign
warnings on platforms where socklen_t is unsigned.
Change the variable type from int to socklen_t to resolve the warning and
ensure type safety across platforms.
warning fixed:
sctp_collision.c:62:70: warning: passing 'int *' to parameter of
type 'socklen_t *' (aka 'unsigned int *') converts between pointers to
integer types with different sign [-Wpointer-sign]
62 | ret = recvfrom(sd, buf, sizeof(buf),
0, (struct sockaddr *)&daddr, &len);
| ^~~~
/usr/include/sys/socket.h:165:27: note: passing argument to
parameter '__addr_len' here
165 | socklen_t *__restrict __addr_len);
| ^
Reviewed-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: Ankit Khushwaha <ankitkhushwaha.linux@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20251028172947.53153-1-ankitkhushwaha.linux@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Russell King says:
====================
net: stmmac: hwif.c cleanups
This series cleans up hwif.c:
- move the reading of the version information out of stmmac_hwif_init()
into its own function, stmmac_get_version(), storing the result in a
new struct.
- simplify stmmac_get_version().
- read the version register once, passing it to stmmac_get_id() and
stmmac_get_dev_id().
- move stmmac_get_id() and stmmac_get_dev_id() into
stmmac_get_version()
- define version register fields and use FIELD_GET() to decode
- start tackling the big loop in stmmac_hwif_init() - provide a
function, stmmac_hwif_find(), which looks up the hwif entry, thus
making a much smaller loop, which improves readability of this code.
- change the use of '^' to '!=' when comparing the dev_id, which is
what is really meant here.
- reorganise the test after calling stmmac_hwif_init() so that we
handle the error case in the indented code, and the success case
with no indent, which is the classical arrangement.
====================
Link: https://patch.msgid.link/aQFZVSGJuv8-_DIo@shell.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Function udp_v4_early_demux() was already declared in 'include/net/udp.h',
no need to keep the extern in 'ip_input.c', which may produce the
following checkpatch warning:
WARNING: externs should be avoided in .c files
#45: FILE: net/ipv4/ip_input.c:322:
+enum skb_drop_reason udp_v4_early_demux(struct sk_buff *skb);
Replace it by including 'net/udp.h'. Do the same for tcp_v4_early_demux().
Signed-off-by: Wang Liang <wangliang74@huawei.com>
Link: https://patch.msgid.link/20251025092637.1020960-1-wangliang74@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The RTL8224 can detect open pairs and short types (in same pair or some
other pair). The distance to this problem can be estimated. This is done
for each of the 4 pairs separately.
It is not meant to be run while there is an active link partner because
this interferes with the active test pulses.
Output with open 50 m cable:
Pair A code Open Circuit, source: TDR
Pair A, fault length: 51.79m, source: TDR
Pair B code Open Circuit, source: TDR
Pair B, fault length: 51.28m, source: TDR
Pair C code Open Circuit, source: TDR
Pair C, fault length: 50.46m, source: TDR
Pair D code Open Circuit, source: TDR
Pair D, fault length: 51.12m, source: TDR
Terminated cable:
Pair A code OK, source: TDR
Pair B code OK, source: TDR
Pair C code OK, source: TDR
Pair D code OK, source: TDR
Shorted cable (both short types are at roughly the same distance)
Pair A code Short to another pair, source: TDR
Pair A, fault length: 2.35m, source: TDR
Pair B code Short to another pair, source: TDR
Pair B, fault length: 2.15m, source: TDR
Pair C code OK, source: TDR
Pair D code Short within Pair, source: TDR
Pair D, fault length: 1.94m, source: TDR
Signed-off-by: Issam Hamdi <ih@simonwunderlich.de>
Co-developed-by: Sven Eckelmann <se@simonwunderlich.de>
Signed-off-by: Sven Eckelmann <se@simonwunderlich.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20251024-rtl8224-cable-test-v1-1-e3cda89ac98f@simonwunderlich.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Tony Nguyen says:
====================
ice: postpone service task disabling
Przemek Kitszel says:
Move service task shutdown to the very end of driver teardown procedure.
This is needed (or at least beneficial) for all unwinding functions that
talk to FW/HW via Admin Queue (so, most of top-level functions, like
ice_deinit_hw()).
Most of the patches move stuff around (I believe it makes it much easier
to review/proof when kept separate) in preparation to defer stopping the
service task to the very end of ice_remove() (and other unwinding flows).
Then last patch fixes duplicate call to ice_init_hw() (actual, but
unlikely to encounter, so -next, given the size of the changes).
First patch is not much related, only by that it was developed together
* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
ice: remove duplicate call to ice_deinit_hw() on error paths
ice: move ice_deinit_dev() to the end of deinit paths
ice: extract ice_init_dev() from ice_init()
ice: move ice_init_pf() out of ice_init_dev()
ice: move udp_tunnel_nic and misc IRQ setup into ice_init_pf()
ice: ice_init_pf: destroy mutexes and xarrays on memory alloc failure
ice: move ice_init_interrupt_scheme() prior ice_init_pf()
ice: move service task start out of ice_init_pf()
ice: enforce RTNL assumption of queue NAPI manipulation
====================
Link: https://patch.msgid.link/20251024204746.3092277-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Fix kernel-doc warnings in tcp_lp.c by adding missing parameter
descriptions for tcp_lp_cong_avoid() and tcp_lp_pkts_acked() when
building with W=1.
Also replace invalid URLs in the file header comment with the currently
valid links to the TCP-LP paper and implementation page.
No functional changes.
Signed-off-by: Rakuram Eswaran <rakuram.e96@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20251025-net_ipv4_tcp_lp_c-v1-1-058cc221499e@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
'struct sctp_sched_ops' is not modified in these drivers.
Constifying this structure moves some data to a read-only section, so
increases overall security, especially when the structure holds some
function pointers.
On a x86_64, with allmodconfig, as an example:
Before:
======
text data bss dec hex filename
8019 568 0 8587 218b net/sctp/stream_sched_fc.o
After:
=====
text data bss dec hex filename
8275 312 0 8587 218b net/sctp/stream_sched_fc.o
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://patch.msgid.link/dce03527eb7b7cc8a3c26d5cdac12bafe3350135.1761377890.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Remove the NET_IOV_MAX workaround from the net_iov_type enum. This entry
was previously added to force the enum size to unsigned long to satisfy
the NET_IOV_ASSERT_OFFSET static assertions.
After commit f3d85c9ee5 ("netmem: introduce struct netmem_desc
mirroring struct page") this approach became unnecessary by placing the
net_iov_type after the netmem_desc. Placing the net_iov_type after
netmem_desc results in the net_iov_type size having no effect on the
position or layout of the fields that mirror the struct page.
The layout before this patch:
struct net_iov {
union {
struct netmem_desc desc; /* 0 48 */
struct {
long unsigned int _flags; /* 0 8 */
long unsigned int pp_magic; /* 8 8 */
struct page_pool * pp; /* 16 8 */
long unsigned int _pp_mapping_pad; /* 24 8 */
long unsigned int dma_addr; /* 32 8 */
atomic_long_t pp_ref_count; /* 40 8 */
}; /* 0 48 */
}; /* 0 48 */
struct net_iov_area * owner; /* 48 8 */
enum net_iov_type type; /* 56 8 */
/* size: 64, cachelines: 1, members: 3 */
};
The layout after this patch:
struct net_iov {
union {
struct netmem_desc desc; /* 0 48 */
struct {
long unsigned int _flags; /* 0 8 */
long unsigned int pp_magic; /* 8 8 */
struct page_pool * pp; /* 16 8 */
long unsigned int _pp_mapping_pad; /* 24 8 */
long unsigned int dma_addr; /* 32 8 */
atomic_long_t pp_ref_count; /* 40 8 */
}; /* 0 48 */
}; /* 0 48 */
struct net_iov_area * owner; /* 48 8 */
enum net_iov_type type; /* 56 4 */
/* size: 64, cachelines: 1, members: 3 */
/* padding: 4 */
};
Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/20251024-b4-devmem-remove-niov-max-v1-1-ba72c68bc869@meta.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
enqueue_to_backlog() is showing up in kernel profiles on hosts
with many cores, when RFS/RPS is used.
The following softnet_data fields need to be updated:
- input_queue_tail
- input_pkt_queue (next, prev, qlen, lock)
- backlog.state (if input_pkt_queue was empty)
Unfortunately they are currenly using two cache lines:
/* --- cacheline 3 boundary (192 bytes) --- */
call_single_data_t csd __attribute__((__aligned__(64))); /* 0xc0 0x20 */
struct softnet_data * rps_ipi_next; /* 0xe0 0x8 */
unsigned int cpu; /* 0xe8 0x4 */
unsigned int input_queue_tail; /* 0xec 0x4 */
struct sk_buff_head input_pkt_queue; /* 0xf0 0x18 */
/* --- cacheline 4 boundary (256 bytes) was 8 bytes ago --- */
struct napi_struct backlog __attribute__((__aligned__(8))); /* 0x108 0x1f0 */
Add one ____cacheline_aligned_in_smp to make sure they now are using
a single cache line.
Also, because napi_struct has written fields, make @state its first field.
We want to make sure that cpus adding packets to sd->input_pkt_queue
are not slowing down cpus processing their backlog because of
false sharing.
After this patch new layout is:
/* --- cacheline 5 boundary (320 bytes) --- */
long int pad[3] __attribute__((__aligned__(64))); /* 0x140 0x18 */
unsigned int input_queue_tail; /* 0x158 0x4 */
/* XXX 4 bytes hole, try to pack */
struct sk_buff_head input_pkt_queue; /* 0x160 0x18 */
struct napi_struct backlog __attribute__((__aligned__(8))); /* 0x178 0x1f0 */
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20251024091240.3292546-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add likely() and unlikely() clauses for the common cases:
Device is running.
Queue is not full.
Queue is less than half capacity.
Add max_backlog parameter to skb_flow_limit() to avoid
a second READ_ONCE(net_hotdata.max_backlog).
skb_flow_limit() does not need the backlog_lock protection,
and can be called before we acquire the lock, for even better
resistance to attacks.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20251024090517.3289181-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In early days of YNL development dumping the NlMsg on errors
was quite useful, as the library itself could have been buggy.
These days increasingly the NlMsg is just taking up screen space
and means nothing to a typical user. Try to format the errors
more in line with how YNL C formats its errors strings.
Before:
$ ynl --family ethtool --do channels-set --json '{}'
Netlink error: Invalid argument
nl_len = 44 (28) nl_flags = 0x300 nl_type = 2
error: -22
extack: {'miss-type': 'header'}
$ ynl --family ethtool --do channels-set --json '{..., "tx-count": 999}'
Netlink error: Invalid argument
nl_len = 88 (72) nl_flags = 0x300 nl_type = 2
error: -22
extack: {'msg': 'requested channel count exceeds maximum', 'bad-attr': '.tx-count'}
After:
$ ynl --family ethtool --do channels-set --json '{}'
Netlink error: Invalid argument {'miss-type': 'header'}
$ ynl --family ethtool --do channels-set --json '{..., "tx-count": 999}'
Netlink error: requested channel count exceeds maximum: Invalid argument {'bad-attr': '.tx-count'}
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20251027192958.2058340-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The DWMAC1000 supports 2 timestamping configurations to configure how
frequency adjustments are made to the ptp_clock, as well as the reported
timestamp values.
There was a previous attempt at upstreaming support for configuring this
mode by Olivier Dautricourt and Julien Beraud a few years back [1]
In a nutshell, the timestamping can be either set in fine mode or in
coarse mode.
In fine mode, which is the default, we use the overflow of an accumulator to
trigger frequency adjustments, but by doing so we lose precision on the
timetamps that are produced by the timestamping unit. The main drawback
is that the sub-second increment value, used to generate timestamps, can't be
set to lower than (2 / ptp_clock_freq).
The "fine" qualification comes from the frequent frequency adjustments we are
able to do, which is perfect for a PTP follower usecase.
In Coarse mode, we don't do frequency adjustments based on an
accumulator overflow. We can therefore have very fine subsecond
increment values, allowing for better timestamping precision. However
this mode works best when the ptp clock frequency is adjusted based on
an external signal, such as a PPS input produced by a GPS clock. This
mode is therefore perfect for a Grand-master usecase.
Introduce a driver-specific devlink parameter "ts_coarse" to enable or
disable coarse mode, keeping the "fine" mode as a default.
This can then be changed with:
devlink dev param set <dev> name ts_coarse value true cmode runtime
The associated documentation is also added.
[1] : https://lore.kernel.org/netdev/20200514102808.31163-1-olivier.dautricourt@orolia.com/
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20251024070720.71174-3-maxime.chevallier@bootlin.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
says:
====================
net: macb: EyeQ5 support
This series' goal is adding support to the MACB driver for EyeQ5 GEM.
The specifics for this compatible are:
- HW cannot add dummy bytes at the start of IP packets for alignment
purposes. The behavior can be detected using DCFG6 so it isn't
attached to compatible data.
- The hardware LSO/TSO is known to be buggy: add a compatible
capability flag to force disable it.
- At init, we have to wiggle two syscon registers that configure the
PHY integration.
In past attempts [0] we did it in macb_config->init() using a syscon
regmap. That was far from ideal so now a generic PHY driver
abstracts that away. We reuse the bp->sgmii_phy field used by some
compatibles.
We have to add a phy_set_mode() call as the PHY power on sequence
depends on whether we do RGMII or SGMII.
[0]: https://lore.kernel.org/lkml/20250627-macb-v2-15-ff8207d0bb77@bootlin.com/
Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com>
---
Changes in v3:
- Drop Fixes: trailer on [2/5]. We don't fix any platform using the
driver currently.
- Improve [5/5] commit message; add info about how an unconditional
phy_set_mode_ext() won't break existing platforms.
- Hardbreak 82 characters line in [2/5]; warning by patchwork.
- Trailers:
- 1x Acked-by: Conor Dooley on [1/5].
- 2x Reviewed-by: Andrew Lunn on [1/5] and [4/5].
- 2x Reviewed-by: Maxime Chevallier on [4/5] and [5/5].
- Link to v2: https://lore.kernel.org/r/20251022-macb-eyeq5-v2-0-7c140abb0581@bootlin.com
Changes in v2:
- Drop non net-next patches.
- Re-run get_maintainers.pl to shorten the To/Cc list.
- Rebase upon latest net-next; no changes. Tested on HW.
- Link to v1: https://lore.kernel.org/r/20251021-macb-eyeq5-v1-0-3b0b5a9d2f85@bootlin.com
Past versions of the MACB EyeQ5 patches:
- March 2025: [PATCH net-next 00/13] Support the Cadence MACB/GEM
instances on Mobileye EyeQ5 SoCs
https://lore.kernel.org/lkml/20250321-macb-v1-0-537b7e37971d@bootlin.com/
- June 2025: [PATCH net-next v2 00/18] Support the Cadence MACB/GEM
instances on Mobileye EyeQ5 SoCs
https://lore.kernel.org/lkml/20250627-macb-v2-0-ff8207d0bb77@bootlin.com/
- August 2025: [PATCH net v3 00/16] net: macb: various fixes & cleanup
https://lore.kernel.org/lkml/20250808-macb-fixes-v3-0-08f1fcb5179f@bootlin.com/
---
Théo Lebrun (5):
dt-bindings: net: cdns,macb: add Mobileye EyeQ5 ethernet interface
net: macb: match skb_reserve(skb, NET_IP_ALIGN) with HW alignment
net: macb: add no LSO capability (MACB_CAPS_NO_LSO)
net: macb: rename bp->sgmii_phy field to bp->phy
net: macb: Add "mobileye,eyeq5-gem" compatible
.../devicetree/bindings/net/cdns,macb.yaml | 10 +++
drivers/net/ethernet/cadence/macb.h | 6 +-
drivers/net/ethernet/cadence/macb_main.c | 94 +++++++++++++++++-----
3 files changed, 91 insertions(+), 19 deletions(-)
---
base-commit: 61b7ade9ba
change-id: 20251020-macb-eyeq5-fe2c0d1edc75
Best regards,
====================
Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-0-af509422c204@bootlin.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Add support for the two GEM instances inside Mobileye EyeQ5 SoCs, using
compatible "mobileye,eyeq5-gem". With it, add a custom init sequence
that must grab a generic PHY and initialise it.
We use bp->phy in both RGMII and SGMII cases. Tell our mode by adding a
phy_set_mode_ext() during macb_open(), before phy_power_on(). We are
the first users of bp->phy that use it in non-SGMII cases.
The phy_set_mode_ext() call is made unconditionally. It cannot cause
issues on platforms where !bp->phy or !bp->phy->ops->set_mode as, in
those cases, the call is a no-op (returning zero). From reading
upstream DTS, we can figure out that no platform has a bp->phy and a
PHY driver that has a .set_mode() implementation:
- cdns,zynqmp-gem: no DTS upstream.
- microchip,mpfs-macb: microchip/mpfs.dtsi, &mac0..1, no PHY attached.
- xlnx,versal-gem: xilinx/versal-net.dtsi, &gem0..1, no PHY attached.
- xlnx,zynqmp-gem: xilinx/zynqmp.dtsi, &gem0..3, PHY attached to
drivers/phy/xilinx/phy-zynqmp.c which has no .set_mode().
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-5-af509422c204@bootlin.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The bp->sgmii_phy field is initialised at probe by init_reset_optional()
if bp->phy_interface == PHY_INTERFACE_MODE_SGMII. It gets used by:
- zynqmp_config: "cdns,zynqmp-gem" or "xlnx,zynqmp-gem" compatibles.
- mpfs_config: "microchip,mpfs-macb" compatible.
- versal_config: "xlnx,versal-gem" compatible.
Make name more generic as EyeQ5 requires the PHY in SGMII & RGMII cases.
Drop "for ZynqMP SGMII mode" comment that is already a lie, as it gets
used on Microchip platforms as well. And soon it won't be SGMII-only.
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: Théo Lebrun <theo.lebrun@bootlin.com>
Link: https://patch.msgid.link/20251023-macb-eyeq5-v3-4-af509422c204@bootlin.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
