* [PATCH net-next 3/5] net: dsa: mv88e6xxx: define .pot_clear() for 6321
From: Marek Behún @ 2026-05-04 15:32 UTC (permalink / raw)
To: Andrew Lunn, Vladimir Oltean, Russell King (Oracle),
Vivien Didelot, Tobias Waldekranz, netdev, Fidan Aliyeva
Cc: Lev Olshvang, Marek Behún
In-Reply-To: <20260504153227.1390546-1-kabel@kernel.org>
Commit 9e907d739cc3 ("net: dsa: mv88e6xxx: add POT operation") did not
add the .pot_clear() method to the 6321 switch operations structure.
Add them now.
Signed-off-by: Marek Behún <kabel@kernel.org>
---
drivers/net/dsa/mv88e6xxx/chip.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 78799666e6b7..b55a9d4a1ee1 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -5244,6 +5244,7 @@ static const struct mv88e6xxx_ops mv88e6321_ops = {
.set_egress_port = mv88e6095_g1_set_egress_port,
.watchdog_ops = &mv88e6390_watchdog_ops,
.mgmt_rsvd2cpu = mv88e6352_g2_mgmt_rsvd2cpu,
+ .pot_clear = mv88e6xxx_g2_pot_clear,
.hardware_reset_pre = mv88e6xxx_g2_eeprom_wait,
.hardware_reset_post = mv88e6xxx_g2_eeprom_wait,
.reset = mv88e6352_g1_reset,
--
2.53.0
^ permalink raw reply related
* [PATCH net-next 2/5] net: dsa: mv88e6xxx: allow SPEED_200 for 6320 family on supported ports
From: Marek Behún @ 2026-05-04 15:32 UTC (permalink / raw)
To: Andrew Lunn, Vladimir Oltean, Russell King (Oracle),
Vivien Didelot, Tobias Waldekranz, netdev, Fidan Aliyeva
Cc: Lev Olshvang, Marek Behún
In-Reply-To: <20260504153227.1390546-1-kabel@kernel.org>
The 6320 family supports the ALT_SPEED bit on ports 2, 5 and 6. Allow
this speed by implementing 6320 family specific .port_set_speed_duplex()
method.
Signed-off-by: Marek Behún <kabel@kernel.org>
---
drivers/net/dsa/mv88e6xxx/chip.c | 4 ++--
drivers/net/dsa/mv88e6xxx/port.c | 15 +++++++++++++++
drivers/net/dsa/mv88e6xxx/port.h | 2 ++
3 files changed, 19 insertions(+), 2 deletions(-)
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 338cba0c8364..78799666e6b7 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -5168,7 +5168,7 @@ static const struct mv88e6xxx_ops mv88e6320_ops = {
.port_set_link = mv88e6xxx_port_set_link,
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6320_port_set_rgmii_delay,
- .port_set_speed_duplex = mv88e6185_port_set_speed_duplex,
+ .port_set_speed_duplex = mv88e6320_port_set_speed_duplex,
.port_tag_remap = mv88e6095_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -5221,7 +5221,7 @@ static const struct mv88e6xxx_ops mv88e6321_ops = {
.port_set_link = mv88e6xxx_port_set_link,
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6320_port_set_rgmii_delay,
- .port_set_speed_duplex = mv88e6185_port_set_speed_duplex,
+ .port_set_speed_duplex = mv88e6320_port_set_speed_duplex,
.port_tag_remap = mv88e6095_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
diff --git a/drivers/net/dsa/mv88e6xxx/port.c b/drivers/net/dsa/mv88e6xxx/port.c
index 74cfc0bb7b5b..ea1fab71968a 100644
--- a/drivers/net/dsa/mv88e6xxx/port.c
+++ b/drivers/net/dsa/mv88e6xxx/port.c
@@ -327,6 +327,21 @@ int mv88e6250_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
duplex);
}
+int mv88e6320_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
+ int speed, int duplex)
+{
+ bool has_200 = (port == 2 || port == 5 || port == 6);
+
+ if (speed > 1000)
+ return -EOPNOTSUPP;
+
+ if (speed == 200 && !has_200)
+ return -EOPNOTSUPP;
+
+ return mv88e6xxx_port_set_speed_duplex(chip, port, speed, has_200,
+ false, duplex);
+}
+
/* Support 10, 100, 200, 1000, 2500 Mbps (e.g. 88E6341) */
int mv88e6341_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex)
diff --git a/drivers/net/dsa/mv88e6xxx/port.h b/drivers/net/dsa/mv88e6xxx/port.h
index 44b100dd618b..5b6cde9f7406 100644
--- a/drivers/net/dsa/mv88e6xxx/port.h
+++ b/drivers/net/dsa/mv88e6xxx/port.h
@@ -493,6 +493,8 @@ int mv88e6185_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex);
int mv88e6250_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex);
+int mv88e6320_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
+ int speed, int duplex);
int mv88e6341_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex);
int mv88e6352_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
--
2.53.0
^ permalink raw reply related
* [PATCH net-next 1/5] net: dsa: mv88e6xxx: fix number of g1 interrupts for 6320 family
From: Marek Behún @ 2026-05-04 15:32 UTC (permalink / raw)
To: Andrew Lunn, Vladimir Oltean, Russell King (Oracle),
Vivien Didelot, Tobias Waldekranz, netdev, Fidan Aliyeva
Cc: Lev Olshvang, Marek Behún
In-Reply-To: <20260504153227.1390546-1-kabel@kernel.org>
The 6320 family has 9 global1 interrupt, not 8. Fix it.
Signed-off-by: Marek Behún <kabel@kernel.org>
---
drivers/net/dsa/mv88e6xxx/chip.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index c618abc75ebb..338cba0c8364 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -6280,7 +6280,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = {
.global1_addr = 0x1b,
.global2_addr = 0x1c,
.age_time_coeff = 15000,
- .g1_irqs = 8,
+ .g1_irqs = 9,
.g2_irqs = 10,
.stats_type = STATS_TYPE_BANK0 | STATS_TYPE_BANK1,
.atu_move_port_mask = 0xf,
@@ -6308,7 +6308,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = {
.global1_addr = 0x1b,
.global2_addr = 0x1c,
.age_time_coeff = 15000,
- .g1_irqs = 8,
+ .g1_irqs = 9,
.g2_irqs = 10,
.stats_type = STATS_TYPE_BANK0 | STATS_TYPE_BANK1,
.atu_move_port_mask = 0xf,
--
2.53.0
^ permalink raw reply related
* [PATCH net-next 0/5] Fixes for mv88e6xxx for 6320/6321 family
From: Marek Behún @ 2026-05-04 15:32 UTC (permalink / raw)
To: Andrew Lunn, Vladimir Oltean, Russell King (Oracle),
Vivien Didelot, Tobias Waldekranz, netdev, Fidan Aliyeva
Cc: Lev Olshvang, Marek Behún
Five fixes for mv88e6xxx for 6320/6321 family, for net-next, without
Fixes tags, as per Andrew's request last year, see
https://lore.kernel.org/netdev/20250313134146.27087-1-kabel@kernel.org/
Marek Behún (5):
net: dsa: mv88e6xxx: fix number of g1 interrupts for 6320 family
net: dsa: mv88e6xxx: allow SPEED_200 for 6320 family on supported
ports
net: dsa: mv88e6xxx: define .pot_clear() for 6321
net: dsa: mv88e6xxx: enable .rmu_disable() for 6320 family
net: dsa: mv88e6xxx: enable devlink ATU hash param for 6320 family
drivers/net/dsa/mv88e6xxx/chip.c | 15 +++++++++++----
drivers/net/dsa/mv88e6xxx/port.c | 15 +++++++++++++++
drivers/net/dsa/mv88e6xxx/port.h | 2 ++
3 files changed, 28 insertions(+), 4 deletions(-)
--
2.53.0
^ permalink raw reply
* [PATCH net v2] xfrm: esp: avoid in-place decrypt on shared skb frags
From: HexRabbit @ 2026-05-04 15:27 UTC (permalink / raw)
To: netdev
Cc: Steffen Klassert, Greg Kroah-Hartman, Herbert Xu, Simon Horman,
David S . Miller, David Ahern, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, Ido Schimmel, Hyunwoo Kim, linux-kernel,
Kuan-Ting Chen, stable
From: Kuan-Ting Chen <h3xrabbit@gmail.com>
MSG_SPLICE_PAGES can attach pages from a pipe directly to an skb. TCP
marks such skbs with SKBFL_SHARED_FRAG after skb_splice_from_iter(),
so later paths that may modify packet data can first make a private
copy. The IPv4/IPv6 datagram append paths did not set this flag when
splicing pages into UDP skbs.
That leaves an ESP-in-UDP packet made from shared pipe pages looking
like an ordinary uncloned nonlinear skb. ESP input then takes the no-COW
fast path for uncloned skbs without a frag_list and decrypts in place
over data that is not owned privately by the skb.
Mark IPv4/IPv6 datagram splice frags with SKBFL_SHARED_FRAG, matching
TCP. Also make ESP input fall back to skb_cow_data() when the flag is
present, so ESP does not decrypt externally backed frags in place.
Private nonlinear skb frags still use the existing fast path.
This intentionally does not change ESP output. In esp_output_head(),
the path that appends the ESP trailer to existing skb tailroom without
calling skb_cow_data() is not reachable for nonlinear skbs:
skb_tailroom() returns zero when skb->data_len is nonzero, while ESP
tailen is positive. Thus ESP output will either use the separate
destination-frag path or fall back to skb_cow_data().
Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible")
Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible")
Fixes: 7da0dde68486 ("ip, udp: Support MSG_SPLICE_PAGES")
Fixes: 6d8192bd69bb ("ip6, udp6: Support MSG_SPLICE_PAGES")
Reported-by: Hyunwoo Kim <imv4bel@gmail.com>
Reported-by: Kuan-Ting Chen <h3xrabbit@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Kuan-Ting Chen <h3xrabbit@gmail.com>
---
v2:
- Add Fixes tags
- Add stable Cc and Reported-by trailers.
net/ipv4/esp4.c | 3 ++-
net/ipv4/ip_output.c | 2 ++
net/ipv6/esp6.c | 3 ++-
net/ipv6/ip6_output.c | 2 ++
4 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 6dfc0bcde..6a5febbdb 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -873,7 +873,8 @@ static int esp_input(struct xfrm_state *x, struct sk_buff *skb)
nfrags = 1;
goto skip_cow;
- } else if (!skb_has_frag_list(skb)) {
+ } else if (!skb_has_frag_list(skb) &&
+ !skb_has_shared_frag(skb)) {
nfrags = skb_shinfo(skb)->nr_frags;
nfrags++;
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index e4790cc7b..5bcd73cbd 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1233,6 +1233,8 @@ static int __ip_append_data(struct sock *sk,
if (err < 0)
goto error;
copy = err;
+ if (!(flags & MSG_NO_SHARED_FRAGS))
+ skb_shinfo(skb)->flags |= SKBFL_SHARED_FRAG;
wmem_alloc_delta += copy;
} else if (!zc) {
int i = skb_shinfo(skb)->nr_frags;
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 9f7531373..9c06c5a14 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -915,7 +915,8 @@ static int esp6_input(struct xfrm_state *x, struct sk_buff *skb)
nfrags = 1;
goto skip_cow;
- } else if (!skb_has_frag_list(skb)) {
+ } else if (!skb_has_frag_list(skb) &&
+ !skb_has_shared_frag(skb)) {
nfrags = skb_shinfo(skb)->nr_frags;
nfrags++;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 7e92909ab..1f2a33fbe 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1794,6 +1794,8 @@ static int __ip6_append_data(struct sock *sk,
if (err < 0)
goto error;
copy = err;
+ if (!(flags & MSG_NO_SHARED_FRAGS))
+ skb_shinfo(skb)->flags |= SKBFL_SHARED_FRAG;
wmem_alloc_delta += copy;
} else if (!zc) {
int i = skb_shinfo(skb)->nr_frags;
--
2.43.0
^ permalink raw reply related
* [PATCH] net: dsa: mv88e6xxx: remove unused .port_max_speed_mode()
From: Marek Behún @ 2026-05-04 15:26 UTC (permalink / raw)
To: Andrew Lunn, Vladimir Oltean, Russell King (Oracle),
Vivien Didelot, Tobias Waldekranz, netdev
Cc: Marek Behún
The .port_max_speed_mode() method is not used anymore since commit
40da0c32c3fc ("net: dsa: mv88e6xxx: remove handling for DSA and CPU ports").
Drop it.
Signed-off-by: Marek Behún <kabel@kernel.org>
---
drivers/net/dsa/mv88e6xxx/chip.c | 9 -------
drivers/net/dsa/mv88e6xxx/chip.h | 4 ----
drivers/net/dsa/mv88e6xxx/port.c | 40 --------------------------------
drivers/net/dsa/mv88e6xxx/port.h | 9 -------
4 files changed, 62 deletions(-)
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 8ca5fd40df92..c618abc75ebb 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -4434,7 +4434,6 @@ static const struct mv88e6xxx_ops mv88e6141_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6341_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6341_port_max_speed_mode,
.port_tag_remap = mv88e6095_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -4826,7 +4825,6 @@ static const struct mv88e6xxx_ops mv88e6190_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6390_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6390_port_max_speed_mode,
.port_tag_remap = mv88e6390_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -4886,7 +4884,6 @@ static const struct mv88e6xxx_ops mv88e6190x_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6390x_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6390x_port_max_speed_mode,
.port_tag_remap = mv88e6390_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -4946,7 +4943,6 @@ static const struct mv88e6xxx_ops mv88e6191_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6390_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6390_port_max_speed_mode,
.port_tag_remap = mv88e6390_port_tag_remap,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
.port_set_ucast_flood = mv88e6352_port_set_ucast_flood,
@@ -5110,7 +5106,6 @@ static const struct mv88e6xxx_ops mv88e6290_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6390_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6390_port_max_speed_mode,
.port_tag_remap = mv88e6390_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -5278,7 +5273,6 @@ static const struct mv88e6xxx_ops mv88e6341_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6341_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6341_port_max_speed_mode,
.port_tag_remap = mv88e6095_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -5499,7 +5493,6 @@ static const struct mv88e6xxx_ops mv88e6390_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6390_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6390_port_max_speed_mode,
.port_tag_remap = mv88e6390_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -5564,7 +5557,6 @@ static const struct mv88e6xxx_ops mv88e6390x_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6390x_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6390x_port_max_speed_mode,
.port_tag_remap = mv88e6390_port_tag_remap,
.port_set_policy = mv88e6352_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
@@ -5627,7 +5619,6 @@ static const struct mv88e6xxx_ops mv88e6393x_ops = {
.port_sync_link = mv88e6xxx_port_sync_link,
.port_set_rgmii_delay = mv88e6390_port_set_rgmii_delay,
.port_set_speed_duplex = mv88e6393x_port_set_speed_duplex,
- .port_max_speed_mode = mv88e6393x_port_max_speed_mode,
.port_tag_remap = mv88e6390_port_tag_remap,
.port_set_policy = mv88e6393x_port_set_policy,
.port_set_frame_mode = mv88e6351_port_set_frame_mode,
diff --git a/drivers/net/dsa/mv88e6xxx/chip.h b/drivers/net/dsa/mv88e6xxx/chip.h
index 2b235ac2c5df..4339f6810535 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.h
+++ b/drivers/net/dsa/mv88e6xxx/chip.h
@@ -585,10 +585,6 @@ struct mv88e6xxx_ops {
int (*port_set_speed_duplex)(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex);
- /* What interface mode should be used for maximum speed? */
- phy_interface_t (*port_max_speed_mode)(struct mv88e6xxx_chip *chip,
- int port);
-
int (*port_tag_remap)(struct mv88e6xxx_chip *chip, int port);
int (*port_set_policy)(struct mv88e6xxx_chip *chip, int port,
diff --git a/drivers/net/dsa/mv88e6xxx/port.c b/drivers/net/dsa/mv88e6xxx/port.c
index 49cd82930b7a..74cfc0bb7b5b 100644
--- a/drivers/net/dsa/mv88e6xxx/port.c
+++ b/drivers/net/dsa/mv88e6xxx/port.c
@@ -344,15 +344,6 @@ int mv88e6341_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
duplex);
}
-phy_interface_t mv88e6341_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port)
-{
- if (port == 5)
- return PHY_INTERFACE_MODE_2500BASEX;
-
- return PHY_INTERFACE_MODE_NA;
-}
-
/* Support 10, 100, 200, 1000 Mbps (e.g. 88E6352 family) */
int mv88e6352_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex)
@@ -384,15 +375,6 @@ int mv88e6390_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
duplex);
}
-phy_interface_t mv88e6390_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port)
-{
- if (port == 9 || port == 10)
- return PHY_INTERFACE_MODE_2500BASEX;
-
- return PHY_INTERFACE_MODE_NA;
-}
-
/* Support 10, 100, 200, 1000, 2500, 10000 Mbps (e.g. 88E6190X) */
int mv88e6390x_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex)
@@ -407,15 +389,6 @@ int mv88e6390x_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
duplex);
}
-phy_interface_t mv88e6390x_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port)
-{
- if (port == 9 || port == 10)
- return PHY_INTERFACE_MODE_XAUI;
-
- return PHY_INTERFACE_MODE_NA;
-}
-
/* Support 10, 100, 200, 1000, 2500, 5000, 10000 Mbps (e.g. 88E6393X)
* Function mv88e6xxx_port_set_speed_duplex() can't be used as the register
* values for speeds 2500 & 5000 conflict.
@@ -509,19 +482,6 @@ int mv88e6393x_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
return 0;
}
-phy_interface_t mv88e6393x_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port)
-{
-
- if (port != 0 && port != 9 && port != 10)
- return PHY_INTERFACE_MODE_NA;
-
- if (chip->info->prod_num == MV88E6XXX_PORT_SWITCH_ID_PROD_6361)
- return PHY_INTERFACE_MODE_2500BASEX;
-
- return PHY_INTERFACE_MODE_10GBASER;
-}
-
static int mv88e6xxx_port_set_cmode(struct mv88e6xxx_chip *chip, int port,
phy_interface_t mode, bool force)
{
diff --git a/drivers/net/dsa/mv88e6xxx/port.h b/drivers/net/dsa/mv88e6xxx/port.h
index a2492cf4d920..44b100dd618b 100644
--- a/drivers/net/dsa/mv88e6xxx/port.h
+++ b/drivers/net/dsa/mv88e6xxx/port.h
@@ -504,15 +504,6 @@ int mv88e6390x_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int mv88e6393x_port_set_speed_duplex(struct mv88e6xxx_chip *chip, int port,
int speed, int duplex);
-phy_interface_t mv88e6341_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port);
-phy_interface_t mv88e6390_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port);
-phy_interface_t mv88e6390x_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port);
-phy_interface_t mv88e6393x_port_max_speed_mode(struct mv88e6xxx_chip *chip,
- int port);
-
int mv88e6xxx_port_set_state(struct mv88e6xxx_chip *chip, int port, u8 state);
int mv88e6xxx_port_set_vlan_map(struct mv88e6xxx_chip *chip, int port, u16 map);
--
2.53.0
^ permalink raw reply related
* Re: [PATCH net-next v1 0/5] net: mdiobus: HIde ACPI implementation
From: Andrew Lunn @ 2026-05-04 15:16 UTC (permalink / raw)
To: Andy Shevchenko
Cc: linux-kernel, netdev, Andrew Lunn, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Heiner Kallweit, Russell King
In-Reply-To: <afij7wOkQUGUN0kM@ashevche-desk.local>
On Mon, May 04, 2026 at 04:49:35PM +0300, Andy Shevchenko wrote:
> On Mon, May 04, 2026 at 03:02:54PM +0200, Andrew Lunn wrote:
> > On Mon, May 04, 2026 at 09:29:51AM +0200, Andy Shevchenko wrote:
> > > This mini-series is dedicated to hiding ACPI implementation details
> > > from the wider users as they (as of today) do not need to know that.
> >
> > Please could you expand on that. ACPI != OF. They have different
> > bindings, so you need different implementations.
>
> As of today the users that want ACPI also have the OF support. Even without
> that if the device is pure ACPI supported one (and somehow never going to DT)
> the proposed API (see the first patch) will be no-op in case when CONFIG_ACPI=n
> or when it's a non-ACPI platform with no support of the device. Hence, the
> pure ACPI (and actually OF as well) do not need to be exposed. The decision is
> made based on the type of firmware node.
I see two different things here:
You are refactoring code into a helper, so reducing the amount of
duplicated code. That in itself is good.
However, we have the problem in general that developers think that OF
and ACPI do the same thing. And that is not true. OF MDIO busses have
support for a GPIO used as a reset. In retrospect, that was a bad
idea. The documented ACPI binding does not have this, and if anybody
was to suggest adding it, i want to NACK it.
Having OF code and ACPI code to instantiate the MDIO bus makes it
clearer they are different things. fwnode gives the impression they
are the same, which is not true. Hence i don't like fwnode at this
level.
Where fwnode makes sense is deeper down in the implementation of the
bindings, for properties which are the same in both bindings. Then
you can use fwnode_ for the shared properties, of_ for the OF only
properties, and acpi_ for the ACPI only properties.
So although i like the reduction in duplicated code, i think overall
it makes the code worse because developers are more likely to wrongly
understand it.
Andrew
^ permalink raw reply
* Re: [PATCH net-next v2 0/2] mv88e6xxx: SERDES on mv88e6321 letter
From: Marek Behún @ 2026-05-04 15:05 UTC (permalink / raw)
To: Fidan Aliyeva
Cc: andrew, olteanv, davem, edumazet, kuba, pabeni, netdev,
linux-kernel, thomas.eckerman.ext
In-Reply-To: <20260430124907.3533344-1-fidan.aliyeva.ext@ericsson.com>
Hello Fidan, Andrew,
last year I worked on 6320+6321 serdes support, but never send it
due to different workload.
Quickly looking at your patches, there are some things I think are
missing in comparison to what I did - I have 5 patches:
1/5: drops serdes methods for 88E6172, which does not have serdes
2/5: is similar to your patch 1/2, adding serdes_get_lane to 6352 family.
The difference is the implementation of mv88e6352_serdes_get_lane(),
which does not read scratch register. Yours is probably better in
this regard, but mv88e6352_g2_scratch_port_has_serdes() is also
called in mv88e6352_pcs_init(). So maybe this is redundant?
Also, I rename MV88E6352_ADDR_SERDES macro to MV88E6352_SERDES_LANE
to be in line with other such macros in serdes.h
3/5: extends 6352 serdes pcs for 6320 family.
In addition to your code I also implement .serdes_irq_mapping().
These depend on other mv88e6xxx fixes I forgot to send, which I
will do now.
4/5: Add hidden register access methods for 6320 and 6352 family.
These can be used to change SerDes modes.
5/5: Add support for changing between sgmii and 1000base-x on 6320 family
via hidden register acces.
This was done by reverse engineering hidden registers (trying
different values), since it is not documented correctly.
Would you prefer for your patches to be applied first and then I can
rebase, or would you rather like to look at my code first?
Marek
On Thu, Apr 30, 2026 at 02:49:05PM +0200, Fidan Aliyeva wrote:
> This patch series add code support to be able to use SERDES feature of
> mv88e6321 version of Marvel mv88e6xxx series. mv88e6321 has 2 ports to
> support high speed SERDES but the support is lacking in the driver.
>
> mv88e6321 version has a similar architecture to mv88e6352 version making it
> possible to reuse its pcs functions. That's why the patch series consist of
> 2 parts:
> 1. Refactor the serdes functions and pcs_init of mv88e6352 to be more
> generic
> 2. Add the SERDES support for mv88e6321 reusing 6352's pcs functions
>
> The final code has been built on top of net-next tree and tested on
> mv88e6321 ethernet device directly by ip ping tests, performance tests and
> also verifying the switch's expected register values.
>
> Referred document: 88E6321/88E6320 Functional Specification
>
> Code has been built with allmodconfig and allyesconfig. checkpatch.pl was
> also run
>
> ---
> Changes in v2:
> - Removed 6321-specific pcs_init and made 6352's pcs_init more generic
> as suggested by Andrew Lunn
> - Added the correct mailing list
>
> ---
> Fidan Aliyeva (2):
> mv88e6xxx: Refactor 6352's serdes functions
> mv88e6xxx: Add SERDES Support for mv88e6321
>
> drivers/net/dsa/mv88e6xxx/chip.c | 8 +++
> drivers/net/dsa/mv88e6xxx/pcs-6352.c | 12 ++--
> drivers/net/dsa/mv88e6xxx/serdes.c | 87 ++++++++++++++++++++++------
> drivers/net/dsa/mv88e6xxx/serdes.h | 5 ++
> 4 files changed, 86 insertions(+), 26 deletions(-)
>
> --
> 2.36.0
>
>
^ permalink raw reply
* [PATCH net-next] atm: solos-pci: Simplify initialisation of pci_device_id array
From: Uwe Kleine-König (The Capable Hub) @ 2026-05-04 15:12 UTC (permalink / raw)
To: Chas Williams
Cc: linux-atm-general, netdev, linux-kernel,
Markus Schneider-Pargmann
Use the convenience macro PCI_DEVICE to initialize .vendor, .device,
.subvendor and .subdevice. Drop explicit zeros that the compiler also
fills in.
Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@baylibre.com>
---
Hello,
The secret plan is to make struct pci_device_id::driver_data an
anonymous union (similar to
https://lore.kernel.org/all/cover.1776579304.git.u.kleine-koenig@baylibre.com/)
and that requires named initializers. But IMHO it's also a nice cleanup
on its own.
The anonymous union will allow changes like the following:
- { PCI_DEVICE(0x8086, 0x8108), .driver_data = (long) &psb_chip_ops },
+ { PCI_DEVICE(0x8086, 0x8108), .driver_data_ptr = &psb_chip_ops },
(together with the respective change in the code when the value is
used). This gets rid of a bunch of casts and thus slightly improves
type safety. But that doesn't affect drivers/atm as here no pointers
are used as driver data. Still this driver needs adaption to work with
the changed definition of pci_device_id.
Best regards
Uwe
drivers/atm/solos-pci.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/atm/solos-pci.c b/drivers/atm/solos-pci.c
index 24c764664c24..bcb1353877e4 100644
--- a/drivers/atm/solos-pci.c
+++ b/drivers/atm/solos-pci.c
@@ -1464,8 +1464,8 @@ static void fpga_remove(struct pci_dev *dev)
}
static const struct pci_device_id fpga_pci_tbl[] = {
- { 0x10ee, 0x0300, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
- { 0, }
+ { PCI_DEVICE(0x10ee, 0x0300) },
+ { }
};
MODULE_DEVICE_TABLE(pci,fpga_pci_tbl);
base-commit: 254f49634ee16a731174d2ae34bc50bd5f45e731
--
2.47.3
^ permalink raw reply related
* Re: [PATCH 3/3 net-next v3] selftests: net: add test for IPv4 devconf netlink notifications
From: Fernando Fernandez Mancera @ 2026-05-04 15:08 UTC (permalink / raw)
To: netdev
Cc: linux-kselftest, horms, pabeni, kuba, edumazet, davem, idosch,
dsahern
In-Reply-To: <20260504123143.6284-3-fmancera@suse.de>
On 5/4/26 2:31 PM, Fernando Fernandez Mancera wrote:
> Introduce a new test, `ipv4_devconf_notify`, to verify that the kernel
> sends the appropriate netlink notifications when IPv4 devconf parameters
> are modified.
>
> Since YNL currently has a bug where it declares an array of u32 values
> instead of the nested attributes expected by the kernel for devconf set
> operations, a temporary hack (`patched_add_attr`) is included to
> pack the netlink attributes correctly.
>
> Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
> ---
> v3: added this patch to the series as requested by Paolo.
> ---
I just learnt about ruff. There are two new warnings. I will repost
fixing them.
Sorry for the noise.
---
pw-bot: cr
^ permalink raw reply
* [PATCH] net: bridge: replace simple_strtoul with kstrtoul
From: Aadarsh Chandra @ 2026-05-04 15:01 UTC (permalink / raw)
To: razor, idosch; +Cc: davem, kuba, netdev, bridge, linux-kernel, Aadarsh Chandra
The simple_strtoul() function is deprecated. It does not handle
errors or overflows correctly. Replace it with kstrtoul() in
brport_store() to ensure that invalid user input is caught and
returned as an error.
Signed-off-by: Aadarsh Chandra <aadarsh.official.xz@gmail.com>
---
net/bridge/br_sysfs_if.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c
index 1f57c36a7fc0..3e948d781970 100644
--- a/net/bridge/br_sysfs_if.c
+++ b/net/bridge/br_sysfs_if.c
@@ -318,7 +318,7 @@ static ssize_t brport_store(struct kobject *kobj,
struct net_bridge_port *p = kobj_to_brport(kobj);
ssize_t ret = -EINVAL;
unsigned long val;
- char *endp;
+ int err;
if (!ns_capable(dev_net(p->dev)->user_ns, CAP_NET_ADMIN))
return -EPERM;
@@ -339,9 +339,11 @@ static ssize_t brport_store(struct kobject *kobj,
spin_unlock_bh(&p->br->lock);
kfree(buf_copy);
} else if (brport_attr->store) {
- val = simple_strtoul(buf, &endp, 0);
- if (endp == buf)
+ err = kstrtoul(buf, 0, &val);
+ if (err) {
+ ret = err;
goto out_unlock;
+ }
spin_lock_bh(&p->br->lock);
ret = brport_attr->store(p, val);
spin_unlock_bh(&p->br->lock);
--
2.54.0
^ permalink raw reply related
* Re: [PATCH v2] ipv6: default IPV6_SIT to m
From: Fernando Fernandez Mancera @ 2026-05-04 15:01 UTC (permalink / raw)
To: Alyssa Ross, David S. Miller, David Ahern, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Simon Horman
Cc: netdev, linux-kernel, Ricardo B . Marlière,
Krzysztof Kozlowski
In-Reply-To: <878q9zdz6s.fsf@alyssa.is>
On 5/4/26 4:30 PM, Alyssa Ross wrote:
> Fernando Fernandez Mancera <fmancera@suse.de> writes:
>
>> On 5/3/26 9:25 PM, Alyssa Ross wrote:
>>> This basically defaulted to m until recently, since IPV6 defaulted to
>>> m. Since IPV6 was changed to a boolean with a default of y, IPV6_SIT
>>> started defaulting to built-in as well. This results in a surprise
>>> sit0 device by default for defconfig (and defconfig-derived config)
>>> users at boot. For me, this broke an (admittedly non-robust) script.
>>> Preserve the behaviour of most configs by avoiding building this
>>> module, that's probably overall seldom used compared to IPv6 as a
>>> whole, into the kernel.
>>>
>>> Fixes: 309b905deee59 ("ipv6: convert CONFIG_IPV6 to built-in only and clean up Kconfigs")
>>> Signed-off-by: Alyssa Ross <hi@alyssa.is>
>>
>> Hi, I am not sure about this. IPV6_SIT was already defaulting to 'y' for
>> most distributions (except for ARM). Sure, if you had CONFIG_IPV6=m then
>> CONFIG_IPV6_SIT was defaulting to 'm'.
>>
>> Hm. Maybe we can aim this for net-next tree instead of net? I just don't
>> think this is a fix but a change in behavior. FWIW, it makes sense to me
>> to have this as a module by default.
>>
>> Let's see what do others think.
>
> Yes, it appears you're right. I had only tested on ARM. In that case,
> perhaps it make sense to instead set this to m in the arm64 defconfig, to
> preserve the previous situation, but I can also see the value in
> reducing platform discongruity by letting it change to y on arm64 too.
Changing it on arm64 defconfig would make sense to me too. We should
probably visit other configs that might be in the same situation. That
in my opinion would qualify a net tree fix as you proposed initially.
^ permalink raw reply
* Re: [PATCH net v5 8/8] xsk: fix u64 descriptor address truncation on 32-bit architectures
From: Stanislav Fomichev @ 2026-05-04 14:59 UTC (permalink / raw)
To: Jason Xing
Cc: davem, edumazet, kuba, pabeni, bjorn, magnus.karlsson,
maciej.fijalkowski, jonathan.lemon, sdf, ast, daniel, hawk,
john.fastabend, horms, andrew+netdev, bpf, netdev, Jason Xing
In-Reply-To: <20260502200722.53960-9-kerneljasonxing@gmail.com>
On 05/02, Jason Xing wrote:
> From: Jason Xing <kernelxing@tencent.com>
>
> In copy mode TX, xsk_skb_destructor_set_addr() stores the 64-bit
> descriptor address into skb_shinfo(skb)->destructor_arg (void *) via a
> uintptr_t cast:
>
> skb_shinfo(skb)->destructor_arg = (void *)((uintptr_t)addr | 0x1UL);
>
> On 32-bit architectures uintptr_t is 32 bits, so the upper 32 bits of
> the descriptor address are silently dropped. In XDP_ZEROCOPY unaligned
> mode the chunk offset is encoded in bits 48-63 of the descriptor
> address (XSK_UNALIGNED_BUF_OFFSET_SHIFT = 48), meaning the offset is
> lost entirely. The completion queue then returns a truncated address to
> userspace, making buffer recycling impossible.
>
> Fix this by handling the 32-bit case directly in
> xsk_skb_destructor_set_addr(): when !CONFIG_64BIT, allocate an
> xsk_addrs struct (the same path already used for multi-descriptor
> SKBs) to store the full u64 address. The existing tagged-pointer logic
> in xsk_skb_destructor_is_addr() stays unchanged: slab pointers returned
> from kmem_cache_zalloc() are always word-aligned and therefore have
> bit 0 clear, which correctly identifies them as a struct pointer
> rather than an inline tagged address on every architecture.
>
> Factor the shared kmem_cache_zalloc + destructor_arg assignment into
> __xsk_addrs_alloc() and add a wrapper xsk_addrs_alloc() that handles
> the inline-to-list upgrade (is_addr check + get_addr + num_descs = 1).
> The three former open-coded kmem_cache_zalloc call sites now reduce to
> a single call each.
>
> Propagate the -ENOMEM from xsk_skb_destructor_set_addr() through
> xsk_skb_init_misc() so the caller can clean up the skb via kfree_skb()
> before skb->destructor is installed.
>
> The overhead is one extra kmem_cache_zalloc per first descriptor on
> 32-bit only; 64-bit builds are completely unchanged.
>
> Closes: https://lore.kernel.org/all/20260419045824.D9E5EC2BCAF@smtp.kernel.org/
> Fixes: 0ebc27a4c67d ("xsk: avoid data corruption on cq descriptor number")
> Signed-off-by: Jason Xing <kernelxing@tencent.com>
LGTM, thanks!
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
^ permalink raw reply
* Re: [PATCH net-next 3/3] net: selftests: add getsockopt_iter regression tests
From: Stanislav Fomichev @ 2026-05-04 14:57 UTC (permalink / raw)
To: Breno Leitao
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefano Garzarella, Shuah Khan, netdev,
linux-kernel, virtualization, linux-kselftest, kernel-team
In-Reply-To: <20260501-getsock_one-v1-3-810ce23ea70e@debian.org>
On 05/01, Breno Leitao wrote:
> Add a single kselftest covering the proto_ops getsockopt_iter
> conversions for AF_NETLINK and AF_VSOCK, using one fixture per protocol:
>
> netlink:
>
> NETLINK_PKTINFO covers the flag-style int path (exact size, oversize
> clamp, undersize -EINVAL); NETLINK_LIST_MEMBERSHIPS covers the
> size-discovery path that always reports the required buffer length back
> via optlen, even when the user buffer is too small to receive any group
> bits.
>
> vsock:
> SO_VM_SOCKETS_BUFFER_SIZE covers the u64 path (exact size, oversize
> clamp, undersize -EINVAL).
>
> Each fixture also exercises an unknown optname and a bogus level so
> the returned-length / errno semantics preserved by the sockopt_t
> conversion are pinned down.
>
> Signed-off-by: Breno Leitao <leitao@debian.org>
A few nits if you happen to respin (or for similar tests in the future):
- add ASSERT_EQ(optlen, xxx), won't hurt
- christmas tree (although not sure how much we care, you'll have to move
optlen = sizeof() assignment)
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
^ permalink raw reply
* [PATCH net-next] e1000e: ethtool: add get_channels support
From: Jon Kohler @ 2026-05-04 15:48 UTC (permalink / raw)
To: Tony Nguyen, Przemek Kitszel, Andrew Lunn, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, intel-wired-lan,
netdev, linux-kernel
Cc: Jon Kohler
e1000e hardware supports a single RX/TX queue pair, add basic support
for ethtool -l (i.e. get_channels), so that callers indeed see a single
queue.
Signed-off-by: Jon Kohler <jon@nutanix.com>
---
drivers/net/ethernet/intel/e1000e/ethtool.c | 26 +++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index dbed30943ef4..3a2c1a77ee74 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -2353,6 +2353,31 @@ static int e1000e_set_priv_flags(struct net_device *netdev, u32 priv_flags)
return 0;
}
+/**
+ * e1000e_get_channels - Report number of network channels
+ * @netdev: network interface device structure
+ * @ch: ethtool channels structure
+ *
+ * e1000e hardware supports a single RX/TX queue pair. When MSI-X is
+ * in use, an additional vector is dedicated to link/status ("other")
+ * events.
+ **/
+static void e1000e_get_channels(struct net_device *netdev,
+ struct ethtool_channels *ch)
+{
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+
+ /* single combined RX/TX channel */
+ ch->max_combined = 1;
+ ch->combined_count = 1;
+
+ /* report "other" vector when MSI-X is in use */
+ if (adapter->msix_entries) {
+ ch->max_other = 1;
+ ch->other_count = 1;
+ }
+}
+
static const struct ethtool_ops e1000_ethtool_ops = {
.supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS,
.get_drvinfo = e1000_get_drvinfo,
@@ -2386,6 +2411,7 @@ static const struct ethtool_ops e1000_ethtool_ops = {
.set_link_ksettings = e1000_set_link_ksettings,
.get_priv_flags = e1000e_get_priv_flags,
.set_priv_flags = e1000e_set_priv_flags,
+ .get_channels = e1000e_get_channels,
};
void e1000e_set_ethtool_ops(struct net_device *netdev)
--
2.43.0
^ permalink raw reply related
* Re: [PATCH] crypto: af_alg - Document the deprecation of AF_ALG
From: Jon Kohler @ 2026-05-04 14:39 UTC (permalink / raw)
To: Eric Biggers
Cc: linux-crypto@vger.kernel.org, Herbert Xu,
linux-doc@vger.kernel.org, linux-api@vger.kernel.org,
linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
Linus Torvalds
In-Reply-To: <20260430011544.31823-1-ebiggers@kernel.org>
> On Apr 29, 2026, at 9:15 PM, Eric Biggers <ebiggers@kernel.org> wrote:
>
> AF_ALG is almost completely unnecessary, and it exposes a massive attack
> surface that hasn't been standing up to modern vulnerability discovery
> tools. The latest one even has its own website, providing a small
> Python script that reliably roots most Linux distros: https://copy.fail/
>
> This isn't sustainable, especially as LLMs have accelerated the rate the
> vulnerabilities are coming in. The effort that is being put into this
> thing is vastly disproportional to the few programs that actually use
> it, and those programs would be better served by userspace code anyway.
>
> These issues have been noted in many mailing list discussions already.
> But until now they haven't been reflected in the documentation or
> kconfig menu itself, and the vulnerabilities are still coming in.
>
> Let's go ahead and document the deprecation.
>
> This isn't intended to change anything overnight. After all, most Linux
> distros won't be able to disable the kconfig options quite yet, mainly
> because of iwd. But this should create a bit more impetus for these
> userspace programs to be fixed, and the documentation update should also
> help prevent more users from appearing.
>
> Signed-off-by: Eric Biggers <ebiggers@kernel.org>
> ---
Quick passing observation
I noticed that when attempting to completely disable these Crypto APIs,
I was experiencing boot failures with fips=1 enabled systems.
Using 6.18-based kernel with an el9-based user space, I see the
following hang in the early boot console from dracut-pre-pivot:
Check integrity of kernel
libkcapi - Error: AF_ALG: socket syscall failed (errno: -97)
Allocation of hmac(sha512) cipher failed (-97)
I haven't looked at every elX version, but at least in el9 and el10,
they use libkcapi-hmaccalc to provide sha512hmac, which dracut [1]
uses to calculate the HMAC value in do_fips().
Digging further, I was only able to disable RNG and AEAD APIs, but
not HASH and SKCIPHER APIs when FIPS was in the picture with el9++.
I’m not sure how other distros do the same, but this could be problematic
elsehwere if other distros went down the libkcapi route.
[1] https://github.com/dracutdevs/dracut/blob/059/modules.d/01fips/fips.sh#L167
^ permalink raw reply
* Re: [PATCH net-next 2/3] vsock: convert to getsockopt_iter
From: Stanislav Fomichev @ 2026-05-04 14:54 UTC (permalink / raw)
To: Breno Leitao
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefano Garzarella, Shuah Khan, netdev,
linux-kernel, virtualization, linux-kselftest, kernel-team
In-Reply-To: <20260501-getsock_one-v1-2-810ce23ea70e@debian.org>
On 05/01, Breno Leitao wrote:
> Convert AF_VSOCK's getsockopt implementation to use the new
> getsockopt_iter callback with sockopt_t. The single
> vsock_connectible_getsockopt() callback is shared by both
> vsock_stream_ops and vsock_seqpacket_ops, so both proto_ops are
> updated to use .getsockopt_iter.
>
> Key changes:
> - Replace (char __user *optval, int __user *optlen) with sockopt_t *opt
> - Use opt->optlen for buffer length (input) and returned size (output)
> - Use copy_to_iter() instead of put_user()/copy_to_user()
>
> Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
^ permalink raw reply
* Re: [PATCH net-next 1/3] netlink: convert to getsockopt_iter
From: Stanislav Fomichev @ 2026-05-04 14:54 UTC (permalink / raw)
To: Breno Leitao
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefano Garzarella, Shuah Khan, netdev,
linux-kernel, virtualization, linux-kselftest, kernel-team
In-Reply-To: <20260501-getsock_one-v1-1-810ce23ea70e@debian.org>
On 05/01, Breno Leitao wrote:
> Convert AF_NETLINK's getsockopt implementation to use the new
> getsockopt_iter callback with sockopt_t.
>
> Key changes:
> - Replace (char __user *optval, int __user *optlen) with sockopt_t *opt
> - Use opt->optlen for buffer length (input) and returned size (output)
> - Use copy_to_iter() instead of put_user()/copy_to_user()
> - For NETLINK_LIST_MEMBERSHIPS: walk the groups bitmap and emit each
> u32 sequentially via copy_to_iter(), then set opt->optlen to the
> total size required (ALIGN(BITS_TO_BYTES(ngroups), sizeof(u32))).
> The wrapper writes opt->optlen back to userspace even on partial
> failure, preserving the existing API that lets userspace discover
> the needed allocation size.
>
> Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
^ permalink raw reply
* [PATCH net v2 2/2] selftests/net: add packetdrill test for locked SO_RCVBUF SWS
From: Ankit Jain @ 2026-05-04 14:49 UTC (permalink / raw)
To: edumazet, kuba, netdev
Cc: davem, pabeni, ncardwell, kuniyu, horms, shuah, quic_subashab,
quic_stranche, linux-kselftest, linux-kernel, karen.badiryan,
ajay.kaher, alexey.makhalov, vamsi-krishna.brahmajosyula,
yin.ding, tapas.kundu, Ankit Jain
In-Reply-To: <20260504144945.13477-1-ankit-aj.jain@broadcom.com>
Add a packetdrill test to verify that TCP does not aggressively penalize
the `scaling_ratio` when an application explicitly locks the receive
buffer via SO_RCVBUF.
This test establishes a connection with a very small MSS, then injects
small payloads. On buggy kernels, the payload-to-truesize memory penalty
drops the `scaling_ratio` to 1, which shrinks the dynamically calculated
window and causes Silly Window Syndrome (SWS).
The test asserts that `tcpi_rcv_ssthresh` (the exposed window clamp)
remains protected and stays at a healthy level.
Signed-off-by: Ankit Jain <ankit-aj.jain@broadcom.com>
---
.../net/packetdrill/tcp_locked_rcvbuf_sws.pkt | 34 +++++++++++++++++++
1 file changed, 34 insertions(+)
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_locked_rcvbuf_sws.pkt
diff --git a/tools/testing/selftests/net/packetdrill/tcp_locked_rcvbuf_sws.pkt b/tools/testing/selftests/net/packetdrill/tcp_locked_rcvbuf_sws.pkt
new file mode 100644
index 000000000000..96d3a0813548
--- /dev/null
+++ b/tools/testing/selftests/net/packetdrill/tcp_locked_rcvbuf_sws.pkt
@@ -0,0 +1,34 @@
+// SPDX-License-Identifier: GPL-2.0
+// Test that TCP does not apply aggressive truesize penalties
+// to the scaling_ratio when the application has explicitly locked
+// the receive buffer, preventing Silly Window Syndrome.
+
+// 1. Create a socket and lock SO_RCVBUF to 32K
+0 socket(..., SOCK_STREAM, IPPROTO_TCP) = 3
++0 setsockopt(3, SOL_SOCKET, SO_RCVBUF, [32768], 4) = 0
++0 bind(3, ..., ...) = 0
++0 listen(3, 1) = 0
+
+// 2. Establish connection with a TINY MSS (48 bytes)
+// This forces the kernel's rcv_mss expectation to initialize at rock bottom.
++0 < S 0:0(0) win 65535 <mss 48,nop,wscale 8>
++0 > S. 0:0(0) ack 1 <...>
++0 < . 1:1(0) ack 1 win 65535
++0 accept(3, ..., ...) = 4
+
+// 3. Inject 100 bytes.
+// 100 is > 48, so this forces tcp_measure_rcv_mss() to recalculate the scaling ratio.
+// Because the payload (100B) is small compared to the sk_buff truesize,
+// kernels without the fix will incorrectly drop the scaling_ratio.
++0.1 < P. 1:101(100) ack 1 win 65535
++0 > . 1:1(0) ack 101 <...>
+
+// 4. Inject 110 bytes to bypass the length jitter optimization and force
+// a second recalculation, driving the scaling_ratio to 1 and crushing the clamp.
++0.1 < P. 101:211(110) ack 1 win 65535
++0 > . 1:1(0) ack 211 <...>
+
+// 5. Assert the window clamp (tcpi_rcv_ssthresh) is protected from the penalty.
++0.1 %{
+assert tcpi_rcv_ssthresh > 15000, f"BUG DETECTED: scaling_ratio crushed! rcv_ssthresh={tcpi_rcv_ssthresh}"
+}%
--
2.53.0
^ permalink raw reply related
* [PATCH net v2 1/2] tcp: protect locked SO_RCVBUF from Silly Window Syndrome
From: Ankit Jain @ 2026-05-04 14:49 UTC (permalink / raw)
To: edumazet, kuba, netdev
Cc: davem, pabeni, ncardwell, kuniyu, horms, shuah, quic_subashab,
quic_stranche, linux-kselftest, linux-kernel, karen.badiryan,
ajay.kaher, alexey.makhalov, vamsi-krishna.brahmajosyula,
yin.ding, tapas.kundu, Ankit Jain
In-Reply-To: <20260504144945.13477-1-ankit-aj.jain@broadcom.com>
When an application locks SO_RCVBUF, it expects strict memory bounds and
disables TCP window auto-tuning. However, recent TCP memory fragmentation
optimizations still apply dynamic truesize penalties to the `scaling_ratio`
of these locked sockets.
For workloads processing small, fragmented packets (like Java's Tomcat),
this penalty drops the scaling_ratio to 1. This shrinks the dynamically
calculated advertised window, leading to Silly Window Syndrome (SWS)
deadlocks and 504 Gateway Timeouts.
This patch fixes the issue by bypassing the truesize penalty for sockets
with `SOCK_RCVBUF_LOCK` set. To ensure the kernel still defends against
memory exhaustion from large aggregate payloads (e.g., GRO), the penalty
is still applied if `skb->len` exceeds the advertised MSS.
Fixes: a2cbb1603943 ("tcp: Update window clamping condition")
Reported-by: Karen Badiryan <karen.badiryan@broadcom.com>
Signed-off-by: Ankit Jain <ankit-aj.jain@broadcom.com>
---
net/ipv4/tcp_input.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index d5c9e65d9760..569299dafa88 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -240,8 +240,14 @@ static void tcp_measure_rcv_mss(struct sock *sk, const struct sk_buff *skb)
/* Note: divides are still a bit expensive.
* For the moment, only adjust scaling_ratio
* when we update icsk_ack.rcv_mss.
+ *
+ * Protect locked SO_RCVBUF from Silly Window Syndrome
+ * due to truesize penalties on small packets. Allow
+ * penalty if aggregate payload (e.g., GRO) exceeds MSS.
*/
- if (unlikely(len != icsk->icsk_ack.rcv_mss)) {
+ if (unlikely(len != icsk->icsk_ack.rcv_mss &&
+ (!(sk->sk_userlocks & SOCK_RCVBUF_LOCK) ||
+ skb->len > tcp_sk(sk)->advmss))) {
u64 val = (u64)skb->len << TCP_RMEM_TO_WIN_SCALE;
u8 old_ratio = tcp_sk(sk)->scaling_ratio;
--
2.53.0
^ permalink raw reply related
* [PATCH net v2 0/2] tcp: protect locked SO_RCVBUF from Silly Window Syndrome
From: Ankit Jain @ 2026-05-04 14:49 UTC (permalink / raw)
To: edumazet, kuba, netdev
Cc: davem, pabeni, ncardwell, kuniyu, horms, shuah, quic_subashab,
quic_stranche, linux-kselftest, linux-kernel, karen.badiryan,
ajay.kaher, alexey.makhalov, vamsi-krishna.brahmajosyula,
yin.ding, tapas.kundu, Ankit Jain
This series fixes a regression where locked SO_RCVBUF sockets suffer from
Silly Window Syndrome (SWS).
Recent memory fragmentation optimizations apply truesize penalties to the
scaling_ratio. For applications locking SO_RCVBUF (like Java/Tomcat) and
processing small packets, this drops the scaling_ratio to 1. This
collapses the advertised window and causes 504 Gateway Timeouts.
Patch 1 bypasses this penalty for locked buffers unless skb->len > advmss.
Patch 2 adds a packetdrill test.
Link to v1:
https://lore.kernel.org/all/20260427152756.1205-1-ankit-aj.jain@broadcom.com/
v1 -> v2:
- Shifted protection from window_clamp to scaling_ratio based on Jakub
Kicinski's feedback.
- Added skb->len > advmss check to ensure large aggregate payloads (GRO)
are still penalized. This allows tcp_rcv_neg_window.pkt to pass.
- Added a new packetdrill test (Patch 2/2).
Testing:
- Verified fix in a live Java/Tomcat environment (504 timeouts resolved).
- Verified deadlock prevention via the new tcp_locked_rcvbuf_sws.pkt.
- Passed upstream regression tests: tcp_rcv_neg_window.pkt,
tcp_rcv_wnd_shrink_allowed.pkt, tcp_rcv_wnd_shrink_nomem.pkt,
tcp_rcv_zero_wnd_fin.pkt, and tcp_rcv_big_endseq.pkt.
Ankit Jain (2):
tcp: protect locked SO_RCVBUF from Silly Window Syndrome
selftests/net: add packetdrill test for locked SO_RCVBUF SWS
net/ipv4/tcp_input.c | 8 ++++-
.../net/packetdrill/tcp_locked_rcvbuf_sws.pkt | 34 +++++++++++++++++++
2 files changed, 41 insertions(+), 1 deletion(-)
create mode 100644 tools/testing/selftests/net/packetdrill/tcp_locked_rcvbuf_sws.pkt
--
2.53.0
^ permalink raw reply
* [PATCH net-next v42 7/8] eea: introduce ethtool support
From: Xuan Zhuo @ 2026-05-04 14:44 UTC (permalink / raw)
To: netdev
Cc: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, Xuan Zhuo, Wen Gu, Philo Lu, Vadim Fedorenko,
Dong Yibo, Mingyu Wang, Heiner Kallweit, Dust Li, Andrew Lunn
In-Reply-To: <20260504144439.29570-1-xuanzhuo@linux.alibaba.com>
Add basic driver framework for the Alibaba Elastic Ethernet Adapter(EEA).
This commit introduces ethtool support.
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Dust Li <dust.li@linux.alibaba.com>
Reviewed-by: Philo Lu <lulie@linux.alibaba.com>
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
---
drivers/net/ethernet/alibaba/eea/Makefile | 1 +
.../net/ethernet/alibaba/eea/eea_ethtool.c | 273 ++++++++++++++++++
.../net/ethernet/alibaba/eea/eea_ethtool.h | 48 +++
drivers/net/ethernet/alibaba/eea/eea_net.c | 1 +
drivers/net/ethernet/alibaba/eea/eea_net.h | 5 +
drivers/net/ethernet/alibaba/eea/eea_rx.c | 32 +-
drivers/net/ethernet/alibaba/eea/eea_tx.c | 27 +-
7 files changed, 380 insertions(+), 7 deletions(-)
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_ethtool.c
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_ethtool.h
diff --git a/drivers/net/ethernet/alibaba/eea/Makefile b/drivers/net/ethernet/alibaba/eea/Makefile
index 5f0961002e57..c38db22cca34 100644
--- a/drivers/net/ethernet/alibaba/eea/Makefile
+++ b/drivers/net/ethernet/alibaba/eea/Makefile
@@ -4,5 +4,6 @@ eea-y := eea_ring.o \
eea_net.o \
eea_pci.o \
eea_adminq.o \
+ eea_ethtool.o \
eea_tx.o \
eea_rx.o
diff --git a/drivers/net/ethernet/alibaba/eea/eea_ethtool.c b/drivers/net/ethernet/alibaba/eea/eea_ethtool.c
new file mode 100644
index 000000000000..479779aa2dd8
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_ethtool.c
@@ -0,0 +1,273 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#include <linux/ethtool.h>
+#include <linux/ethtool_netlink.h>
+#include <linux/rtnetlink.h>
+
+#include "eea_adminq.h"
+#include "eea_net.h"
+#include "eea_pci.h"
+
+struct eea_stat_desc {
+ char desc[ETH_GSTRING_LEN];
+ size_t offset;
+};
+
+#define EEA_TX_STAT(m) {#m, offsetof(struct eea_tx_stats, m)}
+#define EEA_RX_STAT(m) {#m, offsetof(struct eea_rx_stats, m)}
+
+static const struct eea_stat_desc eea_rx_stats_desc[] = {
+ EEA_RX_STAT(descs),
+ EEA_RX_STAT(kicks),
+};
+
+static const struct eea_stat_desc eea_tx_stats_desc[] = {
+ EEA_TX_STAT(descs),
+ EEA_TX_STAT(kicks),
+};
+
+#define EEA_TX_STATS_LEN ARRAY_SIZE(eea_tx_stats_desc)
+#define EEA_RX_STATS_LEN ARRAY_SIZE(eea_rx_stats_desc)
+
+static void eea_get_drvinfo(struct net_device *netdev,
+ struct ethtool_drvinfo *info)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ struct eea_device *edev = enet->edev;
+
+ strscpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+ strscpy(info->bus_info, eea_pci_name(edev), sizeof(info->bus_info));
+}
+
+static void eea_get_ringparam(struct net_device *netdev,
+ struct ethtool_ringparam *ring,
+ struct kernel_ethtool_ringparam *kernel_ring,
+ struct netlink_ext_ack *extack)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+
+ ring->rx_max_pending = enet->cfg_hw.rx_ring_depth;
+ ring->tx_max_pending = enet->cfg_hw.tx_ring_depth;
+ ring->rx_pending = enet->cfg.rx_ring_depth;
+ ring->tx_pending = enet->cfg.tx_ring_depth;
+
+ kernel_ring->tcp_data_split = enet->cfg.split_hdr ?
+ ETHTOOL_TCP_DATA_SPLIT_ENABLED :
+ ETHTOOL_TCP_DATA_SPLIT_DISABLED;
+}
+
+static int eea_set_ringparam(struct net_device *netdev,
+ struct ethtool_ringparam *ring,
+ struct kernel_ethtool_ringparam *kernel_ring,
+ struct netlink_ext_ack *extack)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ struct eea_net_init_ctx ctx;
+ bool need_update = false;
+ struct eea_net_cfg *cfg;
+ bool sh;
+
+ if (ring->rx_pending < EEA_NET_IO_RING_DEPTH_MIN ||
+ ring->tx_pending < EEA_NET_IO_RING_DEPTH_MIN)
+ return -EINVAL;
+
+ if (!is_power_of_2(ring->rx_pending) ||
+ !is_power_of_2(ring->tx_pending))
+ return -EINVAL;
+
+ eea_init_ctx(enet, &ctx);
+
+ cfg = &ctx.cfg;
+
+ if (ring->rx_pending != cfg->rx_ring_depth)
+ need_update = true;
+
+ if (ring->tx_pending != cfg->tx_ring_depth)
+ need_update = true;
+
+ sh = false;
+
+ switch (kernel_ring->tcp_data_split) {
+ case ETHTOOL_TCP_DATA_SPLIT_ENABLED:
+ sh = true;
+ break;
+
+ case ETHTOOL_TCP_DATA_SPLIT_DISABLED:
+ sh = false;
+ break;
+
+ case ETHTOOL_TCP_DATA_SPLIT_UNKNOWN:
+ sh = !!cfg->split_hdr;
+ break;
+ }
+
+ if (sh != !!(cfg->split_hdr))
+ need_update = true;
+
+ if (!need_update)
+ return 0;
+
+ cfg->rx_ring_depth = ring->rx_pending;
+ cfg->tx_ring_depth = ring->tx_pending;
+
+ /* By default, enet->cfg_hw.split_hdr is 128. */
+ cfg->split_hdr = sh ? enet->cfg_hw.split_hdr : 0;
+
+ return eea_reset_hw_resources(enet, &ctx);
+}
+
+static int eea_set_channels(struct net_device *netdev,
+ struct ethtool_channels *channels)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ u16 queue_pairs = channels->combined_count;
+ struct eea_net_init_ctx ctx;
+ struct eea_net_cfg *cfg;
+
+ eea_init_ctx(enet, &ctx);
+
+ cfg = &ctx.cfg;
+
+ cfg->rx_ring_num = queue_pairs;
+ cfg->tx_ring_num = queue_pairs;
+
+ return eea_reset_hw_resources(enet, &ctx);
+}
+
+static void eea_get_channels(struct net_device *netdev,
+ struct ethtool_channels *channels)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+
+ channels->combined_count = enet->cfg.rx_ring_num;
+ channels->max_combined = enet->cfg_hw.rx_ring_num;
+}
+
+static void eea_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ u8 *p = data;
+ u32 i, j;
+
+ if (stringset != ETH_SS_STATS)
+ return;
+
+ for (i = 0; i < enet->cfg.rx_ring_num; i++) {
+ for (j = 0; j < EEA_RX_STATS_LEN; j++)
+ ethtool_sprintf(&p, "rx%u_%s", i,
+ eea_rx_stats_desc[j].desc);
+ }
+
+ for (i = 0; i < enet->cfg.tx_ring_num; i++) {
+ for (j = 0; j < EEA_TX_STATS_LEN; j++)
+ ethtool_sprintf(&p, "tx%u_%s", i,
+ eea_tx_stats_desc[j].desc);
+ }
+}
+
+static int eea_get_sset_count(struct net_device *netdev, int sset)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+
+ if (sset != ETH_SS_STATS)
+ return -EOPNOTSUPP;
+
+ return enet->cfg.rx_ring_num * EEA_RX_STATS_LEN +
+ enet->cfg.tx_ring_num * EEA_TX_STATS_LEN;
+}
+
+static void eea_stats_fill_for_q(struct u64_stats_sync *syncp, u32 num,
+ const struct eea_stat_desc *desc,
+ u64 *data, u32 idx)
+{
+ void *stats_base = syncp;
+ u32 start, i;
+
+ do {
+ start = u64_stats_fetch_begin(syncp);
+ for (i = 0; i < num; i++)
+ data[idx + i] =
+ u64_stats_read(stats_base + desc[i].offset);
+
+ } while (u64_stats_fetch_retry(syncp, start));
+
+ BUILD_BUG_ON(offsetof(struct eea_tx_stats, syncp));
+ BUILD_BUG_ON(offsetof(struct eea_rx_stats, syncp));
+}
+
+static void eea_get_ethtool_stats(struct net_device *netdev,
+ struct ethtool_stats *stats, u64 *data)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ u32 i, idx = 0;
+
+ ASSERT_RTNL();
+
+ if (enet->rx) {
+ for (i = 0; i < enet->cfg.rx_ring_num; i++) {
+ struct eea_net_rx *rx = enet->rx[i];
+
+ eea_stats_fill_for_q(&rx->stats.syncp, EEA_RX_STATS_LEN,
+ eea_rx_stats_desc, data, idx);
+
+ idx += EEA_RX_STATS_LEN;
+ }
+ }
+
+ if (enet->tx) {
+ for (i = 0; i < enet->cfg.tx_ring_num; i++) {
+ struct eea_net_tx *tx = &enet->tx[i];
+
+ eea_stats_fill_for_q(&tx->stats.syncp, EEA_TX_STATS_LEN,
+ eea_tx_stats_desc, data, idx);
+
+ idx += EEA_TX_STATS_LEN;
+ }
+ }
+}
+
+void eea_update_rx_stats(struct eea_rx_stats *rx_stats,
+ struct eea_rx_ctx_stats *stats)
+{
+ u64_stats_update_begin(&rx_stats->syncp);
+ u64_stats_add(&rx_stats->descs, stats->descs);
+ u64_stats_add(&rx_stats->packets, stats->packets);
+ u64_stats_add(&rx_stats->bytes, stats->bytes);
+ u64_stats_add(&rx_stats->drops, stats->drops);
+ u64_stats_add(&rx_stats->split_hdr_bytes, stats->split_hdr_bytes);
+ u64_stats_add(&rx_stats->split_hdr_packets, stats->split_hdr_packets);
+ u64_stats_add(&rx_stats->length_errors, stats->length_errors);
+ u64_stats_add(&rx_stats->kicks, stats->kicks);
+ u64_stats_update_end(&rx_stats->syncp);
+}
+
+static int eea_get_link_ksettings(struct net_device *netdev,
+ struct ethtool_link_ksettings *cmd)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+
+ cmd->base.speed = enet->speed;
+ cmd->base.duplex = enet->duplex;
+ cmd->base.port = PORT_OTHER;
+
+ return 0;
+}
+
+const struct ethtool_ops eea_ethtool_ops = {
+ .supported_ring_params = ETHTOOL_RING_USE_TCP_DATA_SPLIT,
+ .get_drvinfo = eea_get_drvinfo,
+ .get_link = ethtool_op_get_link,
+ .get_ringparam = eea_get_ringparam,
+ .set_ringparam = eea_set_ringparam,
+ .set_channels = eea_set_channels,
+ .get_channels = eea_get_channels,
+ .get_strings = eea_get_strings,
+ .get_sset_count = eea_get_sset_count,
+ .get_ethtool_stats = eea_get_ethtool_stats,
+ .get_link_ksettings = eea_get_link_ksettings,
+};
diff --git a/drivers/net/ethernet/alibaba/eea/eea_ethtool.h b/drivers/net/ethernet/alibaba/eea/eea_ethtool.h
new file mode 100644
index 000000000000..08c824a116de
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_ethtool.h
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#ifndef __EEA_ETHTOOL_H__
+#define __EEA_ETHTOOL_H__
+
+struct eea_tx_stats {
+ struct u64_stats_sync syncp;
+ u64_stats_t descs;
+ u64_stats_t packets;
+ u64_stats_t bytes;
+ u64_stats_t drops;
+ u64_stats_t kicks;
+};
+
+struct eea_rx_ctx_stats {
+ u64 descs;
+ u64 packets;
+ u64 bytes;
+ u64 drops;
+ u64 split_hdr_bytes;
+ u64 split_hdr_packets;
+ u64 kicks;
+ u64 length_errors;
+};
+
+struct eea_rx_stats {
+ struct u64_stats_sync syncp;
+ u64_stats_t descs;
+ u64_stats_t packets;
+ u64_stats_t bytes;
+ u64_stats_t drops;
+ u64_stats_t kicks;
+ u64_stats_t split_hdr_bytes;
+ u64_stats_t split_hdr_packets;
+
+ u64_stats_t length_errors;
+};
+
+void eea_update_rx_stats(struct eea_rx_stats *rx_stats,
+ struct eea_rx_ctx_stats *stats);
+
+extern const struct ethtool_ops eea_ethtool_ops;
+#endif
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.c b/drivers/net/ethernet/alibaba/eea/eea_net.c
index 5d5f87a94e83..c22fdf923130 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_net.c
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.c
@@ -623,6 +623,7 @@ static struct eea_net *eea_netdev_alloc(struct eea_device *edev, u32 pairs)
}
netdev->netdev_ops = &eea_netdev;
+ netdev->ethtool_ops = &eea_ethtool_ops;
SET_NETDEV_DEV(netdev, edev->dma_dev);
enet = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.h b/drivers/net/ethernet/alibaba/eea/eea_net.h
index 1ff9ef1a8fbe..6a5b2f908c76 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_net.h
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.h
@@ -12,6 +12,7 @@
#include <linux/netdevice.h>
#include "eea_adminq.h"
+#include "eea_ethtool.h"
#include "eea_ring.h"
#define EEA_VER_MAJOR 1
@@ -33,6 +34,8 @@ struct eea_net_tx {
u32 index;
char name[16];
+
+ struct eea_tx_stats stats;
};
struct eea_rx_meta {
@@ -87,6 +90,8 @@ struct eea_net_rx {
struct napi_struct *napi;
+ struct eea_rx_stats stats;
+
char name[16];
struct eea_net_rx_pkt_ctx pkt;
diff --git a/drivers/net/ethernet/alibaba/eea/eea_rx.c b/drivers/net/ethernet/alibaba/eea/eea_rx.c
index 9c71a7cf950e..a3f0d2a79ad8 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_rx.c
+++ b/drivers/net/ethernet/alibaba/eea/eea_rx.c
@@ -29,6 +29,8 @@ struct eea_rx_ctx {
bool more;
struct eea_rx_meta *meta;
+
+ struct eea_rx_ctx_stats stats;
};
static struct eea_rx_meta *eea_rx_meta_get(struct eea_net_rx *rx)
@@ -232,6 +234,7 @@ static int eea_harden_check_overflow(struct eea_rx_ctx *ctx,
if (unlikely(ctx->len > max_len)) {
pr_debug("%s: rx error: len %u exceeds truesize %u\n",
enet->netdev->name, ctx->len, max_len);
+ ++ctx->stats.length_errors;
return -EINVAL;
}
@@ -250,6 +253,7 @@ static int eea_harden_check_size(struct eea_rx_ctx *ctx, struct eea_net *enet)
if (unlikely(ctx->hdr_len < ETH_HLEN)) {
pr_debug("%s: short hdr %u\n", enet->netdev->name,
ctx->hdr_len);
+ ++ctx->stats.length_errors;
return -EINVAL;
}
@@ -257,6 +261,7 @@ static int eea_harden_check_size(struct eea_rx_ctx *ctx, struct eea_net *enet)
pr_debug("%s: rx error: hdr len %u exceeds hdr buffer size %u\n",
enet->netdev->name, ctx->hdr_len,
enet->cfg.split_hdr);
+ ++ctx->stats.length_errors;
return -EINVAL;
}
@@ -265,6 +270,7 @@ static int eea_harden_check_size(struct eea_rx_ctx *ctx, struct eea_net *enet)
if (unlikely(ctx->len < ETH_HLEN)) {
pr_debug("%s: short packet %u\n", enet->netdev->name, ctx->len);
+ ++ctx->stats.length_errors;
return -EINVAL;
}
@@ -373,6 +379,7 @@ static void process_remain_buf(struct eea_net_rx *rx, struct eea_rx_ctx *ctx)
err:
dev_kfree_skb(rx->pkt.head_skb);
+ ++ctx->stats.drops;
rx->pkt.do_drop = true;
rx->pkt.head_skb = NULL;
}
@@ -400,6 +407,7 @@ static void process_first_buf(struct eea_net_rx *rx, struct eea_rx_ctx *ctx)
return;
err:
+ ++ctx->stats.drops;
rx->pkt.do_drop = true;
}
@@ -460,6 +468,8 @@ static int eea_rx_desc_to_ctx(struct eea_net_rx *rx,
if (ctx->flags & EEA_DESC_F_SPLIT_HDR) {
ctx->hdr_len = le16_to_cpu(desc->len_ex) &
EEA_RX_CDESC_HDR_LEN_MASK;
+ ctx->stats.split_hdr_bytes += ctx->hdr_len;
+ ++ctx->stats.split_hdr_packets;
}
ctx->more = ctx->flags & EEA_RING_DESC_F_MORE;
@@ -484,8 +494,10 @@ static int eea_cleanrx(struct eea_net_rx *rx, int budget,
if (ctx->meta)
eea_rx_meta_put(rx, ctx->meta);
- if (rx->pkt.head_skb)
+ if (rx->pkt.head_skb) {
dev_kfree_skb(rx->pkt.head_skb);
+ ++ctx->stats.drops;
+ }
/* A hardware error occurred; we are attempting to
* mitigate the impact. Subsequent packets may be
@@ -512,13 +524,17 @@ static int eea_cleanrx(struct eea_net_rx *rx, int budget,
++rx->pkt.idx;
- if (!ctx->more && rx->pkt.head_skb)
+ if (!ctx->more && rx->pkt.head_skb) {
eea_submit_skb(rx, rx->pkt.head_skb, desc);
+ ctx->stats.bytes += rx->pkt.recv_len;
+ ++ctx->stats.packets;
+ }
skip:
eea_rx_meta_put(rx, meta);
ack:
eea_ering_cq_ack_desc(rx->ering, 1);
+ ++ctx->stats.descs;
if (!ctx->more) {
memset(&rx->pkt, 0, sizeof(rx->pkt));
@@ -537,7 +553,7 @@ static void eea_rx_dma_sync_hdr(struct eea_net_rx *rx, dma_addr_t addr)
}
/* Only be called from napi. */
-static void eea_rx_post(struct eea_net_rx *rx)
+static void eea_rx_post(struct eea_net_rx *rx, struct eea_rx_ctx *ctx)
{
u32 tailroom, headroom, room, len;
struct eea_rx_meta *meta;
@@ -586,8 +602,10 @@ static void eea_rx_post(struct eea_net_rx *rx)
++num;
}
- if (num)
+ if (num) {
eea_ering_kick(rx->ering);
+ ++ctx->stats.kicks;
+ }
}
static int eea_poll(struct napi_struct *napi, int budget)
@@ -608,12 +626,14 @@ static int eea_poll(struct napi_struct *napi, int budget)
* buffers are exhausted. Therefore, we should proactively
* pre-fill the buffers to avoid starvation.
*/
- eea_rx_post(rx);
+ eea_rx_post(rx, &ctx);
if (rx->ering->num - rx->ering->num_free < budget)
busy = true;
}
+ eea_update_rx_stats(&rx->stats, &ctx.stats);
+
busy |= received >= budget;
if (busy)
@@ -751,6 +771,8 @@ struct eea_net_rx *eea_alloc_rx(struct eea_net_init_ctx *ctx, u32 idx)
rx->index = idx;
snprintf(rx->name, sizeof(rx->name), "rx.%u", idx);
+ u64_stats_init(&rx->stats.syncp);
+
/* ering */
ering = eea_ering_alloc(idx * 2, ctx->cfg.rx_ring_depth, ctx->edev,
ctx->cfg.rx_sq_desc_size,
diff --git a/drivers/net/ethernet/alibaba/eea/eea_tx.c b/drivers/net/ethernet/alibaba/eea/eea_tx.c
index 1c5a55184ad9..340040634214 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_tx.c
+++ b/drivers/net/ethernet/alibaba/eea/eea_tx.c
@@ -93,6 +93,7 @@ static void eea_meta_free_xmit(struct eea_net_tx *tx,
}
++stats->packets;
+ stats->bytes += meta->skb->len;
napi_consume_skb(meta->skb, budget);
meta->data = NULL;
@@ -137,6 +138,13 @@ static int eea_clean_tx(struct eea_net_tx *tx, int budget)
eea_ering_cq_ack_desc(tx->ering, desc_n);
}
+ if (stats.packets) {
+ u64_stats_update_begin(&tx->stats.syncp);
+ u64_stats_add(&tx->stats.bytes, stats.bytes);
+ u64_stats_add(&tx->stats.packets, stats.packets);
+ u64_stats_update_end(&tx->stats.syncp);
+ }
+
return stats.packets;
}
@@ -341,6 +349,10 @@ static int eea_tx_post_skb(struct eea_net_tx *tx, struct sk_buff *skb)
eea_ering_sq_commit_desc(tx->ering);
+ u64_stats_update_begin(&tx->stats.syncp);
+ u64_stats_add(&tx->stats.descs, meta->num);
+ u64_stats_update_end(&tx->stats.syncp);
+
return 0;
err_cancel:
@@ -353,6 +365,10 @@ static int eea_tx_post_skb(struct eea_net_tx *tx, struct sk_buff *skb)
static void eea_tx_kick(struct eea_net_tx *tx)
{
eea_ering_kick(tx->ering);
+
+ u64_stats_update_begin(&tx->stats.syncp);
+ u64_stats_inc(&tx->stats.kicks);
+ u64_stats_update_end(&tx->stats.syncp);
}
static int eea_tx_check_free_num(struct eea_net_tx *tx,
@@ -382,10 +398,15 @@ netdev_tx_t eea_tx_xmit(struct sk_buff *skb, struct net_device *netdev)
return NETDEV_TX_BUSY;
err = eea_tx_post_skb(tx, skb);
- if (unlikely(err))
+ if (unlikely(err)) {
+ u64_stats_update_begin(&tx->stats.syncp);
+ u64_stats_inc(&tx->stats.drops);
+ u64_stats_update_end(&tx->stats.syncp);
+
dev_kfree_skb_any(skb);
- else
+ } else {
skb_tx_timestamp(skb);
+ }
/* NETDEV_TX_BUSY is expensive. So stop advancing the TX queue. */
eea_tx_check_free_num(tx, txq);
@@ -441,6 +462,8 @@ int eea_alloc_tx(struct eea_net_init_ctx *ctx, struct eea_net_tx *tx, u32 idx)
struct eea_ring *ering;
u32 i;
+ u64_stats_init(&tx->stats.syncp);
+
snprintf(tx->name, sizeof(tx->name), "tx.%u", idx);
ering = eea_ering_alloc(idx * 2 + 1, ctx->cfg.tx_ring_depth, ctx->edev,
--
2.32.0.3.g01195cf9f
^ permalink raw reply related
* [PATCH net-next v42 4/8] eea: create/destroy rx,tx queues for netdevice open and stop
From: Xuan Zhuo @ 2026-05-04 14:44 UTC (permalink / raw)
To: netdev
Cc: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, Xuan Zhuo, Wen Gu, Philo Lu, Vadim Fedorenko,
Dong Yibo, Mingyu Wang, Heiner Kallweit, Dust Li
In-Reply-To: <20260504144439.29570-1-xuanzhuo@linux.alibaba.com>
Add basic driver framework for the Alibaba Elastic Ethernet Adapter(EEA).
This commit introduces the implementation for the netdevice open and
stop.
This commit introduces HA to restore the device when error occurs,
but in HA scenarios the driver can't ensure to restore the status
correctly.
Reviewed-by: Dust Li <dust.li@linux.alibaba.com>
Reviewed-by: Philo Lu <lulie@linux.alibaba.com>
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
---
drivers/net/ethernet/alibaba/eea/Makefile | 4 +-
drivers/net/ethernet/alibaba/eea/eea_net.c | 575 ++++++++++++++++++++-
drivers/net/ethernet/alibaba/eea/eea_net.h | 48 +-
drivers/net/ethernet/alibaba/eea/eea_pci.c | 228 +++++++-
drivers/net/ethernet/alibaba/eea/eea_pci.h | 20 +
drivers/net/ethernet/alibaba/eea/eea_rx.c | 264 ++++++++++
drivers/net/ethernet/alibaba/eea/eea_tx.c | 102 ++++
7 files changed, 1227 insertions(+), 14 deletions(-)
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_rx.c
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_tx.c
diff --git a/drivers/net/ethernet/alibaba/eea/Makefile b/drivers/net/ethernet/alibaba/eea/Makefile
index a842ac416ae8..5f0961002e57 100644
--- a/drivers/net/ethernet/alibaba/eea/Makefile
+++ b/drivers/net/ethernet/alibaba/eea/Makefile
@@ -3,4 +3,6 @@ obj-$(CONFIG_ALIBABA_EEA) += eea.o
eea-y := eea_ring.o \
eea_net.o \
eea_pci.o \
- eea_adminq.o
+ eea_adminq.o \
+ eea_tx.o \
+ eea_rx.o
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.c b/drivers/net/ethernet/alibaba/eea/eea_net.c
index bb8a49f8c6df..5d5f87a94e83 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_net.c
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.c
@@ -18,6 +18,459 @@
#define EEA_SPLIT_HDR_SIZE ALIGN(128, L1_CACHE_BYTES)
+static irqreturn_t eea_irq_handler(int irq, void *data)
+{
+ struct eea_irq_blk *blk = data;
+
+ napi_schedule_irqoff(&blk->napi);
+
+ return IRQ_HANDLED;
+}
+
+static void eea_free_irq_blk(struct eea_net *enet)
+{
+ struct eea_irq_blk *blk;
+ u32 num;
+ int i;
+
+ if (!enet->irq_blks)
+ return;
+
+ num = enet->edev->rx_num;
+
+ for (i = 0; i < num; i++) {
+ blk = &enet->irq_blks[i];
+
+ if (blk->ready)
+ eea_pci_free_irq(blk);
+
+ blk->ready = false;
+ }
+
+ kvfree(enet->irq_blks);
+ enet->irq_blks = NULL;
+}
+
+/* The driver will always attempt to allocate IRQ blocks based on the maximum
+ * possible queue num.
+ */
+static int eea_alloc_irq_blks(struct eea_net *enet)
+{
+ struct eea_device *edev = enet->edev;
+ struct eea_irq_blk *blk, *irq_blks;
+ int i, err, num;
+
+ num = enet->edev->rx_num;
+
+ irq_blks = kvcalloc(num, sizeof(*blk), GFP_KERNEL);
+ if (!irq_blks)
+ return -ENOMEM;
+
+ enet->irq_blks = irq_blks;
+
+ for (i = 0; i < num; i++) {
+ blk = &irq_blks[i];
+ blk->idx = i;
+
+ /* vec 0 is for error notify. */
+ blk->msix_vec = i + 1;
+
+ err = eea_pci_request_irq(edev, blk, eea_irq_handler);
+ if (err)
+ goto err_free_irq_blk;
+
+ blk->ready = true;
+ }
+
+ return 0;
+
+err_free_irq_blk:
+ eea_free_irq_blk(enet);
+ return err;
+}
+
+static int eea_update_queues(struct eea_net *enet)
+{
+ return netif_set_real_num_queues(enet->netdev, enet->cfg.tx_ring_num,
+ enet->cfg.rx_ring_num);
+}
+
+void eea_init_ctx(struct eea_net *enet, struct eea_net_init_ctx *ctx)
+{
+ memset(ctx, 0, sizeof(*ctx));
+
+ ctx->netdev = enet->netdev;
+ ctx->edev = enet->edev;
+ ctx->cfg = enet->cfg;
+}
+
+static void eea_bind_q_and_cfg(struct eea_net *enet,
+ struct eea_net_init_ctx *ctx)
+{
+ struct eea_irq_blk *blk;
+ struct eea_net_rx *rx;
+ struct eea_net_tx *tx;
+ int i;
+
+ enet->cfg = ctx->cfg;
+ enet->rx = ctx->rx;
+ enet->tx = ctx->tx;
+
+ for (i = 0; i < ctx->cfg.rx_ring_num; i++) {
+ blk = &enet->irq_blks[i];
+
+ rx = ctx->rx[i];
+ tx = &ctx->tx[i];
+
+ rx->enet = enet;
+ rx->napi = &blk->napi;
+ rx->ering->msix_vec = blk->msix_vec;
+
+ tx->enet = enet;
+ tx->ering->msix_vec = blk->msix_vec;
+
+ blk->rx = rx;
+ }
+}
+
+static void eea_unbind_q_and_cfg(struct eea_net *enet,
+ struct eea_net_init_ctx *ctx)
+{
+ struct eea_irq_blk *blk;
+ struct eea_net_rx *rx;
+ int i;
+
+ ctx->cfg = enet->cfg;
+ ctx->rx = enet->rx;
+ ctx->tx = enet->tx;
+
+ enet->rx = NULL;
+ enet->tx = NULL;
+
+ for (i = 0; i < ctx->cfg.rx_ring_num; i++) {
+ blk = &enet->irq_blks[i];
+
+ rx = ctx->rx[i];
+
+ rx->napi = NULL;
+
+ blk->rx = NULL;
+ }
+}
+
+static void eea_free_rxtx_q_mem(struct eea_net_init_ctx *ctx)
+{
+ struct eea_net_rx *rx;
+ struct eea_net_tx *tx;
+ int i;
+
+ for (i = 0; i < ctx->cfg.rx_ring_num; i++) {
+ rx = ctx->rx[i];
+ tx = &ctx->tx[i];
+
+ eea_free_rx(rx, &ctx->cfg);
+ eea_free_tx(tx, &ctx->cfg);
+ }
+
+ kvfree(ctx->rx);
+ kvfree(ctx->tx);
+}
+
+/* alloc tx/rx: struct, ring, meta, pp, napi */
+static int eea_alloc_rxtx_q_mem(struct eea_net_init_ctx *ctx)
+{
+ struct eea_net_rx *rx;
+ struct eea_net_tx *tx;
+ int err, i;
+
+ ctx->tx = kvcalloc(ctx->cfg.tx_ring_num, sizeof(*ctx->tx), GFP_KERNEL);
+ if (!ctx->tx)
+ return -ENOMEM;
+
+ ctx->rx = kvcalloc(ctx->cfg.rx_ring_num, sizeof(*ctx->rx), GFP_KERNEL);
+ if (!ctx->rx)
+ goto err_free_tx;
+
+ ctx->cfg.rx_sq_desc_size = sizeof(struct eea_rx_desc);
+ ctx->cfg.rx_cq_desc_size = sizeof(struct eea_rx_cdesc);
+ ctx->cfg.tx_sq_desc_size = sizeof(struct eea_tx_desc);
+ ctx->cfg.tx_cq_desc_size = sizeof(struct eea_tx_cdesc);
+
+ /* ethtool may config this. */
+ if (!ctx->cfg.split_hdr)
+ ctx->cfg.rx_sq_desc_size = sizeof(struct eea_rx_desc_no_hdr);
+
+ for (i = 0; i < ctx->cfg.rx_ring_num; i++) {
+ rx = eea_alloc_rx(ctx, i);
+ if (!rx)
+ goto err_free;
+
+ ctx->rx[i] = rx;
+
+ tx = ctx->tx + i;
+ err = eea_alloc_tx(ctx, tx, i);
+ if (err)
+ goto err_free;
+ }
+
+ return 0;
+
+err_free:
+ for (i = 0; i < ctx->cfg.rx_ring_num; i++) {
+ rx = ctx->rx[i];
+ tx = ctx->tx + i;
+
+ eea_free_rx(rx, &ctx->cfg);
+ eea_free_tx(tx, &ctx->cfg);
+ }
+
+ kvfree(ctx->rx);
+
+err_free_tx:
+ kvfree(ctx->tx);
+ return -ENOMEM;
+}
+
+static int eea_hw_active_ring(struct eea_net *enet)
+{
+ return eea_adminq_create_q(enet, enet->cfg.rx_ring_num
+ + enet->cfg.tx_ring_num, 0);
+}
+
+static int eea_hw_unactive_ring(struct eea_net *enet)
+{
+ int err;
+
+ err = eea_adminq_destroy_all_q(enet);
+ if (err)
+ netdev_warn(enet->netdev, "unactive rxtx ring failed.\n");
+
+ return err;
+}
+
+/* stop rx napi, stop tx queue. */
+static void eea_stop_rxtx(struct net_device *netdev)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ int i;
+
+ netif_tx_disable(netdev);
+
+ for (i = 0; i < enet->cfg.rx_ring_num; i++)
+ enet_rx_stop(enet->rx[i]);
+
+ netif_carrier_off(netdev);
+}
+
+static void eea_start_rxtx(struct eea_net *enet)
+{
+ int i;
+
+ for (i = 0; i < enet->cfg.rx_ring_num; i++)
+ enet_rx_start(enet->rx[i]);
+
+ netif_tx_start_all_queues(enet->netdev);
+ netif_carrier_on(enet->netdev);
+
+ enet->started = true;
+}
+
+static int eea_netdev_stop(struct net_device *netdev)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ struct eea_net_init_ctx ctx;
+
+ /* This function can be called during device anomaly recovery. To
+ * prevent duplicate stop operations, the `started` flag is introduced
+ * for checking.
+ */
+
+ if (!enet->started) {
+ netdev_warn(netdev, "eea netdev stop: but dev is not started.\n");
+ return 0;
+ }
+
+ eea_stop_rxtx(netdev);
+ eea_hw_unactive_ring(enet);
+ eea_unbind_q_and_cfg(enet, &ctx);
+ eea_free_rxtx_q_mem(&ctx);
+
+ enet->started = false;
+
+ return 0;
+}
+
+static int eea_netdev_open(struct net_device *netdev)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ struct eea_net_init_ctx ctx;
+ int err;
+
+ if (enet->link_err) {
+ netdev_err(netdev, "netdev open err, because link error: %d\n",
+ enet->link_err);
+ return -EBUSY;
+ }
+
+ eea_init_ctx(enet, &ctx);
+
+ err = eea_alloc_rxtx_q_mem(&ctx);
+ if (err)
+ goto err_done;
+
+ eea_bind_q_and_cfg(enet, &ctx);
+
+ err = eea_update_queues(enet);
+ if (err)
+ goto err_free_q;
+
+ err = eea_hw_active_ring(enet);
+ if (err)
+ goto err_free_q;
+
+ eea_start_rxtx(enet);
+
+ return 0;
+
+err_free_q:
+ eea_unbind_q_and_cfg(enet, &ctx);
+ eea_free_rxtx_q_mem(&ctx);
+
+err_done:
+ return err;
+}
+
+/* resources: ring, buffers, irq */
+int eea_reset_hw_resources(struct eea_net *enet, struct eea_net_init_ctx *ctx)
+{
+ struct eea_net_init_ctx ctx_old = {0};
+ int err, error;
+
+ if (!netif_running(enet->netdev) || !enet->started) {
+ enet->cfg = ctx->cfg;
+ return 0;
+ }
+
+ err = eea_alloc_rxtx_q_mem(ctx);
+ if (err) {
+ netdev_warn(enet->netdev,
+ "eea reset: alloc q failed. stop reset. err %d\n",
+ err);
+ return err;
+ }
+
+ eea_stop_rxtx(enet->netdev);
+ eea_hw_unactive_ring(enet);
+
+ eea_unbind_q_and_cfg(enet, &ctx_old);
+ eea_bind_q_and_cfg(enet, ctx);
+
+ err = eea_update_queues(enet);
+ if (err) {
+ netdev_err(enet->netdev,
+ "eea reset: set real num queues failed. err %d\n",
+ err);
+ goto err_bind_old;
+ }
+
+ err = eea_hw_active_ring(enet);
+ if (err) {
+ netdev_err(enet->netdev, "eea reset: active new ring. err %d\n",
+ err);
+ eea_unbind_q_and_cfg(enet, ctx);
+ goto err_free_q;
+ }
+
+ eea_start_rxtx(enet);
+ eea_free_rxtx_q_mem(&ctx_old);
+ return 0;
+
+err_bind_old:
+ eea_unbind_q_and_cfg(enet, ctx);
+ eea_bind_q_and_cfg(enet, &ctx_old);
+ error = eea_hw_active_ring(enet);
+ if (error) {
+ netdev_err(enet->netdev, "eea reset: active old ring. err %d\n",
+ error);
+ eea_unbind_q_and_cfg(enet, &ctx_old);
+ err = error;
+ goto err_free_q;
+ }
+
+ eea_start_rxtx(enet);
+ eea_free_rxtx_q_mem(ctx);
+ return err;
+
+err_free_q:
+
+ /* An exception occurred at the hardware level, and there's not much we
+ * can do about it -- we can only release the resources first.
+ */
+ eea_free_rxtx_q_mem(ctx);
+ eea_free_rxtx_q_mem(&ctx_old);
+ enet->started = false;
+ return err;
+}
+
+int eea_queues_check_and_reset(struct eea_device *edev)
+{
+ struct eea_aq_dev_status dstatus = {0};
+ struct eea_aq_queue_status *qstatus;
+ struct eea_aq_queue_status *qs;
+ struct eea_net_init_ctx ctx;
+ bool need_reset = false;
+ int i, err = 0;
+
+ rtnl_lock();
+
+ if (!netif_running(edev->enet->netdev))
+ goto err_unlock;
+
+ /* Maybe stopped by ha. */
+ if (!edev->enet->started || edev->enet->link_err)
+ goto err_unlock;
+
+ err = eea_adminq_dev_status(edev->enet, &dstatus);
+ if (err) {
+ netdev_warn(edev->enet->netdev, "query queue status failed.\n");
+ err = -ENOMEM;
+ goto err_unlock;
+ }
+
+ if (le16_to_cpu(dstatus.status->link_status) == EEA_LINK_DOWN_STATUS) {
+ /* The device is broken, can not be up. */
+ eea_netdev_stop(edev->enet->netdev);
+ edev->enet->link_err = EEA_LINK_ERR_LINK_DOWN;
+ netdev_warn(edev->enet->netdev, "device link is down. stop device.\n");
+ goto err_free;
+ }
+
+ qstatus = dstatus.status->q_status;
+
+ for (i = 0; i < dstatus.num; ++i) {
+ qs = &qstatus[i];
+
+ if (le16_to_cpu(qs->status) == EEA_QUEUE_STATUS_NEED_RESET) {
+ netdev_warn(edev->enet->netdev,
+ "queue status: queue %u needs to reset\n",
+ le16_to_cpu(qs->qidx));
+ need_reset = true;
+ }
+ }
+
+ if (need_reset) {
+ eea_init_ctx(edev->enet, &ctx);
+ err = eea_reset_hw_resources(edev->enet, &ctx);
+ }
+
+err_free:
+ kfree(dstatus.status);
+
+err_unlock:
+ rtnl_unlock();
+ return err;
+}
+
static int eea_update_cfg(struct eea_net *enet,
struct eea_device *edev,
struct eea_aq_cfg *hwcfg)
@@ -149,6 +602,9 @@ static int eea_netdev_init_features(struct net_device *netdev,
}
static const struct net_device_ops eea_netdev = {
+ .ndo_open = eea_netdev_open,
+ .ndo_stop = eea_netdev_stop,
+ .ndo_start_xmit = eea_tx_xmit,
.ndo_validate_addr = eth_validate_addr,
.ndo_features_check = passthru_features_check,
};
@@ -157,6 +613,7 @@ static struct eea_net *eea_netdev_alloc(struct eea_device *edev, u32 pairs)
{
struct net_device *netdev;
struct eea_net *enet;
+ int err;
netdev = alloc_etherdev_mq(sizeof(struct eea_net), pairs);
if (!netdev) {
@@ -173,14 +630,83 @@ static struct eea_net *eea_netdev_alloc(struct eea_device *edev, u32 pairs)
enet->edev = edev;
edev->enet = enet;
+ err = eea_alloc_irq_blks(enet);
+ if (err) {
+ dev_err(edev->dma_dev,
+ "eea_alloc_irq_blks failed with pairs %d\n", pairs);
+ free_netdev(netdev);
+ return NULL;
+ }
+
return enet;
}
+static void eea_update_ts_off(struct eea_device *edev, struct eea_net *enet)
+{
+ u64 ts;
+
+ ts = eea_pci_device_ts(edev);
+
+ enet->hw_ts_offset = ktime_get_real() - ts;
+}
+
+static int eea_net_reprobe(struct eea_device *edev)
+{
+ struct eea_net *enet = edev->enet;
+ int err = 0;
+
+ enet->edev = edev;
+
+ if (!enet->adminq.ring) {
+ err = eea_create_adminq(enet, edev->rx_num + edev->tx_num);
+ if (err)
+ return err;
+ }
+
+ err = eea_alloc_irq_blks(enet);
+ if (err)
+ goto err_destroy_aq;
+
+ eea_update_ts_off(edev, enet);
+
+ rtnl_lock();
+
+ enet->link_err = 0;
+ if (edev->ha_reset_netdev_running &&
+ netif_running(edev->enet->netdev)) {
+ err = eea_netdev_open(enet->netdev);
+ if (err) {
+ enet->link_err = EEA_LINK_ERR_HA_RESET_DEV;
+ rtnl_unlock();
+ goto err_free_irq_blks;
+ }
+ }
+
+ rtnl_unlock();
+
+ enet->wait_pci_ready = false;
+ return 0;
+
+err_free_irq_blks:
+ eea_free_irq_blk(enet);
+
+err_destroy_aq:
+ eea_destroy_adminq(enet);
+
+ return err;
+}
+
int eea_net_probe(struct eea_device *edev)
{
struct eea_net *enet;
int err = -ENOMEM;
+ /* If edev->enet is not null, then this is called from ha reset worker.
+ * Call eea_net_reprobe() directly.
+ */
+ if (edev->enet)
+ return eea_net_reprobe(edev);
+
enet = eea_netdev_alloc(edev, edev->rx_num);
if (!enet)
return -ENOMEM;
@@ -195,6 +721,8 @@ int eea_net_probe(struct eea_device *edev)
if (err)
goto err_reset_dev;
+ eea_update_ts_off(edev, enet);
+
netdev_dbg(enet->netdev, "eea probe success.\n");
/* Queue TX/RX implementation is still in progress. register_netdev is
@@ -208,11 +736,34 @@ int eea_net_probe(struct eea_device *edev)
eea_destroy_adminq(enet);
err_free_netdev:
+ eea_free_irq_blk(enet);
free_netdev(enet->netdev);
return err;
}
-void eea_net_remove(struct eea_device *edev)
+static void eea_net_ha_reset_remove(struct eea_net *enet,
+ struct eea_device *edev)
+{
+ rtnl_lock();
+ edev->ha_reset_netdev_running = false;
+ if (netif_running(enet->netdev)) {
+ eea_netdev_stop(enet->netdev);
+ edev->ha_reset_netdev_running = true;
+ }
+
+ /* Prevent that the user set up the net device. */
+ enet->link_err = EEA_LINK_ERR_HA_RESET_DEV;
+
+ rtnl_unlock();
+
+ eea_device_reset(edev);
+ eea_destroy_adminq(enet);
+ eea_free_irq_blk(enet);
+
+ enet->wait_pci_ready = true;
+}
+
+void eea_net_remove(struct eea_device *edev, bool ha)
{
struct net_device *netdev;
struct eea_net *enet;
@@ -220,11 +771,19 @@ void eea_net_remove(struct eea_device *edev)
enet = edev->enet;
netdev = enet->netdev;
- netdev_dbg(enet->netdev, "eea removed.\n");
+ if (ha) {
+ if (enet->wait_pci_ready)
+ return;
- eea_device_reset(edev);
+ eea_net_ha_reset_remove(enet, edev);
+ return;
+ }
- eea_destroy_adminq(enet);
+ if (!enet->wait_pci_ready) {
+ eea_device_reset(edev);
+ eea_destroy_adminq(enet);
+ eea_free_irq_blk(enet);
+ }
free_netdev(netdev);
}
@@ -241,9 +800,11 @@ void eea_net_shutdown(struct eea_device *edev)
netif_device_detach(netdev);
- eea_device_reset(edev);
-
- eea_destroy_adminq(enet);
+ if (!enet->wait_pci_ready) {
+ eea_device_reset(edev);
+ eea_destroy_adminq(enet);
+ eea_free_irq_blk(enet);
+ }
rtnl_unlock();
}
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.h b/drivers/net/ethernet/alibaba/eea/eea_net.h
index fa0eec8af21b..0d5fc8d1375a 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_net.h
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.h
@@ -18,6 +18,8 @@
#define EEA_VER_MINOR 0
#define EEA_VER_SUB_MINOR 0
+struct eea_tx_meta;
+
struct eea_net_tx {
struct eea_net *enet;
@@ -100,6 +102,18 @@ struct eea_net_cfg {
u8 tx_cq_desc_size;
u32 split_hdr;
+
+ struct hwtstamp_config ts_cfg;
+};
+
+struct eea_net_init_ctx {
+ struct eea_net_cfg cfg;
+
+ struct eea_net_tx *tx;
+ struct eea_net_rx **rx;
+
+ struct net_device *netdev;
+ struct eea_device *edev;
};
enum {
@@ -108,6 +122,17 @@ enum {
EEA_LINK_ERR_LINK_DOWN,
};
+struct eea_irq_blk {
+ struct napi_struct napi;
+ u16 msix_vec;
+ bool ready;
+ struct eea_net_rx *rx;
+ char irq_name[32];
+ int irq;
+ int idx;
+
+};
+
struct eea_net {
struct eea_device *edev;
struct net_device *netdev;
@@ -120,9 +145,12 @@ struct eea_net {
struct eea_net_cfg cfg;
struct eea_net_cfg cfg_hw;
+ struct eea_irq_blk *irq_blks;
+
u32 link_err;
bool started;
+ bool wait_pci_ready;
u8 duplex;
u32 speed;
@@ -131,7 +159,25 @@ struct eea_net {
};
int eea_net_probe(struct eea_device *edev);
-void eea_net_remove(struct eea_device *edev);
+void eea_net_remove(struct eea_device *edev, bool ha);
void eea_net_shutdown(struct eea_device *edev);
+int eea_reset_hw_resources(struct eea_net *enet, struct eea_net_init_ctx *ctx);
+void eea_init_ctx(struct eea_net *enet, struct eea_net_init_ctx *ctx);
+int eea_queues_check_and_reset(struct eea_device *edev);
+
+/* rx apis */
+void enet_rx_stop(struct eea_net_rx *rx);
+void enet_rx_start(struct eea_net_rx *rx);
+
+void eea_free_rx(struct eea_net_rx *rx, struct eea_net_cfg *cfg);
+struct eea_net_rx *eea_alloc_rx(struct eea_net_init_ctx *ctx, u32 idx);
+
+/* tx apis */
+int eea_poll_tx(struct eea_net_tx *tx, int budget);
+netdev_tx_t eea_tx_xmit(struct sk_buff *skb, struct net_device *netdev);
+
+void eea_free_tx(struct eea_net_tx *tx, struct eea_net_cfg *cfg);
+int eea_alloc_tx(struct eea_net_init_ctx *ctx, struct eea_net_tx *tx, u32 idx);
+
#endif
diff --git a/drivers/net/ethernet/alibaba/eea/eea_pci.c b/drivers/net/ethernet/alibaba/eea/eea_pci.c
index e7874979155b..d7af2b26d6e3 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_pci.c
+++ b/drivers/net/ethernet/alibaba/eea/eea_pci.c
@@ -16,6 +16,9 @@
#define EEA_PCI_DB_MAX_SIZE 512
#define EEA_PCI_Q_MAX_NUM 1000
+#define EEA_PCI_CAP_RESET_DEVICE 0xFA
+#define EEA_PCI_CAP_RESET_FLAG BIT(1)
+
struct eea_pci_cfg {
__le32 reserve0;
__le32 reserve1;
@@ -58,9 +61,11 @@ struct eea_pci_device {
void __iomem *db_base;
void __iomem *db_end;
+ struct work_struct ha_handle_work;
char ha_irq_name[32];
u8 reset_pos;
bool shutdown;
+ bool ha_ready;
};
#define cfg_pointer(reg, item) \
@@ -75,6 +80,11 @@ struct eea_pci_device {
#define cfg_read32(reg, item) ioread32(cfg_pointer(reg, item))
#define cfg_read64(reg, item) ioread64(cfg_pointer(reg, item))
+/* Due to circular references, we have to add function definitions here. */
+static int __eea_pci_probe(struct pci_dev *pci_dev,
+ struct eea_pci_device *ep_dev, bool pci_probe);
+static void __eea_pci_remove(struct pci_dev *pci_dev, bool pci_remove);
+
const char *eea_pci_name(struct eea_device *edev)
{
return pci_name(edev->ep_dev->pci_dev);
@@ -201,6 +211,12 @@ static int eea_negotiate(struct eea_device *edev)
static void eea_pci_release_resource(struct eea_pci_device *ep_dev)
{
struct pci_dev *pci_dev = ep_dev->pci_dev;
+ struct eea_device *edev;
+
+ edev = &ep_dev->edev;
+
+ if (edev->status < EEA_PCI_STATUS_READY)
+ return;
if (ep_dev->reg) {
pci_iounmap(pci_dev, ep_dev->reg);
@@ -215,12 +231,16 @@ static void eea_pci_release_resource(struct eea_pci_device *ep_dev)
pci_clear_master(pci_dev);
pci_release_regions(pci_dev);
pci_disable_device(pci_dev);
+
+ edev->status = EEA_PCI_STATUS_NONE;
}
static int eea_pci_setup(struct pci_dev *pci_dev, struct eea_pci_device *ep_dev)
{
int err, n, ret, len;
+ ep_dev->edev.status = EEA_PCI_STATUS_ERR;
+
ep_dev->pci_dev = pci_dev;
err = pci_enable_device(pci_dev);
@@ -309,6 +329,8 @@ static int eea_pci_setup(struct pci_dev *pci_dev, struct eea_pci_device *ep_dev)
ep_dev->msix_vec_n = ret;
+ ep_dev->edev.status = EEA_PCI_STATUS_READY;
+
return 0;
err_clear_master:
@@ -361,6 +383,174 @@ int eea_pci_active_aq(struct eea_ring *ering, int msix_vec)
return 0;
}
+void eea_pci_free_irq(struct eea_irq_blk *blk)
+{
+ irq_update_affinity_hint(blk->irq, NULL);
+ free_irq(blk->irq, blk);
+}
+
+int eea_pci_request_irq(struct eea_device *edev, struct eea_irq_blk *blk,
+ irqreturn_t (*callback)(int irq, void *data))
+{
+ struct eea_pci_device *ep_dev = edev->ep_dev;
+ int irq;
+
+ snprintf(blk->irq_name, sizeof(blk->irq_name), "eea-q%d@%s", blk->idx,
+ pci_name(ep_dev->pci_dev));
+
+ irq = pci_irq_vector(ep_dev->pci_dev, blk->msix_vec);
+
+ blk->irq = irq;
+
+ return request_irq(irq, callback, IRQF_NO_AUTOEN, blk->irq_name, blk);
+}
+
+static void eea_ha_handle_reset(struct eea_pci_device *ep_dev)
+{
+ struct eea_device *edev;
+ struct pci_dev *pci_dev;
+ u16 reset;
+ int err;
+
+ if (!ep_dev->reset_pos) {
+ eea_queues_check_and_reset(&ep_dev->edev);
+ return;
+ }
+
+ edev = &ep_dev->edev;
+
+ pci_read_config_word(ep_dev->pci_dev, ep_dev->reset_pos, &reset);
+
+ /* Clear bits using 0xFFFF and ignore all previous messages. */
+ pci_write_config_word(ep_dev->pci_dev, ep_dev->reset_pos, 0xFFFF);
+
+ if (reset & EEA_PCI_CAP_RESET_FLAG) {
+ dev_warn(&ep_dev->pci_dev->dev, "recv device reset request.\n");
+
+ pci_dev = ep_dev->pci_dev;
+
+ /* The pci remove callback may hold this lock. If the
+ * pci remove callback is called, then we can ignore the
+ * ha interrupt.
+ */
+ if (mutex_trylock(&edev->ha_lock)) {
+ if (edev->status != EEA_PCI_STATUS_DONE) {
+ dev_err(&ep_dev->pci_dev->dev, "ha: reset device: pci status is %d. skip it.\n",
+ edev->status);
+
+ mutex_unlock(&edev->ha_lock);
+ return;
+ }
+
+ __eea_pci_remove(pci_dev, false);
+ err = __eea_pci_probe(pci_dev, ep_dev, false);
+ if (err)
+ /* Currently, for some reason, PCI
+ * initialization or network device re-probing
+ * has failed. Waiting for the PCI subsystem to
+ * call the remove callback to release the
+ * remaining resources.
+ */
+ dev_err(&ep_dev->pci_dev->dev,
+ "ha: re-setup failed.\n");
+
+ mutex_unlock(&edev->ha_lock);
+ } else {
+ /* Device removal is in progress, so return directly. */
+ dev_warn(&ep_dev->pci_dev->dev,
+ "ha device reset: trylock failed.\n");
+ }
+ return;
+ }
+
+ eea_queues_check_and_reset(&ep_dev->edev);
+}
+
+/* ha handle code */
+static void eea_ha_handle_work(struct work_struct *work)
+{
+ struct eea_pci_device *ep_dev;
+
+ ep_dev = container_of(work, struct eea_pci_device, ha_handle_work);
+
+ /* Ha interrupt is triggered, so there maybe some error, we may need to
+ * reset the device or reset some queues.
+ */
+ dev_warn(&ep_dev->pci_dev->dev, "recv ha interrupt.\n");
+
+ eea_ha_handle_reset(ep_dev);
+}
+
+static irqreturn_t eea_pci_ha_handle(int irq, void *data)
+{
+ struct eea_device *edev = data;
+
+ schedule_work(&edev->ep_dev->ha_handle_work);
+
+ return IRQ_HANDLED;
+}
+
+static void eea_pci_free_ha_irq(struct eea_device *edev)
+{
+ struct eea_pci_device *ep_dev = edev->ep_dev;
+ int irq;
+
+ if (ep_dev->ha_ready) {
+ irq = pci_irq_vector(ep_dev->pci_dev, 0);
+ free_irq(irq, edev);
+ ep_dev->ha_ready = false;
+ }
+}
+
+static int eea_pci_ha_init(struct eea_device *edev, struct pci_dev *pci_dev,
+ bool pci_probe)
+{
+ u8 pos, cfg_type_off, type, cfg_drv_off, cfg_dev_off;
+ struct eea_pci_device *ep_dev = edev->ep_dev;
+ int irq, err;
+
+ snprintf(ep_dev->ha_irq_name, sizeof(ep_dev->ha_irq_name), "eea-ha@%s",
+ pci_name(ep_dev->pci_dev));
+
+ irq = pci_irq_vector(ep_dev->pci_dev, 0);
+
+ if (pci_probe)
+ INIT_WORK(&ep_dev->ha_handle_work, eea_ha_handle_work);
+
+ /* This irq is not only work for ha, so request it always. */
+ err = request_irq(irq, eea_pci_ha_handle, 0, ep_dev->ha_irq_name, edev);
+ if (err)
+ return err;
+
+ ep_dev->ha_ready = true;
+ ep_dev->reset_pos = 0;
+
+ cfg_type_off = offsetof(struct eea_pci_cap, cfg_type);
+ cfg_drv_off = offsetof(struct eea_pci_reset_reg, driver);
+ cfg_dev_off = offsetof(struct eea_pci_reset_reg, device);
+
+ for (pos = pci_find_capability(pci_dev, PCI_CAP_ID_VNDR);
+ pos > 0;
+ pos = pci_find_next_capability(pci_dev, pos, PCI_CAP_ID_VNDR)) {
+ pci_read_config_byte(pci_dev, pos + cfg_type_off, &type);
+
+ if (type == EEA_PCI_CAP_RESET_DEVICE) {
+ /* notify device, driver support this feature. */
+ pci_write_config_word(pci_dev, pos + cfg_drv_off,
+ EEA_PCI_CAP_RESET_FLAG);
+ pci_write_config_word(pci_dev, pos + cfg_dev_off,
+ 0xFFFF);
+
+ edev->ep_dev->reset_pos = pos + cfg_dev_off;
+ return 0;
+ }
+ }
+
+ /* irq just for event notify */
+ dev_warn(&edev->ep_dev->pci_dev->dev, "Not Found reset cap.\n");
+ return 0;
+}
+
u64 eea_pci_device_ts(struct eea_device *edev)
{
struct eea_pci_device *ep_dev = edev->ep_dev;
@@ -393,12 +583,16 @@ static int eea_init_device(struct eea_device *edev)
}
static int __eea_pci_probe(struct pci_dev *pci_dev,
- struct eea_pci_device *ep_dev)
+ struct eea_pci_device *ep_dev,
+ bool pci_probe)
{
+ struct eea_device *edev;
int err;
pci_set_drvdata(pci_dev, ep_dev);
+ edev = &ep_dev->edev;
+
err = eea_pci_setup(pci_dev, ep_dev);
if (err)
return err;
@@ -407,20 +601,34 @@ static int __eea_pci_probe(struct pci_dev *pci_dev,
if (err)
goto err_pci_rel;
+ err = eea_pci_ha_init(edev, pci_dev, pci_probe);
+ if (err)
+ goto err_net_rm;
+
+ edev->status = EEA_PCI_STATUS_DONE;
+
return 0;
+err_net_rm:
+ eea_net_remove(edev, !pci_probe);
+
err_pci_rel:
eea_pci_release_resource(ep_dev);
return err;
}
-static void __eea_pci_remove(struct pci_dev *pci_dev)
+static void __eea_pci_remove(struct pci_dev *pci_dev, bool pci_remove)
{
struct eea_pci_device *ep_dev = pci_get_drvdata(pci_dev);
struct device *dev = get_device(&ep_dev->pci_dev->dev);
struct eea_device *edev = &ep_dev->edev;
- eea_net_remove(edev);
+ eea_pci_free_ha_irq(edev);
+
+ if (pci_remove)
+ flush_work(&ep_dev->ha_handle_work);
+
+ eea_net_remove(edev, !pci_remove);
eea_pci_release_resource(ep_dev);
@@ -445,8 +653,11 @@ static int eea_pci_probe(struct pci_dev *pci_dev,
ep_dev->pci_dev = pci_dev;
- err = __eea_pci_probe(pci_dev, ep_dev);
+ mutex_init(&edev->ha_lock);
+
+ err = __eea_pci_probe(pci_dev, ep_dev, true);
if (err) {
+ mutex_destroy(&edev->ha_lock);
pci_set_drvdata(pci_dev, NULL);
kfree(ep_dev);
}
@@ -457,10 +668,17 @@ static int eea_pci_probe(struct pci_dev *pci_dev,
static void eea_pci_remove(struct pci_dev *pci_dev)
{
struct eea_pci_device *ep_dev = pci_get_drvdata(pci_dev);
+ struct eea_device *edev;
- __eea_pci_remove(pci_dev);
+ edev = &ep_dev->edev;
+
+ mutex_lock(&edev->ha_lock);
+ __eea_pci_remove(pci_dev, true);
+ mutex_unlock(&edev->ha_lock);
pci_set_drvdata(pci_dev, NULL);
+
+ mutex_destroy(&edev->ha_lock);
kfree(ep_dev);
}
diff --git a/drivers/net/ethernet/alibaba/eea/eea_pci.h b/drivers/net/ethernet/alibaba/eea/eea_pci.h
index cfd278e2efde..6f6ba7d5514d 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_pci.h
+++ b/drivers/net/ethernet/alibaba/eea/eea_pci.h
@@ -10,8 +10,18 @@
#include <linux/pci.h>
+#include "eea_net.h"
#include "eea_ring.h"
+enum eea_pci_status {
+ EEA_PCI_STATUS_NONE,
+ EEA_PCI_STATUS_ERR,
+ EEA_PCI_STATUS_READY,
+ EEA_PCI_STATUS_DONE,
+};
+
+struct eea_irq_blk;
+
struct eea_pci_cap {
__u8 cap_vndr;
__u8 cap_next;
@@ -34,6 +44,12 @@ struct eea_device {
u64 features;
+ enum eea_pci_status status;
+ bool ha_reset_netdev_running;
+
+ /* ha lock for the race between ha work and pci remove */
+ struct mutex ha_lock;
+
u32 rx_num;
u32 tx_num;
u32 db_blk_size;
@@ -47,6 +63,10 @@ int eea_device_reset(struct eea_device *dev);
int eea_pci_set_aq_up(struct eea_device *dev);
int eea_pci_active_aq(struct eea_ring *ering, int msix_vec);
+int eea_pci_request_irq(struct eea_device *edev, struct eea_irq_blk *blk,
+ irqreturn_t (*callback)(int irq, void *data));
+void eea_pci_free_irq(struct eea_irq_blk *blk);
+
u64 eea_pci_device_ts(struct eea_device *edev);
void __iomem *eea_pci_db_addr(struct eea_device *edev, u32 off);
diff --git a/drivers/net/ethernet/alibaba/eea/eea_rx.c b/drivers/net/ethernet/alibaba/eea/eea_rx.c
new file mode 100644
index 000000000000..b1265048fbc3
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_rx.c
@@ -0,0 +1,264 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#include <net/netdev_rx_queue.h>
+#include <net/page_pool/helpers.h>
+
+#include "eea_adminq.h"
+#include "eea_net.h"
+#include "eea_pci.h"
+#include "eea_ring.h"
+
+#define EEA_ENABLE_F_NAPI BIT(0)
+
+#define EEA_PAGE_FRAGS_NUM 1024
+
+static void eea_free_rx_buffer(struct eea_net_rx *rx, struct eea_rx_meta *meta,
+ bool allow_direct)
+{
+ u32 drain_count;
+
+ drain_count = EEA_PAGE_FRAGS_NUM - meta->frags;
+
+ if (page_pool_unref_page(meta->page, drain_count) == 0)
+ page_pool_put_unrefed_page(rx->pp, meta->page, -1,
+ allow_direct);
+
+ meta->page = NULL;
+}
+
+static void eea_free_rx_hdr(struct eea_net_rx *rx, struct eea_net_cfg *cfg)
+{
+ struct eea_rx_meta *meta;
+ int i;
+
+ for (i = 0; i < cfg->rx_ring_depth; ++i) {
+ meta = &rx->meta[i];
+ meta->hdr_addr = NULL;
+
+ if (!meta->hdr_page)
+ continue;
+
+ dma_unmap_page(rx->dma_dev, meta->hdr_dma, PAGE_SIZE,
+ DMA_FROM_DEVICE);
+ put_page(meta->hdr_page);
+
+ meta->hdr_page = NULL;
+ }
+}
+
+static int eea_alloc_rx_hdr(struct eea_net_init_ctx *ctx, struct eea_net_rx *rx)
+{
+ struct page *hdr_page = NULL;
+ struct eea_rx_meta *meta;
+ u32 offset = 0, hdrsize;
+ struct device *dmadev;
+ dma_addr_t dma;
+ int i;
+
+ dmadev = ctx->edev->dma_dev;
+ hdrsize = ctx->cfg.split_hdr;
+
+ for (i = 0; i < ctx->cfg.rx_ring_depth; ++i) {
+ meta = &rx->meta[i];
+ meta->hdr_page = NULL;
+
+ if (!hdr_page || offset + hdrsize > PAGE_SIZE) {
+ hdr_page = alloc_page(GFP_KERNEL);
+ if (!hdr_page)
+ goto err;
+
+ dma = dma_map_page(dmadev, hdr_page, 0, PAGE_SIZE,
+ DMA_FROM_DEVICE);
+
+ if (unlikely(dma_mapping_error(dmadev, dma))) {
+ put_page(hdr_page);
+ goto err;
+ }
+
+ offset = 0;
+ meta->hdr_page = hdr_page;
+ }
+
+ meta->hdr_dma = dma + offset;
+ meta->hdr_addr = page_address(hdr_page) + offset;
+ offset += hdrsize;
+ }
+
+ return 0;
+
+err:
+ eea_free_rx_hdr(rx, &ctx->cfg);
+ return -ENOMEM;
+}
+
+static int eea_poll(struct napi_struct *napi, int budget)
+{
+ /* Empty function; will be implemented in a subsequent commit. */
+ return 0;
+}
+
+static void eea_free_rx_buffers(struct eea_net_rx *rx, struct eea_net_cfg *cfg)
+{
+ struct eea_rx_meta *meta;
+ u32 i;
+
+ for (i = 0; i < cfg->rx_ring_depth; ++i) {
+ meta = &rx->meta[i];
+ if (!meta->page)
+ continue;
+
+ eea_free_rx_buffer(rx, meta, false);
+ }
+}
+
+static struct page_pool *eea_create_pp(struct eea_net_init_ctx *ctx, u32 idx)
+{
+ struct page_pool_params pp_params = {0};
+
+ pp_params.order = 0;
+ pp_params.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV;
+ pp_params.pool_size = ctx->cfg.rx_ring_depth;
+ pp_params.nid = dev_to_node(ctx->edev->dma_dev);
+ pp_params.dev = ctx->edev->dma_dev;
+ pp_params.netdev = ctx->netdev;
+ pp_params.dma_dir = DMA_FROM_DEVICE;
+ pp_params.max_len = PAGE_SIZE;
+ pp_params.queue_idx = idx;
+
+ return page_pool_create(&pp_params);
+}
+
+static void eea_destroy_page_pool(struct eea_net_rx *rx)
+{
+ if (rx->pp)
+ page_pool_destroy(rx->pp);
+}
+
+void enet_rx_stop(struct eea_net_rx *rx)
+{
+ if (rx->flags & EEA_ENABLE_F_NAPI) {
+ rx->flags &= ~EEA_ENABLE_F_NAPI;
+
+ disable_irq(rx->enet->irq_blks[rx->index].irq);
+ napi_disable(rx->napi);
+
+ page_pool_disable_direct_recycling(rx->pp);
+ netif_napi_del(rx->napi);
+ }
+}
+
+void enet_rx_start(struct eea_net_rx *rx)
+{
+ netif_napi_add(rx->enet->netdev, rx->napi, eea_poll);
+
+ page_pool_enable_direct_recycling(rx->pp, rx->napi);
+
+ napi_enable(rx->napi);
+
+ rx->flags |= EEA_ENABLE_F_NAPI;
+
+ local_bh_disable();
+ napi_schedule(rx->napi);
+ local_bh_enable();
+
+ enable_irq(rx->enet->irq_blks[rx->index].irq);
+}
+
+/* Maybe called before eea_bind_q_and_cfg. So the cfg must be passed. */
+void eea_free_rx(struct eea_net_rx *rx, struct eea_net_cfg *cfg)
+{
+ if (!rx)
+ return;
+
+ if (rx->ering) {
+ eea_ering_free(rx->ering);
+ rx->ering = NULL;
+ }
+
+ if (rx->meta) {
+ eea_free_rx_buffers(rx, cfg);
+ eea_free_rx_hdr(rx, cfg);
+ kvfree(rx->meta);
+ rx->meta = NULL;
+ }
+
+ if (rx->pp) {
+ eea_destroy_page_pool(rx);
+ rx->pp = NULL;
+ }
+
+ kfree(rx);
+}
+
+static void eea_rx_meta_init(struct eea_net_rx *rx, u32 num)
+{
+ struct eea_rx_meta *meta;
+ int i;
+
+ rx->free = NULL;
+
+ for (i = 0; i < num; ++i) {
+ meta = &rx->meta[i];
+ meta->id = i;
+ meta->next = rx->free;
+ rx->free = meta;
+ }
+}
+
+struct eea_net_rx *eea_alloc_rx(struct eea_net_init_ctx *ctx, u32 idx)
+{
+ struct eea_ring *ering;
+ struct eea_net_rx *rx;
+ int err;
+
+ rx = kzalloc(sizeof(*rx), GFP_KERNEL);
+ if (!rx)
+ return rx;
+
+ rx->index = idx;
+ snprintf(rx->name, sizeof(rx->name), "rx.%u", idx);
+
+ /* ering */
+ ering = eea_ering_alloc(idx * 2, ctx->cfg.rx_ring_depth, ctx->edev,
+ ctx->cfg.rx_sq_desc_size,
+ ctx->cfg.rx_cq_desc_size,
+ rx->name);
+ if (!ering)
+ goto err_free_rx;
+
+ rx->ering = ering;
+
+ rx->dma_dev = ctx->edev->dma_dev;
+
+ /* meta */
+ rx->meta = kvcalloc(ctx->cfg.rx_ring_depth,
+ sizeof(*rx->meta), GFP_KERNEL);
+ if (!rx->meta)
+ goto err_free_rx;
+
+ eea_rx_meta_init(rx, ctx->cfg.rx_ring_depth);
+
+ if (ctx->cfg.split_hdr) {
+ err = eea_alloc_rx_hdr(ctx, rx);
+ if (err)
+ goto err_free_rx;
+ }
+
+ rx->pp = eea_create_pp(ctx, idx);
+ if (IS_ERR(rx->pp)) {
+ err = PTR_ERR(rx->pp);
+ rx->pp = NULL;
+ goto err_free_rx;
+ }
+
+ return rx;
+
+err_free_rx:
+ eea_free_rx(rx, &ctx->cfg);
+ return NULL;
+}
diff --git a/drivers/net/ethernet/alibaba/eea/eea_tx.c b/drivers/net/ethernet/alibaba/eea/eea_tx.c
new file mode 100644
index 000000000000..5a8babfce5de
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_tx.c
@@ -0,0 +1,102 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#include <net/netdev_queues.h>
+
+#include "eea_net.h"
+#include "eea_pci.h"
+#include "eea_ring.h"
+
+struct eea_tx_meta {
+ struct eea_tx_meta *next;
+
+ u32 id;
+
+ union {
+ struct sk_buff *skb;
+ void *data;
+ };
+
+ u32 num;
+
+ dma_addr_t dma_addr;
+ struct eea_tx_desc *desc;
+ u32 dma_len;
+};
+
+int eea_poll_tx(struct eea_net_tx *tx, int budget)
+{
+ /* Empty function; will be implemented in a subsequent commit. */
+ return budget;
+}
+
+netdev_tx_t eea_tx_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+ /* Empty function; will be implemented in a subsequent commit. */
+ dev_kfree_skb_any(skb);
+ return NETDEV_TX_OK;
+}
+
+static void eea_free_meta(struct eea_net_tx *tx, struct eea_net_cfg *cfg)
+{
+ kvfree(tx->meta);
+ tx->meta = NULL;
+}
+
+/* Maybe called before eea_bind_q_and_cfg. So the cfg must be passed. */
+void eea_free_tx(struct eea_net_tx *tx, struct eea_net_cfg *cfg)
+{
+ if (!tx)
+ return;
+
+ if (tx->ering) {
+ eea_ering_free(tx->ering);
+ tx->ering = NULL;
+ }
+
+ if (tx->meta)
+ eea_free_meta(tx, cfg);
+}
+
+int eea_alloc_tx(struct eea_net_init_ctx *ctx, struct eea_net_tx *tx, u32 idx)
+{
+ struct eea_tx_meta *meta;
+ struct eea_ring *ering;
+ u32 i;
+
+ snprintf(tx->name, sizeof(tx->name), "tx.%u", idx);
+
+ ering = eea_ering_alloc(idx * 2 + 1, ctx->cfg.tx_ring_depth, ctx->edev,
+ ctx->cfg.tx_sq_desc_size,
+ ctx->cfg.tx_cq_desc_size,
+ tx->name);
+ if (!ering)
+ goto err_free_tx;
+
+ tx->ering = ering;
+ tx->index = idx;
+ tx->dma_dev = ctx->edev->dma_dev;
+
+ /* meta */
+ tx->meta = kvcalloc(ctx->cfg.tx_ring_depth,
+ sizeof(*tx->meta), GFP_KERNEL);
+ if (!tx->meta)
+ goto err_free_tx;
+
+ for (i = 0; i < ctx->cfg.tx_ring_depth; ++i) {
+ meta = &tx->meta[i];
+ meta->id = i;
+ meta->next = tx->free;
+ tx->free = meta;
+ }
+
+ return 0;
+
+err_free_tx:
+ eea_free_tx(tx, &ctx->cfg);
+ return -ENOMEM;
+}
--
2.32.0.3.g01195cf9f
^ permalink raw reply related
* [PATCH net-next v42 8/8] eea: introduce callback for ndo_get_stats64 and register netdev
From: Xuan Zhuo @ 2026-05-04 14:44 UTC (permalink / raw)
To: netdev
Cc: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, Xuan Zhuo, Wen Gu, Philo Lu, Vadim Fedorenko,
Dong Yibo, Mingyu Wang, Heiner Kallweit, Dust Li
In-Reply-To: <20260504144439.29570-1-xuanzhuo@linux.alibaba.com>
This commit adds support for ndo_get_stats64 to provide accurate
interface statistics. With the TX and RX data paths now fully functional,
it is appropriate to register the netdevice and expose the interface to
userspace.
Reviewed-by: Dust Li <dust.li@linux.alibaba.com>
Reviewed-by: Philo Lu <lulie@linux.alibaba.com>
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
---
drivers/net/ethernet/alibaba/eea/eea_net.c | 82 ++++++++++++++++++++--
drivers/net/ethernet/alibaba/eea/eea_net.h | 5 ++
2 files changed, 83 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.c b/drivers/net/ethernet/alibaba/eea/eea_net.c
index c22fdf923130..ade079128b05 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_net.c
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.c
@@ -112,6 +112,11 @@ static void eea_bind_q_and_cfg(struct eea_net *enet,
struct eea_net_tx *tx;
int i;
+ /* Since 'ndo_get_stats64' is not called in softirq context, there is no
+ * need to use 'spin_lock_bh'.
+ */
+ spin_lock(&enet->stats_lock);
+
enet->cfg = ctx->cfg;
enet->rx = ctx->rx;
enet->tx = ctx->tx;
@@ -131,6 +136,8 @@ static void eea_bind_q_and_cfg(struct eea_net *enet,
blk->rx = rx;
}
+
+ spin_unlock(&enet->stats_lock);
}
static void eea_unbind_q_and_cfg(struct eea_net *enet,
@@ -140,6 +147,8 @@ static void eea_unbind_q_and_cfg(struct eea_net *enet,
struct eea_net_rx *rx;
int i;
+ spin_lock(&enet->stats_lock);
+
ctx->cfg = enet->cfg;
ctx->rx = enet->rx;
ctx->tx = enet->tx;
@@ -156,6 +165,8 @@ static void eea_unbind_q_and_cfg(struct eea_net *enet,
blk->rx = NULL;
}
+
+ spin_unlock(&enet->stats_lock);
}
static void eea_free_rxtx_q_mem(struct eea_net_init_ctx *ctx)
@@ -340,6 +351,59 @@ static int eea_netdev_open(struct net_device *netdev)
return err;
}
+/* Statistics may be reset to zero upon device reset. This is expected behavior
+ * for now and will be addressed in the future.
+ */
+static void eea_stats(struct net_device *netdev, struct rtnl_link_stats64 *tot)
+{
+ struct eea_net *enet = netdev_priv(netdev);
+ u64 packets, bytes, drop, lerr;
+ u32 start;
+ int i;
+
+ spin_lock(&enet->stats_lock);
+
+ if (enet->rx) {
+ for (i = 0; i < enet->cfg.rx_ring_num; i++) {
+ struct eea_net_rx *rx = enet->rx[i];
+
+ do {
+ start = u64_stats_fetch_begin(&rx->stats.syncp);
+ packets = u64_stats_read(&rx->stats.packets);
+ bytes = u64_stats_read(&rx->stats.bytes);
+ drop = u64_stats_read(&rx->stats.drops);
+ lerr = u64_stats_read(&rx->stats.length_errors);
+ } while (u64_stats_fetch_retry(&rx->stats.syncp,
+ start));
+
+ tot->rx_packets += packets;
+ tot->rx_bytes += bytes;
+ tot->rx_dropped += drop;
+ tot->rx_length_errors += lerr;
+ }
+ }
+
+ if (enet->tx) {
+ for (i = 0; i < enet->cfg.tx_ring_num; i++) {
+ struct eea_net_tx *tx = &enet->tx[i];
+
+ do {
+ start = u64_stats_fetch_begin(&tx->stats.syncp);
+ packets = u64_stats_read(&tx->stats.packets);
+ bytes = u64_stats_read(&tx->stats.bytes);
+ drop = u64_stats_read(&tx->stats.drops);
+ } while (u64_stats_fetch_retry(&tx->stats.syncp,
+ start));
+
+ tot->tx_packets += packets;
+ tot->tx_bytes += bytes;
+ tot->tx_dropped += drop;
+ }
+ }
+
+ spin_unlock(&enet->stats_lock);
+}
+
/* resources: ring, buffers, irq */
int eea_reset_hw_resources(struct eea_net *enet, struct eea_net_init_ctx *ctx)
{
@@ -347,7 +411,9 @@ int eea_reset_hw_resources(struct eea_net *enet, struct eea_net_init_ctx *ctx)
int err, error;
if (!netif_running(enet->netdev) || !enet->started) {
+ spin_lock(&enet->stats_lock);
enet->cfg = ctx->cfg;
+ spin_unlock(&enet->stats_lock);
return 0;
}
@@ -606,6 +672,7 @@ static const struct net_device_ops eea_netdev = {
.ndo_stop = eea_netdev_stop,
.ndo_start_xmit = eea_tx_xmit,
.ndo_validate_addr = eth_validate_addr,
+ .ndo_get_stats64 = eea_stats,
.ndo_features_check = passthru_features_check,
};
@@ -639,6 +706,8 @@ static struct eea_net *eea_netdev_alloc(struct eea_device *edev, u32 pairs)
return NULL;
}
+ spin_lock_init(&enet->stats_lock);
+
return enet;
}
@@ -724,11 +793,13 @@ int eea_net_probe(struct eea_device *edev)
eea_update_ts_off(edev, enet);
- netdev_dbg(enet->netdev, "eea probe success.\n");
+ netif_carrier_off(enet->netdev);
- /* Queue TX/RX implementation is still in progress. register_netdev is
- * deferred until these are completed in subsequent commits.
- */
+ err = register_netdev(enet->netdev);
+ if (err)
+ goto err_reset_dev;
+
+ netdev_dbg(enet->netdev, "eea probe success.\n");
return 0;
@@ -780,6 +851,8 @@ void eea_net_remove(struct eea_device *edev, bool ha)
return;
}
+ unregister_netdev(netdev);
+
if (!enet->wait_pci_ready) {
eea_device_reset(edev);
eea_destroy_adminq(enet);
@@ -800,6 +873,7 @@ void eea_net_shutdown(struct eea_device *edev)
rtnl_lock();
netif_device_detach(netdev);
+ dev_close(netdev);
if (!enet->wait_pci_ready) {
eea_device_reset(edev);
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.h b/drivers/net/ethernet/alibaba/eea/eea_net.h
index 6a5b2f908c76..848bb90bccf8 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_net.h
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.h
@@ -165,6 +165,11 @@ struct eea_net {
u32 speed;
u64 hw_ts_offset;
+
+ /* Protect the tx and rx of struct eea_net, when eea_stats accesses the
+ * stats from rx and tx queues.
+ */
+ spinlock_t stats_lock;
};
int eea_net_probe(struct eea_device *edev);
--
2.32.0.3.g01195cf9f
^ permalink raw reply related
* [PATCH net-next v42 3/8] eea: probe the netdevice and create adminq
From: Xuan Zhuo @ 2026-05-04 14:44 UTC (permalink / raw)
To: netdev
Cc: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
Paolo Abeni, Xuan Zhuo, Wen Gu, Philo Lu, Vadim Fedorenko,
Dong Yibo, Mingyu Wang, Heiner Kallweit, Dust Li
In-Reply-To: <20260504144439.29570-1-xuanzhuo@linux.alibaba.com>
Add basic driver framework for the Alibaba Elastic Ethernet Adapter(EEA).
This commit creates the netdevice after PCI probe,
and initializes the admin queue to send commands to the device.
Reviewed-by: Dust Li <dust.li@linux.alibaba.com>
Reviewed-by: Philo Lu <lulie@linux.alibaba.com>
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
---
drivers/net/ethernet/alibaba/eea/Makefile | 6 +-
drivers/net/ethernet/alibaba/eea/eea_adminq.c | 518 ++++++++++++++++++
drivers/net/ethernet/alibaba/eea/eea_adminq.h | 83 +++
drivers/net/ethernet/alibaba/eea/eea_net.c | 249 +++++++++
drivers/net/ethernet/alibaba/eea/eea_net.h | 137 +++++
drivers/net/ethernet/alibaba/eea/eea_pci.c | 35 +-
drivers/net/ethernet/alibaba/eea/eea_pci.h | 3 +
7 files changed, 1023 insertions(+), 8 deletions(-)
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_adminq.c
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_adminq.h
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_net.c
create mode 100644 drivers/net/ethernet/alibaba/eea/eea_net.h
diff --git a/drivers/net/ethernet/alibaba/eea/Makefile b/drivers/net/ethernet/alibaba/eea/Makefile
index 7d8e7e8c2f3a..a842ac416ae8 100644
--- a/drivers/net/ethernet/alibaba/eea/Makefile
+++ b/drivers/net/ethernet/alibaba/eea/Makefile
@@ -1,4 +1,6 @@
obj-$(CONFIG_ALIBABA_EEA) += eea.o
-eea-y := eea_ring.o \
- eea_pci.o
+eea-y := eea_ring.o \
+ eea_net.o \
+ eea_pci.o \
+ eea_adminq.o
diff --git a/drivers/net/ethernet/alibaba/eea/eea_adminq.c b/drivers/net/ethernet/alibaba/eea/eea_adminq.c
new file mode 100644
index 000000000000..c73a12757d45
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_adminq.c
@@ -0,0 +1,518 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#include <linux/etherdevice.h>
+#include <linux/iopoll.h>
+#include <linux/utsname.h>
+#include <linux/version.h>
+
+#include "eea_adminq.h"
+#include "eea_net.h"
+#include "eea_pci.h"
+#include "eea_ring.h"
+
+#define EEA_AQ_CMD_CFG_QUERY ((0 << 8) | 0)
+
+#define EEA_AQ_CMD_QUEUE_CREATE ((1 << 8) | 0)
+#define EEA_AQ_CMD_QUEUE_DESTROY_ALL ((1 << 8) | 1)
+
+#define EEA_AQ_CMD_HOST_INFO ((2 << 8) | 0)
+
+#define EEA_AQ_CMD_DEV_STATUS ((3 << 8) | 0)
+
+#define EEA_RING_DESC_F_AQ_PHASE (BIT(15) | BIT(7))
+
+#define EEA_QUEUE_FLAGS_HW_SPLIT_HDR BIT(0)
+#define EEA_QUEUE_FLAGS_SQCQ BIT(1)
+#define EEA_QUEUE_FLAGS_HWTS BIT(2)
+
+struct eea_aq_create {
+ __le32 flags;
+ /* queue index.
+ * rx: 0 == qidx % 2
+ * tx: 1 == qidx % 2
+ */
+ __le16 qidx;
+ /* the depth of the queue */
+ __le16 depth;
+ /* 0: without SPLIT HDR
+ * 1: 128B
+ * 2: 256B
+ * 3: 512B
+ */
+ u8 hdr_buf_size;
+ u8 sq_desc_size;
+ u8 cq_desc_size;
+ u8 reserve0;
+ /* The vector for the irq. rx,tx share the same vector */
+ __le16 msix_vector;
+ __le16 reserve;
+ /* sq ring cfg. */
+ __le32 sq_addr_low;
+ __le32 sq_addr_high;
+ /* cq ring cfg. Just valid when flags include EEA_QUEUE_FLAGS_SQCQ. */
+ __le32 cq_addr_low;
+ __le32 cq_addr_high;
+};
+
+struct eea_aq_queue_drv_status {
+ __le16 qidx;
+
+ __le16 sq_head;
+ __le16 cq_head;
+ __le16 reserved;
+};
+
+#define EEA_OS_DISTRO 0
+#define EEA_DRV_TYPE 0
+#define EEA_OS_LINUX 1
+#define EEA_SPEC_VER_MAJOR 1
+#define EEA_SPEC_VER_MINOR 0
+
+struct eea_aq_host_info_cfg {
+ __le16 os_type;
+ __le16 os_dist;
+ __le16 drv_type;
+
+ __le16 kern_ver_major;
+ __le16 kern_ver_minor;
+ __le16 kern_ver_sub_minor;
+
+ __le16 drv_ver_major;
+ __le16 drv_ver_minor;
+ __le16 drv_ver_sub_minor;
+
+ __le16 spec_ver_major;
+ __le16 spec_ver_minor;
+ __le16 pci_bdf;
+ __le32 pci_domain;
+
+ u8 os_ver_str[64];
+ u8 isa_str[64];
+};
+
+#define EEA_HINFO_MAX_REP_LEN 1024
+#define EEA_HINFO_REP_BAD 2
+
+struct eea_aq_host_info_rep {
+ u8 op_code;
+ u8 has_reply;
+ u8 reply_str[EEA_HINFO_MAX_REP_LEN];
+};
+
+static struct eea_ring *qid_to_ering(struct eea_net *enet, u32 qid)
+{
+ struct eea_ring *ering;
+
+ if (qid % 2 == 0)
+ ering = enet->rx[qid / 2]->ering;
+ else
+ ering = enet->tx[qid / 2].ering;
+
+ return ering;
+}
+
+#define EEA_AQ_TIMEOUT_US (60 * 1000 * 1000)
+
+static int eea_adminq_submit(struct eea_net *enet, u16 cmd,
+ dma_addr_t req_addr, dma_addr_t res_addr,
+ u32 req_size, u32 res_size, u32 *reply_len)
+{
+ struct eea_aq_cdesc *cdesc;
+ struct eea_aq_desc *desc;
+ int ret;
+
+ desc = eea_ering_aq_alloc_desc(enet->adminq.ring);
+
+ desc->classid = cmd >> 8;
+ desc->command = cmd & 0xff;
+
+ desc->data_addr = cpu_to_le64(req_addr);
+ desc->data_len = cpu_to_le32(req_size);
+
+ desc->reply_addr = cpu_to_le64(res_addr);
+ desc->reply_len = cpu_to_le32(res_size);
+
+ /* for update flags */
+ dma_wmb();
+
+ desc->flags = cpu_to_le16(enet->adminq.phase);
+
+ eea_ering_sq_commit_desc(enet->adminq.ring);
+
+ eea_ering_kick(enet->adminq.ring);
+
+ ++enet->adminq.num;
+
+ if ((enet->adminq.num % enet->adminq.ring->num) == 0)
+ enet->adminq.phase ^= EEA_RING_DESC_F_AQ_PHASE;
+
+ ret = read_poll_timeout(eea_ering_cq_get_desc, cdesc, cdesc, 10,
+ EEA_AQ_TIMEOUT_US, false, enet->adminq.ring);
+ if (ret) {
+ netdev_err(enet->netdev,
+ "adminq exec timeout. cmd: %d reset device.\n",
+ cmd);
+ /* The device must be reset before unmapping buffers to avoid
+ * potential DMA writes after the memory is freed.
+ */
+ eea_device_reset(enet->edev);
+ enet->adminq.broken = true;
+ return ret;
+ }
+
+ /* Returns 0 on success, or a negative error code on failure. */
+ ret = le32_to_cpu(cdesc->status);
+
+ eea_ering_cq_ack_desc(enet->adminq.ring, 1);
+
+ if (ret)
+ netdev_err(enet->netdev,
+ "adminq exec failed. cmd: %d ret %d\n", cmd, ret);
+ else
+ *reply_len = le32_to_cpu(cdesc->reply_len);
+
+ return ret;
+}
+
+static int eea_adminq_exec(struct eea_net *enet, u16 cmd,
+ void *req, u32 req_size,
+ void *res, u32 res_size,
+ u32 *reply)
+{
+ dma_addr_t req_addr = 0, res_addr = 0;
+ struct device *dma;
+ u32 reply_len = 0;
+ int ret;
+
+ if (enet->adminq.broken)
+ return -EIO;
+
+ if (reply)
+ *reply = 0;
+
+ dma = enet->edev->dma_dev;
+
+ if (req) {
+ req_addr = dma_map_single(dma, req, req_size, DMA_TO_DEVICE);
+ if (unlikely(dma_mapping_error(dma, req_addr)))
+ return -ENOMEM;
+ }
+
+ if (res) {
+ res_addr = dma_map_single(dma, res, res_size, DMA_FROM_DEVICE);
+ if (unlikely(dma_mapping_error(dma, res_addr))) {
+ ret = -ENOMEM;
+ goto err_unmap_req;
+ }
+ }
+
+ mutex_lock(&enet->adminq.lock);
+ ret = eea_adminq_submit(enet, cmd, req_addr, res_addr,
+ req_size, res_size, &reply_len);
+ mutex_unlock(&enet->adminq.lock);
+ if (res) {
+ dma_unmap_single(dma, res_addr, res_size, DMA_FROM_DEVICE);
+
+ if (ret)
+ memset(res, 0, res_size);
+ else if (res_size > reply_len)
+ memset(res + reply_len, 0, res_size - reply_len);
+
+ if (reply)
+ *reply = reply_len;
+ }
+
+err_unmap_req:
+ if (req)
+ dma_unmap_single(dma, req_addr, req_size, DMA_TO_DEVICE);
+
+ return ret;
+}
+
+void eea_destroy_adminq(struct eea_net *enet)
+{
+ struct eea_aq *aq;
+
+ aq = &enet->adminq;
+
+ if (aq->ring) {
+ eea_ering_free(aq->ring);
+ aq->ring = NULL;
+ aq->phase = 0;
+ }
+
+ kfree(aq->q_req_buf);
+ kfree(aq->q_res_buf);
+
+ aq->q_req_buf = NULL;
+ aq->q_res_buf = NULL;
+}
+
+int eea_create_adminq(struct eea_net *enet, u32 qid)
+{
+ u32 db_size, q_size, num;
+ struct eea_ring *ering;
+ struct eea_aq *aq;
+ int err = -ENOMEM;
+
+ num = enet->edev->rx_num + enet->edev->tx_num;
+ aq = &enet->adminq;
+
+ ering = eea_ering_alloc(qid, 64, enet->edev, sizeof(struct eea_aq_desc),
+ sizeof(struct eea_aq_cdesc), "adminq");
+ if (!ering)
+ return -ENOMEM;
+
+ aq->ring = ering;
+
+ err = eea_pci_active_aq(ering, qid / 2 + 1);
+ if (err)
+ goto err;
+
+ aq->phase = BIT(7);
+ aq->num = 0;
+
+ q_size = sizeof(*aq->q_req_buf) * num;
+ db_size = sizeof(*aq->q_res_buf) * num;
+
+ aq->q_req_size = q_size;
+ aq->q_res_size = db_size;
+
+ err = -ENOMEM;
+
+ aq->q_req_buf = kzalloc(q_size, GFP_KERNEL);
+ if (!aq->q_req_buf)
+ goto err;
+
+ aq->q_res_buf = kzalloc(db_size, GFP_KERNEL);
+ if (!aq->q_res_buf)
+ goto err;
+
+ /* Before we set up the AQ, the device remains in an inactive state, so
+ * there will be no DMA operations. If the 'set up AQ' process fails, we
+ * can safely free the DMA-related memory.
+ */
+ err = eea_pci_set_aq_up(enet->edev);
+ if (err)
+ goto err;
+
+ aq->broken = false;
+
+ mutex_init(&aq->lock);
+
+ return 0;
+
+err:
+ eea_destroy_adminq(enet);
+ return err;
+}
+
+int eea_adminq_query_cfg(struct eea_net *enet, struct eea_aq_cfg *cfg)
+{
+ return eea_adminq_exec(enet, EEA_AQ_CMD_CFG_QUERY, NULL, 0, cfg,
+ sizeof(*cfg), NULL);
+}
+
+static void qcfg_fill(struct eea_aq_create *qcfg, struct eea_ring *ering,
+ u32 flags)
+{
+ qcfg->flags = cpu_to_le32(flags);
+ qcfg->qidx = cpu_to_le16(ering->index);
+ qcfg->depth = cpu_to_le16(ering->num);
+
+ qcfg->hdr_buf_size = flags & EEA_QUEUE_FLAGS_HW_SPLIT_HDR ? 1 : 0;
+ qcfg->sq_desc_size = ering->sq.desc_size;
+ qcfg->cq_desc_size = ering->cq.desc_size;
+ qcfg->msix_vector = cpu_to_le16(ering->msix_vec);
+
+ qcfg->sq_addr_low = cpu_to_le32(lower_32_bits(ering->sq.dma_addr));
+ qcfg->sq_addr_high = cpu_to_le32(upper_32_bits(ering->sq.dma_addr));
+
+ qcfg->cq_addr_low = cpu_to_le32(lower_32_bits(ering->cq.dma_addr));
+ qcfg->cq_addr_high = cpu_to_le32(upper_32_bits(ering->cq.dma_addr));
+}
+
+int eea_adminq_create_q(struct eea_net *enet, u32 num, u32 flags)
+{
+ int i, db_size, q_size, err = -ENOMEM;
+ struct eea_net_cfg *cfg;
+ struct eea_ring *ering;
+ struct eea_aq *aq;
+ u32 reply_len;
+
+ cfg = &enet->cfg;
+ aq = &enet->adminq;
+
+ if (cfg->split_hdr)
+ flags |= EEA_QUEUE_FLAGS_HW_SPLIT_HDR;
+
+ flags |= EEA_QUEUE_FLAGS_SQCQ;
+ flags |= EEA_QUEUE_FLAGS_HWTS;
+
+ q_size = sizeof(*aq->q_req_buf) * num;
+ db_size = sizeof(*aq->q_res_buf) * num;
+
+ for (i = 0; i < num; i++) {
+ ering = qid_to_ering(enet, i);
+ qcfg_fill(aq->q_req_buf + i, ering, flags);
+ }
+
+ err = eea_adminq_exec(enet, EEA_AQ_CMD_QUEUE_CREATE,
+ aq->q_req_buf, q_size,
+ aq->q_res_buf, db_size,
+ &reply_len);
+ if (err)
+ return err;
+
+ if (reply_len != db_size) {
+ eea_adminq_destroy_all_q(enet);
+ netdev_err(enet->netdev, "invalid reply len %u\n", reply_len);
+ return -EINVAL;
+ }
+
+ for (i = 0; i < num; i++) {
+ ering = qid_to_ering(enet, i);
+ ering->db = eea_pci_db_addr(ering->edev,
+ le32_to_cpu(aq->q_res_buf[i]));
+ if (!ering->db) {
+ netdev_err(enet->netdev, "invalid db off %u\n",
+ le32_to_cpu(aq->q_res_buf[i]));
+ goto err;
+ }
+ }
+
+ return err;
+
+err:
+ eea_adminq_destroy_all_q(enet);
+ for (i = 0; i < num; i++) {
+ ering = qid_to_ering(enet, i);
+ ering->db = NULL;
+ }
+
+ return -EIO;
+}
+
+int eea_adminq_destroy_all_q(struct eea_net *enet)
+{
+ return eea_adminq_exec(enet, EEA_AQ_CMD_QUEUE_DESTROY_ALL, NULL, 0,
+ NULL, 0, NULL);
+}
+
+/* The caller must ensure that both the 'rx' and 'tx' arrays are valid. */
+int eea_adminq_dev_status(struct eea_net *enet,
+ struct eea_aq_dev_status *dstatus)
+{
+ struct eea_aq_queue_drv_status *drv_status;
+ struct __eea_aq_dev_status *dev_status;
+ int err, i, io_num, size, q_num;
+ struct eea_ring *ering;
+ void *rep, *req;
+
+ q_num = enet->cfg.rx_ring_num + enet->cfg.tx_ring_num + 1;
+ io_num = enet->cfg.rx_ring_num + enet->cfg.tx_ring_num;
+
+ req = kcalloc(q_num, sizeof(struct eea_aq_queue_drv_status),
+ GFP_KERNEL);
+ if (!req)
+ return -ENOMEM;
+
+ size = struct_size(dev_status, q_status, q_num);
+
+ rep = kzalloc(size, GFP_KERNEL);
+ if (!rep) {
+ kfree(req);
+ return -ENOMEM;
+ }
+
+ drv_status = req;
+ for (i = 0; i < io_num; ++i, ++drv_status) {
+ ering = qid_to_ering(enet, i);
+ drv_status->qidx = cpu_to_le16(i);
+ drv_status->cq_head = cpu_to_le16(ering->cq.head);
+ drv_status->sq_head = cpu_to_le16(ering->sq.head);
+ }
+
+ drv_status->qidx = cpu_to_le16(i);
+ drv_status->cq_head = cpu_to_le16(enet->adminq.ring->cq.head);
+ drv_status->sq_head = cpu_to_le16(enet->adminq.ring->sq.head);
+
+ err = eea_adminq_exec(enet, EEA_AQ_CMD_DEV_STATUS, req,
+ q_num * sizeof(struct eea_aq_queue_drv_status),
+ rep, size, NULL);
+ kfree(req);
+ if (err) {
+ kfree(rep);
+ return err;
+ }
+
+ dstatus->num = q_num;
+ dstatus->status = rep;
+
+ return 0;
+}
+
+void eea_adminq_config_host_info(struct eea_net *enet)
+{
+ struct device *dev = enet->edev->dma_dev;
+ struct eea_aq_host_info_cfg *cfg;
+ struct eea_aq_host_info_rep *rep;
+ int rc = -ENOMEM;
+
+ cfg = kzalloc(sizeof(*cfg), GFP_KERNEL);
+ if (!cfg)
+ return;
+
+ rep = kzalloc(sizeof(*rep), GFP_KERNEL);
+ if (!rep)
+ goto err_free_cfg;
+
+ cfg->os_type = cpu_to_le16(EEA_OS_LINUX);
+ cfg->os_dist = cpu_to_le16(EEA_OS_DISTRO);
+ cfg->drv_type = cpu_to_le16(EEA_DRV_TYPE);
+
+ cfg->kern_ver_major = cpu_to_le16(LINUX_VERSION_MAJOR);
+ cfg->kern_ver_minor = cpu_to_le16(LINUX_VERSION_PATCHLEVEL);
+ cfg->kern_ver_sub_minor = cpu_to_le16(LINUX_VERSION_SUBLEVEL);
+
+ cfg->drv_ver_major = cpu_to_le16(EEA_VER_MAJOR);
+ cfg->drv_ver_minor = cpu_to_le16(EEA_VER_MINOR);
+ cfg->drv_ver_sub_minor = cpu_to_le16(EEA_VER_SUB_MINOR);
+
+ cfg->spec_ver_major = cpu_to_le16(EEA_SPEC_VER_MAJOR);
+ cfg->spec_ver_minor = cpu_to_le16(EEA_SPEC_VER_MINOR);
+
+ cfg->pci_bdf = cpu_to_le16(eea_pci_bdf(enet->edev));
+ cfg->pci_domain = cpu_to_le32(eea_pci_domain_nr(enet->edev));
+
+ strscpy(cfg->os_ver_str, utsname()->release, sizeof(cfg->os_ver_str));
+ strscpy(cfg->isa_str, utsname()->machine, sizeof(cfg->isa_str));
+
+ rc = eea_adminq_exec(enet, EEA_AQ_CMD_HOST_INFO,
+ cfg, sizeof(*cfg), rep, sizeof(*rep), NULL);
+
+ if (!rc) {
+ if (rep->op_code == EEA_HINFO_REP_BAD)
+ dev_warn(dev, "The hardware-driven state validation may be abnormal.\n");
+
+ if (rep->has_reply) {
+ char buf[EEA_HINFO_MAX_REP_LEN];
+
+ rep->reply_str[EEA_HINFO_MAX_REP_LEN - 1] = '\0';
+
+ string_escape_str(rep->reply_str, buf, sizeof(buf),
+ ESCAPE_NP, NULL);
+
+ dev_warn(dev, "Device replied: %s\n", buf);
+ }
+ }
+
+ kfree(rep);
+err_free_cfg:
+ kfree(cfg);
+}
diff --git a/drivers/net/ethernet/alibaba/eea/eea_adminq.h b/drivers/net/ethernet/alibaba/eea/eea_adminq.h
new file mode 100644
index 000000000000..0182f5641fcf
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_adminq.h
@@ -0,0 +1,83 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#ifndef __EEA_ADMINQ_H__
+#define __EEA_ADMINQ_H__
+
+struct eea_aq_cfg {
+ __le32 rx_depth_max;
+ __le32 rx_depth_def;
+
+ __le32 tx_depth_max;
+ __le32 tx_depth_def;
+
+ __le32 max_tso_size;
+ __le32 max_tso_segs;
+
+ u8 mac[ETH_ALEN];
+ __le16 status;
+
+ __le16 mtu;
+ __le16 reserved0;
+ __le16 reserved1;
+ u8 reserved2;
+ u8 reserved3;
+
+ __le16 reserved4;
+ __le16 reserved5;
+ __le16 reserved6;
+};
+
+struct eea_aq_queue_status {
+ __le16 qidx;
+#define EEA_QUEUE_STATUS_OK 0
+#define EEA_QUEUE_STATUS_NEED_RESET 1
+ __le16 status;
+};
+
+struct __eea_aq_dev_status {
+#define EEA_LINK_DOWN_STATUS 0
+#define EEA_LINK_UP_STATUS 1
+ __le16 link_status;
+ __le16 reserved;
+
+ struct eea_aq_queue_status q_status[];
+};
+
+struct eea_aq_dev_status {
+ u32 num;
+ struct __eea_aq_dev_status *status;
+};
+
+struct eea_aq {
+ struct eea_ring *ring;
+ u32 num;
+ bool broken;
+ u16 phase;
+
+ /* lock for adminq exec */
+ struct mutex lock;
+
+ u32 q_req_size;
+ u32 q_res_size;
+ struct eea_aq_create *q_req_buf;
+ __le32 *q_res_buf;
+};
+
+struct eea_net;
+
+int eea_create_adminq(struct eea_net *enet, u32 qid);
+void eea_destroy_adminq(struct eea_net *enet);
+
+int eea_adminq_query_cfg(struct eea_net *enet, struct eea_aq_cfg *cfg);
+
+int eea_adminq_create_q(struct eea_net *enet, u32 num, u32 flags);
+int eea_adminq_destroy_all_q(struct eea_net *enet);
+int eea_adminq_dev_status(struct eea_net *enet,
+ struct eea_aq_dev_status *dstatus);
+void eea_adminq_config_host_info(struct eea_net *enet);
+#endif
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.c b/drivers/net/ethernet/alibaba/eea/eea_net.c
new file mode 100644
index 000000000000..bb8a49f8c6df
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.c
@@ -0,0 +1,249 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#include <linux/etherdevice.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/rtnetlink.h>
+#include <net/netdev_queues.h>
+
+#include "eea_adminq.h"
+#include "eea_net.h"
+#include "eea_pci.h"
+#include "eea_ring.h"
+
+#define EEA_SPLIT_HDR_SIZE ALIGN(128, L1_CACHE_BYTES)
+
+static int eea_update_cfg(struct eea_net *enet,
+ struct eea_device *edev,
+ struct eea_aq_cfg *hwcfg)
+{
+ u32 rx_max = le32_to_cpu(hwcfg->rx_depth_max);
+ u32 tx_max = le32_to_cpu(hwcfg->tx_depth_max);
+ u32 rx_def = le32_to_cpu(hwcfg->rx_depth_def);
+ u32 tx_def = le32_to_cpu(hwcfg->tx_depth_def);
+
+ /* Now, we assert that the rx ring num is equal to the tx ring num. */
+ if (edev->rx_num != edev->tx_num) {
+ dev_err(edev->dma_dev, "Inconsistent ring num: RX %u, TX %u\n",
+ edev->rx_num, edev->tx_num);
+ return -EINVAL;
+ }
+
+ if (rx_max > EEA_NET_IO_HW_RING_DEPTH_MAX ||
+ rx_max < EEA_NET_IO_HW_RING_DEPTH_MIN ||
+ tx_max > EEA_NET_IO_HW_RING_DEPTH_MAX ||
+ tx_max < EEA_NET_IO_HW_RING_DEPTH_MIN) {
+ dev_err(edev->dma_dev, "Invalid HW max depth: RX %u, TX %u\n",
+ rx_max, tx_max);
+ return -EINVAL;
+ }
+
+ if (rx_def > rx_max ||
+ tx_def > tx_max ||
+ rx_def < EEA_NET_IO_HW_RING_DEPTH_MIN ||
+ tx_def < EEA_NET_IO_HW_RING_DEPTH_MIN) {
+ dev_err(edev->dma_dev, "Invalid default depth: RX %u (max %u), TX %u (max %u)\n",
+ rx_def, rx_max, tx_def, tx_max);
+ return -EINVAL;
+ }
+
+ if (!is_power_of_2(rx_max) || !is_power_of_2(tx_max) ||
+ !is_power_of_2(rx_def) || !is_power_of_2(tx_def)) {
+ dev_err(edev->dma_dev, "Ring depth must be power of 2\n");
+ return -EINVAL;
+ }
+
+ enet->cfg_hw.rx_ring_depth = rx_max;
+ enet->cfg_hw.tx_ring_depth = tx_max;
+ enet->cfg_hw.rx_ring_num = edev->rx_num;
+ enet->cfg_hw.tx_ring_num = edev->tx_num;
+ enet->cfg_hw.split_hdr = EEA_SPLIT_HDR_SIZE;
+
+ enet->cfg.rx_ring_depth = rx_def;
+ enet->cfg.tx_ring_depth = tx_def;
+ enet->cfg.rx_ring_num = edev->rx_num;
+ enet->cfg.tx_ring_num = edev->tx_num;
+
+ return 0;
+}
+
+static int eea_netdev_init_features(struct net_device *netdev,
+ struct eea_net *enet,
+ struct eea_device *edev)
+{
+ struct eea_aq_cfg *cfg;
+ int err;
+ u32 mtu;
+
+ cfg = kzalloc(sizeof(*cfg), GFP_KERNEL);
+ if (!cfg)
+ return -ENOMEM;
+
+ err = eea_adminq_query_cfg(enet, cfg);
+ if (err)
+ goto err_free;
+
+ mtu = le16_to_cpu(cfg->mtu);
+ if (mtu < ETH_MIN_MTU) {
+ dev_err(edev->dma_dev, "The device gave us an invalid MTU. Here we can only exit the initialization. %u < %u\n",
+ mtu, ETH_MIN_MTU);
+ err = -EINVAL;
+ goto err_free;
+ }
+
+ err = eea_update_cfg(enet, edev, cfg);
+ if (err)
+ goto err_free;
+
+ netdev->priv_flags |= IFF_UNICAST_FLT;
+ netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
+
+ netdev->hw_features |= NETIF_F_HW_CSUM;
+ netdev->hw_features |= NETIF_F_GRO_HW;
+ netdev->hw_features |= NETIF_F_SG;
+ netdev->hw_features |= NETIF_F_TSO;
+ netdev->hw_features |= NETIF_F_TSO_ECN;
+ netdev->hw_features |= NETIF_F_TSO6;
+ netdev->hw_features |= NETIF_F_GSO_UDP_L4;
+
+ netdev->features |= NETIF_F_HIGHDMA;
+ netdev->features |= NETIF_F_HW_CSUM;
+ netdev->features |= NETIF_F_SG;
+ netdev->features |= NETIF_F_GSO_ROBUST;
+ netdev->features |= netdev->hw_features & NETIF_F_ALL_TSO;
+ netdev->features |= NETIF_F_RXCSUM;
+ netdev->features |= NETIF_F_GRO_HW;
+
+ netdev->vlan_features = netdev->features;
+
+ if (!is_valid_ether_addr(cfg->mac)) {
+ dev_err(edev->dma_dev, "The device gave invalid mac %pM\n",
+ cfg->mac);
+ err = -EINVAL;
+ goto err_free;
+ }
+
+ eth_hw_addr_set(netdev, cfg->mac);
+
+ enet->speed = SPEED_UNKNOWN;
+ enet->duplex = DUPLEX_UNKNOWN;
+
+ netdev->min_mtu = ETH_MIN_MTU;
+
+ netdev->mtu = mtu;
+
+ /* If jumbo frames are already enabled, then the returned MTU will be a
+ * jumbo MTU, and the driver will automatically enable jumbo frame
+ * support by default.
+ */
+ netdev->max_mtu = mtu;
+
+err_free:
+ kfree(cfg);
+ return err;
+}
+
+static const struct net_device_ops eea_netdev = {
+ .ndo_validate_addr = eth_validate_addr,
+ .ndo_features_check = passthru_features_check,
+};
+
+static struct eea_net *eea_netdev_alloc(struct eea_device *edev, u32 pairs)
+{
+ struct net_device *netdev;
+ struct eea_net *enet;
+
+ netdev = alloc_etherdev_mq(sizeof(struct eea_net), pairs);
+ if (!netdev) {
+ dev_err(edev->dma_dev,
+ "alloc_etherdev_mq failed with pairs %d\n", pairs);
+ return NULL;
+ }
+
+ netdev->netdev_ops = &eea_netdev;
+ SET_NETDEV_DEV(netdev, edev->dma_dev);
+
+ enet = netdev_priv(netdev);
+ enet->netdev = netdev;
+ enet->edev = edev;
+ edev->enet = enet;
+
+ return enet;
+}
+
+int eea_net_probe(struct eea_device *edev)
+{
+ struct eea_net *enet;
+ int err = -ENOMEM;
+
+ enet = eea_netdev_alloc(edev, edev->rx_num);
+ if (!enet)
+ return -ENOMEM;
+
+ err = eea_create_adminq(enet, edev->rx_num + edev->tx_num);
+ if (err)
+ goto err_free_netdev;
+
+ eea_adminq_config_host_info(enet);
+
+ err = eea_netdev_init_features(enet->netdev, enet, edev);
+ if (err)
+ goto err_reset_dev;
+
+ netdev_dbg(enet->netdev, "eea probe success.\n");
+
+ /* Queue TX/RX implementation is still in progress. register_netdev is
+ * deferred until these are completed in subsequent commits.
+ */
+
+ return 0;
+
+err_reset_dev:
+ eea_device_reset(edev);
+ eea_destroy_adminq(enet);
+
+err_free_netdev:
+ free_netdev(enet->netdev);
+ return err;
+}
+
+void eea_net_remove(struct eea_device *edev)
+{
+ struct net_device *netdev;
+ struct eea_net *enet;
+
+ enet = edev->enet;
+ netdev = enet->netdev;
+
+ netdev_dbg(enet->netdev, "eea removed.\n");
+
+ eea_device_reset(edev);
+
+ eea_destroy_adminq(enet);
+
+ free_netdev(netdev);
+}
+
+void eea_net_shutdown(struct eea_device *edev)
+{
+ struct net_device *netdev;
+ struct eea_net *enet;
+
+ enet = edev->enet;
+ netdev = enet->netdev;
+
+ rtnl_lock();
+
+ netif_device_detach(netdev);
+
+ eea_device_reset(edev);
+
+ eea_destroy_adminq(enet);
+
+ rtnl_unlock();
+}
diff --git a/drivers/net/ethernet/alibaba/eea/eea_net.h b/drivers/net/ethernet/alibaba/eea/eea_net.h
new file mode 100644
index 000000000000..fa0eec8af21b
--- /dev/null
+++ b/drivers/net/ethernet/alibaba/eea/eea_net.h
@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Driver for Alibaba Elastic Ethernet Adapter.
+ *
+ * Copyright (C) 2025 Alibaba Inc.
+ */
+
+#ifndef __EEA_NET_H__
+#define __EEA_NET_H__
+
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+
+#include "eea_adminq.h"
+#include "eea_ring.h"
+
+#define EEA_VER_MAJOR 1
+#define EEA_VER_MINOR 0
+#define EEA_VER_SUB_MINOR 0
+
+struct eea_net_tx {
+ struct eea_net *enet;
+
+ struct eea_ring *ering;
+
+ struct eea_tx_meta *meta;
+ struct eea_tx_meta *free;
+
+ struct device *dma_dev;
+
+ u32 index;
+
+ char name[16];
+};
+
+struct eea_rx_meta {
+ struct eea_rx_meta *next;
+
+ struct page *page;
+ dma_addr_t dma;
+ u32 offset;
+ u32 frags;
+
+ struct page *hdr_page;
+ void *hdr_addr;
+ dma_addr_t hdr_dma;
+
+ u32 id;
+
+ u32 truesize;
+ u32 headroom;
+ u32 tailroom;
+
+ u32 len;
+};
+
+struct eea_net_rx_pkt_ctx {
+ u16 idx;
+
+ bool data_valid;
+ bool do_drop;
+
+ struct sk_buff *head_skb;
+};
+
+struct eea_net_rx {
+ struct eea_net *enet;
+
+ struct eea_ring *ering;
+
+ struct eea_rx_meta *meta;
+ struct eea_rx_meta *free;
+
+ struct device *dma_dev;
+
+ u32 index;
+
+ u32 flags;
+
+ u32 headroom;
+
+ struct napi_struct *napi;
+
+ char name[16];
+
+ struct eea_net_rx_pkt_ctx pkt;
+
+ struct page_pool *pp;
+};
+
+struct eea_net_cfg {
+ u32 rx_ring_depth;
+ u32 tx_ring_depth;
+ u32 rx_ring_num;
+ u32 tx_ring_num;
+
+ u8 rx_sq_desc_size;
+ u8 rx_cq_desc_size;
+ u8 tx_sq_desc_size;
+ u8 tx_cq_desc_size;
+
+ u32 split_hdr;
+};
+
+enum {
+ EEA_LINK_ERR_NONE,
+ EEA_LINK_ERR_HA_RESET_DEV,
+ EEA_LINK_ERR_LINK_DOWN,
+};
+
+struct eea_net {
+ struct eea_device *edev;
+ struct net_device *netdev;
+
+ struct eea_aq adminq;
+
+ struct eea_net_tx *tx;
+ struct eea_net_rx **rx;
+
+ struct eea_net_cfg cfg;
+ struct eea_net_cfg cfg_hw;
+
+ u32 link_err;
+
+ bool started;
+
+ u8 duplex;
+ u32 speed;
+
+ u64 hw_ts_offset;
+};
+
+int eea_net_probe(struct eea_device *edev);
+void eea_net_remove(struct eea_device *edev);
+void eea_net_shutdown(struct eea_device *edev);
+
+#endif
diff --git a/drivers/net/ethernet/alibaba/eea/eea_pci.c b/drivers/net/ethernet/alibaba/eea/eea_pci.c
index 185774dee38e..e7874979155b 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_pci.c
+++ b/drivers/net/ethernet/alibaba/eea/eea_pci.c
@@ -8,6 +8,7 @@
#include <linux/io-64-nonatomic-lo-hi.h>
#include <linux/iopoll.h>
+#include "eea_net.h"
#include "eea_pci.h"
#define EEA_PCI_DB_OFFSET 4096
@@ -66,7 +67,9 @@ struct eea_pci_device {
((void __iomem *)((reg) + offsetof(struct eea_pci_cfg, item)))
#define cfg_write8(reg, item, val) iowrite8(val, cfg_pointer(reg, item))
+#define cfg_write16(reg, item, val) iowrite16(val, cfg_pointer(reg, item))
#define cfg_write32(reg, item, val) iowrite32(val, cfg_pointer(reg, item))
+#define cfg_write64(reg, item, val) iowrite64_lo_hi(val, cfg_pointer(reg, item))
#define cfg_read8(reg, item) ioread8(cfg_pointer(reg, item))
#define cfg_read32(reg, item) ioread32(cfg_pointer(reg, item))
@@ -339,6 +342,25 @@ void __iomem *eea_pci_db_addr(struct eea_device *edev, u32 off)
return edev->ep_dev->db_base + off;
}
+int eea_pci_active_aq(struct eea_ring *ering, int msix_vec)
+{
+ struct eea_pci_device *ep_dev = ering->edev->ep_dev;
+
+ cfg_write16(ep_dev->reg, aq_size, ering->num);
+ cfg_write16(ep_dev->reg, aq_msix_vector, msix_vec);
+
+ cfg_write64(ep_dev->reg, aq_sq_addr, ering->sq.dma_addr);
+ cfg_write64(ep_dev->reg, aq_cq_addr, ering->cq.dma_addr);
+
+ ering->db = eea_pci_db_addr(ering->edev,
+ cfg_read32(ep_dev->reg, aq_db_off));
+
+ if (!ering->db)
+ return -EIO;
+
+ return 0;
+}
+
u64 eea_pci_device_ts(struct eea_device *edev)
{
struct eea_pci_device *ep_dev = edev->ep_dev;
@@ -360,7 +382,9 @@ static int eea_init_device(struct eea_device *edev)
if (err)
goto err;
- /* do net device probe ... */
+ err = eea_net_probe(edev);
+ if (err)
+ goto err;
return 0;
err:
@@ -394,6 +418,9 @@ static void __eea_pci_remove(struct pci_dev *pci_dev)
{
struct eea_pci_device *ep_dev = pci_get_drvdata(pci_dev);
struct device *dev = get_device(&ep_dev->pci_dev->dev);
+ struct eea_device *edev = &ep_dev->edev;
+
+ eea_net_remove(edev);
eea_pci_release_resource(ep_dev);
@@ -431,8 +458,6 @@ static void eea_pci_remove(struct pci_dev *pci_dev)
{
struct eea_pci_device *ep_dev = pci_get_drvdata(pci_dev);
- eea_device_reset(&ep_dev->edev);
-
__eea_pci_remove(pci_dev);
pci_set_drvdata(pci_dev, NULL);
@@ -448,9 +473,7 @@ static void eea_pci_shutdown(struct pci_dev *pci_dev)
ep_dev->shutdown = true;
- /* do net device stop and clear. */
-
- eea_device_reset(edev);
+ eea_net_shutdown(edev);
pci_clear_master(pci_dev);
}
diff --git a/drivers/net/ethernet/alibaba/eea/eea_pci.h b/drivers/net/ethernet/alibaba/eea/eea_pci.h
index 746cce4dd68e..cfd278e2efde 100644
--- a/drivers/net/ethernet/alibaba/eea/eea_pci.h
+++ b/drivers/net/ethernet/alibaba/eea/eea_pci.h
@@ -10,6 +10,8 @@
#include <linux/pci.h>
+#include "eea_ring.h"
+
struct eea_pci_cap {
__u8 cap_vndr;
__u8 cap_next;
@@ -43,6 +45,7 @@ u16 eea_pci_bdf(struct eea_device *edev);
int eea_device_reset(struct eea_device *dev);
int eea_pci_set_aq_up(struct eea_device *dev);
+int eea_pci_active_aq(struct eea_ring *ering, int msix_vec);
u64 eea_pci_device_ts(struct eea_device *edev);
--
2.32.0.3.g01195cf9f
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox