* Re: [PATCH] af_unix: fix build warning
From: Rami Rosen @ 2013-04-03 12:27 UTC (permalink / raw)
To: dingtianhong; +Cc: David S. Miller, Eric Dumazet, Netdev, Li Zefan
In-Reply-To: <515BF6DA.9010007@huawei.com>
Hi,
Which version of gcc are you using ?
A patch like yours, for the same method, unix_bind(), was sent in the
past, and Stephen Hemminger noted the
with gcc 4.7 the warning you got does not occur.
see:
http://permalink.gmane.org/gmane.linux.network/247807
Best Regards,
Rami Rosen
http://ramirose.wix.com/ramirosen
On Wed, Apr 3, 2013 at 12:31 PM, dingtianhong <dingtianhong@huawei.com> wrote:
> net/unix/af_unix.c: In function ‘unix_bind’:
> net/unix/af_unix.c:892: warning: ‘path.mnt’ may be used uninitialized in this function
> net/unix/af_unix.c:892: warning: ‘path.dentry’ may be used uninitialized in this function
>
> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
> ---
> net/unix/af_unix.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
> index 971282b..3ccc049 100644
> --- a/net/unix/af_unix.c
> +++ b/net/unix/af_unix.c
> @@ -889,7 +889,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
> atomic_set(&addr->refcnt, 1);
>
> if (sun_path[0]) {
> - struct path path;
> + struct path path = {};
> umode_t mode = S_IFSOCK |
> (SOCK_INODE(sock)->i_mode & ~current_umask());
> err = unix_mknod(sun_path, mode, &path);
> --
> 1.8.0
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: 3.7.10 kernel crash
From: Peter Hurley @ 2013-04-03 11:49 UTC (permalink / raw)
To: Fabio Coatti
Cc: linux-kernel, Greg Kroah-Hartman, netdev, Matt Carlson,
Michael Chan
In-Reply-To: <1426784.IFBeqasnTm@calvin>
On Tue, 2013-04-02 at 11:16 +0200, Fabio Coatti wrote:
> Ok, I'll try to get better oopes. However, this is going to be tricky, as the
> machine is remotely administered (via HP iLO) and uses all network interfaces
> (BTW, I'm not even sure to be able to get a network driver related crash using
> netconsole approach)
I use a separate network adapter specifically dedicated for netconsole
(but then I'm trying to crash the machine :)
> So far, no success in using a different resolution for
> boot console.
What distribution is this and what bootloader are you using?
^ permalink raw reply
* [PATCH] af_unix: fix build warning
From: dingtianhong @ 2013-04-03 9:31 UTC (permalink / raw)
To: David S. Miller, Eric Dumazet, netdev, Li Zefan
net/unix/af_unix.c: In function ‘unix_bind’:
net/unix/af_unix.c:892: warning: ‘path.mnt’ may be used uninitialized in this function
net/unix/af_unix.c:892: warning: ‘path.dentry’ may be used uninitialized in this function
Signed-off-by: Ding Tianhong <dingtianhong@huawei.com>
---
net/unix/af_unix.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 971282b..3ccc049 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -889,7 +889,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
atomic_set(&addr->refcnt, 1);
if (sun_path[0]) {
- struct path path;
+ struct path path = {};
umode_t mode = S_IFSOCK |
(SOCK_INODE(sock)->i_mode & ~current_umask());
err = unix_mknod(sun_path, mode, &path);
--
1.8.0
^ permalink raw reply related
* Re: [PATCH 2/4] ARM: kirkwood: add device node entries for the gigabit interfaces
From: Florian Fainelli @ 2013-04-03 9:27 UTC (permalink / raw)
To: Florian Fainelli
Cc: moinejf-GANU6spQydw, jason-NLaQJdtUoK4Be96aLqz0jA,
andrew-g2DYL2Zd6BY, netdev-u79uwXL29TY76Z2rM5mHXA,
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ,
rob.herring-bsGFqQB8/DxBDgjK7y7TUQ,
buytenh-OLH4Qvv75CYX/NnBR394Jw, jm-Pj/HzkgeCk7QXOPxS62xeg,
davem-fT/PcQaiUtIeIZ0/mPfg9Q,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
sebastian.hesselbarth-Re5JQEeQqe8AvxtiuMwx3w
In-Reply-To: <1364580879-4297-3-git-send-email-florian-p3rKhJxN3npAfugRpC6u6w@public.gmane.org>
Le 03/29/13 19:14, Florian Fainelli a écrit :
> This patch modifies kirkwood.dtsi to specify the various gigabit
> interfaces nodes available on kirkwood devices. They are disabled by
> default and should be enabled on a per-board basis. egiga0 and egiga1
> aliases are defined for convenience. The mdio node is also present and
> should be enabled on a per-board basis as well.
>
> Signed-off-by: Florian Fainelli <florian@openwrt.org>
> ---
[snip]
> +
> + ethernet-group@70000 {
This one is off-by 0x2000 and should have been 72000
> + #address-cells = <1>;
> + #size-cells = <0>;
> + compatible = "marvell,mv643xx-eth-block";
> + reg = <0x70000 0x4000>;
Same here
> + tx-csum-limit = <1600>;
> + status = "disabled";
> +
> + egiga0: egiga0@0 {
> + device_type = "network";
> + compatible = "marvell,mv643xx-eth";
> + reg = <0>;
> + interrupts = <11>;
> + clocks = <&gate_clk 0>;
> + clock-names = "0";
> + };
> + };
> +
> + ethernet-group@74000 {
And here, it should be 76000
> + #address-cells = <1>;
> + #size-cells = <0>;
> + compatible = "marvell,mv643xx-eth-block";
> + reg = <0x74000 0x4000>;
And here too.
--
Florian
_______________________________________________
devicetree-discuss mailing list
devicetree-discuss@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/devicetree-discuss
^ permalink raw reply
* [PATCH v2] MPLS: Add limited GSO support
From: Simon Horman @ 2013-04-03 9:11 UTC (permalink / raw)
To: dev, netdev; +Cc: Jesse Gross, jarno.rajahalme, Simon Horman, Pravin B Shelar
In the case where a non-MPLS packet is recieved and an MPLS stack is
added it may well be the case that the original skb is GSO but the
NIC used for transmit does not support GSO of MPLS packets.
The aim of this code is to provide GSO in software for MPLS packets
whose skbs are GSO.
When an implementation adds an MPLS stack to a non-MPLS packet it should do
the following to skb metadata:
* Set skb_mac_header(skb)->protocol to the new MPLS ethertype.
That is, either ETH_P_MPLS_UC or ETH_P_MPLS_MC.
* Leave skb->protocol as the old non-MPLS ethertype.
* Set skb->encapsulation = 1.
This may not strictly be necessary as I believe that checking
skb_mac_header(skb)->protocol and skb->protocol should be necessary and
sufficient.
However, it does seem to fit nicely with the current implementation of
dev_hard_start_xmit() where the more expensive check of
skb_mac_header(skb)->protocol may be guarded by an existing check of
skb->encapsulation.
One aspect of this patch that I am unsure about is the modification I have
made to skb_segment(). This seems to be necessary as checskum accelearation
may no longer be possible as the packet has changed to be MPLS from some
other packet type which may have been supported by the hardware in use.
I will post a patch, "[PATCH v3.24] datapath: Add basic MPLS support to
kernel" which adds MPLS support to the kernel datapath of Open vSwtich.
That patch sets the above requirements in
datapath/actions.c:set_ethertype() and was used to exercise the MPLS GSO
code. The datapath patch is against the Open vSwtich tree but it is
intended that it be added to the Open vSwtich code present in the mainline
Linux kernel at some point.
Suggested by Jesse Gross. Based heavily on "v4 GRE: Add TCP segmentation
offload for GRE" by Pravin B Shelar.
Cc: Jesse Gross <jesse@nicira.com>
Cc: Pravin B Shelar <pshelar@nicira.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
v2
* As suggested by Jarno Rajahalme
- Update NETIF_F_GSO_LAST
- mpls_mc_offload to use ETH_P_MPLS_MC
* Remove NETIF_F_iP_CSUM|NETIF_F_IPV6_CSUM features hack in
mpls_gso_segment()
---
include/linux/netdev_features.h | 4 +-
include/linux/skbuff.h | 15 +++--
net/Kconfig | 1 +
net/Makefile | 1 +
net/core/dev.c | 9 ++-
net/core/ethtool.c | 1 +
net/core/skbuff.c | 3 +-
net/ipv4/af_inet.c | 1 +
net/ipv4/tcp.c | 1 +
net/ipv4/udp.c | 2 +-
net/ipv6/ip6_offload.c | 1 +
net/ipv6/udp_offload.c | 3 +-
net/mpls/Kconfig | 9 +++
net/mpls/Makefile | 4 ++
net/mpls/mpls_gso.c | 124 +++++++++++++++++++++++++++++++++++++++
15 files changed, 170 insertions(+), 9 deletions(-)
create mode 100644 net/mpls/Kconfig
create mode 100644 net/mpls/Makefile
create mode 100644 net/mpls/mpls_gso.c
diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
index d6ee2d0..340f5cd 100644
--- a/include/linux/netdev_features.h
+++ b/include/linux/netdev_features.h
@@ -43,8 +43,9 @@ enum {
NETIF_F_FSO_BIT, /* ... FCoE segmentation */
NETIF_F_GSO_GRE_BIT, /* ... GRE with TSO */
NETIF_F_GSO_UDP_TUNNEL_BIT, /* ... UDP TUNNEL with TSO */
+ NETIF_F_GSO_MPLS_BIT, /* ... MPLS segmentation */
/**/NETIF_F_GSO_LAST = /* last bit, see GSO_MASK */
- NETIF_F_GSO_UDP_TUNNEL_BIT,
+ NETIF_F_GSO_MPLS_BIT,
NETIF_F_FCOE_CRC_BIT, /* FCoE CRC32 */
NETIF_F_SCTP_CSUM_BIT, /* SCTP checksum offload */
@@ -104,6 +105,7 @@ enum {
#define NETIF_F_RXALL __NETIF_F(RXALL)
#define NETIF_F_GSO_GRE __NETIF_F(GSO_GRE)
#define NETIF_F_GSO_UDP_TUNNEL __NETIF_F(GSO_UDP_TUNNEL)
+#define NETIF_F_GSO_MPLS __NETIF_F(GSO_MPLS)
/* Features valid for ethtool to change */
/* = all defined minus driver/device-class-related */
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 878e0ee..9a6c9ca 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -319,6 +319,8 @@ enum {
SKB_GSO_GRE = 1 << 6,
SKB_GSO_UDP_TUNNEL = 1 << 7,
+
+ SKB_GSO_MPLS = 1 << 8,
};
#if BITS_PER_LONG > 32
@@ -2789,12 +2791,17 @@ static inline struct sec_path *skb_sec_path(struct sk_buff *skb)
}
#endif
-/* Keeps track of mac header offset relative to skb->head.
- * It is useful for TSO of Tunneling protocol. e.g. GRE.
- * For non-tunnel skb it points to skb_mac_header() and for
- * tunnel skb it points to outer mac header. */
struct skb_gso_cb {
+ /* Keeps track of mac header offset relative to skb->head.
+ * It is useful for TSO of Tunneling protocol. e.g. GRE.
+ * For non-tunnel skb it points to skb_mac_header() and for
+ * tunnel skb it points to outer mac header. */
int mac_offset;
+
+ /* Keeps track of the ethernet type of an encapsualted
+ * packet where it is otherwise unknown. This is the case
+ * if an MPLS stack is added to a non-MPLS packet. */
+ __be16 encap_eth_type;
};
#define SKB_GSO_CB(skb) ((struct skb_gso_cb *)(skb)->cb)
diff --git a/net/Kconfig b/net/Kconfig
index 2ddc904..d32a7fe 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -218,6 +218,7 @@ source "net/batman-adv/Kconfig"
source "net/openvswitch/Kconfig"
source "net/vmw_vsock/Kconfig"
source "net/netlink/Kconfig"
+source "net/mpls/Kconfig"
config RPS
boolean
diff --git a/net/Makefile b/net/Makefile
index 091e7b04..9492e8c 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -70,3 +70,4 @@ obj-$(CONFIG_BATMAN_ADV) += batman-adv/
obj-$(CONFIG_NFC) += nfc/
obj-$(CONFIG_OPENVSWITCH) += openvswitch/
obj-$(CONFIG_VSOCKETS) += vmw_vsock/
+obj-$(CONFIG_NET_MPLS_GSO) += mpls/
diff --git a/net/core/dev.c b/net/core/dev.c
index 63e2533..3bc68df 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2493,8 +2493,15 @@ int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
* hardware encapsulation features instead of standard
* features for the netdev
*/
- if (skb->encapsulation)
+ if (skb->encapsulation) {
+ struct ethhdr *hdr = (struct ethhdr *)skb_mac_header(skb);
features &= dev->hw_enc_features;
+ if (hdr->h_proto == htons(ETH_P_MPLS_UC) ||
+ hdr->h_proto == htons(ETH_P_MPLS_MC)) {
+ SKB_GSO_CB(skb)->encap_eth_type = skb->protocol;
+ skb->protocol = hdr->h_proto;
+ }
+ }
if (netif_needs_gso(skb, features)) {
if (unlikely(dev_gso_segment(skb, features)))
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index adc1351..f6d5d06 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -79,6 +79,7 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN]
[NETIF_F_FSO_BIT] = "tx-fcoe-segmentation",
[NETIF_F_GSO_GRE_BIT] = "tx-gre-segmentation",
[NETIF_F_GSO_UDP_TUNNEL_BIT] = "tx-udp_tnl-segmentation",
+ [NETIF_F_GSO_MPLS_BIT] = "tx-mpls-segmentation",
[NETIF_F_FCOE_CRC_BIT] = "tx-checksum-fcoe-crc",
[NETIF_F_SCTP_CSUM_BIT] = "tx-checksum-sctp",
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index ba64614..d63292b 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -2824,7 +2824,7 @@ struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features)
doffset + tnl_hlen);
if (fskb != skb_shinfo(skb)->frag_list)
- continue;
+ goto csum;
if (!sg) {
nskb->ip_summed = CHECKSUM_NONE;
@@ -2888,6 +2888,7 @@ skip_fraglist:
nskb->len += nskb->data_len;
nskb->truesize += nskb->data_len;
+csum:
if (!csum) {
nskb->csum = skb_checksum(nskb, doffset,
nskb->len - doffset, 0);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 93824c5..f5b4c9a 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1291,6 +1291,7 @@ static struct sk_buff *inet_gso_segment(struct sk_buff *skb,
SKB_GSO_TCP_ECN |
SKB_GSO_GRE |
SKB_GSO_UDP_TUNNEL |
+ SKB_GSO_MPLS |
0)))
goto out;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index a96f7b5..425c933 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2914,6 +2914,7 @@ struct sk_buff *tcp_tso_segment(struct sk_buff *skb,
SKB_GSO_TCP_ECN |
SKB_GSO_TCPV6 |
SKB_GSO_GRE |
+ SKB_GSO_MPLS |
SKB_GSO_UDP_TUNNEL |
0) ||
!(type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))))
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 7117d14..0b5b57f 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2371,7 +2371,7 @@ struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb,
if (unlikely(type & ~(SKB_GSO_UDP | SKB_GSO_DODGY |
SKB_GSO_UDP_TUNNEL |
- SKB_GSO_GRE) ||
+ SKB_GSO_GRE | SKB_GSO_MPLS) ||
!(type & (SKB_GSO_UDP))))
goto out;
diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c
index 71b766e..a263b99 100644
--- a/net/ipv6/ip6_offload.c
+++ b/net/ipv6/ip6_offload.c
@@ -98,6 +98,7 @@ static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb,
SKB_GSO_TCP_ECN |
SKB_GSO_GRE |
SKB_GSO_UDP_TUNNEL |
+ SKB_GSO_MPLS |
SKB_GSO_TCPV6 |
0)))
goto out;
diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c
index 3bb3a89..76d401a 100644
--- a/net/ipv6/udp_offload.c
+++ b/net/ipv6/udp_offload.c
@@ -63,7 +63,8 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb,
if (unlikely(type & ~(SKB_GSO_UDP |
SKB_GSO_DODGY |
SKB_GSO_UDP_TUNNEL |
- SKB_GSO_GRE) ||
+ SKB_GSO_GRE |
+ SKB_GSO_MPLS) ||
!(type & (SKB_GSO_UDP))))
goto out;
diff --git a/net/mpls/Kconfig b/net/mpls/Kconfig
new file mode 100644
index 0000000..37421db
--- /dev/null
+++ b/net/mpls/Kconfig
@@ -0,0 +1,9 @@
+#
+# MPLS configuration
+#
+config NET_MPLS_GSO
+ tristate "MPLS: GSO support"
+ help
+ This is helper module to allow segmentation of non-MPLS GSO packets
+ that have had MPLS stack entries pushed onto them and thus
+ become MPLS GSO packets.
diff --git a/net/mpls/Makefile b/net/mpls/Makefile
new file mode 100644
index 0000000..0a3c171
--- /dev/null
+++ b/net/mpls/Makefile
@@ -0,0 +1,4 @@
+#
+# Makefile for MPLS.
+#
+obj-y += mpls_gso.o
diff --git a/net/mpls/mpls_gso.c b/net/mpls/mpls_gso.c
new file mode 100644
index 0000000..f96e5b0
--- /dev/null
+++ b/net/mpls/mpls_gso.c
@@ -0,0 +1,124 @@
+/*
+ * MPLS GSO Support
+ *
+ * Authors: Simon Horman (horms@verge.net.au)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Based on: GSO portions of net/ipv4/gre.c
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/netdev_features.h>
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+
+#define MPLS_BOS_MASK 0x00000100 /* Bottom of Stack bit */
+#define MPLS_HLEN (sizeof(__be32))
+
+static struct sk_buff *mpls_gso_segment(struct sk_buff *skb,
+ netdev_features_t features)
+{
+ struct sk_buff *segs = ERR_PTR(-EINVAL);
+ netdev_features_t enc_features;
+ int stack_len = 0;
+ int mac_len = skb->mac_len;
+
+ if (unlikely(skb_shinfo(skb)->gso_type &
+ ~(SKB_GSO_TCPV4 |
+ SKB_GSO_TCPV6 |
+ SKB_GSO_UDP |
+ SKB_GSO_DODGY |
+ SKB_GSO_TCP_ECN |
+ SKB_GSO_GRE |
+ SKB_GSO_MPLS)))
+ goto out;
+
+ while (1) {
+ __be32 lse = *(__be32 *)(skb_network_header(skb) + stack_len);
+
+ stack_len += MPLS_HLEN;
+ if (unlikely(!pskb_may_pull(skb, MPLS_HLEN)))
+ goto out;
+ __skb_pull(skb, MPLS_HLEN);
+
+ if (lse & htonl(MPLS_BOS_MASK))
+ break;
+ }
+
+ /* setup inner skb. */
+ skb_reset_mac_header(skb);
+ skb_reset_network_header(skb);
+ skb->mac_len = 0;
+ skb->protocol = SKB_GSO_CB(skb)->encap_eth_type;
+ skb->encapsulation = 0;
+
+ /* segment inner packet. */
+ enc_features = skb->dev->hw_enc_features & netif_skb_features(skb);
+ segs = skb_mac_gso_segment(skb, enc_features);
+ if (IS_ERR_OR_NULL(segs))
+ goto out;
+
+ skb = segs;
+ do {
+ __skb_push(skb, stack_len + mac_len);
+ skb_reset_mac_header(skb);
+ skb_set_network_header(skb, mac_len);
+ skb->mac_len = mac_len;
+ } while ((skb = skb->next));
+out:
+ return segs;
+}
+
+static int mpls_gso_send_check(struct sk_buff *skb)
+{
+ if (!skb->encapsulation)
+ return -EINVAL;
+ return 0;
+}
+
+static struct packet_offload mpls_mc_offload = {
+ .type = cpu_to_be16(ETH_P_MPLS_MC),
+ .callbacks = {
+ .gso_send_check = mpls_gso_send_check,
+ .gso_segment = mpls_gso_segment,
+ },
+};
+
+static struct packet_offload mpls_uc_offload = {
+ .type = cpu_to_be16(ETH_P_MPLS_UC),
+ .callbacks = {
+ .gso_send_check = mpls_gso_send_check,
+ .gso_segment = mpls_gso_segment,
+ },
+};
+
+static int __init mpls_gso_init(void)
+{
+ pr_info("MPLS GSO support\n");
+
+ dev_add_offload(&mpls_uc_offload);
+ dev_add_offload(&mpls_mc_offload);
+
+ return 0;
+}
+
+static void __exit mpls_gso_exit(void)
+{
+ dev_remove_offload(&mpls_uc_offload);
+ dev_remove_offload(&mpls_mc_offload);
+}
+
+module_init(mpls_gso_init);
+module_exit(mpls_gso_exit);
+
+MODULE_DESCRIPTION("MPLS GSO support");
+MODULE_AUTHOR("Simon Horman (horms@verge.net.au)");
+MODULE_LICENSE("GPL");
+
--
1.7.10.4
^ permalink raw reply related
* Re: [net-next.git 2/7] stmmac: review barriers
From: Shiraz Hashim @ 2013-04-03 8:51 UTC (permalink / raw)
To: Giuseppe CAVALLARO; +Cc: netdev@vger.kernel.org, Deepak Sikri
In-Reply-To: <1364967689-11155-2-git-send-email-peppe.cavallaro@st.com>
Hi Giuseppe,
On Wed, Apr 03, 2013 at 01:41:24PM +0800, Giuseppe CAVALLARO wrote:
> In all my tests performed on SH4 and ARM A9 platforms, I've never met problems
> that can be fixed by using memory barriers. In the past there was some issues
> on SMP ARM but fixed by reviewing xmit spinlock.
The problem which was addressed was not because of SMP IMO. It was
rather due to the fact that the write to the GMAC descriptor (which
happens to be in normal memory) has to be ordered with respect to GMAC
DMA as observer. Isn't it ?
> Further barriers have been added in the commits too: 8e83989106562326bf
>
> This patch is to use the smp_wbm instead of wbm because the driver
^^^^^^^^^^^^^^^^^^^^^^
Perhaps you meant smp_wmb and wmb
> runs on UP systems. Then, IMO it could make sense to only maintain the barriers
> just in places where we touch the dma owner bits (that is the
> only real critical path as we had seen and fixed in the commit:
> eb0dc4bb2e22c04964d).
Replacing wmb by smp_wmb may not be a good idea as we need to order
the store transaction to the descriptor with respect to GMAC DMA and
using smp_* version would just be compiler barrier in uniprocessor
systems.
> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
> Cc: Deepak Sikri <deepak.sikri@st.com>
> Cc: Shiraz Hashim <shiraz.hashim@st.com>
> ---
> drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++------
> 1 files changed, 3 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> index 8b69e3b..c92dcbc 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> @@ -1797,15 +1797,13 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
> priv->tx_skbuff[entry] = NULL;
> priv->hw->desc->prepare_tx_desc(desc, 0, len, csum_insertion,
> priv->mode);
> - wmb();
> + smp_wmb();
> priv->hw->desc->set_tx_owner(desc);
> - wmb();
Since it is a loop, shouldn't we ensure that the ownership of a tx
descriptor is set before next descriptor in chain is programmed ?
> }
>
> /* Finalize the latest segment. */
> priv->hw->desc->close_tx_desc(desc);
>
> - wmb();
> /* According to the coalesce parameter the IC bit for the latest
> * segment could be reset and the timer re-started to invoke the
> * stmmac_tx function. This approach takes care about the fragments.
> @@ -1821,9 +1819,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
> } else
> priv->tx_count_frames = 0;
>
> + smp_wmb();
Please reconsider, may be keeping wmb is better.
> /* To avoid raise condition */
> priv->hw->desc->set_tx_owner(first);
> - wmb();
Not sure about this, perhaps can be removed.
>
> priv->cur_tx++;
>
> @@ -1899,9 +1897,8 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
>
> RX_DBG(KERN_INFO "\trefill entry #%d\n", entry);
> }
> - wmb();
> + smp_wmb();
> priv->hw->desc->set_rx_owner(p);
> - wmb();
Similarly this is a part of a loop, we need to see if set rx owner
should be reflected before next descriptor program.
--
regards
Shiraz
^ permalink raw reply
* [PATCH net-next] 802: fix a possible race condition
From: Cong Wang @ 2013-04-03 7:52 UTC (permalink / raw)
To: netdev
Cc: Eric Dumazet, David S. Miller, David Ward, Jorge Boncompte [DTI2],
Cong Wang
From: Cong Wang <amwang@redhat.com>
(Resend with a better changelog)
garp_pdu_queue() should ways be called with this spin lock.
garp_uninit_applicant() only holds rtnl lock which is not
enough here. A possible race can happen as garp_pdu_rcv()
is called in BH context:
garp_pdu_rcv()
|->garp_pdu_parse_msg()
|->garp_pdu_parse_attr()
|-> garp_gid_event()
Found by code inspection.
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: David Ward <david.ward@ll.mit.edu>
Cc: "Jorge Boncompte [DTI2]" <jorge@dti2.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
net/802/garp.c | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/net/802/garp.c b/net/802/garp.c
index 8456f5d..5d9630a 100644
--- a/net/802/garp.c
+++ b/net/802/garp.c
@@ -609,8 +609,12 @@ void garp_uninit_applicant(struct net_device *dev, struct garp_application *appl
/* Delete timer and generate a final TRANSMIT_PDU event to flush out
* all pending messages before the applicant is gone. */
del_timer_sync(&app->join_timer);
+
+ spin_lock_bh(&app->lock);
garp_gid_event(app, GARP_EVENT_TRANSMIT_PDU);
garp_pdu_queue(app);
+ spin_unlock_bh(&app->lock);
+
garp_queue_xmit(app);
dev_mc_del(dev, appl->proto.group_address);
--
1.7.7.6
^ permalink raw reply related
* Re: [net-next.git 1/7] stmmac: review napi gro support
From: Giuseppe CAVALLARO @ 2013-04-03 7:41 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <1364972715.5113.197.camel@edumazet-glaptop>
On 4/3/2013 9:05 AM, Eric Dumazet wrote:
> On Wed, 2013-04-03 at 07:41 +0200, Giuseppe CAVALLARO wrote:
>> This patch is to:
>> o use napi_gro_flush() before calling __napi_complete()
>> o turn on NETIF_F_GRO by default
>>
>> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
>> ---
>> drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 +++--
>> 1 files changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>> index 6b26d31..8b69e3b 100644
>> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>> @@ -2046,7 +2046,8 @@ static int stmmac_poll(struct napi_struct *napi, int budget)
>>
>> work_done = stmmac_rx(priv, budget);
>> if (work_done < budget) {
>> - napi_complete(napi);
>> + napi_gro_flush(napi, false);
>> + __napi_complete(napi);
>> stmmac_enable_dma_irq(priv);
>> }
>
> Why are you doing this ?
>
> This adds a (fatal) race.
Hmm, I'm in trouble on this :-). Indeed I can understand the (fatal)
race and why napi_complete should be used. Sorry! So my fault and this
patch has to be discarded. I don't understand why I have not seen any
problems while running/stressing on SMP system. Have you got any idea?
Thanks Eric for your prompt feedback.
Let me know if you see other problems so I'll try to fix all soon.
peppe
^ permalink raw reply
* Re: [PATCH net-next] openvswitch: Provide OVS_DP_ATTR_UPCALL_PID in datapath messages
From: Thomas Graf @ 2013-04-03 7:33 UTC (permalink / raw)
To: Jesse Gross; +Cc: dev@openvswitch.org, netdev
In-Reply-To: <CAEP_g=_5+1ZjiXEMCTSXE2Rt6HNkfFx1uR3KNCzwitJrLQYesg@mail.gmail.com>
On 04/02/13 at 04:30pm, Jesse Gross wrote:
> Can you describe the race condition some more? The kernel doesn't
> change the port ID on its own so even needing to request the value
> seems rare.
The upcall nlport is changeable with OVS_VPORT_CMD_SET and may
be received between the OVS_DP_CMD_GET and OVS_VPORT_CMD_GET.
> Assigning the local ports upcall PID through datapath creation is
> really somewhat of a hack since it's port state. I don't disagree
> that it's somewhat asymmetric now but it seems better to move away
> from the current model if possible.
Would you suggest to wait with local vport creation if no upcall
nlport is provided and let new user space binaries create the local
port explicitly?
We can't get rid of the attribute in datapath messages w/o breaking
ABI but we could make its use optional I guess.
^ permalink raw reply
* Re: [net-next.git 3/7] stmmac: review private structure fields
From: Giuseppe CAVALLARO @ 2013-04-03 7:33 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <1364973691.13853.7.camel@edumazet-glaptop>
On 4/3/2013 9:21 AM, Eric Dumazet wrote:
> On Wed, 2013-04-03 at 07:41 +0200, Giuseppe CAVALLARO wrote:
>> recently many new supports have been added in the stmmac driver w/o taking care
>> about where each new field had to be placed inside the private structure for
>> guaranteeing the best cache usage.
>> This is what I wanted in the beginning, so this patch reorganizes all the fields
>> in order to keep adjacent fields for cache effect.
>> I have also tried to optimize them by using pahole.
>>
>> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
>> ---
>> drivers/net/ethernet/stmicro/stmmac/stmmac.h | 70 +++++++++++++-------------
>> 1 files changed, 35 insertions(+), 35 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
>> index 75f997b..8aa28c5 100644
>> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
>> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
>> @@ -35,36 +35,45 @@
>>
>> struct stmmac_priv {
>> /* Frequently used values are kept adjacent for cache effect */
>> - struct dma_desc *dma_tx ____cacheline_aligned; /* Basic TX desc */
>> - struct dma_extended_desc *dma_etx; /* Extended TX descriptor */
>> - dma_addr_t dma_tx_phy;
>> - struct sk_buff **tx_skbuff;
>> - dma_addr_t *tx_skbuff_dma;
>> + struct dma_extended_desc *dma_etx;
>> + struct dma_desc *dma_tx ____cacheline_aligned_in_smp;
>> + struct sk_buff **tx_skbuff ____cacheline_aligned_in_smp;
>
> dma_tx & tx_skbuff are readonly, why put them in separate cache lines ?
I put tx_skbuff in a separate cache line because, when we use extended
descriptors, the driver works with dma_etx instead of dma_tx.
So my idea was to have both dma_etx, dma_tx and tx_skbuff aligned in
any case.
>
> It seems there is an abuse of ____cacheline_aligned_in_smp in this
> driver (especially if this driver only runs on UP arch)
Yes I know that there is this abuse but why do you see an abuse for UP?
In that case we should fall through ____cacheline_aligned (e.g. it is ok
for SH4).
peppe
>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply
* Re: [ovs-dev] [PATCH] MPLS: Add limited GSO support
From: Simon Horman @ 2013-04-03 7:32 UTC (permalink / raw)
To: Rajahalme, Jarno (NSN - FI/Espoo)
Cc: <dev@openvswitch.org>, <netdev@vger.kernel.org>
In-Reply-To: <74B6572A-FB96-41FA-83E8-5206FFABD1D5@nsn.com>
On Wed, Apr 03, 2013 at 06:36:29AM +0000, Rajahalme, Jarno (NSN - FI/Espoo) wrote:
>
> On Apr 3, 2013, at 8:24 , ext Simon Horman wrote:
>
> > In the case where a non-MPLS packet is recieved and an MPLS stack is
> > added it may well be the case that the original skb is GSO but the
> > NIC used for transmit does not support GSO of MPLS packets.
> >
> ...
> > diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
> > index d6ee2d0..e7bffa8 100644
> > --- a/include/linux/netdev_features.h
> > +++ b/include/linux/netdev_features.h
> > @@ -43,6 +43,7 @@ enum {
> > NETIF_F_FSO_BIT, /* ... FCoE segmentation */
> > NETIF_F_GSO_GRE_BIT, /* ... GRE with TSO */
> > NETIF_F_GSO_UDP_TUNNEL_BIT, /* ... UDP TUNNEL with TSO */
> > + NETIF_F_GSO_MPLS_BIT, /* ... MPLS segmentation */
> > /**/NETIF_F_GSO_LAST = /* last bit, see GSO_MASK */
> > NETIF_F_GSO_UDP_TUNNEL_BIT,
> >
>
> NETIF_F_GSO_LAST needs an update here?
Thanks, I will fix that in v2.
> ...
> >
> > diff --git a/net/mpls/mpls_gso.c b/net/mpls/mpls_gso.c
> > new file mode 100644
> > index 0000000..583de9e
> > --- /dev/null
> > +++ b/net/mpls/mpls_gso.c
> > @@ -0,0 +1,125 @@
> > +/*
> > + * MPLS GSO Support
> > + *
> > + * Authors: Simon Horman (horms@verge.net.au)
> > + *
> > + * This program is free software; you can redistribute it and/or
> > + * modify it under the terms of the GNU General Public License
> > + * as published by the Free Software Foundation; either version
> > + * 2 of the License, or (at your option) any later version.
> > + *
> > + * Based on: GSO portions of net/ipv4/gre.c
> > + */
> > +
> > +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> > +
> > +#include <linux/err.h>
> > +#include <linux/module.h>
> > +#include <linux/netdev_features.h>
> > +#include <linux/netdevice.h>
> > +#include <linux/skbuff.h>
> > +
> > +#define MPLS_BOS_MASK 0x00000100 /* Bottom of Stack bit */
> > +#define MPLS_HLEN (sizeof(__be32))
> > +
> > +static struct sk_buff *mpls_gso_segment(struct sk_buff *skb,
> > + netdev_features_t features)
> > +{
> > + struct sk_buff *segs = ERR_PTR(-EINVAL);
> > + netdev_features_t enc_features;
> > + int stack_len = 0;
> > + int mac_len = skb->mac_len;
> > +
> > + if (unlikely(skb_shinfo(skb)->gso_type &
> > + ~(SKB_GSO_TCPV4 |
> > + SKB_GSO_TCPV6 |
> > + SKB_GSO_UDP |
> > + SKB_GSO_DODGY |
> > + SKB_GSO_TCP_ECN |
> > + SKB_GSO_GRE |
> > + SKB_GSO_MPLS)))
> > + goto out;
> > +
> > + while (1) {
> > + __be32 lse = *(__be32 *)(skb_network_header(skb) + stack_len);
> > +
> > + stack_len += MPLS_HLEN;
> > + if (unlikely(!pskb_may_pull(skb, MPLS_HLEN)))
> > + goto out;
> > + __skb_pull(skb, MPLS_HLEN);
> > +
> > + if (lse & htonl(MPLS_BOS_MASK))
> > + break;
> > + }
> > +
> > + /* setup inner skb. */
> > + skb_reset_mac_header(skb);
> > + skb_reset_network_header(skb);
> > + skb->mac_len = 0;
> > + skb->protocol = SKB_GSO_CB(skb)->encap_eth_type;
> > + skb->encapsulation = 0;
> > +
> > + /* segment inner packet. */
> > + enc_features = skb->dev->hw_enc_features & netif_skb_features(skb) &
> > + ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
>
> Add spaces around "|" to make this more readable.
I'm not sure how that crept through.
I believe it should actually be just:
enc_features = skb->dev->hw_enc_features & netif_skb_features(skb);
> > + segs = skb_mac_gso_segment(skb, enc_features);
> > + if (IS_ERR_OR_NULL(segs))
> > + goto out;
> > +
> > + skb = segs;
> > + do {
> > + __skb_push(skb, stack_len + mac_len);
> > + skb_reset_mac_header(skb);
> > + skb_set_network_header(skb, mac_len);
> > + skb->mac_len = mac_len;
> > + } while ((skb = skb->next));
> > +out:
> > + return segs;
> > +}
> > +
> > +static int mpls_gso_send_check(struct sk_buff *skb)
> > +{
> > + if (!skb->encapsulation)
> > + return -EINVAL;
> > + return 0;
> > +}
> > +
> > +static struct packet_offload mpls_mc_offload = {
> > + .type = cpu_to_be16(ETH_P_MPLS_UC),
>
> Should have ETH_P_MPLS_MC here?
Yes, thanks. I'll fix that in v2.
> > + .callbacks = {
> > + .gso_send_check = mpls_gso_send_check,
> > + .gso_segment = mpls_gso_segment,
> > + },
> > +};
> > +
> > +static struct packet_offload mpls_uc_offload = {
> > + .type = cpu_to_be16(ETH_P_MPLS_UC),
> > + .callbacks = {
> > + .gso_send_check = mpls_gso_send_check,
> > + .gso_segment = mpls_gso_segment,
> > + },
> > +};
> > +
> > +static int __init mpls_gso_init(void)
> > +{
> > + pr_info("MPLS GSO support\n");
> > +
> > + dev_add_offload(&mpls_uc_offload);
> > + dev_add_offload(&mpls_mc_offload);
> > +
> > + return 0;
> > +}
> > +
> > +static void __exit mpls_gso_exit(void)
> > +{
> > + dev_remove_offload(&mpls_uc_offload);
> > + dev_remove_offload(&mpls_mc_offload);
> > +}
> > +
> > +module_init(mpls_gso_init);
> > +module_exit(mpls_gso_exit);
> > +
> > +MODULE_DESCRIPTION("MPLS GSO support");
> > +MODULE_AUTHOR("Simon Horman (horms@verge.net.au)");
> > +MODULE_LICENSE("GPL");
> > +
^ permalink raw reply
* Re: [net-next.git 2/7] stmmac: review barriers
From: Giuseppe CAVALLARO @ 2013-04-03 7:28 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <1364973523.13853.4.camel@edumazet-glaptop>
On 4/3/2013 9:18 AM, Eric Dumazet wrote:
> On Wed, 2013-04-03 at 07:41 +0200, Giuseppe CAVALLARO wrote:
>> In all my tests performed on SH4 and ARM A9 platforms, I've never met problems
>> that can be fixed by using memory barriers. In the past there was some issues
>> on SMP ARM but fixed by reviewing xmit spinlock.
>>
>> Further barriers have been added in the commits too: 8e83989106562326bf
>>
>> This patch is to use the smp_wbm instead of wbm because the driver
>> runs on UP systems. Then, IMO it could make sense to only maintain the barriers
>> just in places where we touch the dma owner bits (that is the
>> only real critical path as we had seen and fixed in the commit:
>> eb0dc4bb2e22c04964d).
>>
>> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
>> Cc: Deepak Sikri <deepak.sikri@st.com>
>> Cc: Shiraz Hashim <shiraz.hashim@st.com>
>> ---
>> drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++------
>> 1 files changed, 3 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>> index 8b69e3b..c92dcbc 100644
>> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
>> @@ -1797,15 +1797,13 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
>> priv->tx_skbuff[entry] = NULL;
>> priv->hw->desc->prepare_tx_desc(desc, 0, len, csum_insertion,
>> priv->mode);
>> - wmb();
>> + smp_wmb();
>> priv->hw->desc->set_tx_owner(desc);
>
> This looks pretty bogus to me.
>
> If arch is UP, smp_wmb() is empty.
>
> So why are you using it ?
>
> Barrier here is not to protect the interaction with another cpu, but the
> NIC itself.
>
> If there is no need for barrier as you claim in changelog, don't add a
> fake smp_wmb().
>
ok I'll do it because I do not need any barrier when test on my UP and
SMP. Barriers were added for SPEAr but I have never seen problems
although I have not done too many tests on these platforms. I usually
run on other ST ARM box but, as rule of thumb, similar to some SPEAr
for CPU and MAC.
At any rate, if somebody aims to have them we will discuss in this
mailing list to understand why/where these have to be placed in the code
peppe
^ permalink raw reply
* Re: [net-next.git 3/7] stmmac: review private structure fields
From: Eric Dumazet @ 2013-04-03 7:21 UTC (permalink / raw)
To: Giuseppe CAVALLARO; +Cc: netdev
In-Reply-To: <1364967689-11155-3-git-send-email-peppe.cavallaro@st.com>
On Wed, 2013-04-03 at 07:41 +0200, Giuseppe CAVALLARO wrote:
> recently many new supports have been added in the stmmac driver w/o taking care
> about where each new field had to be placed inside the private structure for
> guaranteeing the best cache usage.
> This is what I wanted in the beginning, so this patch reorganizes all the fields
> in order to keep adjacent fields for cache effect.
> I have also tried to optimize them by using pahole.
>
> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
> ---
> drivers/net/ethernet/stmicro/stmmac/stmmac.h | 70 +++++++++++++-------------
> 1 files changed, 35 insertions(+), 35 deletions(-)
>
> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
> index 75f997b..8aa28c5 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
> @@ -35,36 +35,45 @@
>
> struct stmmac_priv {
> /* Frequently used values are kept adjacent for cache effect */
> - struct dma_desc *dma_tx ____cacheline_aligned; /* Basic TX desc */
> - struct dma_extended_desc *dma_etx; /* Extended TX descriptor */
> - dma_addr_t dma_tx_phy;
> - struct sk_buff **tx_skbuff;
> - dma_addr_t *tx_skbuff_dma;
> + struct dma_extended_desc *dma_etx;
> + struct dma_desc *dma_tx ____cacheline_aligned_in_smp;
> + struct sk_buff **tx_skbuff ____cacheline_aligned_in_smp;
dma_tx & tx_skbuff are readonly, why put them in separate cache lines ?
It seems there is an abuse of ____cacheline_aligned_in_smp in this
driver (especially if this driver only runs on UP arch)
^ permalink raw reply
* Re: [net-next.git 2/7] stmmac: review barriers
From: Eric Dumazet @ 2013-04-03 7:18 UTC (permalink / raw)
To: Giuseppe CAVALLARO; +Cc: netdev, Deepak Sikri, Shiraz Hashim
In-Reply-To: <1364967689-11155-2-git-send-email-peppe.cavallaro@st.com>
On Wed, 2013-04-03 at 07:41 +0200, Giuseppe CAVALLARO wrote:
> In all my tests performed on SH4 and ARM A9 platforms, I've never met problems
> that can be fixed by using memory barriers. In the past there was some issues
> on SMP ARM but fixed by reviewing xmit spinlock.
>
> Further barriers have been added in the commits too: 8e83989106562326bf
>
> This patch is to use the smp_wbm instead of wbm because the driver
> runs on UP systems. Then, IMO it could make sense to only maintain the barriers
> just in places where we touch the dma owner bits (that is the
> only real critical path as we had seen and fixed in the commit:
> eb0dc4bb2e22c04964d).
>
> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
> Cc: Deepak Sikri <deepak.sikri@st.com>
> Cc: Shiraz Hashim <shiraz.hashim@st.com>
> ---
> drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++------
> 1 files changed, 3 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> index 8b69e3b..c92dcbc 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> @@ -1797,15 +1797,13 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
> priv->tx_skbuff[entry] = NULL;
> priv->hw->desc->prepare_tx_desc(desc, 0, len, csum_insertion,
> priv->mode);
> - wmb();
> + smp_wmb();
> priv->hw->desc->set_tx_owner(desc);
This looks pretty bogus to me.
If arch is UP, smp_wmb() is empty.
So why are you using it ?
Barrier here is not to protect the interaction with another cpu, but the
NIC itself.
If there is no need for barrier as you claim in changelog, don't add a
fake smp_wmb().
^ permalink raw reply
* Re: [net-next.git 1/7] stmmac: review napi gro support
From: Eric Dumazet @ 2013-04-03 7:05 UTC (permalink / raw)
To: Giuseppe CAVALLARO; +Cc: netdev
In-Reply-To: <1364967689-11155-1-git-send-email-peppe.cavallaro@st.com>
On Wed, 2013-04-03 at 07:41 +0200, Giuseppe CAVALLARO wrote:
> This patch is to:
> o use napi_gro_flush() before calling __napi_complete()
> o turn on NETIF_F_GRO by default
>
> Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
> ---
> drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 +++--
> 1 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> index 6b26d31..8b69e3b 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> @@ -2046,7 +2046,8 @@ static int stmmac_poll(struct napi_struct *napi, int budget)
>
> work_done = stmmac_rx(priv, budget);
> if (work_done < budget) {
> - napi_complete(napi);
> + napi_gro_flush(napi, false);
> + __napi_complete(napi);
> stmmac_enable_dma_irq(priv);
> }
Why are you doing this ?
This adds a (fatal) race.
^ permalink raw reply
* Re: [PATCH] MPLS: Add limited GSO support
From: Rajahalme, Jarno (NSN - FI/Espoo) @ 2013-04-03 6:36 UTC (permalink / raw)
To: ext Simon Horman
Cc: <dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org>,
<netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
In-Reply-To: <1364966697-20131-1-git-send-email-horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org>
On Apr 3, 2013, at 8:24 , ext Simon Horman wrote:
> In the case where a non-MPLS packet is recieved and an MPLS stack is
> added it may well be the case that the original skb is GSO but the
> NIC used for transmit does not support GSO of MPLS packets.
>
...
> diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
> index d6ee2d0..e7bffa8 100644
> --- a/include/linux/netdev_features.h
> +++ b/include/linux/netdev_features.h
> @@ -43,6 +43,7 @@ enum {
> NETIF_F_FSO_BIT, /* ... FCoE segmentation */
> NETIF_F_GSO_GRE_BIT, /* ... GRE with TSO */
> NETIF_F_GSO_UDP_TUNNEL_BIT, /* ... UDP TUNNEL with TSO */
> + NETIF_F_GSO_MPLS_BIT, /* ... MPLS segmentation */
> /**/NETIF_F_GSO_LAST = /* last bit, see GSO_MASK */
> NETIF_F_GSO_UDP_TUNNEL_BIT,
>
NETIF_F_GSO_LAST needs an update here?
...
>
> diff --git a/net/mpls/mpls_gso.c b/net/mpls/mpls_gso.c
> new file mode 100644
> index 0000000..583de9e
> --- /dev/null
> +++ b/net/mpls/mpls_gso.c
> @@ -0,0 +1,125 @@
> +/*
> + * MPLS GSO Support
> + *
> + * Authors: Simon Horman (horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org)
> + *
> + * This program is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU General Public License
> + * as published by the Free Software Foundation; either version
> + * 2 of the License, or (at your option) any later version.
> + *
> + * Based on: GSO portions of net/ipv4/gre.c
> + */
> +
> +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> +
> +#include <linux/err.h>
> +#include <linux/module.h>
> +#include <linux/netdev_features.h>
> +#include <linux/netdevice.h>
> +#include <linux/skbuff.h>
> +
> +#define MPLS_BOS_MASK 0x00000100 /* Bottom of Stack bit */
> +#define MPLS_HLEN (sizeof(__be32))
> +
> +static struct sk_buff *mpls_gso_segment(struct sk_buff *skb,
> + netdev_features_t features)
> +{
> + struct sk_buff *segs = ERR_PTR(-EINVAL);
> + netdev_features_t enc_features;
> + int stack_len = 0;
> + int mac_len = skb->mac_len;
> +
> + if (unlikely(skb_shinfo(skb)->gso_type &
> + ~(SKB_GSO_TCPV4 |
> + SKB_GSO_TCPV6 |
> + SKB_GSO_UDP |
> + SKB_GSO_DODGY |
> + SKB_GSO_TCP_ECN |
> + SKB_GSO_GRE |
> + SKB_GSO_MPLS)))
> + goto out;
> +
> + while (1) {
> + __be32 lse = *(__be32 *)(skb_network_header(skb) + stack_len);
> +
> + stack_len += MPLS_HLEN;
> + if (unlikely(!pskb_may_pull(skb, MPLS_HLEN)))
> + goto out;
> + __skb_pull(skb, MPLS_HLEN);
> +
> + if (lse & htonl(MPLS_BOS_MASK))
> + break;
> + }
> +
> + /* setup inner skb. */
> + skb_reset_mac_header(skb);
> + skb_reset_network_header(skb);
> + skb->mac_len = 0;
> + skb->protocol = SKB_GSO_CB(skb)->encap_eth_type;
> + skb->encapsulation = 0;
> +
> + /* segment inner packet. */
> + enc_features = skb->dev->hw_enc_features & netif_skb_features(skb) &
> + ~(NETIF_F_IP_CSUM|NETIF_F_IPV6_CSUM);
Add spaces around "|" to make this more readable.
> + segs = skb_mac_gso_segment(skb, enc_features);
> + if (IS_ERR_OR_NULL(segs))
> + goto out;
> +
> + skb = segs;
> + do {
> + __skb_push(skb, stack_len + mac_len);
> + skb_reset_mac_header(skb);
> + skb_set_network_header(skb, mac_len);
> + skb->mac_len = mac_len;
> + } while ((skb = skb->next));
> +out:
> + return segs;
> +}
> +
> +static int mpls_gso_send_check(struct sk_buff *skb)
> +{
> + if (!skb->encapsulation)
> + return -EINVAL;
> + return 0;
> +}
> +
> +static struct packet_offload mpls_mc_offload = {
> + .type = cpu_to_be16(ETH_P_MPLS_UC),
Should have ETH_P_MPLS_MC here?
> + .callbacks = {
> + .gso_send_check = mpls_gso_send_check,
> + .gso_segment = mpls_gso_segment,
> + },
> +};
> +
> +static struct packet_offload mpls_uc_offload = {
> + .type = cpu_to_be16(ETH_P_MPLS_UC),
> + .callbacks = {
> + .gso_send_check = mpls_gso_send_check,
> + .gso_segment = mpls_gso_segment,
> + },
> +};
> +
> +static int __init mpls_gso_init(void)
> +{
> + pr_info("MPLS GSO support\n");
> +
> + dev_add_offload(&mpls_uc_offload);
> + dev_add_offload(&mpls_mc_offload);
> +
> + return 0;
> +}
> +
> +static void __exit mpls_gso_exit(void)
> +{
> + dev_remove_offload(&mpls_uc_offload);
> + dev_remove_offload(&mpls_mc_offload);
> +}
> +
> +module_init(mpls_gso_init);
> +module_exit(mpls_gso_exit);
> +
> +MODULE_DESCRIPTION("MPLS GSO support");
> +MODULE_AUTHOR("Simon Horman (horms-/R6kz+dDXgpPR4JQBCEnsQ@public.gmane.org)");
> +MODULE_LICENSE("GPL");
> +
> --
> 1.7.10.4
>
> _______________________________________________
> dev mailing list
> dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org
> http://openvswitch.org/mailman/listinfo/dev
^ permalink raw reply
* Re: [PATCH v2 net-next 6/8] r8169: add a new chip for RTL8111G
From: Francois Romieu @ 2013-04-03 5:59 UTC (permalink / raw)
To: hayeswang; +Cc: netdev, linux-kernel
In-Reply-To: <5A289D0C2A6C44158FB4CC5226F8E920@realtek.com.tw>
hayeswang <hayeswang@realtek.com> :
[...]
> Excuse me, I don't sure what is your question. Do you mean if the patch
> "[PATCH v2 net-next 5/8] r8169: Update the RTL8111G parameters" is necessary
> before the current patch? According to the settings from our hw engineers,
> some settings of the new chip are the same with the previous 8111g chip
> using the new parameters. It would work without patch #5. However, it is
> necessary to be updated finally, so I add this patch after patch #5.
I was wondering if some part of #5 would be subject to future changes or
reverted in the context of the new chipset (the global description of the
v2 series did not outline the changes in #5).
It completely answers the question. Sorry for the poor wording.
Thanks.
--
Ueimor
^ permalink raw reply
* [net-next.git 7/7] stmmac: prefetch all dma_erx when use extend_desc
From: Giuseppe CAVALLARO @ 2013-04-03 5:41 UTC (permalink / raw)
To: netdev; +Cc: Giuseppe Cavallaro
In-Reply-To: <1364967689-11155-1-git-send-email-peppe.cavallaro@st.com>
This patch is to prefetch, in the stmmac_rx, the whole
dma_erx descriptor in case of using the extended descriptors.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++------
1 files changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 16f5d73..5c6520f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -2001,7 +2001,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
#endif
while (count < limit) {
int status;
- struct dma_desc *p, *p_next;
+ struct dma_desc *p;
if (priv->extend_desc)
p = (struct dma_desc *)(priv->dma_erx + entry);
@@ -2015,12 +2015,9 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
next_entry = (++priv->cur_rx) % rxsize;
if (priv->extend_desc)
- p_next = (struct dma_desc *)(priv->dma_erx +
- next_entry);
+ prefetch(priv->dma_erx + next_entry);
else
- p_next = priv->dma_rx + next_entry;
-
- prefetch(p_next);
+ prefetch(priv->dma_rx + next_entry);
/* read the status of the incoming frame */
status = priv->hw->desc->rx_status(&priv->dev->stats,
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 6/7] stmmac: code tidy-up
From: Giuseppe CAVALLARO @ 2013-04-03 5:41 UTC (permalink / raw)
To: netdev; +Cc: Giuseppe Cavallaro
In-Reply-To: <1364967689-11155-1-git-send-email-peppe.cavallaro@st.com>
This patch tidies up the code. I have run Linden (and verified with checkpatch)
many part of the driver trying to reorganize some sections respecting the
codying-style rules in the points where it was not done.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/chain_mode.c | 12 +-
drivers/net/ethernet/stmicro/stmmac/common.h | 89 ++++---
drivers/net/ethernet/stmicro/stmmac/descs.h | 6 +-
drivers/net/ethernet/stmicro/stmmac/descs_com.h | 3 +-
drivers/net/ethernet/stmicro/stmmac/dwmac1000.h | 69 +++---
.../net/ethernet/stmicro/stmmac/dwmac1000_core.c | 70 +++---
.../net/ethernet/stmicro/stmmac/dwmac1000_dma.c | 23 +-
.../net/ethernet/stmicro/stmmac/dwmac100_core.c | 28 +-
drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c | 26 +-
drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h | 4 +-
drivers/net/ethernet/stmicro/stmmac/mmc.h | 3 +-
drivers/net/ethernet/stmicro/stmmac/norm_desc.c | 4 +-
drivers/net/ethernet/stmicro/stmmac/ring_mode.c | 2 +-
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 6 +-
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 275 ++++++++++----------
drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 2 +-
drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c | 2 +-
17 files changed, 313 insertions(+), 311 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
index 37a3f93..d234ab5 100644
--- a/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/chain_mode.c
@@ -30,7 +30,7 @@
static unsigned int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
{
- struct stmmac_priv *priv = (struct stmmac_priv *) p;
+ struct stmmac_priv *priv = (struct stmmac_priv *)p;
unsigned int txsize = priv->dma_tx_size;
unsigned int entry = priv->cur_tx % txsize;
struct dma_desc *desc = priv->dma_tx + entry;
@@ -103,7 +103,7 @@ static void stmmac_init_dma_chain(void *des, dma_addr_t phy_addr,
dma_addr_t dma_phy = phy_addr;
if (extend_desc) {
- struct dma_extended_desc *p = (struct dma_extended_desc *) des;
+ struct dma_extended_desc *p = (struct dma_extended_desc *)des;
for (i = 0; i < (size - 1); i++) {
dma_phy += sizeof(struct dma_extended_desc);
p->basic.des3 = (unsigned int)dma_phy;
@@ -112,7 +112,7 @@ static void stmmac_init_dma_chain(void *des, dma_addr_t phy_addr,
p->basic.des3 = (unsigned int)phy_addr;
} else {
- struct dma_desc *p = (struct dma_desc *) des;
+ struct dma_desc *p = (struct dma_desc *)des;
for (i = 0; i < (size - 1); i++) {
dma_phy += sizeof(struct dma_desc);
p->des3 = (unsigned int)dma_phy;
@@ -133,7 +133,7 @@ static void stmmac_refill_desc3(void *priv_ptr, struct dma_desc *p)
*/
p->des3 = (unsigned int)(priv->dma_rx_phy +
(((priv->dirty_rx) + 1) %
- priv->dma_rx_size) *
+ priv->dma_rx_size) *
sizeof(struct dma_desc));
}
@@ -148,8 +148,8 @@ static void stmmac_clean_desc3(void *priv_ptr, struct dma_desc *p)
*/
p->des3 = (unsigned int)(priv->dma_tx_phy +
(((priv->dirty_tx + 1) %
- priv->dma_tx_size) *
- sizeof(struct dma_desc)));
+ priv->dma_tx_size) *
+ sizeof(struct dma_desc)));
}
const struct stmmac_chain_mode_ops chain_mode_ops = {
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index ad7e20a..7788fbe 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -174,37 +174,37 @@ struct stmmac_extra_stats {
#define STMMAC_PCS_TBI (1 << 2)
#define STMMAC_PCS_RTBI (1 << 3)
-#define SF_DMA_MODE 1 /* DMA STORE-AND-FORWARD Operation Mode */
+#define SF_DMA_MODE 1 /* DMA STORE-AND-FORWARD Operation Mode */
/* DAM HW feature register fields */
-#define DMA_HW_FEAT_MIISEL 0x00000001 /* 10/100 Mbps Support */
-#define DMA_HW_FEAT_GMIISEL 0x00000002 /* 1000 Mbps Support */
-#define DMA_HW_FEAT_HDSEL 0x00000004 /* Half-Duplex Support */
-#define DMA_HW_FEAT_EXTHASHEN 0x00000008 /* Expanded DA Hash Filter */
-#define DMA_HW_FEAT_HASHSEL 0x00000010 /* HASH Filter */
-#define DMA_HW_FEAT_ADDMACADRSEL 0x00000020 /* Multiple MAC Addr Reg */
-#define DMA_HW_FEAT_PCSSEL 0x00000040 /* PCS registers */
-#define DMA_HW_FEAT_L3L4FLTREN 0x00000080 /* Layer 3 & Layer 4 Feature */
-#define DMA_HW_FEAT_SMASEL 0x00000100 /* SMA(MDIO) Interface */
-#define DMA_HW_FEAT_RWKSEL 0x00000200 /* PMT Remote Wakeup */
-#define DMA_HW_FEAT_MGKSEL 0x00000400 /* PMT Magic Packet */
-#define DMA_HW_FEAT_MMCSEL 0x00000800 /* RMON Module */
-#define DMA_HW_FEAT_TSVER1SEL 0x00001000 /* Only IEEE 1588-2002 Timestamp */
-#define DMA_HW_FEAT_TSVER2SEL 0x00002000 /* IEEE 1588-2008 Adv Timestamp */
-#define DMA_HW_FEAT_EEESEL 0x00004000 /* Energy Efficient Ethernet */
-#define DMA_HW_FEAT_AVSEL 0x00008000 /* AV Feature */
-#define DMA_HW_FEAT_TXCOESEL 0x00010000 /* Checksum Offload in Tx */
-#define DMA_HW_FEAT_RXTYP1COE 0x00020000 /* IP csum Offload(Type 1) in Rx */
-#define DMA_HW_FEAT_RXTYP2COE 0x00040000 /* IP csum Offload(Type 2) in Rx */
-#define DMA_HW_FEAT_RXFIFOSIZE 0x00080000 /* Rx FIFO > 2048 Bytes */
-#define DMA_HW_FEAT_RXCHCNT 0x00300000 /* No. of additional Rx Channels */
-#define DMA_HW_FEAT_TXCHCNT 0x00c00000 /* No. of additional Tx Channels */
-#define DMA_HW_FEAT_ENHDESSEL 0x01000000 /* Alternate (Enhanced Descriptor) */
-#define DMA_HW_FEAT_INTTSEN 0x02000000 /* Timestamping with Internal
- System Time */
-#define DMA_HW_FEAT_FLEXIPPSEN 0x04000000 /* Flexible PPS Output */
-#define DMA_HW_FEAT_SAVLANINS 0x08000000 /* Source Addr or VLAN Insertion */
-#define DMA_HW_FEAT_ACTPHYIF 0x70000000 /* Active/selected PHY interface */
+#define DMA_HW_FEAT_MIISEL 0x00000001 /* 10/100 Mbps Support */
+#define DMA_HW_FEAT_GMIISEL 0x00000002 /* 1000 Mbps Support */
+#define DMA_HW_FEAT_HDSEL 0x00000004 /* Half-Duplex Support */
+#define DMA_HW_FEAT_EXTHASHEN 0x00000008 /* Expanded DA Hash Filter */
+#define DMA_HW_FEAT_HASHSEL 0x00000010 /* HASH Filter */
+#define DMA_HW_FEAT_ADDMAC 0x00000020 /* Multiple MAC Addr Reg */
+#define DMA_HW_FEAT_PCSSEL 0x00000040 /* PCS registers */
+#define DMA_HW_FEAT_L3L4FLTREN 0x00000080 /* Layer 3 & Layer 4 Feature */
+#define DMA_HW_FEAT_SMASEL 0x00000100 /* SMA(MDIO) Interface */
+#define DMA_HW_FEAT_RWKSEL 0x00000200 /* PMT Remote Wakeup */
+#define DMA_HW_FEAT_MGKSEL 0x00000400 /* PMT Magic Packet */
+#define DMA_HW_FEAT_MMCSEL 0x00000800 /* RMON Module */
+#define DMA_HW_FEAT_TSVER1SEL 0x00001000 /* Only IEEE 1588-2002 */
+#define DMA_HW_FEAT_TSVER2SEL 0x00002000 /* IEEE 1588-2008 PTPv2 */
+#define DMA_HW_FEAT_EEESEL 0x00004000 /* Energy Efficient Ethernet */
+#define DMA_HW_FEAT_AVSEL 0x00008000 /* AV Feature */
+#define DMA_HW_FEAT_TXCOESEL 0x00010000 /* Checksum Offload in Tx */
+#define DMA_HW_FEAT_RXTYP1COE 0x00020000 /* IP COE (Type 1) in Rx */
+#define DMA_HW_FEAT_RXTYP2COE 0x00040000 /* IP COE (Type 2) in Rx */
+#define DMA_HW_FEAT_RXFIFOSIZE 0x00080000 /* Rx FIFO > 2048 Bytes */
+#define DMA_HW_FEAT_RXCHCNT 0x00300000 /* No. additional Rx Channels */
+#define DMA_HW_FEAT_TXCHCNT 0x00c00000 /* No. additional Tx Channels */
+#define DMA_HW_FEAT_ENHDESSEL 0x01000000 /* Alternate Descriptor */
+/* Timestamping with Internal System Time */
+#define DMA_HW_FEAT_INTTSEN 0x02000000
+#define DMA_HW_FEAT_FLEXIPPSEN 0x04000000 /* Flexible PPS Output */
+#define DMA_HW_FEAT_SAVLANINS 0x08000000 /* Source Addr or VLAN */
+#define DMA_HW_FEAT_ACTPHYIF 0x70000000 /* Active/selected PHY iface */
#define DEFAULT_DMA_PBL 8
/* Max/Min RI Watchdog Timer count value */
@@ -216,7 +216,8 @@ struct stmmac_extra_stats {
#define STMMAC_TX_MAX_FRAMES 256
#define STMMAC_TX_FRAMES 64
-enum rx_frame_status { /* IPC status */
+/* Rx IPC status */
+enum rx_frame_status {
good_frame = 0,
discard_frame = 1,
csum_none = 2,
@@ -261,9 +262,9 @@ struct dma_features {
unsigned int pmt_remote_wake_up;
unsigned int pmt_magic_frame;
unsigned int rmon;
- /* IEEE 1588-2002*/
+ /* IEEE 1588-2002 */
unsigned int time_stamp;
- /* IEEE 1588-2008*/
+ /* IEEE 1588-2008 */
unsigned int atime_stamp;
/* 802.3az - Energy-Efficient Ethernet (EEE) */
unsigned int eee;
@@ -276,7 +277,7 @@ struct dma_features {
/* TX and RX number of channels */
unsigned int number_rx_channel;
unsigned int number_tx_channel;
- /* Alternate (enhanced) DESC mode*/
+ /* Alternate (enhanced) DESC mode */
unsigned int enh_desc;
};
@@ -344,7 +345,7 @@ struct stmmac_desc_ops {
/* get tx timestamp status */
int (*get_tx_timestamp_status) (struct dma_desc *p);
/* get timestamp value */
- u64 (*get_timestamp) (void *desc, u32 ats);
+ u64(*get_timestamp) (void *desc, u32 ats);
/* get rx timestamp status */
int (*get_rx_timestamp_status) (void *desc, u32 ats);
};
@@ -378,7 +379,7 @@ struct stmmac_dma_ops {
struct stmmac_ops {
/* MAC core initialization */
- void (*core_init) (void __iomem *ioaddr) ____cacheline_aligned;
+ void (*core_init) (void __iomem *ioaddr);
/* Enable and verify that the IPC module is supported */
int (*rx_ipc) (void __iomem *ioaddr);
/* Dump MAC registers */
@@ -410,10 +411,10 @@ struct stmmac_hwtimestamp {
void (*config_hw_tstamping) (void __iomem *ioaddr, u32 data);
void (*config_sub_second_increment) (void __iomem *ioaddr);
int (*init_systime) (void __iomem *ioaddr, u32 sec, u32 nsec);
- int (*config_addend)(void __iomem *ioaddr, u32 addend);
- int (*adjust_systime)(void __iomem *ioaddr, u32 sec, u32 nsec,
- int add_sub);
- u64 (*get_systime)(void __iomem *ioaddr);
+ int (*config_addend) (void __iomem *ioaddr, u32 addend);
+ int (*adjust_systime) (void __iomem *ioaddr, u32 sec, u32 nsec,
+ int add_sub);
+ u64(*get_systime) (void __iomem *ioaddr);
};
struct mac_link {
@@ -446,11 +447,11 @@ struct stmmac_chain_mode_ops {
};
struct mac_device_info {
- const struct stmmac_ops *mac;
- const struct stmmac_desc_ops *desc;
- const struct stmmac_dma_ops *dma;
- const struct stmmac_ring_mode_ops *ring;
- const struct stmmac_chain_mode_ops *chain;
+ const struct stmmac_ops *mac;
+ const struct stmmac_desc_ops *desc;
+ const struct stmmac_dma_ops *dma;
+ const struct stmmac_ring_mode_ops *ring;
+ const struct stmmac_chain_mode_ops *chain;
const struct stmmac_hwtimestamp *ptp;
struct mii_regs mii; /* MII register Addresses */
struct mac_link link;
diff --git a/drivers/net/ethernet/stmicro/stmmac/descs.h b/drivers/net/ethernet/stmicro/stmmac/descs.h
index 2eca0c0..ad39960 100644
--- a/drivers/net/ethernet/stmicro/stmmac/descs.h
+++ b/drivers/net/ethernet/stmicro/stmmac/descs.h
@@ -192,9 +192,9 @@ struct dma_extended_desc {
u32 reserved;
} etx;
} des4;
- unsigned int des5; /* Reserved */
- unsigned int des6; /* Tx/Rx Timestamp Low */
- unsigned int des7; /* Tx/Rx Timestamp High */
+ unsigned int des5; /* Reserved */
+ unsigned int des6; /* Tx/Rx Timestamp Low */
+ unsigned int des7; /* Tx/Rx Timestamp High */
};
/* Transmit checksum insertion control */
diff --git a/drivers/net/ethernet/stmicro/stmmac/descs_com.h b/drivers/net/ethernet/stmicro/stmmac/descs_com.h
index 20f83fc..6f2cc78 100644
--- a/drivers/net/ethernet/stmicro/stmmac/descs_com.h
+++ b/drivers/net/ethernet/stmicro/stmmac/descs_com.h
@@ -117,8 +117,7 @@ static inline void ndesc_rx_set_on_chain(struct dma_desc *p, int end)
p->des01.rx.second_address_chained = 1;
}
-static inline void ndesc_tx_set_on_chain(struct dma_desc *p, int
- ring_size)
+static inline void ndesc_tx_set_on_chain(struct dma_desc *p, int ring_size)
{
p->des01.tx.second_address_chained = 1;
}
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
index 57f4e8f..c12aabb 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
@@ -99,18 +99,18 @@ enum power_event {
#define GMAC_S_R_GMII 0x000000d8 /* SGMII RGMII status */
/* AN Configuration defines */
-#define GMAC_AN_CTRL_RAN 0x00000200 /* Restart Auto-Negotiation */
-#define GMAC_AN_CTRL_ANE 0x00001000 /* Auto-Negotiation Enable */
-#define GMAC_AN_CTRL_ELE 0x00004000 /* External Loopback Enable */
-#define GMAC_AN_CTRL_ECD 0x00010000 /* Enable Comma Detect */
-#define GMAC_AN_CTRL_LR 0x00020000 /* Lock to Reference */
-#define GMAC_AN_CTRL_SGMRAL 0x00040000 /* SGMII RAL Control */
+#define GMAC_AN_CTRL_RAN 0x00000200 /* Restart Auto-Negotiation */
+#define GMAC_AN_CTRL_ANE 0x00001000 /* Auto-Negotiation Enable */
+#define GMAC_AN_CTRL_ELE 0x00004000 /* External Loopback Enable */
+#define GMAC_AN_CTRL_ECD 0x00010000 /* Enable Comma Detect */
+#define GMAC_AN_CTRL_LR 0x00020000 /* Lock to Reference */
+#define GMAC_AN_CTRL_SGMRAL 0x00040000 /* SGMII RAL Control */
/* AN Status defines */
-#define GMAC_AN_STATUS_LS 0x00000004 /* Link Status 0:down 1:up */
-#define GMAC_AN_STATUS_ANA 0x00000008 /* Auto-Negotiation Ability */
-#define GMAC_AN_STATUS_ANC 0x00000020 /* Auto-Negotiation Complete */
-#define GMAC_AN_STATUS_ES 0x00000100 /* Extended Status */
+#define GMAC_AN_STATUS_LS 0x00000004 /* Link Status 0:down 1:up */
+#define GMAC_AN_STATUS_ANA 0x00000008 /* Auto-Negotiation Ability */
+#define GMAC_AN_STATUS_ANC 0x00000020 /* Auto-Negotiation Complete */
+#define GMAC_AN_STATUS_ES 0x00000100 /* Extended Status */
/* Register 54 (SGMII/RGMII status register) */
#define GMAC_S_R_GMII_LINK 0x8
@@ -127,8 +127,8 @@ enum power_event {
#define GMAC_ANE_PSE_SHIFT 7
/* GMAC Configuration defines */
-#define GMAC_CONTROL_TC 0x01000000 /* Transmit Conf. in RGMII/SGMII */
-#define GMAC_CONTROL_WD 0x00800000 /* Disable Watchdog on receive */
+#define GMAC_CONTROL_TC 0x01000000 /* Transmit Conf. in RGMII/SGMII */
+#define GMAC_CONTROL_WD 0x00800000 /* Disable Watchdog on receive */
/* GMAC Configuration defines */
#define GMAC_CONTROL_TC 0x01000000 /* Transmit Conf. in RGMII/SGMII */
@@ -141,19 +141,19 @@ enum inter_frame_gap {
GMAC_CONTROL_IFG_80 = 0x00020000,
GMAC_CONTROL_IFG_40 = 0x000e0000,
};
-#define GMAC_CONTROL_DCRS 0x00010000 /* Disable carrier sense during tx */
-#define GMAC_CONTROL_PS 0x00008000 /* Port Select 0:GMI 1:MII */
-#define GMAC_CONTROL_FES 0x00004000 /* Speed 0:10 1:100 */
-#define GMAC_CONTROL_DO 0x00002000 /* Disable Rx Own */
-#define GMAC_CONTROL_LM 0x00001000 /* Loop-back mode */
-#define GMAC_CONTROL_DM 0x00000800 /* Duplex Mode */
-#define GMAC_CONTROL_IPC 0x00000400 /* Checksum Offload */
-#define GMAC_CONTROL_DR 0x00000200 /* Disable Retry */
-#define GMAC_CONTROL_LUD 0x00000100 /* Link up/down */
-#define GMAC_CONTROL_ACS 0x00000080 /* Automatic Pad/FCS Stripping */
-#define GMAC_CONTROL_DC 0x00000010 /* Deferral Check */
-#define GMAC_CONTROL_TE 0x00000008 /* Transmitter Enable */
-#define GMAC_CONTROL_RE 0x00000004 /* Receiver Enable */
+#define GMAC_CONTROL_DCRS 0x00010000 /* Disable carrier sense */
+#define GMAC_CONTROL_PS 0x00008000 /* Port Select 0:GMI 1:MII */
+#define GMAC_CONTROL_FES 0x00004000 /* Speed 0:10 1:100 */
+#define GMAC_CONTROL_DO 0x00002000 /* Disable Rx Own */
+#define GMAC_CONTROL_LM 0x00001000 /* Loop-back mode */
+#define GMAC_CONTROL_DM 0x00000800 /* Duplex Mode */
+#define GMAC_CONTROL_IPC 0x00000400 /* Checksum Offload */
+#define GMAC_CONTROL_DR 0x00000200 /* Disable Retry */
+#define GMAC_CONTROL_LUD 0x00000100 /* Link up/down */
+#define GMAC_CONTROL_ACS 0x00000080 /* Auto Pad/FCS Stripping */
+#define GMAC_CONTROL_DC 0x00000010 /* Deferral Check */
+#define GMAC_CONTROL_TE 0x00000008 /* Transmitter Enable */
+#define GMAC_CONTROL_RE 0x00000004 /* Receiver Enable */
#define GMAC_CORE_INIT (GMAC_CONTROL_JD | GMAC_CONTROL_PS | GMAC_CONTROL_ACS | \
GMAC_CONTROL_JE | GMAC_CONTROL_BE)
@@ -184,16 +184,16 @@ enum inter_frame_gap {
#define DMA_BUS_MODE_SFT_RESET 0x00000001 /* Software Reset */
#define DMA_BUS_MODE_DA 0x00000002 /* Arbitration scheme */
#define DMA_BUS_MODE_DSL_MASK 0x0000007c /* Descriptor Skip Length */
-#define DMA_BUS_MODE_DSL_SHIFT 2 /* (in DWORDS) */
+#define DMA_BUS_MODE_DSL_SHIFT 2 /* (in DWORDS) */
/* Programmable burst length (passed thorugh platform)*/
#define DMA_BUS_MODE_PBL_MASK 0x00003f00 /* Programmable Burst Len */
#define DMA_BUS_MODE_PBL_SHIFT 8
#define DMA_BUS_MODE_ATDS 0x00000080 /* Alternate Descriptor Size */
enum rx_tx_priority_ratio {
- double_ratio = 0x00004000, /*2:1 */
- triple_ratio = 0x00008000, /*3:1 */
- quadruple_ratio = 0x0000c000, /*4:1 */
+ double_ratio = 0x00004000, /* 2:1 */
+ triple_ratio = 0x00008000, /* 3:1 */
+ quadruple_ratio = 0x0000c000, /* 4:1 */
};
#define DMA_BUS_MODE_FB 0x00010000 /* Fixed burst */
@@ -213,9 +213,10 @@ enum rx_tx_priority_ratio {
#define DMA_BUS_FB 0x00010000 /* Fixed Burst */
/* DMA operation mode defines (start/stop tx/rx are placed in common header)*/
-#define DMA_CONTROL_DT 0x04000000 /* Disable Drop TCP/IP csum error */
-#define DMA_CONTROL_RSF 0x02000000 /* Receive Store and Forward */
-#define DMA_CONTROL_DFF 0x01000000 /* Disaable flushing */
+/* Disable Drop TCP/IP csum error */
+#define DMA_CONTROL_DT 0x04000000
+#define DMA_CONTROL_RSF 0x02000000 /* Receive Store and Forward */
+#define DMA_CONTROL_DFF 0x01000000 /* Disaable flushing */
/* Threshold for Activating the FC */
enum rfa {
act_full_minus_1 = 0x00800000,
@@ -230,7 +231,7 @@ enum rfd {
deac_full_minus_3 = 0x00401000,
deac_full_minus_4 = 0x00401800,
};
-#define DMA_CONTROL_TSF 0x00200000 /* Transmit Store and Forward */
+#define DMA_CONTROL_TSF 0x00200000 /* Transmit Store and Forward */
enum ttc_control {
DMA_CONTROL_TTC_64 = 0x00000000,
@@ -264,7 +265,5 @@ enum rtc_control {
#define GMAC_MMC_TX_INTR 0x108
#define GMAC_MMC_RX_CSUM_OFFLOAD 0x208
-
-
extern const struct stmmac_dma_ops dwmac1000_dma_ops;
#endif /* __DWMAC1000_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
index 29138da..7e05e8d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
@@ -72,22 +72,22 @@ static void dwmac1000_dump_regs(void __iomem *ioaddr)
}
static void dwmac1000_set_umac_addr(void __iomem *ioaddr, unsigned char *addr,
- unsigned int reg_n)
+ unsigned int reg_n)
{
stmmac_set_mac_addr(ioaddr, addr, GMAC_ADDR_HIGH(reg_n),
- GMAC_ADDR_LOW(reg_n));
+ GMAC_ADDR_LOW(reg_n));
}
static void dwmac1000_get_umac_addr(void __iomem *ioaddr, unsigned char *addr,
- unsigned int reg_n)
+ unsigned int reg_n)
{
stmmac_get_mac_addr(ioaddr, addr, GMAC_ADDR_HIGH(reg_n),
- GMAC_ADDR_LOW(reg_n));
+ GMAC_ADDR_LOW(reg_n));
}
static void dwmac1000_set_filter(struct net_device *dev, int id)
{
- void __iomem *ioaddr = (void __iomem *) dev->base_addr;
+ void __iomem *ioaddr = (void __iomem *)dev->base_addr;
unsigned int value = 0;
unsigned int perfect_addr_number;
@@ -97,7 +97,7 @@ static void dwmac1000_set_filter(struct net_device *dev, int id)
if (dev->flags & IFF_PROMISC)
value = GMAC_FRAME_FILTER_PR;
else if ((netdev_mc_count(dev) > HASH_TABLE_SIZE)
- || (dev->flags & IFF_ALLMULTI)) {
+ || (dev->flags & IFF_ALLMULTI)) {
value = GMAC_FRAME_FILTER_PM; /* pass all multi */
writel(0xffffffff, ioaddr + GMAC_HASH_HIGH);
writel(0xffffffff, ioaddr + GMAC_HASH_LOW);
@@ -111,12 +111,13 @@ static void dwmac1000_set_filter(struct net_device *dev, int id)
memset(mc_filter, 0, sizeof(mc_filter));
netdev_for_each_mc_addr(ha, dev) {
/* The upper 6 bits of the calculated CRC are used to
- index the contens of the hash table */
- int bit_nr =
- bitrev32(~crc32_le(~0, ha->addr, 6)) >> 26;
+ * index the contens of the hash table
+ */
+ int bit_nr = bitrev32(~crc32_le(~0, ha->addr, 6)) >> 26;
/* The most significant bit determines the register to
* use (H/L) while the other 5 bits determine the bit
- * within the register. */
+ * within the register.
+ */
mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31);
}
writel(mc_filter[0], ioaddr + GMAC_HASH_LOW);
@@ -129,10 +130,11 @@ static void dwmac1000_set_filter(struct net_device *dev, int id)
else
perfect_addr_number = GMAC_MAX_PERFECT_ADDRESSES / 2;
- /* Handle multiple unicast addresses (perfect filtering)*/
+ /* Handle multiple unicast addresses (perfect filtering) */
if (netdev_uc_count(dev) > perfect_addr_number)
- /* Switch to promiscuous mode is more than 16 addrs
- are required */
+ /* Switch to promiscuous mode if more than 16 addrs
+ * are required
+ */
value |= GMAC_FRAME_FILTER_PR;
else {
int reg = 1;
@@ -150,13 +152,13 @@ static void dwmac1000_set_filter(struct net_device *dev, int id)
#endif
writel(value, ioaddr + GMAC_FRAME_FILTER);
- CHIP_DBG(KERN_INFO "\tFrame Filter reg: 0x%08x\n\tHash regs: "
- "HI 0x%08x, LO 0x%08x\n", readl(ioaddr + GMAC_FRAME_FILTER),
- readl(ioaddr + GMAC_HASH_HIGH), readl(ioaddr + GMAC_HASH_LOW));
+ CHIP_DBG(KERN_INFO "\tFilter: 0x%08x\n\tHash: HI 0x%08x, LO 0x%08x\n",
+ readl(ioaddr + GMAC_FRAME_FILTER),
+ readl(ioaddr + GMAC_HASH_HIGH), readl(ioaddr + GMAC_HASH_LOW));
}
static void dwmac1000_flow_ctrl(void __iomem *ioaddr, unsigned int duplex,
- unsigned int fc, unsigned int pause_time)
+ unsigned int fc, unsigned int pause_time)
{
unsigned int flow = 0;
@@ -203,23 +205,22 @@ static int dwmac1000_irq_status(void __iomem *ioaddr,
/* Not used events (e.g. MMC interrupts) are not handled. */
if ((intr_status & mmc_tx_irq)) {
CHIP_DBG(KERN_INFO "GMAC: MMC tx interrupt: 0x%08x\n",
- readl(ioaddr + GMAC_MMC_TX_INTR));
+ readl(ioaddr + GMAC_MMC_TX_INTR));
x->mmc_tx_irq_n++;
}
if (unlikely(intr_status & mmc_rx_irq)) {
CHIP_DBG(KERN_INFO "GMAC: MMC rx interrupt: 0x%08x\n",
- readl(ioaddr + GMAC_MMC_RX_INTR));
+ readl(ioaddr + GMAC_MMC_RX_INTR));
x->mmc_rx_irq_n++;
}
if (unlikely(intr_status & mmc_rx_csum_offload_irq)) {
CHIP_DBG(KERN_INFO "GMAC: MMC rx csum offload: 0x%08x\n",
- readl(ioaddr + GMAC_MMC_RX_CSUM_OFFLOAD));
+ readl(ioaddr + GMAC_MMC_RX_CSUM_OFFLOAD));
x->mmc_rx_csum_offload_irq_n++;
}
if (unlikely(intr_status & pmt_irq)) {
CHIP_DBG(KERN_INFO "GMAC: received Magic frame\n");
- /* clear the PMT bits 5 and 6 by reading the PMT
- * status register. */
+ /* clear the PMT bits 5 and 6 by reading the PMT status reg */
readl(ioaddr + GMAC_PMT);
x->irq_receive_pmt_irq_n++;
}
@@ -252,14 +253,14 @@ static int dwmac1000_irq_status(void __iomem *ioaddr,
x->irq_pcs_ane_n++;
}
if (intr_status & rgmii_irq) {
- u32 status = readl(ioaddr + GMAC_S_R_GMII);
+ u32 status = readl(ioaddr + GMAC_S_R_GMII);
CHIP_DBG(KERN_INFO "GMAC RGMII/SGMII interrupt\n");
x->irq_rgmii_n++;
/* Save and dump the link status. */
if (status & GMAC_S_R_GMII_LINK) {
int speed_value = (status & GMAC_S_R_GMII_SPEED) >>
- GMAC_S_R_GMII_SPEED_SHIFT;
+ GMAC_S_R_GMII_SPEED_SHIFT;
x->pcs_duplex = (status & GMAC_S_R_GMII_MODE);
if (speed_value == GMAC_S_R_GMII_SPEED_125)
@@ -270,7 +271,7 @@ static int dwmac1000_irq_status(void __iomem *ioaddr,
x->pcs_speed = SPEED_10;
x->pcs_link = 1;
- pr_debug("Link is Up - %d/%s\n", (int) x->pcs_speed,
+ pr_debug("Link is Up - %d/%s\n", (int)x->pcs_speed,
x->pcs_duplex ? "Full" : "Half");
} else {
x->pcs_link = 0;
@@ -281,19 +282,20 @@ static int dwmac1000_irq_status(void __iomem *ioaddr,
return ret;
}
-static void dwmac1000_set_eee_mode(void __iomem *ioaddr)
+static void dwmac1000_set_eee_mode(void __iomem *ioaddr)
{
u32 value;
/* Enable the link status receive on RGMII, SGMII ore SMII
* receive path and instruct the transmit to enter in LPI
- * state. */
+ * state.
+ */
value = readl(ioaddr + LPI_CTRL_STATUS);
value |= LPI_CTRL_STATUS_LPIEN | LPI_CTRL_STATUS_LPITXA;
writel(value, ioaddr + LPI_CTRL_STATUS);
}
-static void dwmac1000_reset_eee_mode(void __iomem *ioaddr)
+static void dwmac1000_reset_eee_mode(void __iomem *ioaddr)
{
u32 value;
@@ -302,7 +304,7 @@ static void dwmac1000_reset_eee_mode(void __iomem *ioaddr)
writel(value, ioaddr + LPI_CTRL_STATUS);
}
-static void dwmac1000_set_eee_pls(void __iomem *ioaddr, int link)
+static void dwmac1000_set_eee_pls(void __iomem *ioaddr, int link)
{
u32 value;
@@ -316,7 +318,7 @@ static void dwmac1000_set_eee_pls(void __iomem *ioaddr, int link)
writel(value, ioaddr + LPI_CTRL_STATUS);
}
-static void dwmac1000_set_eee_timer(void __iomem *ioaddr, int ls, int tw)
+static void dwmac1000_set_eee_timer(void __iomem *ioaddr, int ls, int tw)
{
int value = ((tw & 0xffff)) | ((ls & 0x7ff) << 16);
@@ -375,10 +377,10 @@ static const struct stmmac_ops dwmac1000_ops = {
.pmt = dwmac1000_pmt,
.set_umac_addr = dwmac1000_set_umac_addr,
.get_umac_addr = dwmac1000_get_umac_addr,
- .set_eee_mode = dwmac1000_set_eee_mode,
- .reset_eee_mode = dwmac1000_reset_eee_mode,
- .set_eee_timer = dwmac1000_set_eee_timer,
- .set_eee_pls = dwmac1000_set_eee_pls,
+ .set_eee_mode = dwmac1000_set_eee_mode,
+ .reset_eee_mode = dwmac1000_reset_eee_mode,
+ .set_eee_timer = dwmac1000_set_eee_timer,
+ .set_eee_pls = dwmac1000_set_eee_pls,
.ctrl_ane = dwmac1000_ctrl_ane,
.get_adv = dwmac1000_get_adv,
};
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
index f1c4b2c..2c431b6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
@@ -60,7 +60,7 @@ static int dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
* depending on pbl value.
*/
value = DMA_BUS_MODE_PBL | ((pbl << DMA_BUS_MODE_PBL_SHIFT) |
- (pbl << DMA_BUS_MODE_RPBL_SHIFT));
+ (pbl << DMA_BUS_MODE_RPBL_SHIFT));
/* Set the Fixed burst mode */
if (fb)
@@ -94,14 +94,16 @@ static int dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
*
* For Non Fixed Burst Mode: provide the maximum value of the
* burst length. Any burst equal or below the provided burst
- * length would be allowed to perform. */
+ * length would be allowed to perform.
+ */
writel(burst_len, ioaddr + DMA_AXI_BUS_MODE);
/* Mask interrupts by writing to CSR7 */
writel(DMA_INTR_DEFAULT_MASK, ioaddr + DMA_INTR_ENA);
- /* The base address of the RX/TX descriptor lists must be written into
- * DMA CSR3 and CSR4, respectively. */
+ /* RX/TX descriptor base address lists must be written into
+ * DMA CSR3 and CSR4, respectively
+ */
writel(dma_tx, ioaddr + DMA_TX_BASE_ADDR);
writel(dma_rx, ioaddr + DMA_RCV_BASE_ADDR);
@@ -109,7 +111,7 @@ static int dwmac1000_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
}
static void dwmac1000_dma_operation_mode(void __iomem *ioaddr, int txmode,
- int rxmode)
+ int rxmode)
{
u32 csr6 = readl(ioaddr + DMA_CONTROL);
@@ -118,11 +120,12 @@ static void dwmac1000_dma_operation_mode(void __iomem *ioaddr, int txmode,
/* Transmit COE type 2 cannot be done in cut-through mode. */
csr6 |= DMA_CONTROL_TSF;
/* Operating on second frame increase the performance
- * especially when transmit store-and-forward is used.*/
+ * especially when transmit store-and-forward is used.
+ */
csr6 |= DMA_CONTROL_OSF;
} else {
- CHIP_DBG(KERN_DEBUG "GMAC: disabling TX store and forward mode"
- " (threshold = %d)\n", txmode);
+ CHIP_DBG(KERN_DEBUG "GMAC: disabling TX SF (threshold %d)\n",
+ txmode);
csr6 &= ~DMA_CONTROL_TSF;
csr6 &= DMA_CONTROL_TC_TX_MASK;
/* Set the transmit threshold */
@@ -142,8 +145,8 @@ static void dwmac1000_dma_operation_mode(void __iomem *ioaddr, int txmode,
CHIP_DBG(KERN_DEBUG "GMAC: enable RX store and forward mode\n");
csr6 |= DMA_CONTROL_RSF;
} else {
- CHIP_DBG(KERN_DEBUG "GMAC: disabling RX store and forward mode"
- " (threshold = %d)\n", rxmode);
+ CHIP_DBG(KERN_DEBUG "GMAC: disable RX SF mode (threshold %d)\n",
+ rxmode);
csr6 &= ~DMA_CONTROL_RSF;
csr6 &= DMA_CONTROL_TC_RX_MASK;
if (rxmode <= 32)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c
index cb86a58..007bb2b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c
@@ -47,8 +47,7 @@ static void dwmac100_dump_mac_regs(void __iomem *ioaddr)
{
pr_info("\t----------------------------------------------\n"
"\t DWMAC 100 CSR (base addr = 0x%p)\n"
- "\t----------------------------------------------\n",
- ioaddr);
+ "\t----------------------------------------------\n", ioaddr);
pr_info("\tcontrol reg (offset 0x%x): 0x%08x\n", MAC_CONTROL,
readl(ioaddr + MAC_CONTROL));
pr_info("\taddr HI (offset 0x%x): 0x%08x\n ", MAC_ADDR_HIGH,
@@ -92,7 +91,7 @@ static void dwmac100_get_umac_addr(void __iomem *ioaddr, unsigned char *addr,
static void dwmac100_set_filter(struct net_device *dev, int id)
{
- void __iomem *ioaddr = (void __iomem *) dev->base_addr;
+ void __iomem *ioaddr = (void __iomem *)dev->base_addr;
u32 value = readl(ioaddr + MAC_CONTROL);
if (dev->flags & IFF_PROMISC) {
@@ -113,7 +112,8 @@ static void dwmac100_set_filter(struct net_device *dev, int id)
struct netdev_hw_addr *ha;
/* Perfect filter mode for physical address and Hash
- filter for multicast */
+ * filter for multicast
+ */
value |= MAC_CONTROL_HP;
value &= ~(MAC_CONTROL_PM | MAC_CONTROL_PR |
MAC_CONTROL_IF | MAC_CONTROL_HO);
@@ -121,12 +121,13 @@ static void dwmac100_set_filter(struct net_device *dev, int id)
memset(mc_filter, 0, sizeof(mc_filter));
netdev_for_each_mc_addr(ha, dev) {
/* The upper 6 bits of the calculated CRC are used to
- * index the contens of the hash table */
- int bit_nr =
- ether_crc(ETH_ALEN, ha->addr) >> 26;
+ * index the contens of the hash table
+ */
+ int bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26;
/* The most significant bit determines the register to
* use (H/L) while the other 5 bits determine the bit
- * within the register. */
+ * within the register.
+ */
mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31);
}
writel(mc_filter[0], ioaddr + MAC_HASH_LOW);
@@ -135,10 +136,9 @@ static void dwmac100_set_filter(struct net_device *dev, int id)
writel(value, ioaddr + MAC_CONTROL);
- CHIP_DBG(KERN_INFO "%s: CTRL reg: 0x%08x Hash regs: "
- "HI 0x%08x, LO 0x%08x\n",
- __func__, readl(ioaddr + MAC_CONTROL),
- readl(ioaddr + MAC_HASH_HIGH), readl(ioaddr + MAC_HASH_LOW));
+ CHIP_DBG(KERN_INFO "%s: Filter: 0x%08x Hash: HI 0x%08x, LO 0x%08x\n",
+ __func__, readl(ioaddr + MAC_CONTROL),
+ readl(ioaddr + MAC_HASH_HIGH), readl(ioaddr + MAC_HASH_LOW));
}
static void dwmac100_flow_ctrl(void __iomem *ioaddr, unsigned int duplex,
@@ -151,9 +151,7 @@ static void dwmac100_flow_ctrl(void __iomem *ioaddr, unsigned int duplex,
writel(flow, ioaddr + MAC_FLOW_CTRL);
}
-/* No PMT module supported for this Ethernet Controller.
- * Tested on ST platforms only.
- */
+/* No PMT module supported on ST boards with this Eth chip. */
static void dwmac100_pmt(void __iomem *ioaddr, unsigned long mode)
{
return;
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
index e979a8b..67551c1 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
@@ -52,22 +52,25 @@ static int dwmac100_dma_init(void __iomem *ioaddr, int pbl, int fb, int mb,
/* Enable Application Access by writing to DMA CSR0 */
writel(DMA_BUS_MODE_DEFAULT | (pbl << DMA_BUS_MODE_PBL_SHIFT),
- ioaddr + DMA_BUS_MODE);
+ ioaddr + DMA_BUS_MODE);
/* Mask interrupts by writing to CSR7 */
writel(DMA_INTR_DEFAULT_MASK, ioaddr + DMA_INTR_ENA);
- /* The base address of the RX/TX descriptor lists must be written into
- * DMA CSR3 and CSR4, respectively. */
+ /* RX/TX descriptor base addr lists must be written into
+ * DMA CSR3 and CSR4, respectively
+ */
writel(dma_tx, ioaddr + DMA_TX_BASE_ADDR);
writel(dma_rx, ioaddr + DMA_RCV_BASE_ADDR);
return 0;
}
-/* Store and Forward capability is not used at all..
- * The transmit threshold can be programmed by
- * setting the TTC bits in the DMA control register.*/
+/* Store and Forward capability is not used at all.
+ *
+ * The transmit threshold can be programmed by setting the TTC bits in the DMA
+ * control register.
+ */
static void dwmac100_dma_operation_mode(void __iomem *ioaddr, int txmode,
int rxmode)
{
@@ -90,16 +93,15 @@ static void dwmac100_dump_dma_regs(void __iomem *ioaddr)
CHIP_DBG(KERN_DEBUG "DWMAC 100 DMA CSR\n");
for (i = 0; i < 9; i++)
pr_debug("\t CSR%d (offset 0x%x): 0x%08x\n", i,
- (DMA_BUS_MODE + i * 4),
- readl(ioaddr + DMA_BUS_MODE + i * 4));
+ (DMA_BUS_MODE + i * 4),
+ readl(ioaddr + DMA_BUS_MODE + i * 4));
CHIP_DBG(KERN_DEBUG "\t CSR20 (offset 0x%x): 0x%08x\n",
- DMA_CUR_TX_BUF_ADDR, readl(ioaddr + DMA_CUR_TX_BUF_ADDR));
+ DMA_CUR_TX_BUF_ADDR, readl(ioaddr + DMA_CUR_TX_BUF_ADDR));
CHIP_DBG(KERN_DEBUG "\t CSR21 (offset 0x%x): 0x%08x\n",
- DMA_CUR_RX_BUF_ADDR, readl(ioaddr + DMA_CUR_RX_BUF_ADDR));
+ DMA_CUR_RX_BUF_ADDR, readl(ioaddr + DMA_CUR_RX_BUF_ADDR));
}
-/* DMA controller has two counters to track the number of
- * the receive missed frames. */
+/* DMA controller has two counters to track the number of the missed frames. */
static void dwmac100_dma_diagnostic_fr(void *data, struct stmmac_extra_stats *x,
void __iomem *ioaddr)
{
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
index ab4896e..8e5662c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
@@ -102,7 +102,7 @@
#define DMA_STATUS_TU 0x00000004 /* Transmit Buffer Unavailable */
#define DMA_STATUS_TPS 0x00000002 /* Transmit Process Stopped */
#define DMA_STATUS_TI 0x00000001 /* Transmit Interrupt */
-#define DMA_CONTROL_FTF 0x00100000 /* Flush transmit FIFO */
+#define DMA_CONTROL_FTF 0x00100000 /* Flush transmit FIFO */
extern void dwmac_enable_dma_transmission(void __iomem *ioaddr);
extern void dwmac_enable_dma_irq(void __iomem *ioaddr);
@@ -112,6 +112,6 @@ extern void dwmac_dma_stop_tx(void __iomem *ioaddr);
extern void dwmac_dma_start_rx(void __iomem *ioaddr);
extern void dwmac_dma_stop_rx(void __iomem *ioaddr);
extern int dwmac_dma_interrupt(void __iomem *ioaddr,
- struct stmmac_extra_stats *x);
+ struct stmmac_extra_stats *x);
#endif /* __DWMAC_DMA_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/mmc.h b/drivers/net/ethernet/stmicro/stmmac/mmc.h
index 67995ef..48ec001 100644
--- a/drivers/net/ethernet/stmicro/stmmac/mmc.h
+++ b/drivers/net/ethernet/stmicro/stmmac/mmc.h
@@ -28,8 +28,7 @@
/* MMC control register */
/* When set, all counter are reset */
#define MMC_CNTRL_COUNTER_RESET 0x1
-/* When set, do not roll over zero
- * after reaching the max value*/
+/* When set, do not roll over zero after reaching the max value*/
#define MMC_CNTRL_COUNTER_STOP_ROLLOVER 0x2
#define MMC_CNTRL_RESET_ON_READ 0x4 /* Reset after reading */
#define MMC_CNTRL_COUNTER_FREEZER 0x8 /* Freeze counter values to the
diff --git a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
index 7cbcea3..11775b9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/norm_desc.c
@@ -79,8 +79,8 @@ static int ndesc_get_rx_status(void *data, struct stmmac_extra_stats *x,
struct net_device_stats *stats = (struct net_device_stats *)data;
if (unlikely(p->des01.rx.last_descriptor == 0)) {
- pr_warning("ndesc Error: Oversized Ethernet "
- "frame spanned multiple buffers\n");
+ pr_warn("%s: Oversized frame spanned multiple buffers\n",
+ __func__);
stats->rx_length_errors++;
return discard_frame;
}
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index d0265a7..c9d942a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -30,7 +30,7 @@
static unsigned int stmmac_jumbo_frm(void *p, struct sk_buff *skb, int csum)
{
- struct stmmac_priv *priv = (struct stmmac_priv *) p;
+ struct stmmac_priv *priv = (struct stmmac_priv *)p;
unsigned int txsize = priv->dma_tx_size;
unsigned int entry = priv->cur_tx % txsize;
struct dma_desc *desc = priv->dma_tx + entry;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 8aa28c5..1a3c412 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -52,7 +52,7 @@ struct stmmac_priv {
bool tx_path_in_lpi_mode;
struct timer_list txtimer;
- struct dma_desc *dma_rx ____cacheline_aligned_in_smp;
+ struct dma_desc *dma_rx ____cacheline_aligned_in_smp;
struct dma_extended_desc *dma_erx ____cacheline_aligned_in_smp;
struct sk_buff **rx_skbuff ____cacheline_aligned_in_smp;
unsigned int cur_rx;
@@ -142,6 +142,7 @@ static inline int stmmac_register_platform(void)
return err;
}
+
static inline void stmmac_unregister_platform(void)
{
platform_driver_unregister(&stmmac_pltfr_driver);
@@ -153,6 +154,7 @@ static inline int stmmac_register_platform(void)
return 0;
}
+
static inline void stmmac_unregister_platform(void)
{
}
@@ -170,6 +172,7 @@ static inline int stmmac_register_pci(void)
return err;
}
+
static inline void stmmac_unregister_pci(void)
{
pci_unregister_driver(&stmmac_pci_driver);
@@ -181,6 +184,7 @@ static inline int stmmac_register_pci(void)
return 0;
}
+
static inline void stmmac_unregister_pci(void)
{
}
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 52000b8..16f5d73 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -46,7 +46,7 @@
#ifdef CONFIG_STMMAC_DEBUG_FS
#include <linux/debugfs.h>
#include <linux/seq_file.h>
-#endif
+#endif /* CONFIG_STMMAC_DEBUG_FS */
#include <linux/net_tstamp.h>
#include "stmmac_ptp.h"
#include "stmmac.h"
@@ -192,7 +192,12 @@ static void stmmac_clk_csr_set(struct stmmac_priv *priv)
clk_rate = clk_get_rate(priv->stmmac_clk);
/* Platform provided default clk_csr would be assumed valid
- * for all other cases except for the below mentioned ones. */
+ * for all other cases except for the below mentioned ones.
+ * For values higher than the IEEE 802.3 specified frequency
+ * we can not estimate the proper divider as it is not known
+ * the frequency of clk_csr_i. So we do not change the default
+ * divider.
+ */
if (!(priv->clk_csr & MAC_CSR_H_FRQ_MASK)) {
if (clk_rate < CSR_F_35M)
priv->clk_csr = STMMAC_CSR_20_35M;
@@ -206,10 +211,7 @@ static void stmmac_clk_csr_set(struct stmmac_priv *priv)
priv->clk_csr = STMMAC_CSR_150_250M;
else if ((clk_rate >= CSR_F_250M) && (clk_rate < CSR_F_300M))
priv->clk_csr = STMMAC_CSR_250_300M;
- } /* For values higher than the IEEE 802.3 specified frequency
- * we can not estimate the proper divider as it is not known
- * the frequency of clk_csr_i. So we do not change the default
- * divider. */
+ }
}
#if defined(STMMAC_XMIT_DEBUG) || defined(STMMAC_RX_DEBUG)
@@ -245,8 +247,7 @@ static inline void stmmac_hw_fix_mac_speed(struct stmmac_priv *priv)
struct phy_device *phydev = priv->phydev;
if (likely(priv->plat->fix_mac_speed))
- priv->plat->fix_mac_speed(priv->plat->bsp_priv,
- phydev->speed);
+ priv->plat->fix_mac_speed(priv->plat->bsp_priv, phydev->speed);
}
/**
@@ -351,8 +352,7 @@ static void stmmac_eee_adjust(struct stmmac_priv *priv)
* and also perform some sanity checks.
*/
static void stmmac_get_tx_hwtstamp(struct stmmac_priv *priv,
- unsigned int entry,
- struct sk_buff *skb)
+ unsigned int entry, struct sk_buff *skb)
{
struct skb_shared_hwtstamps shhwtstamp;
u64 ns;
@@ -361,7 +361,7 @@ static void stmmac_get_tx_hwtstamp(struct stmmac_priv *priv,
if (!priv->hwts_tx_en)
return;
- /* if skb doesn't support hw tstamp */
+ /* exit if skb doesn't support hw tstamp */
if (likely(!(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)))
return;
@@ -394,8 +394,7 @@ static void stmmac_get_tx_hwtstamp(struct stmmac_priv *priv,
* and pass it to stack. It also perform some sanity checks.
*/
static void stmmac_get_rx_hwtstamp(struct stmmac_priv *priv,
- unsigned int entry,
- struct sk_buff *skb)
+ unsigned int entry, struct sk_buff *skb)
{
struct skb_shared_hwtstamps *shhwtstamp = NULL;
u64 ns;
@@ -409,7 +408,7 @@ static void stmmac_get_rx_hwtstamp(struct stmmac_priv *priv,
else
desc = (priv->dma_rx + entry);
- /* if rx tstamp is not valid */
+ /* exit if rx tstamp is not valid */
if (!priv->hw->desc->get_rx_timestamp_status(desc, priv->adv_ts))
return;
@@ -456,7 +455,7 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
}
if (copy_from_user(&config, ifr->ifr_data,
- sizeof(struct hwtstamp_config)))
+ sizeof(struct hwtstamp_config)))
return -EFAULT;
pr_debug("%s config flags:0x%x, tx_type:0x%x, rx_filter:0x%x\n",
@@ -479,13 +478,13 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
if (priv->adv_ts) {
switch (config.rx_filter) {
- /* time stamp no incoming packet at all */
case HWTSTAMP_FILTER_NONE:
+ /* time stamp no incoming packet at all */
config.rx_filter = HWTSTAMP_FILTER_NONE;
break;
- /* PTP v1, UDP, any kind of event packet */
case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
+ /* PTP v1, UDP, any kind of event packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT;
/* take time stamp for all event messages */
snap_type_sel = PTP_TCR_SNAPTYPSEL_1;
@@ -494,8 +493,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA;
break;
- /* PTP v1, UDP, Sync packet */
case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+ /* PTP v1, UDP, Sync packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_SYNC;
/* take time stamp for SYNC messages only */
ts_event_en = PTP_TCR_TSEVNTENA;
@@ -504,8 +503,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA;
break;
- /* PTP v1, UDP, Delay_req packet */
case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+ /* PTP v1, UDP, Delay_req packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ;
/* take time stamp for Delay_Req messages only */
ts_master_en = PTP_TCR_TSMSTRENA;
@@ -515,8 +514,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA;
break;
- /* PTP v2, UDP, any kind of event packet */
case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
+ /* PTP v2, UDP, any kind of event packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT;
ptp_v2 = PTP_TCR_TSVER2ENA;
/* take time stamp for all event messages */
@@ -526,8 +525,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA;
break;
- /* PTP v2, UDP, Sync packet */
case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
+ /* PTP v2, UDP, Sync packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_SYNC;
ptp_v2 = PTP_TCR_TSVER2ENA;
/* take time stamp for SYNC messages only */
@@ -537,8 +536,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA;
break;
- /* PTP v2, UDP, Delay_req packet */
case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
+ /* PTP v2, UDP, Delay_req packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ;
ptp_v2 = PTP_TCR_TSVER2ENA;
/* take time stamp for Delay_Req messages only */
@@ -549,8 +548,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA;
break;
- /* PTP v2/802.AS1, any layer, any kind of event packet */
case HWTSTAMP_FILTER_PTP_V2_EVENT:
+ /* PTP v2/802.AS1 any layer, any kind of event packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
ptp_v2 = PTP_TCR_TSVER2ENA;
/* take time stamp for all event messages */
@@ -561,8 +560,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ethernet = PTP_TCR_TSIPENA;
break;
- /* PTP v2/802.AS1, any layer, Sync packet */
case HWTSTAMP_FILTER_PTP_V2_SYNC:
+ /* PTP v2/802.AS1, any layer, Sync packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V2_SYNC;
ptp_v2 = PTP_TCR_TSVER2ENA;
/* take time stamp for SYNC messages only */
@@ -573,8 +572,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ethernet = PTP_TCR_TSIPENA;
break;
- /* PTP v2/802.AS1, any layer, Delay_req packet */
case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
+ /* PTP v2/802.AS1, any layer, Delay_req packet */
config.rx_filter = HWTSTAMP_FILTER_PTP_V2_DELAY_REQ;
ptp_v2 = PTP_TCR_TSVER2ENA;
/* take time stamp for Delay_Req messages only */
@@ -586,8 +585,8 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
ptp_over_ethernet = PTP_TCR_TSIPENA;
break;
- /* time stamp any incoming packet */
case HWTSTAMP_FILTER_ALL:
+ /* time stamp any incoming packet */
config.rx_filter = HWTSTAMP_FILTER_ALL;
tstamp_all = PTP_TCR_TSENALL;
break;
@@ -612,9 +611,9 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
priv->hw->ptp->config_hw_tstamping(priv->ioaddr, 0);
else {
value = (PTP_TCR_TSENA | PTP_TCR_TSCFUPDT | PTP_TCR_TSCTRLSSR |
- tstamp_all | ptp_v2 | ptp_over_ethernet |
- ptp_over_ipv6_udp | ptp_over_ipv4_udp | ts_event_en |
- ts_master_en | snap_type_sel);
+ tstamp_all | ptp_v2 | ptp_over_ethernet |
+ ptp_over_ipv6_udp | ptp_over_ipv4_udp | ts_event_en |
+ ts_master_en | snap_type_sel);
priv->hw->ptp->config_hw_tstamping(priv->ioaddr, value);
@@ -632,7 +631,7 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
* 2^x * y == (y << x), hence
* 2^32 * 50000000 ==> (50000000 << 32)
*/
- temp = (u64)(50000000ULL << 32);
+ temp = (u64) (50000000ULL << 32);
priv->default_addend = div_u64(temp, STMMAC_SYSCLOCK);
priv->hw->ptp->config_addend(priv->ioaddr,
priv->default_addend);
@@ -665,7 +664,8 @@ static int stmmac_init_ptp(struct stmmac_priv *priv)
priv->adv_ts = 0;
}
if (priv->dma_cap.atime_stamp && priv->extend_desc) {
- pr_debug("IEEE 1588-2008 Advanced Time Stamp supported\n");
+ pr_debug
+ ("IEEE 1588-2008 Advanced Time Stamp supported\n");
priv->adv_ts = 1;
}
}
@@ -727,7 +727,7 @@ static void stmmac_adjust_link(struct net_device *dev)
case 1000:
if (likely(priv->plat->has_gmac))
ctrl &= ~priv->hw->link.port;
- stmmac_hw_fix_mac_speed(priv);
+ stmmac_hw_fix_mac_speed(priv);
break;
case 100:
case 10:
@@ -745,8 +745,8 @@ static void stmmac_adjust_link(struct net_device *dev)
break;
default:
if (netif_msg_link(priv))
- pr_warning("%s: Speed (%d) is not 10"
- " or 100!\n", dev->name, phydev->speed);
+ pr_warn("%s: Speed (%d) not 10/100\n",
+ dev->name, phydev->speed);
break;
}
@@ -822,10 +822,10 @@ static int stmmac_init_phy(struct net_device *dev)
if (priv->plat->phy_bus_name)
snprintf(bus_id, MII_BUS_ID_SIZE, "%s-%x",
- priv->plat->phy_bus_name, priv->plat->bus_id);
+ priv->plat->phy_bus_name, priv->plat->bus_id);
else
snprintf(bus_id, MII_BUS_ID_SIZE, "stmmac-%x",
- priv->plat->bus_id);
+ priv->plat->bus_id);
snprintf(phy_id_fmt, MII_BUS_ID_SIZE + 3, PHY_ID_FMT, bus_id,
priv->plat->phy_addr);
@@ -873,23 +873,23 @@ static int stmmac_init_phy(struct net_device *dev)
static void stmmac_display_ring(void *head, int size, int extend_desc)
{
int i;
- struct dma_extended_desc *ep = (struct dma_extended_desc *) head;
- struct dma_desc *p = (struct dma_desc *) head;
+ struct dma_extended_desc *ep = (struct dma_extended_desc *)head;
+ struct dma_desc *p = (struct dma_desc *)head;
for (i = 0; i < size; i++) {
u64 x;
if (extend_desc) {
x = *(u64 *) ep;
pr_info("%d [0x%x]: 0x%x 0x%x 0x%x 0x%x\n",
- i, (unsigned int) virt_to_phys(ep),
- (unsigned int) x, (unsigned int) (x >> 32),
+ i, (unsigned int)virt_to_phys(ep),
+ (unsigned int)x, (unsigned int)(x >> 32),
ep->basic.des2, ep->basic.des3);
ep++;
} else {
x = *(u64 *) p;
pr_info("%d [0x%x]: 0x%x 0x%x 0x%x 0x%x",
- i, (unsigned int) virt_to_phys(p),
- (unsigned int) x, (unsigned int) (x >> 32),
+ i, (unsigned int)virt_to_phys(p),
+ (unsigned int)x, (unsigned int)(x >> 32),
p->des2, p->des3);
p++;
}
@@ -904,9 +904,9 @@ static void stmmac_display_rings(struct stmmac_priv *priv)
if (priv->extend_desc) {
pr_info("Extended RX descriptor ring:\n");
- stmmac_display_ring((void *) priv->dma_erx, rxsize, 1);
+ stmmac_display_ring((void *)priv->dma_erx, rxsize, 1);
pr_info("Extended TX descriptor ring:\n");
- stmmac_display_ring((void *) priv->dma_etx, txsize, 1);
+ stmmac_display_ring((void *)priv->dma_etx, txsize, 1);
} else {
pr_info("RX descriptor ring:\n");
stmmac_display_ring((void *)priv->dma_rx, rxsize, 0);
@@ -1006,7 +1006,8 @@ static void init_dma_desc_rings(struct net_device *dev)
unsigned int bfsize = 0;
/* Set the max buffer size according to the DESC mode
- * and the MTU. Note that RING mode allows 16KiB bsize. */
+ * and the MTU. Note that RING mode allows 16KiB bsize.
+ */
if (priv->mode == STMMAC_RING_MODE)
bfsize = priv->hw->ring->set_16kib_bfsize(dev->mtu);
@@ -1047,7 +1048,7 @@ static void init_dma_desc_rings(struct net_device *dev)
priv->rx_skbuff = kmalloc_array(rxsize, sizeof(struct sk_buff *),
GFP_KERNEL);
priv->tx_skbuff_dma = kmalloc_array(txsize, sizeof(dma_addr_t),
- GFP_KERNEL);
+ GFP_KERNEL);
priv->tx_skbuff = kmalloc_array(txsize, sizeof(struct sk_buff *),
GFP_KERNEL);
if (netif_msg_drv(priv))
@@ -1067,7 +1068,7 @@ static void init_dma_desc_rings(struct net_device *dev)
break;
DBG(probe, INFO, "[%p]\t[%p]\t[%x]\n", priv->rx_skbuff[i],
- priv->rx_skbuff[i]->data, priv->rx_skbuff_dma[i]);
+ priv->rx_skbuff[i]->data, priv->rx_skbuff_dma[i]);
}
priv->cur_rx = 0;
priv->dirty_rx = (unsigned int)(i - rxsize);
@@ -1154,8 +1155,7 @@ static void free_dma_desc_resources(struct stmmac_priv *priv)
dma_free_rx_skbufs(priv);
dma_free_tx_skbufs(priv);
- /* Free the region of consistent memory previously allocated for
- * the DMA */
+ /* Free DMA regions of consistent memory previously allocated */
if (!priv->extend_desc) {
dma_free_coherent(priv->device,
priv->dma_tx_size * sizeof(struct dma_desc),
@@ -1186,7 +1186,7 @@ static void free_dma_desc_resources(struct stmmac_priv *priv)
static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
{
if (likely(priv->plat->force_sf_dma_mode ||
- ((priv->plat->tx_coe) && (!priv->no_csum_insertion)))) {
+ ((priv->plat->tx_coe) && (!priv->no_csum_insertion)))) {
/*
* In case of GMAC, SF mode can be enabled
* to perform the TX COE in HW. This depends on:
@@ -1194,8 +1194,7 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
* 2) There is no bugged Jumbo frame support
* that needs to not insert csum in the TDES.
*/
- priv->hw->dma->dma_mode(priv->ioaddr,
- SF_DMA_MODE, SF_DMA_MODE);
+ priv->hw->dma->dma_mode(priv->ioaddr, SF_DMA_MODE, SF_DMA_MODE);
tc = SF_DMA_MODE;
} else
priv->hw->dma->dma_mode(priv->ioaddr, tc, SF_DMA_MODE);
@@ -1221,7 +1220,7 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
struct dma_desc *p;
if (priv->extend_desc)
- p = (struct dma_desc *) (priv->dma_etx + entry);
+ p = (struct dma_desc *)(priv->dma_etx + entry);
else
p = priv->dma_tx + entry;
@@ -1233,9 +1232,9 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
last = priv->hw->desc->get_tx_ls(p);
if (likely(last)) {
int tx_error =
- priv->hw->desc->tx_status(&priv->dev->stats,
- &priv->xstats, p,
- priv->ioaddr);
+ priv->hw->desc->tx_status(&priv->dev->stats,
+ &priv->xstats, p,
+ priv->ioaddr);
if (likely(tx_error == 0)) {
priv->dev->stats.tx_packets++;
priv->xstats.tx_pkt_n++;
@@ -1245,7 +1244,7 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
stmmac_get_tx_hwtstamp(priv, entry, skb);
}
TX_DBG("%s: curr %d, dirty %d\n", __func__,
- priv->cur_tx, priv->dirty_tx);
+ priv->cur_tx, priv->dirty_tx);
if (likely(priv->tx_skbuff_dma[entry])) {
dma_unmap_single(priv->device,
@@ -1269,7 +1268,7 @@ static void stmmac_tx_clean(struct stmmac_priv *priv)
stmmac_tx_avail(priv) > STMMAC_TX_THRESH(priv))) {
netif_tx_lock(priv->dev);
if (netif_queue_stopped(priv->dev) &&
- stmmac_tx_avail(priv) > STMMAC_TX_THRESH(priv)) {
+ stmmac_tx_avail(priv) > STMMAC_TX_THRESH(priv)) {
TX_DBG("%s: restart transmit\n", __func__);
netif_wake_queue(priv->dev);
}
@@ -1293,7 +1292,6 @@ static inline void stmmac_disable_dma_irq(struct stmmac_priv *priv)
priv->hw->dma->disable_dma_irq(priv->ioaddr);
}
-
/**
* stmmac_tx_err: irq tx error mng function
* @priv: driver private structure
@@ -1363,7 +1361,7 @@ static void stmmac_dma_interrupt(struct stmmac_priv *priv)
static void stmmac_mmc_setup(struct stmmac_priv *priv)
{
unsigned int mode = MMC_CNTRL_RESET_ON_READ | MMC_CNTRL_COUNTER_RESET |
- MMC_CNTRL_PRESET | MMC_CNTRL_FULL_HALF_PRESET;
+ MMC_CNTRL_PRESET | MMC_CNTRL_FULL_HALF_PRESET;
dwmac_mmc_intr_all_mask(priv->ioaddr);
@@ -1378,8 +1376,7 @@ static u32 stmmac_get_synopsys_id(struct stmmac_priv *priv)
{
u32 hwid = priv->hw->synopsys_uid;
- /* Only check valid Synopsys Id because old MAC chips
- * have no HW registers where get the ID */
+ /* Check Synopsys Id (not available on old chips) */
if (likely(hwid)) {
u32 uid = ((hwid & 0x0000ff00) >> 8);
u32 synid = (hwid & 0x000000ff);
@@ -1438,41 +1435,39 @@ static int stmmac_get_hw_features(struct stmmac_priv *priv)
priv->dma_cap.mbps_1000 = (hw_cap & DMA_HW_FEAT_GMIISEL) >> 1;
priv->dma_cap.half_duplex = (hw_cap & DMA_HW_FEAT_HDSEL) >> 2;
priv->dma_cap.hash_filter = (hw_cap & DMA_HW_FEAT_HASHSEL) >> 4;
- priv->dma_cap.multi_addr =
- (hw_cap & DMA_HW_FEAT_ADDMACADRSEL) >> 5;
+ priv->dma_cap.multi_addr = (hw_cap & DMA_HW_FEAT_ADDMAC) >> 5;
priv->dma_cap.pcs = (hw_cap & DMA_HW_FEAT_PCSSEL) >> 6;
priv->dma_cap.sma_mdio = (hw_cap & DMA_HW_FEAT_SMASEL) >> 8;
priv->dma_cap.pmt_remote_wake_up =
- (hw_cap & DMA_HW_FEAT_RWKSEL) >> 9;
+ (hw_cap & DMA_HW_FEAT_RWKSEL) >> 9;
priv->dma_cap.pmt_magic_frame =
- (hw_cap & DMA_HW_FEAT_MGKSEL) >> 10;
+ (hw_cap & DMA_HW_FEAT_MGKSEL) >> 10;
/* MMC */
priv->dma_cap.rmon = (hw_cap & DMA_HW_FEAT_MMCSEL) >> 11;
- /* IEEE 1588-2002*/
+ /* IEEE 1588-2002 */
priv->dma_cap.time_stamp =
- (hw_cap & DMA_HW_FEAT_TSVER1SEL) >> 12;
- /* IEEE 1588-2008*/
+ (hw_cap & DMA_HW_FEAT_TSVER1SEL) >> 12;
+ /* IEEE 1588-2008 */
priv->dma_cap.atime_stamp =
- (hw_cap & DMA_HW_FEAT_TSVER2SEL) >> 13;
+ (hw_cap & DMA_HW_FEAT_TSVER2SEL) >> 13;
/* 802.3az - Energy-Efficient Ethernet (EEE) */
priv->dma_cap.eee = (hw_cap & DMA_HW_FEAT_EEESEL) >> 14;
priv->dma_cap.av = (hw_cap & DMA_HW_FEAT_AVSEL) >> 15;
/* TX and RX csum */
priv->dma_cap.tx_coe = (hw_cap & DMA_HW_FEAT_TXCOESEL) >> 16;
priv->dma_cap.rx_coe_type1 =
- (hw_cap & DMA_HW_FEAT_RXTYP1COE) >> 17;
+ (hw_cap & DMA_HW_FEAT_RXTYP1COE) >> 17;
priv->dma_cap.rx_coe_type2 =
- (hw_cap & DMA_HW_FEAT_RXTYP2COE) >> 18;
+ (hw_cap & DMA_HW_FEAT_RXTYP2COE) >> 18;
priv->dma_cap.rxfifo_over_2048 =
- (hw_cap & DMA_HW_FEAT_RXFIFOSIZE) >> 19;
+ (hw_cap & DMA_HW_FEAT_RXFIFOSIZE) >> 19;
/* TX and RX number of channels */
priv->dma_cap.number_rx_channel =
- (hw_cap & DMA_HW_FEAT_RXCHCNT) >> 20;
+ (hw_cap & DMA_HW_FEAT_RXCHCNT) >> 20;
priv->dma_cap.number_tx_channel =
- (hw_cap & DMA_HW_FEAT_TXCHCNT) >> 22;
- /* Alternate (enhanced) DESC mode*/
- priv->dma_cap.enh_desc =
- (hw_cap & DMA_HW_FEAT_ENHDESSEL) >> 24;
+ (hw_cap & DMA_HW_FEAT_TXCHCNT) >> 22;
+ /* Alternate (enhanced) DESC mode */
+ priv->dma_cap.enh_desc = (hw_cap & DMA_HW_FEAT_ENHDESSEL) >> 24;
}
return hw_cap;
@@ -1491,11 +1486,11 @@ static void stmmac_check_ether_addr(struct stmmac_priv *priv)
priv->hw->mac->get_umac_addr((void __iomem *)
priv->dev->base_addr,
priv->dev->dev_addr, 0);
- if (!is_valid_ether_addr(priv->dev->dev_addr))
+ if (!is_valid_ether_addr(priv->dev->dev_addr))
eth_hw_addr_random(priv->dev);
}
- pr_warning("%s: device MAC address %pM\n", priv->dev->name,
- priv->dev->dev_addr);
+ pr_warn("%s: device MAC address %pM\n", priv->dev->name,
+ priv->dev->dev_addr);
}
/**
@@ -1611,7 +1606,7 @@ static int stmmac_open(struct net_device *dev)
/* Request the IRQ lines */
ret = request_irq(dev->irq, stmmac_interrupt,
- IRQF_SHARED, dev->name, dev);
+ IRQF_SHARED, dev->name, dev);
if (unlikely(ret < 0)) {
pr_err("%s: ERROR: allocating the IRQ %d (error: %d)\n",
__func__, dev->irq, ret);
@@ -1623,8 +1618,8 @@ static int stmmac_open(struct net_device *dev)
ret = request_irq(priv->wol_irq, stmmac_interrupt,
IRQF_SHARED, dev->name, dev);
if (unlikely(ret < 0)) {
- pr_err("%s: ERROR: allocating the ext WoL IRQ %d "
- "(error: %d)\n", __func__, priv->wol_irq, ret);
+ pr_err("%s: ERROR: allocating the WoL IRQ %d (%d)\n",
+ __func__, priv->wol_irq, ret);
goto open_error_wolirq;
}
}
@@ -1659,7 +1654,7 @@ static int stmmac_open(struct net_device *dev)
#ifdef CONFIG_STMMAC_DEBUG_FS
ret = stmmac_init_fs(dev);
if (ret < 0)
- pr_warning("%s: failed debugFS registration\n", __func__);
+ pr_warn("%s: failed debugFS registration\n", __func__);
#endif
/* Start the ball rolling... */
DBG(probe, DEBUG, "%s: DMA RX/TX processes started...\n", dev->name);
@@ -1791,8 +1786,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
if (!netif_queue_stopped(dev)) {
netif_stop_queue(dev);
/* This is a hard error, log it. */
- pr_err("%s: BUG! Tx Ring full when queue awake\n",
- __func__);
+ pr_err("%s: Tx Ring full when queue awake\n", __func__);
}
return NETDEV_TX_BUSY;
}
@@ -1806,10 +1800,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
#ifdef STMMAC_XMIT_DEBUG
if ((skb->len > ETH_FRAME_LEN) || nfrags)
- pr_debug("stmmac xmit: [entry %d]\n"
- "\tskb addr %p - len: %d - nopaged_len: %d\n"
+ pr_debug("%s: [entry %d]: skb addr %p len: %d nopagedlen: %d\n"
"\tn_frags: %d - ip_summed: %d - %s gso\n"
- "\ttx_count_frames %d\n", entry,
+ "\ttx_count_frames %d\n", __func__, entry,
skb, skb->len, nopaged_len, nfrags, skb->ip_summed,
!skb_is_gso(skb) ? "isn't" : "is",
priv->tx_count_frames);
@@ -1818,7 +1811,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
csum_insertion = (skb->ip_summed == CHECKSUM_PARTIAL);
if (priv->extend_desc)
- desc = (struct dma_desc *) (priv->dma_etx + entry);
+ desc = (struct dma_desc *)(priv->dma_etx + entry);
else
desc = priv->dma_tx + entry;
@@ -1841,14 +1834,14 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
csum_insertion);
} else {
is_jumbo = priv->hw->chain->is_jumbo_frm(skb->len,
- priv->plat->enh_desc);
+ priv->plat->enh_desc);
if (unlikely(is_jumbo))
entry = priv->hw->chain->jumbo_frm(priv, skb,
csum_insertion);
}
if (likely(!is_jumbo)) {
desc->des2 = dma_map_single(priv->device, skb->data,
- nopaged_len, DMA_TO_DEVICE);
+ nopaged_len, DMA_TO_DEVICE);
priv->tx_skbuff_dma[entry] = desc->des2;
priv->hw->desc->prepare_tx_desc(desc, 1, nopaged_len,
csum_insertion, priv->mode);
@@ -1861,7 +1854,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
entry = (++priv->cur_tx) % txsize;
if (priv->extend_desc)
- desc = (struct dma_desc *) (priv->dma_etx + entry);
+ desc = (struct dma_desc *)(priv->dma_etx + entry);
else
desc = priv->dma_tx + entry;
@@ -1902,10 +1895,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
#ifdef STMMAC_XMIT_DEBUG
if (netif_msg_pktdata(priv)) {
- pr_info("stmmac xmit: current=%d, dirty=%d, entry=%d, "
- "first=%p, nfrags=%d\n",
- (priv->cur_tx % txsize), (priv->dirty_tx % txsize),
- entry, first, nfrags);
+ pr_info("%s: curr %d dirty=%d entry=%d, first=%p, nfrags=%d"
+ __func__, (priv->cur_tx % txsize),
+ (priv->dirty_tx % txsize), entry, first, nfrags);
if (priv->extend_desc)
stmmac_display_ring((void *)priv->dma_etx, txsize, 1);
else
@@ -1955,7 +1947,7 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
struct dma_desc *p;
if (priv->extend_desc)
- p = (struct dma_desc *) (priv->dma_erx + entry);
+ p = (struct dma_desc *)(priv->dma_erx + entry);
else
p = priv->dma_rx + entry;
@@ -1996,12 +1988,13 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
unsigned int entry = priv->cur_rx % rxsize;
unsigned int next_entry;
unsigned int count = 0;
+ int coe = priv->plat->rx_coe;
#ifdef STMMAC_RX_DEBUG
if (netif_msg_hw(priv)) {
pr_debug(">>> stmmac_rx: descriptor ring:\n");
if (priv->extend_desc)
- stmmac_display_ring((void *) priv->dma_erx, rxsize, 1);
+ stmmac_display_ring((void *)priv->dma_erx, rxsize, 1);
else
stmmac_display_ring((void *)priv->dma_rx, rxsize, 0);
}
@@ -2011,9 +2004,9 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
struct dma_desc *p, *p_next;
if (priv->extend_desc)
- p = (struct dma_desc *) (priv->dma_erx + entry);
+ p = (struct dma_desc *)(priv->dma_erx + entry);
else
- p = priv->dma_rx + entry ;
+ p = priv->dma_rx + entry;
if (priv->hw->desc->get_rx_owner(p))
break;
@@ -2022,8 +2015,8 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
next_entry = (++priv->cur_rx) % rxsize;
if (priv->extend_desc)
- p_next = (struct dma_desc *) (priv->dma_erx +
- next_entry);
+ p_next = (struct dma_desc *)(priv->dma_erx +
+ next_entry);
else
p_next = priv->dma_rx + next_entry;
@@ -2047,32 +2040,34 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
*/
priv->rx_skbuff[entry] = NULL;
dma_unmap_single(priv->device,
- priv->rx_skbuff_dma[entry],
- priv->dma_buf_sz, DMA_FROM_DEVICE);
+ priv->rx_skbuff_dma[entry],
+ priv->dma_buf_sz,
+ DMA_FROM_DEVICE);
}
} else {
struct sk_buff *skb;
int frame_len;
- frame_len = priv->hw->desc->get_rx_frame_len(p,
- priv->plat->rx_coe);
+ frame_len = priv->hw->desc->get_rx_frame_len(p, coe);
+
/* ACS is set; GMAC core strips PAD/FCS for IEEE 802.3
- * Type frames (LLC/LLC-SNAP) */
+ * Type frames (LLC/LLC-SNAP)
+ */
if (unlikely(status != llc_snap))
frame_len -= ETH_FCS_LEN;
#ifdef STMMAC_RX_DEBUG
if (frame_len > ETH_FRAME_LEN)
pr_debug("\tRX frame size %d, COE status: %d\n",
- frame_len, status);
+ frame_len, status);
if (netif_msg_hw(priv))
pr_debug("\tdesc: %p [entry %d] buff=0x%x\n",
- p, entry, p->des2);
+ p, entry, p->des2);
#endif
skb = priv->rx_skbuff[entry];
if (unlikely(!skb)) {
pr_err("%s: Inconsistent Rx descriptor chain\n",
- priv->dev->name);
+ priv->dev->name);
priv->dev->stats.rx_dropped++;
break;
}
@@ -2093,7 +2088,7 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
#endif
skb->protocol = eth_type_trans(skb, priv->dev);
- if (unlikely(!priv->plat->rx_coe))
+ if (unlikely(!coe))
skb_checksum_none_assert(skb);
else
skb->ip_summed = CHECKSUM_UNNECESSARY;
@@ -2162,18 +2157,16 @@ static int stmmac_config(struct net_device *dev, struct ifmap *map)
/* Don't allow changing the I/O address */
if (map->base_addr != dev->base_addr) {
- pr_warning("%s: can't change I/O address\n", dev->name);
+ pr_warn("%s: can't change I/O address\n", dev->name);
return -EOPNOTSUPP;
}
/* Don't allow changing the IRQ */
if (map->irq != dev->irq) {
- pr_warning("%s: can't change IRQ number %d\n",
- dev->name, dev->irq);
+ pr_warn("%s: not change IRQ number %d\n", dev->name, dev->irq);
return -EOPNOTSUPP;
}
- /* ignore other fields */
return 0;
}
@@ -2233,7 +2226,7 @@ static int stmmac_change_mtu(struct net_device *dev, int new_mtu)
}
static netdev_features_t stmmac_fix_features(struct net_device *dev,
- netdev_features_t features)
+ netdev_features_t features)
{
struct stmmac_priv *priv = netdev_priv(dev);
@@ -2247,7 +2240,8 @@ static netdev_features_t stmmac_fix_features(struct net_device *dev,
/* Some GMAC devices have a bugged Jumbo frame support that
* needs to have the Tx COE disabled for oversized frames
* (due to limited buffer sizes). In this case we disable
- * the TX csum insertionin the TDES and not use SF. */
+ * the TX csum insertionin the TDES and not use SF.
+ */
if (priv->plat->bugged_jumbo && (dev->mtu > ETH_DATA_LEN))
features &= ~NETIF_F_ALL_CSUM;
@@ -2294,7 +2288,8 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id)
#ifdef CONFIG_NET_POLL_CONTROLLER
/* Polling receive - used by NETCONSOLE and other diagnostic tools
- * to allow network I/O with interrupts disabled. */
+ * to allow network I/O with interrupts disabled.
+ */
static void stmmac_poll_controller(struct net_device *dev)
{
disable_irq(dev->irq);
@@ -2344,26 +2339,26 @@ static struct dentry *stmmac_rings_status;
static struct dentry *stmmac_dma_cap;
static void sysfs_display_ring(void *head, int size, int extend_desc,
- struct seq_file *seq)
+ struct seq_file *seq)
{
int i;
- struct dma_extended_desc *ep = (struct dma_extended_desc *) head;
- struct dma_desc *p = (struct dma_desc *) head;
+ struct dma_extended_desc *ep = (struct dma_extended_desc *)head;
+ struct dma_desc *p = (struct dma_desc *)head;
for (i = 0; i < size; i++) {
u64 x;
if (extend_desc) {
x = *(u64 *) ep;
seq_printf(seq, "%d [0x%x]: 0x%x 0x%x 0x%x 0x%x\n",
- i, (unsigned int) virt_to_phys(ep),
- (unsigned int) x, (unsigned int) (x >> 32),
+ i, (unsigned int)virt_to_phys(ep),
+ (unsigned int)x, (unsigned int)(x >> 32),
ep->basic.des2, ep->basic.des3);
ep++;
} else {
x = *(u64 *) p;
seq_printf(seq, "%d [0x%x]: 0x%x 0x%x 0x%x 0x%x\n",
- i, (unsigned int) virt_to_phys(ep),
- (unsigned int) x, (unsigned int) (x >> 32),
+ i, (unsigned int)virt_to_phys(ep),
+ (unsigned int)x, (unsigned int)(x >> 32),
p->des2, p->des3);
p++;
}
@@ -2380,9 +2375,9 @@ static int stmmac_sysfs_ring_read(struct seq_file *seq, void *v)
if (priv->extend_desc) {
seq_printf(seq, "Extended RX descriptor ring:\n");
- sysfs_display_ring((void *) priv->dma_erx, rxsize, 1, seq);
+ sysfs_display_ring((void *)priv->dma_erx, rxsize, 1, seq);
seq_printf(seq, "Extended TX descriptor ring:\n");
- sysfs_display_ring((void *) priv->dma_etx, txsize, 1, seq);
+ sysfs_display_ring((void *)priv->dma_etx, txsize, 1, seq);
} else {
seq_printf(seq, "RX descriptor ring:\n");
sysfs_display_ring((void *)priv->dma_rx, rxsize, 0, seq);
@@ -2492,8 +2487,8 @@ static int stmmac_init_fs(struct net_device *dev)
/* Entry to report DMA RX/TX rings */
stmmac_rings_status = debugfs_create_file("descriptors_status",
- S_IRUGO, stmmac_fs_dir, dev,
- &stmmac_rings_status_fops);
+ S_IRUGO, stmmac_fs_dir, dev,
+ &stmmac_rings_status_fops);
if (!stmmac_rings_status || IS_ERR(stmmac_rings_status)) {
pr_info("ERROR creating stmmac ring debugfs file\n");
@@ -2574,7 +2569,7 @@ static int stmmac_hw_init(struct stmmac_priv *priv)
stmmac_selec_desc_mode(priv);
/* To use the chained or ring mode */
- if (chain_mode) {
+ if (chain_mode) {
priv->hw->chain = &chain_mode_ops;
pr_info(" Chain mode enabled\n");
priv->mode = STMMAC_CHAIN_MODE;
@@ -2607,11 +2602,9 @@ static int stmmac_hw_init(struct stmmac_priv *priv)
} else
pr_info(" No HW DMA feature register supported");
- /* Enable the IPC (Checksum Offload) and check if the feature has been
- * enabled during the core configuration. */
ret = priv->hw->mac->rx_ipc(priv->ioaddr);
if (!ret) {
- pr_warning(" RX IPC Checksum Offload not configured.\n");
+ pr_warn(" RX IPC Checksum Offload not configured.\n");
priv->plat->rx_coe = STMMAC_RX_COE_NONE;
}
@@ -2667,7 +2660,8 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device,
stmmac_verify_args();
/* Override with kernel parameters if supplied XXX CRS XXX
- * this needs to have multiple instances */
+ * this needs to have multiple instances
+ */
if ((phyaddr >= 0) && (phyaddr <= 31))
priv->plat->phy_addr = phyaddr;
@@ -2679,7 +2673,7 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device,
ndev->netdev_ops = &stmmac_netdev_ops;
ndev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
- NETIF_F_RXCSUM;
+ NETIF_F_RXCSUM;
ndev->features |= ndev->hw_features | NETIF_F_HIGHDMA | NETIF_F_GRO;
ndev->watchdog_timeo = msecs_to_jiffies(watchdog);
#ifdef STMMAC_VLAN_TAG_USED
@@ -2714,7 +2708,7 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device,
priv->stmmac_clk = clk_get(priv->device, STMMAC_RESOURCE_NAME);
if (IS_ERR(priv->stmmac_clk)) {
- pr_warning("%s: warning: cannot get CSR clock\n", __func__);
+ pr_warn("%s: warning: cannot get CSR clock\n", __func__);
goto error_clk_get;
}
@@ -2831,7 +2825,8 @@ int stmmac_resume(struct net_device *ndev)
* automatically as soon as a magic packet or a Wake-up frame
* is received. Anyway, it's better to manually clear
* this bit because it can generate problems while resuming
- * from another devices (e.g. serial console). */
+ * from another devices (e.g. serial console).
+ */
if (device_may_wakeup(priv->device))
priv->hw->mac->pmt(priv->ioaddr, 0);
else
@@ -2955,7 +2950,7 @@ err:
}
__setup("stmmaceth=", stmmac_cmdline_opt);
-#endif
+#endif /* MODULE */
MODULE_DESCRIPTION("STMMAC 10/100/1000 Ethernet device driver");
MODULE_AUTHOR("Giuseppe Cavallaro <peppe.cavallaro@st.com>");
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
index 0b9829f..cc15039 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
@@ -177,7 +177,7 @@ int stmmac_mdio_register(struct net_device *ndev)
new_bus->write = &stmmac_mdio_write;
new_bus->reset = &stmmac_mdio_reset;
snprintf(new_bus->id, MII_BUS_ID_SIZE, "%s-%x",
- new_bus->name, priv->plat->bus_id);
+ new_bus->name, priv->plat->bus_id);
new_bus->priv = ndev;
new_bus->irq = irqlist;
new_bus->phy_mask = mdio_bus_data->phy_mask;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
index 19b3a25..023b7c2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c
@@ -88,7 +88,7 @@ static int stmmac_pci_probe(struct pci_dev *pdev,
continue;
addr = pci_iomap(pdev, i, 0);
if (addr == NULL) {
- pr_err("%s: ERROR: cannot map register memory, aborting",
+ pr_err("%s: ERROR: cannot map register memory aborting",
__func__);
ret = -EIO;
goto err_out_map_failed;
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 5/7] stmmac: improve/review and fix kernel-doc
From: Giuseppe CAVALLARO @ 2013-04-03 5:41 UTC (permalink / raw)
To: netdev; +Cc: Giuseppe Cavallaro
In-Reply-To: <1364967689-11155-1-git-send-email-peppe.cavallaro@st.com>
this patch reviews/improves and adds some fixes in the code doc.
Also kernel-doc passes w/o any warnings.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 184 ++++++++++++++++-----
drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c | 8 +-
2 files changed, 142 insertions(+), 50 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index c92dcbc..52000b8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -81,14 +81,14 @@
#define JUMBO_LEN 9000
/* Module parameters */
-#define TX_TIMEO 5000 /* default 5 seconds */
+#define TX_TIMEO 5000
static int watchdog = TX_TIMEO;
module_param(watchdog, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(watchdog, "Transmit timeout in milliseconds");
+MODULE_PARM_DESC(watchdog, "Transmit timeout in milliseconds (default 5s)");
-static int debug = -1; /* -1: default, 0: no output, 16: all */
+static int debug = -1;
module_param(debug, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(debug, "Message Level (0: no output, 16: all)");
+MODULE_PARM_DESC(debug, "Message Level (-1: default, 0: no output, 16: all)");
int phyaddr = -1;
module_param(phyaddr, int, S_IRUGO);
@@ -173,6 +173,18 @@ static void stmmac_verify_args(void)
eee_timer = STMMAC_DEFAULT_LPI_TIMER;
}
+/**
+ * stmmac_clk_csr_set - dynamically set the MDC clock
+ * @priv: driver private structure
+ * Description: this is to dynamically set the MDC clock according to the csr
+ * clock input.
+ * Note:
+ * If a specific clk_csr value is passed from the platform
+ * this means that the CSR Clock Range selection cannot be
+ * changed at run-time and it is fixed (as reported in the driver
+ * documentation). Viceversa the driver will try to set the MDC
+ * clock dynamically according to the actual clock input.
+ */
static void stmmac_clk_csr_set(struct stmmac_priv *priv)
{
u32 clk_rate;
@@ -222,8 +234,11 @@ static inline u32 stmmac_tx_avail(struct stmmac_priv *priv)
return priv->dirty_tx + priv->dma_tx_size - priv->cur_tx - 1;
}
-/* On some ST platforms, some HW system configuraton registers have to be
- * set according to the link speed negotiated.
+/**
+ * stmmac_hw_fix_mac_speed: callback for speed selection
+ * @priv: driver private structure
+ * Description: on some platforms (e.g. ST), some HW system configuraton
+ * registers have to be set according to the link speed negotiated.
*/
static inline void stmmac_hw_fix_mac_speed(struct stmmac_priv *priv)
{
@@ -234,6 +249,11 @@ static inline void stmmac_hw_fix_mac_speed(struct stmmac_priv *priv)
phydev->speed);
}
+/**
+ * stmmac_enable_eee_mode: Check and enter in LPI mode
+ * @priv: driver private structure
+ * Description: this function is to verify and enter in LPI mode for EEE.
+ */
static void stmmac_enable_eee_mode(struct stmmac_priv *priv)
{
/* Check and enter in LPI mode */
@@ -242,19 +262,24 @@ static void stmmac_enable_eee_mode(struct stmmac_priv *priv)
priv->hw->mac->set_eee_mode(priv->ioaddr);
}
+/**
+ * stmmac_disable_eee_mode: disable/exit from EEE
+ * @priv: driver private structure
+ * Description: this function is to exit and disable EEE in case of
+ * LPI state is true. This is called by the xmit.
+ */
void stmmac_disable_eee_mode(struct stmmac_priv *priv)
{
- /* Exit and disable EEE in case of we are are in LPI state. */
priv->hw->mac->reset_eee_mode(priv->ioaddr);
del_timer_sync(&priv->eee_ctrl_timer);
priv->tx_path_in_lpi_mode = false;
}
/**
- * stmmac_eee_ctrl_timer
+ * stmmac_eee_ctrl_timer: EEE TX SW timer.
* @arg : data hook
* Description:
- * If there is no data transfer and if we are not in LPI state,
+ * if there is no data transfer and if we are not in LPI state,
* then MAC Transmitter can be moved to LPI state.
*/
static void stmmac_eee_ctrl_timer(unsigned long arg)
@@ -266,8 +291,8 @@ static void stmmac_eee_ctrl_timer(unsigned long arg)
}
/**
- * stmmac_eee_init
- * @priv: private device pointer
+ * stmmac_eee_init: init EEE
+ * @priv: driver private structure
* Description:
* If the EEE support has been enabled while configuring the driver,
* if the GMAC actually supports the EEE (from the HW cap reg) and the
@@ -303,18 +328,22 @@ out:
return ret;
}
+/**
+ * stmmac_eee_adjust: adjust HW EEE according to the speed
+ * @priv: driver private structure
+ * Description:
+ * When the EEE has been already initialised we have to
+ * modify the PLS bit in the LPI ctrl & status reg according
+ * to the PHY link status. For this reason.
+ */
static void stmmac_eee_adjust(struct stmmac_priv *priv)
{
- /* When the EEE has been already initialised we have to
- * modify the PLS bit in the LPI ctrl & status reg according
- * to the PHY link status. For this reason.
- */
if (priv->eee_enabled)
priv->hw->mac->set_eee_pls(priv->ioaddr, priv->phydev->link);
}
-/* stmmac_get_tx_hwtstamp:
- * @priv : pointer to private device structure.
+/* stmmac_get_tx_hwtstamp: get HW TX timestamps
+ * @priv: driver private structure
* @entry : descriptor index to be used.
* @skb : the socket buffer
* Description :
@@ -356,8 +385,8 @@ static void stmmac_get_tx_hwtstamp(struct stmmac_priv *priv,
return;
}
-/* stmmac_get_rx_hwtstamp:
- * @priv : pointer to private device structure.
+/* stmmac_get_rx_hwtstamp: get HW RX timestamps
+ * @priv: driver private structure
* @entry : descriptor index to be used.
* @skb : the socket buffer
* Description :
@@ -618,6 +647,13 @@ static int stmmac_hwtstamp_ioctl(struct net_device *dev, struct ifreq *ifr)
sizeof(struct hwtstamp_config)) ? -EFAULT : 0;
}
+/**
+ * stmmac_init_ptp: init PTP
+ * @priv: driver private structure
+ * Description: this is to verify if the HW supports the PTPv1 or v2.
+ * This is done by looking at the HW cap. register.
+ * Also it registers the ptp driver.
+ */
static int stmmac_init_ptp(struct stmmac_priv *priv)
{
if (!(priv->dma_cap.time_stamp || priv->dma_cap.atime_stamp))
@@ -740,6 +776,13 @@ static void stmmac_adjust_link(struct net_device *dev)
DBG(probe, DEBUG, "stmmac_adjust_link: exiting\n");
}
+/**
+ * stmmac_check_pcs_mode: verify if RGMII/SGMII is supported
+ * @priv: driver private structure
+ * Description: this is to verify if the HW supports the PCS.
+ * Physical Coding Sublayer (PCS) interface that can be used when the MAC is
+ * configured for the TBI, RTBI, or SGMII PHY interface.
+ */
static void stmmac_check_pcs_mode(struct stmmac_priv *priv)
{
int interface = priv->plat->interface;
@@ -821,9 +864,10 @@ static int stmmac_init_phy(struct net_device *dev)
}
/**
- * stmmac_display_ring
- * @p: pointer to the ring.
+ * stmmac_display_ring: display ring
+ * @head: pointer to the head of the ring passed.
* @size: size of the ring.
+ * @extend_desc: to verify if extended descriptors are used.
* Description: display the control/status and buffer descriptors.
*/
static void stmmac_display_ring(void *head, int size, int extend_desc)
@@ -887,6 +931,12 @@ static int stmmac_set_bfsize(int mtu, int bufsize)
return ret;
}
+/**
+ * stmmac_clear_descriptors: clear descriptors
+ * @priv: driver private structure
+ * Description: this function is called to clear the tx and rx descriptors
+ * in case of both basic and extended descriptors are used.
+ */
static void stmmac_clear_descriptors(struct stmmac_priv *priv)
{
int i;
@@ -1129,7 +1179,7 @@ static void free_dma_desc_resources(struct stmmac_priv *priv)
/**
* stmmac_dma_operation_mode - HW DMA operation mode
- * @priv : pointer to the private device structure.
+ * @priv: driver private structure
* Description: it sets the DMA operation mode: tx/rx DMA thresholds
* or Store-And-Forward capability.
*/
@@ -1153,7 +1203,7 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
/**
* stmmac_tx_clean:
- * @priv: private data pointer
+ * @priv: driver private structure
* Description: it reclaims resources after transmission completes.
*/
static void stmmac_tx_clean(struct stmmac_priv *priv)
@@ -1245,8 +1295,8 @@ static inline void stmmac_disable_dma_irq(struct stmmac_priv *priv)
/**
- * stmmac_tx_err:
- * @priv: pointer to the private device structure
+ * stmmac_tx_err: irq tx error mng function
+ * @priv: driver private structure
* Description: it cleans the descriptors and restarts the transmission
* in case of errors.
*/
@@ -1275,6 +1325,14 @@ static void stmmac_tx_err(struct stmmac_priv *priv)
netif_wake_queue(priv->dev);
}
+/**
+ * stmmac_dma_interrupt: DMA ISR
+ * @priv: driver private structure
+ * Description: this is the DMA ISR. It is called by the main ISR.
+ * It calls the dwmac dma routine to understand which type of interrupt
+ * happened. In case of there is a Normal interrupt and either TX or RX
+ * interrupt happened so the NAPI is scheduled.
+ */
static void stmmac_dma_interrupt(struct stmmac_priv *priv)
{
int status;
@@ -1297,13 +1355,16 @@ static void stmmac_dma_interrupt(struct stmmac_priv *priv)
stmmac_tx_err(priv);
}
+/**
+ * stmmac_mmc_setup: setup the Mac Management Counters (MMC)
+ * @priv: driver private structure
+ * Description: this masks the MMC irq, in fact, the counters are managed in SW.
+ */
static void stmmac_mmc_setup(struct stmmac_priv *priv)
{
unsigned int mode = MMC_CNTRL_RESET_ON_READ | MMC_CNTRL_COUNTER_RESET |
MMC_CNTRL_PRESET | MMC_CNTRL_FULL_HALF_PRESET;
- /* Mask MMC irq, counters are managed in SW and registers
- * are cleared on each READ eventually. */
dwmac_mmc_intr_all_mask(priv->ioaddr);
if (priv->dma_cap.rmon) {
@@ -1332,9 +1393,11 @@ static u32 stmmac_get_synopsys_id(struct stmmac_priv *priv)
}
/**
- * stmmac_selec_desc_mode
- * @priv : private structure
- * Description: select the Enhanced/Alternate or Normal descriptors
+ * stmmac_selec_desc_mode: to select among: normal/alternate/extend descriptors
+ * @priv: driver private structure
+ * Description: select the Enhanced/Alternate or Normal descriptors.
+ * In case of Enhanced/Alternate, it looks at the extended descriptors are
+ * supported by the HW cap. register.
*/
static void stmmac_selec_desc_mode(struct stmmac_priv *priv)
{
@@ -1356,8 +1419,8 @@ static void stmmac_selec_desc_mode(struct stmmac_priv *priv)
}
/**
- * stmmac_get_hw_features
- * @priv : private device pointer
+ * stmmac_get_hw_features: get MAC capabilities from the HW cap. register.
+ * @priv: driver private structure
* Description:
* new GMAC chip generations have a new register to indicate the
* presence of the optional feature/functions.
@@ -1415,10 +1478,15 @@ static int stmmac_get_hw_features(struct stmmac_priv *priv)
return hw_cap;
}
+/**
+ * stmmac_check_ether_addr: check if the MAC addr is valid
+ * @priv: driver private structure
+ * Description:
+ * it is to verify if the MAC address is valid, in case of failures it
+ * generates a random MAC address
+ */
static void stmmac_check_ether_addr(struct stmmac_priv *priv)
{
- /* verify if the MAC address is valid, in case of failures it
- * generates a random MAC address */
if (!is_valid_ether_addr(priv->dev->dev_addr)) {
priv->hw->mac->get_umac_addr((void __iomem *)
priv->dev->base_addr,
@@ -1430,15 +1498,20 @@ static void stmmac_check_ether_addr(struct stmmac_priv *priv)
priv->dev->dev_addr);
}
+/**
+ * stmmac_init_dma_engine: DMA init.
+ * @priv: driver private structure
+ * Description:
+ * It inits the DMA invoking the specific MAC/GMAC callback.
+ * Some DMA parameters can be passed from the platform;
+ * in case of these are not passed a default is kept for the MAC or GMAC.
+ */
static int stmmac_init_dma_engine(struct stmmac_priv *priv)
{
int pbl = DEFAULT_DMA_PBL, fixed_burst = 0, burst_len = 0;
int mixed_burst = 0;
int atds = 0;
- /* Some DMA parameters can be passed from the platform;
- * in case of these are not passed we keep a default
- * (good for all the chips) and init the DMA! */
if (priv->plat->dma_cfg) {
pbl = priv->plat->dma_cfg->pbl;
fixed_burst = priv->plat->dma_cfg->fixed_burst;
@@ -1455,7 +1528,7 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv)
}
/**
- * stmmac_tx_timer:
+ * stmmac_tx_timer: mitigation sw timer for tx.
* @data: data pointer
* Description:
* This is the timer handler to directly invoke the stmmac_tx_clean.
@@ -1468,8 +1541,8 @@ static void stmmac_tx_timer(unsigned long data)
}
/**
- * stmmac_tx_timer:
- * @priv: private data structure
+ * stmmac_init_tx_coalesce: init tx mitigation options.
+ * @priv: driver private structure
* Description:
* This inits the transmit coalesce parameters: i.e. timer rate,
* timer handler and default threshold used for enabling the
@@ -1697,10 +1770,12 @@ static int stmmac_release(struct net_device *dev)
}
/**
- * stmmac_xmit:
+ * stmmac_xmit: Tx entry point of the driver
* @skb : the socket buffer
* @dev : device pointer
- * Description : Tx entry point of the driver.
+ * Description : this is the tx entry point of the driver.
+ * It programs the chain or the ring and supports oversized frames
+ * and SG feature.
*/
static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
{
@@ -1864,6 +1939,12 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
return NETDEV_TX_OK;
}
+/**
+ * stmmac_rx_refill: refill used skb preallocated buffers
+ * @priv: driver private structure
+ * Description : this is to reallocate the skb for the reception process
+ * that is based on zero-copy.
+ */
static inline void stmmac_rx_refill(struct stmmac_priv *priv)
{
unsigned int rxsize = priv->dma_rx_size;
@@ -1902,6 +1983,13 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
}
}
+/**
+ * stmmac_rx_refill: refill used skb preallocated buffers
+ * @priv: driver private structure
+ * @limit: napi bugget.
+ * Description : this the function called by the napi poll method.
+ * It gets all the frames inside the ring.
+ */
static int stmmac_rx(struct stmmac_priv *priv, int limit)
{
unsigned int rxsize = priv->dma_rx_size;
@@ -2166,6 +2254,14 @@ static netdev_features_t stmmac_fix_features(struct net_device *dev,
return features;
}
+/**
+ * stmmac_interrupt - main ISR
+ * @irq: interrupt number.
+ * @dev_id: to pass the net device pointer.
+ * Description: this is the main driver interrupt service routine.
+ * It calls the DMA ISR and also the core ISR to manage PMT, MMC, LPI
+ * interrupts.
+ */
static irqreturn_t stmmac_interrupt(int irq, void *dev_id)
{
struct net_device *dev = (struct net_device *)dev_id;
@@ -2214,7 +2310,7 @@ static void stmmac_poll_controller(struct net_device *dev)
* a proprietary structure used to pass information to the driver.
* @cmd: IOCTL command
* Description:
- * Currently it supports just the phy_mii_ioctl(...) and HW time stamping.
+ * Currently it supports the phy_mii_ioctl(...) and HW time stamping.
*/
static int stmmac_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
@@ -2447,7 +2543,7 @@ static const struct net_device_ops stmmac_netdev_ops = {
/**
* stmmac_hw_init - Init the MAC device
- * @priv : pointer to the private device structure.
+ * @priv: driver private structure
* Description: this function detects which MAC device
* (GMAC/MAC10-100) has to attached, checks the HW capability
* (if supported) and sets the driver's features (for example
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c
index 93d4bef..b8b0eee 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c
@@ -174,9 +174,7 @@ static struct ptp_clock_info stmmac_ptp_clock_ops = {
/**
* stmmac_ptp_register
- *
- * @ndev: net device pointer
- *
+ * @priv: driver private structure
* Description: this function will register the ptp clock driver
* to kernel. It also does some house keeping work.
*/
@@ -199,9 +197,7 @@ int stmmac_ptp_register(struct stmmac_priv *priv)
/**
* stmmac_ptp_unregister
- *
- * @ndev: net device pointer
- *
+ * @priv: driver private structure
* Description: this function will remove/unregister the ptp clock driver
* from the kernel.
*/
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 4/7] stmmac: review driver documentation
From: Giuseppe CAVALLARO @ 2013-04-03 5:41 UTC (permalink / raw)
To: netdev; +Cc: Giuseppe Cavallaro
In-Reply-To: <1364967689-11155-1-git-send-email-peppe.cavallaro@st.com>
This patch reviews the driver documentation file;
for example, there were some new fields (in the driver
module parameter section) and the ptp files were
not documented.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
Documentation/networking/stmmac.txt | 12 ++++++++----
1 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/Documentation/networking/stmmac.txt b/Documentation/networking/stmmac.txt
index 8efe0b3..654d2e5 100644
--- a/Documentation/networking/stmmac.txt
+++ b/Documentation/networking/stmmac.txt
@@ -1,6 +1,6 @@
STMicroelectronics 10/100/1000 Synopsys Ethernet driver
-Copyright (C) 2007-2010 STMicroelectronics Ltd
+Copyright (C) 2007-2013 STMicroelectronics Ltd
Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
This is the driver for the MAC 10/100/1000 on-chip Ethernet controllers
@@ -10,7 +10,7 @@ Currently this network device driver is for all STM embedded MAC/GMAC
(i.e. 7xxx/5xxx SoCs), SPEAr (arm), Loongson1B (mips) and XLINX XC2V3000
FF1152AMT0221 D1215994A VIRTEX FPGA board.
-DWC Ether MAC 10/100/1000 Universal version 3.60a (and older) and DWC Ether
+DWC Ether MAC 10/100/1000 Universal version 3.70a (and older) and DWC Ether
MAC 10/100 Universal version 4.0 have been used for developing this driver.
This driver supports both the platform bus and PCI.
@@ -32,6 +32,8 @@ The kernel configuration option is STMMAC_ETH:
watchdog: transmit timeout (in milliseconds);
flow_ctrl: Flow control ability [on/off];
pause: Flow Control Pause Time;
+ eee_timer: tx EEE timer;
+ chain_mode: select chain mode instead of ring.
3) Command line options
Driver parameters can be also passed in command line by using:
@@ -164,12 +166,12 @@ Where:
o bus_setup: perform HW setup of the bus. For example, on some ST platforms
this field is used to configure the AMBA bridge to generate more
efficient STBus traffic.
- o init/exit: callbacks used for calling a custom initialisation;
+ o init/exit: callbacks used for calling a custom initialization;
this is sometime necessary on some platforms (e.g. ST boxes)
where the HW needs to have set some PIO lines or system cfg
registers.
o custom_cfg/custom_data: this is a custom configuration that can be passed
- while initialising the resources.
+ while initializing the resources.
o bsp_priv: another private poiter.
For MDIO bus The we have:
@@ -273,6 +275,8 @@ reset procedure etc).
o norm_desc.c: functions for handling normal descriptors;
o chain_mode.c/ring_mode.c:: functions to manage RING/CHAINED modes;
o mmc_core.c/mmc.h: Management MAC Counters;
+ o stmmac_hwtstamp.c: HW timestamp support for PTP
+ o stmmac_ptp.c: PTP 1588 clock
5) Debug Information
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 2/7] stmmac: review barriers
From: Giuseppe CAVALLARO @ 2013-04-03 5:41 UTC (permalink / raw)
To: netdev; +Cc: Giuseppe Cavallaro, Deepak Sikri, Shiraz Hashim
In-Reply-To: <1364967689-11155-1-git-send-email-peppe.cavallaro@st.com>
In all my tests performed on SH4 and ARM A9 platforms, I've never met problems
that can be fixed by using memory barriers. In the past there was some issues
on SMP ARM but fixed by reviewing xmit spinlock.
Further barriers have been added in the commits too: 8e83989106562326bf
This patch is to use the smp_wbm instead of wbm because the driver
runs on UP systems. Then, IMO it could make sense to only maintain the barriers
just in places where we touch the dma owner bits (that is the
only real critical path as we had seen and fixed in the commit:
eb0dc4bb2e22c04964d).
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Cc: Deepak Sikri <deepak.sikri@st.com>
Cc: Shiraz Hashim <shiraz.hashim@st.com>
---
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 9 +++------
1 files changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 8b69e3b..c92dcbc 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1797,15 +1797,13 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
priv->tx_skbuff[entry] = NULL;
priv->hw->desc->prepare_tx_desc(desc, 0, len, csum_insertion,
priv->mode);
- wmb();
+ smp_wmb();
priv->hw->desc->set_tx_owner(desc);
- wmb();
}
/* Finalize the latest segment. */
priv->hw->desc->close_tx_desc(desc);
- wmb();
/* According to the coalesce parameter the IC bit for the latest
* segment could be reset and the timer re-started to invoke the
* stmmac_tx function. This approach takes care about the fragments.
@@ -1821,9 +1819,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
} else
priv->tx_count_frames = 0;
+ smp_wmb();
/* To avoid raise condition */
priv->hw->desc->set_tx_owner(first);
- wmb();
priv->cur_tx++;
@@ -1899,9 +1897,8 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
RX_DBG(KERN_INFO "\trefill entry #%d\n", entry);
}
- wmb();
+ smp_wmb();
priv->hw->desc->set_rx_owner(p);
- wmb();
}
}
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 3/7] stmmac: review private structure fields
From: Giuseppe CAVALLARO @ 2013-04-03 5:41 UTC (permalink / raw)
To: netdev; +Cc: Giuseppe Cavallaro
In-Reply-To: <1364967689-11155-1-git-send-email-peppe.cavallaro@st.com>
recently many new supports have been added in the stmmac driver w/o taking care
about where each new field had to be placed inside the private structure for
guaranteeing the best cache usage.
This is what I wanted in the beginning, so this patch reorganizes all the fields
in order to keep adjacent fields for cache effect.
I have also tried to optimize them by using pahole.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 70 +++++++++++++-------------
1 files changed, 35 insertions(+), 35 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 75f997b..8aa28c5 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -35,36 +35,45 @@
struct stmmac_priv {
/* Frequently used values are kept adjacent for cache effect */
- struct dma_desc *dma_tx ____cacheline_aligned; /* Basic TX desc */
- struct dma_extended_desc *dma_etx; /* Extended TX descriptor */
- dma_addr_t dma_tx_phy;
- struct sk_buff **tx_skbuff;
- dma_addr_t *tx_skbuff_dma;
+ struct dma_extended_desc *dma_etx;
+ struct dma_desc *dma_tx ____cacheline_aligned_in_smp;
+ struct sk_buff **tx_skbuff ____cacheline_aligned_in_smp;
unsigned int cur_tx;
unsigned int dirty_tx;
unsigned int dma_tx_size;
+ u32 tx_count_frames;
+ u32 tx_coal_frames;
+ u32 tx_coal_timer;
+ dma_addr_t *tx_skbuff_dma;
+ dma_addr_t dma_tx_phy;
int tx_coalesce;
+ int hwts_tx_en;
+ spinlock_t tx_lock;
+ bool tx_path_in_lpi_mode;
+ struct timer_list txtimer;
- struct dma_desc *dma_rx; /* Basic RX descriptor */
- struct dma_extended_desc *dma_erx; /* Extended RX descriptor */
+ struct dma_desc *dma_rx ____cacheline_aligned_in_smp;
+ struct dma_extended_desc *dma_erx ____cacheline_aligned_in_smp;
+ struct sk_buff **rx_skbuff ____cacheline_aligned_in_smp;
unsigned int cur_rx;
unsigned int dirty_rx;
- struct sk_buff **rx_skbuff;
+ unsigned int dma_rx_size;
+ unsigned int dma_buf_sz;
+ u32 rx_riwt;
+ int hwts_rx_en;
dma_addr_t *rx_skbuff_dma;
+ dma_addr_t dma_rx_phy;
+ struct napi_struct napi ____cacheline_aligned_in_smp;
+
+ void __iomem *ioaddr ____cacheline_aligned_in_smp;
struct net_device *dev;
- dma_addr_t dma_rx_phy;
- unsigned int dma_rx_size;
- unsigned int dma_buf_sz;
struct device *device;
struct mac_device_info *hw;
- void __iomem *ioaddr;
-
- struct stmmac_extra_stats xstats;
- struct napi_struct napi;
int no_csum_insertion;
+ spinlock_t lock;
- struct phy_device *phydev;
+ struct phy_device *phydev ____cacheline_aligned_in_smp;
int oldlink;
int speed;
int oldduplex;
@@ -73,39 +82,30 @@ struct stmmac_priv {
struct mii_bus *mii;
int mii_irq[PHY_MAX_ADDR];
- u32 msg_enable;
- spinlock_t lock;
- spinlock_t tx_lock;
- int wolopts;
- int wol_irq;
+ struct stmmac_extra_stats xstats ____cacheline_aligned_in_smp;
struct plat_stmmacenet_data *plat;
- struct stmmac_counters mmc;
struct dma_features dma_cap;
+ struct stmmac_counters mmc;
int hw_cap_support;
+ int synopsys_id;
+ u32 msg_enable;
+ int wolopts;
+ int wol_irq;
struct clk *stmmac_clk;
int clk_csr;
- int synopsys_id;
struct timer_list eee_ctrl_timer;
- bool tx_path_in_lpi_mode;
int lpi_irq;
int eee_enabled;
int eee_active;
int tx_lpi_timer;
- struct timer_list txtimer;
- u32 tx_count_frames;
- u32 tx_coal_frames;
- u32 tx_coal_timer;
- int use_riwt;
- u32 rx_riwt;
+ int pcs;
unsigned int mode;
int extend_desc;
- int pcs;
- int hwts_tx_en;
- int hwts_rx_en;
- unsigned int default_addend;
- u32 adv_ts;
struct ptp_clock *ptp_clock;
struct ptp_clock_info ptp_clock_ops;
+ unsigned int default_addend;
+ u32 adv_ts;
+ int use_riwt;
spinlock_t ptp_lock;
};
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 1/7] stmmac: review napi gro support
From: Giuseppe CAVALLARO @ 2013-04-03 5:41 UTC (permalink / raw)
To: netdev; +Cc: Giuseppe Cavallaro
This patch is to:
o use napi_gro_flush() before calling __napi_complete()
o turn on NETIF_F_GRO by default
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 6b26d31..8b69e3b 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -2046,7 +2046,8 @@ static int stmmac_poll(struct napi_struct *napi, int budget)
work_done = stmmac_rx(priv, budget);
if (work_done < budget) {
- napi_complete(napi);
+ napi_gro_flush(napi, false);
+ __napi_complete(napi);
stmmac_enable_dma_irq(priv);
}
return work_done;
@@ -2586,7 +2587,7 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device,
ndev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
NETIF_F_RXCSUM;
- ndev->features |= ndev->hw_features | NETIF_F_HIGHDMA;
+ ndev->features |= ndev->hw_features | NETIF_F_HIGHDMA | NETIF_F_GRO;
ndev->watchdog_timeo = msecs_to_jiffies(watchdog);
#ifdef STMMAC_VLAN_TAG_USED
/* Both mac100 and gmac support receive VLAN tag detection */
--
1.7.4.4
^ permalink raw reply related
* [PATCH v2.24] datapath: Add basic MPLS support to kernel
From: Simon Horman @ 2013-04-03 5:38 UTC (permalink / raw)
To: dev, netdev; +Cc: Ravi K, Isaku Yamahata, Jesse Gross, Ben Pfaff
Allow datapath to recognize and extract MPLS labels into flow keys
and execute actions which push, pop, and set labels on packets.
Based heavily on work by Leo Alterman and Ravi K.
Cc: Ravi K <rkerur@gmail.com>
Cc: Leo Alterman <lalterman@nicira.com>
Reviewed-by: Isaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
This is the remaining patch of the series "MPLS actions and matches".
It is available in git at:
git://github.com/horms/openvswitch.git devel/mpls-v2.24
v2.24
* Use skb_mac_header() in set_ethertype()
* Set skb->encapsulation in set_ethertype() to support MPLS GSO.
Also add a note about the other requirements for MPLS GSO.
MPLS GSO support will be posted as a patch net-next (Linux mainline)
"MPLS: Add limited GSO support"
* Do not add ETH_TYPE_MIN, it is no longer used
v2.23
* As suggested by Jesse Gross:
- Verify the current ethernet type when validating sample actions
both for the taken and not-taken path if the sample action.
- Document that the OVS_KEY_ATTR_MPLS attribute accepts a list of
struct ovs_key_mpls but that an implementation may restrict
the length it accepts.
- Restrict the array length of the OVS_KEY_ATTR_MPLS to one.
+ Don't add ovs_flow_verify_key_len as it was added to
handle attributes whose values are arrays but there are
no attributes with values that are arrays (of length greater than one).
v2.22
* As suggested by Jesse Gross:
- Fix sparse warning in validate_and_copy_actions()
I have no idea why sparse doesn't show this up this on my system.
- Remove call to skb_cow_head() from push_mpls() as it
is already covered by a call to make_writable()
- Check (key_type > OVS_KEY_ATTR_MAX) in ovs_flow_verify_key_len()
- Disallow set actions on l2.5+ data and MPLS push and pop actions
after an MPLS pop action as there is no verification that the packet
is actually of the new ethernet type. This may later be supported
using recirculation or by other means.
- Do not add spurious debuging message to ovs_flow_cmd_new_or_set()
v2.21
* As suggested by Jesse Gross:
- Verify that l3 and l4 actions always always occur prior to
a push_mpls action and use the network header pointer of an skb
to track the top of the MPLS stack. This avoids adding an l2_size
element to the skb callback.
v2.20
* As suggested by Jesse Gross:
- Do not add ovs_dp_ioctl_hook
+ This appears to be garbage from a rebase
- Do not add skb_cb_set_l2_size. Instead set OVS_CB(skb)->l2_size
in ovs_flow_extract().
- Do not free skb on error in push_mpls(), it is freed in the caller
- Call skb_reset_mac_len() in pop_mpls() and push_mpls()
- Update checksums in pop_mpls(), push_mpls() and set_mpls().
- Rename skb_cb_mpls_bos() as skb_cb_mpls_stack().
It returns the top not the bottom of the stack.
- Track the current eth_type in validate_and_copy_actions
which is initially the eth_type of the flow and may be modified
by push_mpls and pop_mpls actions. Use this to correctly validate
mpls_set actions. This is to allow mpls_set actions to be applied
to a non-MPLS frame after an mpls_push action (although ovs-vswitchd
doesn't currently do that).
Also:
+ Remove the check of the eth_type in set_mpls() as the new validation
scheme should ensure it cannot be incorrect.
+ Use the current eth_type to validate mpls_pop actions and remove
the eth_type check from pop_mpls().
- Move OVS_KEY_ATTR_MPLS to non-upstream group in ovs_key_lens
- Remove unnecessary memset of mpls_key in ovs_flow_to_nlattrs()
- Make a union of the mpls and ip elements of struct sw_flow_key.
Currently the code stops parsing after an MPLS header so it is
not possible for the ip and mpls elements to be used simultaneously
and some space can be saved by using a union.
- Allow an array of MPLS key attributes
+ Currently all but the first element is ignored
+ User-space needs to be updated to accept more than one element,
currently it will treat their presence as an error
- Do not update network header in ovs_flow_extract() for after parsing
the MPLS stack as it is never used because no l3+ processing
occurs on MPLS frames.
- Allow multiple MPLS entries in a match by allowing the OVS_KEY_ATTR_MPLS
to be an array of struct ovs_key_mpls with at least one entry.
Currently only one entry is used which is byte-for-byte compatible with
the previous scheme of having OVS_KEY_ATTR_MPLS as a struct
ovs_key_mpls.
* Make skb writable in pop_mpls(), push_mpls() and set_mpls().
v2.18 - v2.19
* No change
v2.17
* As suggested by Ben Pfaff
- Use consistent terminology for MPLS.
+ Consistently refer to the MPLS component of a packet as the
MPLS label stack and entries in the stack as MPLS label stack entries
(LSE). An MPLS label is a component of an MPLS label stack entry.
The other components are the traffic class (TC), time to live (TTL)
and bottom of stack (BoS) bit.
- Rename compose_.*mpls_ functions as execute_.*mpls_
v2.16
* No change
v2.15
* As suggested by Ben Pfaff
- Use OVS_ACTION_SET to set OVS_KEY_ATTR_MPLS instead of
OVS_ACTION_ATTR_SET_MPLS
v2.14
* Remove include/linux/openvswitch.h portion which added add
new key and action attributes. This
now present in "User-Space MPLS actions and matches"
which is now a dependency of this patch
v2.13
* As suggested by Jarno Rajahalme
- Rename mpls_bos element of ovs_skb_cb as l2_size as it is set and used
regardless of if an MPLS stack is present or not. Update the name of
helper functions and documentation accordingly.
- Ensure that skb_cb_mpls_bos() never returns NULL
* Correct endieness in eth_p_mpls()
v2.12
* Update skb and network header on MPLS extraction in ovs_flow_extract()
* Use NULL in skb_cb_mpls_bos()
* Add eth_p_mpls helper
v2.10 - v2.11
* No change
v2.9
* datapath: Always update the mpls bos if vlan_pop is successful
Regardless of the details of how a successful
vlan_pop is achieved, the mpls bos needs to be updated.
Without this fix it has been observed that the following
results in malformed packets
v2.8
* No change
v2.7
* Rebase
v2.6
* As suggested by Yamahata-san
- Do not guard against label == 0 for
OVS_ACTION_ATTR_SET_MPLS in validate_actions().
A label of 0 is valid
- Remove comment stupulating that if
the top_label element of struct sw_flow_key is 0 then
there is no MPLS label. An MPLS label of 0 is valid
and the correct check if ethertype is
ntohs(ETH_TYPE_MPLS) or ntohs(ETH_TYPE_MPLS_MCAST)
v2.4 - v2.5
* No change
v2.3
* s/mpls_stack/mpls_bos/
This is in keeping with the naming used in the OpenFlow 1.3 specification
v2.2
* Call skb_reset_mac_header() in skb_cb_set_mpls_stack()
eth_hdr(skb) is non-NULL when called in skb_cb_set_mpls_stack().
* Add a call to skb_cb_set_mpls_stack() in ovs_packet_cmd_execute().
I apologise that I have mislaid my notes on this but
it avoids a kernel panic. I can investigate again if necessary.
* Use struct ovs_action_push_mpls instead of
__be16 to decode OVS_ACTION_ATTR_PUSH_MPLS in validate_actions(). This is
consistent with the data format for the attribute.
* Indentation fix in skb_cb_mpls_stack(). [cosmetic]
v2.1
* Manual rebase
---
datapath/actions.c | 106 ++++++++++++++++++
datapath/datapath.c | 148 +++++++++++++++++++++-----
datapath/datapath.h | 2 +
datapath/flow.c | 28 +++++
datapath/flow.h | 25 +++--
datapath/linux/compat/include/linux/skbuff.h | 12 +++
include/linux/openvswitch.h | 6 +-
lib/odp-util.c | 8 +-
8 files changed, 297 insertions(+), 38 deletions(-)
diff --git a/datapath/actions.c b/datapath/actions.c
index 0dac658..e9634fe 100644
--- a/datapath/actions.c
+++ b/datapath/actions.c
@@ -48,6 +48,97 @@ static int make_writable(struct sk_buff *skb, int write_len)
return pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
}
+static void set_ethertype(struct sk_buff *skb, const __be16 ethertype)
+{
+ struct ethhdr *hdr = (struct ethhdr *)skb_mac_header(skb);
+ if (hdr->h_proto == ethertype)
+ return;
+ if (!eth_p_mpls(hdr->h_proto)) {
+ /* This implies we are pushing adding an MPLS label stack
+ * to a previously non-MPLS packet. Set the encapsulation
+ * bit to allow MPLS GSO segmentation. It will make use
+ * of hdr->h_proto, set to the new MPLS ethertype and
+ * skb->protocol which is set to the old non-MPLS ethertype. */
+ skb_set_encapsulation(skb);
+ }
+ hdr->h_proto = ethertype;
+ if (get_ip_summed(skb) == OVS_CSUM_COMPLETE) {
+ __be16 diff[] = { ~hdr->h_proto, ethertype };
+ skb->csum = ~csum_partial((char *)diff, sizeof(diff),
+ ~skb->csum);
+ }
+}
+
+static int push_mpls(struct sk_buff *skb, const struct ovs_action_push_mpls *mpls)
+{
+ __be32 *new_mpls_lse;
+ int err;
+
+ err = make_writable(skb, skb->mac_len + MPLS_HLEN);
+ if (unlikely(err))
+ return err;
+
+ skb_push(skb, MPLS_HLEN);
+ memmove(skb_mac_header(skb) - MPLS_HLEN, skb_mac_header(skb),
+ skb->mac_len);
+ skb_reset_mac_header(skb);
+ skb_set_network_header(skb, skb->mac_len);
+
+ new_mpls_lse = (__be32 *)skb_network_header(skb);
+ *new_mpls_lse = mpls->mpls_lse;
+
+ if (get_ip_summed(skb) == OVS_CSUM_COMPLETE)
+ skb->csum = csum_add(skb->csum, csum_partial(new_mpls_lse,
+ MPLS_HLEN, 0));
+
+ set_ethertype(skb, mpls->mpls_ethertype);
+ return 0;
+}
+
+static int pop_mpls(struct sk_buff *skb, const __be16 *ethertype)
+{
+ int err;
+
+ err = make_writable(skb, skb->mac_len + MPLS_HLEN);
+ if (unlikely(err))
+ return err;
+
+ if (get_ip_summed(skb) == OVS_CSUM_COMPLETE)
+ skb->csum = csum_sub(skb->csum,
+ csum_partial(skb_network_header(skb),
+ MPLS_HLEN, 0));
+
+ memmove(skb_mac_header(skb) + MPLS_HLEN, skb_mac_header(skb),
+ skb->mac_len);
+
+ skb_pull(skb, MPLS_HLEN);
+ skb_reset_mac_header(skb);
+ skb_set_network_header(skb, skb->mac_len);
+
+ set_ethertype(skb, *ethertype);
+ return 0;
+}
+
+static int set_mpls(struct sk_buff *skb, const __be32 *mpls_lse)
+{
+ __be32 *stack = (__be32 *)skb_network_header(skb);
+ int err;
+
+ err = make_writable(skb, skb->mac_len + MPLS_HLEN);
+ if (unlikely(err))
+ return err;
+
+ if (get_ip_summed(skb) == OVS_CSUM_COMPLETE) {
+ __be32 diff[] = { ~(*stack), *mpls_lse };
+ skb->csum = ~csum_partial((char *)diff, sizeof(diff),
+ ~skb->csum);
+ }
+
+ *stack = *mpls_lse;
+
+ return 0;
+}
+
/* remove VLAN header from packet and update csum accordingly. */
static int __pop_vlan_tci(struct sk_buff *skb, __be16 *current_tci)
{
@@ -115,6 +206,9 @@ static int push_vlan(struct sk_buff *skb, const struct ovs_action_push_vlan *vla
if (!__vlan_put_tag(skb, current_tag))
return -ENOMEM;
+ /* update mac_len for MPLS functions */
+ skb_reset_mac_len(skb);
+
if (get_ip_summed(skb) == OVS_CSUM_COMPLETE)
skb->csum = csum_add(skb->csum, csum_partial(skb->data
+ (2 * ETH_ALEN), VLAN_HLEN, 0));
@@ -459,6 +553,10 @@ static int execute_set_action(struct sk_buff *skb,
case OVS_KEY_ATTR_UDP:
err = set_udp(skb, nla_data(nested_attr));
break;
+
+ case OVS_KEY_ATTR_MPLS:
+ err = set_mpls(skb, nla_data(nested_attr));
+ break;
}
return err;
@@ -494,6 +592,14 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb,
output_userspace(dp, skb, a);
break;
+ case OVS_ACTION_ATTR_PUSH_MPLS:
+ err = push_mpls(skb, nla_data(a));
+ break;
+
+ case OVS_ACTION_ATTR_POP_MPLS:
+ err = pop_mpls(skb, nla_data(a));
+ break;
+
case OVS_ACTION_ATTR_PUSH_VLAN:
err = push_vlan(skb, nla_data(a));
if (unlikely(err)) /* skb already freed. */
diff --git a/datapath/datapath.c b/datapath/datapath.c
index 9cd4b0e..e8be795 100644
--- a/datapath/datapath.c
+++ b/datapath/datapath.c
@@ -528,13 +528,26 @@ static inline void add_nested_action_end(struct sw_flow_actions *sfa, int st_off
a->nla_len = sfa->actions_len - st_offset;
}
-static int validate_and_copy_actions(const struct nlattr *attr,
+struct eth_types {
+ size_t depth;
+ __be16 types[SAMPLE_ACTION_DEPTH];
+};
+
+static void eth_types_set(struct eth_types *types, size_t depth, __be16 type)
+{
+ types->depth = depth;
+ types->types[depth] = type;
+}
+
+static int validate_and_copy_actions__(const struct nlattr *attr,
const struct sw_flow_key *key, int depth,
- struct sw_flow_actions **sfa);
+ struct sw_flow_actions **sfa,
+ struct eth_types *eth_types);
static int validate_and_copy_sample(const struct nlattr *attr,
const struct sw_flow_key *key, int depth,
- struct sw_flow_actions **sfa)
+ struct sw_flow_actions **sfa,
+ struct eth_types *eth_types)
{
const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1];
const struct nlattr *probability, *actions;
@@ -570,7 +583,8 @@ static int validate_and_copy_sample(const struct nlattr *attr,
if (st_acts < 0)
return st_acts;
- err = validate_and_copy_actions(actions, key, depth + 1, sfa);
+ err = validate_and_copy_actions__(actions, key, depth + 1, sfa,
+ eth_types);
if (err)
return err;
@@ -580,12 +594,12 @@ static int validate_and_copy_sample(const struct nlattr *attr,
return 0;
}
-static int validate_tp_port(const struct sw_flow_key *flow_key)
+static int validate_tp_port(const struct sw_flow_key *flow_key, __be16 eth_type)
{
- if (flow_key->eth.type == htons(ETH_P_IP)) {
+ if (eth_type == htons(ETH_P_IP)) {
if (flow_key->ipv4.tp.src || flow_key->ipv4.tp.dst)
return 0;
- } else if (flow_key->eth.type == htons(ETH_P_IPV6)) {
+ } else if (eth_type == htons(ETH_P_IPV6)) {
if (flow_key->ipv6.tp.src || flow_key->ipv6.tp.dst)
return 0;
}
@@ -616,7 +630,7 @@ static int validate_and_copy_set_tun(const struct nlattr *attr,
static int validate_set(const struct nlattr *a,
const struct sw_flow_key *flow_key,
struct sw_flow_actions **sfa,
- bool *set_tun)
+ bool *set_tun, struct eth_types *eth_types)
{
const struct nlattr *ovs_key = nla_data(a);
int key_type = nla_type(ovs_key);
@@ -653,9 +667,12 @@ static int validate_set(const struct nlattr *a,
return err;
break;
- case OVS_KEY_ATTR_IPV4:
- if (flow_key->eth.type != htons(ETH_P_IP))
- return -EINVAL;
+ case OVS_KEY_ATTR_IPV4: {
+ size_t i;
+
+ for (i = 0; i < eth_types->depth; i++)
+ if (eth_types->types[i] != htons(ETH_P_IP))
+ return -EINVAL;
if (!flow_key->ip.proto)
return -EINVAL;
@@ -668,10 +685,14 @@ static int validate_set(const struct nlattr *a,
return -EINVAL;
break;
+ }
- case OVS_KEY_ATTR_IPV6:
- if (flow_key->eth.type != htons(ETH_P_IPV6))
- return -EINVAL;
+ case OVS_KEY_ATTR_IPV6: {
+ size_t i;
+
+ for (i = 0; i < eth_types->depth; i++)
+ if (eth_types->types[i] != htons(ETH_P_IPV6))
+ return -EINVAL;
if (!flow_key->ip.proto)
return -EINVAL;
@@ -687,18 +708,37 @@ static int validate_set(const struct nlattr *a,
return -EINVAL;
break;
+ }
+
+ case OVS_KEY_ATTR_TCP: {
+ size_t i;
- case OVS_KEY_ATTR_TCP:
if (flow_key->ip.proto != IPPROTO_TCP)
return -EINVAL;
- return validate_tp_port(flow_key);
+ for (i = 0; i < eth_types->depth; i++)
+ if (validate_tp_port(flow_key, eth_types->types[i]))
+ return -EINVAL;
+ }
- case OVS_KEY_ATTR_UDP:
+ case OVS_KEY_ATTR_UDP: {
+ size_t i;
if (flow_key->ip.proto != IPPROTO_UDP)
return -EINVAL;
- return validate_tp_port(flow_key);
+ for (i = 0; i < eth_types->depth; i++)
+ if (validate_tp_port(flow_key, eth_types->types[i]))
+ return -EINVAL;
+ }
+
+ case OVS_KEY_ATTR_MPLS: {
+ size_t i;
+
+ for (i = 0; i < eth_types->depth; i++)
+ if (!eth_p_mpls(eth_types->types[i]))
+ return -EINVAL;
+ break;
+ }
default:
return -EINVAL;
@@ -742,10 +782,10 @@ static int copy_action(const struct nlattr *from,
return 0;
}
-static int validate_and_copy_actions(const struct nlattr *attr,
- const struct sw_flow_key *key,
- int depth,
- struct sw_flow_actions **sfa)
+static int validate_and_copy_actions__(const struct nlattr *attr,
+ const struct sw_flow_key *key, int depth,
+ struct sw_flow_actions **sfa,
+ struct eth_types *eth_types)
{
const struct nlattr *a;
int rem, err;
@@ -753,11 +793,29 @@ static int validate_and_copy_actions(const struct nlattr *attr,
if (depth >= SAMPLE_ACTION_DEPTH)
return -EOVERFLOW;
+ /* Due to the sample action there may be more than one possibility
+ * for the current ethernet type. They all need to be verified.
+ *
+ * This is handled by tracking a stack of ethernet types, one for
+ * each (sample) depth of validation. Here the ethernet type for
+ * the current depth is pushed onto the stack. It may be modified
+ * as by actions are validated. When a modification occurs the
+ * ethernet types for higher stack-depths are popped off the stack.
+ * All entries on the stack are checked when validating the
+ * ethernet type required by an action.
+ */
+ if (!depth)
+ eth_types_set(eth_types, 0, key->eth.type);
+ else
+ eth_types_set(eth_types, depth, eth_types->types[depth - 1]);
+
nla_for_each_nested(a, attr, rem) {
/* Expected argument lengths, (u32)-1 for variable length. */
static const u32 action_lens[OVS_ACTION_ATTR_MAX + 1] = {
[OVS_ACTION_ATTR_OUTPUT] = sizeof(u32),
[OVS_ACTION_ATTR_USERSPACE] = (u32)-1,
+ [OVS_ACTION_ATTR_PUSH_MPLS] = sizeof(struct ovs_action_push_mpls),
+ [OVS_ACTION_ATTR_POP_MPLS] = sizeof(__be16),
[OVS_ACTION_ATTR_PUSH_VLAN] = sizeof(struct ovs_action_push_vlan),
[OVS_ACTION_ATTR_POP_VLAN] = 0,
[OVS_ACTION_ATTR_SET] = (u32)-1,
@@ -788,6 +846,35 @@ static int validate_and_copy_actions(const struct nlattr *attr,
return -EINVAL;
break;
+ case OVS_ACTION_ATTR_PUSH_MPLS: {
+ const struct ovs_action_push_mpls *mpls = nla_data(a);
+ if (!eth_p_mpls(mpls->mpls_ethertype))
+ return -EINVAL;
+ eth_types_set(eth_types, depth, mpls->mpls_ethertype);
+ break;
+ }
+
+ case OVS_ACTION_ATTR_POP_MPLS: {
+ size_t i;
+
+ for (i = 0; i < eth_types->depth; i++)
+ if (eth_types->types[i] != htons(ETH_P_IP))
+ return -EINVAL;
+
+ /* Disallow subsequent l2.5+ set and mpls_pop actions
+ * as there is no check here to ensure that the new
+ * eth_type is valid and thus set actions could
+ * write off the end of the packet or otherwise
+ * corrupt it.
+ *
+ * Support for these actions that after mpls_pop
+ * using packet recirculation is planned.
+ * are planned to be supported using using packet
+ * recirculation.
+ */
+ eth_types_set(eth_types, depth, ntohs(0));
+ break;
+ }
case OVS_ACTION_ATTR_POP_VLAN:
break;
@@ -801,13 +888,14 @@ static int validate_and_copy_actions(const struct nlattr *attr,
break;
case OVS_ACTION_ATTR_SET:
- err = validate_set(a, key, sfa, &skip_copy);
+ err = validate_set(a, key, sfa, &skip_copy, eth_types);
if (err)
return err;
break;
case OVS_ACTION_ATTR_SAMPLE:
- err = validate_and_copy_sample(a, key, depth, sfa);
+ err = validate_and_copy_sample(a, key, depth, sfa,
+ eth_types);
if (err)
return err;
skip_copy = true;
@@ -829,6 +917,14 @@ static int validate_and_copy_actions(const struct nlattr *attr,
return 0;
}
+static int validate_and_copy_actions(const struct nlattr *attr,
+ const struct sw_flow_key *key,
+ struct sw_flow_actions **sfa)
+{
+ struct eth_types eth_type;
+ return validate_and_copy_actions__(attr, key, 0, sfa, ð_type);
+}
+
static void clear_stats(struct sw_flow *flow)
{
flow->used = 0;
@@ -893,7 +989,7 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info)
if (IS_ERR(acts))
goto err_flow_free;
- err = validate_and_copy_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, 0, &acts);
+ err = validate_and_copy_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, &acts);
rcu_assign_pointer(flow->sf_acts, acts);
if (err)
goto err_flow_free;
@@ -1231,7 +1327,7 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info)
if (IS_ERR(acts))
goto error;
- error = validate_and_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, 0, &acts);
+ error = validate_and_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, &acts);
if (error)
goto err_kfree;
} else if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW) {
diff --git a/datapath/datapath.h b/datapath/datapath.h
index 9bc98fb..7665742 100644
--- a/datapath/datapath.h
+++ b/datapath/datapath.h
@@ -189,4 +189,6 @@ struct sk_buff *ovs_vport_cmd_build_info(struct vport *, u32 portid, u32 seq,
u8 cmd);
int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb);
+
+unsigned char *skb_cb_mpls_stack(const struct sk_buff *skb);
#endif /* datapath.h */
diff --git a/datapath/flow.c b/datapath/flow.c
index 841e8be..8a4cb84 100644
--- a/datapath/flow.c
+++ b/datapath/flow.c
@@ -648,6 +648,7 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key,
return -ENOMEM;
skb_reset_network_header(skb);
+ skb_reset_mac_len(skb);
__skb_push(skb, skb->data - skb_mac_header(skb));
/* Network layer. */
@@ -730,6 +731,13 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key,
memcpy(key->ipv4.arp.tha, arp->ar_tha, ETH_ALEN);
key_len = SW_FLOW_KEY_OFFSET(ipv4.arp);
}
+ } else if (eth_p_mpls(key->eth.type)) {
+ error = check_header(skb, MPLS_HLEN);
+ if (unlikely(error))
+ goto out;
+
+ key_len = SW_FLOW_KEY_OFFSET(mpls.top_lse);
+ memcpy(&key->mpls.top_lse, skb_network_header(skb), MPLS_HLEN);
} else if (key->eth.type == htons(ETH_P_IPV6)) {
int nh_len; /* IPv6 Header + Extensions */
@@ -848,6 +856,9 @@ const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = {
[OVS_KEY_ATTR_ARP] = sizeof(struct ovs_key_arp),
[OVS_KEY_ATTR_ND] = sizeof(struct ovs_key_nd),
[OVS_KEY_ATTR_TUNNEL] = -1,
+
+ /* Not upstream. */
+ [OVS_KEY_ATTR_MPLS] = sizeof(struct ovs_key_mpls),
};
static int ipv4_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_len,
@@ -1254,6 +1265,15 @@ int ovs_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_lenp,
swkey->ip.proto = ntohs(arp_key->arp_op);
memcpy(swkey->ipv4.arp.sha, arp_key->arp_sha, ETH_ALEN);
memcpy(swkey->ipv4.arp.tha, arp_key->arp_tha, ETH_ALEN);
+ } else if (eth_p_mpls(swkey->eth.type)) {
+ const struct ovs_key_mpls *mpls_key;
+ if (!(attrs & (1ULL << OVS_KEY_ATTR_MPLS)))
+ return -EINVAL;
+ attrs &= ~(1ULL << OVS_KEY_ATTR_MPLS);
+
+ key_len = SW_FLOW_KEY_OFFSET(mpls.top_lse);
+ mpls_key = nla_data(a[OVS_KEY_ATTR_MPLS]);
+ swkey->mpls.top_lse = mpls_key->mpls_lse;
}
if (attrs)
@@ -1420,6 +1440,14 @@ int ovs_flow_to_nlattrs(const struct sw_flow_key *swkey, struct sk_buff *skb)
arp_key->arp_op = htons(swkey->ip.proto);
memcpy(arp_key->arp_sha, swkey->ipv4.arp.sha, ETH_ALEN);
memcpy(arp_key->arp_tha, swkey->ipv4.arp.tha, ETH_ALEN);
+ } else if (eth_p_mpls(swkey->eth.type)) {
+ struct ovs_key_mpls *mpls_key;
+
+ nla = nla_reserve(skb, OVS_KEY_ATTR_MPLS, sizeof(*mpls_key));
+ if (!nla)
+ goto nla_put_failure;
+ mpls_key = nla_data(nla);
+ mpls_key->mpls_lse = swkey->mpls.top_lse;
}
if ((swkey->eth.type == htons(ETH_P_IP) ||
diff --git a/datapath/flow.h b/datapath/flow.h
index dba66cf..2e3083b 100644
--- a/datapath/flow.h
+++ b/datapath/flow.h
@@ -72,12 +72,17 @@ struct sw_flow_key {
__be16 tci; /* 0 if no VLAN, VLAN_TAG_PRESENT set otherwise. */
__be16 type; /* Ethernet frame type. */
} eth;
- struct {
- u8 proto; /* IP protocol or lower 8 bits of ARP opcode. */
- u8 tos; /* IP ToS. */
- u8 ttl; /* IP TTL/hop limit. */
- u8 frag; /* One of OVS_FRAG_TYPE_*. */
- } ip;
+ union {
+ struct {
+ __be32 top_lse; /* top label stack entry */
+ } mpls;
+ struct {
+ u8 proto; /* IP protocol or lower 8 bits of ARP opcode. */
+ u8 tos; /* IP ToS. */
+ u8 ttl; /* IP TTL/hop limit. */
+ u8 frag; /* One of OVS_FRAG_TYPE_*. */
+ } ip;
+ };
union {
struct {
struct {
@@ -143,6 +148,8 @@ struct arp_eth_header {
unsigned char ar_tip[4]; /* target IP address */
} __packed;
+#define MPLS_HLEN 4
+
int ovs_flow_init(void);
void ovs_flow_exit(void);
@@ -204,4 +211,10 @@ int ipv4_tun_from_nlattr(const struct nlattr *attr,
int ipv4_tun_to_nlattr(struct sk_buff *skb,
const struct ovs_key_ipv4_tunnel *tun_key);
+static inline bool eth_p_mpls(__be16 eth_type)
+{
+ return eth_type == htons(ETH_P_MPLS_UC) ||
+ eth_type == htons(ETH_P_MPLS_MC);
+}
+
#endif /* flow.h */
diff --git a/datapath/linux/compat/include/linux/skbuff.h b/datapath/linux/compat/include/linux/skbuff.h
index d485b39..938d6e4 100644
--- a/datapath/linux/compat/include/linux/skbuff.h
+++ b/datapath/linux/compat/include/linux/skbuff.h
@@ -251,4 +251,16 @@ static inline void skb_reset_mac_len(struct sk_buff *skb)
skb->mac_len = skb->network_header - skb->mac_header;
}
#endif
+
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,7,0)
+static inline void skb_set_encapsulation(struct sk_buff *skb)
+{
+ skb->encapsulation = 1;
+}
+#else
+/* Encapsulation didn't exist before 3.7.0 but that is ok
+ * because neither did MPLS GSO the only reason the bit is set
+ * by Open vSwtich */
+static inline void skb_set_encapsulation(struct sk_buff *skb) { }
+#endif
#endif
diff --git a/include/linux/openvswitch.h b/include/linux/openvswitch.h
index bd2f05f..e890fd8 100644
--- a/include/linux/openvswitch.h
+++ b/include/linux/openvswitch.h
@@ -287,7 +287,9 @@ enum ovs_key_attr {
OVS_KEY_ATTR_IPV4_TUNNEL, /* struct ovs_key_ipv4_tunnel */
#endif
- OVS_KEY_ATTR_MPLS = 62, /* struct ovs_key_mpls */
+ OVS_KEY_ATTR_MPLS = 62, /* array of struct ovs_key_mpls.
+ * The implementation may restrict
+ * the accepted length of the array. */
__OVS_KEY_ATTR_MAX
};
@@ -330,7 +332,7 @@ struct ovs_key_ethernet {
};
struct ovs_key_mpls {
- __be32 mpls_top_lse;
+ __be32 mpls_lse;
};
struct ovs_key_ipv4 {
diff --git a/lib/odp-util.c b/lib/odp-util.c
index 751c1c9..3206dc9 100644
--- a/lib/odp-util.c
+++ b/lib/odp-util.c
@@ -906,7 +906,7 @@ format_odp_key_attr(const struct nlattr *a, struct ds *ds)
case OVS_KEY_ATTR_MPLS: {
const struct ovs_key_mpls *mpls_key = nl_attr_get(a);
ds_put_char(ds, '(');
- format_mpls_lse(ds, mpls_key->mpls_top_lse);
+ format_mpls_lse(ds, mpls_key->mpls_lse);
ds_put_char(ds, ')');
break;
}
@@ -1231,7 +1231,7 @@ parse_odp_key_attr(const char *s, const struct simap *port_names,
mpls = nl_msg_put_unspec_uninit(key, OVS_KEY_ATTR_MPLS,
sizeof *mpls);
- mpls->mpls_top_lse = mpls_lse_from_components(label, tc, ttl, bos);
+ mpls->mpls_lse = mpls_lse_from_components(label, tc, ttl, bos);
return n;
}
}
@@ -1594,7 +1594,7 @@ odp_flow_key_from_flow(struct ofpbuf *buf, const struct flow *flow,
mpls_key = nl_msg_put_unspec_uninit(buf, OVS_KEY_ATTR_MPLS,
sizeof *mpls_key);
- mpls_key->mpls_top_lse = flow->mpls_lse;
+ mpls_key->mpls_lse = flow->mpls_lse;
}
if (is_ip_any(flow) && !(flow->nw_frag & FLOW_NW_FRAG_LATER)) {
@@ -2250,7 +2250,7 @@ commit_mpls_action(const struct flow *flow, struct flow *base,
} else {
struct ovs_key_mpls mpls_key;
- mpls_key.mpls_top_lse = flow->mpls_lse;
+ mpls_key.mpls_lse = flow->mpls_lse;
commit_set_action(odp_actions, OVS_KEY_ATTR_MPLS,
&mpls_key, sizeof(mpls_key));
}
--
1.7.10.4
^ 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