* [PATCH 0/2] iputils: minor fixes
From: Jan Synacek @ 2012-11-14 12:57 UTC (permalink / raw)
To: yoshfuji; +Cc: netdev, Jan Synacek
Jan Synacek (2):
ping,ping6: Add newline to error message.
ping: Don't free an unintialized value.
ping.c | 8 +++++---
ping_common.c | 2 +-
2 files changed, 6 insertions(+), 4 deletions(-)
--
1.7.11.7
^ permalink raw reply
* [PATCH 1/2] ping,ping6: Add newline to error message.
From: Jan Synacek @ 2012-11-14 12:57 UTC (permalink / raw)
To: yoshfuji; +Cc: netdev, Jan Synacek
In-Reply-To: <1352897836-17603-1-git-send-email-jsynacek@redhat.com>
---
ping_common.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/ping_common.c b/ping_common.c
index 86dbeaa..1605ca9 100644
--- a/ping_common.c
+++ b/ping_common.c
@@ -265,7 +265,7 @@ void common_options(int ch)
char *endp;
mark = (int)strtoul(optarg, &endp, 10);
if (mark < 0 || *endp != '\0') {
- fprintf(stderr, "mark cannot be negative");
+ fprintf(stderr, "mark cannot be negative\n");
exit(2);
}
options |= F_MARK;
--
1.7.11.7
^ permalink raw reply related
* [PATCH 2/2] ping: Don't free an unintialized value.
From: Jan Synacek @ 2012-11-14 12:57 UTC (permalink / raw)
To: yoshfuji; +Cc: netdev, Jan Synacek
In-Reply-To: <1352897836-17603-1-git-send-email-jsynacek@redhat.com>
---
ping.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/ping.c b/ping.c
index fe9ff8a..9de3d08 100644
--- a/ping.c
+++ b/ping.c
@@ -122,7 +122,7 @@ main(int argc, char **argv)
u_char *packet;
char *target;
#ifdef USE_IDN
- char *hnamebuf;
+ char *hnamebuf = NULL;
#else
char hnamebuf[MAX_HOSTNAMELEN];
#endif
@@ -263,8 +263,10 @@ main(int argc, char **argv)
#ifdef USE_IDN
int rc;
- free(hnamebuf);
- hnamebuf = NULL;
+ if (hnamebuf) {
+ free(hnamebuf);
+ hnamebuf = NULL;
+ }
rc = idna_to_ascii_lz(target, &idn, 0);
if (rc != IDNA_SUCCESS) {
--
1.7.11.7
^ permalink raw reply related
* Re: [RFC PATCH 01/13] net: Add generic packet offload infrastructure.
From: Vlad Yasevich @ 2012-11-14 13:03 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, davem
In-Reply-To: <1352859861.4497.0.camel@edumazet-glaptop>
On 11/13/2012 09:24 PM, Eric Dumazet wrote:
> On Tue, 2012-11-13 at 20:24 -0500, Vlad Yasevich wrote:
>> Create a new data structure to contain the GRO/GSO callbacks and add
>> a new registration mechanism.
>>
>> Singed-off-by: Vlad Yasevich <vyasevic@redhat.com>
>> ---
>> include/linux/netdevice.h | 15 ++++++++
>> net/core/dev.c | 80 +++++++++++++++++++++++++++++++++++++++++++++
>> 2 files changed, 95 insertions(+), 0 deletions(-)
>>
>> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>> index f8eda02..d15af51 100644
>> --- a/include/linux/netdevice.h
>> +++ b/include/linux/netdevice.h
>> @@ -1511,6 +1511,18 @@ struct packet_type {
>> struct list_head list;
>> };
>>
>> +struct packet_offload {
>> + __be16 type; /* This is really htons(ether_type). */
>> + struct net_device *dev; /* NULL is wildcarded here */
>
> Shouldnt this dev pointer be removed at some point in the patch serie ?
yes. i was thinking about this as well. I actually shouldn't have been
carried into this struct to begin with since its not really being used
by the offload calls.
-vlad
>
>> + struct sk_buff *(*gso_segment)(struct sk_buff *skb,
>> + netdev_features_t features);
>> + int (*gso_send_check)(struct sk_buff *skb);
>> + struct sk_buff **(*gro_receive)(struct sk_buff **head,
>> + struct sk_buff *skb);
>> + int (*gro_complete)(struct sk_buff *skb);
>> + struct list_head list;
>> +};
>> +
>> #include <linux/notifier.h>
>>
>
^ permalink raw reply
* [PATCH net-next 0/3] myri10ge: LRO to GRO conversion
From: Andrew Gallatin @ 2012-11-14 13:06 UTC (permalink / raw)
To: netdev
Hi,
The following patchset converts myri10ge from using the old inet_lro
interface to GRO.
Note that a naive LRO->GRO conversion of myri10ge will result in a
performance regression for vlan tagged frames. This is because
myri10ge does not offer hardware vlan tag offload, and because GRO
requires hardware vlan tag offload to aggregate vlan tagged frames.
To address this performance regression, I have implemented vlan tag
popping in the myri10ge driver, as it seems to be the lesser of two
evils. As eric.dumazet@gmail.com commented when I asked about this on
netdev last week: "Given GRO assumes NIC does hardware vlan
offloading, I guess I would chose to do that. It seems unfortunate to
add vlan decap in GRO path, already very complex."
Andrew Gallatin (3):
myri10ge: Convert from LRO to GRO
myri10ge: Add vlan rx for better GRO perf.
myri10ge: Use skb_fill_page_desc().
drivers/net/ethernet/myricom/Kconfig | 1 -
drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 281
++++++----------------
2 files changed, 80 insertions(+), 202 deletions(-)
^ permalink raw reply
* [PATCH net-next 2/3] myri10ge: Add vlan rx for better GRO perf.
From: Andrew Gallatin @ 2012-11-14 13:06 UTC (permalink / raw)
To: netdev
Unlike LRO, GRO requires that vlan tags be removed before
aggregation can occur. Since the myri10ge NIC does not support
hardware vlan tag offload, we must remove the tag in the driver
to achieve performance comparable to LRO for vlan tagged frames.
Signed-off-by: Andrew Gallatin <gallatin@myri.com>
---
drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 47
++++++++++++++++++++++
1 file changed, 47 insertions(+)
diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
index a5ab2f2..b9b6dfd 100644
--- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
+++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
@@ -1264,6 +1264,48 @@ myri10ge_unmap_rx_page(struct pci_dev *pdev,
}
}
+/*
+ * GRO does not support acceleration of tagged vlan frames, and
+ * this NIC does not support vlan tag offload, so we must pop
+ * the tag ourselves to be able to achieve GRO performance that
+ * is comparable to LRO.
+ */
+
+static inline void
+myri10ge_vlan_rx(struct net_device *dev, void *addr, struct sk_buff *skb)
+{
+ u8 *va;
+ struct vlan_ethhdr *veh;
+ struct ethhdr *eh;
+ struct skb_frag_struct *frag;
+ u16 proto;
+
+ va = addr;
+ va += MXGEFW_PAD;
+ veh = (struct vlan_ethhdr *) va;
+ if ((dev->features & (NETIF_F_HW_VLAN_RX | NETIF_F_GRO)) ==
+ (NETIF_F_HW_VLAN_RX | NETIF_F_GRO) &&
+ (veh->h_vlan_proto == ntohs(ETH_P_8021Q))) {
+ /* fixup csum if needed */
+ if (skb->ip_summed == CHECKSUM_COMPLETE)
+ skb->csum = csum_sub(skb->csum,
+ csum_partial(va + ETH_HLEN,
+ VLAN_HLEN, 0));
+ /* pop tag */
+ __vlan_hwaccel_put_tag(skb, ntohs(veh->h_vlan_TCI));
+ proto = veh->h_vlan_encapsulated_proto;
+ memmove(va + VLAN_HLEN, va, ETH_HLEN);
+ va += VLAN_HLEN;
+ eh = (struct ethhdr *)va;
+ eh->h_proto = proto;
+ skb->len -= VLAN_HLEN;
+ skb->data_len -= VLAN_HLEN;
+ frag = skb_shinfo(skb)->frags;
+ frag->page_offset += VLAN_HLEN;
+ skb_frag_size_set(frag, skb_frag_size(frag) - VLAN_HLEN);
+ }
+}
+
static inline int
myri10ge_rx_done(struct myri10ge_slice_state *ss, int len, __wsum csum)
{
@@ -1329,6 +1371,7 @@ myri10ge_rx_done(struct myri10ge_slice_state *ss,
int len, __wsum csum)
skb->ip_summed = CHECKSUM_COMPLETE;
skb->csum = csum;
}
+ myri10ge_vlan_rx(mgp->dev, va, skb);
skb_record_rx_queue(skb, ss - &mgp->ss[0]);
napi_gro_frags(&ss->napi);
@@ -3854,6 +3897,10 @@ static int myri10ge_probe(struct pci_dev *pdev,
const struct pci_device_id *ent)
netdev->netdev_ops = &myri10ge_netdev_ops;
netdev->mtu = myri10ge_initial_mtu;
netdev->hw_features = mgp->features | NETIF_F_RXCSUM;
+
+ /* fake NETIF_F_HW_VLAN_RX for good GRO performance */
+ netdev->hw_features |= NETIF_F_HW_VLAN_RX;
+
netdev->features = netdev->hw_features;
if (dac_enabled)
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next 3/3] myri10ge: Use skb_fill_page_desc().
From: Andrew Gallatin @ 2012-11-14 13:06 UTC (permalink / raw)
To: netdev
Now that LRO is gone, the receive routine is much simpler, and
we are able to use the standard skb_fill_page_desc() in myri10ge.
Signed-off-by: Andrew Gallatin <gallatin@myri.com>
---
drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 11 ++++-------
1 file changed, 4 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
index b9b6dfd..82e9dcc 100644
--- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
+++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
@@ -1347,17 +1347,14 @@ myri10ge_rx_done(struct myri10ge_slice_state
*ss, int len, __wsum csum)
/* Fill skb_frag_struct(s) with data from our receive */
for (i = 0, remainder = len; remainder > 0; i++) {
myri10ge_unmap_rx_page(pdev, &rx->info[idx], bytes);
- __skb_frag_set_page(&rx_frags[i], rx->info[idx].page);
- rx_frags[i].page_offset = rx->info[idx].page_offset;
- if (remainder < MYRI10GE_ALLOC_SIZE)
- skb_frag_size_set(&rx_frags[i], remainder);
- else
- skb_frag_size_set(&rx_frags[i], MYRI10GE_ALLOC_SIZE);
+ skb_fill_page_desc(skb, i, rx->info[idx].page,
+ rx->info[idx].page_offset,
+ remainder < MYRI10GE_ALLOC_SIZE ?
+ remainder : MYRI10GE_ALLOC_SIZE);
rx->cnt++;
idx = rx->cnt & rx->mask;
remainder -= MYRI10GE_ALLOC_SIZE;
}
- skb_shinfo(skb)->nr_frags = i;
/* remove padding */
rx_frags[0].page_offset += MXGEFW_PAD;
--
1.7.9.5
^ permalink raw reply related
* Re: [RFC PATCH 03/13] net: Add net protocol offload registration infrustructure
From: Vlad Yasevich @ 2012-11-14 13:08 UTC (permalink / raw)
To: nicolas.dichtel; +Cc: netdev, eric.dumazet, davem
In-Reply-To: <50A354C6.20805@6wind.com>
On 11/14/2012 03:22 AM, Nicolas Dichtel wrote:
> Le 14/11/2012 02:24, Vlad Yasevich a écrit :
>> diff --git a/net/ipv4/protocol.c b/net/ipv4/protocol.c
>> index 8918eff..1278db8 100644
>> --- a/net/ipv4/protocol.c
>> +++ b/net/ipv4/protocol.c
>> @@ -29,6 +29,7 @@
>> #include <net/protocol.h>
>>
>> const struct net_protocol __rcu *inet_protos[MAX_INET_PROTOS]
>> __read_mostly;
>> +const struct net_offload __rcu *inet_offloads[MAX_INET_PROTOS]
>> __read_mostly;
>>
>> /*
>> * Add a protocol handler to the hash tables
>> @@ -41,6 +42,13 @@ int inet_add_protocol(const struct net_protocol
>> *prot, unsigned char protocol)
>> }
>> EXPORT_SYMBOL(inet_add_protocol);
>>
>> +int inet_add_offload(const struct net_offload *prot, unsigned char
>> protocol)
>> +{
>> + return !cmpxchg((const struct net_offload
>> **)&inet_offloads[protocol],
>> + NULL, prot) ? 0 : -1;
>> +}
>> +EXPORT_SYMBOL(inet_add_offload);
>> +
>> /*
>> * Remove a protocol from the hash tables.
>> */
>> @@ -56,4 +64,16 @@ int inet_del_protocol(const struct net_protocol
>> *prot, unsigned char protocol)
>>
>> return ret;
>> }
>> -EXPORT_SYMBOL(inet_del_protocol);
> This line should probably not be removed ;-)
Yep, good catch... thanks...
-vald
^ permalink raw reply
* Re: [RFC PATCH 00/13] Always build GSO/GRO functionality into the kernel
From: Vlad Yasevich @ 2012-11-14 13:10 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, davem
In-Reply-To: <1352859928.4497.1.camel@edumazet-glaptop>
On 11/13/2012 09:25 PM, Eric Dumazet wrote:
> On Tue, 2012-11-13 at 20:24 -0500, Vlad Yasevich wrote:
>> This patch series is a revision suggested by Eric to solve the problem where
>> a host without IPv6 support drops GSO frames from the guest.
>>
>> The problem is that GSO/GRO support is per protocol, and when said protocol
>> is not loaded or is disabled, packets attempting to go through GSO/GRO code paths
>> are dropped. This causes retransmissions and a two orders of magnitude drop in
>> performance.
>>
>> Prior attempt to solve the problem simply enabled enough GSO/GRO functionality
>> for IPv6 protocol when IPv6 was diabled. This did not solve the problem when
>> the protocol was not build in or was blacklisted.
>> To solve the problem, it was suggested that we separate GSO/GRO callback
>> registration from packet processing registrations. That way
>> GSO/GRO callbacks can be built into the kernel and always be there.
>> This patch series attempts to do just that.
>> * Patches 1 and 2 split the GSO/GRO handlers from packet_type and convert
>> to the new structure.
>> * Patches 3, 4 and 5 do the same thing for net_protocol structure.
>> * The rest of the patches try to incrementally move the IPv6 GSO/GRO
>> code out of the module and into the static kernel build. Some IPv6
>> helper functions also had to move as well.
>>
>> I am currently testing the patches, but if folks could look this over
>> and send me any comments, I'd appreciate it.
>
> Seems very nice !
>
> I am just wondering if GSO/GRO is fully enabled at every step ?
>
I think so. I ran basic tests along most of the steps and it seemed to
be enabled. That's why this is a 13 patch series :) Tried to do it
incrementally without impacting functionality.
-vlad
^ permalink raw reply
* [PATCH net-next 1/3] myri10ge: Convert from LRO to GRO
From: Andrew Gallatin @ 2012-11-14 13:06 UTC (permalink / raw)
To: netdev
Convert myri10ge from LRO to GRO, and simplify the driver by removing
various LRO-related code which is no longer needed including
ndo_fix_features op, custom skb building from frags, and LRO
header parsing.
Signed-off-by: Andrew Gallatin <gallatin@myri.com>
---
drivers/net/ethernet/myricom/Kconfig | 1 -
drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 227
+++-------------------
2 files changed, 31 insertions(+), 197 deletions(-)
diff --git a/drivers/net/ethernet/myricom/Kconfig
b/drivers/net/ethernet/myricom/Kconfig
index 540f0c6..3932d08 100644
--- a/drivers/net/ethernet/myricom/Kconfig
+++ b/drivers/net/ethernet/myricom/Kconfig
@@ -23,7 +23,6 @@ config MYRI10GE
depends on PCI && INET
select FW_LOADER
select CRC32
- select INET_LRO
---help---
This driver supports Myricom Myri-10G Dual Protocol interface in
Ethernet mode. If the eeprom on your board is not recent enough,
diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
index 83516e3..a5ab2f2 100644
--- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
+++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
@@ -50,7 +50,6 @@
#include <linux/etherdevice.h>
#include <linux/if_ether.h>
#include <linux/if_vlan.h>
-#include <linux/inet_lro.h>
#include <linux/dca.h>
#include <linux/ip.h>
#include <linux/inet.h>
@@ -96,8 +95,6 @@ MODULE_LICENSE("Dual BSD/GPL");
#define MYRI10GE_EEPROM_STRINGS_SIZE 256
#define MYRI10GE_MAX_SEND_DESC_TSO ((65536 / 2048) * 2)
-#define MYRI10GE_MAX_LRO_DESCRIPTORS 8
-#define MYRI10GE_LRO_MAX_PKTS 64
#define MYRI10GE_NO_CONFIRM_DATA htonl(0xffffffff)
#define MYRI10GE_NO_RESPONSE_RESULT 0xffffffff
@@ -165,8 +162,6 @@ struct myri10ge_rx_done {
dma_addr_t bus;
int cnt;
int idx;
- struct net_lro_mgr lro_mgr;
- struct net_lro_desc lro_desc[MYRI10GE_MAX_LRO_DESCRIPTORS];
};
struct myri10ge_slice_netstats {
@@ -338,11 +333,6 @@ static int myri10ge_debug = -1; /* defaults above */
module_param(myri10ge_debug, int, 0);
MODULE_PARM_DESC(myri10ge_debug, "Debug level (0=none,...,16=all)");
-static int myri10ge_lro_max_pkts = MYRI10GE_LRO_MAX_PKTS;
-module_param(myri10ge_lro_max_pkts, int, S_IRUGO);
-MODULE_PARM_DESC(myri10ge_lro_max_pkts,
- "Number of LRO packets to be aggregated");
-
static int myri10ge_fill_thresh = 256;
module_param(myri10ge_fill_thresh, int, S_IRUGO | S_IWUSR);
MODULE_PARM_DESC(myri10ge_fill_thresh, "Number of empty rx slots
allowed");
@@ -1197,36 +1187,6 @@ static inline void myri10ge_vlan_ip_csum(struct
sk_buff *skb, __wsum hw_csum)
}
}
-static inline void
-myri10ge_rx_skb_build(struct sk_buff *skb, u8 * va,
- struct skb_frag_struct *rx_frags, int len, int hlen)
-{
- struct skb_frag_struct *skb_frags;
-
- skb->len = skb->data_len = len;
- /* attach the page(s) */
-
- skb_frags = skb_shinfo(skb)->frags;
- while (len > 0) {
- memcpy(skb_frags, rx_frags, sizeof(*skb_frags));
- len -= skb_frag_size(rx_frags);
- skb_frags++;
- rx_frags++;
- skb_shinfo(skb)->nr_frags++;
- }
-
- /* pskb_may_pull is not available in irq context, but
- * skb_pull() (for ether_pad and eth_type_trans()) requires
- * the beginning of the packet in skb_headlen(), move it
- * manually */
- skb_copy_to_linear_data(skb, va, hlen);
- skb_shinfo(skb)->frags[0].page_offset += hlen;
- skb_frag_size_sub(&skb_shinfo(skb)->frags[0], hlen);
- skb->data_len -= hlen;
- skb->tail += hlen;
- skb_pull(skb, MXGEFW_PAD);
-}
-
static void
myri10ge_alloc_rx_pages(struct myri10ge_priv *mgp, struct
myri10ge_rx_buf *rx,
int bytes, int watchdog)
@@ -1304,18 +1264,14 @@ myri10ge_unmap_rx_page(struct pci_dev *pdev,
}
}
-#define MYRI10GE_HLEN 64 /* The number of bytes to copy from a
- * page into an skb */
-
static inline int
-myri10ge_rx_done(struct myri10ge_slice_state *ss, int len, __wsum csum,
- bool lro_enabled)
+myri10ge_rx_done(struct myri10ge_slice_state *ss, int len, __wsum csum)
{
struct myri10ge_priv *mgp = ss->mgp;
struct sk_buff *skb;
- struct skb_frag_struct rx_frags[MYRI10GE_MAX_FRAGS_PER_FRAME];
+ struct skb_frag_struct *rx_frags;
struct myri10ge_rx_buf *rx;
- int i, idx, hlen, remainder, bytes;
+ int i, idx, remainder, bytes;
struct pci_dev *pdev = mgp->pdev;
struct net_device *dev = mgp->dev;
u8 *va;
@@ -1332,6 +1288,20 @@ myri10ge_rx_done(struct myri10ge_slice_state *ss,
int len, __wsum csum,
idx = rx->cnt & rx->mask;
va = page_address(rx->info[idx].page) + rx->info[idx].page_offset;
prefetch(va);
+
+ skb = napi_get_frags(&ss->napi);
+ if (unlikely(skb == NULL)) {
+ ss->stats.rx_dropped++;
+ for (i = 0, remainder = len; remainder > 0; i++) {
+ myri10ge_unmap_rx_page(pdev, &rx->info[idx], bytes);
+ put_page(rx->info[idx].page);
+ rx->cnt++;
+ idx = rx->cnt & rx->mask;
+ remainder -= MYRI10GE_ALLOC_SIZE;
+ }
+ return 0;
+ }
+ rx_frags = skb_shinfo(skb)->frags;
/* Fill skb_frag_struct(s) with data from our receive */
for (i = 0, remainder = len; remainder > 0; i++) {
myri10ge_unmap_rx_page(pdev, &rx->info[idx], bytes);
@@ -1345,54 +1315,23 @@ myri10ge_rx_done(struct myri10ge_slice_state
*ss, int len, __wsum csum,
idx = rx->cnt & rx->mask;
remainder -= MYRI10GE_ALLOC_SIZE;
}
+ skb_shinfo(skb)->nr_frags = i;
- if (lro_enabled) {
- rx_frags[0].page_offset += MXGEFW_PAD;
- skb_frag_size_sub(&rx_frags[0], MXGEFW_PAD);
- len -= MXGEFW_PAD;
- lro_receive_frags(&ss->rx_done.lro_mgr, rx_frags,
- /* opaque, will come back in get_frag_header */
- len, len,
- (void *)(__force unsigned long)csum, csum);
+ /* remove padding */
+ rx_frags[0].page_offset += MXGEFW_PAD;
+ rx_frags[0].size -= MXGEFW_PAD;
+ len -= MXGEFW_PAD;
- return 1;
- }
-
- hlen = MYRI10GE_HLEN > len ? len : MYRI10GE_HLEN;
-
- /* allocate an skb to attach the page(s) to. This is done
- * after trying LRO, so as to avoid skb allocation overheads */
-
- skb = netdev_alloc_skb(dev, MYRI10GE_HLEN + 16);
- if (unlikely(skb == NULL)) {
- ss->stats.rx_dropped++;
- do {
- i--;
- __skb_frag_unref(&rx_frags[i]);
- } while (i != 0);
- return 0;
- }
-
- /* Attach the pages to the skb, and trim off any padding */
- myri10ge_rx_skb_build(skb, va, rx_frags, len, hlen);
- if (skb_frag_size(&skb_shinfo(skb)->frags[0]) <= 0) {
- skb_frag_unref(skb, 0);
- skb_shinfo(skb)->nr_frags = 0;
- } else {
- skb->truesize += bytes * skb_shinfo(skb)->nr_frags;
+ skb->len = len;
+ skb->data_len = len;
+ skb->truesize += len;
+ if (dev->features & NETIF_F_RXCSUM) {
+ skb->ip_summed = CHECKSUM_COMPLETE;
+ skb->csum = csum;
}
- skb->protocol = eth_type_trans(skb, dev);
skb_record_rx_queue(skb, ss - &mgp->ss[0]);
- if (dev->features & NETIF_F_RXCSUM) {
- if ((skb->protocol == htons(ETH_P_IP)) ||
- (skb->protocol == htons(ETH_P_IPV6))) {
- skb->csum = csum;
- skb->ip_summed = CHECKSUM_COMPLETE;
- } else
- myri10ge_vlan_ip_csum(skb, csum);
- }
- netif_receive_skb(skb);
+ napi_gro_frags(&ss->napi);
return 1;
}
@@ -1480,18 +1419,11 @@ myri10ge_clean_rx_done(struct
myri10ge_slice_state *ss, int budget)
u16 length;
__wsum checksum;
- /*
- * Prevent compiler from generating more than one ->features memory
- * access to avoid theoretical race condition with functions that
- * change NETIF_F_LRO flag at runtime.
- */
- bool lro_enabled = !!(ACCESS_ONCE(mgp->dev->features) & NETIF_F_LRO);
-
while (rx_done->entry[idx].length != 0 && work_done < budget) {
length = ntohs(rx_done->entry[idx].length);
rx_done->entry[idx].length = 0;
checksum = csum_unfold(rx_done->entry[idx].checksum);
- rx_ok = myri10ge_rx_done(ss, length, checksum, lro_enabled);
+ rx_ok = myri10ge_rx_done(ss, length, checksum);
rx_packets += rx_ok;
rx_bytes += rx_ok * (unsigned long)length;
cnt++;
@@ -1503,9 +1435,6 @@ myri10ge_clean_rx_done(struct myri10ge_slice_state
*ss, int budget)
ss->stats.rx_packets += rx_packets;
ss->stats.rx_bytes += rx_bytes;
- if (lro_enabled)
- lro_flush_all(&rx_done->lro_mgr);
-
/* restock receive rings if needed */
if (ss->rx_small.fill_cnt - ss->rx_small.cnt < myri10ge_fill_thresh)
myri10ge_alloc_rx_pages(mgp, &ss->rx_small,
@@ -1779,7 +1708,6 @@ static const char
myri10ge_gstrings_slice_stats[][ETH_GSTRING_LEN] = {
"tx_pkt_start", "tx_pkt_done", "tx_req", "tx_done",
"rx_small_cnt", "rx_big_cnt",
"wake_queue", "stop_queue", "tx_linearized",
- "LRO aggregated", "LRO flushed", "LRO avg aggr", "LRO no_desc",
};
#define MYRI10GE_NET_STATS_LEN 21
@@ -1880,14 +1808,6 @@ myri10ge_get_ethtool_stats(struct net_device *netdev,
data[i++] = (unsigned int)ss->tx.wake_queue;
data[i++] = (unsigned int)ss->tx.stop_queue;
data[i++] = (unsigned int)ss->tx.linearized;
- data[i++] = ss->rx_done.lro_mgr.stats.aggregated;
- data[i++] = ss->rx_done.lro_mgr.stats.flushed;
- if (ss->rx_done.lro_mgr.stats.flushed)
- data[i++] = ss->rx_done.lro_mgr.stats.aggregated /
- ss->rx_done.lro_mgr.stats.flushed;
- else
- data[i++] = 0;
- data[i++] = ss->rx_done.lro_mgr.stats.no_desc;
}
}
@@ -2271,67 +2191,6 @@ static void myri10ge_free_irq(struct
myri10ge_priv *mgp)
pci_disable_msix(pdev);
}
-static int
-myri10ge_get_frag_header(struct skb_frag_struct *frag, void **mac_hdr,
- void **ip_hdr, void **tcpudp_hdr,
- u64 * hdr_flags, void *priv)
-{
- struct ethhdr *eh;
- struct vlan_ethhdr *veh;
- struct iphdr *iph;
- u8 *va = skb_frag_address(frag);
- unsigned long ll_hlen;
- /* passed opaque through lro_receive_frags() */
- __wsum csum = (__force __wsum) (unsigned long)priv;
-
- /* find the mac header, aborting if not IPv4 */
-
- eh = (struct ethhdr *)va;
- *mac_hdr = eh;
- ll_hlen = ETH_HLEN;
- if (eh->h_proto != htons(ETH_P_IP)) {
- if (eh->h_proto == htons(ETH_P_8021Q)) {
- veh = (struct vlan_ethhdr *)va;
- if (veh->h_vlan_encapsulated_proto != htons(ETH_P_IP))
- return -1;
-
- ll_hlen += VLAN_HLEN;
-
- /*
- * HW checksum starts ETH_HLEN bytes into
- * frame, so we must subtract off the VLAN
- * header's checksum before csum can be used
- */
- csum = csum_sub(csum, csum_partial(va + ETH_HLEN,
- VLAN_HLEN, 0));
- } else {
- return -1;
- }
- }
- *hdr_flags = LRO_IPV4;
-
- iph = (struct iphdr *)(va + ll_hlen);
- *ip_hdr = iph;
- if (iph->protocol != IPPROTO_TCP)
- return -1;
- if (ip_is_fragment(iph))
- return -1;
- *hdr_flags |= LRO_TCP;
- *tcpudp_hdr = (u8 *) (*ip_hdr) + (iph->ihl << 2);
-
- /* verify the IP checksum */
- if (unlikely(ip_fast_csum((u8 *) iph, iph->ihl)))
- return -1;
-
- /* verify the checksum */
- if (unlikely(csum_tcpudp_magic(iph->saddr, iph->daddr,
- ntohs(iph->tot_len) - (iph->ihl << 2),
- IPPROTO_TCP, csum)))
- return -1;
-
- return 0;
-}
-
static int myri10ge_get_txrx(struct myri10ge_priv *mgp, int slice)
{
struct myri10ge_cmd cmd;
@@ -2402,7 +2261,6 @@ static int myri10ge_open(struct net_device *dev)
struct myri10ge_cmd cmd;
int i, status, big_pow2, slice;
u8 *itable;
- struct net_lro_mgr *lro_mgr;
if (mgp->running != MYRI10GE_ETH_STOPPED)
return -EBUSY;
@@ -2513,19 +2371,6 @@ static int myri10ge_open(struct net_device *dev)
goto abort_with_rings;
}
- lro_mgr = &ss->rx_done.lro_mgr;
- lro_mgr->dev = dev;
- lro_mgr->features = LRO_F_NAPI;
- lro_mgr->ip_summed = CHECKSUM_COMPLETE;
- lro_mgr->ip_summed_aggr = CHECKSUM_UNNECESSARY;
- lro_mgr->max_desc = MYRI10GE_MAX_LRO_DESCRIPTORS;
- lro_mgr->lro_arr = ss->rx_done.lro_desc;
- lro_mgr->get_frag_header = myri10ge_get_frag_header;
- lro_mgr->max_aggr = myri10ge_lro_max_pkts;
- lro_mgr->frag_align_pad = 2;
- if (lro_mgr->max_aggr > MAX_SKB_FRAGS)
- lro_mgr->max_aggr = MAX_SKB_FRAGS;
-
/* must happen prior to any irq */
napi_enable(&(ss)->napi);
}
@@ -3143,15 +2988,6 @@ static int myri10ge_set_mac_address(struct
net_device *dev, void *addr)
return 0;
}
-static netdev_features_t myri10ge_fix_features(struct net_device *dev,
- netdev_features_t features)
-{
- if (!(features & NETIF_F_RXCSUM))
- features &= ~NETIF_F_LRO;
-
- return features;
-}
-
static int myri10ge_change_mtu(struct net_device *dev, int new_mtu)
{
struct myri10ge_priv *mgp = netdev_priv(dev);
@@ -3878,7 +3714,6 @@ static const struct net_device_ops
myri10ge_netdev_ops = {
.ndo_get_stats64 = myri10ge_get_stats,
.ndo_validate_addr = eth_validate_addr,
.ndo_change_mtu = myri10ge_change_mtu,
- .ndo_fix_features = myri10ge_fix_features,
.ndo_set_rx_mode = myri10ge_set_multicast_list,
.ndo_set_mac_address = myri10ge_set_mac_address,
};
@@ -4018,7 +3853,7 @@ static int myri10ge_probe(struct pci_dev *pdev,
const struct pci_device_id *ent)
netdev->netdev_ops = &myri10ge_netdev_ops;
netdev->mtu = myri10ge_initial_mtu;
- netdev->hw_features = mgp->features | NETIF_F_LRO | NETIF_F_RXCSUM;
+ netdev->hw_features = mgp->features | NETIF_F_RXCSUM;
netdev->features = netdev->hw_features;
if (dac_enabled)
--
1.7.9.5
^ permalink raw reply related
* iputils-s20121114
From: yoshfuji @ 2012-11-14 13:50 UTC (permalink / raw)
To: netdev
Hello.
iputils-s20121114 comes with a lot of bug fixes and improvements, thanks to
distro bug reports.
Files:
https://sourceforge.net/projects/iputils/files/
http://www.skbuff.net/iputils/
Tree:
http://www.linux-ipv6.org/gitweb/gitweb.cgi?p=gitroot/iputils.git
https://sourceforge.net/p/iputils/code/ci/HEAD/tree/
Regards,
--yoshfuji
Changes between s20121106 and s20121112:
Sergey Fionov (1):
ping,ping6: Fallback to numeric addresses while exiting
YOSHIFUJI Hideaki (19):
ping,ping6: Rework capability support and Make sure -m and -I options work.
ping,tracepath: Spelling fixes in manpages.
ping,ping6: Fix integer overflow with large interval value (-i option).
clockdiff: Make it work with large pid.
ping,ping6: Make in_pr_addr volatile.
arping: Do not quit too early with large deadline value (-w option).
arping: Maintain minimum capabilities for SO_BINDTODEVICE(-I option).
ping: Fix recorded route comparison.
arping: Use getifaddrs() to get broadcast address.
ping6: Fix typo in error message.
ping6: Generate NI Group Address and Subject Name at once.
ping,ping6: Unmask signals on start-up.
arping: Build with USE_CAP=no.
arping,ping,ping6,tracepath,tracepath6,traceroute6: Experimental IDN support.
ping6: IDN support for the Subject Name in NI Query.
tracepath,tracepath6: Introduce -p option for port.
ping6: Add missing definitions/declarations for flowlabel management (-F option).
makefile: Do not include merge commits in RELNOTES.
iputils-s20121112
Changes since iputils-s20121112:
Jan Synacek (2):
clockdiff: remove unused variable
ping: Wrap SO_BINDTODEVICE with the correct capability.
YOSHIFUJI Hideaki (14):
ping: IP_MULTICAST_IF does not need CAP_NET_RAW.
ping6: Check ranges of flowlabel (-F option) and tclass (-Q option) arguments.
ping6: Accept 0x-notation for flowlabel (-F option) and tclass (-Q option) arguments.
ping,ping6: Manual update regarding -F, -Q and -N option.
arping,ping,ping6: Defer exitting to allow users to see usage.
arping,ping,ping6,ninfod: Change euid to uid (non-root) even if capabiliy is enabled.
ninfod: Add configure.
ninfod: libcap support to drop capabilities.
ninfod: Add run as user (-u user) option.
ninfod: Fix usage message.
arping,clockdiff,rarpd,rdisc,tftpd: Change RFC source to tools.ietf.org.
ninfod: Add ninfod(8) manpage.
makefile: Add ninfod, distclean targets.
iputils-s20121114
^ permalink raw reply
* Re: [PATCHES] Networking fixes for -stable.
From: Peter Senna Tschudin @ 2012-11-14 14:02 UTC (permalink / raw)
To: Ben Hutchings; +Cc: David Miller, Greg Kroah-Hartman, stable, netdev
In-Reply-To: <1352836286.16264.102.camel@deadeye.wl.decadent.org.uk>
Hi Ben,
On Tue, Nov 13, 2012 at 8:51 PM, Ben Hutchings <ben@decadent.org.uk> wrote:
> On Mon, 2012-11-12 at 00:25 -0500, David Miller wrote:
>> Please queue up the following networking bug fixes for
>> 3.0.x, 3.2.x, 3.4.x, and 3.6.x, respectively.
> [...]
>> From 2204849a85383fbde75680aa199142abe504adbb Mon Sep 17 00:00:00 2001
>> From: Peter Senna Tschudin <peter.senna@gmail.com>
>> Date: Sun, 28 Oct 2012 06:12:01 +0000
>> Subject: [PATCH 7/9] drivers/net/phy/mdio-bitbang.c: Call mdiobus_unregister
>> before mdiobus_free
>>
>> [ Upstream commit aa731872f7d33dcb8b54dad0cfb82d4e4d195d7e ]
>>
>> Based on commit b27393aecf66199f5ddad37c302d3e0cfadbe6c0
>>
>> Calling mdiobus_free without calling mdiobus_unregister causes
>> BUG_ON(). This patch fixes the issue.
> [...]
>
> This introduces a regresssion, as mdiobus_unregister() is not safe to
> call if the bus isn't registered. Registration is controlled by the
> drivers that use mdio-bitbang, and they should take care of
> unregistration too - and most of them do.
Is mdiobus_free() safe to call if the bus isn't registered?
I found calls to free_mdio_bitbang() after call to
mdiobus_unregister() that may be the worst cases for my patch. But I
also found places that call free_mdio_bitbang() without first calling
mdiobus_unregister(). See:
linux-next/arch/powerpc/platforms/82xx/ep8248e.c:152
linux-next/drivers/net/ethernet/8390/ax88796.c:647
linux-next/drivers/net/ethernet/freescale/fs_enet/mii-bitbang.c:197
linux-next/drivers/net/phy/mdio-gpio.c:157
linux-next/drivers/net/phy/mdio-gpio.c:172
Are those calls to free_mdio_bitbang() wrong by not calling
mdiobus_unregister() before?
>
> This should be reverted in mainline and not applied to any stable
> series.
>
> Ben.
>
> --
> Ben Hutchings
> For every complex problem
> there is a solution that is simple, neat, and wrong.
Thank you for reviewing,
Peter
--
Peter
^ permalink raw reply
* Re: [PATCH v6 2/7] net: mvneta: driver for Marvell Armada 370/XP network unit
From: Thomas Petazzoni @ 2012-11-14 14:04 UTC (permalink / raw)
To: Francois Romieu
Cc: David S. Miller, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
In-Reply-To: <20121113231248.GA30391@electric-eye.fr.zoreil.com>
Francois,
Thanks again for your detailed review. I have a few comments/questions
below.
On Wed, 14 Nov 2012 00:12:48 +0100, Francois Romieu wrote:
> Thomas Petazzoni <thomas.petazzoni@free-electrons.com> :
> [...]
> > +static int mvneta_mac_addr_set(struct mvneta_port *pp, unsigned
> > char *addr,
> > + int queue)
> > +{
> > + unsigned int mac_h;
> > + unsigned int mac_l;
> > +
> > + if (queue >= 1) {
> > + netdev_err(pp->dev, "RX queue #%d is out of
> > range\n", queue);
> > + return -EINVAL;
> > + }
>
> (whence q <= 0)
After changing the code as per your below comments, I have removed this
check.
> > +
> > + if (queue != -1) {
> > + mac_l = (addr[4] << 8) | (addr[5]);
> > + mac_h = (addr[0] << 24) | (addr[1] << 16) |
> > + (addr[2] << 8) | (addr[3] << 0);
> > +
> > + mvreg_write(pp, MVNETA_MAC_ADDR_LOW, mac_l);
> > + mvreg_write(pp, MVNETA_MAC_ADDR_HIGH, mac_h);
> > + }
>
> What is it trying to achieve with the "queue" argument ?
Basically: queue == -1 is a special value to say "I want to discard the
entries in the ucast table". It's used when switching from one MAC
address to another: we remove the old entries by specifying queue=-1,
and then add the new entries by specifying queue=0.
The thing is that the driver does not yet support the multiqueue
aspects of the device, but since many registers have queue-related
aspects, we still have to worry about queues a little bit in the
driver. Complete multiqueue support is planned as a second step.
> [...]
> > +/* Refill processing */
> > +static int mvneta_rx_refill(struct mvneta_port *pp,
> > + struct mvneta_rx_desc *rx_desc)
> > +
> > +{
> > + dma_addr_t phys_addr;
> > + struct sk_buff *skb;
> > +
> > + skb = netdev_alloc_skb(pp->dev, pp->pkt_size);
> > + if (!skb)
> > + return -ENOMEM;
>
> Could netdev_alloc_skb_ip_align be an option ?
If I'm correct netdev_alloc_skb_ip_align() automatically reserve the
two bytes at the beginning of the Ethernet header to ensure that the IP
header will be aligned.
However, with the Marvell hardware, it isn't needed: the hardware
automatically pads with two empty bytes in the receiving RX buffer
before putting the real data (Ethernet header).
> [...]
> > +static int mvneta_rx(struct mvneta_port *pp, int rx_todo,
> > + struct mvneta_rx_queue *rxq)
> > +{
> > + struct net_device *dev = pp->dev;
> > + int rx_done, rx_filled;
> > +
> > + /* Get number of received packets */
> > + rx_done = mvneta_rxq_busy_desc_num_get(pp, rxq);
> > +
> > + if (rx_todo > rx_done)
> > + rx_todo = rx_done;
> > +
> > + rx_done = 0;
> > + rx_filled = 0;
> > +
> > + /* Fairness NAPI loop */
> > + while (rx_done < rx_todo) {
> > + struct mvneta_rx_desc *rx_desc =
> > mvneta_rxq_next_desc_get(rxq);
> > + struct sk_buff *skb;
> > + u32 rx_status;
> > + int rx_bytes, err;
> > +
> > + prefetch(rx_desc);
> > + rx_done++;
> > + rx_filled++;
> > + rx_status = rx_desc->status;
> > + skb = (struct sk_buff *)rx_desc->buf_cookie;
> > +
> > + if (!mvneta_rxq_desc_is_first_last(rx_desc) ||
> > + (rx_status & MVNETA_RXD_ERR_SUMMARY)) {
> > + dev->stats.rx_errors++;
> > + mvneta_rx_error(pp, rx_desc);
> > + mvneta_rx_desc_fill(rx_desc,
> > rx_desc->buf_phys_addr,
> > + (u32)skb);
> > + continue;
> > + }
> > +
> > + dma_unmap_single(pp->dev->dev.parent,
> > rx_desc->buf_phys_addr,
> > + rx_desc->data_size,
> > DMA_FROM_DEVICE); +
> > + rx_bytes = rx_desc->data_size -
> > + (MVNETA_ETH_CRC_SIZE + MVNETA_MH_SIZE);
> > + u64_stats_update_begin(&pp->rx_stats.syncp);
> > + pp->rx_stats.packets++;
> > + pp->rx_stats.bytes += rx_bytes;
> > + u64_stats_update_end(&pp->rx_stats.syncp);
> > +
> > + /* Linux processing */
> > + skb_reserve(skb, MVNETA_MH_SIZE);
> > + skb_put(skb, rx_bytes);
>
> (I suggested to use skb_reserve / skb_put instead of hand-crafted
> code but I may have ignored the elephant in the living room)
>
> I understand the skb_put, ok. What is the purpose of the skb_reserve ?
As explained above, to skip the starting 2 bytes filled with zeroes by
the Marvell hardware, before passing the SKB up to the networking stack.
> Are MVNETA_ETH_CRC_SIZE (4) and MVNETA_MH_SIZE (2) really different
> from ETH_FCS_LEN and NET_IP_ALIGN ?
I have replaced MVNETA_ETH_CRC_SIZE with ETH_FCS_LEN since they are
indeed the same. However, I am not sure using NET_IP_ALIGN directly
instead of MVNETA_MH_SIZE is correct: the Marvell Header (MH) is always
two bytes: those two bytes are filled with zeroes in the normal case,
or otherwise they are used for some custom interaction between the
Marvell Ethernet controller and specific Marvell switches (this
feature is not supported by the driver being submitted). On the other
hand, the NET_IP_ALIGN value is not necessarily zero apparently, so I
have the feeling that those things should remain two separate values.
> > + if (napi_schedule_prep(&pp->napi))
> > + __napi_schedule(&pp->napi);
>
> Hand-crafted napi_schedule.
Agreed, fixed.
> > + ret = mvneta_mac_addr_set(pp, dev->dev_addr, rxq_def);
> > + if (ret < 0) {
> > + netdev_err(dev, "mvneta_mac_addr_set failed\n");
> > + goto mac_addr_set_failure;
> > + }
>
> AFAIU mvneta_mac_addr_set can only fail when (module parameter)
> rxq_def is not set correctly. It imho calls for a check in
> probe/remove and a removal of the failure case in mvneta_mac_addr_set.
Good point, fixed.
> > + /* Connect to port interrupt line */
> > + ret = request_irq(pp->dev->irq, mvneta_isr, IRQF_DISABLED,
> > + MVNETA_DRIVER_NAME, pp);
>
> include/linux/interrupt.h
> [...]
> * IRQF_DISABLED - keep irqs disabled when calling the action handler.
> * DEPRECATED. This flag is a NOOP and scheduled to
> be removed
Fixed.
> > + if (mvneta_init(pp, phy_addr)) {
> > + dev_err(&pdev->dev, "can't init eth hal\n");
> > + err = -ENODEV;
> > + goto err_base;
> > + }
>
> The error code from mvneta_init() should be used.
Fixed.
> > + if (register_netdev(dev)) {
> > + dev_err(&pdev->dev, "failed to register\n");
> > + err = ENOMEM;
> > + goto err_base;
> > + }
>
> - The error code from register_netdev() should be used.
> - Leak: allocations in mvneta_init() are not balanced.
> - Nit: the "err_where_did_it_trigger_first" style of label shows its
> downside when compared to the "err_what_should_be_done" when
> it gets used several times.
Fixed.
> > +/* Device removal routine */
> > +static int __devexit mvneta_remove(struct platform_device *pdev)
> > +{
> > + struct net_device *dev = platform_get_drvdata(pdev);
> > + struct mvneta_port *pp = netdev_priv(dev);
> > +
> > + iounmap(pp->base);
> > +
> > + unregister_netdev(dev);
> > + irq_dispose_mapping(dev->irq);
> > + free_netdev(dev);
> > + mvneta_deinit(pp);
>
> One may expect the same ordering as the mvneta_probe unroll sequence.
Fixed as well!
Thanks,
Thomas
--
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
^ permalink raw reply
* Re: [patch net] net: correct check in dev_addr_del()
From: Eric Dumazet @ 2012-11-14 14:18 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, davem, shemminger, john.r.fastabend
In-Reply-To: <1352897464-832-1-git-send-email-jiri@resnulli.us>
On Wed, 2012-11-14 at 13:51 +0100, Jiri Pirko wrote:
> Check (ha->addr == dev->dev_addr) is always true because dev_addr_init()
> sets this. Correct the check to behave properly on addr removal.
>
> Signed-off-by: Jiri Pirko <jiri@resnulli.us>
> ---
Please add in the changelog which commit added the bug, to ease
stable teams work.
Thanks
^ permalink raw reply
* Re: [PATCH net-next 0/3] myri10ge: LRO to GRO conversion
From: Eric Dumazet @ 2012-11-14 14:27 UTC (permalink / raw)
To: Andrew Gallatin; +Cc: netdev
In-Reply-To: <50A39747.3030207@myri.com>
On Wed, 2012-11-14 at 08:06 -0500, Andrew Gallatin wrote:
> Hi,
>
> The following patchset converts myri10ge from using the old inet_lro
> interface to GRO.
Very nice work, it seems INET_LRO could die eventually ;)
^ permalink raw reply
* Re: [patch net] net: correct check in dev_addr_del()
From: Jiri Pirko @ 2012-11-14 14:29 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, davem, shemminger, john.r.fastabend
In-Reply-To: <1352902715.4497.3.camel@edumazet-glaptop>
Wed, Nov 14, 2012 at 03:18:35PM CET, eric.dumazet@gmail.com wrote:
>On Wed, 2012-11-14 at 13:51 +0100, Jiri Pirko wrote:
>> Check (ha->addr == dev->dev_addr) is always true because dev_addr_init()
>> sets this. Correct the check to behave properly on addr removal.
>>
>> Signed-off-by: Jiri Pirko <jiri@resnulli.us>
>> ---
>
>Please add in the changelog which commit added the bug, to ease
>stable teams work.
bug introduced by:
commit a748ee2426817a95b1f03012d8f339c45c722ae1
Author: Jiri Pirko <jpirko@redhat.com>
Date: Thu Apr 1 21:22:09 2010 +0000
net: move address list functions to a separate file
>
>Thanks
>
>
^ permalink raw reply
* Re: [patch net] net: correct check in dev_addr_del()
From: Eric Dumazet @ 2012-11-14 14:38 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, davem, shemminger, john.r.fastabend
In-Reply-To: <20121114142941.GA2531@minipsycho.brq.redhat.com>
On Wed, 2012-11-14 at 15:29 +0100, Jiri Pirko wrote:
> Wed, Nov 14, 2012 at 03:18:35PM CET, eric.dumazet@gmail.com wrote:
> >On Wed, 2012-11-14 at 13:51 +0100, Jiri Pirko wrote:
> >> Check (ha->addr == dev->dev_addr) is always true because dev_addr_init()
> >> sets this. Correct the check to behave properly on addr removal.
> >>
> >> Signed-off-by: Jiri Pirko <jiri@resnulli.us>
> >> ---
> >
> >Please add in the changelog which commit added the bug, to ease
> >stable teams work.
>
> bug introduced by:
>
> commit a748ee2426817a95b1f03012d8f339c45c722ae1
> Author: Jiri Pirko <jpirko@redhat.com>
> Date: Thu Apr 1 21:22:09 2010 +0000
>
> net: move address list functions to a separate file
Good, the usual way is then to add in the changelog :
Bug added in commit a748ee242681
(net: move address list functions to a separate file)
(No need to give the author/date, and we can shorten the SHA1 to 10 or
12 digits)
Thanks
^ permalink raw reply
* Re: [PATCH net-next 2/3] myri10ge: Add vlan rx for better GRO perf.
From: Eric Dumazet @ 2012-11-14 14:46 UTC (permalink / raw)
To: Andrew Gallatin; +Cc: netdev
In-Reply-To: <50A3975B.7020608@myri.com>
On Wed, 2012-11-14 at 08:06 -0500, Andrew Gallatin wrote:
>
> Unlike LRO, GRO requires that vlan tags be removed before
> aggregation can occur. Since the myri10ge NIC does not support
> hardware vlan tag offload, we must remove the tag in the driver
> to achieve performance comparable to LRO for vlan tagged frames.
>
> Signed-off-by: Andrew Gallatin <gallatin@myri.com>
> ---
> drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 47
> ++++++++++++++++++++++
> 1 file changed, 47 insertions(+)
>
> diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
> b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
> index a5ab2f2..b9b6dfd 100644
> --- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
> +++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
> @@ -1264,6 +1264,48 @@ myri10ge_unmap_rx_page(struct pci_dev *pdev,
> }
> }
>
> +/*
> + * GRO does not support acceleration of tagged vlan frames, and
> + * this NIC does not support vlan tag offload, so we must pop
> + * the tag ourselves to be able to achieve GRO performance that
> + * is comparable to LRO.
> + */
> +
> +static inline void
> +myri10ge_vlan_rx(struct net_device *dev, void *addr, struct sk_buff *skb)
> +{
> + u8 *va;
> + struct vlan_ethhdr *veh;
> + struct ethhdr *eh;
> + struct skb_frag_struct *frag;
> + u16 proto;
> +
> + va = addr;
> + va += MXGEFW_PAD;
> + veh = (struct vlan_ethhdr *) va;
> + if ((dev->features & (NETIF_F_HW_VLAN_RX | NETIF_F_GRO)) ==
> + (NETIF_F_HW_VLAN_RX | NETIF_F_GRO) &&
> + (veh->h_vlan_proto == ntohs(ETH_P_8021Q))) {
> + /* fixup csum if needed */
> + if (skb->ip_summed == CHECKSUM_COMPLETE)
> + skb->csum = csum_sub(skb->csum,
> + csum_partial(va + ETH_HLEN,
> + VLAN_HLEN, 0));
> + /* pop tag */
> + __vlan_hwaccel_put_tag(skb, ntohs(veh->h_vlan_TCI));
> + proto = veh->h_vlan_encapsulated_proto;
I am not sure you need this @proto ?
> + memmove(va + VLAN_HLEN, va, ETH_HLEN);
You could only memmove the mac addresses (2 * ETH_ALEN)
To not touch the proto (and avoid possible aliasing problems)
> + va += VLAN_HLEN;
> + eh = (struct ethhdr *)va;
> + eh->h_proto = proto;
and this should not be needed ?
> + skb->len -= VLAN_HLEN;
> + skb->data_len -= VLAN_HLEN;
> + frag = skb_shinfo(skb)->frags;
> + frag->page_offset += VLAN_HLEN;
> + skb_frag_size_set(frag, skb_frag_size(frag) - VLAN_HLEN);
> + }
> +}
> +
^ permalink raw reply
* [PATCH v7] Network driver for the Armada 370 and Armada XP ARM Marvell SoCs
From: Thomas Petazzoni @ 2012-11-14 14:56 UTC (permalink / raw)
To: David S. Miller
Cc: Francois Romieu, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
David,
This patch set adds a new network driver for the network unit
available in the newest Marvell ARM SoCs Armada 370 and Armada XP, as
well as the necessary Device Tree information to use this driver in
the two evaluation platforms of those SoCs.
The previous versions of this patch set have been sent on September
4th (v1), October 11th (v2), October 23rd (v3), October 26th (v4),
November 12th (v5), November 13th (v6) and now comes the v7 of the
driver. The number of comments over the last versions have been really
small, and I would really appreciate if this driver could land into
the 3.8 kernel release.
People interested in testing this driver can find it at:
git@github.com:MISL-EBU-System-SW/mainline-public.git marvell-neta-v7
In details:
* Patch 1 contains a small driver for the MDIO interface of this
Ethernet controller. Having a separate driver is useful to more
easily handle concurrent accesses on this MDIO interface that is
shared between all Ethernet ports.
* Patch 3 contains the driver itself. The commit log contains a
detailed explanation about why a new driver is needed for this new
Marvell SoC, compared to older Marvell SoCs (Orion, Kirkwood, Dove)
that use the mv643xx_eth driver.
* Patch 4 adds the necessary entry to the MAINTAINERS file.
* Patch 5 adds the SoC-level Device Tree information for Armada 370
and Armada XP.
* Patch 6 adds the board-level Device Tree information for the
Marvell evaluation boards of Armada 370 and Armada XP.
* Patch 6 adds the board-level Device Tree information for the
PlatHome OpenBlocks AX3-4 platform (based on the Armada XP SoC).
* Patch 7 adds the board-level Device Tree information for the
GlobalScale Mirabox platform (based on the Armada 370 SoC).
Changes since v6:
* Add more macros to make the MVNETA_RX_PKT_SIZE() math easier to
understand.
* Use ETH_FCS_LEN instead of our custom
MVNETA_ETH_CRC_SIZE. Suggested by François Romieu.
* Add a comment explaining why we have MVNETA_MH_SIZE.
* Added missing newlines at the end of some error/info
messages. Suggested by Joe Perches.
* Removed the useless error handling of mvneta_mac_addr_set(), and
tested once for all the value of rxq_def at probe() time. Suggested
by Francois Romieu.
* Use napi_schedule() instead of an hand-crafted version of
time. Suggested by Francois Romieu.
* Don't use IRQF_DISABLED. Suggested by Francois Romieu.
* Fix leak in mvneta_probe() on an error path (missing call to
mvneta_deinit()). Re-arranged the goto labels for better
clarity. Suggested by Francois Romieu.
* Use the return value of mvneta_init() and register_netdev() as the
return value of the ->probe() function on failure. Suggested by
Francois Romieu.
Changes since v5:
* Take into account comments from François Romieu (mainly coding
style fixes + addition of a mvneta_rxq_desc_is_first_last() helper
function)
* Fix a wrong argument passed to dma_alloc_coherent():
DMA_BIDIRECTIONAL should have been GFP_KERNEL. Thanks to François
Romieu for having pointed the issue in his review.
* Removed calls to smp_call_function_many() that were useless.
* Ordered alphabetically entries in the Kconfig and Makefile.
Changes since v4:
* Added a separate MDIO driver, which allow to easily handle
concurrent accesses to the MDIO interface.
* The Device Tree now has separate nodes for the PHY devices, which
belong to the MDIO bus handled by the separate MDIO driver.
* Fix tabulation issues in some Device Tree files.
* Rebased on top of 3.7-rc5
* Added the Device Tree code necessary for the GlobalScale Mirabox
platform and the PlatHome OpenBlocks AX3-4 platform.
Changes since v3:
* Use phy_find_first() to get the correct PHY. Suggested by Florian
Fainelli.
* Make pp->cause_rx_tx a simple variable instead of a per-CPU array
since it is not useful. Fixes a comment raised by David Miller.
Changes since v2:
* Change compatible string from 'marvell,neta' to
'marvell,armada-370-neta'. Requested by Rob Herring.
* Rename Ethernet DT nodes from eth@... to ethernet@... Requested by
Rob Herring.
* Remove device_type DT property. Requested by Rob Herring.
* Change the PHY interface for eth0/eth1 to be rgmii-id, which allows
to enable TX/RX delay mechanisms at the PHY level. This fixes CRC
errors on received packets during iperf tests (it was a bug in v2).
* Remove the mvneta_ prefix from module parameters. Requested by
Baruch Siach.
* Many code style improvements suggested by François Romieu.
* Properly stop/restart the TX queue when the number of TX
descriptors available becomes low, instead of returning
NETDEV_TX_BUSY. Requested by François Romieu.
* Properly drop packets on the TX path when DMA mapping functions
return an error, instead of returning NETDEV_TX_BUSY. Requested by
François Romieu.
* Rebased on top of Linux 3.7-rc2.
Changes since v1:
* Reduced the Cc: list in order to make the patch set acceptable for
the netdev@ mailing list.
* Merge the mvneta.h contents into mvneta.c, since the header was
only used by the driver. Requested by Arnd Bergmann.
* Completely reorganize the organization of the register list and
register values, in order to make it more consistent, and hopefully
easier to read (especially easier to match register values with the
corresponding register).
* Integrate with the phylib, as suggested by Florian Fainelli, and
remove the link management code that has become useless as the
result of this integration
* Fix many small details suggested by Florian Fainelli in his review
of the first driver
* Simplify various parts of the driver (descriptors array allocation,
data structures, etc.)
Thanks,
Thomas Petazzoni
^ permalink raw reply
* [PATCH v7 1/6] net: mvmdio: new Marvell MDIO driver
From: Thomas Petazzoni @ 2012-11-14 14:56 UTC (permalink / raw)
To: David S. Miller
Cc: Francois Romieu, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
In-Reply-To: <1352905010-24172-1-git-send-email-thomas.petazzoni@free-electrons.com>
This patch adds a separate driver for the MDIO interface of the
Marvell Ethernet controllers. There are two reasons to have a separate
driver rather than including it inside the MAC driver itself:
*) The MDIO interface is shared by all Ethernet ports, so a driver
must guarantee non-concurrent accesses to this MDIO interface. The
most logical way is to have a separate driver that handles this
single MDIO interface, used by all Ethernet ports.
*) The MDIO interface is the same between the existing mv643xx_eth
driver and the new mvneta driver. Even though it is for now only
used by the mvneta driver, it will in the future be used by the
mv643xx_eth driver as well.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
.../devicetree/bindings/net/marvell-orion-mdio.txt | 35 +++
drivers/net/ethernet/marvell/Kconfig | 11 +
drivers/net/ethernet/marvell/Makefile | 1 +
drivers/net/ethernet/marvell/mvmdio.c | 230 ++++++++++++++++++++
4 files changed, 277 insertions(+)
create mode 100644 Documentation/devicetree/bindings/net/marvell-orion-mdio.txt
create mode 100644 drivers/net/ethernet/marvell/mvmdio.c
diff --git a/Documentation/devicetree/bindings/net/marvell-orion-mdio.txt b/Documentation/devicetree/bindings/net/marvell-orion-mdio.txt
new file mode 100644
index 0000000..34e7aaf
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/marvell-orion-mdio.txt
@@ -0,0 +1,35 @@
+* Marvell MDIO Ethernet Controller interface
+
+The Ethernet controllers of the Marvel Kirkwood, Dove, Orion5x,
+MV78xx0, Armada 370 and Armada XP have an identical unit that provides
+an interface with the MDIO bus. This driver handles this MDIO
+interface.
+
+Required properties:
+- compatible: "marvell,orion-mdio"
+- reg: address and length of the SMI register
+
+The child nodes of the MDIO driver are the individual PHY devices
+connected to this MDIO bus. They must have a "reg" property given the
+PHY address on the MDIO bus.
+
+Example at the SoC level:
+
+mdio {
+ #address-cells = <1>;
+ #size-cells = <0>;
+ compatible = "marvell,orion-mdio";
+ reg = <0xd0072004 0x4>;
+};
+
+And at the board level:
+
+mdio {
+ phy0: ethernet-phy@0 {
+ reg = <0>;
+ };
+
+ phy1: ethernet-phy@1 {
+ reg = <1>;
+ };
+}
diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index 0029934..232ccb3 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -31,6 +31,17 @@ config MV643XX_ETH
Some boards that use the Discovery chipset are the Momenco
Ocelot C and Jaguar ATX and Pegasos II.
+config MVMDIO
+ tristate "Marvell MDIO interface support"
+ ---help---
+ This driver supports the MDIO interface found in the network
+ interface units of the Marvell EBU SoCs (Kirkwood, Orion5x,
+ Dove, Armada 370 and Armada XP).
+
+ For now, this driver is only needed for the MVNETA driver
+ (used on Armada 370 and XP), but it could be used in the
+ future by the MV643XX_ETH driver.
+
config PXA168_ETH
tristate "Marvell pxa168 ethernet support"
depends on CPU_PXA168
diff --git a/drivers/net/ethernet/marvell/Makefile b/drivers/net/ethernet/marvell/Makefile
index 57e3234..0438599 100644
--- a/drivers/net/ethernet/marvell/Makefile
+++ b/drivers/net/ethernet/marvell/Makefile
@@ -3,6 +3,7 @@
#
obj-$(CONFIG_MV643XX_ETH) += mv643xx_eth.o
+obj-$(CONFIG_MVMDIO) += mvmdio.o
obj-$(CONFIG_PXA168_ETH) += pxa168_eth.o
obj-$(CONFIG_SKGE) += skge.o
obj-$(CONFIG_SKY2) += sky2.o
diff --git a/drivers/net/ethernet/marvell/mvmdio.c b/drivers/net/ethernet/marvell/mvmdio.c
new file mode 100644
index 0000000..82fbd23
--- /dev/null
+++ b/drivers/net/ethernet/marvell/mvmdio.c
@@ -0,0 +1,230 @@
+/*
+ * Driver for the MDIO interface of Marvell network interfaces.
+ *
+ * Since the MDIO interface of Marvell network interfaces is shared
+ * between all network interfaces, having a single driver allows to
+ * handle concurrent accesses properly (you may have four Ethernet
+ * ports, but they in fact share the same SMI interface to access the
+ * MDIO bus). Moreover, this MDIO interface code is similar between
+ * the mv643xx_eth driver and the mvneta driver. For now, it is only
+ * used by the mvneta driver, but it could later be used by the
+ * mv643xx_eth driver as well.
+ *
+ * Copyright (C) 2012 Marvell
+ *
+ * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/phy.h>
+#include <linux/of_address.h>
+#include <linux/of_mdio.h>
+#include <linux/platform_device.h>
+
+#include <asm/delay.h>
+
+#define MVMDIO_SMI_DATA_SHIFT 0
+#define MVMDIO_SMI_PHY_ADDR_SHIFT 16
+#define MVMDIO_SMI_PHY_REG_SHIFT 21
+#define MVMDIO_SMI_READ_OPERATION BIT(26)
+#define MVMDIO_SMI_WRITE_OPERATION 0
+#define MVMDIO_SMI_READ_VALID BIT(27)
+#define MVMDIO_SMI_BUSY BIT(28)
+
+struct orion_mdio_dev {
+ struct mutex lock;
+ void __iomem *smireg;
+};
+
+/*
+ * Wait for the SMI unit to be ready for another operation
+ */
+static int orion_mdio_wait_ready(struct mii_bus *bus)
+{
+ struct orion_mdio_dev *dev = bus->priv;
+ int count;
+ u32 val;
+
+ count = 0;
+ while (1) {
+ val = readl(dev->smireg);
+ if (!(val & MVMDIO_SMI_BUSY))
+ break;
+
+ if (count > 100) {
+ dev_err(bus->parent, "Timeout: SMI busy for too long\n");
+ return -ETIMEDOUT;
+ }
+
+ udelay(10);
+ count++;
+ }
+
+ return 0;
+}
+
+static int orion_mdio_read(struct mii_bus *bus, int mii_id,
+ int regnum)
+{
+ struct orion_mdio_dev *dev = bus->priv;
+ int count;
+ u32 val;
+ int ret;
+
+ mutex_lock(&dev->lock);
+
+ ret = orion_mdio_wait_ready(bus);
+ if (ret < 0) {
+ mutex_unlock(&dev->lock);
+ return ret;
+ }
+
+ writel(((mii_id << MVMDIO_SMI_PHY_ADDR_SHIFT) |
+ (regnum << MVMDIO_SMI_PHY_REG_SHIFT) |
+ MVMDIO_SMI_READ_OPERATION),
+ dev->smireg);
+
+ /* Wait for the value to become available */
+ count = 0;
+ while (1) {
+ val = readl(dev->smireg);
+ if (val & MVMDIO_SMI_READ_VALID)
+ break;
+
+ if (count > 100) {
+ dev_err(bus->parent, "Timeout when reading PHY\n");
+ mutex_unlock(&dev->lock);
+ return -ETIMEDOUT;
+ }
+
+ udelay(10);
+ count++;
+ }
+
+ mutex_unlock(&dev->lock);
+
+ return val & 0xFFFF;
+}
+
+static int orion_mdio_write(struct mii_bus *bus, int mii_id,
+ int regnum, u16 value)
+{
+ struct orion_mdio_dev *dev = bus->priv;
+ int ret;
+
+ mutex_lock(&dev->lock);
+
+ ret = orion_mdio_wait_ready(bus);
+ if (ret < 0) {
+ mutex_unlock(&dev->lock);
+ return ret;
+ }
+
+ writel(((mii_id << MVMDIO_SMI_PHY_ADDR_SHIFT) |
+ (regnum << MVMDIO_SMI_PHY_REG_SHIFT) |
+ MVMDIO_SMI_WRITE_OPERATION |
+ (value << MVMDIO_SMI_DATA_SHIFT)),
+ dev->smireg);
+
+ mutex_unlock(&dev->lock);
+
+ return 0;
+}
+
+static int orion_mdio_reset(struct mii_bus *bus)
+{
+ return 0;
+}
+
+static int __devinit orion_mdio_probe(struct platform_device *pdev)
+{
+ struct device_node *np = pdev->dev.of_node;
+ struct mii_bus *bus;
+ struct orion_mdio_dev *dev;
+ int i, ret;
+
+ bus = mdiobus_alloc_size(sizeof(struct orion_mdio_dev));
+ if (!bus) {
+ dev_err(&pdev->dev, "Cannot allocate MDIO bus\n");
+ return -ENOMEM;
+ }
+
+ bus->name = "orion_mdio_bus";
+ bus->read = orion_mdio_read;
+ bus->write = orion_mdio_write;
+ bus->reset = orion_mdio_reset;
+ snprintf(bus->id, MII_BUS_ID_SIZE, "%s-mii",
+ dev_name(&pdev->dev));
+ bus->parent = &pdev->dev;
+
+ bus->irq = kmalloc(sizeof(int) * PHY_MAX_ADDR, GFP_KERNEL);
+ if (!bus->irq) {
+ dev_err(&pdev->dev, "Cannot allocate PHY IRQ array\n");
+ mdiobus_free(bus);
+ return -ENOMEM;
+ }
+
+ for (i = 0; i < PHY_MAX_ADDR; i++)
+ bus->irq[i] = PHY_POLL;
+
+ dev = bus->priv;
+ dev->smireg = of_iomap(pdev->dev.of_node, 0);
+ if (!dev->smireg) {
+ dev_err(&pdev->dev, "No SMI register address given in DT\n");
+ kfree(bus->irq);
+ mdiobus_free(bus);
+ return -ENODEV;
+ }
+
+ mutex_init(&dev->lock);
+
+ ret = of_mdiobus_register(bus, np);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "Cannot register MDIO bus (%d)\n", ret);
+ iounmap(dev->smireg);
+ kfree(bus->irq);
+ mdiobus_free(bus);
+ return ret;
+ }
+
+ platform_set_drvdata(pdev, bus);
+
+ return 0;
+}
+
+static int __devexit orion_mdio_remove(struct platform_device *pdev)
+{
+ struct mii_bus *bus = platform_get_drvdata(pdev);
+ mdiobus_unregister(bus);
+ kfree(bus->irq);
+ mdiobus_free(bus);
+ return 0;
+}
+
+static const struct of_device_id orion_mdio_match[] = {
+ { .compatible = "marvell,orion-mdio" },
+ { }
+};
+MODULE_DEVICE_TABLE(of, orion_mdio_match);
+
+static struct platform_driver orion_mdio_driver = {
+ .probe = orion_mdio_probe,
+ .remove = __devexit_p(orion_mdio_remove),
+ .driver = {
+ .name = "orion-mdio",
+ .of_match_table = orion_mdio_match,
+ },
+};
+
+module_platform_driver(orion_mdio_driver);
+
+MODULE_DESCRIPTION("Marvell MDIO interface driver");
+MODULE_AUTHOR("Thomas Petazzoni <thomas.petazzoni@free-electrons.com>");
+MODULE_LICENSE("GPL");
--
1.7.9.5
^ permalink raw reply related
* [PATCH v7 4/6] arm: mvebu: add Ethernet controllers using mvneta driver for Armada 370/XP
From: Thomas Petazzoni @ 2012-11-14 14:56 UTC (permalink / raw)
To: David S. Miller
Cc: Francois Romieu, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
In-Reply-To: <1352905010-24172-1-git-send-email-thomas.petazzoni@free-electrons.com>
The Armada 370 SoC has two network units, while the Armada XP has four
network units. The first two network units are common to both the
Armada XP and Armada 370, so they are added to armada-370-xp.dtsi,
while the other two network units are specific to the Armada XP and
therefore added to armada-xp.dtsi.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
arch/arm/boot/dts/armada-370-xp.dtsi | 21 +++++++++++++++++++++
arch/arm/boot/dts/armada-xp.dtsi | 14 ++++++++++++++
2 files changed, 35 insertions(+)
diff --git a/arch/arm/boot/dts/armada-370-xp.dtsi b/arch/arm/boot/dts/armada-370-xp.dtsi
index b113e0b..0cd3331 100644
--- a/arch/arm/boot/dts/armada-370-xp.dtsi
+++ b/arch/arm/boot/dts/armada-370-xp.dtsi
@@ -68,6 +68,27 @@
compatible = "marvell,armada-addr-decoding-controller";
reg = <0xd0020000 0x258>;
};
+
+ mdio {
+ #address-cells = <1>;
+ #size-cells = <0>;
+ compatible = "marvell,orion-mdio";
+ reg = <0xd0072004 0x4>;
+ };
+
+ ethernet@d0070000 {
+ compatible = "marvell,armada-370-neta";
+ reg = <0xd0070000 0x2500>;
+ interrupts = <8>;
+ status = "disabled";
+ };
+
+ ethernet@d0074000 {
+ compatible = "marvell,armada-370-neta";
+ reg = <0xd0074000 0x2500>;
+ interrupts = <10>;
+ status = "disabled";
+ };
};
};
diff --git a/arch/arm/boot/dts/armada-xp.dtsi b/arch/arm/boot/dts/armada-xp.dtsi
index 71d6b5d..3bbbccf 100644
--- a/arch/arm/boot/dts/armada-xp.dtsi
+++ b/arch/arm/boot/dts/armada-xp.dtsi
@@ -51,5 +51,19 @@
compatible = "marvell,armada-370-xp-system-controller";
reg = <0xd0018200 0x500>;
};
+
+ ethernet@d0030000 {
+ compatible = "marvell,armada-370-neta";
+ reg = <0xd0030000 0x2500>;
+ interrupts = <12>;
+ status = "disabled";
+ };
+
+ ethernet@d0034000 {
+ compatible = "marvell,armada-370-neta";
+ reg = <0xd0034000 0x2500>;
+ interrupts = <14>;
+ status = "disabled";
+ };
};
};
--
1.7.9.5
^ permalink raw reply related
* [PATCH v7 5/6] arm: mvebu: enable Ethernet controllers on Armada 370/XP eval boards
From: Thomas Petazzoni @ 2012-11-14 14:56 UTC (permalink / raw)
To: David S. Miller
Cc: Francois Romieu, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
In-Reply-To: <1352905010-24172-1-git-send-email-thomas.petazzoni@free-electrons.com>
This patch enables the two network interfaces of the Armada 370
official Marvell evaluation platform, and the four network interfaces
of the Armada XP official Marvell evaluation platform.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
arch/arm/boot/dts/armada-370-db.dts | 23 +++++++++++++++++++
arch/arm/boot/dts/armada-xp-db.dts | 43 +++++++++++++++++++++++++++++++++++
2 files changed, 66 insertions(+)
diff --git a/arch/arm/boot/dts/armada-370-db.dts b/arch/arm/boot/dts/armada-370-db.dts
index fffd5c2..76362f7 100644
--- a/arch/arm/boot/dts/armada-370-db.dts
+++ b/arch/arm/boot/dts/armada-370-db.dts
@@ -38,5 +38,28 @@
clock-frequency = <600000000>;
status = "okay";
};
+
+ mdio {
+ phy0: ethernet-phy@0 {
+ reg = <0>;
+ };
+
+ phy1: ethernet-phy@1 {
+ reg = <1>;
+ };
+ };
+
+ ethernet@d0070000 {
+ clock-frequency = <200000000>;
+ status = "okay";
+ phy = <&phy0>;
+ phy-mode = "rgmii-id";
+ };
+ ethernet@d0074000 {
+ clock-frequency = <200000000>;
+ status = "okay";
+ phy = <&phy1>;
+ phy-mode = "rgmii-id";
+ };
};
};
diff --git a/arch/arm/boot/dts/armada-xp-db.dts b/arch/arm/boot/dts/armada-xp-db.dts
index b1fc728..b614bd0 100644
--- a/arch/arm/boot/dts/armada-xp-db.dts
+++ b/arch/arm/boot/dts/armada-xp-db.dts
@@ -46,5 +46,48 @@
clock-frequency = <250000000>;
status = "okay";
};
+
+ mdio {
+ phy0: ethernet-phy@0 {
+ reg = <0>;
+ };
+
+ phy1: ethernet-phy@1 {
+ reg = <1>;
+ };
+
+ phy2: ethernet-phy@2 {
+ reg = <25>;
+ };
+
+ phy3: ethernet-phy@3 {
+ reg = <27>;
+ };
+ };
+
+ ethernet@d0070000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy = <&phy0>;
+ phy-mode = "rgmii-id";
+ };
+ ethernet@d0074000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy = <&phy1>;
+ phy-mode = "rgmii-id";
+ };
+ ethernet@d0030000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy = <&phy2>;
+ phy-mode = "sgmii";
+ };
+ ethernet@d0034000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy = <&phy3>;
+ phy-mode = "sgmii";
+ };
};
};
--
1.7.9.5
^ permalink raw reply related
* [PATCH v7 6/6] arm: mvebu: enable Ethernet controllers on OpenBlocks AX3-4 platform
From: Thomas Petazzoni @ 2012-11-14 14:56 UTC (permalink / raw)
To: David S. Miller
Cc: Francois Romieu, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
In-Reply-To: <1352905010-24172-1-git-send-email-thomas.petazzoni@free-electrons.com>
The PlatHome OpenBlocks AX3-4 platform has 4 Ethernet ports, connected
to a single quad-port PHY through SGMII.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts | 24 ++++++++++++++++++++++
1 file changed, 24 insertions(+)
diff --git a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
index cb86853..1a07e09 100644
--- a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
+++ b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
@@ -65,5 +65,29 @@
linux,default-trigger = "heartbeat";
};
};
+ ethernet@d0070000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy-mode = "sgmii";
+ phy-addr = <0>;
+ };
+ ethernet@d0074000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy-mode = "sgmii";
+ phy-addr = <1>;
+ };
+ ethernet@d0030000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy-mode = "sgmii";
+ phy-addr = <2>;
+ };
+ ethernet@d0034000 {
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy-mode = "sgmii";
+ phy-addr = <3>;
+ };
};
};
--
1.7.9.5
^ permalink raw reply related
* [PATCH v7 3/6] net: mvneta: update MAINTAINERS file for the mvneta maintainers
From: Thomas Petazzoni @ 2012-11-14 14:56 UTC (permalink / raw)
To: David S. Miller
Cc: Francois Romieu, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
In-Reply-To: <1352905010-24172-1-git-send-email-thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
MAINTAINERS | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 59203e7..3010f1a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4700,6 +4700,12 @@ S: Maintained
F: drivers/net/ethernet/marvell/mv643xx_eth.*
F: include/linux/mv643xx.h
+MARVELL MVNETA ETHERNET DRIVER
+M: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
+L: netdev@vger.kernel.org
+S: Maintained
+F: drivers/net/ethernet/marvell/mvneta.*
+
MARVELL MWIFIEX WIRELESS DRIVER
M: Bing Zhao <bzhao@marvell.com>
L: linux-wireless@vger.kernel.org
--
1.7.9.5
^ permalink raw reply related
* [PATCH v7 2/6] net: mvneta: driver for Marvell Armada 370/XP network unit
From: Thomas Petazzoni @ 2012-11-14 14:56 UTC (permalink / raw)
To: David S. Miller
Cc: Francois Romieu, Lennert Buytenhek, netdev, linux-arm-kernel,
Jason Cooper, Andrew Lunn, Gregory Clement, Lior Amsalem,
Dmitri Epshtein
In-Reply-To: <1352905010-24172-1-git-send-email-thomas.petazzoni@free-electrons.com>
This patch contains a new network driver for the network unit of the
ARM Marvell Armada 370 and the Armada XP. Both SoCs use the PJ4B
processor, a Marvell-developed ARM core that implements the ARMv7
instruction set.
Compared to previous ARM Marvell SoCs (Kirkwood, Orion, Discovery),
the network unit in Armada 370 and Armada XP is highly different. This
is the reason why this new 'mvneta' driver is needed, while the older
ARM Marvell SoCs use the 'mv643xx_eth' driver.
Here is an overview of the most important hardware changes that
require a new, specific, driver for the network unit of Armada 370/XP:
- The new network unit has a completely different design and layout
for the RX and TX descriptors. They are now organized as a simple
array (each RX and TX queue has base address and size of this
array) rather than a linked list as in the old SoCs.
- The new network unit has a different RXQ and TXQ management: this
management is done using special read/write counter registers,
while in the Old SocS, it was done using the Ownership bit in RX
and TX descriptors.
- The new network unit has different interrupt registers
- The new network unit way of cleaning of interrupts is not done by
writing to the cause register, but by updating per-queue counters
- The new network unit has different GMAC registers (link, speed,
duplex configuration) and different WRR registers.
- The new network unit has lots of new units like PnC (Parser and
Classifier), PMT, BM (Memory Buffer Management), xPON, and more.
The driver proposed in the current patch only handles the basic
features. Additional hardware features will progressively be supported
as needed.
This code has originally been written by Rami Rosen
<rosenr@marvell.com>, and then reviewed and cleaned up by Thomas
Petazzoni <thomas.petazzoni@free-electrons.com>.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
.../bindings/net/marvell-armada-370-neta.txt | 23 +
drivers/net/ethernet/marvell/Kconfig | 13 +
drivers/net/ethernet/marvell/Makefile | 1 +
drivers/net/ethernet/marvell/mvneta.c | 2839 ++++++++++++++++++++
4 files changed, 2876 insertions(+)
create mode 100644 Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt
create mode 100644 drivers/net/ethernet/marvell/mvneta.c
diff --git a/Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt b/Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt
new file mode 100644
index 0000000..c4e87f0
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/marvell-armada-370-neta.txt
@@ -0,0 +1,23 @@
+* Marvell Armada 370 / Armada XP Ethernet Controller (NETA)
+
+Required properties:
+- compatible: should be "marvell,armada-370-neta".
+- reg: address and length of the register set for the device.
+- interrupts: interrupt for the device
+- phy: A phandle to a phy node defining the PHY address (as the reg
+ property, a single integer).
+- phy-mode: The interface between the SoC and the PHY (a string that
+ of_get_phy_mode() can understand)
+- clock-frequency: frequency of the peripheral clock of the SoC.
+
+Example:
+
+ethernet@d0070000 {
+ compatible = "marvell,armada-370-neta";
+ reg = <0xd0070000 0x2500>;
+ interrupts = <8>;
+ clock-frequency = <250000000>;
+ status = "okay";
+ phy = <&phy0>;
+ phy-mode = "rgmii-id";
+};
diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index 232ccb3..edfba93 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -42,6 +42,19 @@ config MVMDIO
(used on Armada 370 and XP), but it could be used in the
future by the MV643XX_ETH driver.
+config MVNETA
+ tristate "Marvell Armada 370/XP network interface support"
+ depends on MACH_ARMADA_370_XP
+ select PHYLIB
+ select MVMDIO
+ ---help---
+ This driver supports the network interface units in the
+ Marvell ARMADA XP and ARMADA 370 SoC family.
+
+ Note that this driver is distinct from the mv643xx_eth
+ driver, which should be used for the older Marvell SoCs
+ (Dove, Orion, Discovery, Kirkwood).
+
config PXA168_ETH
tristate "Marvell pxa168 ethernet support"
depends on CPU_PXA168
diff --git a/drivers/net/ethernet/marvell/Makefile b/drivers/net/ethernet/marvell/Makefile
index 0438599..7f63b4a 100644
--- a/drivers/net/ethernet/marvell/Makefile
+++ b/drivers/net/ethernet/marvell/Makefile
@@ -4,6 +4,7 @@
obj-$(CONFIG_MV643XX_ETH) += mv643xx_eth.o
obj-$(CONFIG_MVMDIO) += mvmdio.o
+obj-$(CONFIG_MVNETA) += mvneta.o
obj-$(CONFIG_PXA168_ETH) += pxa168_eth.o
obj-$(CONFIG_SKGE) += skge.o
obj-$(CONFIG_SKY2) += sky2.o
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
new file mode 100644
index 0000000..a7826f0
--- /dev/null
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -0,0 +1,2839 @@
+/*
+ * Driver for Marvell NETA network card for Armada XP and Armada 370 SoCs.
+ *
+ * Copyright (C) 2012 Marvell
+ *
+ * Rami Rosen <rosenr@marvell.com>
+ * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/version.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/platform_device.h>
+#include <linux/skbuff.h>
+#include <linux/inetdevice.h>
+#include <linux/mbus.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
+#include <linux/of.h>
+#include <linux/of_irq.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/of_address.h>
+#include <linux/phy.h>
+
+/* Registers */
+#define MVNETA_RXQ_CONFIG_REG(q) (0x1400 + ((q) << 2))
+#define MVNETA_RXQ_HW_BUF_ALLOC BIT(1)
+#define MVNETA_RXQ_PKT_OFFSET_ALL_MASK (0xf << 8)
+#define MVNETA_RXQ_PKT_OFFSET_MASK(offs) ((offs) << 8)
+#define MVNETA_RXQ_THRESHOLD_REG(q) (0x14c0 + ((q) << 2))
+#define MVNETA_RXQ_NON_OCCUPIED(v) ((v) << 16)
+#define MVNETA_RXQ_BASE_ADDR_REG(q) (0x1480 + ((q) << 2))
+#define MVNETA_RXQ_SIZE_REG(q) (0x14a0 + ((q) << 2))
+#define MVNETA_RXQ_BUF_SIZE_SHIFT 19
+#define MVNETA_RXQ_BUF_SIZE_MASK (0x1fff << 19)
+#define MVNETA_RXQ_STATUS_REG(q) (0x14e0 + ((q) << 2))
+#define MVNETA_RXQ_OCCUPIED_ALL_MASK 0x3fff
+#define MVNETA_RXQ_STATUS_UPDATE_REG(q) (0x1500 + ((q) << 2))
+#define MVNETA_RXQ_ADD_NON_OCCUPIED_SHIFT 16
+#define MVNETA_RXQ_ADD_NON_OCCUPIED_MAX 255
+#define MVNETA_PORT_RX_RESET 0x1cc0
+#define MVNETA_PORT_RX_DMA_RESET BIT(0)
+#define MVNETA_PHY_ADDR 0x2000
+#define MVNETA_PHY_ADDR_MASK 0x1f
+#define MVNETA_MBUS_RETRY 0x2010
+#define MVNETA_UNIT_INTR_CAUSE 0x2080
+#define MVNETA_UNIT_CONTROL 0x20B0
+#define MVNETA_PHY_POLLING_ENABLE BIT(1)
+#define MVNETA_WIN_BASE(w) (0x2200 + ((w) << 3))
+#define MVNETA_WIN_SIZE(w) (0x2204 + ((w) << 3))
+#define MVNETA_WIN_REMAP(w) (0x2280 + ((w) << 2))
+#define MVNETA_BASE_ADDR_ENABLE 0x2290
+#define MVNETA_PORT_CONFIG 0x2400
+#define MVNETA_UNI_PROMISC_MODE BIT(0)
+#define MVNETA_DEF_RXQ(q) ((q) << 1)
+#define MVNETA_DEF_RXQ_ARP(q) ((q) << 4)
+#define MVNETA_TX_UNSET_ERR_SUM BIT(12)
+#define MVNETA_DEF_RXQ_TCP(q) ((q) << 16)
+#define MVNETA_DEF_RXQ_UDP(q) ((q) << 19)
+#define MVNETA_DEF_RXQ_BPDU(q) ((q) << 22)
+#define MVNETA_RX_CSUM_WITH_PSEUDO_HDR BIT(25)
+#define MVNETA_PORT_CONFIG_DEFL_VALUE(q) (MVNETA_DEF_RXQ(q) | \
+ MVNETA_DEF_RXQ_ARP(q) | \
+ MVNETA_DEF_RXQ_TCP(q) | \
+ MVNETA_DEF_RXQ_UDP(q) | \
+ MVNETA_DEF_RXQ_BPDU(q) | \
+ MVNETA_TX_UNSET_ERR_SUM | \
+ MVNETA_RX_CSUM_WITH_PSEUDO_HDR)
+#define MVNETA_PORT_CONFIG_EXTEND 0x2404
+#define MVNETA_MAC_ADDR_LOW 0x2414
+#define MVNETA_MAC_ADDR_HIGH 0x2418
+#define MVNETA_SDMA_CONFIG 0x241c
+#define MVNETA_SDMA_BRST_SIZE_16 4
+#define MVNETA_NO_DESC_SWAP 0x0
+#define MVNETA_RX_BRST_SZ_MASK(burst) ((burst) << 1)
+#define MVNETA_RX_NO_DATA_SWAP BIT(4)
+#define MVNETA_TX_NO_DATA_SWAP BIT(5)
+#define MVNETA_TX_BRST_SZ_MASK(burst) ((burst) << 22)
+#define MVNETA_PORT_STATUS 0x2444
+#define MVNETA_TX_IN_PRGRS BIT(1)
+#define MVNETA_TX_FIFO_EMPTY BIT(8)
+#define MVNETA_RX_MIN_FRAME_SIZE 0x247c
+#define MVNETA_TYPE_PRIO 0x24bc
+#define MVNETA_FORCE_UNI BIT(21)
+#define MVNETA_TXQ_CMD_1 0x24e4
+#define MVNETA_TXQ_CMD 0x2448
+#define MVNETA_TXQ_DISABLE_SHIFT 8
+#define MVNETA_TXQ_ENABLE_MASK 0x000000ff
+#define MVNETA_ACC_MODE 0x2500
+#define MVNETA_CPU_MAP(cpu) (0x2540 + ((cpu) << 2))
+#define MVNETA_CPU_RXQ_ACCESS_ALL_MASK 0x000000ff
+#define MVNETA_CPU_TXQ_ACCESS_ALL_MASK 0x0000ff00
+#define MVNETA_RXQ_TIME_COAL_REG(q) (0x2580 + ((q) << 2))
+#define MVNETA_INTR_NEW_CAUSE 0x25a0
+#define MVNETA_RX_INTR_MASK(nr_rxqs) (((1 << nr_rxqs) - 1) << 8)
+#define MVNETA_INTR_NEW_MASK 0x25a4
+#define MVNETA_INTR_OLD_CAUSE 0x25a8
+#define MVNETA_INTR_OLD_MASK 0x25ac
+#define MVNETA_INTR_MISC_CAUSE 0x25b0
+#define MVNETA_INTR_MISC_MASK 0x25b4
+#define MVNETA_INTR_ENABLE 0x25b8
+#define MVNETA_TXQ_INTR_ENABLE_ALL_MASK 0x0000ff00
+#define MVNETA_RXQ_INTR_ENABLE_ALL_MASK 0xff000000
+#define MVNETA_RXQ_CMD 0x2680
+#define MVNETA_RXQ_DISABLE_SHIFT 8
+#define MVNETA_RXQ_ENABLE_MASK 0x000000ff
+#define MVETH_TXQ_TOKEN_COUNT_REG(q) (0x2700 + ((q) << 4))
+#define MVETH_TXQ_TOKEN_CFG_REG(q) (0x2704 + ((q) << 4))
+#define MVNETA_GMAC_CTRL_0 0x2c00
+#define MVNETA_GMAC_MAX_RX_SIZE_SHIFT 2
+#define MVNETA_GMAC_MAX_RX_SIZE_MASK 0x7ffc
+#define MVNETA_GMAC0_PORT_ENABLE BIT(0)
+#define MVNETA_GMAC_CTRL_2 0x2c08
+#define MVNETA_GMAC2_PSC_ENABLE BIT(3)
+#define MVNETA_GMAC2_PORT_RGMII BIT(4)
+#define MVNETA_GMAC2_PORT_RESET BIT(6)
+#define MVNETA_GMAC_STATUS 0x2c10
+#define MVNETA_GMAC_LINK_UP BIT(0)
+#define MVNETA_GMAC_SPEED_1000 BIT(1)
+#define MVNETA_GMAC_SPEED_100 BIT(2)
+#define MVNETA_GMAC_FULL_DUPLEX BIT(3)
+#define MVNETA_GMAC_RX_FLOW_CTRL_ENABLE BIT(4)
+#define MVNETA_GMAC_TX_FLOW_CTRL_ENABLE BIT(5)
+#define MVNETA_GMAC_RX_FLOW_CTRL_ACTIVE BIT(6)
+#define MVNETA_GMAC_TX_FLOW_CTRL_ACTIVE BIT(7)
+#define MVNETA_GMAC_AUTONEG_CONFIG 0x2c0c
+#define MVNETA_GMAC_FORCE_LINK_DOWN BIT(0)
+#define MVNETA_GMAC_FORCE_LINK_PASS BIT(1)
+#define MVNETA_GMAC_CONFIG_MII_SPEED BIT(5)
+#define MVNETA_GMAC_CONFIG_GMII_SPEED BIT(6)
+#define MVNETA_GMAC_CONFIG_FULL_DUPLEX BIT(12)
+#define MVNETA_MIB_COUNTERS_BASE 0x3080
+#define MVNETA_MIB_LATE_COLLISION 0x7c
+#define MVNETA_DA_FILT_SPEC_MCAST 0x3400
+#define MVNETA_DA_FILT_OTH_MCAST 0x3500
+#define MVNETA_DA_FILT_UCAST_BASE 0x3600
+#define MVNETA_TXQ_BASE_ADDR_REG(q) (0x3c00 + ((q) << 2))
+#define MVNETA_TXQ_SIZE_REG(q) (0x3c20 + ((q) << 2))
+#define MVNETA_TXQ_SENT_THRESH_ALL_MASK 0x3fff0000
+#define MVNETA_TXQ_SENT_THRESH_MASK(coal) ((coal) << 16)
+#define MVNETA_TXQ_UPDATE_REG(q) (0x3c60 + ((q) << 2))
+#define MVNETA_TXQ_DEC_SENT_SHIFT 16
+#define MVNETA_TXQ_STATUS_REG(q) (0x3c40 + ((q) << 2))
+#define MVNETA_TXQ_SENT_DESC_SHIFT 16
+#define MVNETA_TXQ_SENT_DESC_MASK 0x3fff0000
+#define MVNETA_PORT_TX_RESET 0x3cf0
+#define MVNETA_PORT_TX_DMA_RESET BIT(0)
+#define MVNETA_TX_MTU 0x3e0c
+#define MVNETA_TX_TOKEN_SIZE 0x3e14
+#define MVNETA_TX_TOKEN_SIZE_MAX 0xffffffff
+#define MVNETA_TXQ_TOKEN_SIZE_REG(q) (0x3e40 + ((q) << 2))
+#define MVNETA_TXQ_TOKEN_SIZE_MAX 0x7fffffff
+
+#define MVNETA_CAUSE_TXQ_SENT_DESC_ALL_MASK 0xff
+
+/* Descriptor ring Macros */
+#define MVNETA_QUEUE_NEXT_DESC(q, index) \
+ (((index) < (q)->last_desc) ? ((index) + 1) : 0)
+
+/* Various constants */
+
+/* Coalescing */
+#define MVNETA_TXDONE_COAL_PKTS 16
+#define MVNETA_RX_COAL_PKTS 32
+#define MVNETA_RX_COAL_USEC 100
+
+/* Timer */
+#define MVNETA_TX_DONE_TIMER_PERIOD 10
+
+/* Napi polling weight */
+#define MVNETA_RX_POLL_WEIGHT 64
+
+/*
+ * The two bytes Marvell header. Either contains a special value used
+ * by Marvell switches when a specific hardware mode is enabled (not
+ * supported by this driver) or is filled automatically by zeroes on
+ * the RX side. Those two bytes being at the front of the Ethernet
+ * header, they allow to have the IP header aligned on a 4 bytes
+ * boundary automatically: the hardware skips those two bytes on its
+ * own.
+ */
+#define MVNETA_MH_SIZE 2
+
+#define MVNETA_VLAN_TAG_LEN 4
+
+#define MVNETA_CPU_D_CACHE_LINE_SIZE 32
+#define MVNETA_TX_CSUM_MAX_SIZE 9800
+#define MVNETA_ACC_MODE_EXT 1
+
+/* Timeout constants */
+#define MVNETA_TX_DISABLE_TIMEOUT_MSEC 1000
+#define MVNETA_RX_DISABLE_TIMEOUT_MSEC 1000
+#define MVNETA_TX_FIFO_EMPTY_TIMEOUT 10000
+
+#define MVNETA_TX_MTU_MAX 0x3ffff
+
+/* Max number of Rx descriptors */
+#define MVNETA_MAX_RXD 128
+
+/* Max number of Tx descriptors */
+#define MVNETA_MAX_TXD 532
+
+/* descriptor aligned size */
+#define MVNETA_DESC_ALIGNED_SIZE 32
+
+#define MVNETA_RX_PKT_SIZE(mtu) \
+ ALIGN((mtu) + MVNETA_MH_SIZE + MVNETA_VLAN_TAG_LEN + \
+ ETH_HLEN + ETH_FCS_LEN, \
+ MVNETA_CPU_D_CACHE_LINE_SIZE)
+
+#define MVNETA_RX_BUF_SIZE(pkt_size) ((pkt_size) + NET_SKB_PAD)
+
+struct mvneta_stats {
+ struct u64_stats_sync syncp;
+ u64 packets;
+ u64 bytes;
+};
+
+struct mvneta_port {
+ int pkt_size;
+ void __iomem *base;
+ struct mvneta_rx_queue *rxqs;
+ struct mvneta_tx_queue *txqs;
+ struct timer_list tx_done_timer;
+ struct net_device *dev;
+
+ u32 cause_rx_tx;
+ struct napi_struct napi;
+
+ /* Flags */
+ unsigned long flags;
+#define MVNETA_F_TX_DONE_TIMER_BIT 0
+
+ /* Napi weight */
+ int weight;
+
+ /* Core clock */
+ unsigned int clk_rate_hz;
+ u8 mcast_count[256];
+ u16 tx_ring_size;
+ u16 rx_ring_size;
+ struct mvneta_stats tx_stats;
+ struct mvneta_stats rx_stats;
+
+ struct mii_bus *mii_bus;
+ struct phy_device *phy_dev;
+ phy_interface_t phy_interface;
+ struct device_node *phy_node;
+ unsigned int link;
+ unsigned int duplex;
+ unsigned int speed;
+};
+
+/*
+ * The mvneta_tx_desc and mvneta_rx_desc structures describe the
+ * layout of the transmit and reception DMA descriptors, and their
+ * layout is therefore defined by the hardware design
+ */
+struct mvneta_tx_desc {
+ u32 command; /* Options used by HW for packet transmitting.*/
+#define MVNETA_TX_L3_OFF_SHIFT 0
+#define MVNETA_TX_IP_HLEN_SHIFT 8
+#define MVNETA_TX_L4_UDP BIT(16)
+#define MVNETA_TX_L3_IP6 BIT(17)
+#define MVNETA_TXD_IP_CSUM BIT(18)
+#define MVNETA_TXD_Z_PAD BIT(19)
+#define MVNETA_TXD_L_DESC BIT(20)
+#define MVNETA_TXD_F_DESC BIT(21)
+#define MVNETA_TXD_FLZ_DESC (MVNETA_TXD_Z_PAD | \
+ MVNETA_TXD_L_DESC | \
+ MVNETA_TXD_F_DESC)
+#define MVNETA_TX_L4_CSUM_FULL BIT(30)
+#define MVNETA_TX_L4_CSUM_NOT BIT(31)
+
+ u16 reserverd1; /* csum_l4 (for future use) */
+ u16 data_size; /* Data size of transmitted packet in bytes */
+ u32 buf_phys_addr; /* Physical addr of transmitted buffer */
+ u32 reserved2; /* hw_cmd - (for future use, PMT) */
+ u32 reserved3[4]; /* Reserved - (for future use) */
+};
+
+struct mvneta_rx_desc {
+ u32 status; /* Info about received packet */
+#define MVNETA_RXD_ERR_CRC 0x0
+#define MVNETA_RXD_ERR_SUMMARY BIT(16)
+#define MVNETA_RXD_ERR_OVERRUN BIT(17)
+#define MVNETA_RXD_ERR_LEN BIT(18)
+#define MVNETA_RXD_ERR_RESOURCE (BIT(17) | BIT(18))
+#define MVNETA_RXD_ERR_CODE_MASK (BIT(17) | BIT(18))
+#define MVNETA_RXD_L3_IP4 BIT(25)
+#define MVNETA_RXD_FIRST_LAST_DESC (BIT(26) | BIT(27))
+#define MVNETA_RXD_L4_CSUM_OK BIT(30)
+
+ u16 reserved1; /* pnc_info - (for future use, PnC) */
+ u16 data_size; /* Size of received packet in bytes */
+ u32 buf_phys_addr; /* Physical address of the buffer */
+ u32 reserved2; /* pnc_flow_id (for future use, PnC) */
+ u32 buf_cookie; /* cookie for access to RX buffer in rx path */
+ u16 reserved3; /* prefetch_cmd, for future use */
+ u16 reserved4; /* csum_l4 - (for future use, PnC) */
+ u32 reserved5; /* pnc_extra PnC (for future use, PnC) */
+ u32 reserved6; /* hw_cmd (for future use, PnC and HWF) */
+};
+
+struct mvneta_tx_queue {
+ /* Number of this TX queue, in the range 0-7 */
+ u8 id;
+
+ /* Number of TX DMA descriptors in the descriptor ring */
+ int size;
+
+ /* Number of currently used TX DMA descriptor in the
+ * descriptor ring */
+ int count;
+
+ /* Array of transmitted skb */
+ struct sk_buff **tx_skb;
+
+ /* Index of last TX DMA descriptor that was inserted */
+ int txq_put_index;
+
+ /* Index of the TX DMA descriptor to be cleaned up */
+ int txq_get_index;
+
+ u32 done_pkts_coal;
+
+ /* Virtual address of the TX DMA descriptors array */
+ struct mvneta_tx_desc *descs;
+
+ /* DMA address of the TX DMA descriptors array */
+ dma_addr_t descs_phys;
+
+ /* Index of the last TX DMA descriptor */
+ int last_desc;
+
+ /* Index of the next TX DMA descriptor to process */
+ int next_desc_to_proc;
+};
+
+struct mvneta_rx_queue {
+ /* rx queue number, in the range 0-7 */
+ u8 id;
+
+ /* num of rx descriptors in the rx descriptor ring */
+ int size;
+
+ /* counter of times when mvneta_refill() failed */
+ int missed;
+
+ u32 pkts_coal;
+ u32 time_coal;
+
+ /* Virtual address of the RX DMA descriptors array */
+ struct mvneta_rx_desc *descs;
+
+ /* DMA address of the RX DMA descriptors array */
+ dma_addr_t descs_phys;
+
+ /* Index of the last RX DMA descriptor */
+ int last_desc;
+
+ /* Index of the next RX DMA descriptor to process */
+ int next_desc_to_proc;
+};
+
+static int rxq_number = 8;
+static int txq_number = 8;
+
+static int rxq_def;
+static int txq_def;
+
+#define MVNETA_DRIVER_NAME "mvneta"
+#define MVNETA_DRIVER_VERSION "1.0"
+
+/* Utility/helper methods */
+
+/* Write helper method */
+static void mvreg_write(struct mvneta_port *pp, u32 offset, u32 data)
+{
+ writel(data, pp->base + offset);
+}
+
+/* Read helper method */
+static u32 mvreg_read(struct mvneta_port *pp, u32 offset)
+{
+ return readl(pp->base + offset);
+}
+
+/* Increment txq get counter */
+static void mvneta_txq_inc_get(struct mvneta_tx_queue *txq)
+{
+ txq->txq_get_index++;
+ if (txq->txq_get_index == txq->size)
+ txq->txq_get_index = 0;
+}
+
+/* Increment txq put counter */
+static void mvneta_txq_inc_put(struct mvneta_tx_queue *txq)
+{
+ txq->txq_put_index++;
+ if (txq->txq_put_index == txq->size)
+ txq->txq_put_index = 0;
+}
+
+
+/* Clear all MIB counters */
+static void mvneta_mib_counters_clear(struct mvneta_port *pp)
+{
+ int i;
+ u32 dummy;
+
+ /* Perform dummy reads from MIB counters */
+ for (i = 0; i < MVNETA_MIB_LATE_COLLISION; i += 4)
+ dummy = mvreg_read(pp, (MVNETA_MIB_COUNTERS_BASE + i));
+}
+
+/* Get System Network Statistics */
+struct rtnl_link_stats64 *mvneta_get_stats64(struct net_device *dev,
+ struct rtnl_link_stats64 *stats)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ unsigned int start;
+
+ memset(stats, 0, sizeof(struct rtnl_link_stats64));
+
+ do {
+ start = u64_stats_fetch_begin_bh(&pp->rx_stats.syncp);
+ stats->rx_packets = pp->rx_stats.packets;
+ stats->rx_bytes = pp->rx_stats.bytes;
+ } while (u64_stats_fetch_retry_bh(&pp->rx_stats.syncp, start));
+
+
+ do {
+ start = u64_stats_fetch_begin_bh(&pp->tx_stats.syncp);
+ stats->tx_packets = pp->tx_stats.packets;
+ stats->tx_bytes = pp->tx_stats.bytes;
+ } while (u64_stats_fetch_retry_bh(&pp->tx_stats.syncp, start));
+
+ stats->rx_errors = dev->stats.rx_errors;
+ stats->rx_dropped = dev->stats.rx_dropped;
+
+ stats->tx_dropped = dev->stats.tx_dropped;
+
+ return stats;
+}
+
+/* Rx descriptors helper methods */
+
+/*
+ * Checks whether the given RX descriptor is both the first and the
+ * last descriptor for the RX packet. Each RX packet is currently
+ * received through a single RX descriptor, so not having each RX
+ * descriptor with its first and last bits set is an error
+ */
+static int mvneta_rxq_desc_is_first_last(struct mvneta_rx_desc *desc)
+{
+ return (desc->status & MVNETA_RXD_FIRST_LAST_DESC) ==
+ MVNETA_RXD_FIRST_LAST_DESC;
+}
+
+/* Add number of descriptors ready to receive new packets */
+static void mvneta_rxq_non_occup_desc_add(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq,
+ int ndescs)
+{
+ /* Only MVNETA_RXQ_ADD_NON_OCCUPIED_MAX (255) descriptors can
+ * be added at once */
+ while (ndescs > MVNETA_RXQ_ADD_NON_OCCUPIED_MAX) {
+ mvreg_write(pp, MVNETA_RXQ_STATUS_UPDATE_REG(rxq->id),
+ (MVNETA_RXQ_ADD_NON_OCCUPIED_MAX <<
+ MVNETA_RXQ_ADD_NON_OCCUPIED_SHIFT));
+ ndescs -= MVNETA_RXQ_ADD_NON_OCCUPIED_MAX;
+ }
+
+ mvreg_write(pp, MVNETA_RXQ_STATUS_UPDATE_REG(rxq->id),
+ (ndescs << MVNETA_RXQ_ADD_NON_OCCUPIED_SHIFT));
+}
+
+/* Get number of RX descriptors occupied by received packets */
+static int mvneta_rxq_busy_desc_num_get(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_RXQ_STATUS_REG(rxq->id));
+ return val & MVNETA_RXQ_OCCUPIED_ALL_MASK;
+}
+
+/*
+ * Update num of rx desc called upon return from rx path or
+ * from mvneta_rxq_drop_pkts().
+ */
+static void mvneta_rxq_desc_num_update(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq,
+ int rx_done, int rx_filled)
+{
+ u32 val;
+
+ if ((rx_done <= 0xff) && (rx_filled <= 0xff)) {
+ val = rx_done |
+ (rx_filled << MVNETA_RXQ_ADD_NON_OCCUPIED_SHIFT);
+ mvreg_write(pp, MVNETA_RXQ_STATUS_UPDATE_REG(rxq->id), val);
+ return;
+ }
+
+ /* Only 255 descriptors can be added at once */
+ while ((rx_done > 0) || (rx_filled > 0)) {
+ if (rx_done <= 0xff) {
+ val = rx_done;
+ rx_done = 0;
+ } else {
+ val = 0xff;
+ rx_done -= 0xff;
+ }
+ if (rx_filled <= 0xff) {
+ val |= rx_filled << MVNETA_RXQ_ADD_NON_OCCUPIED_SHIFT;
+ rx_filled = 0;
+ } else {
+ val |= 0xff << MVNETA_RXQ_ADD_NON_OCCUPIED_SHIFT;
+ rx_filled -= 0xff;
+ }
+ mvreg_write(pp, MVNETA_RXQ_STATUS_UPDATE_REG(rxq->id), val);
+ }
+}
+
+/* Get pointer to next RX descriptor to be processed by SW */
+static struct mvneta_rx_desc *
+mvneta_rxq_next_desc_get(struct mvneta_rx_queue *rxq)
+{
+ int rx_desc = rxq->next_desc_to_proc;
+
+ rxq->next_desc_to_proc = MVNETA_QUEUE_NEXT_DESC(rxq, rx_desc);
+ return rxq->descs + rx_desc;
+}
+
+/* Change maximum receive size of the port. */
+static void mvneta_max_rx_size_set(struct mvneta_port *pp, int max_rx_size)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_GMAC_CTRL_0);
+ val &= ~MVNETA_GMAC_MAX_RX_SIZE_MASK;
+ val |= ((max_rx_size - MVNETA_MH_SIZE) / 2) <<
+ MVNETA_GMAC_MAX_RX_SIZE_SHIFT;
+ mvreg_write(pp, MVNETA_GMAC_CTRL_0, val);
+}
+
+
+/* Set rx queue offset */
+static void mvneta_rxq_offset_set(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq,
+ int offset)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_RXQ_CONFIG_REG(rxq->id));
+ val &= ~MVNETA_RXQ_PKT_OFFSET_ALL_MASK;
+
+ /* Offset is in */
+ val |= MVNETA_RXQ_PKT_OFFSET_MASK(offset >> 3);
+ mvreg_write(pp, MVNETA_RXQ_CONFIG_REG(rxq->id), val);
+}
+
+
+/* Tx descriptors helper methods */
+
+/* Update HW with number of TX descriptors to be sent */
+static void mvneta_txq_pend_desc_add(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq,
+ int pend_desc)
+{
+ u32 val;
+
+ /* Only 255 descriptors can be added at once ; Assume caller
+ process TX desriptors in quanta less than 256 */
+ val = pend_desc;
+ mvreg_write(pp, MVNETA_TXQ_UPDATE_REG(txq->id), val);
+}
+
+/* Get pointer to next TX descriptor to be processed (send) by HW */
+static struct mvneta_tx_desc *
+mvneta_txq_next_desc_get(struct mvneta_tx_queue *txq)
+{
+ int tx_desc = txq->next_desc_to_proc;
+
+ txq->next_desc_to_proc = MVNETA_QUEUE_NEXT_DESC(txq, tx_desc);
+ return txq->descs + tx_desc;
+}
+
+/* Release the last allocated TX descriptor. Useful to handle DMA
+ * mapping failures in the TX path. */
+static void mvneta_txq_desc_put(struct mvneta_tx_queue *txq)
+{
+ if (txq->next_desc_to_proc == 0)
+ txq->next_desc_to_proc = txq->last_desc - 1;
+ else
+ txq->next_desc_to_proc--;
+}
+
+/* Set rxq buf size */
+static void mvneta_rxq_buf_size_set(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq,
+ int buf_size)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_RXQ_SIZE_REG(rxq->id));
+
+ val &= ~MVNETA_RXQ_BUF_SIZE_MASK;
+ val |= ((buf_size >> 3) << MVNETA_RXQ_BUF_SIZE_SHIFT);
+
+ mvreg_write(pp, MVNETA_RXQ_SIZE_REG(rxq->id), val);
+}
+
+/* Disable buffer management (BM) */
+static void mvneta_rxq_bm_disable(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_RXQ_CONFIG_REG(rxq->id));
+ val &= ~MVNETA_RXQ_HW_BUF_ALLOC;
+ mvreg_write(pp, MVNETA_RXQ_CONFIG_REG(rxq->id), val);
+}
+
+
+
+/* Sets the RGMII Enable bit (RGMIIEn) in port MAC control register */
+static void __devinit mvneta_gmac_rgmii_set(struct mvneta_port *pp, int enable)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
+
+ if (enable)
+ val |= MVNETA_GMAC2_PORT_RGMII;
+ else
+ val &= ~MVNETA_GMAC2_PORT_RGMII;
+
+ mvreg_write(pp, MVNETA_GMAC_CTRL_2, val);
+}
+
+/* Config SGMII port */
+static void __devinit mvneta_port_sgmii_config(struct mvneta_port *pp)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
+ val |= MVNETA_GMAC2_PSC_ENABLE;
+ mvreg_write(pp, MVNETA_GMAC_CTRL_2, val);
+}
+
+/* Start the Ethernet port RX and TX activity */
+static void mvneta_port_up(struct mvneta_port *pp)
+{
+ int queue;
+ u32 q_map;
+
+ /* Enable all initialized TXs. */
+ mvneta_mib_counters_clear(pp);
+ q_map = 0;
+ for (queue = 0; queue < txq_number; queue++) {
+ struct mvneta_tx_queue *txq = &pp->txqs[queue];
+ if (txq->descs != NULL)
+ q_map |= (1 << queue);
+ }
+ mvreg_write(pp, MVNETA_TXQ_CMD, q_map);
+
+ /* Enable all initialized RXQs. */
+ q_map = 0;
+ for (queue = 0; queue < rxq_number; queue++) {
+ struct mvneta_rx_queue *rxq = &pp->rxqs[queue];
+ if (rxq->descs != NULL)
+ q_map |= (1 << queue);
+ }
+
+ mvreg_write(pp, MVNETA_RXQ_CMD, q_map);
+}
+
+/* Stop the Ethernet port activity */
+static void mvneta_port_down(struct mvneta_port *pp)
+{
+ u32 val;
+ int count;
+
+ /* Stop Rx port activity. Check port Rx activity. */
+ val = mvreg_read(pp, MVNETA_RXQ_CMD) & MVNETA_RXQ_ENABLE_MASK;
+
+ /* Issue stop command for active channels only */
+ if (val != 0)
+ mvreg_write(pp, MVNETA_RXQ_CMD,
+ val << MVNETA_RXQ_DISABLE_SHIFT);
+
+ /* Wait for all Rx activity to terminate. */
+ count = 0;
+ do {
+ if (count++ >= MVNETA_RX_DISABLE_TIMEOUT_MSEC) {
+ netdev_warn(pp->dev,
+ "TIMEOUT for RX stopped ! rx_queue_cmd: 0x08%x\n",
+ val);
+ break;
+ }
+ mdelay(1);
+
+ val = mvreg_read(pp, MVNETA_RXQ_CMD);
+ } while (val & 0xff);
+
+ /* Stop Tx port activity. Check port Tx activity. Issue stop
+ command for active channels only */
+ val = (mvreg_read(pp, MVNETA_TXQ_CMD)) & MVNETA_TXQ_ENABLE_MASK;
+
+ if (val != 0)
+ mvreg_write(pp, MVNETA_TXQ_CMD,
+ (val << MVNETA_TXQ_DISABLE_SHIFT));
+
+ /* Wait for all Tx activity to terminate. */
+ count = 0;
+ do {
+ if (count++ >= MVNETA_TX_DISABLE_TIMEOUT_MSEC) {
+ netdev_warn(pp->dev,
+ "TIMEOUT for TX stopped status=0x%08x\n",
+ val);
+ break;
+ }
+ mdelay(1);
+
+ /* Check TX Command reg that all Txqs are stopped */
+ val = mvreg_read(pp, MVNETA_TXQ_CMD);
+
+ } while (val & 0xff);
+
+ /* Double check to verify that TX FIFO is empty */
+ count = 0;
+ do {
+ if (count++ >= MVNETA_TX_FIFO_EMPTY_TIMEOUT) {
+ netdev_warn(pp->dev,
+ "TX FIFO empty timeout status=0x08%x\n",
+ val);
+ break;
+ }
+ mdelay(1);
+
+ val = mvreg_read(pp, MVNETA_PORT_STATUS);
+ } while (!(val & MVNETA_TX_FIFO_EMPTY) &&
+ (val & MVNETA_TX_IN_PRGRS));
+
+ udelay(200);
+}
+
+/* Enable the port by setting the port enable bit of the MAC control register */
+static void mvneta_port_enable(struct mvneta_port *pp)
+{
+ u32 val;
+
+ /* Enable port */
+ val = mvreg_read(pp, MVNETA_GMAC_CTRL_0);
+ val |= MVNETA_GMAC0_PORT_ENABLE;
+ mvreg_write(pp, MVNETA_GMAC_CTRL_0, val);
+}
+
+/* Disable the port and wait for about 200 usec before retuning */
+static void mvneta_port_disable(struct mvneta_port *pp)
+{
+ u32 val;
+
+ /* Reset the Enable bit in the Serial Control Register */
+ val = mvreg_read(pp, MVNETA_GMAC_CTRL_0);
+ val &= ~MVNETA_GMAC0_PORT_ENABLE;
+ mvreg_write(pp, MVNETA_GMAC_CTRL_0, val);
+
+ udelay(200);
+}
+
+/* Multicast tables methods */
+
+/* Set all entries in Unicast MAC Table; queue==-1 means reject all */
+static void mvneta_set_ucast_table(struct mvneta_port *pp, int queue)
+{
+ int offset;
+ u32 val;
+
+ if (queue == -1) {
+ val = 0;
+ } else {
+ val = 0x1 | (queue << 1);
+ val |= (val << 24) | (val << 16) | (val << 8);
+ }
+
+ for (offset = 0; offset <= 0xc; offset += 4)
+ mvreg_write(pp, MVNETA_DA_FILT_UCAST_BASE + offset, val);
+}
+
+/* Set all entries in Special Multicast MAC Table; queue==-1 means reject all */
+static void mvneta_set_special_mcast_table(struct mvneta_port *pp, int queue)
+{
+ int offset;
+ u32 val;
+
+ if (queue == -1) {
+ val = 0;
+ } else {
+ val = 0x1 | (queue << 1);
+ val |= (val << 24) | (val << 16) | (val << 8);
+ }
+
+ for (offset = 0; offset <= 0xfc; offset += 4)
+ mvreg_write(pp, MVNETA_DA_FILT_SPEC_MCAST + offset, val);
+
+}
+
+/* Set all entries in Other Multicast MAC Table. queue==-1 means reject all */
+static void mvneta_set_other_mcast_table(struct mvneta_port *pp, int queue)
+{
+ int offset;
+ u32 val;
+
+ if (queue == -1) {
+ memset(pp->mcast_count, 0, sizeof(pp->mcast_count));
+ val = 0;
+ } else {
+ memset(pp->mcast_count, 1, sizeof(pp->mcast_count));
+ val = 0x1 | (queue << 1);
+ val |= (val << 24) | (val << 16) | (val << 8);
+ }
+
+ for (offset = 0; offset <= 0xfc; offset += 4)
+ mvreg_write(pp, MVNETA_DA_FILT_OTH_MCAST + offset, val);
+}
+
+/* This method sets defaults to the NETA port:
+ * Clears interrupt Cause and Mask registers.
+ * Clears all MAC tables.
+ * Sets defaults to all registers.
+ * Resets RX and TX descriptor rings.
+ * Resets PHY.
+ * This method can be called after mvneta_port_down() to return the port
+ * settings to defaults.
+ */
+static void mvneta_defaults_set(struct mvneta_port *pp)
+{
+ int cpu;
+ int queue;
+ u32 val;
+
+ /* Clear all Cause registers */
+ mvreg_write(pp, MVNETA_INTR_NEW_CAUSE, 0);
+ mvreg_write(pp, MVNETA_INTR_OLD_CAUSE, 0);
+ mvreg_write(pp, MVNETA_INTR_MISC_CAUSE, 0);
+
+ /* Mask all interrupts */
+ mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
+ mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
+ mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
+ mvreg_write(pp, MVNETA_INTR_ENABLE, 0);
+
+ /* Enable MBUS Retry bit16 */
+ mvreg_write(pp, MVNETA_MBUS_RETRY, 0x20);
+
+ /* Set CPU queue access map - all CPUs have access to all RX
+ queues and to all TX queues */
+ for (cpu = 0; cpu < CONFIG_NR_CPUS; cpu++)
+ mvreg_write(pp, MVNETA_CPU_MAP(cpu),
+ (MVNETA_CPU_RXQ_ACCESS_ALL_MASK |
+ MVNETA_CPU_TXQ_ACCESS_ALL_MASK));
+
+ /* Reset RX and TX DMAs */
+ mvreg_write(pp, MVNETA_PORT_RX_RESET, MVNETA_PORT_RX_DMA_RESET);
+ mvreg_write(pp, MVNETA_PORT_TX_RESET, MVNETA_PORT_TX_DMA_RESET);
+
+ /* Disable Legacy WRR, Disable EJP, Release from reset */
+ mvreg_write(pp, MVNETA_TXQ_CMD_1, 0);
+ for (queue = 0; queue < txq_number; queue++) {
+ mvreg_write(pp, MVETH_TXQ_TOKEN_COUNT_REG(queue), 0);
+ mvreg_write(pp, MVETH_TXQ_TOKEN_CFG_REG(queue), 0);
+ }
+
+ mvreg_write(pp, MVNETA_PORT_TX_RESET, 0);
+ mvreg_write(pp, MVNETA_PORT_RX_RESET, 0);
+
+ /* Set Port Acceleration Mode */
+ val = MVNETA_ACC_MODE_EXT;
+ mvreg_write(pp, MVNETA_ACC_MODE, val);
+
+ /* Update val of portCfg register accordingly with all RxQueue types */
+ val = MVNETA_PORT_CONFIG_DEFL_VALUE(rxq_def);
+ mvreg_write(pp, MVNETA_PORT_CONFIG, val);
+
+ val = 0;
+ mvreg_write(pp, MVNETA_PORT_CONFIG_EXTEND, val);
+ mvreg_write(pp, MVNETA_RX_MIN_FRAME_SIZE, 64);
+
+ /* Build PORT_SDMA_CONFIG_REG */
+ val = 0;
+
+ /* Default burst size */
+ val |= MVNETA_TX_BRST_SZ_MASK(MVNETA_SDMA_BRST_SIZE_16);
+ val |= MVNETA_RX_BRST_SZ_MASK(MVNETA_SDMA_BRST_SIZE_16);
+
+ val |= (MVNETA_RX_NO_DATA_SWAP | MVNETA_TX_NO_DATA_SWAP |
+ MVNETA_NO_DESC_SWAP);
+
+ /* Assign port SDMA configuration */
+ mvreg_write(pp, MVNETA_SDMA_CONFIG, val);
+
+ mvneta_set_ucast_table(pp, -1);
+ mvneta_set_special_mcast_table(pp, -1);
+ mvneta_set_other_mcast_table(pp, -1);
+
+ /* Set port interrupt enable register - default enable all */
+ mvreg_write(pp, MVNETA_INTR_ENABLE,
+ (MVNETA_RXQ_INTR_ENABLE_ALL_MASK
+ | MVNETA_TXQ_INTR_ENABLE_ALL_MASK));
+}
+
+/* Set max sizes for tx queues */
+static void mvneta_txq_max_tx_size_set(struct mvneta_port *pp, int max_tx_size)
+
+{
+ u32 val, size, mtu;
+ int queue;
+
+ mtu = max_tx_size * 8;
+ if (mtu > MVNETA_TX_MTU_MAX)
+ mtu = MVNETA_TX_MTU_MAX;
+
+ /* Set MTU */
+ val = mvreg_read(pp, MVNETA_TX_MTU);
+ val &= ~MVNETA_TX_MTU_MAX;
+ val |= mtu;
+ mvreg_write(pp, MVNETA_TX_MTU, val);
+
+ /* TX token size and all TXQs token size must be larger that MTU */
+ val = mvreg_read(pp, MVNETA_TX_TOKEN_SIZE);
+
+ size = val & MVNETA_TX_TOKEN_SIZE_MAX;
+ if (size < mtu) {
+ size = mtu;
+ val &= ~MVNETA_TX_TOKEN_SIZE_MAX;
+ val |= size;
+ mvreg_write(pp, MVNETA_TX_TOKEN_SIZE, val);
+ }
+ for (queue = 0; queue < txq_number; queue++) {
+ val = mvreg_read(pp, MVNETA_TXQ_TOKEN_SIZE_REG(queue));
+
+ size = val & MVNETA_TXQ_TOKEN_SIZE_MAX;
+ if (size < mtu) {
+ size = mtu;
+ val &= ~MVNETA_TXQ_TOKEN_SIZE_MAX;
+ val |= size;
+ mvreg_write(pp, MVNETA_TXQ_TOKEN_SIZE_REG(queue), val);
+ }
+ }
+}
+
+/* Set unicast address */
+static void mvneta_set_ucast_addr(struct mvneta_port *pp, u8 last_nibble,
+ int queue)
+{
+ unsigned int unicast_reg;
+ unsigned int tbl_offset;
+ unsigned int reg_offset;
+
+ /* Locate the Unicast table entry */
+ last_nibble = (0xf & last_nibble);
+
+ /* offset from unicast tbl base */
+ tbl_offset = (last_nibble / 4) * 4;
+
+ /* offset within the above reg */
+ reg_offset = last_nibble % 4;
+
+ unicast_reg = mvreg_read(pp, (MVNETA_DA_FILT_UCAST_BASE + tbl_offset));
+
+ if (queue == -1) {
+ /* Clear accepts frame bit at specified unicast DA tbl entry */
+ unicast_reg &= ~(0xff << (8 * reg_offset));
+ } else {
+ unicast_reg &= ~(0xff << (8 * reg_offset));
+ unicast_reg |= ((0x01 | (queue << 1)) << (8 * reg_offset));
+ }
+
+ mvreg_write(pp, (MVNETA_DA_FILT_UCAST_BASE + tbl_offset), unicast_reg);
+}
+
+/* Set mac address */
+static void mvneta_mac_addr_set(struct mvneta_port *pp, unsigned char *addr,
+ int queue)
+{
+ unsigned int mac_h;
+ unsigned int mac_l;
+
+ if (queue != -1) {
+ mac_l = (addr[4] << 8) | (addr[5]);
+ mac_h = (addr[0] << 24) | (addr[1] << 16) |
+ (addr[2] << 8) | (addr[3] << 0);
+
+ mvreg_write(pp, MVNETA_MAC_ADDR_LOW, mac_l);
+ mvreg_write(pp, MVNETA_MAC_ADDR_HIGH, mac_h);
+ }
+
+ /* Accept frames of this address */
+ mvneta_set_ucast_addr(pp, addr[5], queue);
+}
+
+/*
+ * Set the number of packets that will be received before
+ * RX interrupt will be generated by HW.
+ */
+static void mvneta_rx_pkts_coal_set(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq, u32 value)
+{
+ mvreg_write(pp, MVNETA_RXQ_THRESHOLD_REG(rxq->id),
+ value | MVNETA_RXQ_NON_OCCUPIED(0));
+ rxq->pkts_coal = value;
+}
+
+/*
+ * Set the time delay in usec before
+ * RX interrupt will be generated by HW.
+ */
+static void mvneta_rx_time_coal_set(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq, u32 value)
+{
+ u32 val = (pp->clk_rate_hz / 1000000) * value;
+
+ mvreg_write(pp, MVNETA_RXQ_TIME_COAL_REG(rxq->id), val);
+ rxq->time_coal = value;
+}
+
+/* Set threshold for TX_DONE pkts coalescing */
+static void mvneta_tx_done_pkts_coal_set(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq, u32 value)
+{
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_TXQ_SIZE_REG(txq->id));
+
+ val &= ~MVNETA_TXQ_SENT_THRESH_ALL_MASK;
+ val |= MVNETA_TXQ_SENT_THRESH_MASK(value);
+
+ mvreg_write(pp, MVNETA_TXQ_SIZE_REG(txq->id), val);
+
+ txq->done_pkts_coal = value;
+}
+
+/* Trigger tx done timer in MVNETA_TX_DONE_TIMER_PERIOD msecs */
+static void mvneta_add_tx_done_timer(struct mvneta_port *pp)
+{
+ if (test_and_set_bit(MVNETA_F_TX_DONE_TIMER_BIT, &pp->flags) == 0) {
+ pp->tx_done_timer.expires = jiffies +
+ msecs_to_jiffies(MVNETA_TX_DONE_TIMER_PERIOD);
+ add_timer(&pp->tx_done_timer);
+ }
+}
+
+
+/* Handle rx descriptor fill by setting buf_cookie and buf_phys_addr */
+static void mvneta_rx_desc_fill(struct mvneta_rx_desc *rx_desc,
+ u32 phys_addr, u32 cookie)
+{
+ rx_desc->buf_cookie = cookie;
+ rx_desc->buf_phys_addr = phys_addr;
+}
+
+/* Decrement sent descriptors counter */
+static void mvneta_txq_sent_desc_dec(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq,
+ int sent_desc)
+{
+ u32 val;
+
+ /* Only 255 TX descriptors can be updated at once */
+ while (sent_desc > 0xff) {
+ val = 0xff << MVNETA_TXQ_DEC_SENT_SHIFT;
+ mvreg_write(pp, MVNETA_TXQ_UPDATE_REG(txq->id), val);
+ sent_desc = sent_desc - 0xff;
+ }
+
+ val = sent_desc << MVNETA_TXQ_DEC_SENT_SHIFT;
+ mvreg_write(pp, MVNETA_TXQ_UPDATE_REG(txq->id), val);
+}
+
+/* Get number of TX descriptors already sent by HW */
+static int mvneta_txq_sent_desc_num_get(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq)
+{
+ u32 val;
+ int sent_desc;
+
+ val = mvreg_read(pp, MVNETA_TXQ_STATUS_REG(txq->id));
+ sent_desc = (val & MVNETA_TXQ_SENT_DESC_MASK) >>
+ MVNETA_TXQ_SENT_DESC_SHIFT;
+
+ return sent_desc;
+}
+
+/*
+ * Get number of sent descriptors and decrement counter.
+ * The number of sent descriptors is returned.
+ */
+static int mvneta_txq_sent_desc_proc(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq)
+{
+ int sent_desc;
+
+ /* Get number of sent descriptors */
+ sent_desc = mvneta_txq_sent_desc_num_get(pp, txq);
+
+ /* Decrement sent descriptors counter */
+ if (sent_desc)
+ mvneta_txq_sent_desc_dec(pp, txq, sent_desc);
+
+ return sent_desc;
+}
+
+/* Set TXQ descriptors fields relevant for CSUM calculation */
+static u32 mvneta_txq_desc_csum(int l3_offs, int l3_proto,
+ int ip_hdr_len, int l4_proto)
+{
+ u32 command;
+
+ /* Fields: L3_offset, IP_hdrlen, L3_type, G_IPv4_chk,
+ G_L4_chk, L4_type; required only for checksum
+ calculation */
+ command = l3_offs << MVNETA_TX_L3_OFF_SHIFT;
+ command |= ip_hdr_len << MVNETA_TX_IP_HLEN_SHIFT;
+
+ if (l3_proto == swab16(ETH_P_IP))
+ command |= MVNETA_TXD_IP_CSUM;
+ else
+ command |= MVNETA_TX_L3_IP6;
+
+ if (l4_proto == IPPROTO_TCP)
+ command |= MVNETA_TX_L4_CSUM_FULL;
+ else if (l4_proto == IPPROTO_UDP)
+ command |= MVNETA_TX_L4_UDP | MVNETA_TX_L4_CSUM_FULL;
+ else
+ command |= MVNETA_TX_L4_CSUM_NOT;
+
+ return command;
+}
+
+
+/* Display more error info */
+static void mvneta_rx_error(struct mvneta_port *pp,
+ struct mvneta_rx_desc *rx_desc)
+{
+ u32 status = rx_desc->status;
+
+ if (!mvneta_rxq_desc_is_first_last(rx_desc)) {
+ netdev_err(pp->dev,
+ "bad rx status %08x (buffer oversize), size=%d\n",
+ rx_desc->status, rx_desc->data_size);
+ return;
+ }
+
+ switch (status & MVNETA_RXD_ERR_CODE_MASK) {
+ case MVNETA_RXD_ERR_CRC:
+ netdev_err(pp->dev, "bad rx status %08x (crc error), size=%d\n",
+ status, rx_desc->data_size);
+ break;
+ case MVNETA_RXD_ERR_OVERRUN:
+ netdev_err(pp->dev, "bad rx status %08x (overrun error), size=%d\n",
+ status, rx_desc->data_size);
+ break;
+ case MVNETA_RXD_ERR_LEN:
+ netdev_err(pp->dev, "bad rx status %08x (max frame length error), size=%d\n",
+ status, rx_desc->data_size);
+ break;
+ case MVNETA_RXD_ERR_RESOURCE:
+ netdev_err(pp->dev, "bad rx status %08x (resource error), size=%d\n",
+ status, rx_desc->data_size);
+ break;
+ }
+}
+
+/* Handle RX checksum offload */
+static void mvneta_rx_csum(struct mvneta_port *pp,
+ struct mvneta_rx_desc *rx_desc,
+ struct sk_buff *skb)
+{
+ if ((rx_desc->status & MVNETA_RXD_L3_IP4) &&
+ (rx_desc->status & MVNETA_RXD_L4_CSUM_OK)) {
+ skb->csum = 0;
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+ return;
+ }
+
+ skb->ip_summed = CHECKSUM_NONE;
+}
+
+/* Return tx queue pointer (find last set bit) according to causeTxDone reg */
+static struct mvneta_tx_queue *mvneta_tx_done_policy(struct mvneta_port *pp,
+ u32 cause)
+{
+ int queue = fls(cause) - 1;
+
+ return (queue < 0 || queue >= txq_number) ? NULL : &pp->txqs[queue];
+}
+
+/* Free tx queue skbuffs */
+static void mvneta_txq_bufs_free(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq, int num)
+{
+ int i;
+
+ for (i = 0; i < num; i++) {
+ struct mvneta_tx_desc *tx_desc = txq->descs +
+ txq->txq_get_index;
+ struct sk_buff *skb = txq->tx_skb[txq->txq_get_index];
+
+ mvneta_txq_inc_get(txq);
+
+ if (!skb)
+ continue;
+
+ dma_unmap_single(pp->dev->dev.parent, tx_desc->buf_phys_addr,
+ tx_desc->data_size, DMA_TO_DEVICE);
+ dev_kfree_skb_any(skb);
+ }
+}
+
+/* Handle end of transmission */
+static int mvneta_txq_done(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq)
+{
+ struct netdev_queue *nq = netdev_get_tx_queue(pp->dev, txq->id);
+ int tx_done;
+
+ tx_done = mvneta_txq_sent_desc_proc(pp, txq);
+ if (tx_done == 0)
+ return tx_done;
+ mvneta_txq_bufs_free(pp, txq, tx_done);
+
+ txq->count -= tx_done;
+
+ if (netif_tx_queue_stopped(nq)) {
+ if (txq->size - txq->count >= MAX_SKB_FRAGS + 1)
+ netif_tx_wake_queue(nq);
+ }
+
+ return tx_done;
+}
+
+/* Refill processing */
+static int mvneta_rx_refill(struct mvneta_port *pp,
+ struct mvneta_rx_desc *rx_desc)
+
+{
+ dma_addr_t phys_addr;
+ struct sk_buff *skb;
+
+ skb = netdev_alloc_skb(pp->dev, pp->pkt_size);
+ if (!skb)
+ return -ENOMEM;
+
+ phys_addr = dma_map_single(pp->dev->dev.parent, skb->head,
+ MVNETA_RX_BUF_SIZE(pp->pkt_size),
+ DMA_FROM_DEVICE);
+ if (unlikely(dma_mapping_error(pp->dev->dev.parent, phys_addr))) {
+ dev_kfree_skb(skb);
+ return -ENOMEM;
+ }
+
+ mvneta_rx_desc_fill(rx_desc, phys_addr, (u32)skb);
+
+ return 0;
+}
+
+/* Handle tx checksum */
+static u32 mvneta_skb_tx_csum(struct mvneta_port *pp, struct sk_buff *skb)
+{
+ if (skb->ip_summed == CHECKSUM_PARTIAL) {
+ int ip_hdr_len = 0;
+ u8 l4_proto;
+
+ if (skb->protocol == htons(ETH_P_IP)) {
+ struct iphdr *ip4h = ip_hdr(skb);
+
+ /* Calculate IPv4 checksum and L4 checksum */
+ ip_hdr_len = ip4h->ihl;
+ l4_proto = ip4h->protocol;
+ } else if (skb->protocol == htons(ETH_P_IPV6)) {
+ struct ipv6hdr *ip6h = ipv6_hdr(skb);
+
+ /* Read l4_protocol from one of IPv6 extra headers */
+ if (skb_network_header_len(skb) > 0)
+ ip_hdr_len = (skb_network_header_len(skb) >> 2);
+ l4_proto = ip6h->nexthdr;
+ } else
+ return MVNETA_TX_L4_CSUM_NOT;
+
+ return mvneta_txq_desc_csum(skb_network_offset(skb),
+ skb->protocol, ip_hdr_len, l4_proto);
+ }
+
+ return MVNETA_TX_L4_CSUM_NOT;
+}
+
+/*
+ * Returns rx queue pointer (find last set bit) according to causeRxTx
+ * value
+ */
+static struct mvneta_rx_queue *mvneta_rx_policy(struct mvneta_port *pp,
+ u32 cause)
+{
+ int queue = fls(cause >> 8) - 1;
+
+ return (queue < 0 || queue >= rxq_number) ? NULL : &pp->rxqs[queue];
+}
+
+/* Drop packets received by the RXQ and free buffers */
+static void mvneta_rxq_drop_pkts(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq)
+{
+ int rx_done, i;
+
+ rx_done = mvneta_rxq_busy_desc_num_get(pp, rxq);
+ for (i = 0; i < rxq->size; i++) {
+ struct mvneta_rx_desc *rx_desc = rxq->descs + i;
+ struct sk_buff *skb = (struct sk_buff *)rx_desc->buf_cookie;
+
+ dev_kfree_skb_any(skb);
+ dma_unmap_single(pp->dev->dev.parent, rx_desc->buf_phys_addr,
+ rx_desc->data_size, DMA_FROM_DEVICE);
+ }
+
+ if (rx_done)
+ mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_done);
+}
+
+/* Main rx processing */
+static int mvneta_rx(struct mvneta_port *pp, int rx_todo,
+ struct mvneta_rx_queue *rxq)
+{
+ struct net_device *dev = pp->dev;
+ int rx_done, rx_filled;
+
+ /* Get number of received packets */
+ rx_done = mvneta_rxq_busy_desc_num_get(pp, rxq);
+
+ if (rx_todo > rx_done)
+ rx_todo = rx_done;
+
+ rx_done = 0;
+ rx_filled = 0;
+
+ /* Fairness NAPI loop */
+ while (rx_done < rx_todo) {
+ struct mvneta_rx_desc *rx_desc = mvneta_rxq_next_desc_get(rxq);
+ struct sk_buff *skb;
+ u32 rx_status;
+ int rx_bytes, err;
+
+ prefetch(rx_desc);
+ rx_done++;
+ rx_filled++;
+ rx_status = rx_desc->status;
+ skb = (struct sk_buff *)rx_desc->buf_cookie;
+
+ if (!mvneta_rxq_desc_is_first_last(rx_desc) ||
+ (rx_status & MVNETA_RXD_ERR_SUMMARY)) {
+ dev->stats.rx_errors++;
+ mvneta_rx_error(pp, rx_desc);
+ mvneta_rx_desc_fill(rx_desc, rx_desc->buf_phys_addr,
+ (u32)skb);
+ continue;
+ }
+
+ dma_unmap_single(pp->dev->dev.parent, rx_desc->buf_phys_addr,
+ rx_desc->data_size, DMA_FROM_DEVICE);
+
+ rx_bytes = rx_desc->data_size -
+ (ETH_FCS_LEN + MVNETA_MH_SIZE);
+ u64_stats_update_begin(&pp->rx_stats.syncp);
+ pp->rx_stats.packets++;
+ pp->rx_stats.bytes += rx_bytes;
+ u64_stats_update_end(&pp->rx_stats.syncp);
+
+ /* Linux processing */
+ skb_reserve(skb, MVNETA_MH_SIZE);
+ skb_put(skb, rx_bytes);
+
+ skb->protocol = eth_type_trans(skb, dev);
+
+ mvneta_rx_csum(pp, rx_desc, skb);
+
+ napi_gro_receive(&pp->napi, skb);
+
+ /* Refill processing */
+ err = mvneta_rx_refill(pp, rx_desc);
+ if (err) {
+ netdev_err(pp->dev, "Linux processing - Can't refill\n");
+ rxq->missed++;
+ rx_filled--;
+ }
+ }
+
+ /* Update rxq management counters */
+ mvneta_rxq_desc_num_update(pp, rxq, rx_done, rx_filled);
+
+ return rx_done;
+}
+
+/* Handle tx fragmentation processing */
+static int mvneta_tx_frag_process(struct mvneta_port *pp, struct sk_buff *skb,
+ struct mvneta_tx_queue *txq)
+{
+ struct mvneta_tx_desc *tx_desc;
+ int i;
+
+ for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+ skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+ void *addr = page_address(frag->page.p) + frag->page_offset;
+
+ tx_desc = mvneta_txq_next_desc_get(txq);
+ tx_desc->data_size = frag->size;
+
+ tx_desc->buf_phys_addr =
+ dma_map_single(pp->dev->dev.parent, addr,
+ tx_desc->data_size, DMA_TO_DEVICE);
+
+ if (dma_mapping_error(pp->dev->dev.parent,
+ tx_desc->buf_phys_addr)) {
+ mvneta_txq_desc_put(txq);
+ goto error;
+ }
+
+ if (i == (skb_shinfo(skb)->nr_frags - 1)) {
+ /* Last descriptor */
+ tx_desc->command = MVNETA_TXD_L_DESC | MVNETA_TXD_Z_PAD;
+
+ txq->tx_skb[txq->txq_put_index] = skb;
+
+ mvneta_txq_inc_put(txq);
+ } else {
+ /* Descriptor in the middle: Not First, Not Last */
+ tx_desc->command = 0;
+
+ txq->tx_skb[txq->txq_put_index] = NULL;
+ mvneta_txq_inc_put(txq);
+ }
+ }
+
+ return 0;
+
+error:
+ /* Release all descriptors that were used to map fragments of
+ * this packet, as well as the corresponding DMA mappings */
+ for (i = i - 1; i >= 0; i--) {
+ tx_desc = txq->descs + i;
+ dma_unmap_single(pp->dev->dev.parent,
+ tx_desc->buf_phys_addr,
+ tx_desc->data_size,
+ DMA_TO_DEVICE);
+ mvneta_txq_desc_put(txq);
+ }
+
+ return -ENOMEM;
+}
+
+/* Main tx processing */
+static int mvneta_tx(struct sk_buff *skb, struct net_device *dev)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ struct mvneta_tx_queue *txq = &pp->txqs[txq_def];
+ struct mvneta_tx_desc *tx_desc;
+ struct netdev_queue *nq;
+ int frags = 0;
+ u32 tx_cmd;
+
+ if (!netif_running(dev))
+ goto out;
+
+ frags = skb_shinfo(skb)->nr_frags + 1;
+ nq = netdev_get_tx_queue(dev, txq_def);
+
+ /* Get a descriptor for the first part of the packet */
+ tx_desc = mvneta_txq_next_desc_get(txq);
+
+ tx_cmd = mvneta_skb_tx_csum(pp, skb);
+
+ tx_desc->data_size = skb_headlen(skb);
+
+ tx_desc->buf_phys_addr = dma_map_single(dev->dev.parent, skb->data,
+ tx_desc->data_size,
+ DMA_TO_DEVICE);
+ if (unlikely(dma_mapping_error(dev->dev.parent,
+ tx_desc->buf_phys_addr))) {
+ mvneta_txq_desc_put(txq);
+ frags = 0;
+ goto out;
+ }
+
+ if (frags == 1) {
+ /* First and Last descriptor */
+ tx_cmd |= MVNETA_TXD_FLZ_DESC;
+ tx_desc->command = tx_cmd;
+ txq->tx_skb[txq->txq_put_index] = skb;
+ mvneta_txq_inc_put(txq);
+ } else {
+ /* First but not Last */
+ tx_cmd |= MVNETA_TXD_F_DESC;
+ txq->tx_skb[txq->txq_put_index] = NULL;
+ mvneta_txq_inc_put(txq);
+ tx_desc->command = tx_cmd;
+ /* Continue with other skb fragments */
+ if (mvneta_tx_frag_process(pp, skb, txq)) {
+ dma_unmap_single(dev->dev.parent,
+ tx_desc->buf_phys_addr,
+ tx_desc->data_size,
+ DMA_TO_DEVICE);
+ mvneta_txq_desc_put(txq);
+ frags = 0;
+ goto out;
+ }
+ }
+
+ txq->count += frags;
+ mvneta_txq_pend_desc_add(pp, txq, frags);
+
+ if (txq->size - txq->count < MAX_SKB_FRAGS + 1)
+ netif_tx_stop_queue(nq);
+
+out:
+ if (frags > 0) {
+ u64_stats_update_begin(&pp->tx_stats.syncp);
+ pp->tx_stats.packets++;
+ pp->tx_stats.bytes += skb->len;
+ u64_stats_update_end(&pp->tx_stats.syncp);
+
+ } else {
+ dev->stats.tx_dropped++;
+ dev_kfree_skb_any(skb);
+ }
+
+ if (txq->count >= MVNETA_TXDONE_COAL_PKTS)
+ mvneta_txq_done(pp, txq);
+
+ /* If after calling mvneta_txq_done, count equals
+ frags, we need to set the timer */
+ if (txq->count == frags && frags > 0)
+ mvneta_add_tx_done_timer(pp);
+
+ return NETDEV_TX_OK;
+}
+
+
+/* Free tx resources, when resetting a port */
+static void mvneta_txq_done_force(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq)
+
+{
+ int tx_done = txq->count;
+
+ mvneta_txq_bufs_free(pp, txq, tx_done);
+
+ /* reset txq */
+ txq->count = 0;
+ txq->txq_put_index = 0;
+ txq->txq_get_index = 0;
+}
+
+/* handle tx done - called from tx done timer callback */
+static u32 mvneta_tx_done_gbe(struct mvneta_port *pp, u32 cause_tx_done,
+ int *tx_todo)
+{
+ struct mvneta_tx_queue *txq;
+ u32 tx_done = 0;
+ struct netdev_queue *nq;
+
+ *tx_todo = 0;
+ while (cause_tx_done != 0) {
+ txq = mvneta_tx_done_policy(pp, cause_tx_done);
+ if (!txq)
+ break;
+
+ nq = netdev_get_tx_queue(pp->dev, txq->id);
+ __netif_tx_lock(nq, smp_processor_id());
+
+ if (txq->count) {
+ tx_done += mvneta_txq_done(pp, txq);
+ *tx_todo += txq->count;
+ }
+
+ __netif_tx_unlock(nq);
+ cause_tx_done &= ~((1 << txq->id));
+ }
+
+ return tx_done;
+}
+
+/*
+ * Compute crc8 of the specified address, using a unique algorithm ,
+ * according to hw spec, different than generic crc8 algorithm
+ */
+static int mvneta_addr_crc(unsigned char *addr)
+{
+ int crc = 0;
+ int i;
+
+ for (i = 0; i < ETH_ALEN; i++) {
+ int j;
+
+ crc = (crc ^ addr[i]) << 8;
+ for (j = 7; j >= 0; j--) {
+ if (crc & (0x100 << j))
+ crc ^= 0x107 << j;
+ }
+ }
+
+ return crc;
+}
+
+/* This method controls the net device special MAC multicast support.
+ * The Special Multicast Table for MAC addresses supports MAC of the form
+ * 0x01-00-5E-00-00-XX (where XX is between 0x00 and 0xFF).
+ * The MAC DA[7:0] bits are used as a pointer to the Special Multicast
+ * Table entries in the DA-Filter table. This method set the Special
+ * Multicast Table appropriate entry.
+ */
+static void mvneta_set_special_mcast_addr(struct mvneta_port *pp,
+ unsigned char last_byte,
+ int queue)
+{
+ unsigned int smc_table_reg;
+ unsigned int tbl_offset;
+ unsigned int reg_offset;
+
+ /* Register offset from SMC table base */
+ tbl_offset = (last_byte / 4);
+ /* Entry offset within the above reg */
+ reg_offset = last_byte % 4;
+
+ smc_table_reg = mvreg_read(pp, (MVNETA_DA_FILT_SPEC_MCAST
+ + tbl_offset * 4));
+
+ if (queue == -1)
+ smc_table_reg &= ~(0xff << (8 * reg_offset));
+ else {
+ smc_table_reg &= ~(0xff << (8 * reg_offset));
+ smc_table_reg |= ((0x01 | (queue << 1)) << (8 * reg_offset));
+ }
+
+ mvreg_write(pp, MVNETA_DA_FILT_SPEC_MCAST + tbl_offset * 4,
+ smc_table_reg);
+}
+
+/* This method controls the network device Other MAC multicast support.
+ * The Other Multicast Table is used for multicast of another type.
+ * A CRC-8 is used as an index to the Other Multicast Table entries
+ * in the DA-Filter table.
+ * The method gets the CRC-8 value from the calling routine and
+ * sets the Other Multicast Table appropriate entry according to the
+ * specified CRC-8 .
+ */
+static void mvneta_set_other_mcast_addr(struct mvneta_port *pp,
+ unsigned char crc8,
+ int queue)
+{
+ unsigned int omc_table_reg;
+ unsigned int tbl_offset;
+ unsigned int reg_offset;
+
+ tbl_offset = (crc8 / 4) * 4; /* Register offset from OMC table base */
+ reg_offset = crc8 % 4; /* Entry offset within the above reg */
+
+ omc_table_reg = mvreg_read(pp, MVNETA_DA_FILT_OTH_MCAST + tbl_offset);
+
+ if (queue == -1) {
+ /* Clear accepts frame bit at specified Other DA table entry */
+ omc_table_reg &= ~(0xff << (8 * reg_offset));
+ } else {
+ omc_table_reg &= ~(0xff << (8 * reg_offset));
+ omc_table_reg |= ((0x01 | (queue << 1)) << (8 * reg_offset));
+ }
+
+ mvreg_write(pp, MVNETA_DA_FILT_OTH_MCAST + tbl_offset, omc_table_reg);
+}
+
+/* The network device supports multicast using two tables:
+ * 1) Special Multicast Table for MAC addresses of the form
+ * 0x01-00-5E-00-00-XX (where XX is between 0x00 and 0xFF).
+ * The MAC DA[7:0] bits are used as a pointer to the Special Multicast
+ * Table entries in the DA-Filter table.
+ * 2) Other Multicast Table for multicast of another type. A CRC-8 value
+ * is used as an index to the Other Multicast Table entries in the
+ * DA-Filter table.
+ */
+static int mvneta_mcast_addr_set(struct mvneta_port *pp, unsigned char *p_addr,
+ int queue)
+{
+ unsigned char crc_result = 0;
+
+ if (memcmp(p_addr, "\x01\x00\x5e\x00\x00", 5) == 0) {
+ mvneta_set_special_mcast_addr(pp, p_addr[5], queue);
+ return 0;
+ }
+
+ crc_result = mvneta_addr_crc(p_addr);
+ if (queue == -1) {
+ if (pp->mcast_count[crc_result] == 0) {
+ netdev_info(pp->dev, "No valid Mcast for crc8=0x%02x\n",
+ crc_result);
+ return -EINVAL;
+ }
+
+ pp->mcast_count[crc_result]--;
+ if (pp->mcast_count[crc_result] != 0) {
+ netdev_info(pp->dev,
+ "After delete there are %d valid Mcast for crc8=0x%02x\n",
+ pp->mcast_count[crc_result], crc_result);
+ return -EINVAL;
+ }
+ } else
+ pp->mcast_count[crc_result]++;
+
+ mvneta_set_other_mcast_addr(pp, crc_result, queue);
+
+ return 0;
+}
+
+/* Configure Fitering mode of Ethernet port */
+static void mvneta_rx_unicast_promisc_set(struct mvneta_port *pp,
+ int is_promisc)
+{
+ u32 port_cfg_reg, val;
+
+ port_cfg_reg = mvreg_read(pp, MVNETA_PORT_CONFIG);
+
+ val = mvreg_read(pp, MVNETA_TYPE_PRIO);
+
+ /* Set / Clear UPM bit in port configuration register */
+ if (is_promisc) {
+ /* Accept all Unicast addresses */
+ port_cfg_reg |= MVNETA_UNI_PROMISC_MODE;
+ val |= MVNETA_FORCE_UNI;
+ mvreg_write(pp, MVNETA_MAC_ADDR_LOW, 0xffff);
+ mvreg_write(pp, MVNETA_MAC_ADDR_HIGH, 0xffffffff);
+ } else {
+ /* Reject all Unicast addresses */
+ port_cfg_reg &= ~MVNETA_UNI_PROMISC_MODE;
+ val &= ~MVNETA_FORCE_UNI;
+ }
+
+ mvreg_write(pp, MVNETA_PORT_CONFIG, port_cfg_reg);
+ mvreg_write(pp, MVNETA_TYPE_PRIO, val);
+}
+
+/* register unicast and multicast addresses */
+static void mvneta_set_rx_mode(struct net_device *dev)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ struct netdev_hw_addr *ha;
+
+ if (dev->flags & IFF_PROMISC) {
+ /* Accept all: Multicast + Unicast */
+ mvneta_rx_unicast_promisc_set(pp, 1);
+ mvneta_set_ucast_table(pp, rxq_def);
+ mvneta_set_special_mcast_table(pp, rxq_def);
+ mvneta_set_other_mcast_table(pp, rxq_def);
+ } else {
+ /* Accept single Unicast */
+ mvneta_rx_unicast_promisc_set(pp, 0);
+ mvneta_set_ucast_table(pp, -1);
+ mvneta_mac_addr_set(pp, dev->dev_addr, rxq_def);
+
+ if (dev->flags & IFF_ALLMULTI) {
+ /* Accept all multicast */
+ mvneta_set_special_mcast_table(pp, rxq_def);
+ mvneta_set_other_mcast_table(pp, rxq_def);
+ } else {
+ /* Accept only initialized multicast */
+ mvneta_set_special_mcast_table(pp, -1);
+ mvneta_set_other_mcast_table(pp, -1);
+
+ if (!netdev_mc_empty(dev)) {
+ netdev_for_each_mc_addr(ha, dev) {
+ mvneta_mcast_addr_set(pp, ha->addr,
+ rxq_def);
+ }
+ }
+ }
+ }
+}
+
+/* Interrupt handling - the callback for request_irq() */
+static irqreturn_t mvneta_isr(int irq, void *dev_id)
+{
+ struct mvneta_port *pp = (struct mvneta_port *)dev_id;
+
+ /* Mask all interrupts */
+ mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
+
+ napi_schedule(&pp->napi);
+
+ return IRQ_HANDLED;
+}
+
+/* NAPI handler
+ * Bits 0 - 7 of the causeRxTx register indicate that are transmitted
+ * packets on the corresponding TXQ (Bit 0 is for TX queue 1).
+ * Bits 8 -15 of the cause Rx Tx register indicate that are received
+ * packets on the corresponding RXQ (Bit 8 is for RX queue 0).
+ * Each CPU has its own causeRxTx register
+ */
+static int mvneta_poll(struct napi_struct *napi, int budget)
+{
+ int rx_done = 0;
+ u32 cause_rx_tx;
+ unsigned long flags;
+ struct mvneta_port *pp = netdev_priv(napi->dev);
+
+ if (!netif_running(pp->dev)) {
+ napi_complete(napi);
+ return rx_done;
+ }
+
+ /* Read cause register */
+ cause_rx_tx = mvreg_read(pp, MVNETA_INTR_NEW_CAUSE) &
+ MVNETA_RX_INTR_MASK(rxq_number);
+
+ /*
+ * For the case where the last mvneta_poll did not process all
+ * RX packets
+ */
+ cause_rx_tx |= pp->cause_rx_tx;
+ if (rxq_number > 1) {
+ while ((cause_rx_tx != 0) && (budget > 0)) {
+ int count;
+ struct mvneta_rx_queue *rxq;
+ /* get rx queue number from cause_rx_tx */
+ rxq = mvneta_rx_policy(pp, cause_rx_tx);
+ if (!rxq)
+ break;
+
+ /* process the packet in that rx queue */
+ count = mvneta_rx(pp, budget, rxq);
+ rx_done += count;
+ budget -= count;
+ if (budget > 0) {
+ /* set off the rx bit of the corresponding bit
+ in the cause rx tx register, so that next
+ iteration will find the next rx queue where
+ packets are received on */
+ cause_rx_tx &= ~((1 << rxq->id) << 8);
+ }
+ }
+ } else {
+ rx_done = mvneta_rx(pp, budget, &pp->rxqs[rxq_def]);
+ budget -= rx_done;
+ }
+
+ if (budget > 0) {
+ cause_rx_tx = 0;
+ napi_complete(napi);
+ local_irq_save(flags);
+ mvreg_write(pp, MVNETA_INTR_NEW_MASK,
+ MVNETA_RX_INTR_MASK(rxq_number));
+ local_irq_restore(flags);
+ }
+
+ pp->cause_rx_tx = cause_rx_tx;
+ return rx_done;
+}
+
+/* tx done timer callback */
+static void mvneta_tx_done_timer_callback(unsigned long data)
+{
+ struct net_device *dev = (struct net_device *)data;
+ struct mvneta_port *pp = netdev_priv(dev);
+ int tx_done = 0, tx_todo = 0;
+
+ if (!netif_running(dev))
+ return ;
+
+ clear_bit(MVNETA_F_TX_DONE_TIMER_BIT, &pp->flags);
+
+ tx_done = mvneta_tx_done_gbe(pp,
+ (((1 << txq_number) - 1) &
+ MVNETA_CAUSE_TXQ_SENT_DESC_ALL_MASK),
+ &tx_todo);
+ if (tx_todo > 0)
+ mvneta_add_tx_done_timer(pp);
+}
+
+/* Handle rxq fill: allocates rxq skbs; called when initializing a port */
+static int mvneta_rxq_fill(struct mvneta_port *pp, struct mvneta_rx_queue *rxq,
+ int num)
+{
+ struct net_device *dev = pp->dev;
+ int i;
+
+ for (i = 0; i < num; i++) {
+ struct sk_buff *skb;
+ struct mvneta_rx_desc *rx_desc;
+ unsigned long phys_addr;
+
+ skb = dev_alloc_skb(pp->pkt_size);
+ if (!skb) {
+ netdev_err(dev, "%s:rxq %d, %d of %d buffs filled\n",
+ __func__, rxq->id, i, num);
+ break;
+ }
+
+ rx_desc = rxq->descs + i;
+ memset(rx_desc, 0, sizeof(struct mvneta_rx_desc));
+ phys_addr = dma_map_single(dev->dev.parent, skb->head,
+ MVNETA_RX_BUF_SIZE(pp->pkt_size),
+ DMA_FROM_DEVICE);
+ if (unlikely(dma_mapping_error(dev->dev.parent, phys_addr))) {
+ dev_kfree_skb(skb);
+ break;
+ }
+
+ mvneta_rx_desc_fill(rx_desc, phys_addr, (u32)skb);
+ }
+
+ /* Add this number of RX descriptors as non occupied (ready to
+ get packets) */
+ mvneta_rxq_non_occup_desc_add(pp, rxq, i);
+
+ return i;
+}
+
+/* Free all packets pending transmit from all TXQs and reset TX port */
+static void mvneta_tx_reset(struct mvneta_port *pp)
+{
+ int queue;
+
+ /* free the skb's in the hal tx ring */
+ for (queue = 0; queue < txq_number; queue++)
+ mvneta_txq_done_force(pp, &pp->txqs[queue]);
+
+ mvreg_write(pp, MVNETA_PORT_TX_RESET, MVNETA_PORT_TX_DMA_RESET);
+ mvreg_write(pp, MVNETA_PORT_TX_RESET, 0);
+}
+
+static void mvneta_rx_reset(struct mvneta_port *pp)
+{
+ mvreg_write(pp, MVNETA_PORT_RX_RESET, MVNETA_PORT_RX_DMA_RESET);
+ mvreg_write(pp, MVNETA_PORT_RX_RESET, 0);
+}
+
+/* Rx/Tx queue initialization/cleanup methods */
+
+/* Create a specified RX queue */
+static int mvneta_rxq_init(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq)
+
+{
+ rxq->size = pp->rx_ring_size;
+
+ /* Allocate memory for RX descriptors */
+ rxq->descs = dma_alloc_coherent(pp->dev->dev.parent,
+ rxq->size * MVNETA_DESC_ALIGNED_SIZE,
+ &rxq->descs_phys, GFP_KERNEL);
+ if (rxq->descs == NULL) {
+ netdev_err(pp->dev,
+ "rxq=%d: Can't allocate %d bytes for %d RX descr\n",
+ rxq->id, rxq->size * MVNETA_DESC_ALIGNED_SIZE,
+ rxq->size);
+ return -ENOMEM;
+ }
+
+ BUG_ON(rxq->descs !=
+ PTR_ALIGN(rxq->descs, MVNETA_CPU_D_CACHE_LINE_SIZE));
+
+ rxq->last_desc = rxq->size - 1;
+
+ /* Set Rx descriptors queue starting address */
+ mvreg_write(pp, MVNETA_RXQ_BASE_ADDR_REG(rxq->id), rxq->descs_phys);
+ mvreg_write(pp, MVNETA_RXQ_SIZE_REG(rxq->id), rxq->size);
+
+ /* Set Offset */
+ mvneta_rxq_offset_set(pp, rxq, NET_SKB_PAD);
+
+ /* Set coalescing pkts and time */
+ mvneta_rx_pkts_coal_set(pp, rxq, rxq->pkts_coal);
+ mvneta_rx_time_coal_set(pp, rxq, rxq->time_coal);
+
+ /* Fill RXQ with buffers from RX pool */
+ mvneta_rxq_buf_size_set(pp, rxq, MVNETA_RX_BUF_SIZE(pp->pkt_size));
+ mvneta_rxq_bm_disable(pp, rxq);
+ mvneta_rxq_fill(pp, rxq, rxq->size);
+
+ return 0;
+}
+
+/* Cleanup Rx queue */
+static void mvneta_rxq_deinit(struct mvneta_port *pp,
+ struct mvneta_rx_queue *rxq)
+{
+ mvneta_rxq_drop_pkts(pp, rxq);
+
+ if (rxq->descs)
+ dma_free_coherent(pp->dev->dev.parent,
+ rxq->size * MVNETA_DESC_ALIGNED_SIZE,
+ rxq->descs,
+ rxq->descs_phys);
+
+ rxq->descs = NULL;
+ rxq->last_desc = 0;
+ rxq->next_desc_to_proc = 0;
+ rxq->descs_phys = 0;
+}
+
+/* Create and initialize a tx queue */
+static int mvneta_txq_init(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq)
+{
+ txq->size = pp->tx_ring_size;
+
+ /* Allocate memory for TX descriptors */
+ txq->descs = dma_alloc_coherent(pp->dev->dev.parent,
+ txq->size * MVNETA_DESC_ALIGNED_SIZE,
+ &txq->descs_phys, GFP_KERNEL);
+ if (txq->descs == NULL) {
+ netdev_err(pp->dev,
+ "txQ=%d: Can't allocate %d bytes for %d TX descr\n",
+ txq->id, txq->size * MVNETA_DESC_ALIGNED_SIZE,
+ txq->size);
+ return -ENOMEM;
+ }
+
+ /* Make sure descriptor address is cache line size aligned */
+ BUG_ON(txq->descs !=
+ PTR_ALIGN(txq->descs, MVNETA_CPU_D_CACHE_LINE_SIZE));
+
+ txq->last_desc = txq->size - 1;
+
+ /* Set maximum bandwidth for enabled TXQs */
+ mvreg_write(pp, MVETH_TXQ_TOKEN_CFG_REG(txq->id), 0x03ffffff);
+ mvreg_write(pp, MVETH_TXQ_TOKEN_COUNT_REG(txq->id), 0x3fffffff);
+
+ /* Set Tx descriptors queue starting address */
+ mvreg_write(pp, MVNETA_TXQ_BASE_ADDR_REG(txq->id), txq->descs_phys);
+ mvreg_write(pp, MVNETA_TXQ_SIZE_REG(txq->id), txq->size);
+
+ txq->tx_skb = kmalloc(txq->size * sizeof(*txq->tx_skb), GFP_KERNEL);
+ if (txq->tx_skb == NULL) {
+ dma_free_coherent(pp->dev->dev.parent,
+ txq->size * MVNETA_DESC_ALIGNED_SIZE,
+ txq->descs, txq->descs_phys);
+ return -ENOMEM;
+ }
+ mvneta_tx_done_pkts_coal_set(pp, txq, txq->done_pkts_coal);
+
+ return 0;
+}
+
+/* Free allocated resources when mvneta_txq_init() fails to allocate memory*/
+static void mvneta_txq_deinit(struct mvneta_port *pp,
+ struct mvneta_tx_queue *txq)
+{
+ kfree(txq->tx_skb);
+
+ if (txq->descs)
+ dma_free_coherent(pp->dev->dev.parent,
+ txq->size * MVNETA_DESC_ALIGNED_SIZE,
+ txq->descs, txq->descs_phys);
+
+ txq->descs = NULL;
+ txq->last_desc = 0;
+ txq->next_desc_to_proc = 0;
+ txq->descs_phys = 0;
+
+ /* Set minimum bandwidth for disabled TXQs */
+ mvreg_write(pp, MVETH_TXQ_TOKEN_CFG_REG(txq->id), 0);
+ mvreg_write(pp, MVETH_TXQ_TOKEN_COUNT_REG(txq->id), 0);
+
+ /* Set Tx descriptors queue starting address and size */
+ mvreg_write(pp, MVNETA_TXQ_BASE_ADDR_REG(txq->id), 0);
+ mvreg_write(pp, MVNETA_TXQ_SIZE_REG(txq->id), 0);
+}
+
+/* Cleanup all Tx queues */
+static void mvneta_cleanup_txqs(struct mvneta_port *pp)
+{
+ int queue;
+
+ for (queue = 0; queue < txq_number; queue++)
+ mvneta_txq_deinit(pp, &pp->txqs[queue]);
+}
+
+/* Cleanup all Rx queues */
+static void mvneta_cleanup_rxqs(struct mvneta_port *pp)
+{
+ int queue;
+
+ for (queue = 0; queue < rxq_number; queue++)
+ mvneta_rxq_deinit(pp, &pp->rxqs[queue]);
+}
+
+
+/* Init all Rx queues */
+static int mvneta_setup_rxqs(struct mvneta_port *pp)
+{
+ int queue;
+
+ for (queue = 0; queue < rxq_number; queue++) {
+ int err = mvneta_rxq_init(pp, &pp->rxqs[queue]);
+ if (err) {
+ netdev_err(pp->dev, "%s: can't create rxq=%d\n",
+ __func__, queue);
+ mvneta_cleanup_rxqs(pp);
+ return err;
+ }
+ }
+
+ return 0;
+}
+
+/* Init all tx queues */
+static int mvneta_setup_txqs(struct mvneta_port *pp)
+{
+ int queue;
+
+ for (queue = 0; queue < txq_number; queue++) {
+ int err = mvneta_txq_init(pp, &pp->txqs[queue]);
+ if (err) {
+ netdev_err(pp->dev, "%s: can't create txq=%d\n",
+ __func__, queue);
+ mvneta_cleanup_txqs(pp);
+ return err;
+ }
+ }
+
+ return 0;
+}
+
+static void mvneta_start_dev(struct mvneta_port *pp)
+{
+ mvneta_max_rx_size_set(pp, pp->pkt_size);
+ mvneta_txq_max_tx_size_set(pp, pp->pkt_size);
+
+ /* start the Rx/Tx activity */
+ mvneta_port_enable(pp);
+
+ /* Enable polling on the port */
+ napi_enable(&pp->napi);
+
+ /* Unmask interrupts */
+ mvreg_write(pp, MVNETA_INTR_NEW_MASK,
+ MVNETA_RX_INTR_MASK(rxq_number));
+
+ phy_start(pp->phy_dev);
+ netif_tx_start_all_queues(pp->dev);
+}
+
+static void mvneta_stop_dev(struct mvneta_port *pp)
+{
+ phy_stop(pp->phy_dev);
+
+ napi_disable(&pp->napi);
+
+ netif_carrier_off(pp->dev);
+
+ mvneta_port_down(pp);
+ netif_tx_stop_all_queues(pp->dev);
+
+ /* Stop the port activity */
+ mvneta_port_disable(pp);
+
+ /* Clear all ethernet port interrupts */
+ mvreg_write(pp, MVNETA_INTR_MISC_CAUSE, 0);
+ mvreg_write(pp, MVNETA_INTR_OLD_CAUSE, 0);
+
+ /* Mask all ethernet port interrupts */
+ mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
+ mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
+ mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
+
+ mvneta_tx_reset(pp);
+ mvneta_rx_reset(pp);
+}
+
+/* tx timeout callback - display a message and stop/start the network device */
+static void mvneta_tx_timeout(struct net_device *dev)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ netdev_info(dev, "tx timeout\n");
+ mvneta_stop_dev(pp);
+ mvneta_start_dev(pp);
+}
+
+/* Return positive if MTU is valid */
+static int mvneta_check_mtu_valid(struct net_device *dev, int mtu)
+{
+ if (mtu < 68) {
+ netdev_err(dev, "cannot change mtu to less than 68\n");
+ return -EINVAL;
+ }
+
+ /* 9676 == 9700 - 20 and rounding to 8 */
+ if (mtu > 9676) {
+ netdev_info(dev, "Illegal MTU value %d, round to 9676\n", mtu);
+ mtu = 9676;
+ }
+
+ if (!IS_ALIGNED(MVNETA_RX_PKT_SIZE(mtu), 8)) {
+ netdev_info(dev, "Illegal MTU value %d, rounding to %d\n",
+ mtu, ALIGN(MVNETA_RX_PKT_SIZE(mtu), 8));
+ mtu = ALIGN(MVNETA_RX_PKT_SIZE(mtu), 8);
+ }
+
+ return mtu;
+}
+
+/* Change the device mtu */
+static int mvneta_change_mtu(struct net_device *dev, int mtu)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ int ret;
+
+ mtu = mvneta_check_mtu_valid(dev, mtu);
+ if (mtu < 0)
+ return -EINVAL;
+
+ dev->mtu = mtu;
+
+ if (!netif_running(dev))
+ return 0;
+
+ /*
+ * The interface is running, so we have to force a
+ * reallocation of the RXQs
+ */
+ mvneta_stop_dev(pp);
+
+ mvneta_cleanup_txqs(pp);
+ mvneta_cleanup_rxqs(pp);
+
+ pp->pkt_size = MVNETA_RX_PKT_SIZE(pp->dev->mtu);
+
+ ret = mvneta_setup_rxqs(pp);
+ if (ret) {
+ netdev_err(pp->dev, "unable to setup rxqs after MTU change\n");
+ return ret;
+ }
+
+ mvneta_setup_txqs(pp);
+
+ mvneta_start_dev(pp);
+ mvneta_port_up(pp);
+
+ return 0;
+}
+
+/* Handle setting mac address */
+static int mvneta_set_mac_addr(struct net_device *dev, void *addr)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ u8 *mac = addr + 2;
+ int i;
+
+ if (netif_running(dev))
+ return -EBUSY;
+
+ /* Remove previous address table entry */
+ mvneta_mac_addr_set(pp, dev->dev_addr, -1);
+
+ /* Set new addr in hw */
+ mvneta_mac_addr_set(pp, mac, rxq_def);
+
+ /* Set addr in the device */
+ for (i = 0; i < ETH_ALEN; i++)
+ dev->dev_addr[i] = mac[i];
+
+ return 0;
+}
+
+static void mvneta_adjust_link(struct net_device *ndev)
+{
+ struct mvneta_port *pp = netdev_priv(ndev);
+ struct phy_device *phydev = pp->phy_dev;
+ int status_change = 0;
+
+ if (phydev->link) {
+ if ((pp->speed != phydev->speed) ||
+ (pp->duplex != phydev->duplex)) {
+ u32 val;
+
+ val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+ val &= ~(MVNETA_GMAC_CONFIG_MII_SPEED |
+ MVNETA_GMAC_CONFIG_GMII_SPEED |
+ MVNETA_GMAC_CONFIG_FULL_DUPLEX);
+
+ if (phydev->duplex)
+ val |= MVNETA_GMAC_CONFIG_FULL_DUPLEX;
+
+ if (phydev->speed == SPEED_1000)
+ val |= MVNETA_GMAC_CONFIG_GMII_SPEED;
+ else
+ val |= MVNETA_GMAC_CONFIG_MII_SPEED;
+
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
+
+ pp->duplex = phydev->duplex;
+ pp->speed = phydev->speed;
+ }
+ }
+
+ if (phydev->link != pp->link) {
+ if (!phydev->link) {
+ pp->duplex = -1;
+ pp->speed = 0;
+ }
+
+ pp->link = phydev->link;
+ status_change = 1;
+ }
+
+ if (status_change) {
+ if (phydev->link) {
+ u32 val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+ val |= (MVNETA_GMAC_FORCE_LINK_PASS |
+ MVNETA_GMAC_FORCE_LINK_DOWN);
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
+ mvneta_port_up(pp);
+ netdev_info(pp->dev, "link up\n");
+ } else {
+ mvneta_port_down(pp);
+ netdev_info(pp->dev, "link down\n");
+ }
+ }
+}
+
+static int mvneta_mdio_probe(struct mvneta_port *pp)
+{
+ struct phy_device *phy_dev;
+
+ phy_dev = of_phy_connect(pp->dev, pp->phy_node, mvneta_adjust_link, 0,
+ pp->phy_interface);
+ if (!phy_dev) {
+ netdev_err(pp->dev, "could not find the PHY\n");
+ return -ENODEV;
+ }
+
+ phy_dev->supported &= PHY_GBIT_FEATURES;
+ phy_dev->advertising = phy_dev->supported;
+
+ pp->phy_dev = phy_dev;
+ pp->link = 0;
+ pp->duplex = 0;
+ pp->speed = 0;
+
+ return 0;
+}
+
+static void mvneta_mdio_remove(struct mvneta_port *pp)
+{
+ phy_disconnect(pp->phy_dev);
+ pp->phy_dev = NULL;
+}
+
+static int mvneta_open(struct net_device *dev)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ int ret;
+
+ mvneta_mac_addr_set(pp, dev->dev_addr, rxq_def);
+
+ pp->pkt_size = MVNETA_RX_PKT_SIZE(pp->dev->mtu);
+
+ ret = mvneta_setup_rxqs(pp);
+ if (ret)
+ return ret;
+
+ ret = mvneta_setup_txqs(pp);
+ if (ret)
+ goto err_cleanup_rxqs;
+
+ /* Connect to port interrupt line */
+ ret = request_irq(pp->dev->irq, mvneta_isr, 0,
+ MVNETA_DRIVER_NAME, pp);
+ if (ret) {
+ netdev_err(pp->dev, "cannot request irq %d\n", pp->dev->irq);
+ goto err_cleanup_txqs;
+ }
+
+ /* In default link is down */
+ netif_carrier_off(pp->dev);
+
+ ret = mvneta_mdio_probe(pp);
+ if (ret < 0) {
+ netdev_err(dev, "cannot probe MDIO bus\n");
+ goto err_free_irq;
+ }
+
+ mvneta_start_dev(pp);
+
+ return 0;
+
+err_free_irq:
+ free_irq(pp->dev->irq, pp);
+err_cleanup_txqs:
+ mvneta_cleanup_txqs(pp);
+err_cleanup_rxqs:
+ mvneta_cleanup_rxqs(pp);
+ return ret;
+}
+
+/* Stop the port, free port interrupt line */
+static int mvneta_stop(struct net_device *dev)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ mvneta_stop_dev(pp);
+ mvneta_mdio_remove(pp);
+ free_irq(dev->irq, pp);
+ mvneta_cleanup_rxqs(pp);
+ mvneta_cleanup_txqs(pp);
+ del_timer(&pp->tx_done_timer);
+ clear_bit(MVNETA_F_TX_DONE_TIMER_BIT, &pp->flags);
+
+ return 0;
+}
+
+/* Ethtool methods */
+
+/* Get settings (phy address, speed) for ethtools */
+int mvneta_ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ if (!pp->phy_dev)
+ return -ENODEV;
+
+ return phy_ethtool_gset(pp->phy_dev, cmd);
+}
+
+/* Set settings (phy address, speed) for ethtools */
+int mvneta_ethtool_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ if (!pp->phy_dev)
+ return -ENODEV;
+
+ return phy_ethtool_sset(pp->phy_dev, cmd);
+}
+
+/* Set interrupt coalescing for ethtools */
+static int mvneta_ethtool_set_coalesce(struct net_device *dev,
+ struct ethtool_coalesce *c)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ int queue;
+
+ for (queue = 0; queue < rxq_number; queue++) {
+ struct mvneta_rx_queue *rxq = &pp->rxqs[queue];
+ rxq->time_coal = c->rx_coalesce_usecs;
+ rxq->pkts_coal = c->rx_max_coalesced_frames;
+ mvneta_rx_pkts_coal_set(pp, rxq, rxq->pkts_coal);
+ mvneta_rx_time_coal_set(pp, rxq, rxq->time_coal);
+ }
+
+ for (queue = 0; queue < txq_number; queue++) {
+ struct mvneta_tx_queue *txq = &pp->txqs[queue];
+ txq->done_pkts_coal = c->tx_max_coalesced_frames;
+ mvneta_tx_done_pkts_coal_set(pp, txq, txq->done_pkts_coal);
+ }
+
+ return 0;
+}
+
+/* get coalescing for ethtools */
+static int mvneta_ethtool_get_coalesce(struct net_device *dev,
+ struct ethtool_coalesce *c)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ c->rx_coalesce_usecs = pp->rxqs[0].time_coal;
+ c->rx_max_coalesced_frames = pp->rxqs[0].pkts_coal;
+
+ c->tx_max_coalesced_frames = pp->txqs[0].done_pkts_coal;
+ return 0;
+}
+
+
+static void mvneta_ethtool_get_drvinfo(struct net_device *dev,
+ struct ethtool_drvinfo *drvinfo)
+{
+ strlcpy(drvinfo->driver, MVNETA_DRIVER_NAME,
+ sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, MVNETA_DRIVER_VERSION,
+ sizeof(drvinfo->version));
+ strlcpy(drvinfo->bus_info, dev_name(&dev->dev),
+ sizeof(drvinfo->bus_info));
+}
+
+
+static void mvneta_ethtool_get_ringparam(struct net_device *netdev,
+ struct ethtool_ringparam *ring)
+{
+ struct mvneta_port *pp = netdev_priv(netdev);
+
+ ring->rx_max_pending = MVNETA_MAX_RXD;
+ ring->tx_max_pending = MVNETA_MAX_TXD;
+ ring->rx_pending = pp->rx_ring_size;
+ ring->tx_pending = pp->tx_ring_size;
+}
+
+static int mvneta_ethtool_set_ringparam(struct net_device *dev,
+ struct ethtool_ringparam *ring)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ if ((ring->rx_pending == 0) || (ring->tx_pending == 0))
+ return -EINVAL;
+ pp->rx_ring_size = ring->rx_pending < MVNETA_MAX_RXD ?
+ ring->rx_pending : MVNETA_MAX_RXD;
+ pp->tx_ring_size = ring->tx_pending < MVNETA_MAX_TXD ?
+ ring->tx_pending : MVNETA_MAX_TXD;
+
+ if (netif_running(dev)) {
+ mvneta_stop(dev);
+ if (mvneta_open(dev)) {
+ netdev_err(dev,
+ "error on opening device after ring param change\n");
+ return -ENOMEM;
+ }
+ }
+
+ return 0;
+}
+
+static const struct net_device_ops mvneta_netdev_ops = {
+ .ndo_open = mvneta_open,
+ .ndo_stop = mvneta_stop,
+ .ndo_start_xmit = mvneta_tx,
+ .ndo_set_rx_mode = mvneta_set_rx_mode,
+ .ndo_set_mac_address = mvneta_set_mac_addr,
+ .ndo_change_mtu = mvneta_change_mtu,
+ .ndo_tx_timeout = mvneta_tx_timeout,
+ .ndo_get_stats64 = mvneta_get_stats64,
+};
+
+const struct ethtool_ops mvneta_eth_tool_ops = {
+ .get_link = ethtool_op_get_link,
+ .get_settings = mvneta_ethtool_get_settings,
+ .set_settings = mvneta_ethtool_set_settings,
+ .set_coalesce = mvneta_ethtool_set_coalesce,
+ .get_coalesce = mvneta_ethtool_get_coalesce,
+ .get_drvinfo = mvneta_ethtool_get_drvinfo,
+ .get_ringparam = mvneta_ethtool_get_ringparam,
+ .set_ringparam = mvneta_ethtool_set_ringparam,
+};
+
+/* Initialize hw */
+static int __devinit mvneta_init(struct mvneta_port *pp, int phy_addr)
+{
+ int queue;
+
+ /* Disable port */
+ mvneta_port_disable(pp);
+
+ /* Set port default values */
+ mvneta_defaults_set(pp);
+
+ pp->txqs = kzalloc(txq_number * sizeof(struct mvneta_tx_queue),
+ GFP_KERNEL);
+ if (!pp->txqs)
+ return -ENOMEM;
+
+ /* Initialize TX descriptor rings */
+ for (queue = 0; queue < txq_number; queue++) {
+ struct mvneta_tx_queue *txq = &pp->txqs[queue];
+ txq->id = queue;
+ txq->size = pp->tx_ring_size;
+ txq->done_pkts_coal = MVNETA_TXDONE_COAL_PKTS;
+ }
+
+ pp->rxqs = kzalloc(rxq_number * sizeof(struct mvneta_rx_queue),
+ GFP_KERNEL);
+ if (!pp->rxqs) {
+ kfree(pp->txqs);
+ return -ENOMEM;
+ }
+
+ /* Create Rx descriptor rings */
+ for (queue = 0; queue < rxq_number; queue++) {
+ struct mvneta_rx_queue *rxq = &pp->rxqs[queue];
+ rxq->id = queue;
+ rxq->size = pp->rx_ring_size;
+ rxq->pkts_coal = MVNETA_RX_COAL_PKTS;
+ rxq->time_coal = MVNETA_RX_COAL_USEC;
+ }
+
+ return 0;
+}
+
+static void __devexit mvneta_deinit(struct mvneta_port *pp)
+{
+ kfree(pp->txqs);
+ kfree(pp->rxqs);
+}
+
+/* platform glue : initialize decoding windows */
+static void __devinit
+mvneta_conf_mbus_windows(struct mvneta_port *pp,
+ const struct mbus_dram_target_info *dram)
+{
+ u32 win_enable;
+ u32 win_protect;
+ int i;
+
+ for (i = 0; i < 6; i++) {
+ mvreg_write(pp, MVNETA_WIN_BASE(i), 0);
+ mvreg_write(pp, MVNETA_WIN_SIZE(i), 0);
+
+ if (i < 4)
+ mvreg_write(pp, MVNETA_WIN_REMAP(i), 0);
+ }
+
+ win_enable = 0x3f;
+ win_protect = 0;
+
+ for (i = 0; i < dram->num_cs; i++) {
+ const struct mbus_dram_window *cs = dram->cs + i;
+ mvreg_write(pp, MVNETA_WIN_BASE(i), (cs->base & 0xffff0000) |
+ (cs->mbus_attr << 8) | dram->mbus_dram_target_id);
+
+ mvreg_write(pp, MVNETA_WIN_SIZE(i),
+ (cs->size - 1) & 0xffff0000);
+
+ win_enable &= ~(1 << i);
+ win_protect |= 3 << (2 * i);
+ }
+
+ mvreg_write(pp, MVNETA_BASE_ADDR_ENABLE, win_enable);
+}
+
+/* Power up the port */
+static void __devinit mvneta_port_power_up(struct mvneta_port *pp, int phy_mode)
+{
+ u32 val;
+
+ /* MAC Cause register should be cleared */
+ mvreg_write(pp, MVNETA_UNIT_INTR_CAUSE, 0);
+
+ if (phy_mode == PHY_INTERFACE_MODE_SGMII)
+ mvneta_port_sgmii_config(pp);
+
+ mvneta_gmac_rgmii_set(pp, 1);
+
+ /* Cancel Port Reset */
+ val = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
+ val &= ~MVNETA_GMAC2_PORT_RESET;
+ mvreg_write(pp, MVNETA_GMAC_CTRL_2, val);
+
+ while ((mvreg_read(pp, MVNETA_GMAC_CTRL_2) &
+ MVNETA_GMAC2_PORT_RESET) != 0)
+ continue;
+}
+
+/* Device initialization routine */
+static int __devinit mvneta_probe(struct platform_device *pdev)
+{
+ const struct mbus_dram_target_info *dram_target_info;
+ struct device_node *dn = pdev->dev.of_node;
+ struct device_node *phy_node;
+ u32 phy_addr, clk_rate_hz;
+ struct mvneta_port *pp;
+ struct net_device *dev;
+ const char *mac_addr;
+ int phy_mode;
+ int err;
+
+ /*
+ * Our multiqueue support is not complete, so for now, only
+ * allow the usage of the first RX queue
+ */
+ if (rxq_def != 0) {
+ dev_err(&pdev->dev, "Invalid rxq_def argument: %d\n", rxq_def);
+ return -EINVAL;
+ }
+
+ dev = alloc_etherdev_mq(sizeof(struct mvneta_port), 8);
+ if (!dev)
+ return -ENOMEM;
+
+ dev->irq = irq_of_parse_and_map(dn, 0);
+ if (dev->irq == 0) {
+ err = -EINVAL;
+ goto err_free_netdev;
+ }
+
+ phy_node = of_parse_phandle(dn, "phy", 0);
+ if (!phy_node) {
+ dev_err(&pdev->dev, "no associated PHY\n");
+ err = -ENODEV;
+ goto err_free_irq;
+ }
+
+ phy_mode = of_get_phy_mode(dn);
+ if (phy_mode < 0) {
+ dev_err(&pdev->dev, "incorrect phy-mode\n");
+ err = -EINVAL;
+ goto err_free_irq;
+ }
+
+ if (of_property_read_u32(dn, "clock-frequency", &clk_rate_hz) != 0) {
+ dev_err(&pdev->dev, "could not read clock-frequency\n");
+ err = -EINVAL;
+ goto err_free_irq;
+ }
+
+ mac_addr = of_get_mac_address(dn);
+
+ if (!mac_addr || !is_valid_ether_addr(mac_addr))
+ eth_hw_addr_random(dev);
+ else
+ memcpy(dev->dev_addr, mac_addr, ETH_ALEN);
+
+ dev->tx_queue_len = MVNETA_MAX_TXD;
+ dev->watchdog_timeo = 5 * HZ;
+ dev->netdev_ops = &mvneta_netdev_ops;
+
+ SET_ETHTOOL_OPS(dev, &mvneta_eth_tool_ops);
+
+ pp = netdev_priv(dev);
+
+ pp->tx_done_timer.function = mvneta_tx_done_timer_callback;
+ init_timer(&pp->tx_done_timer);
+ clear_bit(MVNETA_F_TX_DONE_TIMER_BIT, &pp->flags);
+
+ pp->weight = MVNETA_RX_POLL_WEIGHT;
+ pp->clk_rate_hz = clk_rate_hz;
+ pp->phy_node = phy_node;
+ pp->phy_interface = phy_mode;
+
+ pp->base = of_iomap(dn, 0);
+ if (pp->base == NULL) {
+ err = -ENOMEM;
+ goto err_free_irq;
+ }
+
+ pp->tx_done_timer.data = (unsigned long)dev;
+
+ pp->tx_ring_size = MVNETA_MAX_TXD;
+ pp->rx_ring_size = MVNETA_MAX_RXD;
+
+ pp->dev = dev;
+ SET_NETDEV_DEV(dev, &pdev->dev);
+
+ err = mvneta_init(pp, phy_addr);
+ if (err < 0) {
+ dev_err(&pdev->dev, "can't init eth hal\n");
+ goto err_unmap;
+ }
+ mvneta_port_power_up(pp, phy_mode);
+
+ dram_target_info = mv_mbus_dram_info();
+ if (dram_target_info)
+ mvneta_conf_mbus_windows(pp, dram_target_info);
+
+ netif_napi_add(dev, &pp->napi, mvneta_poll, pp->weight);
+
+ err = register_netdev(dev);
+ if (err < 0) {
+ dev_err(&pdev->dev, "failed to register\n");
+ goto err_deinit;
+ }
+
+ dev->features = NETIF_F_SG | NETIF_F_IP_CSUM;
+ dev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM;
+ dev->priv_flags |= IFF_UNICAST_FLT;
+
+ netdev_info(dev, "mac: %pM\n", dev->dev_addr);
+
+ platform_set_drvdata(pdev, pp->dev);
+
+ return 0;
+
+err_deinit:
+ mvneta_deinit(pp);
+err_unmap:
+ iounmap(pp->base);
+err_free_irq:
+ irq_dispose_mapping(dev->irq);
+err_free_netdev:
+ free_netdev(dev);
+ return err;
+}
+
+/* Device removal routine */
+static int __devexit mvneta_remove(struct platform_device *pdev)
+{
+ struct net_device *dev = platform_get_drvdata(pdev);
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ unregister_netdev(dev);
+ mvneta_deinit(pp);
+ iounmap(pp->base);
+ irq_dispose_mapping(dev->irq);
+ free_netdev(dev);
+
+ platform_set_drvdata(pdev, NULL);
+
+ return 0;
+}
+
+static const struct of_device_id mvneta_match[] = {
+ { .compatible = "marvell,armada-370-neta" },
+ { }
+};
+MODULE_DEVICE_TABLE(of, mvneta_match);
+
+static struct platform_driver mvneta_driver = {
+ .probe = mvneta_probe,
+ .remove = __devexit_p(mvneta_remove),
+ .driver = {
+ .name = MVNETA_DRIVER_NAME,
+ .of_match_table = mvneta_match,
+ },
+};
+
+module_platform_driver(mvneta_driver);
+
+MODULE_DESCRIPTION("Marvell NETA Ethernet Driver - www.marvell.com");
+MODULE_AUTHOR("Rami Rosen <rosenr@marvell.com>, Thomas Petazzoni <thomas.petazzoni@free-electrons.com>");
+MODULE_LICENSE("GPL");
+
+module_param(rxq_number, int, S_IRUGO);
+module_param(txq_number, int, S_IRUGO);
+
+module_param(rxq_def, int, S_IRUGO);
+module_param(txq_def, int, S_IRUGO);
--
1.7.9.5
^ 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