* Re: [PATCH 1/2] ipv4: Remove output route check in ipv4_mtu
From: Timo Teras @ 2013-01-17 8:35 UTC (permalink / raw)
To: Steffen Klassert; +Cc: David Miller, Julian Anastasov, luky-37, pupilla, netdev
In-Reply-To: <20130117065500.GG18940@secunet.com>
On Thu, 17 Jan 2013 07:55:01 +0100 Steffen Klassert
<steffen.klassert@secunet.com> wrote:
> The output route check was introduced with git commit 261663b0
> (ipv4: Don't use the cached pmtu informations for input routes)
> during times when we cached the pmtu informations on the
> inetpeer. Now the pmtu informations are back in the routes,
> so this check is obsolete. It also had some unwanted side effects,
> as reported by Timo Teras and Lukas Tribus.
>
> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
> ---
Acked-by: Timo Teräs <timo.teras@iki.fi>
I think this should also go to stable queues which are applicable, since
the original commit caused CLAMPMSS and XFRM pmtu regressions.
The original thread of this is:
http://marc.info/?t=134208398500001&r=1&w=2
My original revert request is at:
http://marc.info/?l=linux-netdev&m=134242561624266&w=2
and additional reasoning at:
http://marc.info/?l=linux-netdev&m=134243626627716&w=2
Thanks,
Timo
^ permalink raw reply
* [Patch net-next 3/2] xfrm: use separated locks to protect pointers of struct xfrm_state_afinfo
From: Cong Wang @ 2013-01-17 8:34 UTC (permalink / raw)
To: netdev; +Cc: Steffen Klassert, Herbert Xu, David S. Miller, Cong Wang
In-Reply-To: <1358323506-19571-2-git-send-email-amwang@redhat.com>
(On top of my previous RCU convert patches)
afinfo->type_map and afinfo->mode_map deserve separated locks,
they are different things.
We should just take RCU read lock to protect afinfo itself,
but not for the inner pointers.
Cc: Steffen Klassert <steffen.klassert@secunet.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index b40baf7..727e357 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -168,57 +168,45 @@ int __xfrm_state_delete(struct xfrm_state *x);
int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol);
void km_state_expired(struct xfrm_state *x, int hard, u32 portid);
-static struct xfrm_state_afinfo *xfrm_state_lock_afinfo(unsigned int family)
-{
- struct xfrm_state_afinfo *afinfo;
- if (unlikely(family >= NPROTO))
- return NULL;
- spin_lock_bh(&xfrm_state_afinfo_lock);
- afinfo = xfrm_state_afinfo[family];
- if (unlikely(!afinfo))
- spin_unlock_bh(&xfrm_state_afinfo_lock);
- return afinfo;
-}
-
-static void xfrm_state_unlock_afinfo(struct xfrm_state_afinfo *afinfo)
-{
- spin_unlock_bh(&xfrm_state_afinfo_lock);
-}
-
+static DEFINE_SPINLOCK(xfrm_type_lock);
int xfrm_register_type(const struct xfrm_type *type, unsigned short family)
{
- struct xfrm_state_afinfo *afinfo = xfrm_state_lock_afinfo(family);
+ struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
const struct xfrm_type **typemap;
int err = 0;
if (unlikely(afinfo == NULL))
return -EAFNOSUPPORT;
typemap = afinfo->type_map;
+ spin_lock_bh(&xfrm_type_lock);
if (likely(typemap[type->proto] == NULL))
typemap[type->proto] = type;
else
err = -EEXIST;
- xfrm_state_unlock_afinfo(afinfo);
+ spin_unlock_bh(&xfrm_type_lock);
+ xfrm_state_put_afinfo(afinfo);
return err;
}
EXPORT_SYMBOL(xfrm_register_type);
int xfrm_unregister_type(const struct xfrm_type *type, unsigned short family)
{
- struct xfrm_state_afinfo *afinfo = xfrm_state_lock_afinfo(family);
+ struct xfrm_state_afinfo *afinfo = xfrm_state_get_afinfo(family);
const struct xfrm_type **typemap;
int err = 0;
if (unlikely(afinfo == NULL))
return -EAFNOSUPPORT;
typemap = afinfo->type_map;
+ spin_lock_bh(&xfrm_type_lock);
if (unlikely(typemap[type->proto] != type))
err = -ENOENT;
else
typemap[type->proto] = NULL;
- xfrm_state_unlock_afinfo(afinfo);
+ spin_unlock_bh(&xfrm_type_lock);
+ xfrm_state_put_afinfo(afinfo);
return err;
}
EXPORT_SYMBOL(xfrm_unregister_type);
@@ -255,6 +243,7 @@ static void xfrm_put_type(const struct xfrm_type *type)
module_put(type->owner);
}
+static DEFINE_SPINLOCK(xfrm_mode_lock);
int xfrm_register_mode(struct xfrm_mode *mode, int family)
{
struct xfrm_state_afinfo *afinfo;
@@ -264,12 +253,13 @@ int xfrm_register_mode(struct xfrm_mode *mode, int family)
if (unlikely(mode->encap >= XFRM_MODE_MAX))
return -EINVAL;
- afinfo = xfrm_state_lock_afinfo(family);
+ afinfo = xfrm_state_get_afinfo(family);
if (unlikely(afinfo == NULL))
return -EAFNOSUPPORT;
err = -EEXIST;
modemap = afinfo->mode_map;
+ spin_lock_bh(&xfrm_mode_lock);
if (modemap[mode->encap])
goto out;
@@ -282,7 +272,8 @@ int xfrm_register_mode(struct xfrm_mode *mode, int family)
err = 0;
out:
- xfrm_state_unlock_afinfo(afinfo);
+ spin_unlock_bh(&xfrm_mode_lock);
+ xfrm_state_put_afinfo(afinfo);
return err;
}
EXPORT_SYMBOL(xfrm_register_mode);
@@ -296,19 +287,21 @@ int xfrm_unregister_mode(struct xfrm_mode *mode, int family)
if (unlikely(mode->encap >= XFRM_MODE_MAX))
return -EINVAL;
- afinfo = xfrm_state_lock_afinfo(family);
+ afinfo = xfrm_state_get_afinfo(family);
if (unlikely(afinfo == NULL))
return -EAFNOSUPPORT;
err = -ENOENT;
modemap = afinfo->mode_map;
+ spin_lock_bh(&xfrm_mode_lock);
if (likely(modemap[mode->encap] == mode)) {
modemap[mode->encap] = NULL;
module_put(mode->afinfo->owner);
err = 0;
}
- xfrm_state_unlock_afinfo(afinfo);
+ spin_unlock_bh(&xfrm_mode_lock);
+ xfrm_state_put_afinfo(afinfo);
return err;
}
EXPORT_SYMBOL(xfrm_unregister_mode);
^ permalink raw reply related
* Re: [PATCH v4 1/1] net: ethernet: davinci_cpdma: Add boundary for rx and tx descriptors
From: Christian Riesch @ 2013-01-17 8:37 UTC (permalink / raw)
To: Mugunthan V N; +Cc: netdev, davem, s.hauer
In-Reply-To: <1358381405-7247-1-git-send-email-mugunthanvnm@ti.com>
Hi,
On 2013-01-17 01:10, Mugunthan V N wrote:
> When there is heavy transmission traffic in the CPDMA, then Rx descriptors
> memory is also utilized as tx desc memory looses all rx descriptors and the
> driver stops working then.
>
> This patch adds boundary for tx and rx descriptors in bd ram dividing the
> descriptor memory to ensure that during heavy transmission tx doesn't use
> rx descriptors.
>
> This patch is already applied to davinci_emac driver, since CPSW and
> davici_dmac shares the same CPDMA, moving the boundry seperation from
> Davinci EMAC driver to CPDMA driver which was done in the following
> commit
>
> commit 86d8c07ff2448eb4e860e50f34ef6ee78e45c40c
> Author: Sascha Hauer <s.hauer@pengutronix.de>
> Date: Tue Jan 3 05:27:47 2012 +0000
>
> net/davinci: do not use all descriptors for tx packets
>
> The driver uses a shared pool for both rx and tx descriptors.
> During open it queues fixed number of 128 descriptors for receive
> packets. For each received packet it tries to queue another
> descriptor. If this fails the descriptor is lost for rx.
> The driver has no limitation on tx descriptors to use, so it
> can happen during a nmap / ping -f attack that the driver
> allocates all descriptors for tx and looses all rx descriptors.
> The driver stops working then.
> To fix this limit the number of tx descriptors used to half of
> the descriptors available, the rx path uses the other half.
>
> Tested on a custom board using nmap / ping -f to the board from
> two different hosts.
>
> Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
> ---
> Changes from v3
> * Changed multiline commenting as per networking style
>
> drivers/net/ethernet/ti/cpsw.c | 9 ++++++
> drivers/net/ethernet/ti/davinci_cpdma.c | 51 ++++++++++++++++++++++++++-----
> drivers/net/ethernet/ti/davinci_cpdma.h | 1 +
> drivers/net/ethernet/ti/davinci_emac.c | 13 ++++----
> 4 files changed, 61 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
> index 3772804..b35e6a7 100644
> --- a/drivers/net/ethernet/ti/cpsw.c
> +++ b/drivers/net/ethernet/ti/cpsw.c
> @@ -374,6 +374,9 @@ void cpsw_tx_handler(void *token, int len, int status)
> struct net_device *ndev = skb->dev;
> struct cpsw_priv *priv = netdev_priv(ndev);
>
> + /* Check whether the queue is stopped due to stalled tx dma, if the
> + * queue is stopped then start the queue as we have free desc for tx
> + */
> if (unlikely(netif_queue_stopped(ndev)))
> netif_start_queue(ndev);
> cpts_tx_timestamp(&priv->cpts, skb);
> @@ -736,6 +739,12 @@ static netdev_tx_t cpsw_ndo_start_xmit(struct sk_buff *skb,
> goto fail;
> }
>
> + /* If there is no more tx desc left free then we need to
> + * tell the kernel to stop sending us tx frames.
> + */
> + if (unlikely(cpdma_check_free_tx_desc(priv->txch)))
> + netif_stop_queue(ndev);
> +
> return NETDEV_TX_OK;
> fail:
> priv->stats.tx_dropped++;
> diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c
> index 4995673..4f6d82f 100644
> --- a/drivers/net/ethernet/ti/davinci_cpdma.c
> +++ b/drivers/net/ethernet/ti/davinci_cpdma.c
> @@ -105,13 +105,13 @@ struct cpdma_ctlr {
> };
>
> struct cpdma_chan {
> + struct cpdma_desc __iomem *head, *tail;
> + void __iomem *hdp, *cp, *rxfree;
> enum cpdma_state state;
> struct cpdma_ctlr *ctlr;
> int chan_num;
> spinlock_t lock;
> - struct cpdma_desc __iomem *head, *tail;
> int count;
> - void __iomem *hdp, *cp, *rxfree;
I still do not see any benefit from this rearranging here.
> u32 mask;
> cpdma_handler_fn handler;
> enum dma_data_direction dir;
> @@ -217,17 +217,29 @@ desc_from_phys(struct cpdma_desc_pool *pool, dma_addr_t dma)
> }
>
> static struct cpdma_desc __iomem *
> -cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc)
> +cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc, bool is_rx)
> {
> unsigned long flags;
> int index;
> + int desc_start;
> + int desc_end;
> struct cpdma_desc __iomem *desc = NULL;
>
> spin_lock_irqsave(&pool->lock, flags);
>
> - index = bitmap_find_next_zero_area(pool->bitmap, pool->num_desc, 0,
> - num_desc, 0);
> - if (index < pool->num_desc) {
> + if (is_rx) {
> + desc_start = 0;
> + desc_end = pool->num_desc/2;
> + index = bitmap_find_next_zero_area(pool->bitmap,
> + pool->num_desc/2, 0, num_desc, 0);
Duplicate calls of bitmap_find_next_zero_area, see a few lines below.
> + } else {
> + desc_start = pool->num_desc/2;
> + desc_end = pool->num_desc;
> + }
> +
> + index = bitmap_find_next_zero_area(pool->bitmap,
> + desc_end, desc_start, num_desc, 0);
> + if (index < desc_end) {
> bitmap_set(pool->bitmap, index, num_desc);
> desc = pool->iomap + pool->desc_size * index;
> pool->used_desc++;
> @@ -660,6 +672,7 @@ int cpdma_chan_submit(struct cpdma_chan *chan, void *token, void *data,
> unsigned long flags;
> u32 mode;
> int ret = 0;
> + bool is_rx;
>
> spin_lock_irqsave(&chan->lock, flags);
>
> @@ -668,7 +681,8 @@ int cpdma_chan_submit(struct cpdma_chan *chan, void *token, void *data,
> goto unlock_ret;
> }
>
> - desc = cpdma_desc_alloc(ctlr->pool, 1);
> + is_rx = (chan->rxfree != 0);
I guess you should use is_rx_chan() here.
> + desc = cpdma_desc_alloc(ctlr->pool, 1, is_rx);
> if (!desc) {
> chan->stats.desc_alloc_fail++;
> ret = -ENOMEM;
> @@ -704,6 +718,29 @@ unlock_ret:
> }
> EXPORT_SYMBOL_GPL(cpdma_chan_submit);
>
> +bool cpdma_check_free_tx_desc(struct cpdma_chan *chan)
> +{
> + unsigned long flags;
> + int index;
> + bool ret;
> + struct cpdma_ctlr *ctlr = chan->ctlr;
> + struct cpdma_desc_pool *pool = ctlr->pool;
> +
> + spin_lock_irqsave(&pool->lock, flags);
> +
> + index = bitmap_find_next_zero_area(pool->bitmap,
> + pool->num_desc, pool->num_desc/2, 1, 0);
I still wonder if using two separate descriptor pools for rx and tx
would be a nicer implementation. In this case you could just use
bitmap_full here, right? And the two pools would have separate
spinlocks, so rx could not lock tx and vice versa (but as I wrote in an
earlier email, I am a newbie here, so I don't know if this would be
beneficial).
Regards, Christian
> +
> + if (index < pool->num_desc)
> + ret = true;
> + else
> + ret = false;
> +
> + spin_unlock_irqrestore(&pool->lock, flags);
> + return ret;
> +}
> +EXPORT_SYMBOL_GPL(cpdma_check_free_tx_desc);
> +
> static void __cpdma_chan_free(struct cpdma_chan *chan,
> struct cpdma_desc __iomem *desc,
> int outlen, int status)
> diff --git a/drivers/net/ethernet/ti/davinci_cpdma.h b/drivers/net/ethernet/ti/davinci_cpdma.h
> index afa19a0..8d2aeb2 100644
> --- a/drivers/net/ethernet/ti/davinci_cpdma.h
> +++ b/drivers/net/ethernet/ti/davinci_cpdma.h
> @@ -88,6 +88,7 @@ int cpdma_chan_process(struct cpdma_chan *chan, int quota);
> int cpdma_ctlr_int_ctrl(struct cpdma_ctlr *ctlr, bool enable);
> void cpdma_ctlr_eoi(struct cpdma_ctlr *ctlr);
> int cpdma_chan_int_ctrl(struct cpdma_chan *chan, bool enable);
> +bool cpdma_check_free_tx_desc(struct cpdma_chan *chan);
>
> enum cpdma_control {
> CPDMA_CMD_IDLE, /* write-only */
> diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c
> index 8478d98..1c97c81 100644
> --- a/drivers/net/ethernet/ti/davinci_emac.c
> +++ b/drivers/net/ethernet/ti/davinci_emac.c
> @@ -120,7 +120,6 @@ static const char emac_version_string[] = "TI DaVinci EMAC Linux v6.1";
> #define EMAC_DEF_TX_CH (0) /* Default 0th channel */
> #define EMAC_DEF_RX_CH (0) /* Default 0th channel */
> #define EMAC_DEF_RX_NUM_DESC (128)
> -#define EMAC_DEF_TX_NUM_DESC (128)
> #define EMAC_DEF_MAX_TX_CH (1) /* Max TX channels configured */
> #define EMAC_DEF_MAX_RX_CH (1) /* Max RX channels configured */
> #define EMAC_POLL_WEIGHT (64) /* Default NAPI poll weight */
> @@ -342,7 +341,6 @@ struct emac_priv {
> u32 mac_hash2;
> u32 multicast_hash_cnt[EMAC_NUM_MULTICAST_BITS];
> u32 rx_addr_type;
> - atomic_t cur_tx;
> const char *phy_id;
> #ifdef CONFIG_OF
> struct device_node *phy_node;
> @@ -1050,10 +1048,10 @@ static void emac_tx_handler(void *token, int len, int status)
> {
> struct sk_buff *skb = token;
> struct net_device *ndev = skb->dev;
> - struct emac_priv *priv = netdev_priv(ndev);
> -
> - atomic_dec(&priv->cur_tx);
>
> + /* Check whether the queue is stopped due to stalled tx dma, if the
> + * queue is stopped then start the queue as we have free desc for tx
> + */
> if (unlikely(netif_queue_stopped(ndev)))
> netif_start_queue(ndev);
> ndev->stats.tx_packets++;
> @@ -1101,7 +1099,10 @@ static int emac_dev_xmit(struct sk_buff *skb, struct net_device *ndev)
> goto fail_tx;
> }
>
> - if (atomic_inc_return(&priv->cur_tx) >= EMAC_DEF_TX_NUM_DESC)
> + /* If there is no more tx desc left free then we need to
> + * tell the kernel to stop sending us tx frames.
> + */
> + if (unlikely(cpdma_check_free_tx_desc(priv->txch)))
> netif_stop_queue(ndev);
>
> return NETDEV_TX_OK;
>
^ permalink raw reply
* Re: [PATCH 1/1] ipv6: fix header length calculation in ip6_append_data()
From: David Miller @ 2013-01-17 8:37 UTC (permalink / raw)
To: r.kuntz; +Cc: netdev, yoshfuji, steffen.klassert, linux-kernel
In-Reply-To: <F2E313A0-60C5-49DE-BBB7-BAEF532C0433@ipflavors.com>
From: Romain KUNTZ <r.kuntz@ipflavors.com>
Date: Wed, 16 Jan 2013 23:47:40 +0100
> Commit 299b0767 (ipv6: Fix IPsec slowpath fragmentation problem)
> has introduced a error in the header length calculation that
> provokes corrupted packets when non-fragmentable extensions
> headers (Destination Option or Routing Header Type 2) are used.
>
> rt->rt6i_nfheader_len is the length of the non-fragmentable
> extension header, and it should be substracted to
> rt->dst.header_len, and not to exthdrlen, as it was done before
> commit 299b0767.
>
> This patch reverts to the original and correct behavior. It has
> been successfully tested with and without IPsec on packets
> that include non-fragmentable extensions headers.
>
> Signed-off-by: Romain Kuntz <r.kuntz@ipflavors.com>
> Acked-by: Steffen Klassert <steffen.klassert@secunet.com>
Applied, thanks.
^ permalink raw reply
* Re: [PATCH 2/2] ipv4: Don't update the pmtu on mtu locked routes
From: Julian Anastasov @ 2013-01-17 8:41 UTC (permalink / raw)
To: Steffen Klassert; +Cc: David Miller, timo.teras, luky-37, pupilla, netdev
In-Reply-To: <20130117065810.GH18940@secunet.com>
Hello,
On Thu, 17 Jan 2013, Steffen Klassert wrote:
> Routes with locked mtu should not use learned pmtu informations,
> so do not update the pmtu on these routes.
>
> Reported-by: Julian Anastasov <ja@ssi.bg>
> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
> ---
> net/ipv4/route.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/net/ipv4/route.c b/net/ipv4/route.c
> index 6e4a89c..259cbee 100644
> --- a/net/ipv4/route.c
> +++ b/net/ipv4/route.c
> @@ -912,6 +912,9 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu)
> struct dst_entry *dst = &rt->dst;
> struct fib_result res;
>
> + if (dst_metric_locked(dst, RTAX_MTU))
> + return;
> +
> if (dst->dev->mtu < mtu)
> return;
>
> --
> 1.7.9.5
Thanks
--
Julian Anastasov <ja@ssi.bg>
^ permalink raw reply
* Re: [PATCH 1/2] ipv4: Remove output route check in ipv4_mtu
From: David Miller @ 2013-01-17 8:40 UTC (permalink / raw)
To: steffen.klassert; +Cc: ja, timo.teras, luky-37, pupilla, netdev
In-Reply-To: <20130117065500.GG18940@secunet.com>
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Thu, 17 Jan 2013 07:55:01 +0100
> The output route check was introduced with git commit 261663b0
> (ipv4: Don't use the cached pmtu informations for input routes)
> during times when we cached the pmtu informations on the
> inetpeer. Now the pmtu informations are back in the routes,
> so this check is obsolete. It also had some unwanted side effects,
> as reported by Timo Teras and Lukas Tribus.
>
> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Applied and queued up for -stable.
^ permalink raw reply
* Re: [PATCH 2/2] ipv4: Don't update the pmtu on mtu locked routes
From: David Miller @ 2013-01-17 8:40 UTC (permalink / raw)
To: steffen.klassert; +Cc: ja, timo.teras, luky-37, pupilla, netdev
In-Reply-To: <20130117065810.GH18940@secunet.com>
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Thu, 17 Jan 2013 07:58:10 +0100
> Routes with locked mtu should not use learned pmtu informations,
> so do not update the pmtu on these routes.
>
> Reported-by: Julian Anastasov <ja@ssi.bg>
> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Applied and queued up for -stable.
^ permalink raw reply
* RE: [RFC Patch net-next] tcp: add a global sysctl to control TCP delayed ack
From: Cong Wang @ 2013-01-17 9:21 UTC (permalink / raw)
To: David Laight
Cc: netdev, Eric Dumazet, Rick Jones, Stephen Hemminger,
David S. Miller, Thomas Graf
In-Reply-To: <AE90C24D6B3A694183C094C60CF0A2F6026B7110@saturn3.aculab.com>
On Wed, 2013-01-16 at 12:22 +0000, David Laight wrote:
> > According to previous discussion, it seems there is no
> > reasonable heuristics.
> >
> > Similar to TCP_QUICK_ACK option, but for people who can't
> > modify the source code and still wants to control
> > TCP delayed ACK behavior.
> >
> > Makes any sense?
>
> A sysctl is a bit of a big hammer, it probably isn't necessary
> to disable delayed acks on all connections.
You mean make this sysctl per-socket? But we don't have per-socket or
per-connection sysctl for networking, do we?
>
> IIRC the related problems I saw were really on the sending
> side when Nagle is disabled and it is doing 'slow start'.
>
> Globally disabling on connections that have Nagle disabled
> might be a possibility - but it is the Nagle parameter
> at the other end that matters.
>
> Perhaps the sending side, after sending 4 small frames immediately,
> could send 1 or 2 additional full sized frames in order to
> provoke an ack (IIRC an ack is sent if there are 2 full sized
> frames of data unacked).
>
> The other problem is that 'slow start' is restarted very
> aggressively - whenever there is no unacked data.
> If you have a very low latency connection and aren't doing
> continuous bulk transfer it is restarted for every short
> burst of transmits - effectively after every received ack.
> There really ought to have to be a moderate idle time
> before 'slow start' is restarted.
>
These situations are not easy at all to detect.
Thanks.
^ permalink raw reply
* Re: [PATCH v4 1/1] net: ethernet: davinci_cpdma: Add boundary for rx and tx descriptors
From: Mugunthan V N @ 2013-01-17 9:30 UTC (permalink / raw)
To: Christian Riesch; +Cc: netdev, davem, s.hauer
In-Reply-To: <50F7B832.6030307@omicron.at>
Christian
On 1/17/2013 2:07 PM, Christian Riesch wrote:
> Hi,
>
> On 2013-01-17 01:10, Mugunthan V N wrote:
>> When there is heavy transmission traffic in the CPDMA, then Rx
>> descriptors
>> memory is also utilized as tx desc memory looses all rx descriptors
>> and the
>> driver stops working then.
>>
>> This patch adds boundary for tx and rx descriptors in bd ram dividing
>> the
>> descriptor memory to ensure that during heavy transmission tx doesn't
>> use
>> rx descriptors.
>>
>> This patch is already applied to davinci_emac driver, since CPSW and
>> davici_dmac shares the same CPDMA, moving the boundry seperation from
>> Davinci EMAC driver to CPDMA driver which was done in the following
>> commit
>>
>> commit 86d8c07ff2448eb4e860e50f34ef6ee78e45c40c
>> Author: Sascha Hauer <s.hauer@pengutronix.de>
>> Date: Tue Jan 3 05:27:47 2012 +0000
>>
>> net/davinci: do not use all descriptors for tx packets
>>
>> The driver uses a shared pool for both rx and tx descriptors.
>> During open it queues fixed number of 128 descriptors for receive
>> packets. For each received packet it tries to queue another
>> descriptor. If this fails the descriptor is lost for rx.
>> The driver has no limitation on tx descriptors to use, so it
>> can happen during a nmap / ping -f attack that the driver
>> allocates all descriptors for tx and looses all rx descriptors.
>> The driver stops working then.
>> To fix this limit the number of tx descriptors used to half of
>> the descriptors available, the rx path uses the other half.
>>
>> Tested on a custom board using nmap / ping -f to the board from
>> two different hosts.
>>
>> Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
>> ---
>> Changes from v3
>> * Changed multiline commenting as per networking style
>>
>> drivers/net/ethernet/ti/cpsw.c | 9 ++++++
>> drivers/net/ethernet/ti/davinci_cpdma.c | 51
>> ++++++++++++++++++++++++++-----
>> drivers/net/ethernet/ti/davinci_cpdma.h | 1 +
>> drivers/net/ethernet/ti/davinci_emac.c | 13 ++++----
>> 4 files changed, 61 insertions(+), 13 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/ti/cpsw.c
>> b/drivers/net/ethernet/ti/cpsw.c
>> index 3772804..b35e6a7 100644
>> --- a/drivers/net/ethernet/ti/cpsw.c
>> +++ b/drivers/net/ethernet/ti/cpsw.c
>> @@ -374,6 +374,9 @@ void cpsw_tx_handler(void *token, int len, int
>> status)
>> struct net_device *ndev = skb->dev;
>> struct cpsw_priv *priv = netdev_priv(ndev);
>>
>> + /* Check whether the queue is stopped due to stalled tx dma, if the
>> + * queue is stopped then start the queue as we have free desc
>> for tx
>> + */
>> if (unlikely(netif_queue_stopped(ndev)))
>> netif_start_queue(ndev);
>> cpts_tx_timestamp(&priv->cpts, skb);
>> @@ -736,6 +739,12 @@ static netdev_tx_t cpsw_ndo_start_xmit(struct
>> sk_buff *skb,
>> goto fail;
>> }
>>
>> + /* If there is no more tx desc left free then we need to
>> + * tell the kernel to stop sending us tx frames.
>> + */
>> + if (unlikely(cpdma_check_free_tx_desc(priv->txch)))
>> + netif_stop_queue(ndev);
>> +
>> return NETDEV_TX_OK;
>> fail:
>> priv->stats.tx_dropped++;
>> diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c
>> b/drivers/net/ethernet/ti/davinci_cpdma.c
>> index 4995673..4f6d82f 100644
>> --- a/drivers/net/ethernet/ti/davinci_cpdma.c
>> +++ b/drivers/net/ethernet/ti/davinci_cpdma.c
>> @@ -105,13 +105,13 @@ struct cpdma_ctlr {
>> };
>>
>> struct cpdma_chan {
>> + struct cpdma_desc __iomem *head, *tail;
>> + void __iomem *hdp, *cp, *rxfree;
>> enum cpdma_state state;
>> struct cpdma_ctlr *ctlr;
>> int chan_num;
>> spinlock_t lock;
>> - struct cpdma_desc __iomem *head, *tail;
>> int count;
>> - void __iomem *hdp, *cp, *rxfree;
>
> I still do not see any benefit from this rearranging here.
This is just code cleanup and doesn't provide any benefit.
>
>> u32 mask;
>> cpdma_handler_fn handler;
>> enum dma_data_direction dir;
>> @@ -217,17 +217,29 @@ desc_from_phys(struct cpdma_desc_pool *pool,
>> dma_addr_t dma)
>> }
>>
>> static struct cpdma_desc __iomem *
>> -cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc)
>> +cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc, bool
>> is_rx)
>> {
>> unsigned long flags;
>> int index;
>> + int desc_start;
>> + int desc_end;
>> struct cpdma_desc __iomem *desc = NULL;
>>
>> spin_lock_irqsave(&pool->lock, flags);
>>
>> - index = bitmap_find_next_zero_area(pool->bitmap, pool->num_desc, 0,
>> - num_desc, 0);
>> - if (index < pool->num_desc) {
>> + if (is_rx) {
>> + desc_start = 0;
>> + desc_end = pool->num_desc/2;
>> + index = bitmap_find_next_zero_area(pool->bitmap,
>> + pool->num_desc/2, 0, num_desc, 0);
>
> Duplicate calls of bitmap_find_next_zero_area, see a few lines below.
Will fix this in next patch version
>
>> + } else {
>> + desc_start = pool->num_desc/2;
>> + desc_end = pool->num_desc;
>> + }
>> +
>> + index = bitmap_find_next_zero_area(pool->bitmap,
>> + desc_end, desc_start, num_desc, 0);
>> + if (index < desc_end) {
>> bitmap_set(pool->bitmap, index, num_desc);
>> desc = pool->iomap + pool->desc_size * index;
>> pool->used_desc++;
>> @@ -660,6 +672,7 @@ int cpdma_chan_submit(struct cpdma_chan *chan,
>> void *token, void *data,
>> unsigned long flags;
>> u32 mode;
>> int ret = 0;
>> + bool is_rx;
>>
>> spin_lock_irqsave(&chan->lock, flags);
>>
>> @@ -668,7 +681,8 @@ int cpdma_chan_submit(struct cpdma_chan *chan,
>> void *token, void *data,
>> goto unlock_ret;
>> }
>>
>> - desc = cpdma_desc_alloc(ctlr->pool, 1);
>> + is_rx = (chan->rxfree != 0);
>
> I guess you should use is_rx_chan() here.
Will change this to MACRO.
>
>> + desc = cpdma_desc_alloc(ctlr->pool, 1, is_rx);
>> if (!desc) {
>> chan->stats.desc_alloc_fail++;
>> ret = -ENOMEM;
>> @@ -704,6 +718,29 @@ unlock_ret:
>> }
>> EXPORT_SYMBOL_GPL(cpdma_chan_submit);
>>
>> +bool cpdma_check_free_tx_desc(struct cpdma_chan *chan)
>> +{
>> + unsigned long flags;
>> + int index;
>> + bool ret;
>> + struct cpdma_ctlr *ctlr = chan->ctlr;
>> + struct cpdma_desc_pool *pool = ctlr->pool;
>> +
>> + spin_lock_irqsave(&pool->lock, flags);
>> +
>> + index = bitmap_find_next_zero_area(pool->bitmap,
>> + pool->num_desc, pool->num_desc/2, 1, 0);
>
> I still wonder if using two separate descriptor pools for rx and tx
> would be a nicer implementation. In this case you could just use
> bitmap_full here, right? And the two pools would have separate
> spinlocks, so rx could not lock tx and vice versa (but as I wrote in
> an earlier email, I am a newbie here, so I don't know if this would be
> beneficial).
>
I don't think having two separate pool with separate lock will improve
performance, i have not seen any issues using one pool with one lock.
Regards
Mugunthan V N
^ permalink raw reply
* [PATCH, resubmit] ax88179_178a: ASIX AX88179_178A USB 3.0/2.0 to gigabit ethernet adapter driver
From: Freddy Xin @ 2013-01-17 9:32 UTC (permalink / raw)
To: netdev; +Cc: linux-usb, linux-kernel, louis, davem, Freddy Xin
This is a resubmission.
The _nopm version usb functions were added to access register in suspend and resume functions.
Serveral variables allocted dynamically were removed and replaced by stack variables.
ax88179_get_eeprom were modified from asix_get_eeprom in asix_common.
This patch adds a driver for ASIX's AX88179 family of USB 3.0/2.0
to gigabit ethernet adapters. It's based on the AX88xxx driver but
the usb commands used to access registers for AX88179 are completely different.
This driver had been verified on x86 system with AX88179/AX88178A and
Sitcomm LN-032 USB dongles.
Signed-off-by: Freddy Xin <freddy@asix.com.tw>
---
drivers/net/usb/Kconfig | 18 +
drivers/net/usb/Makefile | 1 +
drivers/net/usb/ax88179_178a.c | 1424 ++++++++++++++++++++++++++++++++++++++++
3 files changed, 1443 insertions(+)
create mode 100644 drivers/net/usb/ax88179_178a.c
diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig
index ef97621..75af401 100644
--- a/drivers/net/usb/Kconfig
+++ b/drivers/net/usb/Kconfig
@@ -158,6 +158,24 @@ config USB_NET_AX8817X
This driver creates an interface named "ethX", where X depends on
what other networking devices you have in use.
+config USB_NET_AX88179_178A
+ tristate "ASIX AX88179/178A USB 3.0/2.0 to Gigabit Ethernet"
+ depends on USB_USBNET
+ select CRC32
+ select PHYLIB
+ default y
+ help
+ This option adds support for ASIX AX88179 based USB 3.0/2.0
+ to Gigabit Ethernet adapters.
+
+ This driver should work with at least the following devices:
+ * ASIX AX88179
+ * ASIX AX88178A
+ * Sitcomm LN-032
+
+ This driver creates an interface named "ethX", where X depends on
+ what other networking devices you have in use.
+
config USB_NET_CDCETHER
tristate "CDC Ethernet support (smart devices such as cable modems)"
depends on USB_USBNET
diff --git a/drivers/net/usb/Makefile b/drivers/net/usb/Makefile
index 4786913..ce01578 100644
--- a/drivers/net/usb/Makefile
+++ b/drivers/net/usb/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_USB_RTL8150) += rtl8150.o
obj-$(CONFIG_USB_HSO) += hso.o
obj-$(CONFIG_USB_NET_AX8817X) += asix.o
asix-y := asix_devices.o asix_common.o ax88172a.o
+obj-$(CONFIG_USB_NET_AX88179_178A) += ax88179_178a.o
obj-$(CONFIG_USB_NET_CDCETHER) += cdc_ether.o
obj-$(CONFIG_USB_NET_CDC_EEM) += cdc_eem.o
obj-$(CONFIG_USB_NET_DM9601) += dm9601.o
diff --git a/drivers/net/usb/ax88179_178a.c b/drivers/net/usb/ax88179_178a.c
new file mode 100644
index 0000000..4aa9f28
--- /dev/null
+++ b/drivers/net/usb/ax88179_178a.c
@@ -0,0 +1,1424 @@
+/*
+ * ASIX AX88179/178A USB 3.0/2.0 to Gigabit Ethernet Devices
+ *
+ * Copyright (C) 20011-2012 ASIX
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/etherdevice.h>
+#include <linux/mii.h>
+#include <linux/usb.h>
+#include <linux/crc32.h>
+#include <linux/usb/usbnet.h>
+
+
+#define AX88179_PHY_ID 0x03
+#define AX_EEPROM_LEN 0x100
+#define AX_MCAST_FILTER_SIZE 8
+#define AX_MAX_MCAST 64
+#define AX_INT_PPLS_LINK BIT(0)
+#define AX_RXHDR_L4_TYPE_UDP 1
+#define AX_RXHDR_L4_TYPE_TCP 4
+#define AX_ACCESS_MAC 0x01
+#define AX_ACCESS_PHY 0x02
+#define AX_ACCESS_EEPROM 0x04
+#define AX_ACCESS_EFUSE 0x05
+#define AX_PAUSE_WATERLVL_HIGH 0x54
+#define AX_PAUSE_WATERLVL_LOW 0x55
+
+#define PHYSICAL_LINK_STATUS 0x02
+ #define AX_USB_SS 0x04
+ #define AX_USB_HS 0x02
+
+#define GENERAL_STATUS 0x03
+/* Check AX88179 version. UA1:Bit2 = 0, UA2:Bit2 = 1 */
+ #define AX_SECLD 0x04
+
+#define AX_SROM_ADDR 0x07
+#define AX_SROM_CMD 0x0a
+ #define EEP_RD 0x04
+ #define EEP_BUSY 0x10
+
+
+#define AX_SROM_DATA_LOW 0x08
+#define AX_SROM_DATA_HIGH 0x09
+
+#define AX_RX_CTL 0x0b
+ #define AX_RX_CTL_DROPCRCERR 0x0100
+ #define AX_RX_CTL_IPE 0x0200
+ #define AX_RX_CTL_START 0x0080
+ #define AX_RX_CTL_AP 0x0020
+ #define AX_RX_CTL_AM 0x0010
+ #define AX_RX_CTL_AB 0x0008
+ #define AX_RX_CTL_AMALL 0x0002
+ #define AX_RX_CTL_PRO 0x0001
+ #define AX_RX_CTL_STOP 0x0000
+
+#define AX_NODE_ID 0x10
+#define AX_MULTI_FILTER_ARRY 0x16
+
+#define AX_MEDIUM_STATUS_MODE 0x22
+ #define AX_MEDIUM_GIGAMODE 0x01
+ #define AX_MEDIUM_FULL_DUPLEX 0x02
+ #define AX_MEDIUM_ALWAYS_ONE 0x04
+ #define AX_MEDIUM_RXFLOW_CTRLEN 0x10
+ #define AX_MEDIUM_TXFLOW_CTRLEN 0x20
+ #define AX_MEDIUM_RECEIVE_EN 0x100
+ #define AX_MEDIUM_PS 0x200
+
+#define AX_MONITOR_MODE 0x24
+ #define AX_MONITOR_MODE_RWLC 0x02
+ #define AX_MONITOR_MODE_RWMP 0x04
+ #define AX_MONITOR_MODE_PMEPOL 0x20
+ #define AX_MONITOR_MODE_PMETYPE 0x40
+
+#define AX_GPIO_CTRL 0x25
+ #define AX_GPIO_CTRL_GPIO3EN 0x80
+ #define AX_GPIO_CTRL_GPIO2EN 0x40
+ #define AX_GPIO_CTRL_GPIO1EN 0x20
+
+#define AX_PHYPWR_RSTCTL 0x26
+ #define AX_PHYPWR_RSTCTL_BZ 0x0010
+ #define AX_PHYPWR_RSTCTL_IPRL 0x0020
+ #define AX_PHYPWR_RSTCTL_AT 0x1000
+
+#define AX_RX_BULKIN_QCTRL 0x2e
+#define AX_CLK_SELECT 0x33
+ #define AX_CLK_SELECT_BCS 0x01
+ #define AX_CLK_SELECT_ACS 0x02
+ #define AX_CLK_SELECT_ULR 0x08
+
+#define AX_RXCOE_CTL 0x34
+ #define AX_RXCOE_IP 0x01
+ #define AX_RXCOE_TCP 0x02
+ #define AX_RXCOE_UDP 0x04
+ #define AX_RXCOE_TCPV6 0x20
+ #define AX_RXCOE_UDPV6 0x40
+
+#define AX_TXCOE_CTL 0x35
+ #define AX_TXCOE_IP 0x01
+ #define AX_TXCOE_TCP 0x02
+ #define AX_TXCOE_UDP 0x04
+ #define AX_TXCOE_TCPV6 0x20
+ #define AX_TXCOE_UDPV6 0x40
+
+#define AX_LEDCTRL 0x73
+
+#define GMII_PHY_PHYSR 0x11
+ #define GMII_PHY_PHYSR_SMASK 0xc000
+ #define GMII_PHY_PHYSR_GIGA 0x8000
+ #define GMII_PHY_PHYSR_100 0x4000
+ #define GMII_PHY_PHYSR_FULL 0x2000
+ #define GMII_PHY_PHYSR_LINK 0x400
+
+#define GMII_LED_ACTIVE 0x1a
+ #define GMII_LED_ACTIVE_MASK 0xff8fd
+ #define GMII_LED0_ACTIVE BIT(4)
+ #define GMII_LED1_ACTIVE BIT(5)
+ #define GMII_LED2_ACTIVE BIT(6)
+
+#define GMII_LED_LINK 0x1c
+ #define GMII_LED_LINK_MASK 0xf888
+
+ #define GMII_LED0_LINK_10 BIT(0)
+ #define GMII_LED0_LINK_100 BIT(1)
+ #define GMII_LED0_LINK_1000 BIT(2)
+ #define GMII_LED1_LINK_10 BIT(4)
+ #define GMII_LED1_LINK_100 BIT(5)
+ #define GMII_LED1_LINK_1000 BIT(6)
+ #define GMII_LED2_LINK_10 BIT(8)
+ #define GMII_LED2_LINK_100 BIT(9)
+ #define GMII_LED2_LINK_1000 BIT(10)
+
+ #define LED0_ACTIVE BIT(0)
+ #define LED0_LINK_10 BIT(1)
+ #define LED0_LINK_100 BIT(2)
+ #define LED0_LINK_1000 BIT(3)
+ #define LED0_FD BIT(4)
+ #define LED0_USB3_MASK 0x001f
+
+ #define LED1_ACTIVE BIT(5)
+ #define LED1_LINK_10 BIT(6)
+ #define LED1_LINK_100 BIT(7)
+ #define LED1_LINK_1000 BIT(8)
+ #define LED1_FD BIT(9)
+ #define LED1_USB3_MASK 0x03e0
+
+ #define LED2_ACTIVE BIT(10)
+ #define LED2_LINK_1000 BIT(13)
+ #define LED2_LINK_100 BIT(12)
+ #define LED2_LINK_10 BIT(11)
+ #define LED2_FD BIT(14)
+ #define LED_VALID BIT(15)
+ #define LED2_USB3_MASK 0x7c00
+
+#define GMII_PHYPAGE 0x1e
+#define GMII_PHY_PAGE_SELECT 0x1f
+ #define GMII_PHY_PGSEL_EXT 0x0007
+ #define GMII_PHY_PGSEL_PAGE0 0X0000
+
+#define AX88179_EEPROM_MAGIC 0x17900b95
+
+struct ax88179_data {
+ u16 rxctl;
+};
+
+struct ax88179_int_data {
+ u16 res1;
+ u8 link;
+ u16 res2;
+ u8 status;
+ u16 res3;
+} __packed;
+
+struct {unsigned char ctrl, timer_l, timer_h, size, ifg; }
+AX88179_BULKIN_SIZE[] = {
+ {7, 0xa3, 0, 0x14, 0xff},
+ {7, 0x1e, 5, 0x18, 0xff},
+ {7, 0xae, 7, 0x18, 0xff},
+ {7, 0xcc, 0x4c, 0x18, 8},
+};
+
+struct ax88179_rx_pkt_header {
+
+ u8 l4_csum_err:1,
+ l3_csum_err:1,
+ l4_type:3,
+ l3_type:2,
+ ce:1;
+
+ u8 vlan_ind:3,
+ rx_ok:1,
+ pri:3,
+ bmc:1;
+
+ u16 len:13,
+ crc:1,
+ mii:1,
+ drop:1;
+
+} __packed;
+
+static int
+__ax88179_read_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+ u16 size, void *data, int in_pm)
+{
+ int ret;
+ int (*fn)(struct usbnet *, u8, u8, u16, u16, void *, u16);
+
+ if (!in_pm)
+ fn = usbnet_read_cmd;
+ else
+ fn = usbnet_read_cmd_nopm;
+
+ ret = fn(dev, cmd, USB_DIR_IN | USB_TYPE_VENDOR |
+ USB_RECIP_DEVICE, value, index, data, size);
+
+ if (ret != size && ret >= 0)
+ return -EINVAL;
+
+ return ret;
+}
+
+static int
+__ax88179_write_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+ u16 size, void *data, int in_pm)
+{
+ int (*fn)(struct usbnet *, u8, u8, u16, u16, const void *, u16);
+
+ if (!in_pm)
+ fn = usbnet_write_cmd;
+ else
+ fn = usbnet_write_cmd_nopm;
+
+ return fn(dev, cmd, USB_DIR_OUT | USB_TYPE_VENDOR |
+ USB_RECIP_DEVICE, value, index, data, size);
+
+}
+
+static void
+ax88179_write_cmd_async(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+ u16 size, void *data)
+{
+ usbnet_write_cmd_async(dev, cmd,
+ USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+ value, index, data, size);
+}
+
+static int
+ax88179_read_cmd_nopm(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+ u16 size, void *data)
+{
+ return __ax88179_read_cmd(dev, cmd, value, index, size, data, 1);
+}
+
+static int
+ax88179_write_cmd_nopm(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+ u16 size, void *data)
+{
+ return __ax88179_write_cmd(dev, cmd, value, index, size, data, 1);
+}
+
+static int
+ax88179_read_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+ u16 size, void *data)
+{
+ return __ax88179_read_cmd(dev, cmd, value, index, size, data, 0);
+}
+
+static int
+ax88179_write_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+ u16 size, void *data)
+{
+ return __ax88179_write_cmd(dev, cmd, value, index, size, data, 0);
+}
+
+static void ax88179_status(struct usbnet *dev, struct urb *urb)
+{
+ struct ax88179_int_data *event;
+ int link;
+
+ if (urb->actual_length < 8)
+ return;
+
+ event = urb->transfer_buffer;
+ link = event->link & AX_INT_PPLS_LINK;
+ if (netif_carrier_ok(dev->net) != link) {
+ if (link)
+ usbnet_defer_kevent(dev, EVENT_LINK_RESET);
+ else
+ netif_carrier_off(dev->net);
+ netdev_info(dev->net, "ax88179 - Link status is: %d\n", link);
+ }
+}
+
+static int ax88179_mdio_read(struct net_device *netdev, int phy_id, int loc)
+{
+ struct usbnet *dev = netdev_priv(netdev);
+ __u16 res;
+ u16 ret;
+
+ ax88179_read_cmd(dev, AX_ACCESS_PHY, phy_id, (__u16)loc, 2, &res);
+
+ ret = res & 0xffff;
+
+ return le16_to_cpu(ret);
+}
+
+static void
+ax88179_mdio_write(struct net_device *netdev, int phy_id, int loc, int val)
+{
+ struct usbnet *dev = netdev_priv(netdev);
+ __le16 res;
+
+ res = cpu_to_le16(val);
+
+ ax88179_write_cmd(dev, AX_ACCESS_PHY, phy_id, (__u16)loc, 2, &res);
+
+}
+
+static int ax88179_suspend(struct usb_interface *intf, pm_message_t message)
+{
+ struct usbnet *dev = usb_get_intfdata(intf);
+ u16 tmp16;
+
+ usbnet_suspend(intf, message);
+
+ /* Disable RX path */
+ ax88179_read_cmd_nopm(dev, AX_ACCESS_MAC,
+ AX_MEDIUM_STATUS_MODE, 2, 2, &tmp16);
+ tmp16 &= ~AX_MEDIUM_RECEIVE_EN;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC,
+ AX_MEDIUM_STATUS_MODE, 2, 2, &tmp16);
+
+ /* Force bulk-in zero length */
+ ax88179_read_cmd_nopm(dev, AX_ACCESS_MAC,
+ AX_PHYPWR_RSTCTL, 2, 2, &tmp16);
+
+ tmp16 |= AX_PHYPWR_RSTCTL_BZ | AX_PHYPWR_RSTCTL_IPRL;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC,
+ AX_PHYPWR_RSTCTL, 2, 2, &tmp16);
+
+ /* change clock */
+ tmp16 = 0;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC, AX_CLK_SELECT, 1, 1, &tmp16);
+
+ /* Configure RX control register => stop operation */
+ tmp16 = AX_RX_CTL_STOP;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC, AX_RX_CTL, 2, 2, &tmp16);
+
+ return 0;
+}
+
+/* This function is used to enable the autodetach function. */
+/* This function is determined by offset 0x43 of EEPROM */
+static int ax88179_auto_detach(struct usbnet *dev, int in_pm)
+{
+ u16 tmp;
+
+ if (__ax88179_read_cmd(dev, AX_ACCESS_EEPROM, 0x43,
+ 1, 2, &tmp, in_pm) < 0)
+ return 0;
+
+ if ((tmp == 0xFFFF) || (!(tmp & 0x0100)))
+ return 0;
+
+ /* Enable Auto Detach bit */
+ tmp = 0;
+ __ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_CLK_SELECT,
+ 1, 1, &tmp, in_pm);
+ tmp |= AX_CLK_SELECT_ULR;
+ __ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_CLK_SELECT,
+ 1, 1, &tmp, in_pm);
+
+ __ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_PHYPWR_RSTCTL,
+ 2, 2, &tmp, in_pm);
+ tmp |= AX_PHYPWR_RSTCTL_AT;
+ __ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PHYPWR_RSTCTL,
+ 2, 2, &tmp, in_pm);
+
+ return 0;
+}
+
+static int ax88179_resume(struct usb_interface *intf)
+{
+ struct usbnet *dev = usb_get_intfdata(intf);
+ u16 tmp16;
+
+ netif_carrier_off(dev->net);
+
+ /* Power up ethernet PHY */
+ tmp16 = 0;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC,
+ AX_PHYPWR_RSTCTL, 2, 2, &tmp16);
+ udelay(1000);
+
+ tmp16 = AX_PHYPWR_RSTCTL_IPRL;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC,
+ AX_PHYPWR_RSTCTL, 2, 2, &tmp16);
+ msleep(200);
+
+ /* Ethernet PHY Auto Detach*/
+ ax88179_auto_detach(dev, 1);
+
+ /* enable clock */
+ ax88179_read_cmd_nopm(dev, AX_ACCESS_MAC, AX_CLK_SELECT, 1, 1, &tmp16);
+ tmp16 |= AX_CLK_SELECT_ACS | AX_CLK_SELECT_BCS;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC, AX_CLK_SELECT, 1, 1, &tmp16);
+ msleep(100);
+
+ /* Configure RX control register => start operation */
+ tmp16 = AX_RX_CTL_DROPCRCERR | AX_RX_CTL_IPE | AX_RX_CTL_START |
+ AX_RX_CTL_AP | AX_RX_CTL_AMALL | AX_RX_CTL_AB;
+ ax88179_write_cmd_nopm(dev, AX_ACCESS_MAC, AX_RX_CTL, 2, 2, &tmp16);
+
+ return usbnet_resume(intf);
+}
+
+static void
+ax88179_get_wol(struct net_device *net, struct ethtool_wolinfo *wolinfo)
+{
+ struct usbnet *dev = netdev_priv(net);
+ u8 opt;
+
+ wolinfo->supported = 0;
+ wolinfo->wolopts = 0;
+
+ if (ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_MONITOR_MODE, 1, 1, &opt) < 0)
+ return;
+
+ wolinfo->supported = WAKE_PHY | WAKE_MAGIC;
+
+ if (opt & AX_MONITOR_MODE_RWLC)
+ wolinfo->wolopts |= WAKE_PHY;
+ if (opt & AX_MONITOR_MODE_RWMP)
+ wolinfo->wolopts |= WAKE_MAGIC;
+
+}
+
+static int
+ax88179_set_wol(struct net_device *net, struct ethtool_wolinfo *wolinfo)
+{
+ struct usbnet *dev = netdev_priv(net);
+ u8 opt;
+
+ opt = 0;
+ if (ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_MONITOR_MODE, 1, 1, &opt) < 0)
+ return 0;
+
+ if (wolinfo->wolopts & WAKE_PHY)
+ opt |= AX_MONITOR_MODE_RWLC;
+ else
+ opt &= ~AX_MONITOR_MODE_RWLC;
+
+ if (wolinfo->wolopts & WAKE_MAGIC)
+ opt |= AX_MONITOR_MODE_RWMP;
+ else
+ opt &= ~AX_MONITOR_MODE_RWMP;
+
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_MONITOR_MODE, 1, 1, &opt);
+
+ return 0;
+}
+
+static int ax88179_get_eeprom_len(struct net_device *net)
+{
+ return AX_EEPROM_LEN;
+}
+
+int ax88179_get_eeprom(struct net_device *net, struct ethtool_eeprom *eeprom,
+ u8 *data)
+{
+ struct usbnet *dev = netdev_priv(net);
+ u16 *eeprom_buff;
+ int first_word, last_word;
+ int i;
+
+ if (eeprom->len == 0)
+ return -EINVAL;
+
+ eeprom->magic = AX88179_EEPROM_MAGIC;
+
+ first_word = eeprom->offset >> 1;
+ last_word = (eeprom->offset + eeprom->len - 1) >> 1;
+
+ eeprom_buff = kmalloc(sizeof(u16) * (last_word - first_word + 1),
+ GFP_KERNEL);
+ if (!eeprom_buff)
+ return -ENOMEM;
+
+ /* ax88179/178A returns 2 bytes from eeprom on read */
+ for (i = first_word; i <= last_word; i++) {
+ if (ax88179_read_cmd(dev, AX_ACCESS_EEPROM,
+ i, 1, 2, &(eeprom_buff[i - first_word])) < 0) {
+ kfree(eeprom_buff);
+ return -EIO;
+ }
+ }
+
+ memcpy(data, (u8 *)eeprom_buff + (eeprom->offset & 1), eeprom->len);
+ kfree(eeprom_buff);
+ return 0;
+}
+
+static int ax88179_get_settings(struct net_device *net, struct ethtool_cmd *cmd)
+{
+ struct usbnet *dev = netdev_priv(net);
+ return mii_ethtool_gset(&dev->mii, cmd);
+}
+
+static int ax88179_set_settings(struct net_device *net, struct ethtool_cmd *cmd)
+{
+ struct usbnet *dev = netdev_priv(net);
+ return mii_ethtool_sset(&dev->mii, cmd);
+}
+
+
+static int ax88179_ioctl(struct net_device *net, struct ifreq *rq, int cmd)
+{
+ struct usbnet *dev = netdev_priv(net);
+ return generic_mii_ioctl(&dev->mii, if_mii(rq), cmd, NULL);
+}
+
+static struct ethtool_ops ax88179_ethtool_ops = {
+ .get_link = ethtool_op_get_link,
+ .get_msglevel = usbnet_get_msglevel,
+ .set_msglevel = usbnet_set_msglevel,
+ .get_wol = ax88179_get_wol,
+ .set_wol = ax88179_set_wol,
+ .get_eeprom_len = ax88179_get_eeprom_len,
+ .get_eeprom = ax88179_get_eeprom,
+ .get_settings = ax88179_get_settings,
+ .set_settings = ax88179_set_settings,
+ .nway_reset = usbnet_nway_reset,
+};
+
+static void ax88179_set_multicast(struct net_device *net)
+{
+ struct usbnet *dev = netdev_priv(net);
+ struct ax88179_data *data = (struct ax88179_data *)dev->data;
+ u8 *m_filter = ((u8 *)dev->data) + 12;
+ u16 rx_ctl = (AX_RX_CTL_START | AX_RX_CTL_AB | AX_RX_CTL_IPE);
+
+ if (net->flags & IFF_PROMISC) {
+ rx_ctl |= AX_RX_CTL_PRO;
+ } else if (net->flags & IFF_ALLMULTI
+ || netdev_mc_count(net) > AX_MAX_MCAST) {
+ rx_ctl |= AX_RX_CTL_AMALL;
+ } else if (netdev_mc_empty(net)) {
+ /* just broadcast and directed */
+ } else {
+ /* We use the 20 byte dev->data for our 8 byte filter buffer
+ * to avoid allocating memory that is tricky to free later
+ */
+ u32 crc_bits;
+ struct netdev_hw_addr *ha;
+
+ memset(m_filter, 0, AX_MCAST_FILTER_SIZE);
+
+ netdev_for_each_mc_addr(ha, net) {
+ crc_bits = ether_crc(ETH_ALEN, ha->addr) >> 26;
+ *(m_filter + (crc_bits >> 3)) |=
+ 1 << (crc_bits & 7);
+ }
+
+ ax88179_write_cmd_async(dev, AX_ACCESS_MAC,
+ AX_MULTI_FILTER_ARRY, AX_MCAST_FILTER_SIZE,
+ AX_MCAST_FILTER_SIZE, m_filter);
+
+ rx_ctl |= AX_RX_CTL_AM;
+ }
+
+ data->rxctl = rx_ctl;
+ ax88179_write_cmd_async(dev, AX_ACCESS_MAC,
+ AX_RX_CTL, 2, 2, &data->rxctl);
+}
+
+static int
+ax88179_set_features(struct net_device *net, netdev_features_t features)
+{
+ u8 tmp;
+ struct usbnet *dev = netdev_priv(net);
+ netdev_features_t changed = net->features ^ features;
+
+ if (changed & NETIF_F_TSO)
+ net->features ^= NETIF_F_TSO;
+
+ if (changed & NETIF_F_SG)
+ net->features ^= NETIF_F_SG;
+
+ if (changed & NETIF_F_IP_CSUM) {
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
+
+ tmp ^= AX_TXCOE_TCP | AX_TXCOE_UDP;
+
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
+
+ net->features ^= NETIF_F_IP_CSUM;
+ }
+
+ if (changed & NETIF_F_IPV6_CSUM) {
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
+
+ tmp ^= AX_TXCOE_TCPV6 | AX_TXCOE_UDPV6;
+
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, &tmp);
+
+ net->features ^= NETIF_F_IPV6_CSUM;
+ }
+
+ if (changed & NETIF_F_RXCSUM) {
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_RXCOE_CTL, 1, 1, &tmp);
+
+ tmp ^= AX_RXCOE_IP | AX_RXCOE_TCP | AX_RXCOE_UDP |
+ AX_RXCOE_TCPV6 | AX_RXCOE_UDPV6;
+
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_RXCOE_CTL, 1, 1, &tmp);
+
+ net->features ^= NETIF_F_RXCSUM;
+ }
+
+ return 0;
+}
+
+static int ax88179_set_mac_addr(struct net_device *net, void *p)
+{
+ struct usbnet *dev = netdev_priv(net);
+ struct sockaddr *addr = p;
+
+ if (netif_running(net))
+ return -EBUSY;
+ if (!is_valid_ether_addr(addr->sa_data))
+ return -EADDRNOTAVAIL;
+
+ memcpy(net->dev_addr, addr->sa_data, ETH_ALEN);
+
+ /* Set the MAC address */
+ return ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_NODE_ID, ETH_ALEN,
+ ETH_ALEN, net->dev_addr);
+}
+
+static const struct net_device_ops ax88179_netdev_ops = {
+ .ndo_open = usbnet_open,
+ .ndo_stop = usbnet_stop,
+ .ndo_start_xmit = usbnet_start_xmit,
+ .ndo_tx_timeout = usbnet_tx_timeout,
+ .ndo_change_mtu = usbnet_change_mtu,
+ .ndo_set_mac_address = ax88179_set_mac_addr,
+ .ndo_validate_addr = eth_validate_addr,
+ .ndo_do_ioctl = ax88179_ioctl,
+ .ndo_set_rx_mode = ax88179_set_multicast,
+ .ndo_set_features = ax88179_set_features,
+};
+
+static int ax88179_check_eeprom(struct usbnet *dev)
+{
+ u8 i;
+ u8 buf;
+ u8 eeprom[20];
+ u16 csum, delay = HZ / 10;
+ unsigned long jtimeout;
+
+ /* Read EEPROM content */
+ for (i = 0; i < 6; i++) {
+ buf = i;
+ if (ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_ADDR, 1, 1, &buf) < 0)
+ return -EINVAL;
+
+ buf = EEP_RD;
+ if (ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_CMD, 1, 1, &buf) < 0)
+ return -EINVAL;
+
+ jtimeout = jiffies + delay;
+ do {
+ ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_CMD, 1, 1, &buf);
+
+ if (time_after(jiffies, jtimeout))
+ return -EINVAL;
+
+ } while (buf & EEP_BUSY);
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_DATA_LOW, 2, 2, &eeprom[i * 2]);
+
+ if ((i == 0) && (eeprom[0] == 0xFF))
+ return -EINVAL;
+ }
+
+ csum = eeprom[6] + eeprom[7] + eeprom[8] + eeprom[9];
+ csum = (csum >> 8) + (csum & 0xff);
+
+ if ((csum + eeprom[10]) != 0xff)
+ return -EINVAL;
+
+ return 0;
+}
+
+static int ax88179_check_efuse(struct usbnet *dev, u16 *ledmode)
+{
+ u8 i;
+ u8 efuse[64];
+ u16 csum = 0;
+
+ if (ax88179_read_cmd(dev, AX_ACCESS_EFUSE, 0, 64, 64, efuse) < 0)
+ return -EINVAL;
+
+ if (*efuse == 0xFF)
+ return -EINVAL;
+
+ for (i = 0; i < 64; i++)
+ csum = csum + efuse[i];
+
+ while (csum > 255)
+ csum = (csum & 0x00FF) + ((csum >> 8) & 0x00FF);
+
+ if (csum != 0xFF)
+ return -EINVAL;
+
+ *ledmode = (efuse[51] << 8) | efuse[52];
+
+ return 0;
+
+}
+
+static int ax88179_convert_old_led(struct usbnet *dev, u16 *ledvalue)
+{
+ u16 led;
+
+ /* Loaded the old eFuse LED Mode */
+ if (ax88179_read_cmd(dev, AX_ACCESS_EEPROM, 0x3C, 1, 2, &led) < 0)
+ return -EINVAL;
+
+ led >>= 8;
+
+ switch (led) {
+ case 0xFF:
+ led = LED0_ACTIVE | LED1_LINK_10 | LED1_LINK_100 |
+ LED1_LINK_1000 | LED2_ACTIVE | LED2_LINK_10 |
+ LED2_LINK_100 | LED2_LINK_1000 | LED_VALID;
+ break;
+ case 0xFE:
+ led = LED0_ACTIVE | LED1_LINK_1000 | LED2_LINK_100 | LED_VALID;
+ break;
+ case 0xFD:
+ led = LED0_ACTIVE | LED1_LINK_1000 | LED2_LINK_100 |
+ LED2_LINK_10 | LED_VALID;
+ break;
+ case 0xFC:
+ led = LED0_ACTIVE | LED1_ACTIVE | LED1_LINK_1000 | LED2_ACTIVE |
+ LED2_LINK_100 | LED2_LINK_10 | LED_VALID;
+ break;
+ default:
+ led = LED0_ACTIVE | LED1_LINK_10 | LED1_LINK_100 |
+ LED1_LINK_1000 | LED2_ACTIVE | LED2_LINK_10 |
+ LED2_LINK_100 | LED2_LINK_1000 | LED_VALID;
+ break;
+ }
+
+ *ledvalue = led;
+
+ return 0;
+}
+
+static int ax88179_led_setting(struct usbnet *dev)
+{
+ u8 ledfd, value = 0;
+ u16 tmp, ledact, ledlink, ledvalue = 0, delay = HZ / 10;
+ unsigned long jtimeout;
+
+ /* Check AX88179 version. UA1 or UA2*/
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, GENERAL_STATUS, 1, 1, &value);
+
+ if (!(value & AX_SECLD)) { /* UA1 */
+ value = AX_GPIO_CTRL_GPIO3EN | AX_GPIO_CTRL_GPIO2EN |
+ AX_GPIO_CTRL_GPIO1EN;
+ if (ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_GPIO_CTRL, 1, 1, &value) < 0)
+ return -EINVAL;
+ }
+
+ /* Check EEPROM */
+ if (!ax88179_check_eeprom(dev)) {
+ value = 0x42;
+ if (ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_ADDR, 1, 1, &value) < 0)
+ return -EINVAL;
+
+ value = EEP_RD;
+ if (ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_CMD, 1, 1, &value) < 0)
+ return -EINVAL;
+
+ jtimeout = jiffies + delay;
+ do {
+ ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_CMD, 1, 1, &value);
+
+ if (time_after(jiffies, jtimeout))
+ return -EINVAL;
+ } while (value & EEP_BUSY);
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_DATA_HIGH, 1, 1, &value);
+ ledvalue = (value << 8);
+ ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_SROM_DATA_LOW, 1, 1, &value);
+ ledvalue |= value;
+
+ /* load internal ROM for defaule setting */
+ if ((ledvalue == 0xFFFF) || ((ledvalue & LED_VALID) == 0))
+ ax88179_convert_old_led(dev, &ledvalue);
+
+ }
+
+ /* Check eFuse */
+ else if (!ax88179_check_efuse(dev, &ledvalue)) {
+ if ((ledvalue == 0xFFFF) || ((ledvalue & LED_VALID) == 0))
+ ax88179_convert_old_led(dev, &ledvalue);
+ } else
+ ax88179_convert_old_led(dev, &ledvalue);
+
+ tmp = GMII_PHY_PGSEL_EXT;
+ ax88179_write_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_PHY_PAGE_SELECT, 2, &tmp);
+
+ tmp = 0x2c;
+ ax88179_write_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_PHYPAGE, 2, &tmp);
+
+ ax88179_read_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_LED_ACTIVE, 2, &ledact);
+
+ ax88179_read_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_LED_LINK, 2, &ledlink);
+
+ ledact &= GMII_LED_ACTIVE_MASK;
+ ledlink &= GMII_LED_LINK_MASK;
+
+
+ if (ledvalue & LED0_ACTIVE)
+ ledact |= GMII_LED0_ACTIVE;
+
+ if (ledvalue & LED1_ACTIVE)
+ ledact |= GMII_LED1_ACTIVE;
+
+ if (ledvalue & LED2_ACTIVE)
+ ledact |= GMII_LED2_ACTIVE;
+
+ if (ledvalue & LED0_LINK_10)
+ ledlink |= GMII_LED0_LINK_10;
+
+ if (ledvalue & LED1_LINK_10)
+ ledlink |= GMII_LED1_LINK_10;
+
+ if (ledvalue & LED2_LINK_10)
+ ledlink |= GMII_LED2_LINK_10;
+
+ if (ledvalue & LED0_LINK_100)
+ ledlink |= GMII_LED0_LINK_100;
+
+ if (ledvalue & LED1_LINK_100)
+ ledlink |= GMII_LED1_LINK_100;
+
+ if (ledvalue & LED2_LINK_100)
+ ledlink |= GMII_LED2_LINK_100;
+
+ if (ledvalue & LED0_LINK_1000)
+ ledlink |= GMII_LED0_LINK_1000;
+
+ if (ledvalue & LED1_LINK_1000)
+ ledlink |= GMII_LED1_LINK_1000;
+
+ if (ledvalue & LED2_LINK_1000)
+ ledlink |= GMII_LED2_LINK_1000;
+
+
+ tmp = ledact;
+ ax88179_write_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_LED_ACTIVE, 2, &tmp);
+
+ tmp = ledlink;
+ ax88179_write_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_LED_LINK, 2, &tmp);
+
+ tmp = GMII_PHY_PGSEL_PAGE0;
+ ax88179_write_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_PHY_PAGE_SELECT, 2, &tmp);
+
+ /* LED full duplex setting */
+ ledfd = 0;
+ if (ledvalue & LED0_FD)
+ ledfd |= 0x01;
+ else if ((ledvalue & LED0_USB3_MASK) == 0)
+ ledfd |= 0x02;
+
+ if (ledvalue & LED1_FD)
+ ledfd |= 0x04;
+ else if ((ledvalue & LED1_USB3_MASK) == 0)
+ ledfd |= 0x08;
+
+ if (ledvalue & LED2_FD)
+ ledfd |= 0x10;
+ else if ((ledvalue & LED2_USB3_MASK) == 0)
+ ledfd |= 0x20;
+
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_LEDCTRL, 1, 1, &ledfd);
+ return 0;
+}
+
+static int ax88179_bind(struct usbnet *dev, struct usb_interface *intf)
+{
+ u8 buf[5];
+ u16 *tmp16;
+ u8 *tmp;
+ struct ax88179_data *ax179_data =
+ (struct ax88179_data *)dev->data;
+
+ usbnet_get_endpoints(dev, intf);
+
+ tmp16 = (u16 *)buf;
+ tmp = (u8 *)buf;
+
+ memset(ax179_data, 0, sizeof(*ax179_data));
+
+ /* Power up ethernet PHY */
+ *tmp16 = 0;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PHYPWR_RSTCTL, 2, 2, tmp16);
+ *tmp16 = AX_PHYPWR_RSTCTL_IPRL;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PHYPWR_RSTCTL, 2, 2, tmp16);
+ msleep(200);
+
+ *tmp = AX_CLK_SELECT_ACS | AX_CLK_SELECT_BCS;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_CLK_SELECT, 1, 1, tmp);
+ msleep(100);
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_NODE_ID, ETH_ALEN, ETH_ALEN,
+ dev->net->dev_addr);
+ memcpy(dev->net->perm_addr, dev->net->dev_addr, ETH_ALEN);
+
+ /* RX bulk configuration */
+ memcpy(tmp, &AX88179_BULKIN_SIZE[0], 5);
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_RX_BULKIN_QCTRL, 5, 5, tmp);
+ dev->rx_urb_size = 1024 * 20;
+
+ *tmp = 0x34;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PAUSE_WATERLVL_LOW, 1, 1, tmp);
+
+ *tmp = 0x52;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_PAUSE_WATERLVL_HIGH, 1, 1, tmp);
+
+ dev->net->hard_header_len += 12;
+ dev->net->netdev_ops = &ax88179_netdev_ops;
+ dev->net->ethtool_ops = &ax88179_ethtool_ops;
+ dev->net->needed_headroom = 8;
+
+ /* Initialize MII structure */
+ dev->mii.dev = dev->net;
+ dev->mii.mdio_read = ax88179_mdio_read;
+ dev->mii.mdio_write = ax88179_mdio_write;
+ dev->mii.phy_id_mask = 0xff;
+ dev->mii.reg_num_mask = 0xff;
+ dev->mii.phy_id = 0x03;
+ dev->mii.supports_gmii = 1;
+
+ dev->net->features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+ NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_TSO;
+
+ dev->net->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+ NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_TSO;
+
+ /* Enable checksum offload */
+ *tmp = AX_RXCOE_IP | AX_RXCOE_TCP | AX_RXCOE_UDP |
+ AX_RXCOE_TCPV6 | AX_RXCOE_UDPV6;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_RXCOE_CTL, 1, 1, tmp);
+
+ *tmp = AX_TXCOE_IP | AX_TXCOE_TCP | AX_TXCOE_UDP |
+ AX_TXCOE_TCPV6 | AX_TXCOE_UDPV6;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, tmp);
+
+ /* Configure RX control register => start operation */
+ *tmp16 = AX_RX_CTL_DROPCRCERR | AX_RX_CTL_IPE | AX_RX_CTL_START |
+ AX_RX_CTL_AP | AX_RX_CTL_AMALL | AX_RX_CTL_AB;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, 0x0b, 2, 2, tmp16);
+
+ *tmp = AX_MONITOR_MODE_PMETYPE | AX_MONITOR_MODE_PMEPOL |
+ AX_MONITOR_MODE_RWMP;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_MONITOR_MODE, 1, 1, tmp);
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_MONITOR_MODE, 1, 1, tmp);
+
+ /* Configure default medium type => giga */
+ *tmp16 = AX_MEDIUM_RECEIVE_EN | AX_MEDIUM_TXFLOW_CTRLEN |
+ AX_MEDIUM_RXFLOW_CTRLEN | AX_MEDIUM_ALWAYS_ONE |
+ AX_MEDIUM_FULL_DUPLEX | AX_MEDIUM_GIGAMODE;
+
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_MEDIUM_STATUS_MODE, 2, 2, tmp16);
+
+ ax88179_led_setting(dev);
+
+ /* Restart autoneg */
+ mii_nway_restart(&dev->mii);
+
+ netif_carrier_off(dev->net);
+
+ return 0;
+}
+
+static void ax88179_unbind(struct usbnet *dev, struct usb_interface *intf)
+{
+ u16 tmp16;
+
+ /* Configure RX control register => stop operation */
+ tmp16 = AX_RX_CTL_STOP;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_RX_CTL, 2, 2, &tmp16);
+
+ tmp16 = 0x0;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_CLK_SELECT, 1, 1, &tmp16);
+
+ /* Power down ethernet PHY */
+ tmp16 = 0;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_PHYPWR_RSTCTL, 2, 2, &tmp16);
+}
+
+static void
+ax88179_rx_checksum(struct sk_buff *skb, struct ax88179_rx_pkt_header *pkt_hdr)
+{
+ skb->ip_summed = CHECKSUM_NONE;
+
+ /* checksum error bit is set */
+ if (pkt_hdr->l3_csum_err || pkt_hdr->l4_csum_err)
+ return;
+
+ /* It must be a TCP or UDP packet with a valid checksum */
+ if ((pkt_hdr->l4_type == AX_RXHDR_L4_TYPE_TCP) ||
+ (pkt_hdr->l4_type == AX_RXHDR_L4_TYPE_UDP))
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+}
+
+static int ax88179_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
+{
+ struct sk_buff *ax_skb;
+ int pkt_cnt;
+ u32 rx_hdr;
+ u16 hdr_off;
+ struct ax88179_rx_pkt_header *pkt_hdr;
+
+ skb_trim(skb, skb->len - 4);
+ memcpy(&rx_hdr, skb_tail_pointer(skb), sizeof(rx_hdr));
+ le32_to_cpus(&rx_hdr);
+
+ pkt_cnt = (u16)rx_hdr;
+ hdr_off = (u16)(rx_hdr >> 16);
+ pkt_hdr = (struct ax88179_rx_pkt_header *)(skb->data + hdr_off);
+
+ while (pkt_cnt--) {
+
+ u16 pkt_len;
+
+ le32_to_cpus((u32 *)pkt_hdr);
+ pkt_len = pkt_hdr->len;
+
+ /* Check CRC or runt packet */
+ if (pkt_hdr->crc || pkt_hdr->drop) {
+ skb_pull(skb, (pkt_len + 7) & 0xFFF8);
+ pkt_hdr++;
+ continue;
+ }
+
+ if (pkt_cnt == 0) {
+ /* Skip IP alignment psudo header */
+ skb_pull(skb, 2);
+
+ skb->len = pkt_len;
+
+ skb_set_tail_pointer(skb, pkt_len);
+
+ skb->truesize = pkt_len + sizeof(struct sk_buff);
+
+ ax88179_rx_checksum(skb, pkt_hdr);
+
+ return 2;
+ }
+
+ ax_skb = skb_clone(skb, GFP_ATOMIC);
+ if (ax_skb) {
+ ax_skb->len = pkt_len;
+ ax_skb->data = skb->data + 2;
+
+ skb_set_tail_pointer(ax_skb, pkt_len);
+
+ ax_skb->truesize = pkt_len + sizeof(struct sk_buff);
+
+ ax88179_rx_checksum(ax_skb, pkt_hdr);
+
+ usbnet_skb_return(dev, ax_skb);
+ } else {
+ return 0;
+ }
+
+ skb_pull(skb, (pkt_len + 7) & 0xFFF8);
+ pkt_hdr++;
+ }
+ return 0;
+}
+
+static struct sk_buff *
+ax88179_tx_fixup(struct usbnet *dev, struct sk_buff *skb, gfp_t flags)
+{
+ u32 tx_hdr1, tx_hdr2;
+ int frame_size = dev->maxpacket;
+ int mss = skb_shinfo(skb)->gso_size;
+ int headroom;
+ int tailroom;
+
+ tx_hdr1 = skb->len;
+ tx_hdr2 = mss;
+ if (((skb->len + 8) % frame_size) == 0) {
+ tx_hdr2 |= 0x80008000; /* Enable padding */
+ skb->len += 2;
+ }
+
+ skb_linearize(skb);
+ headroom = skb_headroom(skb);
+ tailroom = skb_tailroom(skb);
+
+ if (!skb_header_cloned(skb) && !skb_cloned(skb) &&
+ (headroom + tailroom) >= 8) {
+ if (headroom < 8) {
+ skb->data = memmove(skb->head + 8, skb->data, skb->len);
+ skb_set_tail_pointer(skb, skb->len);
+ }
+ } else {
+ struct sk_buff *skb2;
+ skb2 = skb_copy_expand(skb, 8, 0, flags);
+ dev_kfree_skb_any(skb);
+ skb = skb2;
+ if (!skb)
+ return NULL;
+ }
+
+ skb_push(skb, 4);
+ cpu_to_le32s(&tx_hdr2);
+
+ skb_copy_to_linear_data(skb, &tx_hdr2, 4);
+
+ skb_push(skb, 4);
+ cpu_to_le32s(&tx_hdr1);
+
+ skb_copy_to_linear_data(skb, &tx_hdr1, 4);
+
+ return skb;
+}
+
+static int ax88179_link_reset(struct usbnet *dev)
+{
+ struct ax88179_data *ax179_data = (struct ax88179_data *)dev->data;
+ u8 tmp[5], link_sts;
+ u16 mode, tmp16, delay = HZ / 10;
+ u32 tmp32 = 0x40000000;
+ unsigned long jtimeout;
+
+ jtimeout = jiffies + delay;
+
+ while (tmp32 & 0x40000000) {
+ mode = 0;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_RX_CTL, 2, 2, &mode);
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_RX_CTL, 2, 2, &ax179_data->rxctl);
+
+ /*link up, check the usb device control TX FIFO full or empty*/
+ ax88179_read_cmd(dev, 0x81, 0x8c, 0, 4, &tmp32);
+
+ if (time_after(jiffies, jtimeout))
+ return 0;
+ }
+
+ mode = AX_MEDIUM_RECEIVE_EN | AX_MEDIUM_TXFLOW_CTRLEN |
+ AX_MEDIUM_RXFLOW_CTRLEN | AX_MEDIUM_ALWAYS_ONE;
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ PHYSICAL_LINK_STATUS, 1, 1, &link_sts);
+
+ ax88179_read_cmd(dev, AX_ACCESS_PHY, AX88179_PHY_ID,
+ GMII_PHY_PHYSR, 2, &tmp16);
+
+ if (!(tmp16 & GMII_PHY_PHYSR_LINK)) {
+ return 0;
+ } else if (GMII_PHY_PHYSR_GIGA == (tmp16 & GMII_PHY_PHYSR_SMASK)) {
+ mode |= AX_MEDIUM_GIGAMODE;
+ if (link_sts & AX_USB_SS)
+ memcpy(tmp, &AX88179_BULKIN_SIZE[0], 5);
+ else if (link_sts & AX_USB_HS)
+ memcpy(tmp, &AX88179_BULKIN_SIZE[1], 5);
+ else
+ memcpy(tmp, &AX88179_BULKIN_SIZE[3], 5);
+ } else if (GMII_PHY_PHYSR_100 == (tmp16 & GMII_PHY_PHYSR_SMASK)) {
+ mode |= AX_MEDIUM_PS;
+ if (link_sts & (AX_USB_SS | AX_USB_HS))
+ memcpy(tmp, &AX88179_BULKIN_SIZE[2], 5);
+ else
+ memcpy(tmp, &AX88179_BULKIN_SIZE[3], 5);
+ } else {
+ memcpy(tmp, &AX88179_BULKIN_SIZE[3], 5);
+ }
+
+ /* RX bulk configuration */
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_RX_BULKIN_QCTRL, 5, 5, tmp);
+
+ if (tmp16 & GMII_PHY_PHYSR_FULL)
+ mode |= AX_MEDIUM_FULL_DUPLEX;
+
+ dev->rx_urb_size = (1024 * (*(tmp + 3)));
+
+ mode |= 0x08;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_MEDIUM_STATUS_MODE, 2, 2, &mode);
+
+ netif_carrier_on(dev->net);
+
+ return 0;
+}
+
+static int ax88179_reset(struct usbnet *dev)
+{
+ u8 buf[5];
+ u16 *tmp16;
+ u8 *tmp;
+
+ tmp16 = (u16 *)buf;
+ tmp = (u8 *)buf;
+
+ /* Power up ethernet PHY */
+ *tmp16 = 0;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PHYPWR_RSTCTL, 2, 2, tmp16);
+ *tmp16 = AX_PHYPWR_RSTCTL_IPRL;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PHYPWR_RSTCTL, 2, 2, tmp16);
+ msleep(200);
+ *tmp = AX_CLK_SELECT_ACS | AX_CLK_SELECT_BCS;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_CLK_SELECT, 1, 1, tmp);
+ msleep(100);
+
+ /* Ethernet PHY Auto Detach*/
+ ax88179_auto_detach(dev, 0);
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_NODE_ID, ETH_ALEN,
+ ETH_ALEN, dev->net->dev_addr);
+ memcpy(dev->net->perm_addr, dev->net->dev_addr, ETH_ALEN);
+
+ /* RX bulk configuration */
+ memcpy(tmp, &AX88179_BULKIN_SIZE[0], 5);
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_RX_BULKIN_QCTRL, 5, 5, tmp);
+
+ dev->rx_urb_size = 1024 * 20;
+
+ *tmp = 0x34;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_PAUSE_WATERLVL_LOW, 1, 1, tmp);
+
+ *tmp = 0x52;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_PAUSE_WATERLVL_HIGH, 1, 1, tmp);
+
+ dev->net->features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+ NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_TSO;
+
+ dev->net->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+ NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_TSO;
+
+ /* Enable checksum offload */
+ *tmp = AX_RXCOE_IP | AX_RXCOE_TCP | AX_RXCOE_UDP |
+ AX_RXCOE_TCPV6 | AX_RXCOE_UDPV6;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_RXCOE_CTL, 1, 1, tmp);
+
+ *tmp = AX_TXCOE_IP | AX_TXCOE_TCP | AX_TXCOE_UDP |
+ AX_TXCOE_TCPV6 | AX_TXCOE_UDPV6;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_TXCOE_CTL, 1, 1, tmp);
+
+ /* Configure RX control register => start operation */
+ *tmp16 = AX_RX_CTL_DROPCRCERR | AX_RX_CTL_IPE | AX_RX_CTL_START |
+ AX_RX_CTL_AP | AX_RX_CTL_AMALL | AX_RX_CTL_AB;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, 0x0b, 2, 2, tmp16);
+
+ *tmp = AX_MONITOR_MODE_PMETYPE | AX_MONITOR_MODE_PMEPOL |
+ AX_MONITOR_MODE_RWMP;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC, AX_MONITOR_MODE, 1, 1, tmp);
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC, AX_MONITOR_MODE, 1, 1, tmp);
+
+ /* Configure default medium type => giga */
+ *tmp16 = AX_MEDIUM_RECEIVE_EN | AX_MEDIUM_TXFLOW_CTRLEN |
+ AX_MEDIUM_RXFLOW_CTRLEN | AX_MEDIUM_ALWAYS_ONE |
+ AX_MEDIUM_FULL_DUPLEX | AX_MEDIUM_GIGAMODE;
+
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_MEDIUM_STATUS_MODE, 2, 2, tmp16);
+
+ ax88179_led_setting(dev);
+
+ /* Restart autoneg */
+ mii_nway_restart(&dev->mii);
+
+ netif_carrier_off(dev->net);
+
+ return 0;
+}
+
+static int ax88179_stop(struct usbnet *dev)
+{
+ u16 tmp16;
+
+ ax88179_read_cmd(dev, AX_ACCESS_MAC,
+ AX_MEDIUM_STATUS_MODE, 2, 2, &tmp16);
+ tmp16 &= ~AX_MEDIUM_RECEIVE_EN;
+ ax88179_write_cmd(dev, AX_ACCESS_MAC,
+ AX_MEDIUM_STATUS_MODE, 2, 2, &tmp16);
+
+ return 0;
+}
+
+static const struct driver_info ax88179_info = {
+ .description = "ASIX AX88179 USB 3.0 Gigibit Ethernet",
+ .bind = ax88179_bind,
+ .unbind = ax88179_unbind,
+ .status = ax88179_status,
+ .link_reset = ax88179_link_reset,
+ .reset = ax88179_reset,
+ .stop = ax88179_stop,
+ .flags = FLAG_ETHER | FLAG_FRAMING_AX,
+ .rx_fixup = ax88179_rx_fixup,
+ .tx_fixup = ax88179_tx_fixup,
+};
+
+static const struct driver_info ax88178a_info = {
+ .description = "ASIX AX88178A USB 2.0 Gigibit Ethernet",
+ .bind = ax88179_bind,
+ .unbind = ax88179_unbind,
+ .status = ax88179_status,
+ .link_reset = ax88179_link_reset,
+ .reset = ax88179_reset,
+ .stop = ax88179_stop,
+ .flags = FLAG_ETHER | FLAG_FRAMING_AX,
+ .rx_fixup = ax88179_rx_fixup,
+ .tx_fixup = ax88179_tx_fixup,
+};
+
+static const struct driver_info sitecom_info = {
+ .description = "Sitecom USB 3.0 to Gigabit Adapter",
+ .bind = ax88179_bind,
+ .unbind = ax88179_unbind,
+ .status = ax88179_status,
+ .link_reset = ax88179_link_reset,
+ .reset = ax88179_reset,
+ .stop = ax88179_stop,
+ .flags = FLAG_ETHER | FLAG_FRAMING_AX,
+ .rx_fixup = ax88179_rx_fixup,
+ .tx_fixup = ax88179_tx_fixup,
+};
+
+static const struct usb_device_id products[] = {
+{
+ /* ASIX AX88179 10/100/1000 */
+ USB_DEVICE(0x0b95, 0x1790),
+ .driver_info = (unsigned long)&ax88179_info,
+}, {
+ /* ASIX AX88178A 10/100/1000 */
+ USB_DEVICE(0x0b95, 0x178a),
+ .driver_info = (unsigned long)&ax88178a_info,
+}, {
+ /* Sitecom USB 3.0 to Gigabit Adapter */
+ USB_DEVICE(0x0df6, 0x0072),
+ .driver_info = (unsigned long) &sitecom_info,
+},
+ { },
+};
+MODULE_DEVICE_TABLE(usb, products);
+
+static struct usb_driver asix_driver = {
+ .name = "ax88179_178a",
+ .id_table = products,
+ .probe = usbnet_probe,
+ .suspend = ax88179_suspend,
+ .resume = ax88179_resume,
+ .disconnect = usbnet_disconnect,
+};
+
+static int __init asix_init(void)
+{
+ return usb_register(&asix_driver);
+}
+module_init(asix_init);
+
+static void __exit asix_exit(void)
+{
+ usb_deregister(&asix_driver);
+}
+module_exit(asix_exit);
+
+MODULE_DESCRIPTION("ASIX AX88179/178A based USB 3.0/2.0 Gigabit Ethernet Devices");
+MODULE_LICENSE("GPL");
+
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next] cxgb3: Fix Tx csum stats
From: Vipul Pandya @ 2013-01-17 9:29 UTC (permalink / raw)
To: netdev; +Cc: davem, divy, dm, abhishek, Vipul Pandya, Jay Hernandez
Signed-off-by: Jay Hernandez <jay@chelsio.com>
Signed-off-by: Vipul Pandya <vipul@chelsio.com>
---
drivers/net/ethernet/chelsio/cxgb3/sge.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb3/sge.c b/drivers/net/ethernet/chelsio/cxgb3/sge.c
index dd901c5..9096dc0 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/sge.c
@@ -1278,7 +1278,7 @@ netdev_tx_t t3_eth_xmit(struct sk_buff *skb, struct net_device *dev)
}
/* update port statistics */
- if (skb->ip_summed == CHECKSUM_COMPLETE)
+ if (skb->ip_summed == CHECKSUM_PARTIAL)
qs->port_stats[SGE_PSTAT_TX_CSUM]++;
if (skb_shinfo(skb)->gso_size)
qs->port_stats[SGE_PSTAT_TSO]++;
--
1.7.1
^ permalink raw reply related
* RE: [PATCH 0/5]: soreuseport: Bind multiple sockets to the same port
From: David Laight @ 2013-01-17 9:53 UTC (permalink / raw)
To: Tom Herbert; +Cc: netdev, davem, netdev, eric.dumazet
In-Reply-To: <CA+mtBx-DRiSg4z5OUvp4=pJxZ=Ki4pt-vz716ubsJx1PHnD7Nw@mail.gmail.com>
> We had considered solving this within accept. The problem is that
> there's no way to indicate how much work a thread should do via
> accept. For instance, an event loop usually would look like:
>
> while (1) {
> fd = accept();
> process(fd);
> }
>
> With multiple threads, the number of accepted sockets in a particular
> thread is non-deterministic...
If your loop looks like that then each thread is only processing
a single socket and won't call accept() again until it is idle.
OTOH if each thread is processing multiple requests using
poll/select (or similar) at the top of the loop then a single
thread is likely to pick up a large number of connections.
Given that both poll and select are inefficient with very large
numbers of fds (every call is usually o(n) [1]), the kernel will
support some kind of event mechanism, maybe tweaking that to
signal the waiters in turn would also work - and be more general.
It might also be possible to do something on the user side of
sockets to generate additional fd with their own queue?
(IMHO some of the SCTP stuff should have been done that way).
David
[1] I've known systems where it was actually o(n*n) because
the relevant kernel used a linked list to map fd to kernel
file structures! By the time you got to 1000 fd this dominated.
^ permalink raw reply
* [BUG] "sleeping in invalid" in dev_change_name()
From: Guennadi Liakhovetski @ 2013-01-17 10:22 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller
Hi, getting a BUG dump below from a wireless driver, is probably not
driver specific. Just not seeing it normally, because in most cases the
name is either the same, or the interface is already up and you get a
-EBUSY back.
Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
[ 63.789062] mmc1: queuing unknown CIS tuple 0x80 (5 bytes)
[ 63.789062] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[ 63.804687] mmc1: queuing unknown CIS tuple 0x80 (9 bytes)
[ 63.804687] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[ 63.812500] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[ 63.820312] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[ 63.820312] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[ 63.828125] mmc1: queuing unknown CIS tuple 0x80 (5 bytes)
[ 63.835937] mmc1: queuing unknown CIS tuple 0x80 (9 bytes)
[ 63.843750] mmc1: new SDIO card at address 0001
[ 63.851562] b43-sdio mmc1:0001:1: Chip ID 14e4:4318
[ 63.851562] ssb: Found chip with id 0x4318, rev 0x02 and package 0x00
[ 63.867187] ssb: Core 0 found: ChipCommon (cc 0x800, rev 0x0D, vendor 0x4243)
[ 63.875000] ssb: Core 1 found: IEEE 802.11 (cc 0x812, rev 0x09, vendor 0x4243)
[ 63.890625] ssb: Core 2 found: PCI (cc 0x804, rev 0x0C, vendor 0x4243)
[ 63.890625] ssb: Core 3 found: PCMCIA (cc 0x80D, rev 0x07, vendor 0x4243)
[ 63.906250] b43-phy0: Broadcom 4318 WLAN found (core revision 9)
[ 63.960937] b43-phy0: Found PHY: Analog 3, Type 2 (G), Revision 7
[ 64.015625] ssb: Sonics Silicon Backplane found on SDIO device mmc1:0001:1
[ 64.187500] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
[ 64.203125] BUG: sleeping function called from invalid context at .../linux/mm/slab.c:3179
[ 64.210937] in_atomic(): 1, irqs_disabled(): 0, pid: 940, name: udevd
[ 64.218750] Backtrace:
[ 64.218750] [<c0012724>] (dump_backtrace+0x0/0x110) from [<c0267c80>] (dump_stack+0x18/0x1c)
[ 64.226562] r6:cf1c6bc0 r5:000000d0 r4:cf000080 r3:60000013
[ 64.234375] [<c0267c68>] (dump_stack+0x0/0x1c) from [<c004493c>] (__might_sleep+0xec/0x10c)
[ 64.242187] [<c0044850>] (__might_sleep+0x0/0x10c) from [<c008f988>] (__kmalloc_track_caller+0x80/0x188)
[ 64.250000] [<c008f908>] (__kmalloc_track_caller+0x0/0x188) from [<c0074dac>] (kstrdup+0x38/0x54)
[ 64.257812] r7:000000d0 r6:cf1c6bc0 r5:00000006 r4:c0199af4
[ 64.257812] [<c0074d74>] (kstrdup+0x0/0x54) from [<c0199af4>] (device_rename+0x34/0xa4)
[ 64.265625] r7:c03ca0c0 r6:cf4b9000 r5:cf4b9000 r4:cf4b9218
[ 64.273437] [<c0199ac0>] (device_rename+0x0/0xa4) from [<c01eb9f4>] (dev_change_name+0xdc/0x274)
[ 64.281250] r7:c03ca0c0 r6:00000000 r5:cf4b9000 r4:cf504000
[ 64.289062] [<c01eb918>] (dev_change_name+0x0/0x274) from [<c01ebd5c>] (dev_ifsioc+0x1d0/0x34c)
[ 64.296875] [<c01ebb8c>] (dev_ifsioc+0x0/0x34c) from [<c01ec4a4>] (dev_ioctl+0x5cc/0x748)
[ 64.304687] r7:cf505e70 r6:00000000 r5:00008923 r4:bea7e220
[ 64.312500] [<c01ebed8>] (dev_ioctl+0x0/0x748) from [<c01d8954>] (sock_ioctl+0x270/0x2a4)
[ 64.320312] [<c01d86e4>] (sock_ioctl+0x0/0x2a4) from [<c00a0f84>] (vfs_ioctl+0x28/0x40)
[ 64.328125] r6:00000003 r5:cf1d4340 r4:bea7e220 r3:c01d86e4
[ 64.335937] [<c00a0f5c>] (vfs_ioctl+0x0/0x40) from [<c00a1738>] (do_vfs_ioctl+0x5b0/0x5fc)
[ 64.343750] [<c00a1188>] (do_vfs_ioctl+0x0/0x5fc) from [<c00a17c8>] (sys_ioctl+0x44/0x64)
[ 64.351562] [<c00a1784>] (sys_ioctl+0x0/0x64) from [<c000ec00>] (ret_fast_syscall+0x0/0x30)
[ 64.359375] r8:c000eda8 r7:00000036 r6:b6fb7d98 r5:b6fcf1a8 r4:b6fcf168
[ 84.742187] b43-phy0: Loading OpenSource firmware version 410.31754
[ 84.742187] b43-phy0: Hardware crypto acceleration not supported by firmware
^ permalink raw reply
* [patch net-next] team: do not use -ENOENT
From: Jiri Pirko @ 2013-01-17 10:25 UTC (permalink / raw)
To: netdev; +Cc: davem
Since this error code means "No such file or directory", change this
value in team driver to ones which make more sense.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/net/team/team.c | 4 ++--
drivers/net/team/team_mode_activebackup.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 70d5d6b..3d7cf6e 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -1128,7 +1128,7 @@ static int team_port_del(struct team *team, struct net_device *port_dev)
if (!port || !team_port_find(team, port)) {
netdev_err(dev, "Device %s does not act as a port of this team\n",
portname);
- return -ENOENT;
+ return -ENODEV;
}
__team_option_inst_mark_removed_port(team, port);
@@ -2320,7 +2320,7 @@ static int team_nl_cmd_options_set(struct sk_buff *skb, struct genl_info *info)
list_add(&opt_inst->tmp_list, &opt_inst_list);
}
if (!opt_found) {
- err = -ENOENT;
+ err = -EINVAL;
goto team_put;
}
}
diff --git a/drivers/net/team/team_mode_activebackup.c b/drivers/net/team/team_mode_activebackup.c
index 6262b4d..2792e13 100644
--- a/drivers/net/team/team_mode_activebackup.c
+++ b/drivers/net/team/team_mode_activebackup.c
@@ -81,7 +81,7 @@ static int ab_active_port_set(struct team *team, struct team_gsetter_ctx *ctx)
return 0;
}
}
- return -ENOENT;
+ return -ENODEV;
}
static const struct team_option ab_options[] = {
--
1.8.1
^ permalink raw reply related
* Re: [PATCH 4/4 v3] net/smsc911x: Provide common clock functionality
From: Linus Walleij @ 2013-01-17 10:25 UTC (permalink / raw)
To: Lee Jones
Cc: Russell King - ARM Linux, Steve Glendinning, Robert Marklund,
linus.walleij, arnd, netdev, linux-kernel, linux-arm-kernel
In-Reply-To: <20130109085555.GA31678@gmail.com>
On Wed, Jan 9, 2013 at 9:55 AM, Lee Jones <lee.jones@linaro.org> wrote:
> On Thu, 03 Jan 2013, Linus Walleij wrote:
>
>> On Thu, Jan 3, 2013 at 12:14 PM, Lee Jones <lee.jones@linaro.org> wrote:
>>
>> > Some platforms provide clocks which require enabling before the
>> > SMSC911x chip will power on. This patch uses the new common clk
>> > framework to do just that. If no clock is provided, it will just
>> > be ignored and the driver will continue to assume that no clock
>> > is required for the chip to run successfully.
>> >
>> > Cc: Steve Glendinning <steve.glendinning@shawell.net>
>> > Cc: netdev@vger.kernel.org
>> > Signed-off-by: Lee Jones <lee.jones@linaro.org>
>>
>> Looks all right to me now:
>> Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
>
> I still need a maintiner Ack for this before I can push it.
>
> Anyone?
Steve is there sometimes. But I'd ask David Miller to ACK
it if he's OK with the change.
Yours,
Linus Walleij
^ permalink raw reply
* [PATCH v3 0/2] make mac programming for virtio net more robust
From: akong @ 2013-01-17 10:40 UTC (permalink / raw)
To: mst; +Cc: kvm, netdev, qemu-devel, virtualization, davem
From: Amos Kong <akong@redhat.com>
Currenly mac is programmed byte by byte. This means that we
have an intermediate step where mac is wrong.
Second patch introduced a new vq control command to set mac
address, it's atomic.
V2: check return of sending command, delay eth_mac_addr()
V3: restore software address when fail to set hardware address
Amos Kong (2):
move virtnet_send_command() above virtnet_set_mac_address()
virtio-net: introduce a new control to set macaddr
drivers/net/virtio_net.c | 110 +++++++++++++++++++++++-----------------
include/uapi/linux/virtio_net.h | 8 ++-
2 files changed, 71 insertions(+), 47 deletions(-)
--
1.7.11.7
^ permalink raw reply
* [PATCH v3 1/2] move virtnet_send_command() above virtnet_set_mac_address()
From: akong @ 2013-01-17 10:40 UTC (permalink / raw)
To: mst; +Cc: kvm, netdev, qemu-devel, virtualization, davem
In-Reply-To: <1358419212-32356-1-git-send-email-akong@redhat.com>
From: Amos Kong <akong@redhat.com>
We will send vq command to set mac address in virtnet_set_mac_address()
a little fix of coding style
Signed-off-by: Amos Kong <akong@redhat.com>
---
drivers/net/virtio_net.c | 89 ++++++++++++++++++++++++------------------------
1 file changed, 44 insertions(+), 45 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index a6fcf15..395ab4f 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -753,6 +753,50 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
return NETDEV_TX_OK;
}
+/*
+ * Send command via the control virtqueue and check status. Commands
+ * supported by the hypervisor, as indicated by feature bits, should
+ * never fail unless improperly formated.
+ */
+static bool virtnet_send_command(struct virtnet_info *vi, u8 class, u8 cmd,
+ struct scatterlist *data, int out, int in)
+{
+ struct scatterlist *s, sg[VIRTNET_SEND_COMMAND_SG_MAX + 2];
+ struct virtio_net_ctrl_hdr ctrl;
+ virtio_net_ctrl_ack status = ~0;
+ unsigned int tmp;
+ int i;
+
+ /* Caller should know better */
+ BUG_ON(!virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ||
+ (out + in > VIRTNET_SEND_COMMAND_SG_MAX));
+
+ out++; /* Add header */
+ in++; /* Add return status */
+
+ ctrl.class = class;
+ ctrl.cmd = cmd;
+
+ sg_init_table(sg, out + in);
+
+ sg_set_buf(&sg[0], &ctrl, sizeof(ctrl));
+ for_each_sg(data, s, out + in - 2, i)
+ sg_set_buf(&sg[i + 1], sg_virt(s), s->length);
+ sg_set_buf(&sg[out + in - 1], &status, sizeof(status));
+
+ BUG_ON(virtqueue_add_buf(vi->cvq, sg, out, in, vi, GFP_ATOMIC) < 0);
+
+ virtqueue_kick(vi->cvq);
+
+ /* Spin for a response, the kick causes an ioport write, trapping
+ * into the hypervisor, so the request should be handled immediately.
+ */
+ while (!virtqueue_get_buf(vi->cvq, &tmp))
+ cpu_relax();
+
+ return status == VIRTIO_NET_OK;
+}
+
static int virtnet_set_mac_address(struct net_device *dev, void *p)
{
struct virtnet_info *vi = netdev_priv(dev);
@@ -819,51 +863,6 @@ static void virtnet_netpoll(struct net_device *dev)
}
#endif
-/*
- * Send command via the control virtqueue and check status. Commands
- * supported by the hypervisor, as indicated by feature bits, should
- * never fail unless improperly formated.
- */
-static bool virtnet_send_command(struct virtnet_info *vi, u8 class, u8 cmd,
- struct scatterlist *data, int out, int in)
-{
- struct scatterlist *s, sg[VIRTNET_SEND_COMMAND_SG_MAX + 2];
- struct virtio_net_ctrl_hdr ctrl;
- virtio_net_ctrl_ack status = ~0;
- unsigned int tmp;
- int i;
-
- /* Caller should know better */
- BUG_ON(!virtio_has_feature(vi->vdev, VIRTIO_NET_F_CTRL_VQ) ||
- (out + in > VIRTNET_SEND_COMMAND_SG_MAX));
-
- out++; /* Add header */
- in++; /* Add return status */
-
- ctrl.class = class;
- ctrl.cmd = cmd;
-
- sg_init_table(sg, out + in);
-
- sg_set_buf(&sg[0], &ctrl, sizeof(ctrl));
- for_each_sg(data, s, out + in - 2, i)
- sg_set_buf(&sg[i + 1], sg_virt(s), s->length);
- sg_set_buf(&sg[out + in - 1], &status, sizeof(status));
-
- BUG_ON(virtqueue_add_buf(vi->cvq, sg, out, in, vi, GFP_ATOMIC) < 0);
-
- virtqueue_kick(vi->cvq);
-
- /*
- * Spin for a response, the kick causes an ioport write, trapping
- * into the hypervisor, so the request should be handled immediately.
- */
- while (!virtqueue_get_buf(vi->cvq, &tmp))
- cpu_relax();
-
- return status == VIRTIO_NET_OK;
-}
-
static void virtnet_ack_link_announce(struct virtnet_info *vi)
{
rtnl_lock();
--
1.7.11.7
^ permalink raw reply related
* [PATCH v3 2/2] virtio-net: introduce a new control to set macaddr
From: akong @ 2013-01-17 10:40 UTC (permalink / raw)
To: mst; +Cc: kvm, netdev, qemu-devel, virtualization, davem
In-Reply-To: <1358419212-32356-1-git-send-email-akong@redhat.com>
From: Amos Kong <akong@redhat.com>
Currently we write MAC address to pci config space byte by byte,
this means that we have an intermediate step where mac is wrong.
This patch introduced a new control command to set MAC address,
it's atomic.
VIRTIO_NET_F_CTRL_MAC_ADDR is a new feature bit for compatibility.
Signed-off-by: Amos Kong <akong@redhat.com>
---
drivers/net/virtio_net.c | 21 ++++++++++++++++++++-
include/uapi/linux/virtio_net.h | 8 +++++++-
2 files changed, 27 insertions(+), 2 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 395ab4f..837c978 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -802,14 +802,32 @@ static int virtnet_set_mac_address(struct net_device *dev, void *p)
struct virtnet_info *vi = netdev_priv(dev);
struct virtio_device *vdev = vi->vdev;
int ret;
+ struct scatterlist sg;
+ char save_addr[ETH_ALEN];
+ unsigned char save_aatype;
+
+ memcpy(save_addr, dev->dev_addr, ETH_ALEN);
+ save_aatype = dev->addr_assign_type;
ret = eth_mac_addr(dev, p);
if (ret)
return ret;
- if (virtio_has_feature(vdev, VIRTIO_NET_F_MAC))
+ if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_MAC_ADDR)) {
+ sg_init_one(&sg, dev->dev_addr, dev->addr_len);
+ if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_MAC,
+ VIRTIO_NET_CTRL_MAC_ADDR_SET,
+ &sg, 1, 0)) {
+ dev_warn(&vdev->dev,
+ "Failed to set mac address by vq command.\n");
+ memcpy(dev->dev_addr, save_addr, ETH_ALEN);
+ dev->addr_assign_type = save_aatype;
+ return -EINVAL;
+ }
+ } else if (virtio_has_feature(vdev, VIRTIO_NET_F_MAC)) {
vdev->config->set(vdev, offsetof(struct virtio_net_config, mac),
dev->dev_addr, dev->addr_len);
+ }
return 0;
}
@@ -1627,6 +1645,7 @@ static unsigned int features[] = {
VIRTIO_NET_F_MRG_RXBUF, VIRTIO_NET_F_STATUS, VIRTIO_NET_F_CTRL_VQ,
VIRTIO_NET_F_CTRL_RX, VIRTIO_NET_F_CTRL_VLAN,
VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ,
+ VIRTIO_NET_F_CTRL_MAC_ADDR,
};
static struct virtio_driver virtio_net_driver = {
diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h
index 848e358..a5a8c88 100644
--- a/include/uapi/linux/virtio_net.h
+++ b/include/uapi/linux/virtio_net.h
@@ -53,6 +53,7 @@
* network */
#define VIRTIO_NET_F_MQ 22 /* Device supports Receive Flow
* Steering */
+#define VIRTIO_NET_F_CTRL_MAC_ADDR 23 /* Set MAC address */
#define VIRTIO_NET_S_LINK_UP 1 /* Link is up */
#define VIRTIO_NET_S_ANNOUNCE 2 /* Announcement is needed */
@@ -127,7 +128,7 @@ typedef __u8 virtio_net_ctrl_ack;
#define VIRTIO_NET_CTRL_RX_NOBCAST 5
/*
- * Control the MAC filter table.
+ * Control the MAC
*
* The MAC filter table is managed by the hypervisor, the guest should
* assume the size is infinite. Filtering should be considered
@@ -140,6 +141,10 @@ typedef __u8 virtio_net_ctrl_ack;
* first sg list contains unicast addresses, the second is for multicast.
* This functionality is present if the VIRTIO_NET_F_CTRL_RX feature
* is available.
+ *
+ * The ADDR_SET command requests one out scatterlist, it contains a
+ * 6 bytes MAC address. This functionality is present if the
+ * VIRTIO_NET_F_CTRL_MAC_ADDR feature is available.
*/
struct virtio_net_ctrl_mac {
__u32 entries;
@@ -148,6 +153,7 @@ struct virtio_net_ctrl_mac {
#define VIRTIO_NET_CTRL_MAC 1
#define VIRTIO_NET_CTRL_MAC_TABLE_SET 0
+ #define VIRTIO_NET_CTRL_MAC_ADDR_SET 1
/*
* Control VLAN filtering
--
1.7.11.7
^ permalink raw reply related
* Re: [PATCH 4/4 v3] net/smsc911x: Provide common clock functionality
From: Lee Jones @ 2013-01-17 10:47 UTC (permalink / raw)
To: davem
Cc: Russell King - ARM Linux, Steve Glendinning, Robert Marklund,
linus.walleij, arnd, netdev, linux-kernel, linux-arm-kernel,
linus.walleij
In-Reply-To: <CACRpkdY+C8jNT7LsJvBkauN4+x4k4gGJKSe0ynY_HLTpmpgFPA@mail.gmail.com>
> >> > Some platforms provide clocks which require enabling before the
> >> > SMSC911x chip will power on. This patch uses the new common clk
> >> > framework to do just that. If no clock is provided, it will just
> >> > be ignored and the driver will continue to assume that no clock
> >> > is required for the chip to run successfully.
> >> >
> >> > Cc: Steve Glendinning <steve.glendinning@shawell.net>
> >> > Cc: netdev@vger.kernel.org
> >> > Signed-off-by: Lee Jones <lee.jones@linaro.org>
> >>
> >> Looks all right to me now:
> >> Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
> >
> > I still need a maintiner Ack for this before I can push it.
> >
> > Anyone?
>
> Steve is there sometimes. But I'd ask David Miller to ACK
> it if he's OK with the change.
>
> Yours,
> Linus Walleij
Hi David,
Does this [1] patch look okay to you too?
It's the last Ack I require before I can push the set.
[1] https://patchwork.kernel.org/patch/1926971/
--
Lee Jones
Linaro ST-Ericsson Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
^ permalink raw reply
* Re: Redefinition of struct in6_addr in <netinet/in.h> and <linux/in6.h>
From: Jan Engelhardt @ 2013-01-17 10:57 UTC (permalink / raw)
To: David Miller
Cc: carlos, bhutchings, yoshfuji, amwang, tmb, eblake, netdev,
linux-kernel, libvirt-list, tgraf, libc-alpha, schwab
In-Reply-To: <20130116.210512.2230612692562212911.davem@davemloft.net>
On Thursday 2013-01-17 03:05, David Miller wrote:
>From: Carlos O'Donell <carlos@systemhalted.org>
>Date: Wed, 16 Jan 2013 20:58:47 -0500
>
>> So I just went down the rabbit hole, and the further I get the
>> closer I get to having two exact copies of the same definitions
>> in both glibc and the kernel and using whichever one was included
>> first.
>>
>> Is anyone opposed to that kind of solution?
>
>Sounds interesting, please share :-)
iptables has the same issue, and solved it its way.
(uapi/)linux/netfilter.h is used to get at things like union
nf_inet_addr. This union contains struct in6_addr. There is no include
for in6_addr in netfilter.h itself. This may break the "standalone
compilation" test, but at least allows for specifying the
environment-specific header for in6_addr in the C file:
a. userspace: #include <netinet/in.h> before <linux/netfilter.h>
b. kernel parts: #include <linux/in6.h> before <linux/netfilter.h>
^ permalink raw reply
* Re: [PATCH] ipv6: do not create neighbor entries for local delivery
From: Marcelo Ricardo Leitner @ 2013-01-17 11:17 UTC (permalink / raw)
To: David Miller, yoshfuji; +Cc: netdev
In-Reply-To: <20130116.162813.1734293658687337225.davem@davemloft.net>
Em 16-01-2013 19:28, David Miller escreveu:
> From: Marcelo Ricardo Leitner <mleitner@redhat.com>
> Date: Wed, 16 Jan 2013 11:31:54 -0200
>
>> diff --git a/net/ipv6/route.c b/net/ipv6/route.c
>> index
>> 7c34c01b515b7ac370f8021ba6f5ec4c654e9e41..b63e33bcbdfbc1bd78af0b98ffc2d211805214ea
>> 100644
>
> Patch corrupted by your email client.
Doh, really sorry about that.
> Note also that YOSHIFUJI Hideaki's patches to remove the cached neighbour
> entirely from ipv6 routes will have the same effect, so your patch won't
> be needed.
Yes but seems it needs some time yet. Mind if I submit it again (right
this time)?
Thanks,
Marcelo.
^ permalink raw reply
* [net-next 02/14] e1000e: Use standard #defines for PCIe Capability ASPM fields
From: Jeff Kirsher @ 2013-01-17 11:35 UTC (permalink / raw)
To: davem; +Cc: e1000-devel, netdev, gospo, Bjorn Helgaas, sassmann
In-Reply-To: <1358422519-20981-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Bjorn Helgaas <bhelgaas@google.com>
Use the standard #defines for PCIe Capability ASPM fields.
Previously we used PCIE_LINK_STATE_L0S and PCIE_LINK_STATE_L1 directly, but
these are defined for the Linux ASPM interfaces, e.g.,
pci_disable_link_state(), and only coincidentally match the actual register
bits. PCIE_LINK_STATE_CLKPM, also part of that interface, does not match
the register bit.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
CC: e1000-devel@lists.sourceforge.net
Acked-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/e1000e/netdev.c | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 05538e6..bf2c84cf 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -5548,14 +5548,21 @@ static void __e1000e_disable_aspm(struct pci_dev *pdev, u16 state)
#else
static void __e1000e_disable_aspm(struct pci_dev *pdev, u16 state)
{
+ u16 aspm_ctl = 0;
+
+ if (state & PCIE_LINK_STATE_L0S)
+ aspm_ctl |= PCI_EXP_LNKCTL_ASPM_L0S;
+ if (state & PCIE_LINK_STATE_L1)
+ aspm_ctl |= PCI_EXP_LNKCTL_ASPM_L1;
+
/* Both device and parent should have the same ASPM setting.
* Disable ASPM in downstream component first and then upstream.
*/
- pcie_capability_clear_word(pdev, PCI_EXP_LNKCTL, state);
+ pcie_capability_clear_word(pdev, PCI_EXP_LNKCTL, aspm_ctl);
if (pdev->bus->self)
pcie_capability_clear_word(pdev->bus->self, PCI_EXP_LNKCTL,
- state);
+ aspm_ctl);
}
#endif
static void e1000e_disable_aspm(struct pci_dev *pdev, u16 state)
--
1.7.11.7
------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122712
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
^ permalink raw reply related
* [net-next 01/14] e1000e: add ethtool .get_eee/.set_eee
From: Jeff Kirsher @ 2013-01-17 11:35 UTC (permalink / raw)
To: davem; +Cc: Bruce Allan, netdev, gospo, sassmann, Jeff Kirsher
In-Reply-To: <1358422519-20981-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Bruce Allan <bruce.w.allan@intel.com>
Add the ability to query and set Energy Efficient Ethernet parameters via
ethtool for applicable devices.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/e1000e/defines.h | 30 +++++++
drivers/net/ethernet/intel/e1000e/e1000.h | 1 +
drivers/net/ethernet/intel/e1000e/ethtool.c | 134 ++++++++++++++++++++++++++++
drivers/net/ethernet/intel/e1000e/hw.h | 1 +
drivers/net/ethernet/intel/e1000e/ich8lan.c | 24 +----
5 files changed, 167 insertions(+), 23 deletions(-)
diff --git a/drivers/net/ethernet/intel/e1000e/defines.h b/drivers/net/ethernet/intel/e1000e/defines.h
index 7326ea2..3041a54 100644
--- a/drivers/net/ethernet/intel/e1000e/defines.h
+++ b/drivers/net/ethernet/intel/e1000e/defines.h
@@ -383,6 +383,9 @@
#define E1000_KABGTXD_BGSQLBIAS 0x00050000
+/* Low Power IDLE Control */
+#define E1000_LPIC_LPIET_SHIFT 24 /* Low Power Idle Entry Time */
+
/* PBA constants */
#define E1000_PBA_8K 0x0008 /* 8KB */
#define E1000_PBA_16K 0x0010 /* 16KB */
@@ -799,6 +802,33 @@
/* BME1000 PHY Specific Control Register */
#define BME1000_PSCR_ENABLE_DOWNSHIFT 0x0800 /* 1 = enable downshift */
+/* PHY Low Power Idle Control */
+#define I82579_LPI_CTRL PHY_REG(772, 20)
+#define I82579_LPI_CTRL_100_ENABLE 0x2000
+#define I82579_LPI_CTRL_1000_ENABLE 0x4000
+#define I82579_LPI_CTRL_ENABLE_MASK 0x6000
+#define I82579_LPI_CTRL_FORCE_PLL_LOCK_COUNT 0x80
+
+/* Extended Management Interface (EMI) Registers */
+#define I82579_EMI_ADDR 0x10
+#define I82579_EMI_DATA 0x11
+#define I82579_LPI_UPDATE_TIMER 0x4805 /* in 40ns units + 40 ns base value */
+#define I82579_MSE_THRESHOLD 0x084F /* 82579 Mean Square Error Threshold */
+#define I82577_MSE_THRESHOLD 0x0887 /* 82577 Mean Square Error Threshold */
+#define I82579_MSE_LINK_DOWN 0x2411 /* MSE count before dropping link */
+#define I82579_EEE_PCS_STATUS 0x182D /* IEEE MMD Register 3.1 >> 8 */
+#define I82579_EEE_CAPABILITY 0x0410 /* IEEE MMD Register 3.20 */
+#define I82579_EEE_ADVERTISEMENT 0x040E /* IEEE MMD Register 7.60 */
+#define I82579_EEE_LP_ABILITY 0x040F /* IEEE MMD Register 7.61 */
+#define I82579_EEE_100_SUPPORTED (1 << 1) /* 100BaseTx EEE supported */
+#define I82579_EEE_1000_SUPPORTED (1 << 2) /* 1000BaseTx EEE supported */
+#define I217_EEE_PCS_STATUS 0x9401 /* IEEE MMD Register 3.1 */
+#define I217_EEE_CAPABILITY 0x8000 /* IEEE MMD Register 3.20 */
+#define I217_EEE_ADVERTISEMENT 0x8001 /* IEEE MMD Register 7.60 */
+#define I217_EEE_LP_ABILITY 0x8002 /* IEEE MMD Register 7.61 */
+
+#define E1000_EEE_RX_LPI_RCVD 0x0400 /* Tx LP idle received */
+#define E1000_EEE_TX_LPI_RCVD 0x0800 /* Rx LP idle received */
#define PHY_PAGE_SHIFT 5
#define PHY_REG(page, reg) (((page) << PHY_PAGE_SHIFT) | \
diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h
index 3f8bbc3..b89b181 100644
--- a/drivers/net/ethernet/intel/e1000e/e1000.h
+++ b/drivers/net/ethernet/intel/e1000e/e1000.h
@@ -659,6 +659,7 @@ extern s32 e1000_check_polarity_ife(struct e1000_hw *hw);
extern s32 e1000_phy_force_speed_duplex_ife(struct e1000_hw *hw);
extern s32 e1000_check_polarity_igp(struct e1000_hw *hw);
extern bool e1000_check_phy_82574(struct e1000_hw *hw);
+extern s32 e1000_read_emi_reg_locked(struct e1000_hw *hw, u16 addr, u16 *data);
static inline s32 e1000_phy_hw_reset(struct e1000_hw *hw)
{
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index 2225603..636ba09 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -35,6 +35,7 @@
#include <linux/slab.h>
#include <linux/delay.h>
#include <linux/vmalloc.h>
+#include <linux/mdio.h>
#include "e1000.h"
@@ -2050,6 +2051,137 @@ static int e1000_get_rxnfc(struct net_device *netdev,
}
}
+static int e1000e_get_eee(struct net_device *netdev, struct ethtool_eee *edata)
+{
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+ struct e1000_hw *hw = &adapter->hw;
+ u16 cap_addr, adv_addr, lpa_addr, pcs_stat_addr, phy_data, lpi_ctrl;
+ u32 status, ret_val;
+
+ if (!(adapter->flags & FLAG_IS_ICH) ||
+ !(adapter->flags2 & FLAG2_HAS_EEE))
+ return -EOPNOTSUPP;
+
+ switch (hw->phy.type) {
+ case e1000_phy_82579:
+ cap_addr = I82579_EEE_CAPABILITY;
+ adv_addr = I82579_EEE_ADVERTISEMENT;
+ lpa_addr = I82579_EEE_LP_ABILITY;
+ pcs_stat_addr = I82579_EEE_PCS_STATUS;
+ break;
+ case e1000_phy_i217:
+ cap_addr = I217_EEE_CAPABILITY;
+ adv_addr = I217_EEE_ADVERTISEMENT;
+ lpa_addr = I217_EEE_LP_ABILITY;
+ pcs_stat_addr = I217_EEE_PCS_STATUS;
+ break;
+ default:
+ return -EOPNOTSUPP;
+ }
+
+ ret_val = hw->phy.ops.acquire(hw);
+ if (ret_val)
+ return -EBUSY;
+
+ /* EEE Capability */
+ ret_val = e1000_read_emi_reg_locked(hw, cap_addr, &phy_data);
+ if (ret_val)
+ goto release;
+ edata->supported = mmd_eee_cap_to_ethtool_sup_t(phy_data);
+
+ /* EEE Advertised */
+ ret_val = e1000_read_emi_reg_locked(hw, adv_addr, &phy_data);
+ if (ret_val)
+ goto release;
+ edata->advertised = mmd_eee_adv_to_ethtool_adv_t(phy_data);
+
+ /* EEE Link Partner Advertised */
+ ret_val = e1000_read_emi_reg_locked(hw, lpa_addr, &phy_data);
+ if (ret_val)
+ goto release;
+ edata->lp_advertised = mmd_eee_adv_to_ethtool_adv_t(phy_data);
+
+ /* EEE PCS Status */
+ ret_val = e1000_read_emi_reg_locked(hw, pcs_stat_addr, &phy_data);
+ if (hw->phy.type == e1000_phy_82579)
+ phy_data <<= 8;
+
+release:
+ hw->phy.ops.release(hw);
+ if (ret_val)
+ return -ENODATA;
+
+ e1e_rphy(hw, I82579_LPI_CTRL, &lpi_ctrl);
+ status = er32(STATUS);
+
+ /* Result of the EEE auto negotiation - there is no register that
+ * has the status of the EEE negotiation so do a best-guess based
+ * on whether both Tx and Rx LPI indications have been received or
+ * base it on the link speed, the EEE advertised speeds on both ends
+ * and the speeds on which EEE is enabled locally.
+ */
+ if (((phy_data & E1000_EEE_TX_LPI_RCVD) &&
+ (phy_data & E1000_EEE_RX_LPI_RCVD)) ||
+ ((status & E1000_STATUS_SPEED_100) &&
+ (edata->advertised & ADVERTISED_100baseT_Full) &&
+ (edata->lp_advertised & ADVERTISED_100baseT_Full) &&
+ (lpi_ctrl & I82579_LPI_CTRL_100_ENABLE)) ||
+ ((status & E1000_STATUS_SPEED_1000) &&
+ (edata->advertised & ADVERTISED_1000baseT_Full) &&
+ (edata->lp_advertised & ADVERTISED_1000baseT_Full) &&
+ (lpi_ctrl & I82579_LPI_CTRL_1000_ENABLE)))
+ edata->eee_active = true;
+
+ edata->eee_enabled = !hw->dev_spec.ich8lan.eee_disable;
+ edata->tx_lpi_enabled = true;
+ edata->tx_lpi_timer = er32(LPIC) >> E1000_LPIC_LPIET_SHIFT;
+
+ return 0;
+}
+
+static int e1000e_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
+{
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+ struct e1000_hw *hw = &adapter->hw;
+ struct ethtool_eee eee_curr;
+ s32 ret_val;
+
+ if (!(adapter->flags & FLAG_IS_ICH) ||
+ !(adapter->flags2 & FLAG2_HAS_EEE))
+ return -EOPNOTSUPP;
+
+ ret_val = e1000e_get_eee(netdev, &eee_curr);
+ if (ret_val)
+ return ret_val;
+
+ if (eee_curr.advertised != edata->advertised) {
+ e_err("Setting EEE advertisement is not supported\n");
+ return -EINVAL;
+ }
+
+ if (eee_curr.tx_lpi_enabled != edata->tx_lpi_enabled) {
+ e_err("Setting EEE tx-lpi is not supported\n");
+ return -EINVAL;
+ }
+
+ if (eee_curr.tx_lpi_timer != edata->tx_lpi_timer) {
+ e_err("Setting EEE Tx LPI timer is not supported\n");
+ return -EINVAL;
+ }
+
+ if (hw->dev_spec.ich8lan.eee_disable != !edata->eee_enabled) {
+ hw->dev_spec.ich8lan.eee_disable = !edata->eee_enabled;
+
+ /* reset the link */
+ if (netif_running(netdev))
+ e1000e_reinit_locked(adapter);
+ else
+ e1000e_reset(adapter);
+ }
+
+ return 0;
+}
+
static const struct ethtool_ops e1000_ethtool_ops = {
.get_settings = e1000_get_settings,
.set_settings = e1000_set_settings,
@@ -2078,6 +2210,8 @@ static const struct ethtool_ops e1000_ethtool_ops = {
.set_coalesce = e1000_set_coalesce,
.get_rxnfc = e1000_get_rxnfc,
.get_ts_info = ethtool_op_get_ts_info,
+ .get_eee = e1000e_get_eee,
+ .set_eee = e1000e_set_eee,
};
void e1000e_set_ethtool_ops(struct net_device *netdev)
diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h
index 06239fe..81afad5 100644
--- a/drivers/net/ethernet/intel/e1000e/hw.h
+++ b/drivers/net/ethernet/intel/e1000e/hw.h
@@ -62,6 +62,7 @@ enum e1e_registers {
E1000_IVAR = 0x000E4, /* Interrupt Vector Allocation - RW */
E1000_EITR_82574_BASE = 0x000E8, /* Interrupt Throttling - RW */
#define E1000_EITR_82574(_n) (E1000_EITR_82574_BASE + (_n << 2))
+ E1000_LPIC = 0x000FC, /* Low Power Idle Control - RW */
E1000_RCTL = 0x00100, /* Rx Control - RW */
E1000_FCTTV = 0x00170, /* Flow Control Transmit Timer Value - RW */
E1000_TXCW = 0x00178, /* Tx Configuration Word - RW */
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index 051dfda..3829f7f 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -148,28 +148,6 @@
#define HV_PM_CTRL PHY_REG(770, 17)
#define HV_PM_CTRL_PLL_STOP_IN_K1_GIGA 0x100
-/* PHY Low Power Idle Control */
-#define I82579_LPI_CTRL PHY_REG(772, 20)
-#define I82579_LPI_CTRL_100_ENABLE 0x2000
-#define I82579_LPI_CTRL_1000_ENABLE 0x4000
-#define I82579_LPI_CTRL_ENABLE_MASK 0x6000
-#define I82579_LPI_CTRL_FORCE_PLL_LOCK_COUNT 0x80
-
-/* Extended Management Interface (EMI) Registers */
-#define I82579_EMI_ADDR 0x10
-#define I82579_EMI_DATA 0x11
-#define I82579_LPI_UPDATE_TIMER 0x4805 /* in 40ns units + 40 ns base value */
-#define I82579_MSE_THRESHOLD 0x084F /* 82579 Mean Square Error Threshold */
-#define I82577_MSE_THRESHOLD 0x0887 /* 82577 Mean Square Error Threshold */
-#define I82579_MSE_LINK_DOWN 0x2411 /* MSE count before dropping link */
-#define I82579_EEE_PCS_STATUS 0x182D /* IEEE MMD Register 3.1 >> 8 */
-#define I82579_EEE_LP_ABILITY 0x040F /* IEEE MMD Register 7.61 */
-#define I82579_EEE_100_SUPPORTED (1 << 1) /* 100BaseTx EEE supported */
-#define I82579_EEE_1000_SUPPORTED (1 << 2) /* 1000BaseTx EEE supported */
-#define I217_EEE_PCS_STATUS 0x9401 /* IEEE MMD Register 3.1 */
-#define I217_EEE_ADVERTISEMENT 0x8001 /* IEEE MMD Register 7.60 */
-#define I217_EEE_LP_ABILITY 0x8002 /* IEEE MMD Register 7.61 */
-
/* Intel Rapid Start Technology Support */
#define I217_PROXY_CTRL BM_PHY_REG(BM_WUC_PAGE, 70)
#define I217_PROXY_CTRL_AUTO_DISABLE 0x0080
@@ -829,7 +807,7 @@ static s32 __e1000_access_emi_reg_locked(struct e1000_hw *hw, u16 address,
*
* Assumes the SW/FW/HW Semaphore is already acquired.
**/
-static s32 e1000_read_emi_reg_locked(struct e1000_hw *hw, u16 addr, u16 *data)
+s32 e1000_read_emi_reg_locked(struct e1000_hw *hw, u16 addr, u16 *data)
{
return __e1000_access_emi_reg_locked(hw, addr, data, true);
}
--
1.7.11.7
^ permalink raw reply related
* [net-next 00/14][pull request] Intel Wired LAN Driver Updates
From: Jeff Kirsher @ 2013-01-17 11:35 UTC (permalink / raw)
To: davem; +Cc: Jeff Kirsher, netdev, gospo, sassmann
This series contains updates to e1000e and igb. Most notably is the
added timestamp support in e1000e and additional software timestamp
support in igb. As well as, the added thermal data support and SR-IOV
configuration support in igb.
The following are changes since commit d59577b6ffd313d0ab3be39cb1ab47e29bdc9182:
sk-filter: Add ability to lock a socket filter program
and are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master
Alexander Duyck (1):
igb: Report L4 Rx hash via skb->l4_rxhash
Bjorn Helgaas (1):
e1000e: Use standard #defines for PCIe Capability ASPM fields
Bruce Allan (3):
e1000e: add ethtool .get_eee/.set_eee
e1000e: add support for hardware timestamping on some devices
e1000e: add support for IEEE-1588 PTP
Carolyn Wyborny (3):
igb: Add i2c interface to igb.
igb: Add support functions to access thermal data.
igb: Enable hwmon data output for thermal sensors via I2C.
Greg Rose (1):
igb: Enable SR-IOV configuration via PCI sysfs interface
Matthew Vick (5):
igb: Add support for SW timestamping
igb: Add timeout for PTP Tx work item
igb: Add mechanism for detecting latched hardware Rx timestamp
igb: Use in-kernel PTP_EV_PORT #define
igb: Free any held skb that should have been timestamped on remove
drivers/net/ethernet/intel/Kconfig | 14 +
drivers/net/ethernet/intel/e1000e/82571.c | 2 +
drivers/net/ethernet/intel/e1000e/Makefile | 2 +-
drivers/net/ethernet/intel/e1000e/defines.h | 55 +++
drivers/net/ethernet/intel/e1000e/e1000.h | 56 ++-
drivers/net/ethernet/intel/e1000e/ethtool.c | 171 ++++++-
drivers/net/ethernet/intel/e1000e/hw.h | 13 +
drivers/net/ethernet/intel/e1000e/ich8lan.c | 26 +-
drivers/net/ethernet/intel/e1000e/netdev.c | 487 +++++++++++++++++-
drivers/net/ethernet/intel/e1000e/ptp.c | 276 +++++++++++
drivers/net/ethernet/intel/igb/Makefile | 2 +-
drivers/net/ethernet/intel/igb/e1000_82575.c | 140 ++++++
drivers/net/ethernet/intel/igb/e1000_82575.h | 26 +-
drivers/net/ethernet/intel/igb/e1000_defines.h | 13 +
drivers/net/ethernet/intel/igb/e1000_hw.h | 20 +
drivers/net/ethernet/intel/igb/e1000_regs.h | 16 +
drivers/net/ethernet/intel/igb/igb.h | 46 +-
drivers/net/ethernet/intel/igb/igb_ethtool.c | 11 +
drivers/net/ethernet/intel/igb/igb_hwmon.c | 242 +++++++++
drivers/net/ethernet/intel/igb/igb_main.c | 653 +++++++++++++++++++++----
drivers/net/ethernet/intel/igb/igb_ptp.c | 65 ++-
21 files changed, 2191 insertions(+), 145 deletions(-)
create mode 100644 drivers/net/ethernet/intel/e1000e/ptp.c
create mode 100644 drivers/net/ethernet/intel/igb/igb_hwmon.c
--
1.7.11.7
^ permalink raw reply
* [net-next 03/14] e1000e: add support for hardware timestamping on some devices
From: Jeff Kirsher @ 2013-01-17 11:35 UTC (permalink / raw)
To: davem; +Cc: Bruce Allan, netdev, gospo, sassmann, Jeff Kirsher
In-Reply-To: <1358422519-20981-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Bruce Allan <bruce.w.allan@intel.com>
On 82574, 82583, 82579, I217 and I218 add support for hardware time
stamping of all or no Rx packets and Tx packets which have the
SKBTX_HW_TSTAMP flag set. Update the .get_ts_info ethtool operation to
report the supported time stamping modes, and enable and disable hardware
time stamping with the SIOCSHWTSTAMP ioctl.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/e1000e/82571.c | 2 +
drivers/net/ethernet/intel/e1000e/defines.h | 14 ++
drivers/net/ethernet/intel/e1000e/e1000.h | 46 +++-
drivers/net/ethernet/intel/e1000e/ethtool.c | 25 +-
drivers/net/ethernet/intel/e1000e/hw.h | 10 +
drivers/net/ethernet/intel/e1000e/ich8lan.c | 2 +
drivers/net/ethernet/intel/e1000e/netdev.c | 371 +++++++++++++++++++++++++++-
7 files changed, 462 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/intel/e1000e/82571.c b/drivers/net/ethernet/intel/e1000e/82571.c
index c77d010..587890d 100644
--- a/drivers/net/ethernet/intel/e1000e/82571.c
+++ b/drivers/net/ethernet/intel/e1000e/82571.c
@@ -2044,6 +2044,7 @@ const struct e1000_info e1000_82574_info = {
| FLAG_HAS_MSIX
| FLAG_HAS_JUMBO_FRAMES
| FLAG_HAS_WOL
+ | FLAG_HAS_HW_TIMESTAMP
| FLAG_APME_IN_CTRL3
| FLAG_HAS_SMART_POWER_DOWN
| FLAG_HAS_AMT
@@ -2065,6 +2066,7 @@ const struct e1000_info e1000_82583_info = {
.mac = e1000_82583,
.flags = FLAG_HAS_HW_VLAN_FILTER
| FLAG_HAS_WOL
+ | FLAG_HAS_HW_TIMESTAMP
| FLAG_APME_IN_CTRL3
| FLAG_HAS_SMART_POWER_DOWN
| FLAG_HAS_AMT
diff --git a/drivers/net/ethernet/intel/e1000e/defines.h b/drivers/net/ethernet/intel/e1000e/defines.h
index 3041a54..36f9fad 100644
--- a/drivers/net/ethernet/intel/e1000e/defines.h
+++ b/drivers/net/ethernet/intel/e1000e/defines.h
@@ -107,6 +107,7 @@
#define E1000_RXD_ERR_RXE 0x80 /* Rx Data Error */
#define E1000_RXD_SPC_VLAN_MASK 0x0FFF /* VLAN ID is in lower 12 bits */
+#define E1000_RXDEXT_STATERR_TST 0x00000100 /* Time Stamp taken */
#define E1000_RXDEXT_STATERR_CE 0x01000000
#define E1000_RXDEXT_STATERR_SE 0x02000000
#define E1000_RXDEXT_STATERR_SEQ 0x04000000
@@ -318,6 +319,7 @@
#define E1000_TXD_CMD_IP 0x02000000 /* IP packet */
#define E1000_TXD_CMD_TSE 0x04000000 /* TCP Seg enable */
#define E1000_TXD_STAT_TC 0x00000004 /* Tx Underrun */
+#define E1000_TXD_EXTCMD_TSTAMP 0x00000010 /* IEEE1588 Timestamp packet */
/* Transmit Control */
#define E1000_TCTL_EN 0x00000002 /* enable Tx */
@@ -536,6 +538,18 @@
#define E1000_RXCW_C 0x20000000 /* Receive config */
#define E1000_RXCW_SYNCH 0x40000000 /* Receive config synch */
+#define E1000_TSYNCTXCTL_VALID 0x00000001 /* Tx timestamp valid */
+#define E1000_TSYNCRXCTL_TYPE_ALL 0x08
+#define E1000_TSYNCTXCTL_ENABLED 0x00000010 /* enable Tx timestamping */
+
+#define E1000_TSYNCRXCTL_VALID 0x00000001 /* Rx timestamp valid */
+#define E1000_TSYNCRXCTL_TYPE_MASK 0x0000000E /* Rx type mask */
+#define E1000_TSYNCRXCTL_ENABLED 0x00000010 /* enable Rx timestamping */
+#define E1000_TSYNCRXCTL_SYSCFI 0x00000020 /* Sys clock frequency */
+
+#define E1000_TIMINCA_INCPERIOD_SHIFT 24
+#define E1000_TIMINCA_INCVALUE_MASK 0x00FFFFFF
+
/* PCI Express Control */
#define E1000_GCR_RXD_NO_SNOOP 0x00000001
#define E1000_GCR_RXDSCW_NO_SNOOP 0x00000002
diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h
index b89b181..dea9e55 100644
--- a/drivers/net/ethernet/intel/e1000e/e1000.h
+++ b/drivers/net/ethernet/intel/e1000e/e1000.h
@@ -41,6 +41,8 @@
#include <linux/pci-aspm.h>
#include <linux/crc32.h>
#include <linux/if_vlan.h>
+#include <linux/clocksource.h>
+#include <linux/net_tstamp.h>
#include "hw.h"
@@ -353,6 +355,7 @@ struct e1000_adapter {
u64 gorc_old;
u32 alloc_rx_buff_failed;
u32 rx_dma_failed;
+ u32 rx_hwtstamp_cleared;
unsigned int rx_ps_pages;
u16 rx_ps_bsize0;
@@ -402,6 +405,14 @@ struct e1000_adapter {
u16 tx_ring_count;
u16 rx_ring_count;
+
+ struct hwtstamp_config hwtstamp_config;
+ struct delayed_work systim_overflow_work;
+ struct sk_buff *tx_hwtstamp_skb;
+ struct work_struct tx_hwtstamp_work;
+ spinlock_t systim_lock; /* protects SYSTIML/H regsters */
+ struct cyclecounter cc;
+ struct timecounter tc;
};
struct e1000_info {
@@ -416,6 +427,38 @@ struct e1000_info {
const struct e1000_nvm_operations *nvm_ops;
};
+/* The system time is maintained by a 64-bit counter comprised of the 32-bit
+ * SYSTIMH and SYSTIML registers. How the counter increments (and therefore
+ * its resolution) is based on the contents of the TIMINCA register - it
+ * increments every incperiod (bits 31:24) clock ticks by incvalue (bits 23:0).
+ * For the best accuracy, the incperiod should be as small as possible. The
+ * incvalue is scaled by a factor as large as possible (while still fitting
+ * in bits 23:0) so that relatively small clock corrections can be made.
+ *
+ * As a result, a shift of INCVALUE_SHIFT_n is used to fit a value of
+ * INCVALUE_n into the TIMINCA register allowing 32+8+(24-INCVALUE_SHIFT_n)
+ * bits to count nanoseconds leaving the rest for fractional nonseconds.
+ */
+#define INCVALUE_96MHz 125
+#define INCVALUE_SHIFT_96MHz 17
+#define INCPERIOD_SHIFT_96MHz 2
+#define INCPERIOD_96MHz (12 >> INCPERIOD_SHIFT_96MHz)
+
+#define INCVALUE_25MHz 40
+#define INCVALUE_SHIFT_25MHz 18
+#define INCPERIOD_25MHz 1
+
+/* Another drawback of scaling the incvalue by a large factor is the
+ * 64-bit SYSTIM register overflows more quickly. This is dealt with
+ * by simply reading the clock before it overflows.
+ *
+ * Clock ns bits Overflows after
+ * ~~~~~~ ~~~~~~~ ~~~~~~~~~~~~~~~
+ * 96MHz 47-bit 2^(47-INCPERIOD_SHIFT_96MHz) / 10^9 / 3600 = 9.77 hrs
+ * 25MHz 46-bit 2^46 / 10^9 / 3600 = 19.55 hours
+ */
+#define E1000_SYSTIM_OVERFLOW_PERIOD (HZ * 60 * 60 * 4)
+
/* hardware capability, feature, and workaround flags */
#define FLAG_HAS_AMT (1 << 0)
#define FLAG_HAS_FLASH (1 << 1)
@@ -431,7 +474,7 @@ struct e1000_info {
#define FLAG_HAS_SMART_POWER_DOWN (1 << 11)
#define FLAG_IS_QUAD_PORT_A (1 << 12)
#define FLAG_IS_QUAD_PORT (1 << 13)
-/* reserved bit14 */
+#define FLAG_HAS_HW_TIMESTAMP (1 << 14)
#define FLAG_APME_IN_WUC (1 << 15)
#define FLAG_APME_IN_CTRL3 (1 << 16)
#define FLAG_APME_CHECK_PORT_B (1 << 17)
@@ -463,6 +506,7 @@ struct e1000_info {
#define FLAG2_NO_DISABLE_RX (1 << 10)
#define FLAG2_PCIM2PCI_ARBITER_WA (1 << 11)
#define FLAG2_DFLT_CRC_STRIPPING (1 << 12)
+#define FLAG2_CHECK_RX_HWTSTAMP (1 << 13)
#define E1000_RX_DESC_PS(R, i) \
(&(((union e1000_rx_desc_packet_split *)((R).desc))[i]))
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index 636ba09..f268cbc 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -108,6 +108,7 @@ static const struct e1000_stats e1000_gstrings_stats[] = {
E1000_STAT("dropped_smbus", stats.mgpdc),
E1000_STAT("rx_dma_failed", rx_dma_failed),
E1000_STAT("tx_dma_failed", tx_dma_failed),
+ E1000_STAT("rx_hwtstamp_cleared", rx_hwtstamp_cleared),
};
#define E1000_GLOBAL_STATS_LEN ARRAY_SIZE(e1000_gstrings_stats)
@@ -2182,6 +2183,28 @@ static int e1000e_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
return 0;
}
+static int e1000e_get_ts_info(struct net_device *netdev,
+ struct ethtool_ts_info *info)
+{
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+
+ ethtool_op_get_ts_info(netdev, info);
+
+ if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP))
+ return 0;
+
+ info->so_timestamping |= (SOF_TIMESTAMPING_TX_HARDWARE |
+ SOF_TIMESTAMPING_RX_HARDWARE |
+ SOF_TIMESTAMPING_RAW_HARDWARE);
+
+ info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
+
+ info->rx_filters = ((1 << HWTSTAMP_FILTER_NONE) |
+ (1 << HWTSTAMP_FILTER_ALL));
+
+ return 0;
+}
+
static const struct ethtool_ops e1000_ethtool_ops = {
.get_settings = e1000_get_settings,
.set_settings = e1000_set_settings,
@@ -2209,7 +2232,7 @@ static const struct ethtool_ops e1000_ethtool_ops = {
.get_coalesce = e1000_get_coalesce,
.set_coalesce = e1000_set_coalesce,
.get_rxnfc = e1000_get_rxnfc,
- .get_ts_info = ethtool_op_get_ts_info,
+ .get_ts_info = e1000e_get_ts_info,
.get_eee = e1000e_get_eee,
.set_eee = e1000e_set_eee,
};
diff --git a/drivers/net/ethernet/intel/e1000e/hw.h b/drivers/net/ethernet/intel/e1000e/hw.h
index 81afad5..8e7e803 100644
--- a/drivers/net/ethernet/intel/e1000e/hw.h
+++ b/drivers/net/ethernet/intel/e1000e/hw.h
@@ -60,6 +60,7 @@ enum e1e_registers {
E1000_EIAC_82574 = 0x000DC, /* Ext. Interrupt Auto Clear - RW */
E1000_IAM = 0x000E0, /* Interrupt Acknowledge Auto Mask */
E1000_IVAR = 0x000E4, /* Interrupt Vector Allocation - RW */
+ E1000_FEXTNVM7 = 0x000E4, /* Future Extended NVM 7 - RW */
E1000_EITR_82574_BASE = 0x000E8, /* Interrupt Throttling - RW */
#define E1000_EITR_82574(_n) (E1000_EITR_82574_BASE + (_n << 2))
E1000_LPIC = 0x000FC, /* Low Power Idle Control - RW */
@@ -241,6 +242,15 @@ enum e1e_registers {
#define E1000_PCH_RAICC(_n) (E1000_PCH_RAICC_BASE + ((_n) * 4))
#define E1000_CRC_OFFSET E1000_PCH_RAICC_BASE
E1000_HICR = 0x08F00, /* Host Interface Control */
+ E1000_SYSTIML = 0x0B600, /* System time register Low - RO */
+ E1000_SYSTIMH = 0x0B604, /* System time register High - RO */
+ E1000_TIMINCA = 0x0B608, /* Increment attributes register - RW */
+ E1000_TSYNCTXCTL = 0x0B614, /* Tx Time Sync Control register - RW */
+ E1000_TXSTMPL = 0x0B618, /* Tx timestamp value Low - RO */
+ E1000_TXSTMPH = 0x0B61C, /* Tx timestamp value High - RO */
+ E1000_TSYNCRXCTL = 0x0B620, /* Rx Time Sync Control register - RW */
+ E1000_RXSTMPL = 0x0B624, /* Rx timestamp Low - RO */
+ E1000_RXSTMPH = 0x0B628, /* Rx timestamp High - RO */
};
#define E1000_MAX_PHY_ADDR 4
diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c
index 3829f7f..50935ef 100644
--- a/drivers/net/ethernet/intel/e1000e/ich8lan.c
+++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c
@@ -4601,6 +4601,7 @@ const struct e1000_info e1000_pch2_info = {
.mac = e1000_pch2lan,
.flags = FLAG_IS_ICH
| FLAG_HAS_WOL
+ | FLAG_HAS_HW_TIMESTAMP
| FLAG_HAS_CTRLEXT_ON_LOAD
| FLAG_HAS_AMT
| FLAG_HAS_FLASH
@@ -4620,6 +4621,7 @@ const struct e1000_info e1000_pch_lpt_info = {
.mac = e1000_pch_lpt,
.flags = FLAG_IS_ICH
| FLAG_HAS_WOL
+ | FLAG_HAS_HW_TIMESTAMP
| FLAG_HAS_CTRLEXT_ON_LOAD
| FLAG_HAS_AMT
| FLAG_HAS_FLASH
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index bf2c84cf..c15b7e4 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -488,20 +488,87 @@ static int e1000_desc_unused(struct e1000_ring *ring)
}
/**
+ * e1000e_systim_to_hwtstamp - convert system time value to hw time stamp
+ * @adapter: board private structure
+ * @hwtstamps: time stamp structure to update
+ * @systim: unsigned 64bit system time value.
+ *
+ * Convert the system time value stored in the RX/TXSTMP registers into a
+ * hwtstamp which can be used by the upper level time stamping functions.
+ *
+ * The 'systim_lock' spinlock is used to protect the consistency of the
+ * system time value. This is needed because reading the 64 bit time
+ * value involves reading two 32 bit registers. The first read latches the
+ * value.
+ **/
+static void e1000e_systim_to_hwtstamp(struct e1000_adapter *adapter,
+ struct skb_shared_hwtstamps *hwtstamps,
+ u64 systim)
+{
+ u64 ns;
+ unsigned long flags;
+
+ spin_lock_irqsave(&adapter->systim_lock, flags);
+ ns = timecounter_cyc2time(&adapter->tc, systim);
+ spin_unlock_irqrestore(&adapter->systim_lock, flags);
+
+ memset(hwtstamps, 0, sizeof(*hwtstamps));
+ hwtstamps->hwtstamp = ns_to_ktime(ns);
+}
+
+/**
+ * e1000e_rx_hwtstamp - utility function which checks for Rx time stamp
+ * @adapter: board private structure
+ * @status: descriptor extended error and status field
+ * @skb: particular skb to include time stamp
+ *
+ * If the time stamp is valid, convert it into the timecounter ns value
+ * and store that result into the shhwtstamps structure which is passed
+ * up the network stack.
+ **/
+static void e1000e_rx_hwtstamp(struct e1000_adapter *adapter, u32 status,
+ struct sk_buff *skb)
+{
+ struct e1000_hw *hw = &adapter->hw;
+ u64 rxstmp;
+
+ if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP) ||
+ !(status & E1000_RXDEXT_STATERR_TST) ||
+ !(er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_VALID))
+ return;
+
+ /* The Rx time stamp registers contain the time stamp. No other
+ * received packet will be time stamped until the Rx time stamp
+ * registers are read. Because only one packet can be time stamped
+ * at a time, the register values must belong to this packet and
+ * therefore none of the other additional attributes need to be
+ * compared.
+ */
+ rxstmp = (u64)er32(RXSTMPL);
+ rxstmp |= (u64)er32(RXSTMPH) << 32;
+ e1000e_systim_to_hwtstamp(adapter, skb_hwtstamps(skb), rxstmp);
+
+ adapter->flags2 &= ~FLAG2_CHECK_RX_HWTSTAMP;
+}
+
+/**
* e1000_receive_skb - helper function to handle Rx indications
* @adapter: board private structure
- * @status: descriptor status field as written by hardware
+ * @staterr: descriptor extended error and status field as written by hardware
* @vlan: descriptor vlan field as written by hardware (no le/be conversion)
* @skb: pointer to sk_buff to be indicated to stack
**/
static void e1000_receive_skb(struct e1000_adapter *adapter,
struct net_device *netdev, struct sk_buff *skb,
- u8 status, __le16 vlan)
+ u32 staterr, __le16 vlan)
{
u16 tag = le16_to_cpu(vlan);
+
+ e1000e_rx_hwtstamp(adapter, staterr, skb);
+
skb->protocol = eth_type_trans(skb, netdev);
- if (status & E1000_RXD_STAT_VP)
+ if (staterr & E1000_RXD_STAT_VP)
__vlan_hwaccel_put_tag(skb, tag);
napi_gro_receive(&adapter->napi, skb);
@@ -1092,6 +1159,41 @@ static void e1000_print_hw_hang(struct work_struct *work)
}
/**
+ * e1000e_tx_hwtstamp_work - check for Tx time stamp
+ * @work: pointer to work struct
+ *
+ * This work function polls the TSYNCTXCTL valid bit to determine when a
+ * timestamp has been taken for the current stored skb. The timestamp must
+ * be for this skb because only one such packet is allowed in the queue.
+ */
+static void e1000e_tx_hwtstamp_work(struct work_struct *work)
+{
+ struct e1000_adapter *adapter = container_of(work, struct e1000_adapter,
+ tx_hwtstamp_work);
+ struct e1000_hw *hw = &adapter->hw;
+
+ if (!adapter->tx_hwtstamp_skb)
+ return;
+
+ if (er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_VALID) {
+ struct skb_shared_hwtstamps shhwtstamps;
+ u64 txstmp;
+
+ txstmp = er32(TXSTMPL);
+ txstmp |= (u64)er32(TXSTMPH) << 32;
+
+ e1000e_systim_to_hwtstamp(adapter, &shhwtstamps, txstmp);
+
+ skb_tstamp_tx(adapter->tx_hwtstamp_skb, &shhwtstamps);
+ dev_kfree_skb_any(adapter->tx_hwtstamp_skb);
+ adapter->tx_hwtstamp_skb = NULL;
+ } else {
+ /* reschedule to check later */
+ schedule_work(&adapter->tx_hwtstamp_work);
+ }
+}
+
+/**
* e1000_clean_tx_irq - Reclaim resources after transmit completes
* @tx_ring: Tx descriptor ring
*
@@ -1345,8 +1447,8 @@ copydone:
cpu_to_le16(E1000_RXDPS_HDRSTAT_HDRSP))
adapter->rx_hdr_split++;
- e1000_receive_skb(adapter, netdev, skb,
- staterr, rx_desc->wb.middle.vlan);
+ e1000_receive_skb(adapter, netdev, skb, staterr,
+ rx_desc->wb.middle.vlan);
next_desc:
rx_desc->wb.middle.status_error &= cpu_to_le32(~0xFF);
@@ -3304,6 +3406,159 @@ static void e1000e_setup_rss_hash(struct e1000_adapter *adapter)
}
/**
+ * e1000e_get_base_timinca - get default SYSTIM time increment attributes
+ * @adapter: board private structure
+ * @timinca: pointer to returned time increment attributes
+ *
+ * Get attributes for incrementing the System Time Register SYSTIML/H at
+ * the default base frequency, and set the cyclecounter shift value.
+ **/
+static s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
+{
+ struct e1000_hw *hw = &adapter->hw;
+ u32 incvalue, incperiod, shift;
+
+ /* Make sure clock is enabled on I217 before checking the frequency */
+ if ((hw->mac.type == e1000_pch_lpt) &&
+ !(er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_ENABLED) &&
+ !(er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_ENABLED)) {
+ u32 fextnvm7 = er32(FEXTNVM7);
+
+ if (!(fextnvm7 & (1 << 0))) {
+ ew32(FEXTNVM7, fextnvm7 | (1 << 0));
+ e1e_flush();
+ }
+ }
+
+ switch (hw->mac.type) {
+ case e1000_pch2lan:
+ case e1000_pch_lpt:
+ /* On I217, the clock frequency is 25MHz or 96MHz as
+ * indicated by the System Clock Frequency Indication
+ */
+ if ((hw->mac.type != e1000_pch_lpt) ||
+ (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI)) {
+ /* Stable 96MHz frequency */
+ incperiod = INCPERIOD_96MHz;
+ incvalue = INCVALUE_96MHz;
+ shift = INCVALUE_SHIFT_96MHz;
+ adapter->cc.shift = shift + INCPERIOD_SHIFT_96MHz;
+ break;
+ }
+ /* fall-through */
+ case e1000_82574:
+ case e1000_82583:
+ /* Stable 25MHz frequency */
+ incperiod = INCPERIOD_25MHz;
+ incvalue = INCVALUE_25MHz;
+ shift = INCVALUE_SHIFT_25MHz;
+ adapter->cc.shift = shift;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ *timinca = ((incperiod << E1000_TIMINCA_INCPERIOD_SHIFT) |
+ ((incvalue << shift) & E1000_TIMINCA_INCVALUE_MASK));
+
+ return 0;
+}
+
+/**
+ * e1000e_config_hwtstamp - configure the hwtstamp registers and enable/disable
+ * @adapter: board private structure
+ *
+ * Outgoing time stamping can be enabled and disabled. Play nice and
+ * disable it when requested, although it shouldn't cause any overhead
+ * when no packet needs it. At most one packet in the queue may be
+ * marked for time stamping, otherwise it would be impossible to tell
+ * for sure to which packet the hardware time stamp belongs.
+ *
+ * Incoming time stamping has to be configured via the hardware filters.
+ * Not all combinations are supported, in particular event type has to be
+ * specified. Matching the kind of event packet is not supported, with the
+ * exception of "all V2 events regardless of level 2 or 4".
+ **/
+static int e1000e_config_hwtstamp(struct e1000_adapter *adapter)
+{
+ struct e1000_hw *hw = &adapter->hw;
+ struct hwtstamp_config *config = &adapter->hwtstamp_config;
+ u32 tsync_tx_ctl = E1000_TSYNCTXCTL_ENABLED;
+ u32 tsync_rx_ctl = E1000_TSYNCRXCTL_ENABLED;
+ u32 regval;
+ s32 ret_val;
+
+ if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP))
+ return -EINVAL;
+
+ /* flags reserved for future extensions - must be zero */
+ if (config->flags)
+ return -EINVAL;
+
+ switch (config->tx_type) {
+ case HWTSTAMP_TX_OFF:
+ tsync_tx_ctl = 0;
+ break;
+ case HWTSTAMP_TX_ON:
+ break;
+ default:
+ return -ERANGE;
+ }
+
+ switch (config->rx_filter) {
+ case HWTSTAMP_FILTER_NONE:
+ tsync_rx_ctl = 0;
+ break;
+ case HWTSTAMP_FILTER_ALL:
+ tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_ALL;
+ config->rx_filter = HWTSTAMP_FILTER_ALL;
+ break;
+ default:
+ return -ERANGE;
+ }
+
+ /* enable/disable Tx h/w time stamping */
+ regval = er32(TSYNCTXCTL);
+ regval &= ~E1000_TSYNCTXCTL_ENABLED;
+ regval |= tsync_tx_ctl;
+ ew32(TSYNCTXCTL, regval);
+ if ((er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_ENABLED) !=
+ (regval & E1000_TSYNCTXCTL_ENABLED)) {
+ e_err("Timesync Tx Control register not set as expected\n");
+ return -EAGAIN;
+ }
+
+ /* enable/disable Rx h/w time stamping */
+ regval = er32(TSYNCRXCTL);
+ regval &= ~(E1000_TSYNCRXCTL_ENABLED | E1000_TSYNCRXCTL_TYPE_MASK);
+ regval |= tsync_rx_ctl;
+ ew32(TSYNCRXCTL, regval);
+ if ((er32(TSYNCRXCTL) & (E1000_TSYNCRXCTL_ENABLED |
+ E1000_TSYNCRXCTL_TYPE_MASK)) !=
+ (regval & (E1000_TSYNCRXCTL_ENABLED |
+ E1000_TSYNCRXCTL_TYPE_MASK))) {
+ e_err("Timesync Rx Control register not set as expected\n");
+ return -EAGAIN;
+ }
+
+ /* Clear TSYNCRXCTL_VALID & TSYNCTXCTL_VALID bit */
+ regval = er32(RXSTMPH);
+ regval = er32(TXSTMPH);
+
+ /* Get and set the System Time Register SYSTIM base frequency */
+ ret_val = e1000e_get_base_timinca(adapter, ®val);
+ if (ret_val)
+ return ret_val;
+ ew32(TIMINCA, regval);
+
+ /* reset the ns time counter */
+ timecounter_init(&adapter->tc, &adapter->cc,
+ ktime_to_ns(ktime_get_real()));
+
+ return 0;
+}
+
+/**
* e1000_configure - configure the hardware for Rx and Tx
* @adapter: private board structure
**/
@@ -3529,6 +3784,9 @@ void e1000e_reset(struct e1000_adapter *adapter)
e1000e_reset_adaptive(hw);
+ /* initialize systim and reset the ns time counter */
+ e1000e_config_hwtstamp(adapter);
+
if (!netif_running(adapter->netdev) &&
!test_bit(__E1000_TESTING, &adapter->state)) {
e1000_power_down_phy(adapter);
@@ -3665,6 +3923,24 @@ void e1000e_reinit_locked(struct e1000_adapter *adapter)
}
/**
+ * e1000e_cyclecounter_read - read raw cycle counter (used by time counter)
+ * @cc: cyclecounter structure
+ **/
+static cycle_t e1000e_cyclecounter_read(const struct cyclecounter *cc)
+{
+ struct e1000_adapter *adapter = container_of(cc, struct e1000_adapter,
+ cc);
+ struct e1000_hw *hw = &adapter->hw;
+ cycle_t systim;
+
+ /* latch SYSTIMH on read of SYSTIML */
+ systim = (cycle_t)er32(SYSTIML);
+ systim |= (cycle_t)er32(SYSTIMH) << 32;
+
+ return systim;
+}
+
+/**
* e1000_sw_init - Initialize general software structures (struct e1000_adapter)
* @adapter: board private structure to initialize
*
@@ -3690,6 +3966,17 @@ static int e1000_sw_init(struct e1000_adapter *adapter)
if (e1000_alloc_queues(adapter))
return -ENOMEM;
+ /* Setup hardware time stamping cyclecounter */
+ if (adapter->flags & FLAG_HAS_HW_TIMESTAMP) {
+ adapter->cc.read = e1000e_cyclecounter_read;
+ adapter->cc.mask = CLOCKSOURCE_MASK(64);
+ adapter->cc.mult = 1;
+ /* cc.shift set in e1000e_get_base_tininca() */
+
+ spin_lock_init(&adapter->systim_lock);
+ INIT_WORK(&adapter->tx_hwtstamp_work, e1000e_tx_hwtstamp_work);
+ }
+
/* Explicitly disable IRQ since the NIC can be in any state. */
e1000_irq_disable(adapter);
@@ -4597,6 +4884,17 @@ link_up:
if (adapter->flags2 & FLAG2_CHECK_PHY_HANG)
e1000e_check_82574_phy_workaround(adapter);
+ /* Clear valid timestamp stuck in RXSTMPL/H due to a Rx error */
+ if (adapter->hwtstamp_config.rx_filter != HWTSTAMP_FILTER_NONE) {
+ if ((adapter->flags2 & FLAG2_CHECK_RX_HWTSTAMP) &&
+ (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_VALID)) {
+ er32(RXSTMPH);
+ adapter->rx_hwtstamp_cleared++;
+ } else {
+ adapter->flags2 |= FLAG2_CHECK_RX_HWTSTAMP;
+ }
+ }
+
/* Reset the timer */
if (!test_bit(__E1000_DOWN, &adapter->state))
mod_timer(&adapter->watchdog_timer,
@@ -4608,6 +4906,7 @@ link_up:
#define E1000_TX_FLAGS_TSO 0x00000004
#define E1000_TX_FLAGS_IPV4 0x00000008
#define E1000_TX_FLAGS_NO_FCS 0x00000010
+#define E1000_TX_FLAGS_HWTSTAMP 0x00000020
#define E1000_TX_FLAGS_VLAN_MASK 0xffff0000
#define E1000_TX_FLAGS_VLAN_SHIFT 16
@@ -4866,6 +5165,11 @@ static void e1000_tx_queue(struct e1000_ring *tx_ring, int tx_flags, int count)
if (unlikely(tx_flags & E1000_TX_FLAGS_NO_FCS))
txd_lower &= ~(E1000_TXD_CMD_IFCS);
+ if (unlikely(tx_flags & E1000_TX_FLAGS_HWTSTAMP)) {
+ txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D;
+ txd_upper |= E1000_TXD_EXTCMD_TSTAMP;
+ }
+
i = tx_ring->next_to_use;
do {
@@ -5089,7 +5393,15 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
count = e1000_tx_map(tx_ring, skb, first, adapter->tx_fifo_limit,
nr_frags);
if (count) {
- skb_tx_timestamp(skb);
+ if (unlikely((skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
+ !adapter->tx_hwtstamp_skb)) {
+ skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+ tx_flags |= E1000_TX_FLAGS_HWTSTAMP;
+ adapter->tx_hwtstamp_skb = skb_get(skb);
+ schedule_work(&adapter->tx_hwtstamp_work);
+ } else {
+ skb_tx_timestamp(skb);
+ }
netdev_sent_queue(netdev, skb->len);
e1000_tx_queue(tx_ring, tx_flags, count);
@@ -5317,6 +5629,43 @@ static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
return 0;
}
+/**
+ * e1000e_hwtstamp_ioctl - control hardware time stamping
+ * @netdev: network interface device structure
+ * @ifreq: interface request
+ *
+ * Outgoing time stamping can be enabled and disabled. Play nice and
+ * disable it when requested, although it shouldn't cause any overhead
+ * when no packet needs it. At most one packet in the queue may be
+ * marked for time stamping, otherwise it would be impossible to tell
+ * for sure to which packet the hardware time stamp belongs.
+ *
+ * Incoming time stamping has to be configured via the hardware filters.
+ * Not all combinations are supported, in particular event type has to be
+ * specified. Matching the kind of event packet is not supported, with the
+ * exception of "all V2 events regardless of level 2 or 4".
+ **/
+static int e1000e_hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr)
+{
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+ struct hwtstamp_config config;
+ int ret_val;
+
+ if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
+ return -EFAULT;
+
+ adapter->hwtstamp_config = config;
+
+ ret_val = e1000e_config_hwtstamp(adapter);
+ if (ret_val)
+ return ret_val;
+
+ config = adapter->hwtstamp_config;
+
+ return copy_to_user(ifr->ifr_data, &config,
+ sizeof(config)) ? -EFAULT : 0;
+}
+
static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
{
switch (cmd) {
@@ -5324,6 +5673,8 @@ static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
case SIOCGMIIREG:
case SIOCSMIIREG:
return e1000_mii_ioctl(netdev, ifr, cmd);
+ case SIOCSHWTSTAMP:
+ return e1000e_hwtstamp_ioctl(netdev, ifr);
default:
return -EOPNOTSUPP;
}
@@ -6380,6 +6731,14 @@ static void e1000_remove(struct pci_dev *pdev)
cancel_work_sync(&adapter->update_phy_task);
cancel_work_sync(&adapter->print_hang_task);
+ if (adapter->flags & FLAG_HAS_HW_TIMESTAMP) {
+ cancel_work_sync(&adapter->tx_hwtstamp_work);
+ if (adapter->tx_hwtstamp_skb) {
+ dev_kfree_skb_any(adapter->tx_hwtstamp_skb);
+ adapter->tx_hwtstamp_skb = NULL;
+ }
+ }
+
if (!(netdev->flags & IFF_UP))
e1000_power_down_phy(adapter);
--
1.7.11.7
^ 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