* [PATCH net] net: qmi_wwan: add TP-LINK HSUPA Modem MA180
From: Bjørn Mork @ 2013-01-15 9:19 UTC (permalink / raw)
To: netdev; +Cc: linux-usb, hockseng leow, Thomas Schäfer, Bjørn Mork
In-Reply-To: <201301122122.07307.tschaefer@t-online.de>
The driver description files gives these names to the vendor specific
functions on this modem:
Diagnostics VID_2357&PID_0201&MI_00
NMEA VID_2357&PID_0201&MI_01
Modem VID_2357&PID_0201&MI_03
Networkcard VID_2357&PID_0201&MI_04
The "Networkcard" function has been verified to support these QMI
services:
ctl (1.3)
wds (1.3)
dms (1.2)
nas (1.0)
Reported-by: Thomas Schäfer <tschaefer@t-online.de>
Signed-off-by: Bjørn Mork <bjorn@mork.no>
---
drivers/net/usb/qmi_wwan.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c
index 6a1ca50..c434108 100644
--- a/drivers/net/usb/qmi_wwan.c
+++ b/drivers/net/usb/qmi_wwan.c
@@ -459,6 +459,7 @@ static const struct usb_device_id products[] = {
{QMI_FIXED_INTF(0x1199, 0x68a2, 19)}, /* Sierra Wireless MC7710 in QMI mode */
{QMI_FIXED_INTF(0x1199, 0x901c, 8)}, /* Sierra Wireless EM7700 */
{QMI_FIXED_INTF(0x1bbb, 0x011e, 4)}, /* Telekom Speedstick LTE II (Alcatel One Touch L100V LTE) */
+ {QMI_FIXED_INTF(0x2357, 0x0201, 4)}, /* TP-LINK HSUPA Modem MA180 */
/* 4. Gobi 1000 devices */
{QMI_GOBI1K_DEVICE(0x05c6, 0x9212)}, /* Acer Gobi Modem Device */
--
1.7.10.4
^ permalink raw reply related
* what is the use of poll function int a net_device struct
From: horseriver @ 2013-01-14 14:21 UTC (permalink / raw)
To: netdev
hi:
Where is the handler function defined for an netcard's IRQ ?
When an IRQ has reached, how to wake_up the wait_queue of the socket?
thanks1
^ permalink raw reply
* Re: [PATCH 15/15] batman-adv: unbloat batadv_priv if debug is not enabled
From: Antonio Quartulli @ 2013-01-15 8:53 UTC (permalink / raw)
To: Joe Perches
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
b.a.t.m.a.n-ZwoEplunGu2X36UT3dwllkB+6BGkLq7r, Marek Lindner,
davem-fT/PcQaiUtIeIZ0/mPfg9Q
In-Reply-To: <1358185000.19400.15.camel@joe-AO722>
[-- Attachment #1: Type: text/plain, Size: 1220 bytes --]
Hi Joe,
On Mon, Jan 14, 2013 at 09:36:40 -0800, Joe Perches wrote:
> On Mon, 2013-01-14 at 09:41 +1000, Antonio Quartulli wrote:
> > diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
> []
> > @@ -284,12 +284,16 @@ struct batadv_priv {
> > atomic_t gw_bandwidth; /* gw bandwidth */
> > atomic_t orig_interval; /* uint */
> > atomic_t hop_penalty; /* uint */
> > +#ifdef CONFIG_BATMAN_ADV_DEBUG
> > atomic_t log_level; /* uint */
> > +#endif
> > atomic_t bcast_seqno;
> > atomic_t bcast_queue_left;
> > atomic_t batman_queue_left;
> > char num_ifaces;
> > +#ifdef CONFIG_BATMAN_ADV_DEBUG
> > struct batadv_debug_log *debug_log;
> > +#endif
> > struct kobject *mesh_obj;
> > struct dentry *debug_dir;
> > struct hlist_head forw_bat_list;
>
> How about moving one of these so
> there's only one #ifdef block.
Yeah. This is the main struct and actually it still needs some more housekeeping
(and some love).
It was initially sorted in another way, but as soon as we added new features it
lost its shape.
We will take care of that.
Thanks a lot!
Cheers,
--
Antonio Quartulli
..each of us alone is worth nothing..
Ernesto "Che" Guevara
[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply
* Re: [PATCH] CDC_NCM adding support IFF_NOARP for infineon modem platform
From: Bjørn Mork @ 2013-01-15 8:34 UTC (permalink / raw)
To: David Miller; +Cc: dcbw, cpuwolf, gregkh, alexey.orishko, linux-usb, netdev
In-Reply-To: <20130114.133412.479351928049237212.davem@davemloft.net>
David Miller <davem@davemloft.net> writes:
> From: Dan Williams <dcbw@redhat.com>
>
>> IFF_NOARP is already done for other WWAN devices (sierra_net, hso,
>> cdc-ether, cdc-phonet, lg-vl600, etc) so there is some precedent. Some
>> drivers (phonet, hso) set *both* POINTTOPOINT and NOARP. Is that
>> redundant, and should all WWAN drivers be moved to only POINTTOPOINT?
>>
>> (aside: usbnet has FLAG_POINTTOPOINT, but that's nothing to do with
>> IFF_POINTTOPOINT, it only controls whether the interface is named usbX
>> or ethX. Confusing.)
>
> I can't answer any of your questions unless you tell me what the
> real limitation of these devices is.
>
> For the second time, is the problem that these devices cannot
> support broadcast packets properly?
The main problem is that these devices don't support ethernet. They
support IP (v4 and _maybe_ v6) with an ethernet header. Many of them
will do ARP (and IPv6 ND) as well to complete the picture, but some of
them don't and that's what these drivers try to deal with.
Note that most of the devices will run a DHCP server, so there is some
sort of IP broadcast support. Whether that qualifies as proper ethernet
broadcast support is another question...
These devices are attempting to bridge an IP-only point-to-point
interface and an ethernet over USB interface, with the intention to make
the point-to-point interface look like ethernet to applications and
users. This is of course always going to be imperfect. But I believe
that we should aim to help the firmware achive this goal when writing
drivers instead of working against it. Setting IFF_NOARP and not
IFF_POINTTOPOINT is one way to do that.
Bjørn
^ permalink raw reply
* Re: [PATCH net-next 09/10] bnx2x: Added FW GRO bridging support
From: Yuval Mintz @ 2013-01-15 7:28 UTC (permalink / raw)
To: Eric Dumazet; +Cc: davem, netdev, eilong, ariele
In-Reply-To: <1358189075.8744.3320.camel@edumazet-glaptop>
On 01/14/2013 08:44 PM, Eric Dumazet wrote:
> On Mon, 2013-01-14 at 09:22 -0800, Eric Dumazet wrote:
>
> What is the value of gso_segs ?
>
> The reason I am pointing this out is the recent change in commit
> 1def9238d4aa2146924994aa4b7dc861f03b9362
> (net_sched: more precise pkt_len computation)
>
> bnx2x not setting gso_segs means that qdisc accounting on ingress is
> completely wrong.
Hi Eric,
First I just want to state that you're totally correct about the gso_segs -
bnx2x is not setting it correctly (it's currently totally omitted), and so
it would incorrectly affect the accounting.
However, notice this behaviour was not introduced in this patch -
Since 621b4d6 bnx2x is using FW GRO, overriding the kernel's GRO implementation:
As the bnx2x driver is supplied with the aggregated packet from its FW,
the bnx2x passes a value in the `gso_size' field of its skb, causing
`skb_is_gso' to return `true'.
This will cause the aggregated skb to override the GRO machinations
(in `dev_gro_receive'), overriding all calls to `gro_receive' and thus also
the call to `skb_gro_receive' which whould have incremented `gso_segs'.
This patch actually tries to correct said behaviour, obviously failing
with the gso_segs but still improving the current state of bnx2x GRO
in bridging scenarios.
>> This looks weird to me. This should be called by GRO stack only.
I think this is the main question - we could try and implement this
inside the network-core itself, but as said behaviour is unique to the
bnx2x driver (correct me if I'm wrong, but I'm unaware of any other
driver which does GRO without the kernel GRO implementation), the
solution is specially tailored for the bnx2x driver.
We could either:
1. Continue with this patch, later sending a patch correcting gso_segs,
as this is not a new issue.
2. Send a V2 patch-series which will also set gso_segs correctly.
3. Send a V2 patch-series which omits this patch, and later send an RFC
for some kernel implementation which fixes the issue.
Your thoughts on this matter will be greatly appreciated.
Thanks,
Yuval
^ permalink raw reply
* Re: [PATCH net-next] ipip: do not use 'err' variable for setting return value
From: David Miller @ 2013-01-15 8:20 UTC (permalink / raw)
To: sakiwit; +Cc: netdev
In-Reply-To: <1358237650-15953-1-git-send-email-sakiwit@gmail.com>
From: Jean Sacren <sakiwit@gmail.com>
Date: Tue, 15 Jan 2013 01:14:10 -0700
> It's a bit confusing to set 'err' variable inside and outside of the
> block for the return value of ipip_err(). We don't even need this
> variable, so clean it all up to spare declaration as well as bunch of
> unnecessary initializations and jumps.
>
> Signed-off-by: Jean Sacren <sakiwit@gmail.com>
It makes the non-error path straight line code.
The code as-is is fine, and I'm not applying this patch, sorry.
^ permalink raw reply
* [PATCH net-next] ipip: do not use 'err' variable for setting return value
From: Jean Sacren @ 2013-01-15 8:14 UTC (permalink / raw)
To: netdev
It's a bit confusing to set 'err' variable inside and outside of the
block for the return value of ipip_err(). We don't even need this
variable, so clean it all up to spare declaration as well as bunch of
unnecessary initializations and jumps.
Signed-off-by: Jean Sacren <sakiwit@gmail.com>
---
net/ipv4/ipip.c | 20 +++++++-------------
1 files changed, 7 insertions(+), 13 deletions(-)
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 191fc24..d604824 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -343,7 +343,6 @@ static int ipip_err(struct sk_buff *skb, u32 info)
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct ip_tunnel *t;
- int err;
switch (type) {
default:
@@ -372,40 +371,35 @@ static int ipip_err(struct sk_buff *skb, u32 info)
break;
}
- err = -ENOENT;
t = ipip_tunnel_lookup(dev_net(skb->dev), iph->daddr, iph->saddr);
if (t == NULL)
- goto out;
+ return -ENOENT;
if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
ipv4_update_pmtu(skb, dev_net(skb->dev), info,
t->dev->ifindex, 0, IPPROTO_IPIP, 0);
- err = 0;
- goto out;
+ return 0;
}
if (type == ICMP_REDIRECT) {
ipv4_redirect(skb, dev_net(skb->dev), t->dev->ifindex, 0,
IPPROTO_IPIP, 0);
- err = 0;
- goto out;
+ return 0;
}
if (t->parms.iph.daddr == 0)
- goto out;
+ return -ENOENT;
- err = 0;
if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
- goto out;
+ return 0;
if (time_before(jiffies, t->err_time + IPTUNNEL_ERR_TIMEO))
t->err_count++;
else
t->err_count = 1;
- t->err_time = jiffies;
-out:
- return err;
+ t->err_time = jiffies;
+ return 0;
}
static int ipip_rcv(struct sk_buff *skb)
^ permalink raw reply related
* Re: r8169 rx_missed increasing in bursts (regression)
From: Timo Teras @ 2013-01-15 8:11 UTC (permalink / raw)
To: Francois Romieu; +Cc: netdev
In-Reply-To: <20130109191456.0888ac75@vostro>
On Wed, 9 Jan 2013 19:14:56 +0200 Timo Teras <timo.teras@iki.fi> wrote:
> It seems that the rx_missed is not directly related to traffic amount.
> At times the box is handling easily 10000+ pps, while packet loss can
> happen at other times on 4000-8000pps levels.
>
> Generally time_squeeze does not happen, and the box is at 20-30%
> softirq. Some times time_squeeze bumps up with one (within a one
> second interval) or two and packet loss does not happen.
>
> When rx_missed is getting bumped, time_squeeze goes up with 1-3, and
> rx_missed goes up with 50-1000 packets. Usually around 200 packets. (1
> second sampling period)
>
> I did find a strong correlation that rx_misses happen usually when the
> box has dropped a packet due to iptables DROP/REJECT rule, or some
> other reason (e.g. I'm seeing once in a while dmesg contain:
> "nf_ct_sip: dropping packet").
>
> Any ideas why a netfilter packet drop might cause netdevice rx to
> stall long enough to saturate the hardware receive queue?
Ok, found the main culprit. Embarassingly, there was LOG targets, which
wound up writing to ttyS0. That apparently is synchronous operation, so
the printk's made the napi poller choke whenever packets got dropped.
However, I do still observe minor packet drops on specific loads.
Increasing the rx fifo size helped noticeably, but even at 512 it still
does some drops. Seems that Realtek's r8168 driver has rx fifo size of
1024.
Would it be feasible to make the fifo size a module parameter, or
somehow autotuned according to available system memory?
I also noticed that since recent commits, the dirty_rx is always
identical cur_rx. Basically, dirty_rx can be just removed. Should I
send a patch for this?
- Timo
^ permalink raw reply
* Re: Information needed on the kernel version which supports ----
From: Stephen Hemminger @ 2013-01-15 6:10 UTC (permalink / raw)
To: sandeepacs
Cc: vyasevic, mst, netdev, stephen, bridge, shmulik.ladkani,
shemminger, davem
In-Reply-To: <50F3D5FD.6090608@tataelxsi.co.in>
[-- Attachment #1: Type: text/plain, Size: 1701 bytes --]
1. VLAN filtering in bridge is still under review, changes are requested.
Please see netdev patchwork to check for state of upcoming stuff.
2. I don't remember IGMP snooping is part of that work.
On Mon, Jan 14, 2013 at 1:55 AM, Sandeepa C S <sandeepacs@tataelxsi.co.in>wrote:
> Dear All,
>
> We are trying to add vlan support for the IGMP snooping feature in the
> kernel version 2.6.39
> But from the below thread discussion we understood that vlan support is
> already in the bridge.
> http://www.spinics.net/lists/**netdev/msg221800.html<http://www.spinics.net/lists/netdev/msg221800.html>
>
> Could you please let us know the below informations,
> 1) On what kernel version the VLAN support for multicast groups is being
> done
> 2)On what flavors it is supported, is it available with Debian
> 3)How can we get/download the supported patch files for reference.
>
> --
> Thanks & Regards,
> Sandeepa cs.
>
> Notice: The information contained in this e-mail message and/or
> attachments to it may contain confidential or privileged information. If
> you are not the intended recipient, any dissemination, use, review,
> distribution, printing or copying of the information contained in this
> e-mail message and/or attachments to it are strictly prohibited. If you
> have received this communication in error, please notify us by reply e-mail
> or telephone and immediately and permanently delete the message and any
> attachments. Thank you
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/**majordomo-info.html<http://vger.kernel.org/majordomo-info.html>
>
[-- Attachment #2: Type: text/html, Size: 2277 bytes --]
^ permalink raw reply
* [PATCH] usbnet: dm9601: Fix incorrect command
From: Tushar Behera @ 2013-01-15 5:39 UTC (permalink / raw)
To: linux-usb, linux-kernel, netdev; +Cc: gregkh, jacmet, patches, Ming Lei
commit 24b1042c4eb2 ("usbnet: dm9601: apply introduced usb command
APIs") removes the distiction between DM_WRITE_REG and DM_WRITE_REGS
command. The distiction is reintroduced to the driver so that the
functionality of the driver remains same.
CC: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Tushar Behera <tushar.behera@linaro.org>
---
drivers/net/usb/dm9601.c | 22 ++++++----------------
1 files changed, 6 insertions(+), 16 deletions(-)
diff --git a/drivers/net/usb/dm9601.c b/drivers/net/usb/dm9601.c
index 3f554c1..011410f 100644
--- a/drivers/net/usb/dm9601.c
+++ b/drivers/net/usb/dm9601.c
@@ -53,7 +53,6 @@
#define DM_RX_OVERHEAD 7 /* 3 byte header + 4 byte crc tail */
#define DM_TIMEOUT 1000
-
static int dm_read(struct usbnet *dev, u8 reg, u16 length, void *data)
{
int err;
@@ -84,32 +83,23 @@ static int dm_write(struct usbnet *dev, u8 reg, u16 length, void *data)
static int dm_write_reg(struct usbnet *dev, u8 reg, u8 value)
{
- return usbnet_write_cmd(dev, DM_WRITE_REGS,
+ return usbnet_write_cmd(dev, DM_WRITE_REG,
USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
value, reg, NULL, 0);
}
-static void dm_write_async_helper(struct usbnet *dev, u8 reg, u8 value,
- u16 length, void *data)
+static void dm_write_async(struct usbnet *dev, u8 reg, u16 length, void *data)
{
usbnet_write_cmd_async(dev, DM_WRITE_REGS,
USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
- value, reg, data, length);
-}
-
-static void dm_write_async(struct usbnet *dev, u8 reg, u16 length, void *data)
-{
- netdev_dbg(dev->net, "dm_write_async() reg=0x%02x length=%d\n", reg, length);
-
- dm_write_async_helper(dev, reg, 0, length, data);
+ 0, reg, data, length);
}
static void dm_write_reg_async(struct usbnet *dev, u8 reg, u8 value)
{
- netdev_dbg(dev->net, "dm_write_reg_async() reg=0x%02x value=0x%02x\n",
- reg, value);
-
- dm_write_async_helper(dev, reg, value, 0, NULL);
+ usbnet_write_cmd_async(dev, DM_WRITE_REG,
+ USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+ value, reg, NULL, 0);
}
static int dm_read_shared_word(struct usbnet *dev, int phy, u8 reg, __le16 *value)
--
1.7.4.1
^ permalink raw reply related
* Re: Information needed on the kernel version which supports ----
From: Sandeepa C S @ 2013-01-15 5:11 UTC (permalink / raw)
To: vyasevic
Cc: Balaji Foss, shemminger, mst, netdev, stephen, bridge,
shmulik.ladkani, davem, Satyanarayana Reddy M (D&D)
In-Reply-To: <50F41CB1.3040909@redhat.com>
Hi,
Even i could not find the changed files.
Could you please share if the changed files are available with you.
Thanks and Regards,
Sandeepa CS.
Vlad Yasevich wrote:
> On 01/14/2013 06:35 AM, Balaji Foss wrote:
>> >1) On what kernel version the VLAN support for multicast groups is
>> being
>>> done
>> Its currently in the net-next tree which would become 3.8
>
> I did not see any "Applied" messages from upstream, so I doubt its in
> any tree yet. The code is still in development.
>
> -vlad
>
>>
>>> 2)On what flavors it is supported, is it available with Debian
>>
>> If you want to test those changes right away, you will have to do a git
>> clone of the net-next repo and test those patches, else wait for the
>> distributions to package the 3.8 kernel.
>>
>>> 3)How can we get/download the supported patch files for reference.
>> you will have to make a clone of the net-next repo. For more information
>> look up tutorials on git.
>>
>> Cheers,
>> - Balaji
>>
>> On Mon, Jan 14, 2013 at 3:25 PM, Sandeepa C S
>> <sandeepacs@tataelxsi.co.in>wrote:
>>
>>> Dear All,
>>>
>>> We are trying to add vlan support for the IGMP snooping feature in the
>>> kernel version 2.6.39
>>> But from the below thread discussion we understood that vlan support is
>>> already in the bridge.
>>> http://www.spinics.net/lists/**netdev/msg221800.html<http://www.spinics.net/lists/netdev/msg221800.html>
>>>
>>>
>>> Could you please let us know the below informations,
>>> 1) On what kernel version the VLAN support for multicast groups is
>>> being
>>> done
>>> 2)On what flavors it is supported, is it available with Debian
>>> 3)How can we get/download the supported patch files for reference.
>>>
>>> --
>>> Thanks & Regards,
>>> Sandeepa cs.
>>>
>>> Notice: The information contained in this e-mail message and/or
>>> attachments to it may contain confidential or privileged
>>> information. If
>>> you are not the intended recipient, any dissemination, use, review,
>>> distribution, printing or copying of the information contained in this
>>> e-mail message and/or attachments to it are strictly prohibited. If you
>>> have received this communication in error, please notify us by reply
>>> e-mail
>>> or telephone and immediately and permanently delete the message and any
>>> attachments. Thank you
>>>
>>>
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe netdev" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at
>>> http://vger.kernel.org/**majordomo-info.html<http://vger.kernel.org/majordomo-info.html>
>>>
>>>
>>
>
>
> .
>
--
Thanks & Regards,
Sandeepa cs.
Notice: The information contained in this e-mail message and/or attachments to it may contain confidential or privileged information. If you are not the intended recipient, any dissemination, use, review, distribution, printing or copying of the information contained in this e-mail message and/or attachments to it are strictly prohibited. If you have received this communication in error, please notify us by reply e-mail or telephone and immediately and permanently delete the message and any attachments. Thank you
^ permalink raw reply
* Re: [PATCH 1/1] net: ipv4: refine the logic for determining daddr in ip_queue_xmit.
From: Eric Dumazet @ 2013-01-15 4:57 UTC (permalink / raw)
To: Harvey Yang
Cc: David S. Miller, Alexey Kuznetsov, James Morris,
Hideaki YOSHIFUJI, netdev, linux-kernel
In-Reply-To: <1358225255-30973-1-git-send-email-harvey.huawei.yang@gmail.com>
On Tue, 2013-01-15 at 12:47 +0800, Harvey Yang wrote:
> The destination address daddr is faddr if source route option is set, otherwise it is inet_daddr. So use if-else to assign the value.
>
> Signed-off-by: Harvey Yang <harvey.huawei.yang@gmail.com>
> ---
> net/ipv4/ip_output.c | 4 ++--
> 1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
> index 3e98ed2..22b738c 100644
> --- a/net/ipv4/ip_output.c
> +++ b/net/ipv4/ip_output.c
> @@ -349,10 +349,10 @@ int ip_queue_xmit(struct sk_buff *skb, struct flowi *fl)
> __be32 daddr;
>
> /* Use correct destination address if we have options. */
> - daddr = inet->inet_daddr;
> if (inet_opt && inet_opt->opt.srr)
> daddr = inet_opt->opt.faddr;
> -
> + else
> + daddr = inet->inet_daddr;
> /* If this fails, retransmit mechanism of transport layer will
> * keep trying until route appears or the connection times
> * itself out.
Current code is faster, srr is not often used.
^ permalink raw reply
* Re: [PATCH 1/1] net: ipv4: refine the logic for determining daddr in ip_queue_xmit.
From: David Miller @ 2013-01-15 4:48 UTC (permalink / raw)
To: harvey.huawei.yang; +Cc: kuznet, jmorris, yoshfuji, netdev, linux-kernel
In-Reply-To: <1358225255-30973-1-git-send-email-harvey.huawei.yang@gmail.com>
From: Harvey Yang <harvey.huawei.yang@gmail.com>
Date: Tue, 15 Jan 2013 12:47:35 +0800
> The destination address daddr is faddr if source route option is set, otherwise it is inet_daddr. So use if-else to assign the value.
>
> Signed-off-by: Harvey Yang <harvey.huawei.yang@gmail.com>
You are changing nothing in how this code behaves. And stylistically
I know it was coded this way on purpose, because I'm pretty sure I
wrote it.
^ permalink raw reply
* [PATCH 1/1] net: ipv4: refine the logic for determining daddr in ip_queue_xmit.
From: Harvey Yang @ 2013-01-15 4:47 UTC (permalink / raw)
To: David S. Miller, Alexey Kuznetsov, James Morris,
Hideaki YOSHIFUJI
Cc: netdev, linux-kernel, Harvey Yang
The destination address daddr is faddr if source route option is set, otherwise it is inet_daddr. So use if-else to assign the value.
Signed-off-by: Harvey Yang <harvey.huawei.yang@gmail.com>
---
net/ipv4/ip_output.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 3e98ed2..22b738c 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -349,10 +349,10 @@ int ip_queue_xmit(struct sk_buff *skb, struct flowi *fl)
__be32 daddr;
/* Use correct destination address if we have options. */
- daddr = inet->inet_daddr;
if (inet_opt && inet_opt->opt.srr)
daddr = inet_opt->opt.faddr;
-
+ else
+ daddr = inet->inet_daddr;
/* If this fails, retransmit mechanism of transport layer will
* keep trying until route appears or the connection times
* itself out.
--
1.7.1
^ permalink raw reply related
* Re: [PATCH 2/2 net] tg3: Fix crc errors on jumbo frame receive
From: David Miller @ 2013-01-15 3:09 UTC (permalink / raw)
To: mchan; +Cc: netdev, nsujir
In-Reply-To: <1358219460-5993-2-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 14 Jan 2013 19:11:00 -0800
> From: Nithin Nayak Sujir <nsujir@broadcom.com>
>
> TG3_PHY_AUXCTL_SMDSP_ENABLE/DISABLE macros do a blind write to the phy
> auxiliary control register and overwrite the EXT_PKT_LEN (bit 14) resulting
> in intermittent crc errors on jumbo frames with some link partners. Change
> the code to do a read/modify/write.
>
> Signed-off-by: Nithin Nayak Sujir <nsujir@broadcom.com>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
Applied.
^ permalink raw reply
* Re: [PATCH 1/2 net] tg3: Avoid null pointer dereference in tg3_interrupt in netconsole mode
From: David Miller @ 2013-01-15 3:09 UTC (permalink / raw)
To: mchan; +Cc: netdev, nsujir
In-Reply-To: <1358219460-5993-1-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 14 Jan 2013 19:10:59 -0800
> From: Nithin Nayak Sujir <nsujir@broadcom.com>
>
> When netconsole is enabled, logging messages generated during tg3_open
> can result in a null pointer dereference for the uninitialized tg3
> status block. Use the irq_sync flag to disable polling in the early
> stages. irq_sync is cleared when the driver is enabling interrupts after
> all initialization is completed.
>
> Signed-off-by: Nithin Nayak Sujir <nsujir@broadcom.com>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
Applied.
^ permalink raw reply
* [PATCH 1/2 net] tg3: Avoid null pointer dereference in tg3_interrupt in netconsole mode
From: Michael Chan @ 2013-01-15 3:10 UTC (permalink / raw)
To: davem; +Cc: netdev, nsujir
From: Nithin Nayak Sujir <nsujir@broadcom.com>
When netconsole is enabled, logging messages generated during tg3_open
can result in a null pointer dereference for the uninitialized tg3
status block. Use the irq_sync flag to disable polling in the early
stages. irq_sync is cleared when the driver is enabling interrupts after
all initialization is completed.
Signed-off-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
---
drivers/net/ethernet/broadcom/tg3.c | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 78ea90c..d326d92 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -6950,6 +6950,9 @@ static void tg3_poll_controller(struct net_device *dev)
int i;
struct tg3 *tp = netdev_priv(dev);
+ if (tg3_irq_sync(tp))
+ return;
+
for (i = 0; i < tp->irq_cnt; i++)
tg3_interrupt(tp->napi[i].irq_vec, &tp->napi[i]);
}
@@ -16367,6 +16370,7 @@ static int tg3_init_one(struct pci_dev *pdev,
tp->pm_cap = pm_cap;
tp->rx_mode = TG3_DEF_RX_MODE;
tp->tx_mode = TG3_DEF_TX_MODE;
+ tp->irq_sync = 1;
if (tg3_debug > 0)
tp->msg_enable = tg3_debug;
--
1.7.1
^ permalink raw reply related
* [PATCH 2/2 net] tg3: Fix crc errors on jumbo frame receive
From: Michael Chan @ 2013-01-15 3:11 UTC (permalink / raw)
To: davem; +Cc: netdev, nsujir
In-Reply-To: <1358219460-5993-1-git-send-email-mchan@broadcom.com>
From: Nithin Nayak Sujir <nsujir@broadcom.com>
TG3_PHY_AUXCTL_SMDSP_ENABLE/DISABLE macros do a blind write to the phy
auxiliary control register and overwrite the EXT_PKT_LEN (bit 14) resulting
in intermittent crc errors on jumbo frames with some link partners. Change
the code to do a read/modify/write.
Signed-off-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
---
---
drivers/net/ethernet/broadcom/tg3.c | 58 +++++++++++++++++++++--------------
1 files changed, 35 insertions(+), 23 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index d326d92..bdb0869 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -1283,14 +1283,26 @@ static int tg3_phy_auxctl_write(struct tg3 *tp, int reg, u32 set)
return tg3_writephy(tp, MII_TG3_AUX_CTRL, set | reg);
}
-#define TG3_PHY_AUXCTL_SMDSP_ENABLE(tp) \
- tg3_phy_auxctl_write((tp), MII_TG3_AUXCTL_SHDWSEL_AUXCTL, \
- MII_TG3_AUXCTL_ACTL_SMDSP_ENA | \
- MII_TG3_AUXCTL_ACTL_TX_6DB)
+static int tg3_phy_toggle_auxctl_smdsp(struct tg3 *tp, bool enable)
+{
+ u32 val;
+ int err;
+
+ err = tg3_phy_auxctl_read(tp, MII_TG3_AUXCTL_SHDWSEL_AUXCTL, &val);
-#define TG3_PHY_AUXCTL_SMDSP_DISABLE(tp) \
- tg3_phy_auxctl_write((tp), MII_TG3_AUXCTL_SHDWSEL_AUXCTL, \
- MII_TG3_AUXCTL_ACTL_TX_6DB);
+ if (err)
+ return err;
+ if (enable)
+
+ val |= MII_TG3_AUXCTL_ACTL_SMDSP_ENA;
+ else
+ val &= ~MII_TG3_AUXCTL_ACTL_SMDSP_ENA;
+
+ err = tg3_phy_auxctl_write((tp), MII_TG3_AUXCTL_SHDWSEL_AUXCTL,
+ val | MII_TG3_AUXCTL_ACTL_TX_6DB);
+
+ return err;
+}
static int tg3_bmcr_reset(struct tg3 *tp)
{
@@ -2223,7 +2235,7 @@ static void tg3_phy_apply_otp(struct tg3 *tp)
otp = tp->phy_otp;
- if (TG3_PHY_AUXCTL_SMDSP_ENABLE(tp))
+ if (tg3_phy_toggle_auxctl_smdsp(tp, true))
return;
phy = ((otp & TG3_OTP_AGCTGT_MASK) >> TG3_OTP_AGCTGT_SHIFT);
@@ -2248,7 +2260,7 @@ static void tg3_phy_apply_otp(struct tg3 *tp)
((otp & TG3_OTP_RCOFF_MASK) >> TG3_OTP_RCOFF_SHIFT);
tg3_phydsp_write(tp, MII_TG3_DSP_EXP97, phy);
- TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ tg3_phy_toggle_auxctl_smdsp(tp, false);
}
static void tg3_phy_eee_adjust(struct tg3 *tp, u32 current_link_up)
@@ -2284,9 +2296,9 @@ static void tg3_phy_eee_adjust(struct tg3 *tp, u32 current_link_up)
if (!tp->setlpicnt) {
if (current_link_up == 1 &&
- !TG3_PHY_AUXCTL_SMDSP_ENABLE(tp)) {
+ !tg3_phy_toggle_auxctl_smdsp(tp, true)) {
tg3_phydsp_write(tp, MII_TG3_DSP_TAP26, 0x0000);
- TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ tg3_phy_toggle_auxctl_smdsp(tp, false);
}
val = tr32(TG3_CPMU_EEE_MODE);
@@ -2302,11 +2314,11 @@ static void tg3_phy_eee_enable(struct tg3 *tp)
(GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5717 ||
GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5719 ||
tg3_flag(tp, 57765_CLASS)) &&
- !TG3_PHY_AUXCTL_SMDSP_ENABLE(tp)) {
+ !tg3_phy_toggle_auxctl_smdsp(tp, true)) {
val = MII_TG3_DSP_TAP26_ALNOKO |
MII_TG3_DSP_TAP26_RMRXSTO;
tg3_phydsp_write(tp, MII_TG3_DSP_TAP26, val);
- TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ tg3_phy_toggle_auxctl_smdsp(tp, false);
}
val = tr32(TG3_CPMU_EEE_MODE);
@@ -2450,7 +2462,7 @@ static int tg3_phy_reset_5703_4_5(struct tg3 *tp)
tg3_writephy(tp, MII_CTRL1000,
CTL1000_AS_MASTER | CTL1000_ENABLE_MASTER);
- err = TG3_PHY_AUXCTL_SMDSP_ENABLE(tp);
+ err = tg3_phy_toggle_auxctl_smdsp(tp, true);
if (err)
return err;
@@ -2471,7 +2483,7 @@ static int tg3_phy_reset_5703_4_5(struct tg3 *tp)
tg3_writephy(tp, MII_TG3_DSP_ADDRESS, 0x8200);
tg3_writephy(tp, MII_TG3_DSP_CONTROL, 0x0000);
- TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ tg3_phy_toggle_auxctl_smdsp(tp, false);
tg3_writephy(tp, MII_CTRL1000, phy9_orig);
@@ -2572,10 +2584,10 @@ static int tg3_phy_reset(struct tg3 *tp)
out:
if ((tp->phy_flags & TG3_PHYFLG_ADC_BUG) &&
- !TG3_PHY_AUXCTL_SMDSP_ENABLE(tp)) {
+ !tg3_phy_toggle_auxctl_smdsp(tp, true)) {
tg3_phydsp_write(tp, 0x201f, 0x2aaa);
tg3_phydsp_write(tp, 0x000a, 0x0323);
- TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ tg3_phy_toggle_auxctl_smdsp(tp, false);
}
if (tp->phy_flags & TG3_PHYFLG_5704_A0_BUG) {
@@ -2584,14 +2596,14 @@ out:
}
if (tp->phy_flags & TG3_PHYFLG_BER_BUG) {
- if (!TG3_PHY_AUXCTL_SMDSP_ENABLE(tp)) {
+ if (!tg3_phy_toggle_auxctl_smdsp(tp, true)) {
tg3_phydsp_write(tp, 0x000a, 0x310b);
tg3_phydsp_write(tp, 0x201f, 0x9506);
tg3_phydsp_write(tp, 0x401f, 0x14e2);
- TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ tg3_phy_toggle_auxctl_smdsp(tp, false);
}
} else if (tp->phy_flags & TG3_PHYFLG_JITTER_BUG) {
- if (!TG3_PHY_AUXCTL_SMDSP_ENABLE(tp)) {
+ if (!tg3_phy_toggle_auxctl_smdsp(tp, true)) {
tg3_writephy(tp, MII_TG3_DSP_ADDRESS, 0x000a);
if (tp->phy_flags & TG3_PHYFLG_ADJUST_TRIM) {
tg3_writephy(tp, MII_TG3_DSP_RW_PORT, 0x110b);
@@ -2600,7 +2612,7 @@ out:
} else
tg3_writephy(tp, MII_TG3_DSP_RW_PORT, 0x010b);
- TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ tg3_phy_toggle_auxctl_smdsp(tp, false);
}
}
@@ -4009,7 +4021,7 @@ static int tg3_phy_autoneg_cfg(struct tg3 *tp, u32 advertise, u32 flowctrl)
tw32(TG3_CPMU_EEE_MODE,
tr32(TG3_CPMU_EEE_MODE) & ~TG3_CPMU_EEEMD_LPI_ENABLE);
- err = TG3_PHY_AUXCTL_SMDSP_ENABLE(tp);
+ err = tg3_phy_toggle_auxctl_smdsp(tp, true);
if (!err) {
u32 err2;
@@ -4042,7 +4054,7 @@ static int tg3_phy_autoneg_cfg(struct tg3 *tp, u32 advertise, u32 flowctrl)
MII_TG3_DSP_CH34TP2_HIBW01);
}
- err2 = TG3_PHY_AUXCTL_SMDSP_DISABLE(tp);
+ err2 = tg3_phy_toggle_auxctl_smdsp(tp, false);
if (!err)
err = err2;
}
--
1.7.1
^ permalink raw reply related
* Re: [RFC net-next] ipv6 route: Do not attach neighbour on route.
From: Cong Wang @ 2013-01-15 2:37 UTC (permalink / raw)
To: netdev
In-Reply-To: <50F463E5.7070206@linux-ipv6.org>
On Mon, 14 Jan 2013 at 20:00 GMT, YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org> wrote:
> Not tested, just an RFC.
> Depends on previous new_neigh removal from netevent patch (sorry).
>
Forgot to define __ipv6_neigh_lookup_noref()?
^ permalink raw reply
* Re: [RFC Patch net-next] ipv6: look up neighbours on demand in ip6_finish_output2()
From: Cong Wang @ 2013-01-15 2:28 UTC (permalink / raw)
To: David Miller; +Cc: netdev, roland, Eric Dumazet
In-Reply-To: <20130114.133004.19237632940616685.davem@davemloft.net>
On 01/15/2013 02:30 AM, David Miller wrote:
>
> This is a very incomplete patch.
>
> If this change were so simple, it would have been done by someone
> else a long time ago.
>
> You must, in addition to the incredibly obvious changes in the packet
> output path, completely eliminate the caching of the neighbour entry
> in the ipv6 routes themselves.
>
> This means replacing every rt6->n access or test with something
> equivalent.
>
THanks, David and Eric!
I knew this is probably incomplete, that is why I added "RFC".
Fortunately, YOSHIFUJI just sent a more complete patch:
[RFC net-next] ipv6 route: Do not attach neighbour on route.
So, please ignore mine and use his. :)
^ permalink raw reply
* Re: [PATCH net-next] drivers/net: delete orphaned MCA ibmlana driver content
From: David Miller @ 2013-01-15 2:26 UTC (permalink / raw)
To: paul.gortmaker; +Cc: netdev
In-Reply-To: <1358216245-3956-1-git-send-email-paul.gortmaker@windriver.com>
From: Paul Gortmaker <paul.gortmaker@windriver.com>
Date: Mon, 14 Jan 2013 21:17:25 -0500
> In commit a5e371f61ad33c07b28e7c9b60c78d71fdd34e2a ("drivers/net: delete
> all code/drivers depending on CONFIG_MCA") most of the MCA drivers went,
> including the Kconfig/Makefile hooks for ibmlana, but it seems that I
> missed the "git rm" on these actual driver files, and with the namespace
> overlap with machine check architecture, it got missed by various git
> grep type checking done at that time.
>
> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Applied, thanks Paul.
^ permalink raw reply
* Re: [Patch net-next v3] netpoll: fix a rtnl lock assertion failure
From: Cong Wang @ 2013-01-15 2:23 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, Jiri Pirko, David S. Miller
In-Reply-To: <1358183396.8744.3100.camel@edumazet-glaptop>
On Mon, 2013-01-14 at 09:09 -0800, Eric Dumazet wrote:
> On Mon, 2013-01-14 at 09:07 -0800, Eric Dumazet wrote:
>
> > Strange, why dont we call netdev_master_upper_dev_get() instead ?
> >
> >
>
> What I meant is : the result of this check might be of little use, if we
> don't hold any lock preventing another thread to change things behind
> us.
>
>
Even if so, enslaving this device can still happen after we release the
rtnl lock, unless we take this lock for the whole netpoll_setup().
Am I missing anything?
^ permalink raw reply
* [PATCH net-next] drivers/net: delete orphaned MCA ibmlana driver content
From: Paul Gortmaker @ 2013-01-15 2:17 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Paul Gortmaker
In commit a5e371f61ad33c07b28e7c9b60c78d71fdd34e2a ("drivers/net: delete
all code/drivers depending on CONFIG_MCA") most of the MCA drivers went,
including the Kconfig/Makefile hooks for ibmlana, but it seems that I
missed the "git rm" on these actual driver files, and with the namespace
overlap with machine check architecture, it got missed by various git
grep type checking done at that time.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
---
drivers/net/ethernet/natsemi/ibmlana.c | 1075 --------------------------------
drivers/net/ethernet/natsemi/ibmlana.h | 278 ---------
2 files changed, 1353 deletions(-)
delete mode 100644 drivers/net/ethernet/natsemi/ibmlana.c
delete mode 100644 drivers/net/ethernet/natsemi/ibmlana.h
diff --git a/drivers/net/ethernet/natsemi/ibmlana.c b/drivers/net/ethernet/natsemi/ibmlana.c
deleted file mode 100644
index 923e640..0000000
--- a/drivers/net/ethernet/natsemi/ibmlana.c
+++ /dev/null
@@ -1,1075 +0,0 @@
-/*
-net-3-driver for the IBM LAN Adapter/A
-
-This is an extension to the Linux operating system, and is covered by the
-same GNU General Public License that covers that work.
-
-Copyright 1999 by Alfred Arnold (alfred@ccac.rwth-aachen.de,
- alfred.arnold@lancom.de)
-
-This driver is based both on the SK_MCA driver, which is itself based on the
-SK_G16 and 3C523 driver.
-
-paper sources:
- 'PC Hardware: Aufbau, Funktionsweise, Programmierung' by
- Hans-Peter Messmer for the basic Microchannel stuff
-
- 'Linux Geraetetreiber' by Allesandro Rubini, Kalle Dalheimer
- for help on Ethernet driver programming
-
- 'DP83934CVUL-20/25 MHz SONIC-T Ethernet Controller Datasheet' by National
- Semiconductor for info on the MAC chip
-
- 'LAN Technical Reference Ethernet Adapter Interface Version 1 Release 1.0
- Document Number SC30-3661-00' by IBM for info on the adapter itself
-
- Also see http://www.national.com/analog
-
-special acknowledgements to:
- - Bob Eager for helping me out with documentation from IBM
- - Jim Shorney for his endless patience with me while I was using
- him as a beta tester to trace down the address filter bug ;-)
-
- Missing things:
-
- -> set debug level via ioctl instead of compile-time switches
- -> I didn't follow the development of the 2.1.x kernels, so my
- assumptions about which things changed with which kernel version
- are probably nonsense
-
-History:
- Nov 6th, 1999
- startup from SK_MCA driver
- Dec 6th, 1999
- finally got docs about the card. A big thank you to Bob Eager!
- Dec 12th, 1999
- first packet received
- Dec 13th, 1999
- recv queue done, tcpdump works
- Dec 15th, 1999
- transmission part works
- Dec 28th, 1999
- added usage of the isa_functions for Linux 2.3 . Things should
- still work with 2.0.x....
- Jan 28th, 2000
- in Linux 2.2.13, the version.h file mysteriously didn't get
- included. Added a workaround for this. Furthermore, it now
- not only compiles as a modules ;-)
- Jan 30th, 2000
- newer kernels automatically probe more than one board, so the
- 'startslot' as a variable is also needed here
- Apr 12th, 2000
- the interrupt mask register is not set 'hard' instead of individually
- setting registers, since this seems to set bits that shouldn't be
- set
- May 21st, 2000
- reset interrupt status immediately after CAM load
- add a recovery delay after releasing the chip's reset line
- May 24th, 2000
- finally found the bug in the address filter setup - damned signed
- chars!
- June 1st, 2000
- corrected version codes, added support for the latest 2.3 changes
- Oct 28th, 2002
- cleaned up for the 2.5 tree <alan@lxorguk.ukuu.org.uk>
-
- *************************************************************************/
-
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/ioport.h>
-#include <linux/interrupt.h>
-#include <linux/delay.h>
-#include <linux/time.h>
-#include <linux/mca.h>
-#include <linux/module.h>
-#include <linux/netdevice.h>
-#include <linux/etherdevice.h>
-#include <linux/if_ether.h>
-#include <linux/skbuff.h>
-#include <linux/bitops.h>
-
-#include <asm/processor.h>
-#include <asm/io.h>
-
-#define _IBM_LANA_DRIVER_
-#include "ibmlana.h"
-
-#undef DEBUG
-
-#define DRV_NAME "ibmlana"
-
-/* ------------------------------------------------------------------------
- * global static data - not more since we can handle multiple boards and
- * have to pack all state info into the device struct!
- * ------------------------------------------------------------------------ */
-
-static char *MediaNames[Media_Count] = {
- "10BaseT", "10Base5", "Unknown", "10Base2"
-};
-
-/* ------------------------------------------------------------------------
- * private subfunctions
- * ------------------------------------------------------------------------ */
-
-#ifdef DEBUG
- /* dump all registers */
-
-static void dumpregs(struct net_device *dev)
-{
- int z;
-
- for (z = 0; z < 160; z += 2) {
- if (!(z & 15))
- printk("REGS: %04x:", z);
- printk(" %04x", inw(dev->base_addr + z));
- if ((z & 15) == 14)
- printk("\n");
- }
-}
-
-/* dump parts of shared memory - only needed during debugging */
-
-static void dumpmem(struct net_device *dev, u32 start, u32 len)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- int z;
-
- printk("Address %04x:\n", start);
- for (z = 0; z < len; z++) {
- if ((z & 15) == 0)
- printk("%04x:", z);
- printk(" %02x", readb(priv->base + start + z));
- if ((z & 15) == 15)
- printk("\n");
- }
- if ((z & 15) != 0)
- printk("\n");
-}
-
-/* print exact time - ditto */
-
-static void PrTime(void)
-{
- struct timeval tv;
-
- do_gettimeofday(&tv);
- printk("%9d:%06d: ", (int) tv.tv_sec, (int) tv.tv_usec);
-}
-#endif /* DEBUG */
-
-/* deduce resources out of POS registers */
-
-static void getaddrs(struct mca_device *mdev, int *base, int *memlen,
- int *iobase, int *irq, ibmlana_medium *medium)
-{
- u_char pos0, pos1;
-
- pos0 = mca_device_read_stored_pos(mdev, 2);
- pos1 = mca_device_read_stored_pos(mdev, 3);
-
- *base = 0xc0000 + ((pos1 & 0xf0) << 9);
- *memlen = (pos1 & 0x01) ? 0x8000 : 0x4000;
- *iobase = (pos0 & 0xe0) << 7;
- switch (pos0 & 0x06) {
- case 0:
- *irq = 5;
- break;
- case 2:
- *irq = 15;
- break;
- case 4:
- *irq = 10;
- break;
- case 6:
- *irq = 11;
- break;
- }
- *medium = (pos0 & 0x18) >> 3;
-}
-
-/* wait on register value with mask and timeout */
-
-static int wait_timeout(struct net_device *dev, int regoffs, u16 mask,
- u16 value, int timeout)
-{
- unsigned long fin = jiffies + timeout;
-
- while (time_before(jiffies,fin))
- if ((inw(dev->base_addr + regoffs) & mask) == value)
- return 1;
-
- return 0;
-}
-
-
-/* reset the whole board */
-
-static void ResetBoard(struct net_device *dev)
-{
- unsigned char bcmval;
-
- /* read original board control value */
-
- bcmval = inb(dev->base_addr + BCMREG);
-
- /* set reset bit for a while */
-
- bcmval |= BCMREG_RESET;
- outb(bcmval, dev->base_addr + BCMREG);
- udelay(10);
- bcmval &= ~BCMREG_RESET;
- outb(bcmval, dev->base_addr + BCMREG);
-
- /* switch over to RAM again */
-
- bcmval |= BCMREG_RAMEN | BCMREG_RAMWIN;
- outb(bcmval, dev->base_addr + BCMREG);
-}
-
-/* calculate RAM layout & set up descriptors in RAM */
-
-static void InitDscrs(struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- u32 addr, baddr, raddr;
- int z;
- tda_t tda;
- rda_t rda;
- rra_t rra;
-
- /* initialize RAM */
-
- memset_io(priv->base, 0xaa,
- dev->mem_start - dev->mem_start); /* XXX: typo? */
-
- /* setup n TX descriptors - independent of RAM size */
-
- priv->tdastart = addr = 0;
- priv->txbufstart = baddr = sizeof(tda_t) * TXBUFCNT;
- for (z = 0; z < TXBUFCNT; z++) {
- tda.status = 0;
- tda.config = 0;
- tda.length = 0;
- tda.fragcount = 1;
- tda.startlo = baddr;
- tda.starthi = 0;
- tda.fraglength = 0;
- if (z == TXBUFCNT - 1)
- tda.link = priv->tdastart;
- else
- tda.link = addr + sizeof(tda_t);
- tda.link |= 1;
- memcpy_toio(priv->base + addr, &tda, sizeof(tda_t));
- addr += sizeof(tda_t);
- baddr += PKTSIZE;
- }
-
- /* calculate how many receive buffers fit into remaining memory */
-
- priv->rxbufcnt = (dev->mem_end - dev->mem_start - baddr) / (sizeof(rra_t) + sizeof(rda_t) + PKTSIZE);
-
- /* calculate receive addresses */
-
- priv->rrastart = raddr = priv->txbufstart + (TXBUFCNT * PKTSIZE);
- priv->rdastart = addr = priv->rrastart + (priv->rxbufcnt * sizeof(rra_t));
- priv->rxbufstart = baddr = priv->rdastart + (priv->rxbufcnt * sizeof(rda_t));
-
- for (z = 0; z < priv->rxbufcnt; z++) {
- rra.startlo = baddr;
- rra.starthi = 0;
- rra.cntlo = PKTSIZE >> 1;
- rra.cnthi = 0;
- memcpy_toio(priv->base + raddr, &rra, sizeof(rra_t));
-
- rda.status = 0;
- rda.length = 0;
- rda.startlo = 0;
- rda.starthi = 0;
- rda.seqno = 0;
- if (z < priv->rxbufcnt - 1)
- rda.link = addr + sizeof(rda_t);
- else
- rda.link = 1;
- rda.inuse = 1;
- memcpy_toio(priv->base + addr, &rda, sizeof(rda_t));
-
- baddr += PKTSIZE;
- raddr += sizeof(rra_t);
- addr += sizeof(rda_t);
- }
-
- /* initialize current pointers */
-
- priv->nextrxdescr = 0;
- priv->lastrxdescr = priv->rxbufcnt - 1;
- priv->nexttxdescr = 0;
- priv->currtxdescr = 0;
- priv->txusedcnt = 0;
- memset(priv->txused, 0, sizeof(priv->txused));
-}
-
-/* set up Rx + Tx descriptors in SONIC */
-
-static int InitSONIC(struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
-
- /* set up start & end of resource area */
-
- outw(0, SONIC_URRA);
- outw(priv->rrastart, dev->base_addr + SONIC_RSA);
- outw(priv->rrastart + (priv->rxbufcnt * sizeof(rra_t)), dev->base_addr + SONIC_REA);
- outw(priv->rrastart, dev->base_addr + SONIC_RRP);
- outw(priv->rrastart, dev->base_addr + SONIC_RWP);
-
- /* set EOBC so that only one packet goes into one buffer */
-
- outw((PKTSIZE - 4) >> 1, dev->base_addr + SONIC_EOBC);
-
- /* let SONIC read the first RRA descriptor */
-
- outw(CMDREG_RRRA, dev->base_addr + SONIC_CMDREG);
- if (!wait_timeout(dev, SONIC_CMDREG, CMDREG_RRRA, 0, 2)) {
- printk(KERN_ERR "%s: SONIC did not respond on RRRA command - giving up.", dev->name);
- return 0;
- }
-
- /* point SONIC to the first RDA */
-
- outw(0, dev->base_addr + SONIC_URDA);
- outw(priv->rdastart, dev->base_addr + SONIC_CRDA);
-
- /* set upper half of TDA address */
-
- outw(0, dev->base_addr + SONIC_UTDA);
-
- return 1;
-}
-
-/* stop SONIC so we can reinitialize it */
-
-static void StopSONIC(struct net_device *dev)
-{
- /* disable interrupts */
-
- outb(inb(dev->base_addr + BCMREG) & (~BCMREG_IEN), dev->base_addr + BCMREG);
- outb(0, dev->base_addr + SONIC_IMREG);
-
- /* reset the SONIC */
-
- outw(CMDREG_RST, dev->base_addr + SONIC_CMDREG);
- udelay(10);
- outw(CMDREG_RST, dev->base_addr + SONIC_CMDREG);
-}
-
-/* initialize card and SONIC for proper operation */
-
-static void putcam(camentry_t * cams, int *camcnt, char *addr)
-{
- camentry_t *pcam = cams + (*camcnt);
- u8 *uaddr = (u8 *) addr;
-
- pcam->index = *camcnt;
- pcam->addr0 = (((u16) uaddr[1]) << 8) | uaddr[0];
- pcam->addr1 = (((u16) uaddr[3]) << 8) | uaddr[2];
- pcam->addr2 = (((u16) uaddr[5]) << 8) | uaddr[4];
- (*camcnt)++;
-}
-
-static void InitBoard(struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- int camcnt;
- camentry_t cams[16];
- u32 cammask;
- struct netdev_hw_addr *ha;
- u16 rcrval;
-
- /* reset the SONIC */
-
- outw(CMDREG_RST, dev->base_addr + SONIC_CMDREG);
- udelay(10);
-
- /* clear all spurious interrupts */
-
- outw(inw(dev->base_addr + SONIC_ISREG), dev->base_addr + SONIC_ISREG);
-
- /* set up the SONIC's bus interface - constant for this adapter -
- must be done while the SONIC is in reset */
-
- outw(DCREG_USR1 | DCREG_USR0 | DCREG_WC1 | DCREG_DW32, dev->base_addr + SONIC_DCREG);
- outw(0, dev->base_addr + SONIC_DCREG2);
-
- /* remove reset form the SONIC */
-
- outw(0, dev->base_addr + SONIC_CMDREG);
- udelay(10);
-
- /* data sheet requires URRA to be programmed before setting up the CAM contents */
-
- outw(0, dev->base_addr + SONIC_URRA);
-
- /* program the CAM entry 0 to the device address */
-
- camcnt = 0;
- putcam(cams, &camcnt, dev->dev_addr);
-
- /* start putting the multicast addresses into the CAM list. Stop if
- it is full. */
-
- netdev_for_each_mc_addr(ha, dev) {
- putcam(cams, &camcnt, ha->addr);
- if (camcnt == 16)
- break;
- }
-
- /* calculate CAM mask */
-
- cammask = (1 << camcnt) - 1;
-
- /* feed CDA into SONIC, initialize RCR value (always get broadcasts) */
-
- memcpy_toio(priv->base, cams, sizeof(camentry_t) * camcnt);
- memcpy_toio(priv->base + (sizeof(camentry_t) * camcnt), &cammask, sizeof(cammask));
-
-#ifdef DEBUG
- printk("CAM setup:\n");
- dumpmem(dev, 0, sizeof(camentry_t) * camcnt + sizeof(cammask));
-#endif
-
- outw(0, dev->base_addr + SONIC_CAMPTR);
- outw(camcnt, dev->base_addr + SONIC_CAMCNT);
- outw(CMDREG_LCAM, dev->base_addr + SONIC_CMDREG);
- if (!wait_timeout(dev, SONIC_CMDREG, CMDREG_LCAM, 0, 2)) {
- printk(KERN_ERR "%s:SONIC did not respond on LCAM command - giving up.", dev->name);
- return;
- } else {
- /* clear interrupt condition */
-
- outw(ISREG_LCD, dev->base_addr + SONIC_ISREG);
-
-#ifdef DEBUG
- printk("Loading CAM done, address pointers %04x:%04x\n",
- inw(dev->base_addr + SONIC_URRA),
- inw(dev->base_addr + SONIC_CAMPTR));
- {
- int z;
-
- printk("\n-->CAM: PTR %04x CNT %04x\n",
- inw(dev->base_addr + SONIC_CAMPTR),
- inw(dev->base_addr + SONIC_CAMCNT));
- outw(CMDREG_RST, dev->base_addr + SONIC_CMDREG);
- for (z = 0; z < camcnt; z++) {
- outw(z, dev->base_addr + SONIC_CAMEPTR);
- printk("Entry %d: %04x %04x %04x\n", z,
- inw(dev->base_addr + SONIC_CAMADDR0),
- inw(dev->base_addr + SONIC_CAMADDR1),
- inw(dev->base_addr + SONIC_CAMADDR2));
- }
- outw(0, dev->base_addr + SONIC_CMDREG);
- }
-#endif
- }
-
- rcrval = RCREG_BRD | RCREG_LB_NONE;
-
- /* if still multicast addresses left or ALLMULTI is set, set the multicast
- enable bit */
-
- if ((dev->flags & IFF_ALLMULTI) || netdev_mc_count(dev) > camcnt)
- rcrval |= RCREG_AMC;
-
- /* promiscuous mode ? */
-
- if (dev->flags & IFF_PROMISC)
- rcrval |= RCREG_PRO;
-
- /* program receive mode */
-
- outw(rcrval, dev->base_addr + SONIC_RCREG);
-#ifdef DEBUG
- printk("\nRCRVAL: %04x\n", rcrval);
-#endif
-
- /* set up descriptors in shared memory + feed them into SONIC registers */
-
- InitDscrs(dev);
- if (!InitSONIC(dev))
- return;
-
- /* reset all pending interrupts */
-
- outw(0xffff, dev->base_addr + SONIC_ISREG);
-
- /* enable transmitter + receiver interrupts */
-
- outw(CMDREG_RXEN, dev->base_addr + SONIC_CMDREG);
- outw(IMREG_PRXEN | IMREG_RBEEN | IMREG_PTXEN | IMREG_TXEREN, dev->base_addr + SONIC_IMREG);
-
- /* turn on card interrupts */
-
- outb(inb(dev->base_addr + BCMREG) | BCMREG_IEN, dev->base_addr + BCMREG);
-
-#ifdef DEBUG
- printk("Register dump after initialization:\n");
- dumpregs(dev);
-#endif
-}
-
-/* start transmission of a descriptor */
-
-static void StartTx(struct net_device *dev, int descr)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- int addr;
-
- addr = priv->tdastart + (descr * sizeof(tda_t));
-
- /* put descriptor address into SONIC */
-
- outw(addr, dev->base_addr + SONIC_CTDA);
-
- /* trigger transmitter */
-
- priv->currtxdescr = descr;
- outw(CMDREG_TXP, dev->base_addr + SONIC_CMDREG);
-}
-
-/* ------------------------------------------------------------------------
- * interrupt handler(s)
- * ------------------------------------------------------------------------ */
-
-/* receive buffer area exhausted */
-
-static void irqrbe_handler(struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
-
- /* point the SONIC back to the RRA start */
-
- outw(priv->rrastart, dev->base_addr + SONIC_RRP);
- outw(priv->rrastart, dev->base_addr + SONIC_RWP);
-}
-
-/* receive interrupt */
-
-static void irqrx_handler(struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- rda_t rda;
- u32 rdaaddr, lrdaaddr;
-
- /* loop until ... */
-
- while (1) {
- /* read descriptor that was next to be filled by SONIC */
-
- rdaaddr = priv->rdastart + (priv->nextrxdescr * sizeof(rda_t));
- lrdaaddr = priv->rdastart + (priv->lastrxdescr * sizeof(rda_t));
- memcpy_fromio(&rda, priv->base + rdaaddr, sizeof(rda_t));
-
- /* iron out upper word halves of fields we use - SONIC will duplicate
- bits 0..15 to 16..31 */
-
- rda.status &= 0xffff;
- rda.length &= 0xffff;
- rda.startlo &= 0xffff;
-
- /* stop if the SONIC still owns it, i.e. there is no data for us */
-
- if (rda.inuse)
- break;
-
- /* good packet? */
-
- else if (rda.status & RCREG_PRX) {
- struct sk_buff *skb;
-
- /* fetch buffer */
-
- skb = netdev_alloc_skb(dev, rda.length + 2);
- if (skb == NULL)
- dev->stats.rx_dropped++;
- else {
- /* copy out data */
-
- memcpy_fromio(skb_put(skb, rda.length),
- priv->base +
- rda.startlo, rda.length);
-
- /* set up skb fields */
-
- skb->protocol = eth_type_trans(skb, dev);
- skb_checksum_none_assert(skb);
-
- /* bookkeeping */
- dev->stats.rx_packets++;
- dev->stats.rx_bytes += rda.length;
-
- /* pass to the upper layers */
- netif_rx(skb);
- }
- }
-
- /* otherwise check error status bits and increase statistics */
-
- else {
- dev->stats.rx_errors++;
- if (rda.status & RCREG_FAER)
- dev->stats.rx_frame_errors++;
- if (rda.status & RCREG_CRCR)
- dev->stats.rx_crc_errors++;
- }
-
- /* descriptor processed, will become new last descriptor in queue */
-
- rda.link = 1;
- rda.inuse = 1;
- memcpy_toio(priv->base + rdaaddr, &rda,
- sizeof(rda_t));
-
- /* set up link and EOL = 0 in currently last descriptor. Only write
- the link field since the SONIC may currently already access the
- other fields. */
-
- memcpy_toio(priv->base + lrdaaddr + 20, &rdaaddr, 4);
-
- /* advance indices */
-
- priv->lastrxdescr = priv->nextrxdescr;
- if ((++priv->nextrxdescr) >= priv->rxbufcnt)
- priv->nextrxdescr = 0;
- }
-}
-
-/* transmit interrupt */
-
-static void irqtx_handler(struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- tda_t tda;
-
- /* fetch descriptor (we forgot the size ;-) */
- memcpy_fromio(&tda, priv->base + priv->tdastart + (priv->currtxdescr * sizeof(tda_t)), sizeof(tda_t));
-
- /* update statistics */
- dev->stats.tx_packets++;
- dev->stats.tx_bytes += tda.length;
-
- /* update our pointers */
- priv->txused[priv->currtxdescr] = 0;
- priv->txusedcnt--;
-
- /* if there are more descriptors present in RAM, start them */
- if (priv->txusedcnt > 0)
- StartTx(dev, (priv->currtxdescr + 1) % TXBUFCNT);
-
- /* tell the upper layer we can go on transmitting */
- netif_wake_queue(dev);
-}
-
-static void irqtxerr_handler(struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- tda_t tda;
-
- /* fetch descriptor to check status */
- memcpy_fromio(&tda, priv->base + priv->tdastart + (priv->currtxdescr * sizeof(tda_t)), sizeof(tda_t));
-
- /* update statistics */
- dev->stats.tx_errors++;
- if (tda.status & (TCREG_NCRS | TCREG_CRSL))
- dev->stats.tx_carrier_errors++;
- if (tda.status & TCREG_EXC)
- dev->stats.tx_aborted_errors++;
- if (tda.status & TCREG_OWC)
- dev->stats.tx_window_errors++;
- if (tda.status & TCREG_FU)
- dev->stats.tx_fifo_errors++;
-
- /* update our pointers */
- priv->txused[priv->currtxdescr] = 0;
- priv->txusedcnt--;
-
- /* if there are more descriptors present in RAM, start them */
- if (priv->txusedcnt > 0)
- StartTx(dev, (priv->currtxdescr + 1) % TXBUFCNT);
-
- /* tell the upper layer we can go on transmitting */
- netif_wake_queue(dev);
-}
-
-/* general interrupt entry */
-
-static irqreturn_t irq_handler(int dummy, void *device)
-{
- struct net_device *dev = device;
- u16 ival;
-
- /* in case we're not meant... */
- if (!(inb(dev->base_addr + BCMREG) & BCMREG_IPEND))
- return IRQ_NONE;
-
- /* loop through the interrupt bits until everything is clear */
- while (1) {
- ival = inw(dev->base_addr + SONIC_ISREG);
-
- if (ival & ISREG_RBE) {
- irqrbe_handler(dev);
- outw(ISREG_RBE, dev->base_addr + SONIC_ISREG);
- }
- if (ival & ISREG_PKTRX) {
- irqrx_handler(dev);
- outw(ISREG_PKTRX, dev->base_addr + SONIC_ISREG);
- }
- if (ival & ISREG_TXDN) {
- irqtx_handler(dev);
- outw(ISREG_TXDN, dev->base_addr + SONIC_ISREG);
- }
- if (ival & ISREG_TXER) {
- irqtxerr_handler(dev);
- outw(ISREG_TXER, dev->base_addr + SONIC_ISREG);
- }
- break;
- }
- return IRQ_HANDLED;
-}
-
-/* ------------------------------------------------------------------------
- * driver methods
- * ------------------------------------------------------------------------ */
-
-/* MCA info */
-
-#if 0 /* info available elsewhere, but this is kept for reference */
-static int ibmlana_getinfo(char *buf, int slot, void *d)
-{
- int len = 0, i;
- struct net_device *dev = (struct net_device *) d;
- ibmlana_priv *priv;
-
- /* can't say anything about an uninitialized device... */
-
- if (dev == NULL)
- return len;
- priv = netdev_priv(dev);
-
- /* print info */
-
- len += sprintf(buf + len, "IRQ: %d\n", priv->realirq);
- len += sprintf(buf + len, "I/O: %#lx\n", dev->base_addr);
- len += sprintf(buf + len, "Memory: %#lx-%#lx\n", dev->mem_start, dev->mem_end - 1);
- len += sprintf(buf + len, "Transceiver: %s\n", MediaNames[priv->medium]);
- len += sprintf(buf + len, "Device: %s\n", dev->name);
- len += sprintf(buf + len, "MAC address:");
- for (i = 0; i < 6; i++)
- len += sprintf(buf + len, " %02x", dev->dev_addr[i]);
- buf[len++] = '\n';
- buf[len] = 0;
-
- return len;
-}
-#endif
-
-/* open driver. Means also initialization and start of LANCE */
-
-static int ibmlana_open(struct net_device *dev)
-{
- int result;
- ibmlana_priv *priv = netdev_priv(dev);
-
- /* register resources - only necessary for IRQ */
-
- result = request_irq(priv->realirq, irq_handler, IRQF_SHARED,
- dev->name, dev);
- if (result != 0) {
- printk(KERN_ERR "%s: failed to register irq %d\n", dev->name, dev->irq);
- return result;
- }
- dev->irq = priv->realirq;
-
- /* set up the card and SONIC */
- InitBoard(dev);
-
- /* initialize operational flags */
- netif_start_queue(dev);
- return 0;
-}
-
-/* close driver. Shut down board and free allocated resources */
-
-static int ibmlana_close(struct net_device *dev)
-{
- /* turn off board */
-
- /* release resources */
- if (dev->irq != 0)
- free_irq(dev->irq, dev);
- dev->irq = 0;
- return 0;
-}
-
-/* transmit a block. */
-
-static netdev_tx_t ibmlana_tx(struct sk_buff *skb, struct net_device *dev)
-{
- ibmlana_priv *priv = netdev_priv(dev);
- int tmplen, addr;
- unsigned long flags;
- tda_t tda;
- int baddr;
-
- /* find out if there are free slots for a frame to transmit. If not,
- the upper layer is in deep desperation and we simply ignore the frame. */
-
- if (priv->txusedcnt >= TXBUFCNT) {
- dev->stats.tx_dropped++;
- goto tx_done;
- }
-
- /* copy the frame data into the next free transmit buffer - fillup missing */
- tmplen = skb->len;
- if (tmplen < 60)
- tmplen = 60;
- baddr = priv->txbufstart + (priv->nexttxdescr * PKTSIZE);
- memcpy_toio(priv->base + baddr, skb->data, skb->len);
-
- /* copy filler into RAM - in case we're filling up...
- we're filling a bit more than necessary, but that doesn't harm
- since the buffer is far larger...
- Sorry Linus for the filler string but I couldn't resist ;-) */
-
- if (tmplen > skb->len) {
- char *fill = "NetBSD is a nice OS too! ";
- unsigned int destoffs = skb->len, l = strlen(fill);
-
- while (destoffs < tmplen) {
- memcpy_toio(priv->base + baddr + destoffs, fill, l);
- destoffs += l;
- }
- }
-
- /* set up the new frame descriptor */
- addr = priv->tdastart + (priv->nexttxdescr * sizeof(tda_t));
- memcpy_fromio(&tda, priv->base + addr, sizeof(tda_t));
- tda.length = tda.fraglength = tmplen;
- memcpy_toio(priv->base + addr, &tda, sizeof(tda_t));
-
- /* if there were no active descriptors, trigger the SONIC */
- spin_lock_irqsave(&priv->lock, flags);
-
- priv->txusedcnt++;
- priv->txused[priv->nexttxdescr] = 1;
-
- /* are all transmission slots used up ? */
- if (priv->txusedcnt >= TXBUFCNT)
- netif_stop_queue(dev);
-
- if (priv->txusedcnt == 1)
- StartTx(dev, priv->nexttxdescr);
- priv->nexttxdescr = (priv->nexttxdescr + 1) % TXBUFCNT;
-
- spin_unlock_irqrestore(&priv->lock, flags);
-tx_done:
- dev_kfree_skb(skb);
- return NETDEV_TX_OK;
-}
-
-/* switch receiver mode. */
-
-static void ibmlana_set_multicast_list(struct net_device *dev)
-{
- /* first stop the SONIC... */
- StopSONIC(dev);
- /* ...then reinit it with the new flags */
- InitBoard(dev);
-}
-
-/* ------------------------------------------------------------------------
- * hardware check
- * ------------------------------------------------------------------------ */
-
-static int ibmlana_irq;
-static int ibmlana_io;
-static int startslot; /* counts through slots when probing multiple devices */
-
-static short ibmlana_adapter_ids[] __initdata = {
- IBM_LANA_ID,
- 0x0000
-};
-
-static char *ibmlana_adapter_names[] = {
- "IBM LAN Adapter/A",
- NULL
-};
-
-
-static const struct net_device_ops ibmlana_netdev_ops = {
- .ndo_open = ibmlana_open,
- .ndo_stop = ibmlana_close,
- .ndo_start_xmit = ibmlana_tx,
- .ndo_set_rx_mode = ibmlana_set_multicast_list,
- .ndo_change_mtu = eth_change_mtu,
- .ndo_set_mac_address = eth_mac_addr,
- .ndo_validate_addr = eth_validate_addr,
-};
-
-static int ibmlana_init_one(struct device *kdev)
-{
- struct mca_device *mdev = to_mca_device(kdev);
- struct net_device *dev;
- int slot = mdev->slot, z, rc;
- int base = 0, irq = 0, iobase = 0, memlen = 0;
- ibmlana_priv *priv;
- ibmlana_medium medium;
-
- dev = alloc_etherdev(sizeof(ibmlana_priv));
- if (!dev)
- return -ENOMEM;
-
- dev->irq = ibmlana_irq;
- dev->base_addr = ibmlana_io;
-
- base = dev->mem_start;
- irq = dev->irq;
-
- /* deduce card addresses */
- getaddrs(mdev, &base, &memlen, &iobase, &irq, &medium);
-
- /* were we looking for something different ? */
- if (dev->irq && dev->irq != irq) {
- rc = -ENODEV;
- goto err_out;
- }
- if (dev->mem_start && dev->mem_start != base) {
- rc = -ENODEV;
- goto err_out;
- }
-
- /* announce success */
- printk(KERN_INFO "%s: IBM LAN Adapter/A found in slot %d\n", dev->name, slot + 1);
-
- /* try to obtain I/O range */
- if (!request_region(iobase, IBM_LANA_IORANGE, DRV_NAME)) {
- printk(KERN_ERR "%s: cannot allocate I/O range at %#x!\n", DRV_NAME, iobase);
- startslot = slot + 1;
- rc = -EBUSY;
- goto err_out;
- }
-
- priv = netdev_priv(dev);
- priv->slot = slot;
- priv->realirq = mca_device_transform_irq(mdev, irq);
- priv->medium = medium;
- spin_lock_init(&priv->lock);
-
- /* set base + irq for this device (irq not allocated so far) */
-
- dev->irq = 0;
- dev->mem_start = base;
- dev->mem_end = base + memlen;
- dev->base_addr = iobase;
-
- priv->base = ioremap(base, memlen);
- if (!priv->base) {
- printk(KERN_ERR "%s: cannot remap memory!\n", DRV_NAME);
- startslot = slot + 1;
- rc = -EBUSY;
- goto err_out_reg;
- }
-
- mca_device_set_name(mdev, ibmlana_adapter_names[mdev->index]);
- mca_device_set_claim(mdev, 1);
-
- /* set methods */
- dev->netdev_ops = &ibmlana_netdev_ops;
- dev->flags |= IFF_MULTICAST;
-
- /* copy out MAC address */
-
- for (z = 0; z < ETH_ALEN; z++)
- dev->dev_addr[z] = inb(dev->base_addr + MACADDRPROM + z);
-
- /* print config */
-
- printk(KERN_INFO "%s: IRQ %d, I/O %#lx, memory %#lx-%#lx, "
- "MAC address %pM.\n",
- dev->name, priv->realirq, dev->base_addr,
- dev->mem_start, dev->mem_end - 1,
- dev->dev_addr);
- printk(KERN_INFO "%s: %s medium\n", dev->name, MediaNames[priv->medium]);
-
- /* reset board */
-
- ResetBoard(dev);
-
- /* next probe will start at next slot */
-
- startslot = slot + 1;
-
- rc = register_netdev(dev);
- if (rc)
- goto err_out_claimed;
-
- dev_set_drvdata(kdev, dev);
- return 0;
-
-err_out_claimed:
- mca_device_set_claim(mdev, 0);
- iounmap(priv->base);
-err_out_reg:
- release_region(iobase, IBM_LANA_IORANGE);
-err_out:
- free_netdev(dev);
- return rc;
-}
-
-static int ibmlana_remove_one(struct device *kdev)
-{
- struct mca_device *mdev = to_mca_device(kdev);
- struct net_device *dev = dev_get_drvdata(kdev);
- ibmlana_priv *priv = netdev_priv(dev);
-
- unregister_netdev(dev);
- /*DeinitBoard(dev); */
- release_region(dev->base_addr, IBM_LANA_IORANGE);
- mca_device_set_claim(mdev, 0);
- iounmap(priv->base);
- free_netdev(dev);
- return 0;
-}
-
-/* ------------------------------------------------------------------------
- * modularization support
- * ------------------------------------------------------------------------ */
-
-module_param_named(irq, ibmlana_irq, int, 0);
-module_param_named(io, ibmlana_io, int, 0);
-MODULE_PARM_DESC(irq, "IBM LAN/A IRQ number");
-MODULE_PARM_DESC(io, "IBM LAN/A I/O base address");
-MODULE_LICENSE("GPL");
-
-static struct mca_driver ibmlana_driver = {
- .id_table = ibmlana_adapter_ids,
- .driver = {
- .name = "ibmlana",
- .bus = &mca_bus_type,
- .probe = ibmlana_init_one,
- .remove = ibmlana_remove_one,
- },
-};
-
-static int __init ibmlana_init_module(void)
-{
- return mca_register_driver(&ibmlana_driver);
-}
-
-static void __exit ibmlana_cleanup_module(void)
-{
- mca_unregister_driver(&ibmlana_driver);
-}
-
-module_init(ibmlana_init_module);
-module_exit(ibmlana_cleanup_module);
diff --git a/drivers/net/ethernet/natsemi/ibmlana.h b/drivers/net/ethernet/natsemi/ibmlana.h
deleted file mode 100644
index accd5ef..0000000
--- a/drivers/net/ethernet/natsemi/ibmlana.h
+++ /dev/null
@@ -1,278 +0,0 @@
-#ifndef _IBM_LANA_INCLUDE_
-#define _IBM_LANA_INCLUDE_
-
-#ifdef _IBM_LANA_DRIVER_
-
-/* maximum packet size */
-
-#define PKTSIZE 1524
-
-/* number of transmit buffers */
-
-#define TXBUFCNT 4
-
-/* Adapter ID's */
-#define IBM_LANA_ID 0xffe0
-
-/* media enumeration - defined in a way that it fits onto the LAN/A's
- POS registers... */
-
-typedef enum {
- Media_10BaseT, Media_10Base5,
- Media_Unknown, Media_10Base2, Media_Count
-} ibmlana_medium;
-
-/* private structure */
-
-typedef struct {
- unsigned int slot; /* MCA-Slot-# */
- int realirq; /* memorizes actual IRQ, even when
- currently not allocated */
- ibmlana_medium medium; /* physical cannector */
- u32 tdastart, txbufstart, /* addresses */
- rrastart, rxbufstart, rdastart, rxbufcnt, txusedcnt;
- int nextrxdescr, /* next rx descriptor to be used */
- lastrxdescr, /* last free rx descriptor */
- nexttxdescr, /* last tx descriptor to be used */
- currtxdescr, /* tx descriptor currently tx'ed */
- txused[TXBUFCNT]; /* busy flags */
- void __iomem *base;
- spinlock_t lock;
-} ibmlana_priv;
-
-/* this card uses quite a lot of I/O ports...luckily the MCA bus decodes
- a full 64K I/O range... */
-
-#define IBM_LANA_IORANGE 0xa0
-
-/* Command Register: */
-
-#define SONIC_CMDREG 0x00
-#define CMDREG_HTX 0x0001 /* halt transmission */
-#define CMDREG_TXP 0x0002 /* start transmission */
-#define CMDREG_RXDIS 0x0004 /* disable receiver */
-#define CMDREG_RXEN 0x0008 /* enable receiver */
-#define CMDREG_STP 0x0010 /* stop timer */
-#define CMDREG_ST 0x0020 /* start timer */
-#define CMDREG_RST 0x0080 /* software reset */
-#define CMDREG_RRRA 0x0100 /* force SONIC to read first RRA */
-#define CMDREG_LCAM 0x0200 /* force SONIC to read CAM descrs */
-
-/* Data Configuration Register */
-
-#define SONIC_DCREG 0x02
-#define DCREG_EXBUS 0x8000 /* Extended Bus Mode */
-#define DCREG_LBR 0x2000 /* Latched Bus Retry */
-#define DCREG_PO1 0x1000 /* Programmable Outputs */
-#define DCREG_PO0 0x0800
-#define DCREG_SBUS 0x0400 /* Synchronous Bus Mode */
-#define DCREG_USR1 0x0200 /* User Definable Pins */
-#define DCREG_USR0 0x0100
-#define DCREG_WC0 0x0000 /* 0..3 Wait States */
-#define DCREG_WC1 0x0040
-#define DCREG_WC2 0x0080
-#define DCREG_WC3 0x00c0
-#define DCREG_DW16 0x0000 /* 16 bit Bus Mode */
-#define DCREG_DW32 0x0020 /* 32 bit Bus Mode */
-#define DCREG_BMS 0x0010 /* Block Mode Select */
-#define DCREG_RFT4 0x0000 /* 4/8/16/24 bytes RX Threshold */
-#define DCREG_RFT8 0x0004
-#define DCREG_RFT16 0x0008
-#define DCREG_RFT24 0x000c
-#define DCREG_TFT8 0x0000 /* 8/16/24/28 bytes TX Threshold */
-#define DCREG_TFT16 0x0001
-#define DCREG_TFT24 0x0002
-#define DCREG_TFT28 0x0003
-
-/* Receive Control Register */
-
-#define SONIC_RCREG 0x04
-#define RCREG_ERR 0x8000 /* accept damaged and collided pkts */
-#define RCREG_RNT 0x4000 /* accept packets that are < 64 */
-#define RCREG_BRD 0x2000 /* accept broadcasts */
-#define RCREG_PRO 0x1000 /* promiscuous mode */
-#define RCREG_AMC 0x0800 /* accept all multicasts */
-#define RCREG_LB_NONE 0x0000 /* no loopback */
-#define RCREG_LB_MAC 0x0200 /* MAC loopback */
-#define RCREG_LB_ENDEC 0x0400 /* ENDEC loopback */
-#define RCREG_LB_XVR 0x0600 /* Transceiver loopback */
-#define RCREG_MC 0x0100 /* Multicast received */
-#define RCREG_BC 0x0080 /* Broadcast received */
-#define RCREG_LPKT 0x0040 /* last packet in RBA */
-#define RCREG_CRS 0x0020 /* carrier sense present */
-#define RCREG_COL 0x0010 /* recv'd packet with collision */
-#define RCREG_CRCR 0x0008 /* recv'd packet with CRC error */
-#define RCREG_FAER 0x0004 /* recv'd packet with inv. framing */
-#define RCREG_LBK 0x0002 /* recv'd loopback packet */
-#define RCREG_PRX 0x0001 /* recv'd packet is OK */
-
-/* Transmit Control Register */
-
-#define SONIC_TCREG 0x06
-#define TCREG_PINT 0x8000 /* generate interrupt after TDA read */
-#define TCREG_POWC 0x4000 /* timer start out of window detect */
-#define TCREG_CRCI 0x2000 /* inhibit CRC generation */
-#define TCREG_EXDIS 0x1000 /* disable excessive deferral timer */
-#define TCREG_EXD 0x0400 /* excessive deferral occurred */
-#define TCREG_DEF 0x0200 /* single deferral occurred */
-#define TCREG_NCRS 0x0100 /* no carrier detected */
-#define TCREG_CRSL 0x0080 /* carrier lost */
-#define TCREG_EXC 0x0040 /* excessive collisions occurred */
-#define TCREG_OWC 0x0020 /* out of window collision occurred */
-#define TCREG_PMB 0x0008 /* packet monitored bad */
-#define TCREG_FU 0x0004 /* FIFO underrun */
-#define TCREG_BCM 0x0002 /* byte count mismatch of fragments */
-#define TCREG_PTX 0x0001 /* packet transmitted OK */
-
-/* Interrupt Mask Register */
-
-#define SONIC_IMREG 0x08
-#define IMREG_BREN 0x4000 /* interrupt when bus retry occurred */
-#define IMREG_HBLEN 0x2000 /* interrupt when heartbeat lost */
-#define IMREG_LCDEN 0x1000 /* interrupt when CAM loaded */
-#define IMREG_PINTEN 0x0800 /* interrupt when PINT in TDA set */
-#define IMREG_PRXEN 0x0400 /* interrupt when packet received */
-#define IMREG_PTXEN 0x0200 /* interrupt when packet was sent */
-#define IMREG_TXEREN 0x0100 /* interrupt when send failed */
-#define IMREG_TCEN 0x0080 /* interrupt when timer completed */
-#define IMREG_RDEEN 0x0040 /* interrupt when RDA exhausted */
-#define IMREG_RBEEN 0x0020 /* interrupt when RBA exhausted */
-#define IMREG_RBAEEN 0x0010 /* interrupt when RBA too short */
-#define IMREG_CRCEN 0x0008 /* interrupt when CRC counter rolls */
-#define IMREG_FAEEN 0x0004 /* interrupt when FAE counter rolls */
-#define IMREG_MPEN 0x0002 /* interrupt when MP counter rolls */
-#define IMREG_RFOEN 0x0001 /* interrupt when Rx FIFO overflows */
-
-/* Interrupt Status Register */
-
-#define SONIC_ISREG 0x0a
-#define ISREG_BR 0x4000 /* bus retry occurred */
-#define ISREG_HBL 0x2000 /* heartbeat lost */
-#define ISREG_LCD 0x1000 /* CAM loaded */
-#define ISREG_PINT 0x0800 /* PINT in TDA set */
-#define ISREG_PKTRX 0x0400 /* packet received */
-#define ISREG_TXDN 0x0200 /* packet was sent */
-#define ISREG_TXER 0x0100 /* send failed */
-#define ISREG_TC 0x0080 /* timer completed */
-#define ISREG_RDE 0x0040 /* RDA exhausted */
-#define ISREG_RBE 0x0020 /* RBA exhausted */
-#define ISREG_RBAE 0x0010 /* RBA too short for received frame */
-#define ISREG_CRC 0x0008 /* CRC counter rolls over */
-#define ISREG_FAE 0x0004 /* FAE counter rolls over */
-#define ISREG_MP 0x0002 /* MP counter rolls over */
-#define ISREG_RFO 0x0001 /* Rx FIFO overflows */
-
-#define SONIC_UTDA 0x0c /* current transmit descr address */
-#define SONIC_CTDA 0x0e
-
-#define SONIC_URDA 0x1a /* current receive descr address */
-#define SONIC_CRDA 0x1c
-
-#define SONIC_CRBA0 0x1e /* current receive buffer address */
-#define SONIC_CRBA1 0x20
-
-#define SONIC_RBWC0 0x22 /* word count in receive buffer */
-#define SONIC_RBWC1 0x24
-
-#define SONIC_EOBC 0x26 /* minimum space to be free in RBA */
-
-#define SONIC_URRA 0x28 /* upper address of CDA & Recv Area */
-
-#define SONIC_RSA 0x2a /* start of receive resource area */
-
-#define SONIC_REA 0x2c /* end of receive resource area */
-
-#define SONIC_RRP 0x2e /* resource read pointer */
-
-#define SONIC_RWP 0x30 /* resource write pointer */
-
-#define SONIC_CAMEPTR 0x42 /* CAM entry pointer */
-
-#define SONIC_CAMADDR2 0x44 /* CAM address ports */
-#define SONIC_CAMADDR1 0x46
-#define SONIC_CAMADDR0 0x48
-
-#define SONIC_CAMPTR 0x4c /* lower address of CDA */
-
-#define SONIC_CAMCNT 0x4e /* # of CAM descriptors to load */
-
-/* Data Configuration Register 2 */
-
-#define SONIC_DCREG2 0x7e
-#define DCREG2_EXPO3 0x8000 /* extended programmable outputs */
-#define DCREG2_EXPO2 0x4000
-#define DCREG2_EXPO1 0x2000
-#define DCREG2_EXPO0 0x1000
-#define DCREG2_HD 0x0800 /* heartbeat disable */
-#define DCREG2_JD 0x0200 /* jabber timer disable */
-#define DCREG2_AUTO 0x0100 /* enable AUI/TP auto selection */
-#define DCREG2_XWRAP 0x0040 /* TP transceiver loopback */
-#define DCREG2_PH 0x0010 /* HOLD request timing */
-#define DCREG2_PCM 0x0004 /* packet compress when matched */
-#define DCREG2_PCNM 0x0002 /* packet compress when not matched */
-#define DCREG2_RJCM 0x0001 /* inverse packet match via CAM */
-
-/* Board Control Register: Enable RAM, Interrupts... */
-
-#define BCMREG 0x80
-#define BCMREG_RAMEN 0x80 /* switch over to RAM */
-#define BCMREG_IPEND 0x40 /* interrupt pending ? */
-#define BCMREG_RESET 0x08 /* reset board */
-#define BCMREG_16BIT 0x04 /* adapter in 16-bit slot */
-#define BCMREG_RAMWIN 0x02 /* enable RAM window */
-#define BCMREG_IEN 0x01 /* interrupt enable */
-
-/* MAC Address PROM */
-
-#define MACADDRPROM 0x92
-
-/* structure of a CAM entry */
-
-typedef struct {
- u32 index; /* pointer into CAM area */
- u32 addr0; /* address part (bits 0..15 used) */
- u32 addr1;
- u32 addr2;
-} camentry_t;
-
-/* structure of a receive resource */
-
-typedef struct {
- u32 startlo; /* start address (bits 0..15 used) */
- u32 starthi;
- u32 cntlo; /* size in 16-bit quantities */
- u32 cnthi;
-} rra_t;
-
-/* structure of a receive descriptor */
-
-typedef struct {
- u32 status; /* packet status */
- u32 length; /* length in bytes */
- u32 startlo; /* start address */
- u32 starthi;
- u32 seqno; /* frame sequence */
- u32 link; /* pointer to next descriptor */
- /* bit 0 = EOL */
- u32 inuse; /* !=0 --> free for SONIC to write */
-} rda_t;
-
-/* structure of a transmit descriptor */
-
-typedef struct {
- u32 status; /* transmit status */
- u32 config; /* value for TCR */
- u32 length; /* total length */
- u32 fragcount; /* number of fragments */
- u32 startlo; /* start address of fragment */
- u32 starthi;
- u32 fraglength; /* length of this fragment */
- /* more address/length triplets may */
- /* follow here */
- u32 link; /* pointer to next descriptor */
- /* bit 0 = EOL */
-} tda_t;
-
-#endif /* _IBM_LANA_DRIVER_ */
-
-#endif /* _IBM_LANA_INCLUDE_ */
--
1.8.1
^ permalink raw reply related
* Re: [PATCH 0/5]: soreuseport: Bind multiple sockets to the same port
From: Tom Herbert @ 2013-01-15 1:33 UTC (permalink / raw)
To: Vijay Subramanian; +Cc: David Miller, netdev, netdev, eric.dumazet
In-Reply-To: <CAGK4HS8tHrNUBfd8jmRtWNwFwBARsj7jyxbhD8C7OhhHnpqfQQ@mail.gmail.com>
On Mon, Jan 14, 2013 at 3:35 PM, Vijay Subramanian
<subramanian.vijay@gmail.com> wrote:
> On 14 January 2013 12:29, David Miller <davem@davemloft.net> wrote:
>> From: Tom Herbert <therbert@google.com>
>> Date: Mon, 14 Jan 2013 12:00:16 -0800 (PST)
>>
>>> Rebasing the soreuseport patches to 3.8. No material changes since
>>> first posted.
>
> Tom,
> I am not sure if this series was just RFC or you already are aware of
> this but I got the following errors when compiling with
> CONFIG_NETFILTER_TPROXY=m
>
Thanks,
I'll fix that.
Tom
> This is because the definitions of inet_lookup_listener and
> inet6_lookup_listener have changed with your patch.
>
> In file included from net/netfilter/nf_tproxy_core.c:19:
> include/net/netfilter/nf_tproxy_core.h: In function ‘nf_tproxy_get_sock_v4’:
> include/net/netfilter/nf_tproxy_core.h:86: error: too few arguments to
> function ‘inet_lookup_listener’
> include/net/netfilter/nf_tproxy_core.h: In function ‘nf_tproxy_get_sock_v6’:
> include/net/netfilter/nf_tproxy_core.h:155: warning: passing argument
> 5 of ‘inet6_lookup_listener’ makes pointer from integer without a cast
> include/net/inet6_hashtables.h:72: note: expected ‘const struct
> in6_addr *’ but argument is of type ‘int’
> include/net/netfilter/nf_tproxy_core.h:155: error: too few arguments
> to function ‘inet6_lookup_listener’
> make[1]: *** [net/netfilter/nf_tproxy_core.o] Error 1
> make[1]: *** Waiting for unfinished jobs....
>
>
> Thanks,
> Vijay
^ permalink raw reply
* Re: [PATCH 1/2] usbnet: allow status interrupt URB to always be active
From: Ming Lei @ 2013-01-15 1:13 UTC (permalink / raw)
To: Dan Williams
Cc: Oliver Neukum, Elina Pasheva, netdev, linux-usb, Rory Filer,
Phil Sutter
In-Reply-To: <1358184201.1713.30.camel@dcbw.foobar.com>
On Tue, Jan 15, 2013 at 1:23 AM, Dan Williams <dcbw@redhat.com> wrote:
> On Fri, 2013-01-11 at 11:06 +0800, Ming Lei wrote:
>> I am curious who are interested in the 'custom Restart indication'
>> information after the interface is closed.
>
> It's actually before the interface is even opened. It's really just a
> sync signal that's part of the driver's setup/initialization of the
> device.
OK, got it.
Considered that it is only required by Sierra, could you submit
the status URB in driver->bind and cancel it in driver->reset so
we can avoid touching usbnet core code(one simple change
might be to move init_status() before calling driver->info()
inside usbnet_probe())?
>
>> If sierra_net provides ways(such as read registers) to query the
>> indication event, you can just query the information and setup
>> the device in driver_info->reset() during device open, so you can
>> avoid submitting interrupt URB always.
>
> As far as I know, it does not, or at least Sierra hasn't released such
> information about the firmware API of their devices.
You are unlucky, :-)
In fact, many usbnet devices provide read command to query this kind of
information.
Thanks,
--
Ming Lei
^ permalink raw reply
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