* [PATCH V2 0/3] handle polling errors in vhost/vhost_net
From: Jason Wang @ 2013-01-05 9:34 UTC (permalink / raw)
To: davem, mst, netdev, linux-kernel; +Cc: Jason Wang
This is an update version of last version to fix the handling of polling errors
in vhost/vhost_net.
Currently, vhost and vhost_net ignores errors of polling which can lead kernel
crashing for buggy/malicious userspace. Fix this by extending the idea of tx
polling state to all other vhost_polls and changing the state only when polling
succeed.
Jason Wang (3):
vhost_net: correct error handling in vhost_net_set_backend()
vhost: handle polling errors
tuntap: don't add to waitqueue when POLLERR
drivers/net/tun.c | 5 +--
drivers/vhost/net.c | 89 ++++++++++++++++++++----------------------------
drivers/vhost/vhost.c | 25 +++++++++++---
drivers/vhost/vhost.h | 11 +++++-
4 files changed, 68 insertions(+), 62 deletions(-)
^ permalink raw reply
* Re: [3.8-rc] regression: NETDEV WATCHDOG: eth0 (r8169): transmit queue 0 timed out
From: Francois Romieu @ 2013-01-05 9:37 UTC (permalink / raw)
To: Jörg Otte; +Cc: Linux Kernel Mailing List, David S. Miller, netdev
In-Reply-To: <CADDKRnBqxSwC7HUP_aN1QELrv2+7jnncXj7EP0Pqv8E8FFTpPw@mail.gmail.com>
Jörg Otte <jrg.otte@gmail.com> :
[...]
> It's a regression, it never happend before 3.8-rc.
Please check that 'dmesg | grep XID' exhibits a 8168evl.
I'll showe and dig it. It's epidemic.
--
Ueimor
^ permalink raw reply
* Re: [3.8-rc] regression: NETDEV WATCHDOG: eth0 (r8169): transmit queue 0 timed out
From: Jörg Otte @ 2013-01-05 10:15 UTC (permalink / raw)
To: Francois Romieu; +Cc: Linux Kernel Mailing List, David S. Miller, netdev
In-Reply-To: <20130105093718.GA938@electric-eye.fr.zoreil.com>
2013/1/5 Francois Romieu <romieu@fr.zoreil.com>:
> Jörg Otte <jrg.otte@gmail.com> :
> [...]
>> It's a regression, it never happend before 3.8-rc.
>
> Please check that 'dmesg | grep XID' exhibits a 8168evl.
jojo@ahorn:~$ dmesg | grep XID
[ 1.808847] r8169 0000:02:00.0 eth0: RTL8168evl/8111evl at
0xffffc90000054000, 5c:9a:d8:69:2b:39, XID 0c900800 IRQ 42
jojo@ahorn:~$
>
> I'll showe and dig it. It's epidemic.
>
Thanks, Jörg
^ permalink raw reply
* [PATCH] ethtool: Add version check for et131x regs
From: Mark Einon @ 2013-01-05 9:57 UTC (permalink / raw)
To: bhutchings; +Cc: netdev, Mark Einon
Added a version check for the et131x reg dump, in case the dump format
needs to be changed later.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
---
et131x.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/et131x.c b/et131x.c
index b36c184..8cdbec0 100644
--- a/et131x.c
+++ b/et131x.c
@@ -4,8 +4,12 @@
int et131x_dump_regs(struct ethtool_drvinfo *info, struct ethtool_regs *regs)
{
+ u8 version = (u8)(regs->version >> 24);
u32 *reg = (u32 *)regs->data;
+ if(version != 1)
+ return -1;
+
fprintf(stdout, "PHY Registers\n");
fprintf(stdout, "0x0, Basic Control Reg = 0x%04X\n", *reg++);
fprintf(stdout, "0x1, Basic Status Reg = 0x%04X\n", *reg++);
--
1.7.9.5
^ permalink raw reply related
* Re: [PATCH 1/2] usbnet: allow status interrupt URB to always be active
From: Bjørn Mork @ 2013-01-05 10:59 UTC (permalink / raw)
To: Dan Williams
Cc: Oliver Neukum, Elina Pasheva, netdev-u79uwXL29TY76Z2rM5mHXA,
linux-usb-u79uwXL29TY76Z2rM5mHXA, Rory Filer, Phil Sutter
In-Reply-To: <1357349193.19684.3.camel-wKZy7rqYPVb5EHUCmHmTqw@public.gmane.org>
Dan Williams <dcbw-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> writes:
> On Fri, 2013-01-04 at 23:16 +0100, Oliver Neukum wrote:
>> On Friday 04 January 2013 10:48:16 Dan Williams wrote:
>> > Some drivers (ex sierra_net) need the status interrupt URB
>> > active even when the device is closed, because they receive
>> > custom indications from firmware. Allow sub-drivers to set
>> > a flag that submits the status interrupt URB on probe and
>> > keeps the URB alive over device open/close. The URB is still
>> > killed/re-submitted for suspend/resume, as before.
>> >
>> > Signed-off-by: Dan Williams <dcbw-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
>> > ---
>> > Oliver: alternatively, is there a problem with *always*
>> > submitting the interrupt URB, and then simply not calling
>> > the subdriver's .status function when the netdev is
>> > closed? That would be a much simpler patch.
>>
>> That is quite radical. We have no idea what a device
>> does when we do not react to a status update. I would
>> much prefer to not take the risk.
>> Besides, we don't use bandwidth if we don't have to.
>
> Ok, so scratch the alternative. Thus, does the posted patch look like
> the right course of action?
>
> If I wasn't clear enough before, sierra_net needs to listen to the
> status interrupt URB to receive the custom Restart indication as part of
> the driver's device setup. Thus for sierra_net at least, tying the
> status interrupt URB submission to device open/close isn't right.
>
> I'd previously done a patch to handle this all in sierra_net, but the
> problem there is suspend/resume: without directly accessing the usbnet
> structure's ->suspend_count member (icky!) sierra_net can't correctly
> kill/submit the URB itself. So I went with a flag to usbnet that Sierra
> can set.
Just a few random thoughts...
The sierra_net driver would have been an excellent candidate for the
cdc-wdm subdriver model if that had existed when sierra_net was written.
I assume the current driver only implements a minimal subset of the
DirectIP HIP protocol embedded in CDC, and exporting this to userspace
instead would have made it possible to extend that support without
changing the driver, in addtion to making the driver much more robust
against firmware differences. It would have eliminated the problem you
are facing and removed the minidriver workqueue complexity.
But I guess it's too late to change this now. In theory we could
implement the cdc-wdm hooks that were initially proposed for cdc_mbim
and rewrite sierra_net to use it while being backwards compatible.
Don't think anyone wants to do either, so forget about it...
You can still use a trick similar to what qmi_wwan and cdc_mbim does to
take over the status endpoint from usbnet: By not implementing .status,
and possibly setting dev->status to NULL in .bind, you are free to
handle the status endpoint entirely inside the minidriver. Not sure if
that is smart though. You would have to reimplement init_status and
intr_complete from usbnet, and kill or resubmit the interrupt urb on
suspend/resume/disconnect yourself.
The new usbnet flag is probably a better solution.
FWIW, I agree with Oliver that always submitting the interrupt URB is
both risky and will cause too much unnecessary USB activity for most
usbnet devices.
Bjørn
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH 1/2] usbnet: allow status interrupt URB to always be active
From: Oliver Neukum @ 2013-01-05 11:01 UTC (permalink / raw)
To: Dan Williams; +Cc: Elina Pasheva, netdev, linux-usb, Rory Filer, Phil Sutter
In-Reply-To: <1357349193.19684.3.camel@dcbw.foobar.com>
On Friday 04 January 2013 19:26:33 Dan Williams wrote:
> On Fri, 2013-01-04 at 23:16 +0100, Oliver Neukum wrote:
> > On Friday 04 January 2013 10:48:16 Dan Williams wrote:
> > > Some drivers (ex sierra_net) need the status interrupt URB
> > > active even when the device is closed, because they receive
> > > custom indications from firmware. Allow sub-drivers to set
> > > a flag that submits the status interrupt URB on probe and
> > > keeps the URB alive over device open/close. The URB is still
> > > killed/re-submitted for suspend/resume, as before.
> > >
> > > Signed-off-by: Dan Williams <dcbw@redhat.com>
> > > ---
> > > Oliver: alternatively, is there a problem with *always*
> > > submitting the interrupt URB, and then simply not calling
> > > the subdriver's .status function when the netdev is
> > > closed? That would be a much simpler patch.
> >
> > That is quite radical. We have no idea what a device
> > does when we do not react to a status update. I would
> > much prefer to not take the risk.
> > Besides, we don't use bandwidth if we don't have to.
>
> Ok, so scratch the alternative. Thus, does the posted patch look like
> the right course of action?
In principle yes.
> If I wasn't clear enough before, sierra_net needs to listen to the
> status interrupt URB to receive the custom Restart indication as part of
> the driver's device setup. Thus for sierra_net at least, tying the
> status interrupt URB submission to device open/close isn't right.
So, there seems to be an inevitable race before probe() is called.
Have you looked at FLAG_AVOID_UNLINK_URBS ?
> I'd previously done a patch to handle this all in sierra_net, but the
> problem there is suspend/resume: without directly accessing the usbnet
> structure's ->suspend_count member (icky!) sierra_net can't correctly
> kill/submit the URB itself. So I went with a flag to usbnet that Sierra
> can set.
That is absolutely the right way to do it.
Regards
Oliver
^ permalink raw reply
* Re: [PATCH 1/2] usbnet: allow status interrupt URB to always be active
From: Bjørn Mork @ 2013-01-05 11:44 UTC (permalink / raw)
To: Oliver Neukum
Cc: Dan Williams, Elina Pasheva, netdev, linux-usb, Rory Filer,
Phil Sutter
In-Reply-To: <3476529.YzSzbMTfeu@linux-5eaq.site>
Oliver Neukum <oliver@neukum.org> writes:
> On Friday 04 January 2013 19:26:33 Dan Williams wrote:
>
>> I'd previously done a patch to handle this all in sierra_net, but the
>> problem there is suspend/resume: without directly accessing the usbnet
>> structure's ->suspend_count member (icky!) sierra_net can't correctly
>> kill/submit the URB itself. So I went with a flag to usbnet that Sierra
>> can set.
>
> That is absolutely the right way to do it.
Yes.
Just a comment regarding the ->suspend_count: Are you absolutely sure
you need to look at that, Dan? usbnet uses it to handle suspend/resume
for minidrivers with an unknown number of interfaces, without knowing
whether it is the control or data interface which is suspended or
resumed first. By using the counter it can ensure that the correct
action is taken exactly once regardless of this.
The sierra_net minidriver has the advantage of knowing that there always
is only *one* USB interface being suspended and resumed. So you don't
have to care about ->suspend_count. Just do whatever you need to do on
suspend and resume.
Bjørn
^ permalink raw reply
* Re: [PATCH] bgmac: driver for GBit MAC core on BCMA bus
From: Rafał Miłecki @ 2013-01-05 12:08 UTC (permalink / raw)
To: Francois Romieu; +Cc: netdev, David S. Miller
In-Reply-To: <20130103203027.GA4305@electric-eye.fr.zoreil.com>
2013/1/3 Francois Romieu <romieu@fr.zoreil.com>:
> Rafał Miłecki <zajec5@gmail.com> :
>> > The driver should stop queueing after its processing if it detects that
>> > it can't handle more requests.
>>
>> I think it's what the code does... netif_stop_queue and return
>> NETDEV_TX_BUSY. Anything wrong about it?
>
> The driver does not hold its promise to send the packet. It should
> netif_stop_queue and return NETDEV_TX_OK after the last transmitted
> packet.
We really need some good explanation for NETDEV_TX_BUSY somewhere
(documentation!).
I've checked some drivers and:
1) b44.c
In case of:
> if (unlikely(TX_BUFFS_AVAIL(bp) < 1)) {
it does netif_stop_queue and returns NETDEV_TX_BUSY
2) sky2.c
In case of:
> if (unlikely(tx_avail(sky2) < tx_le_req(skb)))
it just returns NETDEV_TX_BUSY
3) jme.2
In case of:
> if (unlikely(idx < 0)) {
it does netif_stop_queue and returns NETDEV_TX_BUSY
4) de2104x.c
In case of:
> if (tx_free == 0) {
it does netif_stop_queue and returns NETDEV_TX_BUSY
So there are many drivers doing the same mistake I did.
--
Rafał
^ permalink raw reply
* [PATCH net-next] team: use strlcpy with ethtool_drvinfo fields
From: Flavio Leitner @ 2013-01-05 12:53 UTC (permalink / raw)
To: netdev; +Cc: Ben Hutchings, Jiri Pirko, Flavio Leitner
The fields must be null-terminated.
Signed-off-by: Flavio Leitner <fbl@redhat.com>
---
drivers/net/team/team.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 78c7d87..e27ac5c 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -1747,8 +1747,8 @@ static const struct net_device_ops team_netdev_ops = {
static void team_ethtool_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, DRV_NAME, 32);
- strncpy(drvinfo->version, UTS_RELEASE, 32);
+ strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, UTS_RELEASE, sizeof(drvinfo->version));
}
static const struct ethtool_ops team_ethtool_ops = {
--
1.8.0.1
^ permalink raw reply related
* [PATCH V3] bgmac: driver for GBit MAC core on BCMA bus
From: Rafał Miłecki @ 2013-01-05 12:52 UTC (permalink / raw)
To: netdev, David S. Miller
Cc: Joe Perches, Francois Romieu, Rafał Miłecki
BCMA is a Broadcom specific bus with devices AKA cores. All recent BCMA
based SoCs have gigabit ethernet provided by the GBit MAC core. This
patch adds driver for such a cores registering itself as a netdev. It
has been tested on a BCM4706 and BCM4718 chipsets.
In the kernel tree there is already b44 driver which has some common
things with bgmac, however there are many differences that has led to
the decision or writing a new driver:
1) GBit MAC cores appear on BCMA bus (not SSB as in case of b44)
2) There is 64bit DMA engine which differs from 32bit one
3) There is no CAM (Content Addressable Memory) in GBit MAC
4) We have 4 TX queues on GBit MAC devices (instead of 1)
5) Many registers have different addresses/values
6) RX header flags are also different
The driver in it's state is functional how, however there is of course
place for improvements:
1) Supporting more net_device_ops
2) SUpporting more ethtool_ops
3) Unaligned addressing in DMA
4) Writing separated PHY driver
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
---
V2: fix typos
improve stat_xmit (netif_stop_queue)
change scope of some variables
switch to dev_kfree_skb
change net_dev->stats.rx_dropped
drop "bool tx;" from DMA ring struct
bgmac_open: fail on IRQ request error
use dev<level> for printing (this is great!)
V3: fix strings (long lines)
fix returned values in xmit
some extra info message about PHY addr
---
drivers/bcma/driver_chipcommon_pmu.c | 3 +-
drivers/net/ethernet/broadcom/Kconfig | 9 +
drivers/net/ethernet/broadcom/Makefile | 1 +
drivers/net/ethernet/broadcom/bgmac.c | 1412 +++++++++++++++++++++++++++
drivers/net/ethernet/broadcom/bgmac.h | 447 +++++++++
include/linux/bcma/bcma_driver_chipcommon.h | 2 +
6 files changed, 1873 insertions(+), 1 deletions(-)
create mode 100644 drivers/net/ethernet/broadcom/bgmac.c
create mode 100644 drivers/net/ethernet/broadcom/bgmac.h
diff --git a/drivers/bcma/driver_chipcommon_pmu.c b/drivers/bcma/driver_chipcommon_pmu.c
index c62c788..932b101 100644
--- a/drivers/bcma/driver_chipcommon_pmu.c
+++ b/drivers/bcma/driver_chipcommon_pmu.c
@@ -264,7 +264,7 @@ static u32 bcma_pmu_pll_clock_bcm4706(struct bcma_drv_cc *cc, u32 pll0, u32 m)
}
/* query bus clock frequency for PMU-enabled chipcommon */
-static u32 bcma_pmu_get_bus_clock(struct bcma_drv_cc *cc)
+u32 bcma_pmu_get_bus_clock(struct bcma_drv_cc *cc)
{
struct bcma_bus *bus = cc->core->bus;
@@ -293,6 +293,7 @@ static u32 bcma_pmu_get_bus_clock(struct bcma_drv_cc *cc)
}
return BCMA_CC_PMU_HT_CLOCK;
}
+EXPORT_SYMBOL_GPL(bcma_pmu_get_bus_clock);
/* query cpu clock frequency for PMU-enabled chipcommon */
u32 bcma_pmu_get_cpu_clock(struct bcma_drv_cc *cc)
diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig
index f552673..42c0913 100644
--- a/drivers/net/ethernet/broadcom/Kconfig
+++ b/drivers/net/ethernet/broadcom/Kconfig
@@ -121,4 +121,13 @@ config BNX2X
To compile this driver as a module, choose M here: the module
will be called bnx2x. This is recommended.
+config BGMAC
+ tristate "BCMA bus GBit core support"
+ depends on BCMA_HOST_SOC && HAS_DMA
+ ---help---
+ This driver supports GBit MAC and BCM4706 GBit MAC cores on BCMA bus.
+ They can be found on BCM47xx SoCs and provide gigabit ethernet.
+ In case of using this driver on BCM4706 it's also requires to enable
+ BCMA_DRIVER_GMAC_CMN to make it work.
+
endif # NET_VENDOR_BROADCOM
diff --git a/drivers/net/ethernet/broadcom/Makefile b/drivers/net/ethernet/broadcom/Makefile
index b789605..68efa1a 100644
--- a/drivers/net/ethernet/broadcom/Makefile
+++ b/drivers/net/ethernet/broadcom/Makefile
@@ -9,3 +9,4 @@ obj-$(CONFIG_CNIC) += cnic.o
obj-$(CONFIG_BNX2X) += bnx2x/
obj-$(CONFIG_SB1250_MAC) += sb1250-mac.o
obj-$(CONFIG_TIGON3) += tg3.o
+obj-$(CONFIG_BGMAC) += bgmac.o
diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
new file mode 100644
index 0000000..2d06c1c
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -0,0 +1,1412 @@
+/*
+ * Driver for (BCM4706)? GBit MAC core on BCMA bus.
+ *
+ * Copyright (C) 2012 Rafał Miłecki <zajec5@gmail.com>
+ *
+ * Licensed under the GNU/GPL. See COPYING for details.
+ */
+
+#include "bgmac.h"
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/etherdevice.h>
+#include <linux/mii.h>
+#include <linux/interrupt.h>
+#include <linux/dma-mapping.h>
+#include <asm/mach-bcm47xx/nvram.h>
+
+static const struct bcma_device_id bgmac_bcma_tbl[] = {
+ BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_4706_MAC_GBIT, BCMA_ANY_REV, BCMA_ANY_CLASS),
+ BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_MAC_GBIT, BCMA_ANY_REV, BCMA_ANY_CLASS),
+ BCMA_CORETABLE_END
+};
+MODULE_DEVICE_TABLE(bcma, bgmac_bcma_tbl);
+
+static bool bgmac_wait_value(struct bcma_device *core, u16 reg, u32 mask,
+ u32 value, int timeout)
+{
+ u32 val;
+ int i;
+
+ for (i = 0; i < timeout / 10; i++) {
+ val = bcma_read32(core, reg);
+ if ((val & mask) == value)
+ return true;
+ udelay(10);
+ }
+ pr_err("Timeout waiting for reg 0x%X\n", reg);
+ return false;
+}
+
+/**************************************************
+ * DMA
+ **************************************************/
+
+static void bgmac_dma_tx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
+{
+ u32 val;
+ int i;
+
+ if (!ring->mmio_base)
+ return;
+
+ /* Suspend DMA TX ring first.
+ * bgmac_wait_value doesn't support waiting for any of few values, so
+ * implement whole loop here.
+ */
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL,
+ BGMAC_DMA_TX_SUSPEND);
+ for (i = 0; i < 10000 / 10; i++) {
+ val = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS);
+ val &= BGMAC_DMA_TX_STAT;
+ if (val == BGMAC_DMA_TX_STAT_DISABLED ||
+ val == BGMAC_DMA_TX_STAT_IDLEWAIT ||
+ val == BGMAC_DMA_TX_STAT_STOPPED) {
+ i = 0;
+ break;
+ }
+ udelay(10);
+ }
+ if (i)
+ bgmac_err(bgmac, "Timeout suspending DMA TX ring 0x%X (BGMAC_DMA_TX_STAT: 0x%08X)\n",
+ ring->mmio_base, val);
+
+ /* Remove SUSPEND bit */
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL, 0);
+ if (!bgmac_wait_value(bgmac->core,
+ ring->mmio_base + BGMAC_DMA_TX_STATUS,
+ BGMAC_DMA_TX_STAT, BGMAC_DMA_TX_STAT_DISABLED,
+ 10000)) {
+ bgmac_warn(bgmac, "DMA TX ring 0x%X wasn't disabled on time, waiting additional 300us\n",
+ ring->mmio_base);
+ udelay(300);
+ val = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS);
+ if ((val & BGMAC_DMA_TX_STAT) != BGMAC_DMA_TX_STAT_DISABLED)
+ bgmac_err(bgmac, "Reset of DMA TX ring 0x%X failed\n",
+ ring->mmio_base);
+ }
+}
+
+static void bgmac_dma_tx_enable(struct bgmac *bgmac,
+ struct bgmac_dma_ring *ring)
+{
+ u32 ctl;
+
+ ctl = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL);
+ ctl |= BGMAC_DMA_TX_ENABLE;
+ ctl |= BGMAC_DMA_TX_PARITY_DISABLE;
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL, ctl);
+}
+
+static netdev_tx_t bgmac_dma_tx_add(struct bgmac *bgmac,
+ struct bgmac_dma_ring *ring,
+ struct sk_buff *skb)
+{
+ struct device *dma_dev = bgmac->core->dma_dev;
+ struct bgmac_dma_desc *dma_desc;
+ struct bgmac_slot_info *slot;
+ u32 ctl0, ctl1;
+ int free_slots;
+
+ if (skb->len > BGMAC_DESC_CTL1_LEN) {
+ bgmac_err(bgmac, "Too long skb (%d)\n", skb->len);
+ goto err_stop_drop;
+ }
+
+ if (ring->start <= ring->end)
+ free_slots = ring->start - ring->end + BGMAC_TX_RING_SLOTS;
+ else
+ free_slots = ring->start - ring->end;
+ if (free_slots <= 1) {
+ bgmac_err(bgmac, "No free slots on ring 0x%X!\n",
+ ring->mmio_base);
+ goto err_stop_drop;
+ }
+
+ slot = &ring->slots[ring->end];
+ slot->skb = skb;
+ slot->dma_addr = dma_map_single(dma_dev, skb->data, skb->len,
+ DMA_TO_DEVICE);
+ if (dma_mapping_error(dma_dev, slot->dma_addr)) {
+ bgmac_err(bgmac, "Mapping error of skb on ring 0x%X\n",
+ ring->mmio_base);
+ goto err_stop_drop;
+ }
+
+ ctl0 = BGMAC_DESC_CTL0_IOC | BGMAC_DESC_CTL0_SOF | BGMAC_DESC_CTL0_EOF;
+ if (ring->end == ring->num_slots - 1)
+ ctl0 |= BGMAC_DESC_CTL0_EOT;
+ ctl1 = skb->len & BGMAC_DESC_CTL1_LEN;
+
+ dma_desc = ring->cpu_base;
+ dma_desc += ring->end;
+ dma_desc->addr_low = cpu_to_le32(lower_32_bits(slot->dma_addr));
+ dma_desc->addr_high = cpu_to_le32(upper_32_bits(slot->dma_addr));
+ dma_desc->ctl0 = cpu_to_le32(ctl0);
+ dma_desc->ctl1 = cpu_to_le32(ctl1);
+
+ wmb();
+
+ /* Increase ring->end to point empty slot. We tell hardware the first
+ * slot it should *not* read.
+ */
+ if (++ring->end >= BGMAC_TX_RING_SLOTS)
+ ring->end = 0;
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_INDEX,
+ ring->end * sizeof(struct bgmac_dma_desc));
+
+ return NETDEV_TX_OK;
+
+err_stop_drop:
+ netif_stop_queue(bgmac->net_dev);
+ dev_kfree_skb(skb);
+ return NETDEV_TX_OK;
+}
+
+/* Free transmitted packets */
+static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
+{
+ struct device *dma_dev = bgmac->core->dma_dev;
+ int empty_slot;
+
+ /* The last slot that hardware didn't consume yet */
+ empty_slot = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS);
+ empty_slot &= BGMAC_DMA_TX_STATDPTR;
+ empty_slot /= sizeof(struct bgmac_dma_desc);
+
+ while (ring->start != empty_slot) {
+ struct bgmac_slot_info *slot = &ring->slots[ring->start];
+
+ if (slot->skb) {
+ /* Unmap no longer used buffer */
+ dma_unmap_single(dma_dev, slot->dma_addr,
+ slot->skb->len, DMA_TO_DEVICE);
+ slot->dma_addr = 0;
+
+ /* Free memory! :) */
+ dev_kfree_skb(slot->skb);
+ slot->skb = NULL;
+ } else {
+ bgmac_err(bgmac, "Hardware reported transmission for empty TX ring slot %d! End of ring: %d\n",
+ ring->start, ring->end);
+ }
+
+ if (++ring->start >= BGMAC_TX_RING_SLOTS)
+ ring->start = 0;
+ }
+}
+
+static void bgmac_dma_rx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
+{
+ if (!ring->mmio_base)
+ return;
+
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL, 0);
+ if (!bgmac_wait_value(bgmac->core,
+ ring->mmio_base + BGMAC_DMA_RX_STATUS,
+ BGMAC_DMA_RX_STAT, BGMAC_DMA_RX_STAT_DISABLED,
+ 10000))
+ bgmac_err(bgmac, "Reset of ring 0x%X RX failed\n",
+ ring->mmio_base);
+}
+
+static void bgmac_dma_rx_enable(struct bgmac *bgmac,
+ struct bgmac_dma_ring *ring)
+{
+ u32 ctl;
+
+ ctl = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL);
+ ctl &= BGMAC_DMA_RX_ADDREXT_MASK;
+ ctl |= BGMAC_DMA_RX_ENABLE;
+ ctl |= BGMAC_DMA_RX_PARITY_DISABLE;
+ ctl |= BGMAC_DMA_RX_OVERFLOW_CONT;
+ ctl |= BGMAC_RX_FRAME_OFFSET << BGMAC_DMA_RX_FRAME_OFFSET_SHIFT;
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL, ctl);
+}
+
+static int bgmac_dma_rx_skb_for_slot(struct bgmac *bgmac,
+ struct bgmac_slot_info *slot)
+{
+ struct device *dma_dev = bgmac->core->dma_dev;
+ struct bgmac_rx_header *rx;
+
+ /* Alloc skb */
+ slot->skb = netdev_alloc_skb(bgmac->net_dev, BGMAC_RX_BUF_SIZE);
+ if (!slot->skb) {
+ bgmac_err(bgmac, "Allocation of skb failed!\n");
+ return -ENOMEM;
+ }
+
+ /* Poison - if everything goes fine, hardware will overwrite it */
+ rx = (struct bgmac_rx_header *)slot->skb->data;
+ rx->len = cpu_to_le16(0xdead);
+ rx->flags = cpu_to_le16(0xbeef);
+
+ /* Map skb for the DMA */
+ slot->dma_addr = dma_map_single(dma_dev, slot->skb->data,
+ BGMAC_RX_BUF_SIZE, DMA_FROM_DEVICE);
+ if (dma_mapping_error(dma_dev, slot->dma_addr)) {
+ bgmac_err(bgmac, "DMA mapping error\n");
+ return -ENOMEM;
+ }
+ if (slot->dma_addr & 0xC0000000)
+ bgmac_warn(bgmac, "DMA address using 0xC0000000 bit(s), it may need translation trick\n");
+
+ return 0;
+}
+
+static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring,
+ int weight)
+{
+ u32 end_slot;
+ int handled = 0;
+
+ end_slot = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_STATUS);
+ end_slot &= BGMAC_DMA_RX_STATDPTR;
+ end_slot /= sizeof(struct bgmac_dma_desc);
+
+ ring->end = end_slot;
+
+ while (ring->start != ring->end) {
+ struct device *dma_dev = bgmac->core->dma_dev;
+ struct bgmac_slot_info *slot = &ring->slots[ring->start];
+ struct sk_buff *skb = slot->skb;
+ struct sk_buff *new_skb;
+ struct bgmac_rx_header *rx;
+ u16 len, flags;
+
+ /* Unmap buffer to make it accessible to the CPU */
+ dma_sync_single_for_cpu(dma_dev, slot->dma_addr,
+ BGMAC_RX_BUF_SIZE, DMA_FROM_DEVICE);
+
+ /* Get info from the header */
+ rx = (struct bgmac_rx_header *)skb->data;
+ len = le16_to_cpu(rx->len);
+ flags = le16_to_cpu(rx->flags);
+
+ /* Check for poison and drop or pass the packet */
+ if (len == 0xdead && flags == 0xbeef) {
+ bgmac_err(bgmac, "Found poisoned packet at slot %d, DMA issue!\n",
+ ring->start);
+ } else {
+ new_skb = netdev_alloc_skb(bgmac->net_dev, len);
+ if (new_skb) {
+ skb_put(new_skb, len);
+ skb_copy_from_linear_data_offset(skb, BGMAC_RX_FRAME_OFFSET,
+ new_skb->data,
+ len);
+ new_skb->protocol =
+ eth_type_trans(new_skb, bgmac->net_dev);
+ netif_receive_skb(new_skb);
+ handled++;
+ } else {
+ bgmac->net_dev->stats.rx_dropped++;
+ bgmac_err(bgmac, "Allocation of skb for copying packet failed!\n");
+ }
+
+ /* Poison the old skb */
+ rx->len = cpu_to_le16(0xdead);
+ rx->flags = cpu_to_le16(0xbeef);
+ }
+
+ /* Make it back accessible to the hardware */
+ dma_sync_single_for_device(dma_dev, slot->dma_addr,
+ BGMAC_RX_BUF_SIZE, DMA_FROM_DEVICE);
+
+ if (++ring->start >= BGMAC_RX_RING_SLOTS)
+ ring->start = 0;
+
+ if (handled >= weight) /* Should never be greater */
+ break;
+ }
+
+ return handled;
+}
+
+/* Does ring support unaligned addressing? */
+static bool bgmac_dma_unaligned(struct bgmac *bgmac,
+ struct bgmac_dma_ring *ring,
+ enum bgmac_dma_ring_type ring_type)
+{
+ switch (ring_type) {
+ case BGMAC_DMA_RING_TX:
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGLO,
+ 0xff0);
+ if (bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGLO))
+ return true;
+ break;
+ case BGMAC_DMA_RING_RX:
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGLO,
+ 0xff0);
+ if (bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGLO))
+ return true;
+ break;
+ }
+ return false;
+}
+
+static void bgmac_dma_ring_free(struct bgmac *bgmac,
+ struct bgmac_dma_ring *ring)
+{
+ struct device *dma_dev = bgmac->core->dma_dev;
+ struct bgmac_slot_info *slot;
+ int size;
+ int i;
+
+ for (i = 0; i < ring->num_slots; i++) {
+ slot = &ring->slots[i];
+ if (slot->skb) {
+ if (slot->dma_addr)
+ dma_unmap_single(dma_dev, slot->dma_addr,
+ slot->skb->len, DMA_TO_DEVICE);
+ dev_kfree_skb(slot->skb);
+ }
+ }
+
+ if (ring->cpu_base) {
+ /* Free ring of descriptors */
+ size = ring->num_slots * sizeof(struct bgmac_dma_desc);
+ dma_free_coherent(dma_dev, size, ring->cpu_base,
+ ring->dma_base);
+ }
+}
+
+static void bgmac_dma_free(struct bgmac *bgmac)
+{
+ int i;
+
+ for (i = 0; i < BGMAC_MAX_TX_RINGS; i++)
+ bgmac_dma_ring_free(bgmac, &bgmac->tx_ring[i]);
+ for (i = 0; i < BGMAC_MAX_RX_RINGS; i++)
+ bgmac_dma_ring_free(bgmac, &bgmac->rx_ring[i]);
+}
+
+static int bgmac_dma_alloc(struct bgmac *bgmac)
+{
+ struct device *dma_dev = bgmac->core->dma_dev;
+ struct bgmac_dma_ring *ring;
+ static const u16 ring_base[] = { BGMAC_DMA_BASE0, BGMAC_DMA_BASE1,
+ BGMAC_DMA_BASE2, BGMAC_DMA_BASE3, };
+ int size; /* ring size: different for Tx and Rx */
+ int err;
+ int i;
+
+ BUILD_BUG_ON(BGMAC_MAX_TX_RINGS > ARRAY_SIZE(ring_base));
+ BUILD_BUG_ON(BGMAC_MAX_RX_RINGS > ARRAY_SIZE(ring_base));
+
+ if (!(bcma_aread32(bgmac->core, BCMA_IOST) & BCMA_IOST_DMA64)) {
+ bgmac_err(bgmac, "Core does not report 64-bit DMA\n");
+ return -ENOTSUPP;
+ }
+
+ for (i = 0; i < BGMAC_MAX_TX_RINGS; i++) {
+ ring = &bgmac->tx_ring[i];
+ ring->num_slots = BGMAC_TX_RING_SLOTS;
+ ring->mmio_base = ring_base[i];
+ if (bgmac_dma_unaligned(bgmac, ring, BGMAC_DMA_RING_TX))
+ bgmac_warn(bgmac, "TX on ring 0x%X supports unaligned addressing but this feature is not implemented\n",
+ ring->mmio_base);
+
+ /* Alloc ring of descriptors */
+ size = ring->num_slots * sizeof(struct bgmac_dma_desc);
+ ring->cpu_base = dma_zalloc_coherent(dma_dev, size,
+ &ring->dma_base,
+ GFP_KERNEL);
+ if (!ring->cpu_base) {
+ bgmac_err(bgmac, "Allocation of TX ring 0x%X failed\n",
+ ring->mmio_base);
+ goto err_dma_free;
+ }
+ if (ring->dma_base & 0xC0000000)
+ bgmac_warn(bgmac, "DMA address using 0xC0000000 bit(s), it may need translation trick\n");
+
+ /* No need to alloc TX slots yet */
+ }
+
+ for (i = 0; i < BGMAC_MAX_RX_RINGS; i++) {
+ ring = &bgmac->rx_ring[i];
+ ring->num_slots = BGMAC_RX_RING_SLOTS;
+ ring->mmio_base = ring_base[i];
+ if (bgmac_dma_unaligned(bgmac, ring, BGMAC_DMA_RING_RX))
+ bgmac_warn(bgmac, "RX on ring 0x%X supports unaligned addressing but this feature is not implemented\n",
+ ring->mmio_base);
+
+ /* Alloc ring of descriptors */
+ size = ring->num_slots * sizeof(struct bgmac_dma_desc);
+ ring->cpu_base = dma_zalloc_coherent(dma_dev, size,
+ &ring->dma_base,
+ GFP_KERNEL);
+ if (!ring->cpu_base) {
+ bgmac_err(bgmac, "Allocation of RX ring 0x%X failed\n",
+ ring->mmio_base);
+ err = -ENOMEM;
+ goto err_dma_free;
+ }
+ if (ring->dma_base & 0xC0000000)
+ bgmac_warn(bgmac, "DMA address using 0xC0000000 bit(s), it may need translation trick\n");
+
+ /* Alloc RX slots */
+ for (i = 0; i < ring->num_slots; i++) {
+ err = bgmac_dma_rx_skb_for_slot(bgmac, &ring->slots[i]);
+ if (err) {
+ bgmac_err(bgmac, "Can't allocate skb for slot in RX ring\n");
+ goto err_dma_free;
+ }
+ }
+ }
+
+ return 0;
+
+err_dma_free:
+ bgmac_dma_free(bgmac);
+ return -ENOMEM;
+}
+
+static void bgmac_dma_init(struct bgmac *bgmac)
+{
+ struct bgmac_dma_ring *ring;
+ struct bgmac_dma_desc *dma_desc;
+ u32 ctl0, ctl1;
+ int i;
+
+ for (i = 0; i < BGMAC_MAX_TX_RINGS; i++) {
+ ring = &bgmac->tx_ring[i];
+
+ /* We don't implement unaligned addressing, so enable first */
+ bgmac_dma_tx_enable(bgmac, ring);
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGLO,
+ lower_32_bits(ring->dma_base));
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGHI,
+ upper_32_bits(ring->dma_base));
+
+ ring->start = 0;
+ ring->end = 0; /* Points the slot that should *not* be read */
+ }
+
+ for (i = 0; i < BGMAC_MAX_RX_RINGS; i++) {
+ ring = &bgmac->rx_ring[i];
+
+ /* We don't implement unaligned addressing, so enable first */
+ bgmac_dma_rx_enable(bgmac, ring);
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGLO,
+ lower_32_bits(ring->dma_base));
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGHI,
+ upper_32_bits(ring->dma_base));
+
+ for (i = 0, dma_desc = ring->cpu_base; i < ring->num_slots;
+ i++, dma_desc++) {
+ ctl0 = ctl1 = 0;
+
+ if (i == ring->num_slots - 1)
+ ctl0 |= BGMAC_DESC_CTL0_EOT;
+ ctl1 |= BGMAC_RX_BUF_SIZE & BGMAC_DESC_CTL1_LEN;
+ /* Is there any BGMAC device that requires extension? */
+ /* ctl1 |= (addrext << B43_DMA64_DCTL1_ADDREXT_SHIFT) &
+ * B43_DMA64_DCTL1_ADDREXT_MASK;
+ */
+
+ dma_desc->addr_low = cpu_to_le32(lower_32_bits(ring->slots[i].dma_addr));
+ dma_desc->addr_high = cpu_to_le32(upper_32_bits(ring->slots[i].dma_addr));
+ dma_desc->ctl0 = cpu_to_le32(ctl0);
+ dma_desc->ctl1 = cpu_to_le32(ctl1);
+ }
+
+ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_INDEX,
+ ring->num_slots * sizeof(struct bgmac_dma_desc));
+
+ ring->start = 0;
+ ring->end = 0;
+ }
+}
+
+/**************************************************
+ * PHY ops
+ **************************************************/
+
+u16 bgmac_phy_read(struct bgmac *bgmac, u8 phyaddr, u8 reg)
+{
+ struct bcma_device *core;
+ u16 phy_access_addr;
+ u16 phy_ctl_addr;
+ u32 tmp;
+
+ BUILD_BUG_ON(BGMAC_PA_DATA_MASK != BCMA_GMAC_CMN_PA_DATA_MASK);
+ BUILD_BUG_ON(BGMAC_PA_ADDR_MASK != BCMA_GMAC_CMN_PA_ADDR_MASK);
+ BUILD_BUG_ON(BGMAC_PA_ADDR_SHIFT != BCMA_GMAC_CMN_PA_ADDR_SHIFT);
+ BUILD_BUG_ON(BGMAC_PA_REG_MASK != BCMA_GMAC_CMN_PA_REG_MASK);
+ BUILD_BUG_ON(BGMAC_PA_REG_SHIFT != BCMA_GMAC_CMN_PA_REG_SHIFT);
+ BUILD_BUG_ON(BGMAC_PA_WRITE != BCMA_GMAC_CMN_PA_WRITE);
+ BUILD_BUG_ON(BGMAC_PA_START != BCMA_GMAC_CMN_PA_START);
+ BUILD_BUG_ON(BGMAC_PC_EPA_MASK != BCMA_GMAC_CMN_PC_EPA_MASK);
+ BUILD_BUG_ON(BGMAC_PC_MCT_MASK != BCMA_GMAC_CMN_PC_MCT_MASK);
+ BUILD_BUG_ON(BGMAC_PC_MCT_SHIFT != BCMA_GMAC_CMN_PC_MCT_SHIFT);
+ BUILD_BUG_ON(BGMAC_PC_MTE != BCMA_GMAC_CMN_PC_MTE);
+
+ if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT) {
+ core = bgmac->core->bus->drv_gmac_cmn.core;
+ phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS;
+ phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL;
+ } else {
+ core = bgmac->core;
+ phy_access_addr = BGMAC_PHY_ACCESS;
+ phy_ctl_addr = BGMAC_PHY_CNTL;
+ }
+
+ tmp = bcma_read32(core, phy_ctl_addr);
+ tmp &= ~BGMAC_PC_EPA_MASK;
+ tmp |= phyaddr;
+ bcma_write32(core, phy_ctl_addr, tmp);
+
+ tmp = BGMAC_PA_START;
+ tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT;
+ tmp |= reg << BGMAC_PA_REG_SHIFT;
+ bcma_write32(core, phy_access_addr, tmp);
+
+ if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000)) {
+ bgmac_err(bgmac, "Reading PHY %d register 0x%X failed\n",
+ phyaddr, reg);
+ return 0xffff;
+ }
+
+ return bcma_read32(core, phy_access_addr) & BGMAC_PA_DATA_MASK;
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphywr */
+void bgmac_phy_write(struct bgmac *bgmac, u8 phyaddr, u8 reg, u16 value)
+{
+ struct bcma_device *core;
+ u16 phy_access_addr;
+ u16 phy_ctl_addr;
+ u32 tmp;
+
+ if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT) {
+ core = bgmac->core->bus->drv_gmac_cmn.core;
+ phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS;
+ phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL;
+ } else {
+ core = bgmac->core;
+ phy_access_addr = BGMAC_PHY_ACCESS;
+ phy_ctl_addr = BGMAC_PHY_CNTL;
+ }
+
+ tmp = bcma_read32(core, phy_ctl_addr);
+ tmp &= ~BGMAC_PC_EPA_MASK;
+ tmp |= phyaddr;
+ bcma_write32(core, phy_ctl_addr, tmp);
+
+ bgmac_write(bgmac, BGMAC_INT_STATUS, BGMAC_IS_MDIO);
+ if (bgmac_read(bgmac, BGMAC_INT_STATUS) & BGMAC_IS_MDIO)
+ bgmac_warn(bgmac, "Error setting MDIO int\n");
+
+ tmp = BGMAC_PA_START;
+ tmp |= BGMAC_PA_WRITE;
+ tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT;
+ tmp |= reg << BGMAC_PA_REG_SHIFT;
+ tmp |= value;
+ bcma_write32(core, phy_access_addr, tmp);
+
+ if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000))
+ bgmac_err(bgmac, "Writing to PHY %d register 0x%X failed\n",
+ phyaddr, reg);
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyforce */
+static void bgmac_phy_force(struct bgmac *bgmac)
+{
+ u16 ctl;
+ u16 mask = ~(BGMAC_PHY_CTL_SPEED | BGMAC_PHY_CTL_SPEED_MSB |
+ BGMAC_PHY_CTL_ANENAB | BGMAC_PHY_CTL_DUPLEX);
+
+ if (bgmac->phyaddr == BGMAC_PHY_NOREGS)
+ return;
+
+ if (bgmac->autoneg)
+ return;
+
+ ctl = bgmac_phy_read(bgmac, bgmac->phyaddr, BGMAC_PHY_CTL);
+ ctl &= mask;
+ if (bgmac->full_duplex)
+ ctl |= BGMAC_PHY_CTL_DUPLEX;
+ if (bgmac->speed == BGMAC_SPEED_100)
+ ctl |= BGMAC_PHY_CTL_SPEED_100;
+ else if (bgmac->speed == BGMAC_SPEED_1000)
+ ctl |= BGMAC_PHY_CTL_SPEED_1000;
+ bgmac_phy_write(bgmac, bgmac->phyaddr, BGMAC_PHY_CTL, ctl);
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyadvertise */
+static void bgmac_phy_advertise(struct bgmac *bgmac)
+{
+ u16 adv;
+
+ if (bgmac->phyaddr == BGMAC_PHY_NOREGS)
+ return;
+
+ if (!bgmac->autoneg)
+ return;
+
+ /* Adv selected 10/100 speeds */
+ adv = bgmac_phy_read(bgmac, bgmac->phyaddr, BGMAC_PHY_ADV);
+ adv &= ~(BGMAC_PHY_ADV_10HALF | BGMAC_PHY_ADV_10FULL |
+ BGMAC_PHY_ADV_100HALF | BGMAC_PHY_ADV_100FULL);
+ if (!bgmac->full_duplex && bgmac->speed & BGMAC_SPEED_10)
+ adv |= BGMAC_PHY_ADV_10HALF;
+ if (!bgmac->full_duplex && bgmac->speed & BGMAC_SPEED_100)
+ adv |= BGMAC_PHY_ADV_100HALF;
+ if (bgmac->full_duplex && bgmac->speed & BGMAC_SPEED_10)
+ adv |= BGMAC_PHY_ADV_10FULL;
+ if (bgmac->full_duplex && bgmac->speed & BGMAC_SPEED_100)
+ adv |= BGMAC_PHY_ADV_100FULL;
+ bgmac_phy_write(bgmac, bgmac->phyaddr, BGMAC_PHY_ADV, adv);
+
+ /* Adv selected 1000 speeds */
+ adv = bgmac_phy_read(bgmac, bgmac->phyaddr, BGMAC_PHY_ADV2);
+ adv &= ~(BGMAC_PHY_ADV2_1000HALF | BGMAC_PHY_ADV2_1000FULL);
+ if (!bgmac->full_duplex && bgmac->speed & BGMAC_SPEED_1000)
+ adv |= BGMAC_PHY_ADV2_1000HALF;
+ if (bgmac->full_duplex && bgmac->speed & BGMAC_SPEED_1000)
+ adv |= BGMAC_PHY_ADV2_1000FULL;
+ bgmac_phy_write(bgmac, bgmac->phyaddr, BGMAC_PHY_ADV2, adv);
+
+ /* Restart */
+ bgmac_phy_write(bgmac, bgmac->phyaddr, BGMAC_PHY_CTL,
+ bgmac_phy_read(bgmac, bgmac->phyaddr, BGMAC_PHY_CTL) |
+ BGMAC_PHY_CTL_RESTART);
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyinit */
+static void bgmac_phy_init(struct bgmac *bgmac)
+{
+ struct bcma_chipinfo *ci = &bgmac->core->bus->chipinfo;
+ struct bcma_drv_cc *cc = &bgmac->core->bus->drv_cc;
+ u8 i;
+
+ if (ci->id == BCMA_CHIP_ID_BCM5356) {
+ for (i = 0; i < 5; i++) {
+ bgmac_phy_write(bgmac, i, 0x1f, 0x008b);
+ bgmac_phy_write(bgmac, i, 0x15, 0x0100);
+ bgmac_phy_write(bgmac, i, 0x1f, 0x000f);
+ bgmac_phy_write(bgmac, i, 0x12, 0x2aaa);
+ bgmac_phy_write(bgmac, i, 0x1f, 0x000b);
+ }
+ }
+ if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg != 10) ||
+ (ci->id == BCMA_CHIP_ID_BCM4749 && ci->pkg != 10) ||
+ (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg != 9)) {
+ bcma_chipco_chipctl_maskset(cc, 2, ~0xc0000000, 0);
+ bcma_chipco_chipctl_maskset(cc, 4, ~0x80000000, 0);
+ for (i = 0; i < 5; i++) {
+ bgmac_phy_write(bgmac, i, 0x1f, 0x000f);
+ bgmac_phy_write(bgmac, i, 0x16, 0x5284);
+ bgmac_phy_write(bgmac, i, 0x1f, 0x000b);
+ bgmac_phy_write(bgmac, i, 0x17, 0x0010);
+ bgmac_phy_write(bgmac, i, 0x1f, 0x000f);
+ bgmac_phy_write(bgmac, i, 0x16, 0x5296);
+ bgmac_phy_write(bgmac, i, 0x17, 0x1073);
+ bgmac_phy_write(bgmac, i, 0x17, 0x9073);
+ bgmac_phy_write(bgmac, i, 0x16, 0x52b6);
+ bgmac_phy_write(bgmac, i, 0x17, 0x9273);
+ bgmac_phy_write(bgmac, i, 0x1f, 0x000b);
+ }
+ }
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyreset */
+static void bgmac_phy_reset(struct bgmac *bgmac)
+{
+ if (bgmac->phyaddr == BGMAC_PHY_NOREGS)
+ return;
+
+ bgmac_phy_write(bgmac, bgmac->phyaddr, BGMAC_PHY_CTL,
+ BGMAC_PHY_CTL_RESET);
+ udelay(100);
+ if (bgmac_phy_read(bgmac, bgmac->phyaddr, BGMAC_PHY_CTL) &
+ BGMAC_PHY_CTL_RESET)
+ bgmac_err(bgmac, "PHY reset failed\n");
+ bgmac_phy_init(bgmac);
+}
+
+/**************************************************
+ * Chip ops
+ **************************************************/
+
+/* TODO: can we just drop @force? Can we don't reset MAC at all if there is
+ * nothing to change? Try if after stabilizng driver.
+ */
+static void bgmac_cmdcfg_maskset(struct bgmac *bgmac, u32 mask, u32 set,
+ bool force)
+{
+ u32 cmdcfg = bgmac_read(bgmac, BGMAC_CMDCFG);
+ u32 new_val = (cmdcfg & mask) | set;
+
+ bgmac_set(bgmac, BGMAC_CMDCFG, BGMAC_CMDCFG_SR);
+ udelay(2);
+
+ if (new_val != cmdcfg || force)
+ bgmac_write(bgmac, BGMAC_CMDCFG, new_val);
+
+ bgmac_mask(bgmac, BGMAC_CMDCFG, ~BGMAC_CMDCFG_SR);
+ udelay(2);
+}
+
+#if 0 /* We don't use that regs yet */
+static void bgmac_chip_stats_update(struct bgmac *bgmac)
+{
+ int i;
+
+ if (bgmac->core->id.id != BCMA_CORE_4706_MAC_GBIT) {
+ for (i = 0; i < BGMAC_NUM_MIB_TX_REGS; i++)
+ bgmac->mib_tx_regs[i] =
+ bgmac_read(bgmac,
+ BGMAC_TX_GOOD_OCTETS + (i * 4));
+ for (i = 0; i < BGMAC_NUM_MIB_RX_REGS; i++)
+ bgmac->mib_rx_regs[i] =
+ bgmac_read(bgmac,
+ BGMAC_RX_GOOD_OCTETS + (i * 4));
+ }
+
+ /* TODO: what else? how to handle BCM4706? Specs are needed */
+}
+#endif
+
+static void bgmac_clear_mib(struct bgmac *bgmac)
+{
+ int i;
+
+ if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT)
+ return;
+
+ bgmac_set(bgmac, BGMAC_DEV_CTL, BGMAC_DC_MROR);
+ for (i = 0; i < BGMAC_NUM_MIB_TX_REGS; i++)
+ bgmac_read(bgmac, BGMAC_TX_GOOD_OCTETS + (i * 4));
+ for (i = 0; i < BGMAC_NUM_MIB_RX_REGS; i++)
+ bgmac_read(bgmac, BGMAC_RX_GOOD_OCTETS + (i * 4));
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_speed */
+static void bgmac_speed(struct bgmac *bgmac, int speed)
+{
+ u32 mask = ~(BGMAC_CMDCFG_ES_MASK | BGMAC_CMDCFG_HD);
+ u32 set = 0;
+
+ if (speed & BGMAC_SPEED_10)
+ set |= BGMAC_CMDCFG_ES_10;
+ if (speed & BGMAC_SPEED_100)
+ set |= BGMAC_CMDCFG_ES_100;
+ if (speed & BGMAC_SPEED_1000)
+ set |= BGMAC_CMDCFG_ES_1000;
+ if (!bgmac->full_duplex)
+ set |= BGMAC_CMDCFG_HD;
+ bgmac_cmdcfg_maskset(bgmac, mask, set, true);
+}
+
+static void bgmac_miiconfig(struct bgmac *bgmac)
+{
+ u8 imode = (bgmac_read(bgmac, BGMAC_DEV_STATUS) & BGMAC_DS_MM_MASK) >>
+ BGMAC_DS_MM_SHIFT;
+ if (imode == 0 || imode == 1) {
+ if (bgmac->autoneg)
+ bgmac_speed(bgmac, BGMAC_SPEED_100);
+ else
+ bgmac_speed(bgmac, bgmac->speed);
+ }
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipreset */
+static void bgmac_chip_reset(struct bgmac *bgmac)
+{
+ struct bcma_device *core = bgmac->core;
+ struct bcma_bus *bus = core->bus;
+ struct bcma_chipinfo *ci = &bus->chipinfo;
+ u32 flags = 0;
+ u32 iost;
+ int i;
+
+ if (bcma_core_is_enabled(core)) {
+ if (!bgmac->stats_grabbed) {
+ /* bgmac_chip_stats_update(bgmac); */
+ bgmac->stats_grabbed = true;
+ }
+
+ for (i = 0; i < BGMAC_MAX_TX_RINGS; i++)
+ bgmac_dma_tx_reset(bgmac, &bgmac->tx_ring[i]);
+
+ bgmac_cmdcfg_maskset(bgmac, ~0, BGMAC_CMDCFG_ML, false);
+ udelay(1);
+
+ for (i = 0; i < BGMAC_MAX_RX_RINGS; i++)
+ bgmac_dma_rx_reset(bgmac, &bgmac->rx_ring[i]);
+
+ /* TODO: Clear software multicast filter list */
+ }
+
+ iost = bcma_aread32(core, BCMA_IOST);
+ if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg == 10) ||
+ (ci->id == BCMA_CHIP_ID_BCM4749 && ci->pkg == 10) ||
+ (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg == 9))
+ iost &= ~BGMAC_BCMA_IOST_ATTACHED;
+
+ if (iost & BGMAC_BCMA_IOST_ATTACHED) {
+ flags = BGMAC_BCMA_IOCTL_SW_CLKEN;
+ if (!bgmac->has_robosw)
+ flags |= BGMAC_BCMA_IOCTL_SW_RESET;
+ }
+
+ bcma_core_enable(core, flags);
+
+ if (core->id.rev > 2) {
+ bgmac_set(bgmac, BCMA_CLKCTLST, 1 << 8);
+ bgmac_wait_value(bgmac->core, BCMA_CLKCTLST, 1 << 24, 1 << 24,
+ 1000);
+ }
+
+ if (ci->id == BCMA_CHIP_ID_BCM5357 || ci->id == BCMA_CHIP_ID_BCM4749 ||
+ ci->id == BCMA_CHIP_ID_BCM53572) {
+ struct bcma_drv_cc *cc = &bgmac->core->bus->drv_cc;
+ u8 et_swtype = 0;
+ u8 sw_type = BGMAC_CHIPCTL_1_SW_TYPE_EPHY |
+ BGMAC_CHIPCTL_1_IF_TYPE_RMII;
+ char buf[2];
+
+ if (nvram_getenv("et_swtype", buf, 1) > 0) {
+ if (kstrtou8(buf, 0, &et_swtype))
+ bgmac_err(bgmac, "Failed to parse et_swtype (%s)\n",
+ buf);
+ et_swtype &= 0x0f;
+ et_swtype <<= 4;
+ sw_type = et_swtype;
+ } else if (ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg == 9) {
+ sw_type = BGMAC_CHIPCTL_1_SW_TYPE_EPHYRMII;
+ } else if (0) {
+ /* TODO */
+ }
+ bcma_chipco_chipctl_maskset(cc, 1,
+ ~(BGMAC_CHIPCTL_1_IF_TYPE_MASK |
+ BGMAC_CHIPCTL_1_SW_TYPE_MASK),
+ sw_type);
+ }
+
+ if (iost & BGMAC_BCMA_IOST_ATTACHED && !bgmac->has_robosw)
+ bcma_awrite32(core, BCMA_IOCTL,
+ bcma_aread32(core, BCMA_IOCTL) &
+ ~BGMAC_BCMA_IOCTL_SW_RESET);
+
+ /* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_reset
+ * Specs don't say about using BGMAC_CMDCFG_SR, but in this routine
+ * BGMAC_CMDCFG is read _after_ putting chip in a reset. So it has to
+ * be keps until taking MAC out of the reset.
+ */
+ bgmac_cmdcfg_maskset(bgmac,
+ ~(BGMAC_CMDCFG_TE |
+ BGMAC_CMDCFG_RE |
+ BGMAC_CMDCFG_RPI |
+ BGMAC_CMDCFG_TAI |
+ BGMAC_CMDCFG_HD |
+ BGMAC_CMDCFG_ML |
+ BGMAC_CMDCFG_CFE |
+ BGMAC_CMDCFG_RL |
+ BGMAC_CMDCFG_RED |
+ BGMAC_CMDCFG_PE |
+ BGMAC_CMDCFG_TPI |
+ BGMAC_CMDCFG_PAD_EN |
+ BGMAC_CMDCFG_PF),
+ BGMAC_CMDCFG_PROM |
+ BGMAC_CMDCFG_NLC |
+ BGMAC_CMDCFG_CFE |
+ BGMAC_CMDCFG_SR,
+ false);
+
+ bgmac_clear_mib(bgmac);
+ if (core->id.id == BCMA_CORE_4706_MAC_GBIT)
+ bcma_maskset32(bgmac->cmn, BCMA_GMAC_CMN_PHY_CTL, ~0,
+ BCMA_GMAC_CMN_PC_MTE);
+ else
+ bgmac_set(bgmac, BGMAC_PHY_CNTL, BGMAC_PC_MTE);
+ bgmac_miiconfig(bgmac);
+ bgmac_phy_init(bgmac);
+
+ bgmac->int_status = 0;
+}
+
+static void bgmac_chip_intrs_on(struct bgmac *bgmac)
+{
+ bgmac_write(bgmac, BGMAC_INT_MASK, bgmac->int_mask);
+}
+
+static void bgmac_chip_intrs_off(struct bgmac *bgmac)
+{
+ bgmac_write(bgmac, BGMAC_INT_MASK, 0);
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_enable */
+static void bgmac_enable(struct bgmac *bgmac)
+{
+ struct bcma_chipinfo *ci = &bgmac->core->bus->chipinfo;
+ u32 cmdcfg;
+ u32 mode;
+ u32 rxq_ctl;
+ u32 fl_ctl;
+ u16 bp_clk;
+ u8 mdp;
+
+ cmdcfg = bgmac_read(bgmac, BGMAC_CMDCFG);
+ bgmac_cmdcfg_maskset(bgmac, ~(BGMAC_CMDCFG_TE | BGMAC_CMDCFG_RE),
+ BGMAC_CMDCFG_SR, true);
+ udelay(2);
+ cmdcfg |= BGMAC_CMDCFG_TE | BGMAC_CMDCFG_RE;
+ bgmac_write(bgmac, BGMAC_CMDCFG, cmdcfg);
+
+ mode = (bgmac_read(bgmac, BGMAC_DEV_STATUS) & BGMAC_DS_MM_MASK) >>
+ BGMAC_DS_MM_SHIFT;
+ if (ci->id != BCMA_CHIP_ID_BCM47162 || mode != 0)
+ bgmac_set(bgmac, BCMA_CLKCTLST, BCMA_CLKCTLST_FORCEHT);
+ if (ci->id == BCMA_CHIP_ID_BCM47162 && mode == 2)
+ bcma_chipco_chipctl_maskset(&bgmac->core->bus->drv_cc, 1, ~0,
+ BGMAC_CHIPCTL_1_RXC_DLL_BYPASS);
+
+ switch (ci->id) {
+ case BCMA_CHIP_ID_BCM5357:
+ case BCMA_CHIP_ID_BCM4749:
+ case BCMA_CHIP_ID_BCM53572:
+ case BCMA_CHIP_ID_BCM4716:
+ case BCMA_CHIP_ID_BCM47162:
+ fl_ctl = 0x03cb04cb;
+ if (ci->id == BCMA_CHIP_ID_BCM5357 ||
+ ci->id == BCMA_CHIP_ID_BCM4749 ||
+ ci->id == BCMA_CHIP_ID_BCM53572)
+ fl_ctl = 0x2300e1;
+ bgmac_write(bgmac, BGMAC_FLOW_CTL_THRESH, fl_ctl);
+ bgmac_write(bgmac, BGMAC_PAUSE_CTL, 0x27fff);
+ break;
+ }
+
+ rxq_ctl = bgmac_read(bgmac, BGMAC_RXQ_CTL);
+ rxq_ctl &= ~BGMAC_RXQ_CTL_MDP_MASK;
+ bp_clk = bcma_pmu_get_bus_clock(&bgmac->core->bus->drv_cc) / 1000000;
+ mdp = (bp_clk * 128 / 1000) - 3;
+ rxq_ctl |= (mdp << BGMAC_RXQ_CTL_MDP_SHIFT);
+ bgmac_write(bgmac, BGMAC_RXQ_CTL, rxq_ctl);
+}
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipinit */
+static void bgmac_chip_init(struct bgmac *bgmac, bool full_init)
+{
+ struct bgmac_dma_ring *ring;
+ u8 *mac = bgmac->net_dev->dev_addr;
+ u32 tmp;
+ int i;
+
+ /* 1 interrupt per received frame */
+ bgmac_write(bgmac, BGMAC_INT_RECV_LAZY, 1 << BGMAC_IRL_FC_SHIFT);
+
+ /* Enable 802.3x tx flow control (honor received PAUSE frames) */
+ bgmac_cmdcfg_maskset(bgmac, ~BGMAC_CMDCFG_RPI, 0, true);
+
+ if (bgmac->net_dev->flags & IFF_PROMISC)
+ bgmac_cmdcfg_maskset(bgmac, ~0, BGMAC_CMDCFG_PROM, false);
+ else
+ bgmac_cmdcfg_maskset(bgmac, ~BGMAC_CMDCFG_PROM, 0, false);
+
+ /* Set MAC addr */
+ tmp = (mac[0] << 24) | (mac[1] << 16) | (mac[2] << 8) | mac[3];
+ bgmac_write(bgmac, BGMAC_MACADDR_HIGH, tmp);
+ tmp = (mac[4] << 8) | mac[5];
+ bgmac_write(bgmac, BGMAC_MACADDR_LOW, tmp);
+
+ if (bgmac->loopback)
+ bgmac_cmdcfg_maskset(bgmac, ~0, BGMAC_CMDCFG_ML, true);
+ else
+ bgmac_cmdcfg_maskset(bgmac, ~BGMAC_CMDCFG_ML, 0, true);
+
+ bgmac_write(bgmac, BGMAC_RXMAX_LENGTH, 32 + ETHER_MAX_LEN);
+
+ if (!bgmac->autoneg) {
+ bgmac_speed(bgmac, bgmac->speed);
+ bgmac_phy_force(bgmac);
+ } else if (bgmac->speed) { /* if there is anything to adv */
+ bgmac_phy_advertise(bgmac);
+ }
+
+ if (full_init) {
+ bgmac_dma_init(bgmac);
+ if (1) /* FIXME: is there any case we don't want IRQs? */
+ bgmac_chip_intrs_on(bgmac);
+ } else {
+ for (i = 0; i < BGMAC_MAX_RX_RINGS; i++) {
+ ring = &bgmac->rx_ring[i];
+ bgmac_dma_rx_enable(bgmac, ring);
+ }
+ }
+
+ bgmac_enable(bgmac);
+}
+
+static irqreturn_t bgmac_interrupt(int irq, void *dev_id)
+{
+ struct bgmac *bgmac = netdev_priv(dev_id);
+
+ u32 int_status = bgmac_read(bgmac, BGMAC_INT_STATUS);
+ int_status &= bgmac->int_mask;
+
+ if (!int_status)
+ return IRQ_NONE;
+
+ /* Ack */
+ bgmac_write(bgmac, BGMAC_INT_STATUS, int_status);
+
+ /* Disable new interrupts until handling existing ones */
+ bgmac_chip_intrs_off(bgmac);
+
+ bgmac->int_status = int_status;
+
+ napi_schedule(&bgmac->napi);
+
+ return IRQ_HANDLED;
+}
+
+static int bgmac_poll(struct napi_struct *napi, int weight)
+{
+ struct bgmac *bgmac = container_of(napi, struct bgmac, napi);
+ struct bgmac_dma_ring *ring;
+ int handled = 0;
+
+ if (bgmac->int_status & BGMAC_IS_TX0) {
+ ring = &bgmac->tx_ring[0];
+ bgmac_dma_tx_free(bgmac, ring);
+ bgmac->int_status &= ~BGMAC_IS_TX0;
+ }
+
+ if (bgmac->int_status & BGMAC_IS_RX) {
+ ring = &bgmac->rx_ring[0];
+ handled += bgmac_dma_rx_read(bgmac, ring, weight);
+ bgmac->int_status &= ~BGMAC_IS_RX;
+ }
+
+ if (bgmac->int_status) {
+ bgmac_err(bgmac, "Unknown IRQs: 0x%08X\n", bgmac->int_status);
+ bgmac->int_status = 0;
+ }
+
+ if (handled < weight)
+ napi_complete(napi);
+
+ bgmac_chip_intrs_on(bgmac);
+
+ return handled;
+}
+
+/**************************************************
+ * net_device_ops
+ **************************************************/
+
+static int bgmac_open(struct net_device *net_dev)
+{
+ struct bgmac *bgmac = netdev_priv(net_dev);
+ int err = 0;
+
+ bgmac_chip_reset(bgmac);
+ /* Specs say about reclaiming rings here, but we do that in DMA init */
+ bgmac_chip_init(bgmac, true);
+
+ err = request_irq(bgmac->core->irq, bgmac_interrupt, IRQF_SHARED,
+ KBUILD_MODNAME, net_dev);
+ if (err < 0) {
+ bgmac_err(bgmac, "IRQ request error: %d!\n", err);
+ goto err_out;
+ }
+ napi_enable(&bgmac->napi);
+
+ netif_carrier_on(net_dev);
+
+err_out:
+ return err;
+}
+
+static int bgmac_stop(struct net_device *net_dev)
+{
+ struct bgmac *bgmac = netdev_priv(net_dev);
+
+ netif_carrier_off(net_dev);
+
+ napi_disable(&bgmac->napi);
+ bgmac_chip_intrs_off(bgmac);
+ free_irq(bgmac->core->irq, net_dev);
+
+ bgmac_chip_reset(bgmac);
+
+ return 0;
+}
+
+static netdev_tx_t bgmac_start_xmit(struct sk_buff *skb,
+ struct net_device *net_dev)
+{
+ struct bgmac *bgmac = netdev_priv(net_dev);
+ struct bgmac_dma_ring *ring;
+
+ /* No QOS support yet */
+ ring = &bgmac->tx_ring[0];
+ return bgmac_dma_tx_add(bgmac, ring, skb);
+}
+
+static int bgmac_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd)
+{
+ struct bgmac *bgmac = netdev_priv(net_dev);
+ struct mii_ioctl_data *data = if_mii(ifr);
+
+ switch (cmd) {
+ case SIOCGMIIPHY:
+ data->phy_id = bgmac->phyaddr;
+ /* fallthru */
+ case SIOCGMIIREG:
+ if (!netif_running(net_dev))
+ return -EAGAIN;
+ data->val_out = bgmac_phy_read(bgmac, bgmac->phyaddr,
+ data->reg_num & 0x1f);
+ return 0;
+ case SIOCSMIIREG:
+ if (!netif_running(net_dev))
+ return -EAGAIN;
+ bgmac_phy_write(bgmac, bgmac->phyaddr, data->reg_num & 0x1f,
+ data->val_in);
+ return 0;
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
+static const struct net_device_ops bgmac_netdev_ops = {
+ .ndo_open = bgmac_open,
+ .ndo_stop = bgmac_stop,
+ .ndo_start_xmit = bgmac_start_xmit,
+ .ndo_set_mac_address = eth_mac_addr, /* generic, sets dev_addr */
+ .ndo_do_ioctl = bgmac_ioctl,
+};
+
+/**************************************************
+ * ethtool_ops
+ **************************************************/
+
+static int bgmac_get_settings(struct net_device *net_dev,
+ struct ethtool_cmd *cmd)
+{
+ struct bgmac *bgmac = netdev_priv(net_dev);
+
+ cmd->supported = SUPPORTED_10baseT_Half |
+ SUPPORTED_10baseT_Full |
+ SUPPORTED_100baseT_Half |
+ SUPPORTED_100baseT_Full |
+ SUPPORTED_1000baseT_Half |
+ SUPPORTED_1000baseT_Full |
+ SUPPORTED_Autoneg;
+
+ if (bgmac->autoneg) {
+ WARN_ON(cmd->advertising);
+ if (bgmac->full_duplex) {
+ if (bgmac->speed & BGMAC_SPEED_10)
+ cmd->advertising |= ADVERTISED_10baseT_Full;
+ if (bgmac->speed & BGMAC_SPEED_100)
+ cmd->advertising |= ADVERTISED_100baseT_Full;
+ if (bgmac->speed & BGMAC_SPEED_1000)
+ cmd->advertising |= ADVERTISED_1000baseT_Full;
+ } else {
+ if (bgmac->speed & BGMAC_SPEED_10)
+ cmd->advertising |= ADVERTISED_10baseT_Half;
+ if (bgmac->speed & BGMAC_SPEED_100)
+ cmd->advertising |= ADVERTISED_100baseT_Half;
+ if (bgmac->speed & BGMAC_SPEED_1000)
+ cmd->advertising |= ADVERTISED_1000baseT_Half;
+ }
+ } else {
+ switch (bgmac->speed) {
+ case BGMAC_SPEED_10:
+ ethtool_cmd_speed_set(cmd, SPEED_10);
+ break;
+ case BGMAC_SPEED_100:
+ ethtool_cmd_speed_set(cmd, SPEED_100);
+ break;
+ case BGMAC_SPEED_1000:
+ ethtool_cmd_speed_set(cmd, SPEED_1000);
+ break;
+ }
+ }
+
+ cmd->duplex = bgmac->full_duplex ? DUPLEX_FULL : DUPLEX_HALF;
+
+ cmd->autoneg = bgmac->autoneg;
+
+ return 0;
+}
+
+#if 0
+static int bgmac_set_settings(struct net_device *net_dev,
+ struct ethtool_cmd *cmd)
+{
+ struct bgmac *bgmac = netdev_priv(net_dev);
+
+ return -1;
+}
+#endif
+
+static void bgmac_get_drvinfo(struct net_device *net_dev,
+ struct ethtool_drvinfo *info)
+{
+ strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+ strlcpy(info->bus_info, "BCMA", sizeof(info->bus_info));
+}
+
+static const struct ethtool_ops bgmac_ethtool_ops = {
+ .get_settings = bgmac_get_settings,
+ .get_drvinfo = bgmac_get_drvinfo,
+};
+
+/**************************************************
+ * BCMA bus ops
+ **************************************************/
+
+/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipattach */
+static int bgmac_probe(struct bcma_device *core)
+{
+ struct net_device *net_dev;
+ struct bgmac *bgmac;
+ struct ssb_sprom *sprom = &core->bus->sprom;
+ u8 *mac = core->core_unit ? sprom->et1mac : sprom->et0mac;
+ int err;
+
+ /* We don't support 2nd, 3rd, ... units, SPROM has to be adjusted */
+ if (core->core_unit > 1) {
+ pr_err("Unsupported core_unit %d\n", core->core_unit);
+ return -ENOTSUPP;
+ }
+
+ /* Allocation and references */
+ net_dev = alloc_etherdev(sizeof(*bgmac));
+ if (!net_dev)
+ return -ENOMEM;
+ net_dev->netdev_ops = &bgmac_netdev_ops;
+ net_dev->irq = core->irq;
+ SET_ETHTOOL_OPS(net_dev, &bgmac_ethtool_ops);
+ bgmac = netdev_priv(net_dev);
+ bgmac->net_dev = net_dev;
+ bgmac->core = core;
+ bcma_set_drvdata(core, bgmac);
+
+ /* Defaults */
+ bgmac->autoneg = true;
+ bgmac->full_duplex = true;
+ bgmac->speed = BGMAC_SPEED_10 | BGMAC_SPEED_100 | BGMAC_SPEED_1000;
+ memcpy(bgmac->net_dev->dev_addr, mac, ETH_ALEN);
+
+ /* On BCM4706 we need common core to access PHY */
+ if (core->id.id == BCMA_CORE_4706_MAC_GBIT &&
+ !core->bus->drv_gmac_cmn.core) {
+ bgmac_err(bgmac, "GMAC CMN core not found (required for BCM4706)\n");
+ err = -ENODEV;
+ goto err_netdev_free;
+ }
+ bgmac->cmn = core->bus->drv_gmac_cmn.core;
+
+ bgmac->phyaddr = core->core_unit ? sprom->et1phyaddr :
+ sprom->et0phyaddr;
+ bgmac->phyaddr &= BGMAC_PHY_MASK;
+ if (bgmac->phyaddr == BGMAC_PHY_MASK) {
+ bgmac_err(bgmac, "No PHY found\n");
+ err = -ENODEV;
+ goto err_netdev_free;
+ }
+ bgmac_info(bgmac, "Found PHY addr: %d%s\n", bgmac->phyaddr,
+ bgmac->phyaddr == BGMAC_PHY_NOREGS ? " (NOREGS)" : "");
+
+ if (core->bus->hosttype == BCMA_HOSTTYPE_PCI) {
+ bgmac_err(bgmac, "PCI setup not implemented\n");
+ err = -ENOTSUPP;
+ goto err_netdev_free;
+ }
+
+ bgmac_chip_reset(bgmac);
+
+ err = bgmac_dma_alloc(bgmac);
+ if (err) {
+ bgmac_err(bgmac, "Unable to alloc memory for DMA\n");
+ goto err_netdev_free;
+ }
+
+ bgmac->int_mask = BGMAC_IS_ERRMASK | BGMAC_IS_RX | BGMAC_IS_TX_MASK;
+ if (nvram_getenv("et0_no_txint", NULL, 0) == 0)
+ bgmac->int_mask &= ~BGMAC_IS_TX_MASK;
+
+ /* TODO: reset the external phy. Specs are needed */
+ bgmac_phy_reset(bgmac);
+
+ bgmac->has_robosw = !!(core->bus->sprom.boardflags_lo &
+ BGMAC_BFL_ENETROBO);
+ if (bgmac->has_robosw)
+ bgmac_warn(bgmac, "Support for Roboswitch not implemented\n");
+
+ if (core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETADM)
+ bgmac_warn(bgmac, "Support for ADMtek ethernet switch not implemented\n");
+
+ err = register_netdev(bgmac->net_dev);
+ if (err) {
+ bgmac_err(bgmac, "Cannot register net device\n");
+ err = -ENOTSUPP;
+ goto err_dma_free;
+ }
+
+ netif_carrier_off(net_dev);
+
+ netif_napi_add(net_dev, &bgmac->napi, bgmac_poll, BGMAC_WEIGHT);
+
+ return 0;
+
+err_dma_free:
+ bgmac_dma_free(bgmac);
+
+err_netdev_free:
+ bcma_set_drvdata(core, NULL);
+ free_netdev(net_dev);
+
+ return err;
+}
+
+static void bgmac_remove(struct bcma_device *core)
+{
+ struct bgmac *bgmac = bcma_get_drvdata(core);
+
+ netif_napi_del(&bgmac->napi);
+ unregister_netdev(bgmac->net_dev);
+ bgmac_dma_free(bgmac);
+ bcma_set_drvdata(core, NULL);
+ free_netdev(bgmac->net_dev);
+}
+
+static struct bcma_driver bgmac_bcma_driver = {
+ .name = KBUILD_MODNAME,
+ .id_table = bgmac_bcma_tbl,
+ .probe = bgmac_probe,
+ .remove = bgmac_remove,
+};
+
+static int __init bgmac_init(void)
+{
+ int err;
+
+ err = bcma_driver_register(&bgmac_bcma_driver);
+ if (err)
+ return err;
+ pr_info("Broadcom 47xx GBit MAC driver loaded\n");
+
+ return 0;
+}
+
+static void __exit bgmac_exit(void)
+{
+ bcma_driver_unregister(&bgmac_bcma_driver);
+}
+
+module_init(bgmac_init)
+module_exit(bgmac_exit)
+
+MODULE_AUTHOR("Rafał Miłecki");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h
new file mode 100644
index 0000000..b22a9b7
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bgmac.h
@@ -0,0 +1,447 @@
+#ifndef _BGMAC_H
+#define _BGMAC_H
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#define bgmac_err(bgmac, fmt, ...) \
+ dev_err(&(bgmac)->core->dev, fmt, ##__VA_ARGS__)
+#define bgmac_warn(bgmac, fmt, ...) \
+ dev_warn(&(bgmac)->core->dev, fmt, ##__VA_ARGS__)
+#define bgmac_info(bgmac, fmt, ...) \
+ dev_info(&(bgmac)->core->dev, fmt, ##__VA_ARGS__)
+#define bgmac_dbg(bgmac, fmt, ...) \
+ dev_dbg(&(bgmac)->core->dev, fmt, ##__VA_ARGS__)
+
+#include <linux/bcma/bcma.h>
+#include <linux/netdevice.h>
+
+#define BGMAC_DEV_CTL 0x000
+#define BGMAC_DC_TSM 0x00000002
+#define BGMAC_DC_CFCO 0x00000004
+#define BGMAC_DC_RLSS 0x00000008
+#define BGMAC_DC_MROR 0x00000010
+#define BGMAC_DC_FCM_MASK 0x00000060
+#define BGMAC_DC_FCM_SHIFT 5
+#define BGMAC_DC_NAE 0x00000080
+#define BGMAC_DC_TF 0x00000100
+#define BGMAC_DC_RDS_MASK 0x00030000
+#define BGMAC_DC_RDS_SHIFT 16
+#define BGMAC_DC_TDS_MASK 0x000c0000
+#define BGMAC_DC_TDS_SHIFT 18
+#define BGMAC_DEV_STATUS 0x004 /* Configuration of the interface */
+#define BGMAC_DS_RBF 0x00000001
+#define BGMAC_DS_RDF 0x00000002
+#define BGMAC_DS_RIF 0x00000004
+#define BGMAC_DS_TBF 0x00000008
+#define BGMAC_DS_TDF 0x00000010
+#define BGMAC_DS_TIF 0x00000020
+#define BGMAC_DS_PO 0x00000040
+#define BGMAC_DS_MM_MASK 0x00000300 /* Mode of the interface */
+#define BGMAC_DS_MM_SHIFT 8
+#define BGMAC_BIST_STATUS 0x00c
+#define BGMAC_INT_STATUS 0x020 /* Interrupt status */
+#define BGMAC_IS_MRO 0x00000001
+#define BGMAC_IS_MTO 0x00000002
+#define BGMAC_IS_TFD 0x00000004
+#define BGMAC_IS_LS 0x00000008
+#define BGMAC_IS_MDIO 0x00000010
+#define BGMAC_IS_MR 0x00000020
+#define BGMAC_IS_MT 0x00000040
+#define BGMAC_IS_TO 0x00000080
+#define BGMAC_IS_DESC_ERR 0x00000400 /* Descriptor error */
+#define BGMAC_IS_DATA_ERR 0x00000800 /* Data error */
+#define BGMAC_IS_DESC_PROT_ERR 0x00001000 /* Descriptor protocol error */
+#define BGMAC_IS_RX_DESC_UNDERF 0x00002000 /* Receive descriptor underflow */
+#define BGMAC_IS_RX_F_OVERF 0x00004000 /* Receive FIFO overflow */
+#define BGMAC_IS_TX_F_UNDERF 0x00008000 /* Transmit FIFO underflow */
+#define BGMAC_IS_RX 0x00010000 /* Interrupt for RX queue 0 */
+#define BGMAC_IS_TX0 0x01000000 /* Interrupt for TX queue 0 */
+#define BGMAC_IS_TX1 0x02000000 /* Interrupt for TX queue 1 */
+#define BGMAC_IS_TX2 0x04000000 /* Interrupt for TX queue 2 */
+#define BGMAC_IS_TX3 0x08000000 /* Interrupt for TX queue 3 */
+#define BGMAC_IS_TX_MASK 0x0f000000
+#define BGMAC_IS_INTMASK 0x0f01fcff
+#define BGMAC_IS_ERRMASK 0x0000fc00
+#define BGMAC_INT_MASK 0x024 /* Interrupt mask */
+#define BGMAC_GP_TIMER 0x028
+#define BGMAC_INT_RECV_LAZY 0x100
+#define BGMAC_IRL_TO_MASK 0x00ffffff
+#define BGMAC_IRL_FC_MASK 0xff000000
+#define BGMAC_IRL_FC_SHIFT 24 /* Shift the number of interrupts triggered per received frame */
+#define BGMAC_FLOW_CTL_THRESH 0x104 /* Flow control thresholds */
+#define BGMAC_WRRTHRESH 0x108
+#define BGMAC_GMAC_IDLE_CNT_THRESH 0x10c
+#define BGMAC_PHY_ACCESS 0x180 /* PHY access address */
+#define BGMAC_PA_DATA_MASK 0x0000ffff
+#define BGMAC_PA_ADDR_MASK 0x001f0000
+#define BGMAC_PA_ADDR_SHIFT 16
+#define BGMAC_PA_REG_MASK 0x1f000000
+#define BGMAC_PA_REG_SHIFT 24
+#define BGMAC_PA_WRITE 0x20000000
+#define BGMAC_PA_START 0x40000000
+#define BGMAC_PHY_CNTL 0x188 /* PHY control address */
+#define BGMAC_PC_EPA_MASK 0x0000001f
+#define BGMAC_PC_MCT_MASK 0x007f0000
+#define BGMAC_PC_MCT_SHIFT 16
+#define BGMAC_PC_MTE 0x00800000
+#define BGMAC_TXQ_CTL 0x18c
+#define BGMAC_TXQ_CTL_DBT_MASK 0x00000fff
+#define BGMAC_TXQ_CTL_DBT_SHIFT 0
+#define BGMAC_RXQ_CTL 0x190
+#define BGMAC_RXQ_CTL_DBT_MASK 0x00000fff
+#define BGMAC_RXQ_CTL_DBT_SHIFT 0
+#define BGMAC_RXQ_CTL_PTE 0x00001000
+#define BGMAC_RXQ_CTL_MDP_MASK 0x3f000000
+#define BGMAC_RXQ_CTL_MDP_SHIFT 24
+#define BGMAC_GPIO_SELECT 0x194
+#define BGMAC_GPIO_OUTPUT_EN 0x198
+/* For 0x1e0 see BCMA_CLKCTLST */
+#define BGMAC_HW_WAR 0x1e4
+#define BGMAC_PWR_CTL 0x1e8
+#define BGMAC_DMA_BASE0 0x200 /* Tx and Rx controller */
+#define BGMAC_DMA_BASE1 0x240 /* Tx controller only */
+#define BGMAC_DMA_BASE2 0x280 /* Tx controller only */
+#define BGMAC_DMA_BASE3 0x2C0 /* Tx controller only */
+#define BGMAC_TX_GOOD_OCTETS 0x300
+#define BGMAC_TX_GOOD_OCTETS_HIGH 0x304
+#define BGMAC_TX_GOOD_PKTS 0x308
+#define BGMAC_TX_OCTETS 0x30c
+#define BGMAC_TX_OCTETS_HIGH 0x310
+#define BGMAC_TX_PKTS 0x314
+#define BGMAC_TX_BROADCAST_PKTS 0x318
+#define BGMAC_TX_MULTICAST_PKTS 0x31c
+#define BGMAC_TX_LEN_64 0x320
+#define BGMAC_TX_LEN_65_TO_127 0x324
+#define BGMAC_TX_LEN_128_TO_255 0x328
+#define BGMAC_TX_LEN_256_TO_511 0x32c
+#define BGMAC_TX_LEN_512_TO_1023 0x330
+#define BGMAC_TX_LEN_1024_TO_1522 0x334
+#define BGMAC_TX_LEN_1523_TO_2047 0x338
+#define BGMAC_TX_LEN_2048_TO_4095 0x33c
+#define BGMAC_TX_LEN_4095_TO_8191 0x340
+#define BGMAC_TX_LEN_8192_TO_MAX 0x344
+#define BGMAC_TX_JABBER_PKTS 0x348 /* Error */
+#define BGMAC_TX_OVERSIZE_PKTS 0x34c /* Error */
+#define BGMAC_TX_FRAGMENT_PKTS 0x350
+#define BGMAC_TX_UNDERRUNS 0x354 /* Error */
+#define BGMAC_TX_TOTAL_COLS 0x358
+#define BGMAC_TX_SINGLE_COLS 0x35c
+#define BGMAC_TX_MULTIPLE_COLS 0x360
+#define BGMAC_TX_EXCESSIVE_COLS 0x364 /* Error */
+#define BGMAC_TX_LATE_COLS 0x368 /* Error */
+#define BGMAC_TX_DEFERED 0x36c
+#define BGMAC_TX_CARRIER_LOST 0x370
+#define BGMAC_TX_PAUSE_PKTS 0x374
+#define BGMAC_TX_UNI_PKTS 0x378
+#define BGMAC_TX_Q0_PKTS 0x37c
+#define BGMAC_TX_Q0_OCTETS 0x380
+#define BGMAC_TX_Q0_OCTETS_HIGH 0x384
+#define BGMAC_TX_Q1_PKTS 0x388
+#define BGMAC_TX_Q1_OCTETS 0x38c
+#define BGMAC_TX_Q1_OCTETS_HIGH 0x390
+#define BGMAC_TX_Q2_PKTS 0x394
+#define BGMAC_TX_Q2_OCTETS 0x398
+#define BGMAC_TX_Q2_OCTETS_HIGH 0x39c
+#define BGMAC_TX_Q3_PKTS 0x3a0
+#define BGMAC_TX_Q3_OCTETS 0x3a4
+#define BGMAC_TX_Q3_OCTETS_HIGH 0x3a8
+#define BGMAC_RX_GOOD_OCTETS 0x3b0
+#define BGMAC_RX_GOOD_OCTETS_HIGH 0x3b4
+#define BGMAC_RX_GOOD_PKTS 0x3b8
+#define BGMAC_RX_OCTETS 0x3bc
+#define BGMAC_RX_OCTETS_HIGH 0x3c0
+#define BGMAC_RX_PKTS 0x3c4
+#define BGMAC_RX_BROADCAST_PKTS 0x3c8
+#define BGMAC_RX_MULTICAST_PKTS 0x3cc
+#define BGMAC_RX_LEN_64 0x3d0
+#define BGMAC_RX_LEN_65_TO_127 0x3d4
+#define BGMAC_RX_LEN_128_TO_255 0x3d8
+#define BGMAC_RX_LEN_256_TO_511 0x3dc
+#define BGMAC_RX_LEN_512_TO_1023 0x3e0
+#define BGMAC_RX_LEN_1024_TO_1522 0x3e4
+#define BGMAC_RX_LEN_1523_TO_2047 0x3e8
+#define BGMAC_RX_LEN_2048_TO_4095 0x3ec
+#define BGMAC_RX_LEN_4095_TO_8191 0x3f0
+#define BGMAC_RX_LEN_8192_TO_MAX 0x3f4
+#define BGMAC_RX_JABBER_PKTS 0x3f8 /* Error */
+#define BGMAC_RX_OVERSIZE_PKTS 0x3fc /* Error */
+#define BGMAC_RX_FRAGMENT_PKTS 0x400
+#define BGMAC_RX_MISSED_PKTS 0x404 /* Error */
+#define BGMAC_RX_CRC_ALIGN_ERRS 0x408 /* Error */
+#define BGMAC_RX_UNDERSIZE 0x40c /* Error */
+#define BGMAC_RX_CRC_ERRS 0x410 /* Error */
+#define BGMAC_RX_ALIGN_ERRS 0x414 /* Error */
+#define BGMAC_RX_SYMBOL_ERRS 0x418 /* Error */
+#define BGMAC_RX_PAUSE_PKTS 0x41c
+#define BGMAC_RX_NONPAUSE_PKTS 0x420
+#define BGMAC_RX_SACHANGES 0x424
+#define BGMAC_RX_UNI_PKTS 0x428
+#define BGMAC_UNIMAC_VERSION 0x800
+#define BGMAC_HDBKP_CTL 0x804
+#define BGMAC_CMDCFG 0x808 /* Configuration */
+#define BGMAC_CMDCFG_TE 0x00000001 /* Set to activate TX */
+#define BGMAC_CMDCFG_RE 0x00000002 /* Set to activate RX */
+#define BGMAC_CMDCFG_ES_MASK 0x0000000c /* Ethernet speed see gmac_speed */
+#define BGMAC_CMDCFG_ES_10 0x00000000
+#define BGMAC_CMDCFG_ES_100 0x00000004
+#define BGMAC_CMDCFG_ES_1000 0x00000008
+#define BGMAC_CMDCFG_PROM 0x00000010 /* Set to activate promiscuous mode */
+#define BGMAC_CMDCFG_PAD_EN 0x00000020
+#define BGMAC_CMDCFG_CF 0x00000040
+#define BGMAC_CMDCFG_PF 0x00000080
+#define BGMAC_CMDCFG_RPI 0x00000100 /* Unset to enable 802.3x tx flow control */
+#define BGMAC_CMDCFG_TAI 0x00000200
+#define BGMAC_CMDCFG_HD 0x00000400 /* Set if in half duplex mode */
+#define BGMAC_CMDCFG_HD_SHIFT 10
+#define BGMAC_CMDCFG_SR 0x00000800 /* Set to reset mode */
+#define BGMAC_CMDCFG_ML 0x00008000 /* Set to activate mac loopback mode */
+#define BGMAC_CMDCFG_AE 0x00400000
+#define BGMAC_CMDCFG_CFE 0x00800000
+#define BGMAC_CMDCFG_NLC 0x01000000
+#define BGMAC_CMDCFG_RL 0x02000000
+#define BGMAC_CMDCFG_RED 0x04000000
+#define BGMAC_CMDCFG_PE 0x08000000
+#define BGMAC_CMDCFG_TPI 0x10000000
+#define BGMAC_CMDCFG_AT 0x20000000
+#define BGMAC_MACADDR_HIGH 0x80c /* High 4 octets of own mac address */
+#define BGMAC_MACADDR_LOW 0x810 /* Low 2 octets of own mac address */
+#define BGMAC_RXMAX_LENGTH 0x814 /* Max receive frame length with vlan tag */
+#define BGMAC_PAUSEQUANTA 0x818
+#define BGMAC_MAC_MODE 0x844
+#define BGMAC_OUTERTAG 0x848
+#define BGMAC_INNERTAG 0x84c
+#define BGMAC_TXIPG 0x85c
+#define BGMAC_PAUSE_CTL 0xb30
+#define BGMAC_TX_FLUSH 0xb34
+#define BGMAC_RX_STATUS 0xb38
+#define BGMAC_TX_STATUS 0xb3c
+
+#define BGMAC_PHY_CTL 0x00
+#define BGMAC_PHY_CTL_SPEED_MSB 0x0040
+#define BGMAC_PHY_CTL_DUPLEX 0x0100 /* duplex mode */
+#define BGMAC_PHY_CTL_RESTART 0x0200 /* restart autonegotiation */
+#define BGMAC_PHY_CTL_ANENAB 0x1000 /* enable autonegotiation */
+#define BGMAC_PHY_CTL_SPEED 0x2000
+#define BGMAC_PHY_CTL_LOOP 0x4000 /* loopback */
+#define BGMAC_PHY_CTL_RESET 0x8000 /* reset */
+/* Helpers */
+#define BGMAC_PHY_CTL_SPEED_10 0
+#define BGMAC_PHY_CTL_SPEED_100 BGMAC_PHY_CTL_SPEED
+#define BGMAC_PHY_CTL_SPEED_1000 BGMAC_PHY_CTL_SPEED_MSB
+#define BGMAC_PHY_ADV 0x04
+#define BGMAC_PHY_ADV_10HALF 0x0020 /* advertise 10MBits/s half duplex */
+#define BGMAC_PHY_ADV_10FULL 0x0040 /* advertise 10MBits/s full duplex */
+#define BGMAC_PHY_ADV_100HALF 0x0080 /* advertise 100MBits/s half duplex */
+#define BGMAC_PHY_ADV_100FULL 0x0100 /* advertise 100MBits/s full duplex */
+#define BGMAC_PHY_ADV2 0x09
+#define BGMAC_PHY_ADV2_1000HALF 0x0100 /* advertise 1000MBits/s half duplex */
+#define BGMAC_PHY_ADV2_1000FULL 0x0200 /* advertise 1000MBits/s full duplex */
+
+/* BCMA GMAC core specific IO Control (BCMA_IOCTL) flags */
+#define BGMAC_BCMA_IOCTL_SW_CLKEN 0x00000004 /* PHY Clock Enable */
+#define BGMAC_BCMA_IOCTL_SW_RESET 0x00000008 /* PHY Reset */
+
+/* BCMA GMAC core specific IO status (BCMA_IOST) flags */
+#define BGMAC_BCMA_IOST_ATTACHED 0x00000800
+
+#define BGMAC_NUM_MIB_TX_REGS \
+ (((BGMAC_TX_Q3_OCTETS_HIGH - BGMAC_TX_GOOD_OCTETS) / 4) + 1)
+#define BGMAC_NUM_MIB_RX_REGS \
+ (((BGMAC_RX_UNI_PKTS - BGMAC_RX_GOOD_OCTETS) / 4) + 1)
+
+#define BGMAC_DMA_TX_CTL 0x00
+#define BGMAC_DMA_TX_ENABLE 0x00000001
+#define BGMAC_DMA_TX_SUSPEND 0x00000002
+#define BGMAC_DMA_TX_LOOPBACK 0x00000004
+#define BGMAC_DMA_TX_FLUSH 0x00000010
+#define BGMAC_DMA_TX_PARITY_DISABLE 0x00000800
+#define BGMAC_DMA_TX_ADDREXT_MASK 0x00030000
+#define BGMAC_DMA_TX_ADDREXT_SHIFT 16
+#define BGMAC_DMA_TX_INDEX 0x04
+#define BGMAC_DMA_TX_RINGLO 0x08
+#define BGMAC_DMA_TX_RINGHI 0x0C
+#define BGMAC_DMA_TX_STATUS 0x10
+#define BGMAC_DMA_TX_STATDPTR 0x00001FFF
+#define BGMAC_DMA_TX_STAT 0xF0000000
+#define BGMAC_DMA_TX_STAT_DISABLED 0x00000000
+#define BGMAC_DMA_TX_STAT_ACTIVE 0x10000000
+#define BGMAC_DMA_TX_STAT_IDLEWAIT 0x20000000
+#define BGMAC_DMA_TX_STAT_STOPPED 0x30000000
+#define BGMAC_DMA_TX_STAT_SUSP 0x40000000
+#define BGMAC_DMA_TX_ERROR 0x14
+#define BGMAC_DMA_TX_ERRDPTR 0x0001FFFF
+#define BGMAC_DMA_TX_ERR 0xF0000000
+#define BGMAC_DMA_TX_ERR_NOERR 0x00000000
+#define BGMAC_DMA_TX_ERR_PROT 0x10000000
+#define BGMAC_DMA_TX_ERR_UNDERRUN 0x20000000
+#define BGMAC_DMA_TX_ERR_TRANSFER 0x30000000
+#define BGMAC_DMA_TX_ERR_DESCREAD 0x40000000
+#define BGMAC_DMA_TX_ERR_CORE 0x50000000
+#define BGMAC_DMA_RX_CTL 0x20
+#define BGMAC_DMA_RX_ENABLE 0x00000001
+#define BGMAC_DMA_RX_FRAME_OFFSET_MASK 0x000000FE
+#define BGMAC_DMA_RX_FRAME_OFFSET_SHIFT 1
+#define BGMAC_DMA_RX_DIRECT_FIFO 0x00000100
+#define BGMAC_DMA_RX_OVERFLOW_CONT 0x00000400
+#define BGMAC_DMA_RX_PARITY_DISABLE 0x00000800
+#define BGMAC_DMA_RX_ADDREXT_MASK 0x00030000
+#define BGMAC_DMA_RX_ADDREXT_SHIFT 16
+#define BGMAC_DMA_RX_INDEX 0x24
+#define BGMAC_DMA_RX_RINGLO 0x28
+#define BGMAC_DMA_RX_RINGHI 0x2C
+#define BGMAC_DMA_RX_STATUS 0x30
+#define BGMAC_DMA_RX_STATDPTR 0x00001FFF
+#define BGMAC_DMA_RX_STAT 0xF0000000
+#define BGMAC_DMA_RX_STAT_DISABLED 0x00000000
+#define BGMAC_DMA_RX_STAT_ACTIVE 0x10000000
+#define BGMAC_DMA_RX_STAT_IDLEWAIT 0x20000000
+#define BGMAC_DMA_RX_STAT_STOPPED 0x30000000
+#define BGMAC_DMA_RX_STAT_SUSP 0x40000000
+#define BGMAC_DMA_RX_ERROR 0x34
+#define BGMAC_DMA_RX_ERRDPTR 0x0001FFFF
+#define BGMAC_DMA_RX_ERR 0xF0000000
+#define BGMAC_DMA_RX_ERR_NOERR 0x00000000
+#define BGMAC_DMA_RX_ERR_PROT 0x10000000
+#define BGMAC_DMA_RX_ERR_UNDERRUN 0x20000000
+#define BGMAC_DMA_RX_ERR_TRANSFER 0x30000000
+#define BGMAC_DMA_RX_ERR_DESCREAD 0x40000000
+#define BGMAC_DMA_RX_ERR_CORE 0x50000000
+
+#define BGMAC_DESC_CTL0_EOT 0x10000000 /* End of ring */
+#define BGMAC_DESC_CTL0_IOC 0x20000000 /* IRQ on complete */
+#define BGMAC_DESC_CTL0_SOF 0x40000000 /* Start of frame */
+#define BGMAC_DESC_CTL0_EOF 0x80000000 /* End of frame */
+#define BGMAC_DESC_CTL1_LEN 0x00001FFF
+
+#define BGMAC_PHY_NOREGS 0x1E
+#define BGMAC_PHY_MASK 0x1F
+
+#define BGMAC_MAX_TX_RINGS 4
+#define BGMAC_MAX_RX_RINGS 1
+
+#define BGMAC_TX_RING_SLOTS 128
+#define BGMAC_RX_RING_SLOTS 512 - 1 /* Why -1? Well, Broadcom does that... */
+
+#define BGMAC_RX_HEADER_LEN 28 /* Last 24 bytes are unused. Well... */
+#define BGMAC_RX_FRAME_OFFSET 30 /* There are 2 unused bytes between header and real data */
+#define BGMAC_RX_MAX_FRAME_SIZE 1536 /* Copied from b44/tg3 */
+#define BGMAC_RX_BUF_SIZE (BGMAC_RX_FRAME_OFFSET + BGMAC_RX_MAX_FRAME_SIZE)
+
+#define BGMAC_BFL_ENETROBO 0x0010 /* has ephy roboswitch spi */
+#define BGMAC_BFL_ENETADM 0x0080 /* has ADMtek switch */
+#define BGMAC_BFL_ENETVLAN 0x0100 /* can do vlan */
+
+#define BGMAC_CHIPCTL_1_IF_TYPE_MASK 0x00000030
+#define BGMAC_CHIPCTL_1_IF_TYPE_RMII 0x00000000
+#define BGMAC_CHIPCTL_1_IF_TYPE_MI 0x00000010
+#define BGMAC_CHIPCTL_1_IF_TYPE_RGMII 0x00000020
+#define BGMAC_CHIPCTL_1_SW_TYPE_MASK 0x000000C0
+#define BGMAC_CHIPCTL_1_SW_TYPE_EPHY 0x00000000
+#define BGMAC_CHIPCTL_1_SW_TYPE_EPHYMII 0x00000040
+#define BGMAC_CHIPCTL_1_SW_TYPE_EPHYRMII 0x00000080
+#define BGMAC_CHIPCTL_1_SW_TYPE_RGMI 0x000000C0
+#define BGMAC_CHIPCTL_1_RXC_DLL_BYPASS 0x00010000
+
+#define BGMAC_SPEED_10 0x0001
+#define BGMAC_SPEED_100 0x0002
+#define BGMAC_SPEED_1000 0x0004
+
+#define BGMAC_WEIGHT 64
+
+#define ETHER_MAX_LEN 1518
+
+struct bgmac_slot_info {
+ struct sk_buff *skb;
+ dma_addr_t dma_addr;
+};
+
+struct bgmac_dma_desc {
+ __le32 ctl0;
+ __le32 ctl1;
+ __le32 addr_low;
+ __le32 addr_high;
+} __packed;
+
+enum bgmac_dma_ring_type {
+ BGMAC_DMA_RING_TX,
+ BGMAC_DMA_RING_RX,
+};
+
+struct bgmac_dma_ring {
+ u16 num_slots;
+ u16 start;
+ u16 end;
+
+ u16 mmio_base;
+ struct bgmac_dma_desc *cpu_base;
+ dma_addr_t dma_base;
+
+ struct bgmac_slot_info slots[BGMAC_RX_RING_SLOTS];
+};
+
+struct bgmac_rx_header {
+ __le16 len;
+ __le16 flags;
+ __le16 pad[12];
+};
+
+struct bgmac {
+ struct bcma_device *core;
+ struct bcma_device *cmn; /* Reference to CMN core for BCM4706 */
+ struct net_device *net_dev;
+ struct napi_struct napi;
+
+ /* DMA */
+ struct bgmac_dma_ring tx_ring[BGMAC_MAX_TX_RINGS];
+ struct bgmac_dma_ring rx_ring[BGMAC_MAX_RX_RINGS];
+
+ /* Stats */
+ bool stats_grabbed;
+ u32 mib_tx_regs[BGMAC_NUM_MIB_TX_REGS];
+ u32 mib_rx_regs[BGMAC_NUM_MIB_RX_REGS];
+
+ /* Int */
+ u32 int_mask;
+ u32 int_status;
+
+ /* Speed-related */
+ int speed;
+ bool autoneg;
+ bool full_duplex;
+
+ u8 phyaddr;
+ bool has_robosw;
+
+ bool loopback;
+};
+
+static inline u32 bgmac_read(struct bgmac *bgmac, u16 offset)
+{
+ return bcma_read32(bgmac->core, offset);
+}
+
+static inline void bgmac_write(struct bgmac *bgmac, u16 offset, u32 value)
+{
+ bcma_write32(bgmac->core, offset, value);
+}
+
+static inline void bgmac_maskset(struct bgmac *bgmac, u16 offset, u32 mask,
+ u32 set)
+{
+ bgmac_write(bgmac, offset, (bgmac_read(bgmac, offset) & mask) | set);
+}
+
+static inline void bgmac_mask(struct bgmac *bgmac, u16 offset, u32 mask)
+{
+ bgmac_maskset(bgmac, offset, mask, 0);
+}
+
+static inline void bgmac_set(struct bgmac *bgmac, u16 offset, u32 set)
+{
+ bgmac_maskset(bgmac, offset, ~0, set);
+}
+
+u16 bgmac_phy_read(struct bgmac *bgmac, u8 phyaddr, u8 reg);
+void bgmac_phy_write(struct bgmac *bgmac, u8 phyaddr, u8 reg, u16 value);
+
+#endif /* _BGMAC_H */
diff --git a/include/linux/bcma/bcma_driver_chipcommon.h b/include/linux/bcma/bcma_driver_chipcommon.h
index 9a0e3fa..ee332fa 100644
--- a/include/linux/bcma/bcma_driver_chipcommon.h
+++ b/include/linux/bcma/bcma_driver_chipcommon.h
@@ -634,4 +634,6 @@ extern void bcma_chipco_regctl_maskset(struct bcma_drv_cc *cc,
u32 offset, u32 mask, u32 set);
extern void bcma_pmu_spuravoid_pllupdate(struct bcma_drv_cc *cc, int spuravoid);
+extern u32 bcma_pmu_get_bus_clock(struct bcma_drv_cc *cc);
+
#endif /* LINUX_BCMA_DRIVER_CC_H_ */
--
1.7.7
^ permalink raw reply related
* Re: [PATCH net-next] team: use strlcpy with ethtool_drvinfo fields
From: Jiri Pirko @ 2013-01-05 13:17 UTC (permalink / raw)
To: Flavio Leitner; +Cc: netdev, Ben Hutchings
In-Reply-To: <1357390390-16549-1-git-send-email-fbl@redhat.com>
Sat, Jan 05, 2013 at 01:53:10PM CET, fbl@redhat.com wrote:
>The fields must be null-terminated.
>
>Signed-off-by: Flavio Leitner <fbl@redhat.com>
>---
> drivers/net/team/team.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
>diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
>index 78c7d87..e27ac5c 100644
>--- a/drivers/net/team/team.c
>+++ b/drivers/net/team/team.c
>@@ -1747,8 +1747,8 @@ static const struct net_device_ops team_netdev_ops = {
> static void team_ethtool_get_drvinfo(struct net_device *dev,
> struct ethtool_drvinfo *drvinfo)
> {
>- strncpy(drvinfo->driver, DRV_NAME, 32);
>- strncpy(drvinfo->version, UTS_RELEASE, 32);
>+ strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
>+ strlcpy(drvinfo->version, UTS_RELEASE, sizeof(drvinfo->version));
> }
>
> static const struct ethtool_ops team_ethtool_ops = {
>--
>1.8.0.1
>
Acked-by: Jiri Pirko <jiri@resnulli.us>
^ permalink raw reply
* [PATCH] drivers/net/wireless/ipw2x00: use strlcpy instead of strncpy
From: Chen Gang @ 2013-01-05 13:41 UTC (permalink / raw)
To: stas.yakovlev, linville; +Cc: linux-wireless, netdev
The fields must be null-terminated, or IPW_DEBUG_ASSOC will cause issue.
Signed-off-by: Chen Gang <gang.chen@asianux.com>
---
drivers/net/wireless/ipw2x00/ipw2200.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/wireless/ipw2x00/ipw2200.c b/drivers/net/wireless/ipw2x00/ipw2200.c
index 844f201..c85261b 100644
--- a/drivers/net/wireless/ipw2x00/ipw2200.c
+++ b/drivers/net/wireless/ipw2x00/ipw2200.c
@@ -5558,7 +5558,7 @@ static int ipw_find_adhoc_network(struct ipw_priv *priv,
min(network->ssid_len, priv->essid_len)))) {
char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
- strncpy(escaped,
+ strlcpy(escaped,
print_ssid(ssid, network->ssid,
network->ssid_len),
sizeof(escaped));
@@ -5771,7 +5771,7 @@ static int ipw_best_network(struct ipw_priv *priv,
memcmp(network->ssid, priv->essid,
min(network->ssid_len, priv->essid_len)))) {
char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
- strncpy(escaped,
+ strlcpy(escaped,
print_ssid(ssid, network->ssid,
network->ssid_len),
sizeof(escaped));
@@ -5788,7 +5788,7 @@ static int ipw_best_network(struct ipw_priv *priv,
* testing everything else. */
if (match->network && match->network->stats.rssi > network->stats.rssi) {
char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
- strncpy(escaped,
+ strlcpy(escaped,
print_ssid(ssid, network->ssid, network->ssid_len),
sizeof(escaped));
IPW_DEBUG_ASSOC("Network '%s (%pM)' excluded because "
--
1.7.10.4
^ permalink raw reply related
* Re: [PATCH net-next] team: use strlcpy with ethtool_drvinfo fields
From: Ben Hutchings @ 2013-01-05 14:11 UTC (permalink / raw)
To: Flavio Leitner; +Cc: netdev, Jiri Pirko
In-Reply-To: <1357390390-16549-1-git-send-email-fbl@redhat.com>
On Sat, 2013-01-05 at 10:53 -0200, Flavio Leitner wrote:
> The fields must be null-terminated.
>
> Signed-off-by: Flavio Leitner <fbl@redhat.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
> ---
> drivers/net/team/team.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
> index 78c7d87..e27ac5c 100644
> --- a/drivers/net/team/team.c
> +++ b/drivers/net/team/team.c
> @@ -1747,8 +1747,8 @@ static const struct net_device_ops team_netdev_ops = {
> static void team_ethtool_get_drvinfo(struct net_device *dev,
> struct ethtool_drvinfo *drvinfo)
> {
> - strncpy(drvinfo->driver, DRV_NAME, 32);
> - strncpy(drvinfo->version, UTS_RELEASE, 32);
> + strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
> + strlcpy(drvinfo->version, UTS_RELEASE, sizeof(drvinfo->version));
> }
>
> static const struct ethtool_ops team_ethtool_ops = {
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH] drivers/net/wireless/ipw2x00: use strlcpy instead of strncpy
From: Joe Perches @ 2013-01-05 14:42 UTC (permalink / raw)
To: Chen Gang; +Cc: stas.yakovlev, linville, linux-wireless, netdev
In-Reply-To: <50E82D7D.4090107@asianux.com>
On Sat, 2013-01-05 at 21:41 +0800, Chen Gang wrote:
> The fields must be null-terminated, or IPW_DEBUG_ASSOC will cause issue.
[]
> diff --git a/drivers/net/wireless/ipw2x00/ipw2200.c b/drivers/net/wireless/ipw2x00/ipw2200.c
[]
> @@ -5558,7 +5558,7 @@ static int ipw_find_adhoc_network(struct ipw_priv *priv,
> min(network->ssid_len, priv->essid_len)))) {
> char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
>
> - strncpy(escaped,
> + strlcpy(escaped,
> print_ssid(ssid, network->ssid,
> network->ssid_len),
> sizeof(escaped));
> @@ -5771,7 +5771,7 @@ static int ipw_best_network(struct ipw_priv *priv,
> memcmp(network->ssid, priv->essid,
> min(network->ssid_len, priv->essid_len)))) {
> char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
> - strncpy(escaped,
> + strlcpy(escaped,
> print_ssid(ssid, network->ssid,
> network->ssid_len),
> sizeof(escaped));
> @@ -5788,7 +5788,7 @@ static int ipw_best_network(struct ipw_priv *priv,
> * testing everything else. */
> if (match->network && match->network->stats.rssi > network->stats.rssi) {
> char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
> - strncpy(escaped,
> + strlcpy(escaped,
> print_ssid(ssid, network->ssid, network->ssid_len),
> sizeof(escaped));
> IPW_DEBUG_ASSOC("Network '%s (%pM)' excluded because "
This happens because escaped is declared the wrong size.
It'd be better to change
char escaped[IW_ESSID_MAX_SIZE * 2 + 1];
to
DECLARE_SSID_BUF(escaped);
and use
print_ssid(escaped, network->ssid, network->ssid_len)
in the debug.
^ permalink raw reply
* [PATCH] tcp: make sysctl_tcp_ecn namespace aware
From: Hannes Frederic Sowa @ 2013-01-05 15:01 UTC (permalink / raw)
To: netdev; +Cc: eric.dumazet, davem, shemminger
As per suggestion from Eric Dumazet this patch makes tcp_ecn sysctl aware.
The reason behind this patch is to ease the testing of ecn problems on
the internet and allows applications to tune their own use of ecn.
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
---
include/net/netns/ipv4.h | 2 ++
include/net/tcp.h | 7 ++++---
net/ipv4/syncookies.c | 6 ++++--
net/ipv4/sysctl_net_ipv4.c | 16 +++++++++-------
net/ipv4/tcp_input.c | 2 --
net/ipv4/tcp_ipv4.c | 4 +++-
net/ipv4/tcp_output.c | 1 +
net/ipv6/syncookies.c | 3 ++-
net/ipv6/tcp_ipv6.c | 7 +++++--
9 files changed, 30 insertions(+), 18 deletions(-)
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 2ae2b83..9b78862 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -61,6 +61,8 @@ struct netns_ipv4 {
int sysctl_icmp_ratemask;
int sysctl_icmp_errors_use_inbound_ifaddr;
+ int sysctl_tcp_ecn;
+
kgid_t sysctl_ping_group_range[2];
long sysctl_tcp_mem[3];
diff --git a/include/net/tcp.h b/include/net/tcp.h
index aed42c7..a808c59 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -266,7 +266,6 @@ extern int sysctl_tcp_abort_on_overflow;
extern int sysctl_tcp_max_orphans;
extern int sysctl_tcp_fack;
extern int sysctl_tcp_reordering;
-extern int sysctl_tcp_ecn;
extern int sysctl_tcp_dsack;
extern int sysctl_tcp_wmem[3];
extern int sysctl_tcp_rmem[3];
@@ -504,7 +503,8 @@ static inline __u32 cookie_v4_init_sequence(struct sock *sk,
#endif
extern __u32 cookie_init_timestamp(struct request_sock *req);
-extern bool cookie_check_timestamp(struct tcp_options_received *opt, bool *);
+extern bool cookie_check_timestamp(struct tcp_options_received *opt,
+ int sysctl_tcp_ecn, bool *ecn_ok);
/* From net/ipv6/syncookies.c */
extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb);
@@ -728,7 +728,8 @@ struct tcp_skb_cb {
* notifications, we disable TCP ECN negociation.
*/
static inline void
-TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb)
+TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb,
+ int sysctl_tcp_ecn)
{
const struct tcphdr *th = tcp_hdr(skb);
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index b236ef0..f51c3f5 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -232,7 +232,8 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb,
*
* return false if we decode an option that should not be.
*/
-bool cookie_check_timestamp(struct tcp_options_received *tcp_opt, bool *ecn_ok)
+bool cookie_check_timestamp(struct tcp_options_received *tcp_opt,
+ int sysctl_tcp_ecn, bool *ecn_ok)
{
/* echoed timestamp, lowest bits contain options */
u32 options = tcp_opt->rcv_tsecr & TSMASK;
@@ -279,6 +280,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
__u8 rcv_wscale;
bool ecn_ok = false;
struct flowi4 fl4;
+ int sysctl_tcp_ecn = sock_net(sk)->ipv4.sysctl_tcp_ecn;
if (!sysctl_tcp_syncookies || !th->ack || th->rst)
goto out;
@@ -295,7 +297,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
memset(&tcp_opt, 0, sizeof(tcp_opt));
tcp_parse_options(skb, &tcp_opt, &hash_location, 0, NULL);
- if (!cookie_check_timestamp(&tcp_opt, &ecn_ok))
+ if (!cookie_check_timestamp(&tcp_opt, sysctl_tcp_ecn, &ecn_ok))
goto out;
ret = NULL;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index d84400b..7716d40 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -538,13 +538,6 @@ static struct ctl_table ipv4_table[] = {
.proc_handler = proc_dointvec
},
{
- .procname = "tcp_ecn",
- .data = &sysctl_tcp_ecn,
- .maxlen = sizeof(int),
- .mode = 0644,
- .proc_handler = proc_dointvec
- },
- {
.procname = "tcp_dsack",
.data = &sysctl_tcp_dsack,
.maxlen = sizeof(int),
@@ -850,6 +843,13 @@ static struct ctl_table ipv4_net_table[] = {
.proc_handler = ipv4_ping_group_range,
},
{
+ .procname = "tcp_ecn",
+ .data = &init_net.ipv4.sysctl_tcp_ecn,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = proc_dointvec
+ },
+ {
.procname = "tcp_mem",
.maxlen = sizeof(init_net.ipv4.sysctl_tcp_mem),
.mode = 0644,
@@ -882,6 +882,8 @@ static __net_init int ipv4_sysctl_init_net(struct net *net)
&net->ipv4.sysctl_icmp_ratemask;
table[6].data =
&net->ipv4.sysctl_ping_group_range;
+ table[7].data =
+ &net->ipv4.sysctl_tcp_ecn;
/* Don't export sysctls to unprivileged users */
if (net->user_ns != &init_user_ns)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a28e4db..38e1184 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -81,8 +81,6 @@ int sysctl_tcp_sack __read_mostly = 1;
int sysctl_tcp_fack __read_mostly = 1;
int sysctl_tcp_reordering __read_mostly = TCP_FASTRETRANS_THRESH;
EXPORT_SYMBOL(sysctl_tcp_reordering);
-int sysctl_tcp_ecn __read_mostly = 2;
-EXPORT_SYMBOL(sysctl_tcp_ecn);
int sysctl_tcp_dsack __read_mostly = 1;
int sysctl_tcp_app_win __read_mostly = 31;
int sysctl_tcp_adv_win_scale __read_mostly = 1;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 54139fa..70215e7 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1568,7 +1568,8 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
goto drop_and_free;
if (!want_cookie || tmp_opt.tstamp_ok)
- TCP_ECN_create_request(req, skb);
+ TCP_ECN_create_request(req, skb,
+ sock_net(sk)->ipv4.sysctl_tcp_ecn);
if (want_cookie) {
isn = cookie_v4_init_sequence(sk, skb, &req->mss);
@@ -2888,6 +2889,7 @@ EXPORT_SYMBOL(tcp_prot);
static int __net_init tcp_sk_init(struct net *net)
{
+ net->ipv4.sysctl_tcp_ecn = 2;
return 0;
}
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 5d45159..a407f92 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -312,6 +312,7 @@ static inline void TCP_ECN_send_synack(const struct tcp_sock *tp, struct sk_buff
static inline void TCP_ECN_send_syn(struct sock *sk, struct sk_buff *skb)
{
struct tcp_sock *tp = tcp_sk(sk);
+ int sysctl_tcp_ecn = sock_net(sk)->ipv4.sysctl_tcp_ecn;
tp->ecn_flags = 0;
if (sysctl_tcp_ecn == 1) {
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 4016197..2a632ed 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -163,6 +163,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
struct dst_entry *dst;
__u8 rcv_wscale;
bool ecn_ok = false;
+ int sysctl_tcp_ecn = sock_net(sk)->ipv4.sysctl_tcp_ecn;
if (!sysctl_tcp_syncookies || !th->ack || th->rst)
goto out;
@@ -179,7 +180,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
memset(&tcp_opt, 0, sizeof(tcp_opt));
tcp_parse_options(skb, &tcp_opt, &hash_location, 0, NULL);
- if (!cookie_check_timestamp(&tcp_opt, &ecn_ok))
+ if (!cookie_check_timestamp(&tcp_opt, sysctl_tcp_ecn, &ecn_ok))
goto out;
ret = NULL;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 93825dd..932d667 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1026,8 +1026,11 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
treq = inet6_rsk(req);
treq->rmt_addr = ipv6_hdr(skb)->saddr;
treq->loc_addr = ipv6_hdr(skb)->daddr;
- if (!want_cookie || tmp_opt.tstamp_ok)
- TCP_ECN_create_request(req, skb);
+ if (!want_cookie || tmp_opt.tstamp_ok) {
+ /* we reuse the ipv4 sysctl for ecn here */
+ TCP_ECN_create_request(req, skb,
+ sock_net(sk)->ipv4.sysctl_tcp_ecn);
+ }
treq->iif = sk->sk_bound_dev_if;
--
1.7.11.7
^ permalink raw reply related
* [PATCH 2/2] ipv6: fix packet corruption when Dest/RT2 options are used
From: Romain KUNTZ @ 2013-01-05 16:19 UTC (permalink / raw)
To: netdev; +Cc: yoshfuji, davem
In-Reply-To: <0CC79564-4AF2-42F9-8D06-1BCC912A1AF7@ipflavors.com>
Commit 299b0767 (ipv6: Fix IPsec slowpath fragmentation problem)
has introduced a bug that provokes corrupted packets when Destination
Options or Routing Header Type 2 are used (such as with Mobile IPv6):
rt->rt6i_nfheader_len should be substracted to rt->dst.header_len,
and not to exthdrlen.
This patch reverts to the original and correct behavior. Successfully
tested with and without IPsec activated for MH packets.
Signed-off-by: Romain Kuntz <r.kuntz@ipflavors.com>
---
net/ipv6/ip6_output.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 5552d13..0c7c03d 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1213,10 +1213,10 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
if (dst_allfrag(rt->dst.path))
cork->flags |= IPCORK_ALLFRAG;
cork->length = 0;
- exthdrlen = (opt ? opt->opt_flen : 0) - rt->rt6i_nfheader_len;
+ exthdrlen = (opt ? opt->opt_flen : 0);
length += exthdrlen;
transhdrlen += exthdrlen;
- dst_exthdrlen = rt->dst.header_len;
+ dst_exthdrlen = rt->dst.header_len - rt->rt6i_nfheader_len;
} else {
rt = (struct rt6_info *)cork->dst;
fl6 = &inet->cork.fl.u.ip6;
--
1.7.2.5
^ permalink raw reply related
* [PATCH 1/2] ipv6: avoid blackhole and prohibited entries upon prefix purge
From: Romain KUNTZ @ 2013-01-05 16:13 UTC (permalink / raw)
To: netdev; +Cc: yoshfuji, davem
Mobile IPv6 provokes a kernel Oops since commit 64c6d08e (ipv6:
del unreachable route when an addr is deleted on lo), because
ip6_route_lookup() may also return blackhole and prohibited
entry. However, these entries have a NULL rt6i_table argument,
which provokes an Oops in __ip6_del_rt() when trying to lock
rt6i_table->tb6_lock.
Beside, when purging a prefix, blakhole and prohibited entries
should not be selected because they are not what we are looking
for.
We fix this by adding two new lookup flags (RT6_LOOKUP_F_NO_BLK_HOLE
and RT6_LOOKUP_F_NO_PROHIBIT) in order to ensure that such entries
are skipped during lookup and that the correct entry is returned.
Signed-off-by: Romain Kuntz <r.kuntz@ipflavors.com>
---
include/net/ip6_route.h | 2 ++
net/ipv6/addrconf.c | 4 +++-
net/ipv6/fib6_rules.c | 4 ++++
3 files changed, 9 insertions(+), 1 deletions(-)
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 27d8318..3c93743 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -30,6 +30,8 @@ struct route_info {
#define RT6_LOOKUP_F_SRCPREF_TMP 0x00000008
#define RT6_LOOKUP_F_SRCPREF_PUBLIC 0x00000010
#define RT6_LOOKUP_F_SRCPREF_COA 0x00000020
+#define RT6_LOOKUP_F_NO_BLK_HOLE 0x00000040
+#define RT6_LOOKUP_F_NO_PROHIBIT 0x00000080
/*
* rt6_srcprefs2flags() and rt6_flags2srcprefs() translate
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 408cac4a..1891e23 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -948,7 +948,9 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
fl6.flowi6_oif = ifp->idev->dev->ifindex;
fl6.daddr = prefix;
rt = (struct rt6_info *)ip6_route_lookup(net, &fl6,
- RT6_LOOKUP_F_IFACE);
+ RT6_LOOKUP_F_IFACE |
+ RT6_LOOKUP_F_NO_BLK_HOLE |
+ RT6_LOOKUP_F_NO_PROHIBIT);
if (rt != net->ipv6.ip6_null_entry &&
addrconf_is_prefix_route(rt)) {
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 2e1a432..d290da5 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -64,9 +64,13 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
goto discard_pkt;
default:
case FR_ACT_BLACKHOLE:
+ if (flags & RT6_LOOKUP_F_NO_BLK_HOLE)
+ goto again;
rt = net->ipv6.ip6_blk_hole_entry;
goto discard_pkt;
case FR_ACT_PROHIBIT:
+ if (flags & RT6_LOOKUP_F_NO_PROHIBIT)
+ goto again;
rt = net->ipv6.ip6_prohibit_entry;
goto discard_pkt;
}
--
1.7.2.5
^ permalink raw reply related
* Re: [3.8-rc] regression: NETDEV WATCHDOG: eth0 (r8169): transmit queue 0 timed out
From: Francois Romieu @ 2013-01-05 16:57 UTC (permalink / raw)
To: Jörg Otte; +Cc: Linux Kernel Mailing List, David S. Miller, netdev
In-Reply-To: <CADDKRnBCrDz2X=5qpGE+r92ZFX5cA+jnSGt3wS8TR0RE8hi1HA@mail.gmail.com>
Jörg Otte <jrg.otte@gmail.com> :
[...]
> jojo@ahorn:~$ dmesg | grep XID
> [ 1.808847] r8169 0000:02:00.0 eth0: RTL8168evl/8111evl at
> 0xffffc90000054000, 5c:9a:d8:69:2b:39, XID 0c900800 IRQ 42
Can you check if things improve with v3.8-rc2 after removing :
1. 9ecb9aabaf634677c77af467f4e3028b09d7bcda
r8169: workaround for missing extended GigaMAC registers
2. d64ec841517a25f6d468bde9f67e5b4cffdc67c7
r8169: enable internal ASPM and clock request settings
3. e0c075577965d1c01b30038d38bf637b027a1df3
r8169: enable ALDPS for power saving
(you can directly try v3.7 r8169.c with v3.8-rc2 if it worked for you
so far)
If the regression is still there, please apply the patch below to both
v3.8-rc2 unpatched and a known working version then send me their dmesg
after you 'ip link set dev eth0 up'.
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index ed96f30..3d2d2446 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -90,10 +90,28 @@ static const int multicast_filter_limit = 32;
#define RTL8169_TX_TIMEOUT (6*HZ)
#define RTL8169_PHY_TIMEOUT (10*HZ)
+static void rw8(void __iomem *ioaddr, u8 b)
+{
+ printk(KERN_DEBUG PFX "w %p %02x\n", ioaddr, b);
+ writeb(b, ioaddr);
+}
+
+static void rw16(void __iomem *ioaddr, u16 w)
+{
+ printk(KERN_DEBUG PFX "w %p %04x\n", ioaddr, w);
+ writew(w, ioaddr);
+}
+
+static void rw32(void __iomem *ioaddr, u32 d)
+{
+ printk(KERN_DEBUG PFX "w %p %08x\n", ioaddr, d);
+ writel(d, ioaddr);
+}
+
/* write/read MMIO register */
-#define RTL_W8(reg, val8) writeb ((val8), ioaddr + (reg))
-#define RTL_W16(reg, val16) writew ((val16), ioaddr + (reg))
-#define RTL_W32(reg, val32) writel ((val32), ioaddr + (reg))
+#define RTL_W8(reg, val8) rw8(ioaddr + (reg), (val8))
+#define RTL_W16(reg, val16) rw16(ioaddr + (reg), (val16))
+#define RTL_W32(reg, val32) rw32(ioaddr + (reg), (val32))
#define RTL_R8(reg) readb (ioaddr + (reg))
#define RTL_R16(reg) readw (ioaddr + (reg))
#define RTL_R32(reg) readl (ioaddr + (reg))
^ permalink raw reply related
* Re: [PATCH] bgmac: driver for GBit MAC core on BCMA bus
From: Francois Romieu @ 2013-01-05 16:57 UTC (permalink / raw)
To: Rafał Miłecki; +Cc: netdev, David S. Miller
In-Reply-To: <CACna6ry4fQCvh_hSKooNi1Lq=ftjVD=tNeDijnfW1gnnvGzLeQ@mail.gmail.com>
Rafał Miłecki <zajec5@gmail.com> :
[...]
> We really need some good explanation for NETDEV_TX_BUSY somewhere
> (documentation!).
It is already explained in Documentation/networking/driver.txt:
<snip>
1) The ndo_start_xmit method must not return NETDEV_TX_BUSY under
any normal circumstances. It is considered a hard error unless
there is no way your device can tell ahead of time when it's
transmit function will become busy.
Instead it must maintain the queue properly. For example,
for a driver implementing scatter-gather this means:
</snip>
Please compare your V3:bgmac_dma_tx_add and the sample code after the
aforementioned paragraph.
> I've checked some drivers and:
>
> 1) b44.c
> In case of:
> > if (unlikely(TX_BUFFS_AVAIL(bp) < 1)) {
> it does netif_stop_queue and returns NETDEV_TX_BUSY
The driver labels it as bug checking and it stops queueing when need
arises at the end of the xmit function.
> 2) sky2.c
> In case of:
> > if (unlikely(tx_avail(sky2) < tx_le_req(skb)))
> it just returns NETDEV_TX_BUSY
Same thing as above. Less documented though.
> 3) jme.2
> In case of:
> > if (unlikely(idx < 0)) {
> it does netif_stop_queue and returns NETDEV_TX_BUSY
The driver labels it as bug checking. It stops queueing when need
arises at the end of the xmit function (see jme_stop_queue_if_full).
> 4) de2104x.c
> In case of:
> > if (tx_free == 0) {
> it does netif_stop_queue and returns NETDEV_TX_BUSY
It stops queueing at the end of the start_xmit handler as well.
> So there are many drivers doing the same mistake I did.
After a packet has been handed to hardware, each of these drivers
checks if queueing should be disabled before returning from start_xmit.
V3 bgmac doesn't.
--
Ueimor
^ permalink raw reply
* Re: [PATCH] bgmac: driver for GBit MAC core on BCMA bus
From: Rafał Miłecki @ 2013-01-05 18:28 UTC (permalink / raw)
To: Francois Romieu; +Cc: netdev, David S. Miller
In-Reply-To: <20130105165715.GA4906@electric-eye.fr.zoreil.com>
2013/1/5 Francois Romieu <romieu@fr.zoreil.com>:
> Rafał Miłecki <zajec5@gmail.com> :
> [...]
>> We really need some good explanation for NETDEV_TX_BUSY somewhere
>> (documentation!).
>
> It is already explained in Documentation/networking/driver.txt:
> <snip>
> 1) The ndo_start_xmit method must not return NETDEV_TX_BUSY under
> any normal circumstances. It is considered a hard error unless
> there is no way your device can tell ahead of time when it's
> transmit function will become busy.
>
> Instead it must maintain the queue properly. For example,
> for a driver implementing scatter-gather this means:
> </snip>
>
> Please compare your V3:bgmac_dma_tx_add and the sample code after the
> aforementioned paragraph.
>
>> I've checked some drivers and:
>>
>> 1) b44.c
>> In case of:
>> > if (unlikely(TX_BUFFS_AVAIL(bp) < 1)) {
>> it does netif_stop_queue and returns NETDEV_TX_BUSY
>
> The driver labels it as bug checking and it stops queueing when need
> arises at the end of the xmit function.
>
>> 2) sky2.c
>> In case of:
>> > if (unlikely(tx_avail(sky2) < tx_le_req(skb)))
>> it just returns NETDEV_TX_BUSY
>
> Same thing as above. Less documented though.
>
>> 3) jme.2
>> In case of:
>> > if (unlikely(idx < 0)) {
>> it does netif_stop_queue and returns NETDEV_TX_BUSY
>
> The driver labels it as bug checking. It stops queueing when need
> arises at the end of the xmit function (see jme_stop_queue_if_full).
>
>> 4) de2104x.c
>> In case of:
>> > if (tx_free == 0) {
>> it does netif_stop_queue and returns NETDEV_TX_BUSY
>
> It stops queueing at the end of the start_xmit handler as well.
>
>> So there are many drivers doing the same mistake I did.
>
> After a packet has been handed to hardware, each of these drivers
> checks if queueing should be disabled before returning from start_xmit.
>
> V3 bgmac doesn't.
How did I miss that?! Thanks for pointing that out!
--
Rafał
^ permalink raw reply
* Re: [PATCH] tcp: make sysctl_tcp_ecn namespace aware
From: Stephen Hemminger @ 2013-01-05 19:00 UTC (permalink / raw)
To: Hannes Frederic Sowa; +Cc: netdev, eric.dumazet, davem
In-Reply-To: <20130105150130.GC4031@order.stressinduktion.org>
Rather than passing sysctl_tcp_ecn around as a parameter, I think it
would be clearer and more efficient to make the ECN functions namespace
aware.
Instead of:
> @@ -728,7 +728,8 @@ struct tcp_skb_cb {
> * notifications, we disable TCP ECN negociation.
> */
> static inline void
> -TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb)
> +TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb,
> + int sysctl_tcp_ecn)
> {
> const struct tcphdr *th = tcp_hdr(skb);
>
@@ -731,8 +730,9 @@ static inline void
TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb)
{
const struct tcphdr *th = tcp_hdr(skb);
-
- if (sysctl_tcp_ecn && th->ece && th->cwr &&
+
+ if (sock_net(req->sk)->ipv4.sysctl_tcp_ecn &&
+ th->ece && th->cwr &&
INET_ECN_is_not_ect(TCP_SKB_CB(skb)->ip_dsfield))
inet_rsk(req)->ecn_ok = 1;
}
^ permalink raw reply
* Re: [PATCH] tcp: make sysctl_tcp_ecn namespace aware
From: Hannes Frederic Sowa @ 2013-01-05 19:05 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev, eric.dumazet, davem
In-Reply-To: <20130105110005.2bf230eb@nehalam.linuxnetplumber.net>
On Sat, Jan 05, 2013 at 11:00:05AM -0800, Stephen Hemminger wrote:
> Rather than passing sysctl_tcp_ecn around as a parameter, I think it
> would be clearer and more efficient to make the ECN functions namespace
> aware.
>
> Instead of:
> > @@ -728,7 +728,8 @@ struct tcp_skb_cb {
> > * notifications, we disable TCP ECN negociation.
> > */
> > static inline void
> > -TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb)
> > +TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb,
> > + int sysctl_tcp_ecn)
> > {
> > const struct tcphdr *th = tcp_hdr(skb);
> >
>
> @@ -731,8 +730,9 @@ static inline void
> TCP_ECN_create_request(struct request_sock *req, const struct sk_buff *skb)
> {
> const struct tcphdr *th = tcp_hdr(skb);
> -
> - if (sysctl_tcp_ecn && th->ece && th->cwr &&
> +
> + if (sock_net(req->sk)->ipv4.sysctl_tcp_ecn &&
> + th->ece && th->cwr &&
> INET_ECN_is_not_ect(TCP_SKB_CB(skb)->ip_dsfield))
> inet_rsk(req)->ecn_ok = 1;
Having a quick look, I think req->sk is not initialized at that moment, but
I'll verify later.
Thanks!
^ permalink raw reply
* Re: [PATCH V2 3/3] tuntap: don't add to waitqueue when POLLERR
From: Eric Dumazet @ 2013-01-05 19:37 UTC (permalink / raw)
To: Jason Wang; +Cc: davem, mst, netdev, linux-kernel
In-Reply-To: <1357378446-19656-4-git-send-email-jasowang@redhat.com>
On Sat, 2013-01-05 at 17:34 +0800, Jason Wang wrote:
> Currently, tun_chr_poll() returns POLLERR after waitqueue adding during device
> unregistration. This would confuse some of its user such as vhost which assume
> when POLLERR is returned, it wasn't added to the waitqueue. Fix this by
> returning POLLERR before adding to waitqueue.
>
> Signed-off-by: Jason Wang <jasowang@redhat.com>
> ---
> drivers/net/tun.c | 5 +----
> 1 files changed, 1 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/tun.c b/drivers/net/tun.c
> index fbd106e..f9c0049 100644
> --- a/drivers/net/tun.c
> +++ b/drivers/net/tun.c
> @@ -886,7 +886,7 @@ static unsigned int tun_chr_poll(struct file *file, poll_table *wait)
> struct sock *sk;
> unsigned int mask = 0;
>
> - if (!tun)
> + if (!tun || tun->dev->reg_state != NETREG_REGISTERED)
> return POLLERR;
>
> sk = tfile->socket.sk;
> @@ -903,9 +903,6 @@ static unsigned int tun_chr_poll(struct file *file, poll_table *wait)
> sock_writeable(sk)))
> mask |= POLLOUT | POLLWRNORM;
>
> - if (tun->dev->reg_state != NETREG_REGISTERED)
> - mask = POLLERR;
> -
> tun_put(tun);
> return mask;
> }
This patch is buggy.
First, the caller assuming POLLERR means poll_wait() was not called is
wrong.
Secondly, you add a ref leak.
^ permalink raw reply
* [patch net-next] ethtool: fix drvinfo strings set in drivers
From: Jiri Pirko @ 2013-01-05 19:48 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, shemminger, fbl, sathya.perla,
subbu.seetharaman, ajit.khaparde, roland, sean.hefty,
hal.rosenstock
Use strlcpy where possible to ensure the string is \0 terminated.
Use always sizeof(string) instead of 32 and custom defines.
Also use snprintf instead of strcpy in nes_nic.
Also do not do unnecessary init of drvinfo in spider_net_ethtool.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/infiniband/hw/nes/nes_nic.c | 12 +++++++-----
drivers/infiniband/ulp/ipoib/ipoib_ethtool.c | 2 +-
drivers/net/bonding/bond_main.c | 9 +++++----
drivers/net/ethernet/broadcom/bcm63xx_enet.c | 9 +++++----
drivers/net/ethernet/emulex/benet/be_ethtool.c | 12 ++++++------
drivers/net/ethernet/freescale/gianfar.h | 2 --
drivers/net/ethernet/freescale/gianfar_ethtool.c | 9 +++++----
drivers/net/ethernet/freescale/ucc_geth_ethtool.c | 8 ++++----
drivers/net/ethernet/tehuti/tehuti.c | 8 ++++----
drivers/net/ethernet/toshiba/spider_net_ethtool.c | 12 +++++++-----
drivers/net/macvlan.c | 4 ++--
net/dsa/slave.c | 8 ++++----
12 files changed, 50 insertions(+), 45 deletions(-)
diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c
index 9542e16..177cf76 100644
--- a/drivers/infiniband/hw/nes/nes_nic.c
+++ b/drivers/infiniband/hw/nes/nes_nic.c
@@ -1317,11 +1317,13 @@ static void nes_netdev_get_drvinfo(struct net_device *netdev,
struct nes_vnic *nesvnic = netdev_priv(netdev);
struct nes_adapter *nesadapter = nesvnic->nesdev->nesadapter;
- strcpy(drvinfo->driver, DRV_NAME);
- strcpy(drvinfo->bus_info, pci_name(nesvnic->nesdev->pcidev));
- sprintf(drvinfo->fw_version, "%u.%u", nesadapter->firmware_version>>16,
- nesadapter->firmware_version & 0x000000ff);
- strcpy(drvinfo->version, DRV_VERSION);
+ strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->bus_info, pci_name(nesvnic->nesdev->pcidev),
+ sizeof(drvinfo->bus_info));
+ snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+ "%u.%u", nesadapter->firmware_version >> 16,
+ nesadapter->firmware_version & 0x000000ff);
+ strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
drvinfo->testinfo_len = 0;
drvinfo->eedump_len = 0;
drvinfo->regdump_len = 0;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
index 29bc7b5..ca13133 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
@@ -39,7 +39,7 @@
static void ipoib_get_drvinfo(struct net_device *netdev,
struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, "ipoib", sizeof(drvinfo->driver) - 1);
+ strlcpy(drvinfo->driver, "ipoib", sizeof(drvinfo->driver));
}
static int ipoib_get_coalesce(struct net_device *dev,
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index c495298..564cf42 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -4330,11 +4330,12 @@ void bond_set_mode_ops(struct bonding *bond, int mode)
}
static void bond_ethtool_get_drvinfo(struct net_device *bond_dev,
- struct ethtool_drvinfo *drvinfo)
+ struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, DRV_NAME, 32);
- strncpy(drvinfo->version, DRV_VERSION, 32);
- snprintf(drvinfo->fw_version, 32, "%d", BOND_ABI_VERSION);
+ strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
+ snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%d",
+ BOND_ABI_VERSION);
}
static const struct ethtool_ops bond_ethtool_ops = {
diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
index 39387d6..d8a1510 100644
--- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c
+++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
@@ -1227,10 +1227,11 @@ static const u32 unused_mib_regs[] = {
static void bcm_enet_get_drvinfo(struct net_device *netdev,
struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, bcm_enet_driver_name, 32);
- strncpy(drvinfo->version, bcm_enet_driver_version, 32);
- strncpy(drvinfo->fw_version, "N/A", 32);
- strncpy(drvinfo->bus_info, "bcm63xx", 32);
+ strlcpy(drvinfo->driver, bcm_enet_driver_name, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, bcm_enet_driver_version,
+ sizeof(drvinfo->version));
+ strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
+ strlcpy(drvinfo->bus_info, "bcm63xx", sizeof(drvinfo->bus_info));
drvinfo->n_stats = BCM_ENET_STATS_LEN;
}
diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c
index 00454a1..76b302f 100644
--- a/drivers/net/ethernet/emulex/benet/be_ethtool.c
+++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c
@@ -183,12 +183,12 @@ static void be_get_drvinfo(struct net_device *netdev,
strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
strlcpy(drvinfo->version, DRV_VER, sizeof(drvinfo->version));
- strncpy(drvinfo->fw_version, adapter->fw_ver, FW_VER_LEN);
- if (memcmp(adapter->fw_ver, fw_on_flash, FW_VER_LEN) != 0) {
- strcat(drvinfo->fw_version, " [");
- strcat(drvinfo->fw_version, fw_on_flash);
- strcat(drvinfo->fw_version, "]");
- }
+ if (!memcmp(adapter->fw_ver, fw_on_flash, FW_VER_LEN))
+ strlcpy(drvinfo->fw_version, adapter->fw_ver,
+ sizeof(drvinfo->fw_version));
+ else
+ snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+ "%s [%s]", adapter->fw_ver, fw_on_flash);
strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
sizeof(drvinfo->bus_info));
diff --git a/drivers/net/ethernet/freescale/gianfar.h b/drivers/net/ethernet/freescale/gianfar.h
index 22eabc1..1b6a67c 100644
--- a/drivers/net/ethernet/freescale/gianfar.h
+++ b/drivers/net/ethernet/freescale/gianfar.h
@@ -649,8 +649,6 @@ struct gfar_extra_stats {
/* Number of stats in the stats structure (ignore car and cam regs)*/
#define GFAR_STATS_LEN (GFAR_RMON_LEN + GFAR_EXTRA_STATS_LEN)
-#define GFAR_INFOSTR_LEN 32
-
struct gfar_stats {
u64 extra[GFAR_EXTRA_STATS_LEN];
u64 rmon[GFAR_RMON_LEN];
diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c
index ab6762c..d0fe53c 100644
--- a/drivers/net/ethernet/freescale/gianfar_ethtool.c
+++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c
@@ -184,10 +184,11 @@ static int gfar_sset_count(struct net_device *dev, int sset)
static void gfar_gdrvinfo(struct net_device *dev,
struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, DRV_NAME, GFAR_INFOSTR_LEN);
- strncpy(drvinfo->version, gfar_driver_version, GFAR_INFOSTR_LEN);
- strncpy(drvinfo->fw_version, "N/A", GFAR_INFOSTR_LEN);
- strncpy(drvinfo->bus_info, "N/A", GFAR_INFOSTR_LEN);
+ strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, gfar_driver_version,
+ sizeof(drvinfo->version));
+ strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
+ strlcpy(drvinfo->bus_info, "N/A", sizeof(drvinfo->bus_info));
drvinfo->regdump_len = 0;
drvinfo->eedump_len = 0;
}
diff --git a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
index 37b0353..1ebf712 100644
--- a/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
+++ b/drivers/net/ethernet/freescale/ucc_geth_ethtool.c
@@ -350,10 +350,10 @@ static void
uec_get_drvinfo(struct net_device *netdev,
struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, DRV_NAME, 32);
- strncpy(drvinfo->version, DRV_VERSION, 32);
- strncpy(drvinfo->fw_version, "N/A", 32);
- strncpy(drvinfo->bus_info, "QUICC ENGINE", 32);
+ strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
+ strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
+ strlcpy(drvinfo->bus_info, "QUICC ENGINE", sizeof(drvinfo->bus_info));
drvinfo->eedump_len = 0;
drvinfo->regdump_len = uec_get_regs_len(netdev);
}
diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c
index 1e4d743..e15cc71 100644
--- a/drivers/net/ethernet/tehuti/tehuti.c
+++ b/drivers/net/ethernet/tehuti/tehuti.c
@@ -2179,10 +2179,10 @@ bdx_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
{
struct bdx_priv *priv = netdev_priv(netdev);
- strlcat(drvinfo->driver, BDX_DRV_NAME, sizeof(drvinfo->driver));
- strlcat(drvinfo->version, BDX_DRV_VERSION, sizeof(drvinfo->version));
- strlcat(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
- strlcat(drvinfo->bus_info, pci_name(priv->pdev),
+ strlcpy(drvinfo->driver, BDX_DRV_NAME, sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, BDX_DRV_VERSION, sizeof(drvinfo->version));
+ strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
+ strlcpy(drvinfo->bus_info, pci_name(priv->pdev),
sizeof(drvinfo->bus_info));
drvinfo->n_stats = ((priv->stats_flag) ? ARRAY_SIZE(bdx_stat_names) : 0);
diff --git a/drivers/net/ethernet/toshiba/spider_net_ethtool.c b/drivers/net/ethernet/toshiba/spider_net_ethtool.c
index 9c288cd..ffe5193 100644
--- a/drivers/net/ethernet/toshiba/spider_net_ethtool.c
+++ b/drivers/net/ethernet/toshiba/spider_net_ethtool.c
@@ -72,11 +72,13 @@ spider_net_ethtool_get_drvinfo(struct net_device *netdev,
card = netdev_priv(netdev);
/* clear and fill out info */
- memset(drvinfo, 0, sizeof(struct ethtool_drvinfo));
- strncpy(drvinfo->driver, spider_net_driver_name, 32);
- strncpy(drvinfo->version, VERSION, 32);
- strcpy(drvinfo->fw_version, "no information");
- strncpy(drvinfo->bus_info, pci_name(card->pdev), 32);
+ strlcpy(drvinfo->driver, spider_net_driver_name,
+ sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, VERSION, sizeof(drvinfo->version));
+ strlcpy(drvinfo->fw_version, "no information",
+ sizeof(drvinfo->fw_version));
+ strlcpy(drvinfo->bus_info, pci_name(card->pdev),
+ sizeof(drvinfo->bus_info));
}
static void
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 0ae4518..1047e58 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -585,8 +585,8 @@ static int macvlan_fdb_del(struct ndmsg *ndm,
static void macvlan_ethtool_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *drvinfo)
{
- snprintf(drvinfo->driver, 32, "macvlan");
- snprintf(drvinfo->version, 32, "0.1");
+ strlcpy(drvinfo->driver, "macvlan", sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, "0.1", sizeof(drvinfo->version));
}
static int macvlan_ethtool_get_settings(struct net_device *dev,
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index e32083d..f795b0c 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -203,10 +203,10 @@ dsa_slave_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
static void dsa_slave_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, "dsa", 32);
- strncpy(drvinfo->version, dsa_driver_version, 32);
- strncpy(drvinfo->fw_version, "N/A", 32);
- strncpy(drvinfo->bus_info, "platform", 32);
+ strlcpy(drvinfo->driver, "dsa", sizeof(drvinfo->driver));
+ strlcpy(drvinfo->version, dsa_driver_version, sizeof(drvinfo->version));
+ strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version));
+ strlcpy(drvinfo->bus_info, "platform", sizeof(drvinfo->bus_info));
}
static int dsa_slave_nway_reset(struct net_device *dev)
--
1.8.1
^ permalink raw reply related
* Re: [PATCH 1/2] ipv6: avoid blackhole and prohibited entries upon prefix purge
From: Eric Dumazet @ 2013-01-05 19:59 UTC (permalink / raw)
To: Romain KUNTZ; +Cc: netdev, yoshfuji, davem
In-Reply-To: <0CC79564-4AF2-42F9-8D06-1BCC912A1AF7@ipflavors.com>
On Sat, 2013-01-05 at 17:13 +0100, Romain KUNTZ wrote:
> Mobile IPv6 provokes a kernel Oops since commit 64c6d08e (ipv6:
> del unreachable route when an addr is deleted on lo), because
> ip6_route_lookup() may also return blackhole and prohibited
> entry. However, these entries have a NULL rt6i_table argument,
> which provokes an Oops in __ip6_del_rt() when trying to lock
> rt6i_table->tb6_lock.
>
> Beside, when purging a prefix, blakhole and prohibited entries
> should not be selected because they are not what we are looking
> for.
>
> We fix this by adding two new lookup flags (RT6_LOOKUP_F_NO_BLK_HOLE
> and RT6_LOOKUP_F_NO_PROHIBIT) in order to ensure that such entries
> are skipped during lookup and that the correct entry is returned.
>
> Signed-off-by: Romain Kuntz <r.kuntz@ipflavors.com>
> ---
> include/net/ip6_route.h | 2 ++
> net/ipv6/addrconf.c | 4 +++-
> net/ipv6/fib6_rules.c | 4 ++++
> 3 files changed, 9 insertions(+), 1 deletions(-)
>
> diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
> index 27d8318..3c93743 100644
> --- a/include/net/ip6_route.h
> +++ b/include/net/ip6_route.h
> @@ -30,6 +30,8 @@ struct route_info {
> #define RT6_LOOKUP_F_SRCPREF_TMP 0x00000008
> #define RT6_LOOKUP_F_SRCPREF_PUBLIC 0x00000010
> #define RT6_LOOKUP_F_SRCPREF_COA 0x00000020
> +#define RT6_LOOKUP_F_NO_BLK_HOLE 0x00000040
> +#define RT6_LOOKUP_F_NO_PROHIBIT 0x00000080
>
> /*
> * rt6_srcprefs2flags() and rt6_flags2srcprefs() translate
> diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
> index 408cac4a..1891e23 100644
> --- a/net/ipv6/addrconf.c
> +++ b/net/ipv6/addrconf.c
> @@ -948,7 +948,9 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
> fl6.flowi6_oif = ifp->idev->dev->ifindex;
> fl6.daddr = prefix;
> rt = (struct rt6_info *)ip6_route_lookup(net, &fl6,
> - RT6_LOOKUP_F_IFACE);
> + RT6_LOOKUP_F_IFACE |
> + RT6_LOOKUP_F_NO_BLK_HOLE |
> + RT6_LOOKUP_F_NO_PROHIBIT);
>
> if (rt != net->ipv6.ip6_null_entry &&
> addrconf_is_prefix_route(rt)) {
> diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
> index 2e1a432..d290da5 100644
> --- a/net/ipv6/fib6_rules.c
> +++ b/net/ipv6/fib6_rules.c
> @@ -64,9 +64,13 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
> goto discard_pkt;
> default:
> case FR_ACT_BLACKHOLE:
> + if (flags & RT6_LOOKUP_F_NO_BLK_HOLE)
> + goto again;
goto out; ???
> rt = net->ipv6.ip6_blk_hole_entry;
> goto discard_pkt;
> case FR_ACT_PROHIBIT:
> + if (flags & RT6_LOOKUP_F_NO_PROHIBIT)
> + goto again;
goto out;
> rt = net->ipv6.ip6_prohibit_entry;
> goto discard_pkt;
> }
^ 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