* [PATCH net-next 2/2] ptp: reduce stack usage when measuring the system time offset
From: Richard Cochran @ 2012-11-26 11:44 UTC (permalink / raw)
To: netdev; +Cc: David Miller
In-Reply-To: <1146e32bcb835ebf394bab91db3161600ab5213a.1353929829.git.richardcochran@gmail.com>
This patch removes the large buffer from the stack of the system
offset ioctl and replaces it with a kmalloced buffer.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
---
drivers/ptp/ptp_chardev.c | 21 ++++++++++++++-------
1 files changed, 14 insertions(+), 7 deletions(-)
diff --git a/drivers/ptp/ptp_chardev.c b/drivers/ptp/ptp_chardev.c
index 9d7542e..34a0c60 100644
--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -34,7 +34,7 @@ long ptp_ioctl(struct posix_clock *pc, unsigned int cmd, unsigned long arg)
{
struct ptp_clock_caps caps;
struct ptp_clock_request req;
- struct ptp_sys_offset sysoff;
+ struct ptp_sys_offset *sysoff = NULL;
struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
struct ptp_clock_info *ops = ptp->info;
struct ptp_clock_time *pct;
@@ -94,17 +94,22 @@ long ptp_ioctl(struct posix_clock *pc, unsigned int cmd, unsigned long arg)
break;
case PTP_SYS_OFFSET:
- if (copy_from_user(&sysoff, (void __user *)arg,
- sizeof(sysoff))) {
+ sysoff = kmalloc(sizeof(*sysoff), GFP_KERNEL);
+ if (!sysoff) {
+ err = -ENOMEM;
+ break;
+ }
+ if (copy_from_user(sysoff, (void __user *)arg,
+ sizeof(*sysoff))) {
err = -EFAULT;
break;
}
- if (sysoff.n_samples > PTP_MAX_SAMPLES) {
+ if (sysoff->n_samples > PTP_MAX_SAMPLES) {
err = -EINVAL;
break;
}
- pct = &sysoff.ts[0];
- for (i = 0; i < sysoff.n_samples; i++) {
+ pct = &sysoff->ts[0];
+ for (i = 0; i < sysoff->n_samples; i++) {
getnstimeofday(&ts);
pct->sec = ts.tv_sec;
pct->nsec = ts.tv_nsec;
@@ -117,7 +122,7 @@ long ptp_ioctl(struct posix_clock *pc, unsigned int cmd, unsigned long arg)
getnstimeofday(&ts);
pct->sec = ts.tv_sec;
pct->nsec = ts.tv_nsec;
- if (copy_to_user((void __user *)arg, &sysoff, sizeof(sysoff)))
+ if (copy_to_user((void __user *)arg, sysoff, sizeof(*sysoff)))
err = -EFAULT;
break;
@@ -125,6 +130,8 @@ long ptp_ioctl(struct posix_clock *pc, unsigned int cmd, unsigned long arg)
err = -ENOTTY;
break;
}
+
+ kfree(sysoff);
return err;
}
--
1.7.2.5
^ permalink raw reply related
* [PATCH net-next 1/2] ptp: reduce stack usage when reading external time stamps
From: Richard Cochran @ 2012-11-26 11:44 UTC (permalink / raw)
To: netdev; +Cc: David Miller
In-Reply-To: <20121102020631.GI16883@yliu-dev.sh.intel.com>
This patch removes the large buffer from the stack of the read file
operation and replaces it with a kmalloced buffer.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
---
drivers/ptp/ptp_chardev.c | 22 +++++++++++++++++-----
1 files changed, 17 insertions(+), 5 deletions(-)
diff --git a/drivers/ptp/ptp_chardev.c b/drivers/ptp/ptp_chardev.c
index 4f8ae80..9d7542e 100644
--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -21,6 +21,7 @@
#include <linux/posix-clock.h>
#include <linux/poll.h>
#include <linux/sched.h>
+#include <linux/slab.h>
#include "ptp_private.h"
@@ -136,20 +137,23 @@ unsigned int ptp_poll(struct posix_clock *pc, struct file *fp, poll_table *wait)
return queue_cnt(&ptp->tsevq) ? POLLIN : 0;
}
+#define EXTTS_BUFSIZE (PTP_BUF_TIMESTAMPS * sizeof(struct ptp_extts_event))
+
ssize_t ptp_read(struct posix_clock *pc,
uint rdflags, char __user *buf, size_t cnt)
{
struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
struct timestamp_event_queue *queue = &ptp->tsevq;
- struct ptp_extts_event event[PTP_BUF_TIMESTAMPS];
+ struct ptp_extts_event *event;
unsigned long flags;
size_t qcnt, i;
+ int result;
if (cnt % sizeof(struct ptp_extts_event) != 0)
return -EINVAL;
- if (cnt > sizeof(event))
- cnt = sizeof(event);
+ if (cnt > EXTTS_BUFSIZE)
+ cnt = EXTTS_BUFSIZE;
cnt = cnt / sizeof(struct ptp_extts_event);
@@ -167,6 +171,12 @@ ssize_t ptp_read(struct posix_clock *pc,
return -ENODEV;
}
+ event = kmalloc(EXTTS_BUFSIZE, GFP_KERNEL);
+ if (!event) {
+ mutex_unlock(&ptp->tsevq_mux);
+ return -ENOMEM;
+ }
+
spin_lock_irqsave(&queue->lock, flags);
qcnt = queue_cnt(queue);
@@ -185,8 +195,10 @@ ssize_t ptp_read(struct posix_clock *pc,
mutex_unlock(&ptp->tsevq_mux);
+ result = cnt;
if (copy_to_user(buf, event, cnt))
- return -EFAULT;
+ result = -EFAULT;
- return cnt;
+ kfree(event);
+ return result;
}
--
1.7.2.5
^ permalink raw reply related
* Re: [PATCH] net/macb: Use non-coherent memory for rx buffers
From: Nicolas Ferre @ 2012-11-26 10:44 UTC (permalink / raw)
To: Joachim Eastwood
Cc: David S. Miller, netdev, linux-arm-kernel, linux-kernel,
Jean-Christophe PLAGNIOL-VILLARD, Havard Skinnemoen
In-Reply-To: <CAGhQ9VwWmP_B17LnAEqDHcL8Yd-s0C-Bf35z3+=vnJmK_NoCSw@mail.gmail.com>
On 11/23/2012 05:12 PM, Joachim Eastwood :
> Hi Nicolas,
>
> On 23 November 2012 14:50, Nicolas Ferre <nicolas.ferre@atmel.com> wrote:
>> From: Havard Skinnemoen <havard@skinnemoen.net>
>>
>> Allocate regular pages to use as backing for the RX ring and use the
>> DMA API to sync the caches. This should give a bit better performance
>> since it allows the CPU to do burst transfers from memory. It is also
>> a necessary step on the way to reduce the amount of copying done by
>> the driver.
>>
>> Signed-off-by: Havard Skinnemoen <havard@skinnemoen.net>
>> [nicolas.ferre@atmel.com: adapt to newer kernel]
>> Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
>> ---
>> drivers/net/ethernet/cadence/macb.c | 206 +++++++++++++++++++++++-------------
>> drivers/net/ethernet/cadence/macb.h | 20 +++-
>> 2 files changed, 148 insertions(+), 78 deletions(-)
>
> <snip>
>
>> diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
>> index 570908b..74e68a3 100644
>> --- a/drivers/net/ethernet/cadence/macb.h
>> +++ b/drivers/net/ethernet/cadence/macb.h
>> @@ -453,6 +453,23 @@ struct macb_dma_desc {
>> #define MACB_TX_USED_SIZE 1
>>
>> /**
>> + * struct macb_rx_page - data associated with a page used as RX buffers
>> + * @page: Physical page used as storage for the buffers
>> + * @phys: DMA address of the page
>> + *
>> + * Each page is used to provide %MACB_RX_BUFFERS_PER_PAGE RX buffers.
>> + * The page gets an initial reference when it is inserted into the
>> + * ring, and an additional reference each time it is passed up the
>> + * stack as a fragment. When all the buffers have been used, we drop
>> + * the initial reference and allocate a new page. Any additional
>> + * references are dropped when the higher layers free the skb.
>> + */
>> +struct macb_rx_page {
>> + struct page *page;
>> + dma_addr_t phys;
>> +};
>> +
>> +/**
>> * struct macb_tx_skb - data about an skb which is being transmitted
>> * @skb: skb currently being transmitted
>> * @mapping: DMA address of the skb's data buffer
>> @@ -543,7 +560,7 @@ struct macb {
>>
>> unsigned int rx_tail;
>> struct macb_dma_desc *rx_ring;
>> - void *rx_buffers;
>> + struct macb_rx_page *rx_page;
>>
>> unsigned int tx_head, tx_tail;
>> struct macb_dma_desc *tx_ring;
>> @@ -564,7 +581,6 @@ struct macb {
>>
>> dma_addr_t rx_ring_dma;
>> dma_addr_t tx_ring_dma;
>> - dma_addr_t rx_buffers_dma;
>>
>> struct mii_bus *mii_bus;
>> struct phy_device *phy_dev;
>> --
>
> struct macb is shared between at91_ether and macb. Removing
> rx_buffers_dma and rx_buffers will break compilation on at91_ether.
OMG, you are absolutely right.
> So please either leave the two struct members alone, for now, or fix
> up at91_ether at the same time.
Well, I do not plan to touch at91_ether driver for the moment, so I
certainly will keep the two struct members for now.
I will wait a little more feedback before sending a v2 patch with these
changes.
Best regards,
--
Nicolas Ferre
^ permalink raw reply
* Re: [PATCH 1/3] net: stmmac: change GMAC control register for SGMII
From: Giuseppe CAVALLARO @ 2012-11-26 10:31 UTC (permalink / raw)
To: Byungho An; +Cc: davem, jeffrey.t.kirsher, netdev, kgene.kim, linux-kernel
In-Reply-To: <004b01cdc959$80af8030$820e8090$%an@samsung.com>
On 11/23/2012 10:04 AM, Byungho An wrote:
>
> This patch changes GMAC control register (TC(Transmit
> Configuration) and PS(Port Selection) bit for SGMII.
> In case of SGMII, TC bit is '1' and PS bit is 0.
IMO this new support that should be released for net-next and further
effort is actually needed.
The availability of the PCS registers is given by looking at the HW
feature register. In fact, these are optional registers.
I don't want to break the compatibility with old chips.
I do not see why we have to use Kconfig macro to select ANE etc
(as you do in your patches).
The driver could directly manage the phy device by itself if possible
and the stmmac_init_phy should be reworked.
There are several things that need to be implemented. For example:
The ISR (e.g. priv->hw->mac->host_irq_status) should be able to manage
these new interrupts.
The code has to be able to maintain the user interface.
For example if you want to enable ANE or manage Advertisement caps.
> Signed-off-by: Byungho An <bh74.an@samsung.com>
> ---
[snip]
> + if (priv->phydev->interface == PHY_INTERFACE_MODE_SGMII) {
> + value = readl(priv->ioaddr);
> + /* GMAC_CONTROL_TC : transmit config in RGMII/SGMII */
> + value |= 0x1000000;
> + /* GMAC_CONTROL_PS : Port Selection for GMII */
> + value &= ~(0x8000);
> + writel(value, priv->ioaddr);
> + }
> +
This parts of code have to be moved in
drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
Pls, do not use value |= 0x1000000 but provide the appropriate defines.
> /* Request the IRQ lines */
> ret = request_irq(dev->irq, stmmac_interrupt,
> IRQF_SHARED, dev->name, dev);
>
^ permalink raw reply
* WARNING VIRUS DETECTED
From: WEBMAIL UPGRADE @ 2012-11-26 9:47 UTC (permalink / raw)
To: Recipients
Strong virus has been detected in your email account, which might make you lose your email account if not deleted. Use the below web link to Delete the virus in your email account:
Click link below:
https://docs.google.com/spreadsheet/viewform?formkey=dFRMRm9TOHlwS2lVNWtEZEhYTjVoMGc6MQ
Thank you for using our email.
Copyright ©2012 Email Helpdesk Centre
^ permalink raw reply
* Re: BQL support in gianfar causes network hickup
From: Tino Keitel @ 2012-11-26 10:01 UTC (permalink / raw)
To: Eric Dumazet, Paul Gortmaker; +Cc: netdev, Keitel, Tino (ALC NetworX GmbH)
In-Reply-To: <1353800616.2590.4562.camel@edumazet-glaptop>
On Sat, Nov 24, 2012 at 15:43:36 -0800, Eric Dumazet wrote:
[...]
> Hmm, I wonder if BQL makes a particular bug showing more often.
>
> I see gianfar uses a very small watchdog_timeo of 1 second, while many
> drivers use 5 seconds.
>
> What happens if you change this to 5 seconds ?
I still got the trace and a failing ptp client.
Regards,
Tino
^ permalink raw reply
* linux-next: manual merge of the arm-soc tree with the net-next tree
From: Stephen Rothwell @ 2012-11-26 9:45 UTC (permalink / raw)
To: Olof Johansson, Arnd Bergmann, linux-arm-kernel
Cc: linux-next, linux-kernel, Mugunthan V N, Vaibhav Hiremath,
David Miller, netdev, AnilKumar Ch, Benoit Cousson, Jon Hunter,
Afzal Mohammed, Philip, Avinash, Ajay Kumar Gupta
[-- Attachment #1: Type: text/plain, Size: 4866 bytes --]
Hi all,
Today's linux-next merge of the arm-soc tree got a conflict in
arch/arm/boot/dts/am33xx.dtsi between commit 1a39a65cba08 ("arm/dts:
am33xx: Add CPSW and MDIO module nodes for AM33XX") from the net-next
tree and commits 059b185d5345 ("ARM: dts: AM33XX: Add D_CAN device tree
data") and 4c94ac29b5c1 ("ARM: dts: OMAP: Move interrupt-parent to the
root node to avoid duplication") (and a few others that added more later
nodes) from the arm-soc tree.
I fixed it up (see below) and can carry the fix as necessary (no action
is required).
--
Cheers,
Stephen Rothwell sfr@canb.auug.org.au
diff --cc arch/arm/boot/dts/am33xx.dtsi
index a4615b4,20a3f29..0000000
--- a/arch/arm/boot/dts/am33xx.dtsi
+++ b/arch/arm/boot/dts/am33xx.dtsi
@@@ -210,53 -222,121 +222,168 @@@
interrupts = <91>;
};
+ mac: ethernet@4a100000 {
+ compatible = "ti,cpsw";
+ ti,hwmods = "cpgmac0";
+ cpdma_channels = <8>;
+ ale_entries = <1024>;
+ bd_ram_size = <0x2000>;
+ no_bd_ram = <0>;
+ rx_descs = <64>;
+ mac_control = <0x20>;
+ slaves = <2>;
+ cpts_active_slave = <0>;
+ cpts_clock_mult = <0x80000000>;
+ cpts_clock_shift = <29>;
+ reg = <0x4a100000 0x800
+ 0x4a101200 0x100>;
+ #address-cells = <1>;
+ #size-cells = <1>;
- interrupt-parent = <&intc>;
+ /*
+ * c0_rx_thresh_pend
+ * c0_rx_pend
+ * c0_tx_pend
+ * c0_misc_pend
+ */
+ interrupts = <40 41 42 43>;
+ ranges;
+
+ davinci_mdio: mdio@4a101000 {
+ compatible = "ti,davinci_mdio";
+ #address-cells = <1>;
+ #size-cells = <0>;
+ ti,hwmods = "davinci_mdio";
+ bus_freq = <1000000>;
+ reg = <0x4a101000 0x100>;
+ };
+
+ cpsw_emac0: slave@4a100200 {
+ /* Filled in by U-Boot */
+ mac-address = [ 00 00 00 00 00 00 ];
+ };
+
+ cpsw_emac1: slave@4a100300 {
+ /* Filled in by U-Boot */
+ mac-address = [ 00 00 00 00 00 00 ];
+ };
+
+ };
++
+ dcan0: d_can@481cc000 {
+ compatible = "bosch,d_can";
+ ti,hwmods = "d_can0";
+ reg = <0x481cc000 0x2000>;
+ interrupts = <52>;
+ status = "disabled";
+ };
+
+ dcan1: d_can@481d0000 {
+ compatible = "bosch,d_can";
+ ti,hwmods = "d_can1";
+ reg = <0x481d0000 0x2000>;
+ interrupts = <55>;
+ status = "disabled";
+ };
+
+ timer1: timer@44e31000 {
+ compatible = "ti,omap2-timer";
+ reg = <0x44e31000 0x400>;
+ interrupts = <67>;
+ ti,hwmods = "timer1";
+ ti,timer-alwon;
+ };
+
+ timer2: timer@48040000 {
+ compatible = "ti,omap2-timer";
+ reg = <0x48040000 0x400>;
+ interrupts = <68>;
+ ti,hwmods = "timer2";
+ };
+
+ timer3: timer@48042000 {
+ compatible = "ti,omap2-timer";
+ reg = <0x48042000 0x400>;
+ interrupts = <69>;
+ ti,hwmods = "timer3";
+ };
+
+ timer4: timer@48044000 {
+ compatible = "ti,omap2-timer";
+ reg = <0x48044000 0x400>;
+ interrupts = <92>;
+ ti,hwmods = "timer4";
+ ti,timer-pwm;
+ };
+
+ timer5: timer@48046000 {
+ compatible = "ti,omap2-timer";
+ reg = <0x48046000 0x400>;
+ interrupts = <93>;
+ ti,hwmods = "timer5";
+ ti,timer-pwm;
+ };
+
+ timer6: timer@48048000 {
+ compatible = "ti,omap2-timer";
+ reg = <0x48048000 0x400>;
+ interrupts = <94>;
+ ti,hwmods = "timer6";
+ ti,timer-pwm;
+ };
+
+ timer7: timer@4804a000 {
+ compatible = "ti,omap2-timer";
+ reg = <0x4804a000 0x400>;
+ interrupts = <95>;
+ ti,hwmods = "timer7";
+ ti,timer-pwm;
+ };
+
+ rtc@44e3e000 {
+ compatible = "ti,da830-rtc";
+ reg = <0x44e3e000 0x1000>;
+ interrupts = <75
+ 76>;
+ ti,hwmods = "rtc";
+ };
+
+ spi0: spi@48030000 {
+ compatible = "ti,omap4-mcspi";
+ #address-cells = <1>;
+ #size-cells = <0>;
+ reg = <0x48030000 0x400>;
+ interrupt = <65>;
+ ti,spi-num-cs = <2>;
+ ti,hwmods = "spi0";
+ status = "disabled";
+ };
+
+ spi1: spi@481a0000 {
+ compatible = "ti,omap4-mcspi";
+ #address-cells = <1>;
+ #size-cells = <0>;
+ reg = <0x481a0000 0x400>;
+ interrupt = <125>;
+ ti,spi-num-cs = <2>;
+ ti,hwmods = "spi1";
+ status = "disabled";
+ };
+
+ usb@47400000 {
+ compatible = "ti,musb-am33xx";
+ reg = <0x47400000 0x1000 /* usbss */
+ 0x47401000 0x800 /* musb instance 0 */
+ 0x47401800 0x800>; /* musb instance 1 */
+ interrupts = <17 /* usbss */
+ 18 /* musb instance 0 */
+ 19>; /* musb instance 1 */
+ multipoint = <1>;
+ num-eps = <16>;
+ ram-bits = <12>;
+ port0-mode = <3>;
+ port1-mode = <3>;
+ power = <250>;
+ ti,hwmods = "usb_otg_hs";
+ };
};
};
[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply
* Re: [PATCH] ewrk3: silence GCC warning
From: Paul Bolle @ 2012-11-26 9:41 UTC (permalink / raw)
To: richard -rw- weinberger; +Cc: netdev, linux-kernel
In-Reply-To: <CAFLxGvyfH3WdNpsGzV2OgvyT_8hu4U03+HEuv3u9DvSkOmNBtA@mail.gmail.com>
On Sun, 2012-11-25 at 11:45 +0100, richard -rw- weinberger wrote:
> > While we're touching this code also drop an outdated comment, that
> > should have been dropped with the patch named "MODULE_PARM conversions"
> > from early 2005.
>
> Please send this change as an additional patch.
Sure. Although it will be a trivial patch I suppose it's easiest for
everyone if I still send it to netdev (and not to trivial, which
normally handles stuff like that).
Paul Bolle
^ permalink raw reply
* RE: [PATCH 080/493] fddi: remove use of __devexit_p
From: David Laight @ 2012-11-26 9:33 UTC (permalink / raw)
To: Maciej W. Rozycki, Greg KH; +Cc: Bill Pemberton, netdev
In-Reply-To: <alpine.LFD.2.02.1211260425240.25230@eddie.linux-mips.org>
> > Even those don't care about less than 5k of memory, do they?
>
> I guess you're right. As long as it's not 5k per driver +
> who_knows_how_much per platform for some generic stuff that is. Of course
> this is still a waste, but I can accept it as a compromise between the use
If you are worried about that much memory (eg for a small embedded Linux)
I'r worry about the sizes of all the hash tables that keep popping up.
At least some of them really need converting to some kind of tree.
David
^ permalink raw reply
* Re: [PATCH] arping: Fix find_device_by_ifaddrs()
From: Jan Synacek @ 2012-11-26 9:26 UTC (permalink / raw)
To: yoshfuji; +Cc: netdev
In-Reply-To: <1353574275-21742-1-git-send-email-jsynacek@redhat.com>
On 11/22/2012 09:51 AM, Jan Synacek wrote:
> Look for another device if the device name and the currently found one are the
> same, not different.
>
> Also, make checking for the device's flags nonfatal.
>
> Signed-off-by: Jan Synacek <jsynacek@redhat.com>
> ---
> arping.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arping.c b/arping.c
> index ff77bec..d0edccf 100644
> --- a/arping.c
> +++ b/arping.c
> @@ -550,10 +550,10 @@ static int find_device_by_ifaddrs(void)
> continue;
> if (ifa->ifa_addr->sa_family != AF_PACKET)
> continue;
> - if (device.name && ifa->ifa_name && strcmp(ifa->ifa_name, device.name))
> + if (device.name && ifa->ifa_name && !strcmp(ifa->ifa_name, device.name))
> continue;
>
> - if (check_ifflags(ifa->ifa_flags, device.name != NULL) < 0)
> + if (check_ifflags(ifa->ifa_flags, 0) < 0)
> continue;
>
> if (!((struct sockaddr_ll *)ifa->ifa_addr)->sll_halen)
>
Please, disregard this patch. I found a problem with it and will send a proper
patch soon.
Cheers,
--
Jan Synacek
Software Engineer, BaseOS team Brno, Red Hat
^ permalink raw reply
* Re: private netdev flags into UAPI?
From: David Howells @ 2012-11-26 9:22 UTC (permalink / raw)
To: Or Gerlitz; +Cc: dhowells, netdev
In-Reply-To: <50B1CC1B.20104@mellanox.com>
Or Gerlitz <ogerlitz@mellanox.com> wrote:
> not sure if this has been brought up before, but I realized that the private
> IFF_yyy netdevice flags which weren't exposed to user space so far have been
> moved to include/uapi/linux/if.h, isn't that wrong?
They were exposed to userspace already. On my Fedora 16 test box, I see:
[root@andromeda ~]# grep -r IFF_EBRIDGE /usr/include/linux
/usr/include/linux/if.h:#define IFF_EBRIDGE 0x2 /* Ethernet bridging device. */
which is from:
kernel-headers-3.3.6-3.fc16.x86_64
which was built before UAPI proper started going in:
[root@andromeda ~]# rpm -qfi /usr/include/linux/if.h | grep "Build Date"
Build Date : Wed 16 May 2012 23:05:59 BST
Looking at commit 607ca46e97a1b6594b29647d98a32d545c24bdff, I see:
--------------------------- include/uapi/linux/if.h ---------------------------
similarity index 100%
rename from include/linux/if.h
rename to include/uapi/linux/if.h
so my script just moved it without alteration.
David
^ permalink raw reply
* Re: [RFC net-next PATCH V1 9/9] net: frag remove readers-writer lock (hack)
From: Florian Westphal @ 2012-11-26 9:18 UTC (permalink / raw)
To: Jesper Dangaard Brouer
Cc: Eric Dumazet, David S. Miller, Florian Westphal, netdev,
Pablo Neira Ayuso, Thomas Graf, Cong Wang, Patrick McHardy,
Paul E. McKenney, Herbert Xu
In-Reply-To: <20121123130847.18764.87682.stgit@dragon>
Jesper Dangaard Brouer <brouer@redhat.com> wrote:
> After all the other patches, the rw_lock is now the contention point.
>
> This is a quick hack, that remove the readers-writer lock, by
> disabling/breaking hash rebuilding. Just to see how big the
> performance gain would be.
>
> 2x10G size(4416) result: 6481+6764 = 13245 Mbit/s (gen: 7652+8077 Mbit/s)
>
> 4x10G size(4416) result:(5610+6283+5735+5238)=22866 Mbit/s
> (gen: 6530+7860+5967+5238 =25595 Mbit/s)
>
> And the results show, that its a big win. With 4x10G size(4416)
> before: 17923 Mbit/s -> now: 22866 Mbit/s increase 4943 Mbit/s.
> With 2x10G size(4416) before 10689 Mbit/s -> 13245 Mbit/s
> increase 2556 Mbit/s.
>
> I'll work on a real solution for removing the rw_lock while still
> supporting hash rebuilding. Suggestions and ideas are welcome.
<devils advocate>
Why not kill it altogether, and just set new secret_interval
without moving frag queues to new location?
The only consequence is that all fragments queued at the time of
changing secret_interval will be lost, and free'd by evictor/timer.
Default secret rebuild interval is 10 minutes, should we care about
small packet loss every 10 minutes?
</devils advocate>
^ permalink raw reply
* RE: Re: RTL 8169 linux driver question
From: David Laight @ 2012-11-26 9:15 UTC (permalink / raw)
To: Stéphane ANCELOT, netdev; +Cc: sancelot
In-Reply-To: <50AFCB1D.8080002@free.fr>
> I had problem with it, my application sends a frame that is immediately
> transmitted back by some slaves, there was abnormally 100us lost
> between the send and receive call.
>
> Finally I found it was coming from the following register setup in the
> driver :
>
> RTL_W16(IntrMitigate, 0x5151);
>
> Can you give me some details about it, since I do not have the RTL8169
> programming guide.
That sounds like an 'interrupt mitigation' setting - which will cause
RX interrupts to be delayed a short time in order to reduce the
interrupt load on the kernel.
There is usually an 'ethtool' setting to disable interrupt mitigation.
David
^ permalink raw reply
* [net-next.git 6/6] stmmac: update the driver version to Nov_2012
From: Giuseppe CAVALLARO @ 2012-11-26 9:10 UTC (permalink / raw)
To: netdev; +Cc: bhutchings, davem, Giuseppe Cavallaro
In-Reply-To: <1353921046-5121-1-git-send-email-peppe.cavallaro@st.com>
Many new feauture have been introduced in the driver:
ethtool coalesce options, Rx HW watchdog... so this patch updates the
driver's version.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index fe0535c..023a4fb 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -24,7 +24,7 @@
#define __STMMAC_H__
#define STMMAC_RESOURCE_NAME "stmmaceth"
-#define DRV_MODULE_VERSION "March_2012"
+#define DRV_MODULE_VERSION "Nov_2012"
#include <linux/clk.h>
#include <linux/stmmac.h>
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 4/6 (V2)] stmmac: get/set coalesce parameters via ethtool
From: Giuseppe CAVALLARO @ 2012-11-26 9:10 UTC (permalink / raw)
To: netdev; +Cc: bhutchings, davem, Giuseppe Cavallaro
In-Reply-To: <1353921046-5121-1-git-send-email-peppe.cavallaro@st.com>
This patch is to get/set the tx/rx coalesce parameters
via ethtool interface.
Tests have been done on several platform with different GMAC chips w/o and w/
RX watchdog feature.
V2: reject coalesce settings that are not supported.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Reviewed-by: Ben Hutchings <bhutchings@solarflare.com>
---
.../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 83 ++++++++++++++++++++
1 files changed, 83 insertions(+), 0 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index 240acd2..f2aa013 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -524,6 +524,87 @@ static int stmmac_ethtool_op_set_eee(struct net_device *dev,
return phy_ethtool_set_eee(priv->phydev, edata);
}
+static u32 stmmac_usec2riwt(u32 usec, struct stmmac_priv *priv)
+{
+ unsigned long clk = clk_get_rate(priv->stmmac_clk);
+
+ if (!clk)
+ return 0;
+
+ return (usec * (clk / 1000000)) / 256;
+}
+
+static u32 stmmac_riwt2usec(u32 riwt, struct stmmac_priv *priv)
+{
+ unsigned long clk = clk_get_rate(priv->stmmac_clk);
+
+ if (!clk)
+ return 0;
+
+ return (riwt * 256) / (clk / 1000000);
+}
+
+static int stmmac_get_coalesce(struct net_device *dev,
+ struct ethtool_coalesce *ec)
+{
+ struct stmmac_priv *priv = netdev_priv(dev);
+
+ ec->tx_coalesce_usecs = priv->tx_coal_timer;
+ ec->tx_max_coalesced_frames = priv->tx_coal_frames;
+
+ if (priv->use_riwt)
+ ec->rx_coalesce_usecs = stmmac_riwt2usec(priv->rx_riwt, priv);
+
+ return 0;
+}
+
+static int stmmac_set_coalesce(struct net_device *dev,
+ struct ethtool_coalesce *ec)
+{
+ struct stmmac_priv *priv = netdev_priv(dev);
+ unsigned int rx_riwt;
+
+ /* Check not supported parameters */
+ if ((ec->rx_max_coalesced_frames) || (ec->rx_coalesce_usecs_irq) ||
+ (ec->rx_max_coalesced_frames_irq) || (ec->tx_coalesce_usecs_irq) ||
+ (ec->use_adaptive_rx_coalesce) || (ec->use_adaptive_tx_coalesce) ||
+ (ec->pkt_rate_low) || (ec->rx_coalesce_usecs_low) ||
+ (ec->rx_max_coalesced_frames_low) || (ec->tx_coalesce_usecs_high) ||
+ (ec->tx_max_coalesced_frames_low) || (ec->pkt_rate_high) ||
+ (ec->tx_coalesce_usecs_low) || (ec->rx_coalesce_usecs_high) ||
+ (ec->rx_max_coalesced_frames_high) ||
+ (ec->tx_max_coalesced_frames_irq) ||
+ (ec->stats_block_coalesce_usecs) ||
+ (ec->tx_max_coalesced_frames_high) || (ec->rate_sample_interval))
+ return -EOPNOTSUPP;
+
+ if (ec->rx_coalesce_usecs == 0)
+ return -EINVAL;
+
+ if ((ec->tx_coalesce_usecs == 0) &&
+ (ec->tx_max_coalesced_frames == 0))
+ return -EINVAL;
+
+ if ((ec->tx_coalesce_usecs > STMMAC_COAL_TX_TIMER) ||
+ (ec->tx_max_coalesced_frames > STMMAC_TX_MAX_FRAMES))
+ return -EINVAL;
+
+ rx_riwt = stmmac_usec2riwt(ec->rx_coalesce_usecs, priv);
+
+ if ((rx_riwt > MAX_DMA_RIWT) || (rx_riwt < MIN_DMA_RIWT))
+ return -EINVAL;
+ else if (!priv->use_riwt)
+ return -EOPNOTSUPP;
+
+ /* Only copy relevant parameters, ignore all others. */
+ priv->tx_coal_frames = ec->tx_max_coalesced_frames;
+ priv->tx_coal_timer = ec->tx_coalesce_usecs;
+ priv->rx_riwt = rx_riwt;
+ priv->hw->dma->rx_watchdog(priv->ioaddr, priv->rx_riwt);
+
+ return 0;
+}
+
static const struct ethtool_ops stmmac_ethtool_ops = {
.begin = stmmac_check_if_running,
.get_drvinfo = stmmac_ethtool_getdrvinfo,
@@ -544,6 +625,8 @@ static const struct ethtool_ops stmmac_ethtool_ops = {
.set_eee = stmmac_ethtool_op_set_eee,
.get_sset_count = stmmac_get_sset_count,
.get_ts_info = ethtool_op_get_ts_info,
+ .get_coalesce = stmmac_get_coalesce,
+ .set_coalesce = stmmac_set_coalesce,
};
void stmmac_set_ethtool_ops(struct net_device *netdev)
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 5/6] stmmac: update the doc with new IRQ mitigation
From: Giuseppe CAVALLARO @ 2012-11-26 9:10 UTC (permalink / raw)
To: netdev; +Cc: bhutchings, davem, Giuseppe Cavallaro
In-Reply-To: <1353921046-5121-1-git-send-email-peppe.cavallaro@st.com>
This patch updates the stmmac.txt adding some information
about the new rx/tx mitigation schema adopted in the driver.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
Documentation/networking/stmmac.txt | 28 +++++++++++++++-------------
1 files changed, 15 insertions(+), 13 deletions(-)
diff --git a/Documentation/networking/stmmac.txt b/Documentation/networking/stmmac.txt
index ef9ee71..f9fa6db 100644
--- a/Documentation/networking/stmmac.txt
+++ b/Documentation/networking/stmmac.txt
@@ -29,11 +29,9 @@ The kernel configuration option is STMMAC_ETH:
dma_txsize: DMA tx ring size;
buf_sz: DMA buffer size;
tc: control the HW FIFO threshold;
- tx_coe: Enable/Disable Tx Checksum Offload engine;
watchdog: transmit timeout (in milliseconds);
flow_ctrl: Flow control ability [on/off];
pause: Flow Control Pause Time;
- tmrate: timer period (only if timer optimisation is configured).
3) Command line options
Driver parameters can be also passed in command line by using:
@@ -60,17 +58,19 @@ Then the poll method will be scheduled at some future point.
The incoming packets are stored, by the DMA, in a list of pre-allocated socket
buffers in order to avoid the memcpy (Zero-copy).
-4.3) Timer-Driver Interrupt
-Instead of having the device that asynchronously notifies the frame receptions,
-the driver configures a timer to generate an interrupt at regular intervals.
-Based on the granularity of the timer, the frames that are received by the
-device will experience different levels of latency. Some NICs have dedicated
-timer device to perform this task. STMMAC can use either the RTC device or the
-TMU channel 2 on STLinux platforms.
-The timers frequency can be passed to the driver as parameter; when change it,
-take care of both hardware capability and network stability/performance impact.
-Several performance tests on STM platforms showed this optimisation allows to
-spare the CPU while having the maximum throughput.
+4.3) Interrupt Mitigation
+The driver is able to mitigate the number of its DMA interrupts
+using NAPI for the reception on chips older than the 3.50.
+New chips have an HW RX-Watchdog used for this mitigation.
+
+On Tx-side, the mitigation schema is based on a SW timer that calls the
+tx function (stmmac_tx) to reclaim the resource after transmitting the
+frames.
+Also there is another parameter (like a threshold) used to program
+the descriptors avoiding to set the interrupt on completion bit in
+when the frame is sent (xmit).
+
+Mitigation parameters can be tuned by ethtool.
4.4) WOL
Wake up on Lan feature through Magic and Unicast frames are supported for the
@@ -121,6 +121,7 @@ struct plat_stmmacenet_data {
int bugged_jumbo;
int pmt;
int force_sf_dma_mode;
+ int riwt_off;
void (*fix_mac_speed)(void *priv, unsigned int speed);
void (*bus_setup)(void __iomem *ioaddr);
int (*init)(struct platform_device *pdev);
@@ -156,6 +157,7 @@ Where:
o pmt: core has the embedded power module (optional).
o force_sf_dma_mode: force DMA to use the Store and Forward mode
instead of the Threshold.
+ o riwt_off: force to disable the RX watchdog feature and switch to NAPI mode.
o fix_mac_speed: this callback is used for modifying some syscfg registers
(on ST SoCs) according to the link speed negotiated by the
physical layer .
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 2/6 (V3)] stmmac: add the initial tx coalesce schema
From: Giuseppe CAVALLARO @ 2012-11-26 9:10 UTC (permalink / raw)
To: netdev; +Cc: bhutchings, davem, Giuseppe Cavallaro
In-Reply-To: <1353921046-5121-1-git-send-email-peppe.cavallaro@st.com>
This patch adds a new schema used for mitigating the
number of transmit interrupts.
It is based on a SW timer and a threshold value.
The timer is used to periodically call the stmmac_tx_clean
function; the threshold is used for setting the IC (Interrupt
on Completion bit). The ISR will then invoke the poll method.
Also the patch improves some ethtool stat fields.
V2: review the logic to manage the IC bit in the TDESC
that was bugged because it didn't take care about the
fragments. Also fix the tx_count_frames that has not to be
limited to TX DMA ring. Thanks to Ben Hutchings.
V3: removed the spin_lock irqsave/restore as D. Miller suggested.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/common.h | 22 +++-
drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c | 7 +-
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 4 +
.../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 8 +-
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 132 ++++++++++++--------
5 files changed, 111 insertions(+), 62 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 719be39..d17eeef 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -95,9 +95,12 @@ struct stmmac_extra_stats {
unsigned long threshold;
unsigned long tx_pkt_n;
unsigned long rx_pkt_n;
- unsigned long poll_n;
- unsigned long sched_timer_n;
unsigned long normal_irq_n;
+ unsigned long rx_normal_irq_n;
+ unsigned long napi_poll;
+ unsigned long tx_normal_irq_n;
+ unsigned long tx_clean;
+ unsigned long tx_reset_ic_bit;
unsigned long mmc_tx_irq_n;
unsigned long mmc_rx_irq_n;
unsigned long mmc_rx_csum_offload_irq_n;
@@ -162,6 +165,12 @@ struct stmmac_extra_stats {
#define DMA_HW_FEAT_ACTPHYIF 0x70000000 /* Active/selected PHY interface */
#define DEFAULT_DMA_PBL 8
+/* Tx coalesce parameters */
+#define STMMAC_COAL_TX_TIMER 40000
+#define STMMAC_MAX_COAL_TX_TICK 100000
+#define STMMAC_TX_MAX_FRAMES 256
+#define STMMAC_TX_FRAMES 64
+
enum rx_frame_status { /* IPC status */
good_frame = 0,
discard_frame = 1,
@@ -169,10 +178,11 @@ enum rx_frame_status { /* IPC status */
llc_snap = 4,
};
-enum tx_dma_irq_status {
- tx_hard_error = 1,
- tx_hard_error_bump_tc = 2,
- handle_tx_rx = 3,
+enum dma_irq_status {
+ tx_hard_error = 0x1,
+ tx_hard_error_bump_tc = 0x2,
+ handle_rx = 0x4,
+ handle_tx = 0x8,
};
enum core_specific_irq_mask {
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
index 4e0e18a..73766e6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
@@ -206,9 +206,10 @@ int dwmac_dma_interrupt(void __iomem *ioaddr,
/* TX/RX NORMAL interrupts */
if (intr_status & DMA_STATUS_NIS) {
x->normal_irq_n++;
- if (likely((intr_status & DMA_STATUS_RI) ||
- (intr_status & (DMA_STATUS_TI))))
- ret = handle_tx_rx;
+ if (likely(intr_status & DMA_STATUS_RI))
+ ret |= handle_rx;
+ if (intr_status & (DMA_STATUS_TI))
+ ret |= handle_tx;
}
/* Optional hardware blocks, interrupts should be disabled */
if (unlikely(intr_status &
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 5f89415..01b3451 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -87,6 +87,10 @@ struct stmmac_priv {
int eee_enabled;
int eee_active;
int tx_lpi_timer;
+ struct timer_list txtimer;
+ u32 tx_count_frames;
+ u32 tx_coal_frames;
+ u32 tx_coal_timer;
};
extern int phyaddr;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index 76fd61a..de2210c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -90,10 +90,12 @@ static const struct stmmac_stats stmmac_gstrings_stats[] = {
STMMAC_STAT(threshold),
STMMAC_STAT(tx_pkt_n),
STMMAC_STAT(rx_pkt_n),
- STMMAC_STAT(poll_n),
- STMMAC_STAT(sched_timer_n),
- STMMAC_STAT(normal_irq_n),
STMMAC_STAT(normal_irq_n),
+ STMMAC_STAT(rx_normal_irq_n),
+ STMMAC_STAT(napi_poll),
+ STMMAC_STAT(tx_normal_irq_n),
+ STMMAC_STAT(tx_clean),
+ STMMAC_STAT(tx_reset_ic_bit),
STMMAC_STAT(mmc_tx_irq_n),
STMMAC_STAT(mmc_rx_irq_n),
STMMAC_STAT(mmc_rx_csum_offload_irq_n),
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 777234c..d40cf9e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -137,6 +137,8 @@ static int stmmac_init_fs(struct net_device *dev);
static void stmmac_exit_fs(void);
#endif
+#define STMMAC_COAL_TIMER(x) (jiffies + usecs_to_jiffies(x))
+
/**
* stmmac_verify_args - verify the driver parameters.
* Description: it verifies if some wrong parameter is passed to the driver.
@@ -688,16 +690,18 @@ static void stmmac_dma_operation_mode(struct stmmac_priv *priv)
}
/**
- * stmmac_tx:
- * @priv: private driver structure
+ * stmmac_tx_clean:
+ * @priv: private data pointer
* Description: it reclaims resources after transmission completes.
*/
-static void stmmac_tx(struct stmmac_priv *priv)
+static void stmmac_tx_clean(struct stmmac_priv *priv)
{
unsigned int txsize = priv->dma_tx_size;
spin_lock(&priv->tx_lock);
+ priv->xstats.tx_clean++;
+
while (priv->dirty_tx != priv->cur_tx) {
int last;
unsigned int entry = priv->dirty_tx % txsize;
@@ -757,40 +761,16 @@ static void stmmac_tx(struct stmmac_priv *priv)
spin_unlock(&priv->tx_lock);
}
-static inline void stmmac_enable_irq(struct stmmac_priv *priv)
+static inline void stmmac_enable_dma_irq(struct stmmac_priv *priv)
{
priv->hw->dma->enable_dma_irq(priv->ioaddr);
}
-static inline void stmmac_disable_irq(struct stmmac_priv *priv)
+static inline void stmmac_disable_dma_irq(struct stmmac_priv *priv)
{
priv->hw->dma->disable_dma_irq(priv->ioaddr);
}
-static int stmmac_has_work(struct stmmac_priv *priv)
-{
- unsigned int has_work = 0;
- int rxret, tx_work = 0;
-
- rxret = priv->hw->desc->get_rx_owner(priv->dma_rx +
- (priv->cur_rx % priv->dma_rx_size));
-
- if (priv->dirty_tx != priv->cur_tx)
- tx_work = 1;
-
- if (likely(!rxret || tx_work))
- has_work = 1;
-
- return has_work;
-}
-
-static inline void _stmmac_schedule(struct stmmac_priv *priv)
-{
- if (likely(stmmac_has_work(priv))) {
- stmmac_disable_irq(priv);
- napi_schedule(&priv->napi);
- }
-}
/**
* stmmac_tx_err:
@@ -813,16 +793,18 @@ static void stmmac_tx_err(struct stmmac_priv *priv)
netif_wake_queue(priv->dev);
}
-
static void stmmac_dma_interrupt(struct stmmac_priv *priv)
{
int status;
status = priv->hw->dma->dma_interrupt(priv->ioaddr, &priv->xstats);
- if (likely(status == handle_tx_rx))
- _stmmac_schedule(priv);
-
- else if (unlikely(status == tx_hard_error_bump_tc)) {
+ if (likely((status & handle_rx)) || (status & handle_tx)) {
+ if (likely(napi_schedule_prep(&priv->napi))) {
+ stmmac_disable_dma_irq(priv);
+ __napi_schedule(&priv->napi);
+ }
+ }
+ if (unlikely(status & tx_hard_error_bump_tc)) {
/* Try to bump up the dma threshold on this failure */
if (unlikely(tc != SF_DMA_MODE) && (tc <= 256)) {
tc += 64;
@@ -938,7 +920,6 @@ static int stmmac_get_hw_features(struct stmmac_priv *priv)
/* Alternate (enhanced) DESC mode*/
priv->dma_cap.enh_desc =
(hw_cap & DMA_HW_FEAT_ENHDESSEL) >> 24;
-
}
return hw_cap;
@@ -980,6 +961,38 @@ static int stmmac_init_dma_engine(struct stmmac_priv *priv)
}
/**
+ * stmmac_tx_timer:
+ * @data: data pointer
+ * Description:
+ * This is the timer handler to directly invoke the stmmac_tx_clean.
+ */
+static void stmmac_tx_timer(unsigned long data)
+{
+ struct stmmac_priv *priv = (struct stmmac_priv *)data;
+
+ stmmac_tx_clean(priv);
+}
+
+/**
+ * stmmac_tx_timer:
+ * @priv: private data structure
+ * Description:
+ * This inits the transmit coalesce parameters: i.e. timer rate,
+ * timer handler and default threshold used for enabling the
+ * interrupt on completion bit.
+ */
+static void stmmac_init_tx_coalesce(struct stmmac_priv *priv)
+{
+ priv->tx_coal_frames = STMMAC_TX_FRAMES;
+ priv->tx_coal_timer = STMMAC_COAL_TX_TIMER;
+ init_timer(&priv->txtimer);
+ priv->txtimer.expires = STMMAC_COAL_TIMER(priv->tx_coal_timer);
+ priv->txtimer.data = (unsigned long)priv;
+ priv->txtimer.function = stmmac_tx_timer;
+ add_timer(&priv->txtimer);
+}
+
+/**
* stmmac_open - open entry point of the driver
* @dev : pointer to the device structure.
* Description:
@@ -1091,6 +1104,8 @@ static int stmmac_open(struct net_device *dev)
priv->tx_lpi_timer = STMMAC_DEFAULT_TWT_LS_TIMER;
priv->eee_enabled = stmmac_eee_init(priv);
+ stmmac_init_tx_coalesce(priv);
+
napi_enable(&priv->napi);
netif_start_queue(dev);
@@ -1136,6 +1151,8 @@ static int stmmac_release(struct net_device *dev)
napi_disable(&priv->napi);
+ del_timer_sync(&priv->txtimer);
+
/* Free the IRQ lines */
free_irq(dev->irq, dev);
if (priv->wol_irq != dev->irq)
@@ -1198,11 +1215,13 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
#ifdef STMMAC_XMIT_DEBUG
if ((skb->len > ETH_FRAME_LEN) || nfrags)
- pr_info("stmmac xmit:\n"
- "\tskb addr %p - len: %d - nopaged_len: %d\n"
- "\tn_frags: %d - ip_summed: %d - %s gso\n",
- skb, skb->len, nopaged_len, nfrags, skb->ip_summed,
- !skb_is_gso(skb) ? "isn't" : "is");
+ pr_debug("stmmac xmit: [entry %d]\n"
+ "\tskb addr %p - len: %d - nopaged_len: %d\n"
+ "\tn_frags: %d - ip_summed: %d - %s gso\n"
+ "\ttx_count_frames %d\n", entry,
+ skb, skb->len, nopaged_len, nfrags, skb->ip_summed,
+ !skb_is_gso(skb) ? "isn't" : "is",
+ priv->tx_count_frames);
#endif
csum_insertion = (skb->ip_summed == CHECKSUM_PARTIAL);
@@ -1212,9 +1231,9 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
#ifdef STMMAC_XMIT_DEBUG
if ((nfrags > 0) || (skb->len > ETH_FRAME_LEN))
- pr_debug("stmmac xmit: skb len: %d, nopaged_len: %d,\n"
- "\t\tn_frags: %d, ip_summed: %d\n",
- skb->len, nopaged_len, nfrags, skb->ip_summed);
+ pr_debug("\tskb len: %d, nopaged_len: %d,\n"
+ "\t\tn_frags: %d, ip_summed: %d\n",
+ skb->len, nopaged_len, nfrags, skb->ip_summed);
#endif
priv->tx_skbuff[entry] = skb;
@@ -1245,10 +1264,24 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
wmb();
}
- /* Interrupt on completition only for the latest segment */
+ /* Finalize the latest segment. */
priv->hw->desc->close_tx_desc(desc);
wmb();
+ /* According to the coalesce parameter the IC bit for the latest
+ * segment could be reset and the timer re-started to invoke the
+ * stmmac_tx function. This approach takes care about the fragments.
+ */
+ priv->tx_count_frames += nfrags + 1;
+ if (priv->tx_coal_frames > priv->tx_count_frames) {
+ priv->hw->desc->clear_tx_ic(desc);
+ priv->xstats.tx_reset_ic_bit++;
+ TX_DBG("\t[entry %d]: tx_count_frames %d\n", entry,
+ priv->tx_count_frames);
+ mod_timer(&priv->txtimer,
+ STMMAC_COAL_TIMER(priv->tx_coal_timer));
+ } else
+ priv->tx_count_frames = 0;
/* To avoid raise condition */
priv->hw->desc->set_tx_owner(first);
@@ -1419,21 +1452,20 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
* @budget : maximum number of packets that the current CPU can receive from
* all interfaces.
* Description :
- * This function implements the the reception process.
- * Also it runs the TX completion thread
+ * To look at the incoming frames and clear the tx resources.
*/
static int stmmac_poll(struct napi_struct *napi, int budget)
{
struct stmmac_priv *priv = container_of(napi, struct stmmac_priv, napi);
int work_done = 0;
- priv->xstats.poll_n++;
- stmmac_tx(priv);
- work_done = stmmac_rx(priv, budget);
+ priv->xstats.napi_poll++;
+ stmmac_tx_clean(priv);
+ work_done = stmmac_rx(priv, budget);
if (work_done < budget) {
napi_complete(napi);
- stmmac_enable_irq(priv);
+ stmmac_enable_dma_irq(priv);
}
return work_done;
}
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 0/6 (V5)] stmmac: remove dead code for STMMAC_TIMER and add new mitigation schema
From: Giuseppe CAVALLARO @ 2012-11-26 9:10 UTC (permalink / raw)
To: netdev; +Cc: bhutchings, davem, Giuseppe Cavallaro
These patch series remove the STMMAC_TIMER option no longer updated
and never used and add a new mitigation schema.
Having removed the Timer opt, this has made the driver slim.
On top of this work, it has been easier to introduce the new
mitigation schema based on HW RX-watchdog (available in new cores).
In fact, 3.50 and newer cores have an HW RX-Watchdog that can be used for
mitigating the Rx-interrupts and first results look promising.
Running n-u-t-t-c-p with the following parameters:
Throughput: 500Mbps
UDP Buffer size: 1328bytes
TCP Buffer size: 65536bytes
for example, I got on ST box (arm-based) these improvements:
--------------------------------------------------------------------
Original | With New Mitigation patch
--------------------------------------------------------------------
Test CPU usage pkt/loss | CPU usage pkt/loss
Type Mbps % % |Mbps % %
--------------------------------------------------------------------
UDP-RX 395.5065 95 20.89 |499.8811 23 0.00
UDP-TX 499.5578 100 0.08915 |499.1340 99 0.00
TCP-RX 499.9221 77 |499.8776 26
TCP-TX 389.5719 99 |499.9771 79
--------------------------------------------------------------------
... no regression on ST boxes (SH based) I always test.
This is a brief explanation of the new mitigation schema although there
is a patch that updates the driver's documentation.
o On Rx-side I have:
New GMACs will use the RX-watchdog timer; old ones will continue to
use NAPI to mitigate the RX DMA interrupts.
For the RX-watchdog, there is a parameter that is the RI Watchdog
Timer count. It indicates the number of system clock cycles and can be
set via *ethtool*.
o On Tx-side, the mitigation schema is based on a SW timer
that calls the tx function (stmmac_tx) to reclaim the resource after
transmitting the frames.
Also there is another parameter (a threshold) used to program
the descriptors avoiding to set the interrupt on completion bit in
when the frame is sent (xmit).
V2: these patches add the ethtool support to get/set coalesce parameters
and totally remove the sysFS support added in the first patches.
V3: added several fixes: for example NAPI and RX-watchdog work together
while in the previous implementation the HW RX-watchdog disabled NAPI.
On the tx side, erroneously the tx coalesce frame parameter was limited
to the ring size and the driver didn't take care of the segment numbers
when enable/disable the IC bit in the TDES.
V4: reject not supported coalesce settings in ethtool.
V5: in these new patch series, I am releasing two patches that
add the mitigation for tx and rx where I collected the feedback received
from the mailing list. Concerning the tx, I've also reworked and
fixed the spinlock irqsave and restore as D. Miller had suggested (thx).
Giuseppe Cavallaro (6):
stmmac: remove dead code for STMMAC_TIMER support
stmmac: add the initial tx coalesce schema
stmmac: add Rx watchdog support to mitigate the DMA irqs
stmmac: get/set coalesce parameters via ethtool
stmmac: update the doc with new IRQ mitigation
stmmac: update the driver version to Nov_2012
Documentation/networking/stmmac.txt | 28 ++-
drivers/net/ethernet/stmicro/stmmac/Kconfig | 25 --
drivers/net/ethernet/stmicro/stmmac/Makefile | 1 -
drivers/net/ethernet/stmicro/stmmac/common.h | 39 +++-
drivers/net/ethernet/stmicro/stmmac/dwmac1000.h | 3 -
.../net/ethernet/stmicro/stmmac/dwmac1000_dma.c | 6 +
drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h | 5 +-
drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c | 20 ++-
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 14 +-
.../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 100 +++++++-
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 261 ++++++++------------
drivers/net/ethernet/stmicro/stmmac/stmmac_timer.c | 134 ----------
drivers/net/ethernet/stmicro/stmmac/stmmac_timer.h | 46 ----
include/linux/stmmac.h | 1 +
14 files changed, 281 insertions(+), 402 deletions(-)
delete mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_timer.c
delete mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_timer.h
--
1.7.4.4
^ permalink raw reply
* [net-next.git 3/6 (V4)] stmmac: add Rx watchdog support to mitigate the DMA irqs
From: Giuseppe CAVALLARO @ 2012-11-26 9:10 UTC (permalink / raw)
To: netdev; +Cc: bhutchings, davem, Giuseppe Cavallaro
In-Reply-To: <1353921046-5121-1-git-send-email-peppe.cavallaro@st.com>
GMAC devices newer than databook 3.40 has an embedded timer
that can be used for mitigating the number of interrupts.
So this patch adds this optimizations.
At any rate, the Rx watchdog can be disable (on bugged HW) by
passing from the platform the riwt_off field.
In this implementation the rx timer stored in the Reg9 is fixed
to the max value. This will be tuned by using ethtool.
V2: added a platform parameter to force to disable the rx-watchdog
for example on new core where it is bugged.
V3: do not disable NAPI when Rx watchdog is used.
V4: a new extra statistic field has been added to show the early
receive status in the interrupt handler.
This patch also adds an extra check to avoid to call
napi_schedule when the DMA_INTR_ENA_RIE bit is disabled in the
Interrupt Mask register.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/common.h | 17 +++++++++--
drivers/net/ethernet/stmicro/stmmac/dwmac1000.h | 3 --
.../net/ethernet/stmicro/stmmac/dwmac1000_dma.c | 6 ++++
drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h | 5 ++-
drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c | 19 ++++++++++--
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 +
.../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 9 ++++--
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 30 ++++++++++++++++----
include/linux/stmmac.h | 1 +
9 files changed, 72 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index d17eeef..98203f6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -48,6 +48,10 @@
#define CHIP_DBG(fmt, args...) do { } while (0)
#endif
+/* Synopsys Core versions */
+#define DWMAC_CORE_3_40 0x34
+#define DWMAC_CORE_3_50 0x35
+
#undef FRAME_FILTER_DEBUG
/* #define FRAME_FILTER_DEBUG */
@@ -81,7 +85,7 @@ struct stmmac_extra_stats {
unsigned long rx_missed_cntr;
unsigned long rx_overflow_cntr;
unsigned long rx_vlan;
- /* Tx/Rx IRQ errors */
+ /* Tx/Rx IRQ error info */
unsigned long tx_undeflow_irq;
unsigned long tx_process_stopped_irq;
unsigned long tx_jabber_irq;
@@ -91,7 +95,8 @@ struct stmmac_extra_stats {
unsigned long rx_watchdog_irq;
unsigned long tx_early_irq;
unsigned long fatal_bus_error_irq;
- /* Extra info */
+ /* Tx/Rx IRQ Events */
+ unsigned long rx_early_irq;
unsigned long threshold;
unsigned long tx_pkt_n;
unsigned long rx_pkt_n;
@@ -101,11 +106,12 @@ struct stmmac_extra_stats {
unsigned long tx_normal_irq_n;
unsigned long tx_clean;
unsigned long tx_reset_ic_bit;
+ unsigned long irq_receive_pmt_irq_n;
+ /* MMC info */
unsigned long mmc_tx_irq_n;
unsigned long mmc_rx_irq_n;
unsigned long mmc_rx_csum_offload_irq_n;
/* EEE */
- unsigned long irq_receive_pmt_irq_n;
unsigned long irq_tx_path_in_lpi_mode_n;
unsigned long irq_tx_path_exit_lpi_mode_n;
unsigned long irq_rx_path_in_lpi_mode_n;
@@ -165,6 +171,9 @@ struct stmmac_extra_stats {
#define DMA_HW_FEAT_ACTPHYIF 0x70000000 /* Active/selected PHY interface */
#define DEFAULT_DMA_PBL 8
+/* Max/Min RI Watchdog Timer count value */
+#define MAX_DMA_RIWT 0xff
+#define MIN_DMA_RIWT 0x20
/* Tx coalesce parameters */
#define STMMAC_COAL_TX_TIMER 40000
#define STMMAC_MAX_COAL_TX_TICK 100000
@@ -306,6 +315,8 @@ struct stmmac_dma_ops {
struct stmmac_extra_stats *x);
/* If supported then get the optional core features */
unsigned int (*get_hw_feature) (void __iomem *ioaddr);
+ /* Program the HW RX Watchdog */
+ void (*rx_watchdog) (void __iomem *ioaddr, u32 riwt);
};
struct stmmac_ops {
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
index 0e4cace..7ad56af 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h
@@ -230,8 +230,5 @@ enum rtc_control {
#define GMAC_MMC_TX_INTR 0x108
#define GMAC_MMC_RX_CSUM_OFFLOAD 0x208
-/* Synopsys Core versions */
-#define DWMAC_CORE_3_40 0x34
-
extern const struct stmmac_dma_ops dwmac1000_dma_ops;
#endif /* __DWMAC1000_H__ */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
index 0335000..bf83c03 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
@@ -174,6 +174,11 @@ static unsigned int dwmac1000_get_hw_feature(void __iomem *ioaddr)
return readl(ioaddr + DMA_HW_FEATURE);
}
+static void dwmac1000_rx_watchdog(void __iomem *ioaddr, u32 riwt)
+{
+ writel(riwt, ioaddr + DMA_RX_WATCHDOG);
+}
+
const struct stmmac_dma_ops dwmac1000_dma_ops = {
.init = dwmac1000_dma_init,
.dump_regs = dwmac1000_dump_dma_regs,
@@ -187,4 +192,5 @@ const struct stmmac_dma_ops dwmac1000_dma_ops = {
.stop_rx = dwmac_dma_stop_rx,
.dma_interrupt = dwmac_dma_interrupt,
.get_hw_feature = dwmac1000_get_hw_feature,
+ .rx_watchdog = dwmac1000_rx_watchdog,
};
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
index e49c9a0..807f303 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_dma.h
@@ -35,7 +35,10 @@
#define DMA_CONTROL 0x00001018 /* Ctrl (Operational Mode) */
#define DMA_INTR_ENA 0x0000101c /* Interrupt Enable */
#define DMA_MISSED_FRAME_CTR 0x00001020 /* Missed Frame Counter */
-#define DMA_AXI_BUS_MODE 0x00001028 /* AXI Bus Mode */
+/* Rx watchdog register */
+#define DMA_RX_WATCHDOG 0x00001024
+/* AXI Bus Mode */
+#define DMA_AXI_BUS_MODE 0x00001028
#define DMA_CUR_TX_BUF_ADDR 0x00001050 /* Current Host Tx Buffer */
#define DMA_CUR_RX_BUF_ADDR 0x00001054 /* Current Host Rx Buffer */
#define DMA_HW_FEATURE 0x00001058 /* HW Feature Register */
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
index 73766e6..491d7e9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac_lib.c
@@ -204,17 +204,28 @@ int dwmac_dma_interrupt(void __iomem *ioaddr,
}
}
/* TX/RX NORMAL interrupts */
- if (intr_status & DMA_STATUS_NIS) {
+ if (likely(intr_status & DMA_STATUS_NIS)) {
x->normal_irq_n++;
- if (likely(intr_status & DMA_STATUS_RI))
- ret |= handle_rx;
- if (intr_status & (DMA_STATUS_TI))
+ if (likely(intr_status & DMA_STATUS_RI)) {
+ u32 value = readl(ioaddr + DMA_INTR_ENA);
+ /* to schedule NAPI on real RIE event. */
+ if (likely(value & DMA_INTR_ENA_RIE)) {
+ x->rx_normal_irq_n++;
+ ret |= handle_rx;
+ }
+ }
+ if (likely(intr_status & DMA_STATUS_TI)) {
+ x->tx_normal_irq_n++;
ret |= handle_tx;
+ }
+ if (unlikely(intr_status & DMA_STATUS_ERI))
+ x->rx_early_irq++;
}
/* Optional hardware blocks, interrupts should be disabled */
if (unlikely(intr_status &
(DMA_STATUS_GPI | DMA_STATUS_GMI | DMA_STATUS_GLI)))
pr_info("%s: unexpected status %08x\n", __func__, intr_status);
+
/* Clear the interrupt by writing a logic 1 to the CSR5[15-0] */
writel((intr_status & 0x1ffff), ioaddr + DMA_STATUS);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 01b3451..fe0535c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -91,6 +91,8 @@ struct stmmac_priv {
u32 tx_count_frames;
u32 tx_coal_frames;
u32 tx_coal_timer;
+ int use_riwt;
+ u32 rx_riwt;
};
extern int phyaddr;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index de2210c..240acd2 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -76,7 +76,7 @@ static const struct stmmac_stats stmmac_gstrings_stats[] = {
STMMAC_STAT(rx_missed_cntr),
STMMAC_STAT(rx_overflow_cntr),
STMMAC_STAT(rx_vlan),
- /* Tx/Rx IRQ errors */
+ /* Tx/Rx IRQ error info */
STMMAC_STAT(tx_undeflow_irq),
STMMAC_STAT(tx_process_stopped_irq),
STMMAC_STAT(tx_jabber_irq),
@@ -86,7 +86,8 @@ static const struct stmmac_stats stmmac_gstrings_stats[] = {
STMMAC_STAT(rx_watchdog_irq),
STMMAC_STAT(tx_early_irq),
STMMAC_STAT(fatal_bus_error_irq),
- /* Extra info */
+ /* Tx/Rx IRQ Events */
+ STMMAC_STAT(rx_early_irq),
STMMAC_STAT(threshold),
STMMAC_STAT(tx_pkt_n),
STMMAC_STAT(rx_pkt_n),
@@ -96,10 +97,12 @@ static const struct stmmac_stats stmmac_gstrings_stats[] = {
STMMAC_STAT(tx_normal_irq_n),
STMMAC_STAT(tx_clean),
STMMAC_STAT(tx_reset_ic_bit),
+ STMMAC_STAT(irq_receive_pmt_irq_n),
+ /* MMC info */
STMMAC_STAT(mmc_tx_irq_n),
STMMAC_STAT(mmc_rx_irq_n),
STMMAC_STAT(mmc_rx_csum_offload_irq_n),
- STMMAC_STAT(irq_receive_pmt_irq_n),
+ /* EEE */
STMMAC_STAT(irq_tx_path_in_lpi_mode_n),
STMMAC_STAT(irq_tx_path_exit_lpi_mode_n),
STMMAC_STAT(irq_rx_path_in_lpi_mode_n),
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index d40cf9e..2fb0e5f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -603,6 +603,8 @@ static void init_dma_desc_rings(struct net_device *dev)
priv->dirty_tx = 0;
priv->cur_tx = 0;
+ if (priv->use_riwt)
+ dis_ic = 1;
/* Clear the Rx/Tx descriptors */
priv->hw->desc->init_rx_desc(priv->dma_rx, rxsize, dis_ic);
priv->hw->desc->init_tx_desc(priv->dma_tx, txsize);
@@ -1106,6 +1108,11 @@ static int stmmac_open(struct net_device *dev)
stmmac_init_tx_coalesce(priv);
+ if ((priv->use_riwt) && (priv->hw->dma->rx_watchdog)) {
+ priv->rx_riwt = MAX_DMA_RIWT;
+ priv->hw->dma->rx_watchdog(priv->ioaddr, MAX_DMA_RIWT);
+ }
+
napi_enable(&priv->napi);
netif_start_queue(dev);
@@ -1423,14 +1430,12 @@ static int stmmac_rx(struct stmmac_priv *priv, int limit)
#endif
skb->protocol = eth_type_trans(skb, priv->dev);
- if (unlikely(!priv->plat->rx_coe)) {
- /* No RX COE for old mac10/100 devices */
+ if (unlikely(!priv->plat->rx_coe))
skb_checksum_none_assert(skb);
- netif_receive_skb(skb);
- } else {
+ else
skb->ip_summed = CHECKSUM_UNNECESSARY;
- napi_gro_receive(&priv->napi, skb);
- }
+
+ napi_gro_receive(&priv->napi, skb);
priv->dev->stats.rx_packets++;
priv->dev->stats.rx_bytes += frame_len;
@@ -2001,6 +2006,16 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device,
if (flow_ctrl)
priv->flow_ctrl = FLOW_AUTO; /* RX/TX pause on */
+ /* Rx Watchdog is available in the COREs newer than the 3.40.
+ * In some case, for example on bugged HW this feature
+ * has to be disable and this can be done by passing the
+ * riwt_off field from the platform.
+ */
+ if ((priv->synopsys_id >= DWMAC_CORE_3_50) && (!priv->plat->riwt_off)) {
+ priv->use_riwt = 1;
+ pr_info(" Enable RX Mitigation via HW Watchdog Timer\n");
+ }
+
netif_napi_add(ndev, &priv->napi, stmmac_poll, 64);
spin_lock_init(&priv->lock);
@@ -2092,6 +2107,9 @@ int stmmac_suspend(struct net_device *ndev)
netif_device_detach(ndev);
netif_stop_queue(ndev);
+ if (priv->use_riwt)
+ dis_ic = 1;
+
napi_disable(&priv->napi);
/* Stop TX/RX DMA */
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index a1547ea..de5b2f8 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -104,6 +104,7 @@ struct plat_stmmacenet_data {
int bugged_jumbo;
int pmt;
int force_sf_dma_mode;
+ int riwt_off;
void (*fix_mac_speed)(void *priv, unsigned int speed);
void (*bus_setup)(void __iomem *ioaddr);
int (*init)(struct platform_device *pdev);
--
1.7.4.4
^ permalink raw reply related
* [net-next.git 1/6] stmmac: remove dead code for STMMAC_TIMER support
From: Giuseppe CAVALLARO @ 2012-11-26 9:10 UTC (permalink / raw)
To: netdev; +Cc: bhutchings, davem, Giuseppe Cavallaro
In-Reply-To: <1353921046-5121-1-git-send-email-peppe.cavallaro@st.com>
The TIMER option is not longer supported and this
code can be considered dead for this driver in
the new kernel series.
In fact, It was not updated at all and never used.
Signed-off-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
---
drivers/net/ethernet/stmicro/stmmac/Kconfig | 25 ----
drivers/net/ethernet/stmicro/stmmac/Makefile | 1 -
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 6 -
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 101 +--------------
drivers/net/ethernet/stmicro/stmmac/stmmac_timer.c | 134 --------------------
drivers/net/ethernet/stmicro/stmmac/stmmac_timer.h | 46 -------
6 files changed, 3 insertions(+), 310 deletions(-)
delete mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_timer.c
delete mode 100644 drivers/net/ethernet/stmicro/stmmac/stmmac_timer.h
diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig
index 9f44827..1164930 100644
--- a/drivers/net/ethernet/stmicro/stmmac/Kconfig
+++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig
@@ -54,31 +54,6 @@ config STMMAC_DA
By default, the DMA arbitration scheme is based on Round-robin
(rx:tx priority is 1:1).
-config STMMAC_TIMER
- bool "STMMAC Timer optimisation"
- default n
- depends on RTC_HCTOSYS_DEVICE
- ---help---
- Use an external timer for mitigating the number of network
- interrupts. Currently, for SH architectures, it is possible
- to use the TMU channel 2 and the SH-RTC device.
-
-choice
- prompt "Select Timer device"
- depends on STMMAC_TIMER
-
-config STMMAC_TMU_TIMER
- bool "TMU channel 2"
- depends on CPU_SH4
- ---help---
-
-config STMMAC_RTC_TIMER
- bool "Real time clock"
- depends on RTC_CLASS
- ---help---
-
-endchoice
-
choice
prompt "Select the DMA TX/RX descriptor operating modes"
depends on STMMAC_ETH
diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile
index bc965ac..c8e8ea6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/Makefile
+++ b/drivers/net/ethernet/stmicro/stmmac/Makefile
@@ -1,5 +1,4 @@
obj-$(CONFIG_STMMAC_ETH) += stmmac.o
-stmmac-$(CONFIG_STMMAC_TIMER) += stmmac_timer.o
stmmac-$(CONFIG_STMMAC_RING) += ring_mode.o
stmmac-$(CONFIG_STMMAC_CHAINED) += chain_mode.o
stmmac-$(CONFIG_STMMAC_PLATFORM) += stmmac_platform.o
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index 7d51a65..5f89415 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -31,9 +31,6 @@
#include <linux/phy.h>
#include <linux/pci.h>
#include "common.h"
-#ifdef CONFIG_STMMAC_TIMER
-#include "stmmac_timer.h"
-#endif
struct stmmac_priv {
/* Frequently used values are kept adjacent for cache effect */
@@ -77,9 +74,6 @@ struct stmmac_priv {
spinlock_t tx_lock;
int wolopts;
int wol_irq;
-#ifdef CONFIG_STMMAC_TIMER
- struct stmmac_timer *tm;
-#endif
struct plat_stmmacenet_data *plat;
struct stmmac_counters mmc;
struct dma_features dma_cap;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index c6cdbc4..777234c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -115,16 +115,6 @@ static int tc = TC_DEFAULT;
module_param(tc, int, S_IRUGO | S_IWUSR);
MODULE_PARM_DESC(tc, "DMA threshold control value");
-/* Pay attention to tune this parameter; take care of both
- * hardware capability and network stabitily/performance impact.
- * Many tests showed that ~4ms latency seems to be good enough. */
-#ifdef CONFIG_STMMAC_TIMER
-#define DEFAULT_PERIODIC_RATE 256
-static int tmrate = DEFAULT_PERIODIC_RATE;
-module_param(tmrate, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(tmrate, "External timer freq. (default: 256Hz)");
-#endif
-
#define DMA_BUFFER_SIZE BUF_SIZE_2KiB
static int buf_sz = DMA_BUFFER_SIZE;
module_param(buf_sz, int, S_IRUGO | S_IWUSR);
@@ -536,12 +526,6 @@ static void init_dma_desc_rings(struct net_device *dev)
else
bfsize = stmmac_set_bfsize(dev->mtu, priv->dma_buf_sz);
-#ifdef CONFIG_STMMAC_TIMER
- /* Disable interrupts on completion for the reception if timer is on */
- if (likely(priv->tm->enable))
- dis_ic = 1;
-#endif
-
DBG(probe, INFO, "stmmac: txsize %d, rxsize %d, bfsize %d\n",
txsize, rxsize, bfsize);
@@ -775,22 +759,12 @@ static void stmmac_tx(struct stmmac_priv *priv)
static inline void stmmac_enable_irq(struct stmmac_priv *priv)
{
-#ifdef CONFIG_STMMAC_TIMER
- if (likely(priv->tm->enable))
- priv->tm->timer_start(tmrate);
- else
-#endif
- priv->hw->dma->enable_dma_irq(priv->ioaddr);
+ priv->hw->dma->enable_dma_irq(priv->ioaddr);
}
static inline void stmmac_disable_irq(struct stmmac_priv *priv)
{
-#ifdef CONFIG_STMMAC_TIMER
- if (likely(priv->tm->enable))
- priv->tm->timer_stop();
- else
-#endif
- priv->hw->dma->disable_dma_irq(priv->ioaddr);
+ priv->hw->dma->disable_dma_irq(priv->ioaddr);
}
static int stmmac_has_work(struct stmmac_priv *priv)
@@ -818,25 +792,6 @@ static inline void _stmmac_schedule(struct stmmac_priv *priv)
}
}
-#ifdef CONFIG_STMMAC_TIMER
-void stmmac_schedule(struct net_device *dev)
-{
- struct stmmac_priv *priv = netdev_priv(dev);
-
- priv->xstats.sched_timer_n++;
-
- _stmmac_schedule(priv);
-}
-
-static void stmmac_no_timer_started(unsigned int x)
-{;
-};
-
-static void stmmac_no_timer_stopped(void)
-{;
-};
-#endif
-
/**
* stmmac_tx_err:
* @priv: pointer to the private device structure
@@ -1038,23 +993,6 @@ static int stmmac_open(struct net_device *dev)
struct stmmac_priv *priv = netdev_priv(dev);
int ret;
-#ifdef CONFIG_STMMAC_TIMER
- priv->tm = kzalloc(sizeof(struct stmmac_timer *), GFP_KERNEL);
- if (unlikely(priv->tm == NULL))
- return -ENOMEM;
-
- priv->tm->freq = tmrate;
-
- /* Test if the external timer can be actually used.
- * In case of failure continue without timer. */
- if (unlikely((stmmac_open_ext_timer(dev, priv->tm)) < 0)) {
- pr_warning("stmmaceth: cannot attach the external timer.\n");
- priv->tm->freq = 0;
- priv->tm->timer_start = stmmac_no_timer_started;
- priv->tm->timer_stop = stmmac_no_timer_stopped;
- } else
- priv->tm->enable = 1;
-#endif
clk_prepare_enable(priv->stmmac_clk);
stmmac_check_ether_addr(priv);
@@ -1141,10 +1079,6 @@ static int stmmac_open(struct net_device *dev)
priv->hw->dma->start_tx(priv->ioaddr);
priv->hw->dma->start_rx(priv->ioaddr);
-#ifdef CONFIG_STMMAC_TIMER
- priv->tm->timer_start(tmrate);
-#endif
-
/* Dump DMA/MAC registers */
if (netif_msg_hw(priv)) {
priv->hw->mac->dump_regs(priv->ioaddr);
@@ -1170,9 +1104,6 @@ open_error_wolirq:
free_irq(dev->irq, dev);
open_error:
-#ifdef CONFIG_STMMAC_TIMER
- kfree(priv->tm);
-#endif
if (priv->phydev)
phy_disconnect(priv->phydev);
@@ -1203,12 +1134,6 @@ static int stmmac_release(struct net_device *dev)
netif_stop_queue(dev);
-#ifdef CONFIG_STMMAC_TIMER
- /* Stop and release the timer */
- stmmac_close_ext_timer();
- if (priv->tm != NULL)
- kfree(priv->tm);
-#endif
napi_disable(&priv->napi);
/* Free the IRQ lines */
@@ -1323,12 +1248,6 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
/* Interrupt on completition only for the latest segment */
priv->hw->desc->close_tx_desc(desc);
-#ifdef CONFIG_STMMAC_TIMER
- /* Clean IC while using timer */
- if (likely(priv->tm->enable))
- priv->hw->desc->clear_tx_ic(desc);
-#endif
-
wmb();
/* To avoid raise condition */
@@ -1523,7 +1442,7 @@ static int stmmac_poll(struct napi_struct *napi, int budget)
* stmmac_tx_timeout
* @dev : Pointer to net device structure
* Description: this function is called when a packet transmission fails to
- * complete within a reasonable tmrate. The driver will mark the error in the
+ * complete within a reasonable time. The driver will mark the error in the
* netdev structure and arrange for the device to be reset to a sane state
* in order to transmit a new packet.
*/
@@ -2141,11 +2060,6 @@ int stmmac_suspend(struct net_device *ndev)
netif_device_detach(ndev);
netif_stop_queue(ndev);
-#ifdef CONFIG_STMMAC_TIMER
- priv->tm->timer_stop();
- if (likely(priv->tm->enable))
- dis_ic = 1;
-#endif
napi_disable(&priv->napi);
/* Stop TX/RX DMA */
@@ -2196,10 +2110,6 @@ int stmmac_resume(struct net_device *ndev)
priv->hw->dma->start_tx(priv->ioaddr);
priv->hw->dma->start_rx(priv->ioaddr);
-#ifdef CONFIG_STMMAC_TIMER
- if (likely(priv->tm->enable))
- priv->tm->timer_start(tmrate);
-#endif
napi_enable(&priv->napi);
netif_start_queue(ndev);
@@ -2295,11 +2205,6 @@ static int __init stmmac_cmdline_opt(char *str)
} else if (!strncmp(opt, "eee_timer:", 6)) {
if (kstrtoint(opt + 10, 0, &eee_timer))
goto err;
-#ifdef CONFIG_STMMAC_TIMER
- } else if (!strncmp(opt, "tmrate:", 7)) {
- if (kstrtoint(opt + 7, 0, &tmrate))
- goto err;
-#endif
}
}
return 0;
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_timer.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_timer.c
deleted file mode 100644
index 4ccd4e2..0000000
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_timer.c
+++ /dev/null
@@ -1,134 +0,0 @@
-/*******************************************************************************
- STMMAC external timer support.
-
- Copyright (C) 2007-2009 STMicroelectronics Ltd
-
- This program is free software; you can redistribute it and/or modify it
- under the terms and conditions of the GNU General Public License,
- version 2, as published by the Free Software Foundation.
-
- This program is distributed in the hope it will be useful, but WITHOUT
- ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
- more details.
-
- You should have received a copy of the GNU General Public License along with
- this program; if not, write to the Free Software Foundation, Inc.,
- 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
- The full GNU General Public License is included in this distribution in
- the file called "COPYING".
-
- Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
-*******************************************************************************/
-
-#include <linux/kernel.h>
-#include <linux/etherdevice.h>
-#include "stmmac_timer.h"
-
-static void stmmac_timer_handler(void *data)
-{
- struct net_device *dev = (struct net_device *)data;
-
- stmmac_schedule(dev);
-}
-
-#define STMMAC_TIMER_MSG(timer, freq) \
-printk(KERN_INFO "stmmac_timer: %s Timer ON (freq %dHz)\n", timer, freq);
-
-#if defined(CONFIG_STMMAC_RTC_TIMER)
-#include <linux/rtc.h>
-static struct rtc_device *stmmac_rtc;
-static rtc_task_t stmmac_task;
-
-static void stmmac_rtc_start(unsigned int new_freq)
-{
- rtc_irq_set_freq(stmmac_rtc, &stmmac_task, new_freq);
- rtc_irq_set_state(stmmac_rtc, &stmmac_task, 1);
-}
-
-static void stmmac_rtc_stop(void)
-{
- rtc_irq_set_state(stmmac_rtc, &stmmac_task, 0);
-}
-
-int stmmac_open_ext_timer(struct net_device *dev, struct stmmac_timer *tm)
-{
- stmmac_task.private_data = dev;
- stmmac_task.func = stmmac_timer_handler;
-
- stmmac_rtc = rtc_class_open(CONFIG_RTC_HCTOSYS_DEVICE);
- if (stmmac_rtc == NULL) {
- pr_err("open rtc device failed\n");
- return -ENODEV;
- }
-
- rtc_irq_register(stmmac_rtc, &stmmac_task);
-
- /* Periodic mode is not supported */
- if ((rtc_irq_set_freq(stmmac_rtc, &stmmac_task, tm->freq) < 0)) {
- pr_err("set periodic failed\n");
- rtc_irq_unregister(stmmac_rtc, &stmmac_task);
- rtc_class_close(stmmac_rtc);
- return -1;
- }
-
- STMMAC_TIMER_MSG(CONFIG_RTC_HCTOSYS_DEVICE, tm->freq);
-
- tm->timer_start = stmmac_rtc_start;
- tm->timer_stop = stmmac_rtc_stop;
-
- return 0;
-}
-
-int stmmac_close_ext_timer(void)
-{
- rtc_irq_set_state(stmmac_rtc, &stmmac_task, 0);
- rtc_irq_unregister(stmmac_rtc, &stmmac_task);
- rtc_class_close(stmmac_rtc);
- return 0;
-}
-
-#elif defined(CONFIG_STMMAC_TMU_TIMER)
-#include <linux/clk.h>
-#define TMU_CHANNEL "tmu2_clk"
-static struct clk *timer_clock;
-
-static void stmmac_tmu_start(unsigned int new_freq)
-{
- clk_set_rate(timer_clock, new_freq);
- clk_prepare_enable(timer_clock);
-}
-
-static void stmmac_tmu_stop(void)
-{
- clk_disable_unprepare(timer_clock);
-}
-
-int stmmac_open_ext_timer(struct net_device *dev, struct stmmac_timer *tm)
-{
- timer_clock = clk_get(NULL, TMU_CHANNEL);
-
- if (IS_ERR(timer_clock))
- return -1;
-
- if (tmu2_register_user(stmmac_timer_handler, (void *)dev) < 0) {
- timer_clock = NULL;
- return -1;
- }
-
- STMMAC_TIMER_MSG("TMU2", tm->freq);
- tm->timer_start = stmmac_tmu_start;
- tm->timer_stop = stmmac_tmu_stop;
-
- return 0;
-}
-
-int stmmac_close_ext_timer(void)
-{
- clk_disable_unprepare(timer_clock);
- tmu2_unregister_user();
- clk_put(timer_clock);
- return 0;
-}
-#endif
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_timer.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_timer.h
deleted file mode 100644
index aea9b14..0000000
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_timer.h
+++ /dev/null
@@ -1,46 +0,0 @@
-/*******************************************************************************
- STMMAC external timer Header File.
-
- Copyright (C) 2007-2009 STMicroelectronics Ltd
-
- This program is free software; you can redistribute it and/or modify it
- under the terms and conditions of the GNU General Public License,
- version 2, as published by the Free Software Foundation.
-
- This program is distributed in the hope it will be useful, but WITHOUT
- ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
- more details.
-
- You should have received a copy of the GNU General Public License along with
- this program; if not, write to the Free Software Foundation, Inc.,
- 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
-
- The full GNU General Public License is included in this distribution in
- the file called "COPYING".
-
- Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
-*******************************************************************************/
-#ifndef __STMMAC_TIMER_H__
-#define __STMMAC_TIMER_H__
-
-struct stmmac_timer {
- void (*timer_start) (unsigned int new_freq);
- void (*timer_stop) (void);
- unsigned int freq;
- unsigned int enable;
-};
-
-/* Open the HW timer device and return 0 in case of success */
-int stmmac_open_ext_timer(struct net_device *dev, struct stmmac_timer *tm);
-/* Stop the timer and release it */
-int stmmac_close_ext_timer(void);
-/* Function used for scheduling task within the stmmac */
-void stmmac_schedule(struct net_device *dev);
-
-#if defined(CONFIG_STMMAC_TMU_TIMER)
-extern int tmu2_register_user(void *fnt, void *data);
-extern void tmu2_unregister_user(void);
-#endif
-
-#endif /* __STMMAC_TIMER_H__ */
--
1.7.4.4
^ permalink raw reply related
* [PATCH net-next 2/2] tun: put correct method name in a debug message.
From: Rami Rosen @ 2012-11-26 8:07 UTC (permalink / raw)
To: davem; +Cc: maxk, netdev, Rami Rosen
In-Reply-To: <1353917261-2974-1-git-send-email-ramirose@gmail.com>
This patch puts the correct method name, tun_do_read, in a debug message.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
---
drivers/net/tun.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 1dfb135..607a3a5 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1296,7 +1296,7 @@ static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
struct sk_buff *skb;
ssize_t ret = 0;
- tun_debug(KERN_INFO, tun, "tun_chr_read\n");
+ tun_debug(KERN_INFO, tun, "tun_do_read\n");
if (unlikely(!noblock))
add_wait_queue(&tfile->wq.wait, &wait);
--
1.7.11.7
^ permalink raw reply related
* [PATCH net-next 1/2] vtun: fix typos.
From: Rami Rosen @ 2012-11-26 8:07 UTC (permalink / raw)
To: davem; +Cc: maxk, netdev, Rami Rosen
In-Reply-To: <1353917261-2974-1-git-send-email-ramirose@gmail.com>
This patch fixes four typos in drivers/net/vtun.c.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
---
drivers/net/tun.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 3bd9932..1dfb135 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -121,11 +121,11 @@ struct tap_filter {
* also contains all socket related strctures (except sock_fprog and tap_filter)
* to serve as one transmit queue for tuntap device. The sock_fprog and
* tap_filter were kept in tun_struct since they were used for filtering for the
- * netdevice not for a specific queue (at least I didn't see the reqirement for
+ * netdevice not for a specific queue (at least I didn't see the requirement for
* this).
*
* RCU usage:
- * The tun_file and tun_struct are loosely coupled, the pointer from on to the
+ * The tun_file and tun_struct are loosely coupled, the pointer from one to the
* other can only be read while rcu_read_lock or rtnl_lock is held.
*/
struct tun_file {
@@ -153,7 +153,7 @@ struct tun_flow_entry {
#define TUN_NUM_FLOW_ENTRIES 1024
/* Since the socket were moved to tun_file, to preserve the behavior of persist
- * device, socket fileter, sndbuf and vnet header size were restore when the
+ * device, socket filter, sndbuf and vnet header size were restore when the
* file were attached to a persist device.
*/
struct tun_struct {
@@ -689,7 +689,7 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
sk_filter(tfile->socket.sk, skb))
goto drop;
- /* Limit the number of packets queued by divining txq length with the
+ /* Limit the number of packets queued by dividing txq length with the
* number of queues.
*/
if (skb_queue_len(&tfile->socket.sk->sk_receive_queue)
--
1.7.11.7
^ permalink raw reply related
* [PATCH net-next 0/2] tun: trivial fixes.
From: Rami Rosen @ 2012-11-26 8:07 UTC (permalink / raw)
To: davem; +Cc: maxk, netdev, Rami Rosen
These two patches contain trivial fixes to the tun driver.
First patch fixes 4 typos.
Second patch puts the correct method name in a debug message in tun_do_read().
Rami Rosen (2):
[PATCH net-next 1/2] vtun: fix typos.
[PATCH net-next 2/2] tun: put correct method name in a debug message.
drivers/net/tun.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
--
1.7.11.7
^ permalink raw reply
* [net-next RFC] pktgen: don't wait for the device who doesn't free skb immediately after sent
From: Jason Wang @ 2012-11-26 7:56 UTC (permalink / raw)
To: rusty, mst, davem, netdev, linux-kernel, virtualization; +Cc: shemminger
Some deivces do not free the old tx skbs immediately after it has been sent
(usually in tx interrupt). One such example is virtio-net which optimizes for
virt and only free the possible old tx skbs during the next packet sending. This
would lead the pktgen to wait forever in the refcount of the skb if no other
pakcet will be sent afterwards.
Solving this issue by introducing a new flag IFF_TX_SKB_FREE_DELAY which could
notify the pktgen that the device does not free skb immediately after it has
been sent and let it not to wait for the refcount to be one.
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
Another choice is to introduce an new .ndo which could be called by pktgen
during the waiting to free the old tx skbs.
drivers/net/virtio_net.c | 3 ++-
include/uapi/linux/if.h | 2 ++
net/core/pktgen.c | 3 ++-
3 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 26c502e..8262232 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1058,7 +1058,8 @@ static int virtnet_probe(struct virtio_device *vdev)
return -ENOMEM;
/* Set up network device as normal. */
- dev->priv_flags |= IFF_UNICAST_FLT | IFF_LIVE_ADDR_CHANGE;
+ dev->priv_flags |= IFF_UNICAST_FLT | IFF_LIVE_ADDR_CHANGE |
+ IFF_TX_SKB_FREE_DELAY;
dev->netdev_ops = &virtnet_netdev;
dev->features = NETIF_F_HIGHDMA;
diff --git a/include/uapi/linux/if.h b/include/uapi/linux/if.h
index 1ec407b..a0d2168 100644
--- a/include/uapi/linux/if.h
+++ b/include/uapi/linux/if.h
@@ -83,6 +83,8 @@
#define IFF_SUPP_NOFCS 0x80000 /* device supports sending custom FCS */
#define IFF_LIVE_ADDR_CHANGE 0x100000 /* device supports hardware address
* change when it's running */
+#define IFF_TX_SKB_FREE_DELAY 0x200000 /* device does free the tx skb
+ * immediately when it is sent */
#define IF_GET_IFACE 0x0001 /* for querying only */
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index b29dacf..85d4e53 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -3269,7 +3269,8 @@ unlock:
/* If pkt_dev->count is zero, then run forever */
if ((pkt_dev->count != 0) && (pkt_dev->sofar >= pkt_dev->count)) {
- pktgen_wait_for_skb(pkt_dev);
+ if (!(pkt_dev->odev->priv_flags & IFF_TX_SKB_FREE_DELAY))
+ pktgen_wait_for_skb(pkt_dev);
/* Done with this */
pktgen_stop_device(pkt_dev);
--
1.7.1
^ permalink raw reply related
* Re: Fwd: Re: RTL 8169 linux driver question
From: Stéphane ANCELOT @ 2012-11-26 7:35 UTC (permalink / raw)
To: Francois Romieu; +Cc: netdev, sancelot, Hayes Wang
In-Reply-To: <20121123192056.GA30290@electric-eye.fr.zoreil.com>
On 23/11/2012 20:20, Francois Romieu wrote:
> Stéphane ANCELOT <sancelot@free.fr> :
> [...]
>> I have an adapted version of this driver for a realtime linux kernel
>> and a 8168/811B rev 2 component (as listed by lspci).
> You should grep for the XID line in the kernel dmesg to identify
> the chipset version.
XID 1c4000c0
>> I had problem with it, my application sends a frame that is
>> immediately transmitted back by some slaves, there was abnormally
>> 100us lost between the send and receive call.
>>
>> Finally I found it was coming from the following register setup in
>> the driver :
>>
>> RTL_W16(IntrMitigate, 0x5151);
>>
>> Can you give me some details about it, since I do not have the
>> RTL8169 programming guide.
> "Reserved" in my 2007 8168c rev1.0 datasheet.
>
> I merged it long ago from Realtek's driver. It has now changed to 0x5f51.
>
> On the old PCI 8169, bits 15..8 relate to Tx and 7..0 to Rx. Bits 7..4
> count in units of 125 us and bits 0..3 in packet units. You may give
> 0x..00 a try.
>
> Hayes knows better for the 8168 line.
>
>> /100us is important since this component acts as an Ethercat Master
>> running at 1ms./
> Which realtime kernel is it ?
xenomai.
^ 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