* Re: [PATCH net-next 06/10] {NET,IB}/mlx4: Add device managed flow steering firmware API
From: David Miller @ 2012-07-01 21:42 UTC (permalink / raw)
To: ogerlitz; +Cc: roland, yevgenyp, oren, netdev, hadarh
In-Reply-To: <4FF042B5.1000303@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
Date: Sun, 1 Jul 2012 15:29:41 +0300
> And for the time being use the module param, makes sense?
This talk about making an alternative, clean, mechanism later
is non-sense. Because you'll never be able to get rid of this
module param once you add it.
Module parameters stink because every driver is going to provide the
knob differently, with a different name, and different semantics.
This creates a terrible user experience, and I will not allow it.
^ permalink raw reply
* Re: [BUG, regression, bisected] Marvell 88E8055 NIC (sky2) fails to detect link after resume from S3
From: Michal Zatloukal @ 2012-07-01 20:58 UTC (permalink / raw)
To: Francois Romieu; +Cc: Stephen Hemminger, netdev
In-Reply-To: <CAKKZj2DZprXcEg1tRDnt9fnFahLnOEMQNno9hS2UJzQ-n13JHA@mail.gmail.com>
Traces (7z-compressed to ~250k) attached to ubuntu bug, direct link:
<https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1007841/+attachment/3210143/+files/traces.7z>
MZ
On Sun, Jul 1, 2012 at 3:30 PM, Michal Zatloukal <myxal.mxl@gmail.com> wrote:
> Breaking news:
> Fiddling with mmiotrace, I was able to get the "phy I/O error" on
> previously-thought-good kernel (non-reliably; it took several
> suspend-resume cycles), and also vice versa - in vty, I was able to
> suspend and resume the "bad" kernel with the NIC still working (also
> not reliably - in GUI, it still flaked out after first
> suspend-resume).
>
> Anyway, I've gathered traces of both good and bad behaviour, each from
> the respective kernel (not that it seems to matter much). Now, what's
> the recommended way of posting them? Each is over 3MB in size and
> pastebin won't take them. :-(
>
> MZ
>
> On Sat, Jun 30, 2012 at 10:02 PM, Francois Romieu <romieu@fr.zoreil.com> wrote:
>> Michal Zatloukal <myxal.mxl@gmail.com> :
>> [...]
>>> Is there something I can try?
>>
>> I have not used it for quite some time but comparing mmiotrace output
>> (see Documentation/trace/mmiotrace.txt) before and after the regression
>> commit may give some hint.
>>
>> Otherwise I would ask for help on linux-pm@vger.kernel.org
>>
>> --
>> Ueimor
^ permalink raw reply
* Re: [PATCH net-next 09/10] net/mlx4_en: Manage flow steering rules with ethtool
From: Joe Perches @ 2012-07-01 19:52 UTC (permalink / raw)
To: Andreas Schwab
Cc: Ben Hutchings, Or Gerlitz, davem, roland, yevgenyp, oren, netdev,
Hadar Hen Zion
In-Reply-To: <m2ipe72u78.fsf@igel.home>
On Sun, 2012-07-01 at 20:48 +0200, Andreas Schwab wrote:
> Joe Perches <joe@perches.com> writes:
> > On Sun, 2012-07-01 at 09:38 -0700, Joe Perches wrote:
> >> Perhaps it's better as:
> >>
> >> #define all_zeros_or_all_ones(field) \
> >> ({ \
> >> field && (typeof(field))~field; \
> >> })
> >
> > Umm, or not.
> >
> > It helps when I actually test the code not just type
> > it into an email client.
> >
> > !(field && (typeof(field))~field)
>
> Or write it as (!field || !(typeof(field))~field) which more closely
> resembles what the macro name expresses.
Better still, or maybe:
field == 0 || field == (typeof field)~0
^ permalink raw reply
* Re: [PATCH v5] bonding support for IPv6 transmit hashing
From: John Eaglesham @ 2012-07-01 19:01 UTC (permalink / raw)
To: David Miller; +Cc: netdev
In-Reply-To: <20120701.033354.539821533277159165.davem@davemloft.net>
On 7/1/2012 3:33 AM, David Miller wrote:
> From: John Eaglesham <linux@8192.net>
> Date: Sun, 1 Jul 2012 01:07:43 -0700
>
>> + v6hash =
>> + (ipv6h->saddr.s6_addr32[1] ^ ipv6h->daddr.s6_addr32[1]) ^
>> + (ipv6h->saddr.s6_addr32[2] ^ ipv6h->daddr.s6_addr32[2]) ^
>> + (ipv6h->saddr.s6_addr32[3] ^ ipv6h->daddr.s6_addr32[3]);
>
> You completely ignored my feedback:
>
> This is rediculous, just put &ipv6h->saddr into a local
> pointer named 's' and then you won't have use such gymnastics
> to indent the code.
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
My apologies, I completely misunderstood your feedback. I will make the
correction and re-submit.
Thanks.
John
^ permalink raw reply
* Re: [PATCH net-next 09/10] net/mlx4_en: Manage flow steering rules with ethtool
From: Andreas Schwab @ 2012-07-01 18:48 UTC (permalink / raw)
To: Joe Perches
Cc: Ben Hutchings, Or Gerlitz, davem, roland, yevgenyp, oren, netdev,
Hadar Hen Zion
In-Reply-To: <1341163872.2032.29.camel@joe2Laptop>
Joe Perches <joe@perches.com> writes:
> On Sun, 2012-07-01 at 09:38 -0700, Joe Perches wrote:
>
>> I think this macro is suboptimal because
>> negated names are easy to misuse.
>>
>> I think type is also unnecessary and too
>> easy to mismatch or keep up to date with
>> field type changes.
>>
>> Perhaps it's better as:
>>
>> #define all_zeros_or_all_ones(field) \
>> ({ \
>> field && (typeof(field))~field; \
>> })
>
> Umm, or not.
>
> It helps when I actually test the code not just type
> it into an email client.
>
> !(field && (typeof(field))~field)
Or write it as (!field || !(typeof(field))~field) which more closely
resembles what the macro name expresses.
Andreas.
--
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply
* [PATCH 0/3 v2] net: various tilegx networking fixes
From: Chris Metcalf @ 2012-07-01 18:43 UTC (permalink / raw)
To: David S. Miller, netdev, linux-kernel
Revised series to have the dates on the patches correct. Considering
I've gotten this wrong before, I promise this is the last time. :-)
This series of fixes corrects a few bugs or style issues observed
in the tilegx network driver.
David, once any potential issues with these patches are resolved,
would you prefer to take them from email, pull them from my net tree,
or have me push them as part of the tile tree?
The tree is at:
git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile.git net
Chris Metcalf (3):
net: tilegx driver bugfix (be explicit about percpu queue number)
tilegx net driver: handle payload data not in frags
tilegx net: use eth_hw_addr_random(), not random_ether_addr()
drivers/net/ethernet/tile/tilegx.c | 61 ++++++++++++++++++++----------------
1 file changed, 34 insertions(+), 27 deletions(-)
Thanks!
--
Chris Metcalf, Tilera Corp.
http://www.tilera.com
^ permalink raw reply
* [PATCH 1/3] net: tilegx driver bugfix (be explicit about percpu queue number)
From: Chris Metcalf @ 2012-07-01 18:43 UTC (permalink / raw)
To: David S. Miller, netdev, linux-kernel
In-Reply-To: <201207181640.q6IGet7P007227@lab-41.internal.tilera.com>
Avoid packets belonging to queue/cpu A trying to transmit on cpu B.
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
---
drivers/net/ethernet/tile/tilegx.c | 23 +++++++++++++++--------
1 file changed, 15 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/tile/tilegx.c b/drivers/net/ethernet/tile/tilegx.c
index 83b4b38..c7bde28 100644
--- a/drivers/net/ethernet/tile/tilegx.c
+++ b/drivers/net/ethernet/tile/tilegx.c
@@ -123,6 +123,7 @@ struct tile_net_comps {
/* The transmit wake timer for a given cpu and echannel. */
struct tile_net_tx_wake {
+ int tx_queue_idx;
struct hrtimer timer;
struct net_device *dev;
};
@@ -573,12 +574,14 @@ static void add_comp(gxio_mpipe_equeue_t *equeue,
comps->comp_next++;
}
-static void tile_net_schedule_tx_wake_timer(struct net_device *dev)
+static void tile_net_schedule_tx_wake_timer(struct net_device *dev,
+ int tx_queue_idx)
{
- struct tile_net_info *info = &__get_cpu_var(per_cpu_info);
+ struct tile_net_info *info = &per_cpu(per_cpu_info, tx_queue_idx);
struct tile_net_priv *priv = netdev_priv(dev);
+ struct tile_net_tx_wake *tx_wake = &info->tx_wake[priv->echannel];
- hrtimer_start(&info->tx_wake[priv->echannel].timer,
+ hrtimer_start(&tx_wake->timer,
ktime_set(0, TX_TIMER_DELAY_USEC * 1000UL),
HRTIMER_MODE_REL_PINNED);
}
@@ -587,7 +590,7 @@ static enum hrtimer_restart tile_net_handle_tx_wake_timer(struct hrtimer *t)
{
struct tile_net_tx_wake *tx_wake =
container_of(t, struct tile_net_tx_wake, timer);
- netif_wake_subqueue(tx_wake->dev, smp_processor_id());
+ netif_wake_subqueue(tx_wake->dev, tx_wake->tx_queue_idx);
return HRTIMER_NORESTART;
}
@@ -1218,6 +1221,7 @@ static int tile_net_open(struct net_device *dev)
hrtimer_init(&tx_wake->timer, CLOCK_MONOTONIC,
HRTIMER_MODE_REL);
+ tx_wake->tx_queue_idx = cpu;
tx_wake->timer.function = tile_net_handle_tx_wake_timer;
tx_wake->dev = dev;
}
@@ -1291,6 +1295,7 @@ static inline void *tile_net_frag_buf(skb_frag_t *f)
* stop the queue and schedule the tx_wake timer.
*/
static s64 tile_net_equeue_try_reserve(struct net_device *dev,
+ int tx_queue_idx,
struct tile_net_comps *comps,
gxio_mpipe_equeue_t *equeue,
int num_edescs)
@@ -1313,8 +1318,8 @@ static s64 tile_net_equeue_try_reserve(struct net_device *dev,
}
/* Still nothing; give up and stop the queue for a short while. */
- netif_stop_subqueue(dev, smp_processor_id());
- tile_net_schedule_tx_wake_timer(dev);
+ netif_stop_subqueue(dev, tx_queue_idx);
+ tile_net_schedule_tx_wake_timer(dev, tx_queue_idx);
return -1;
}
@@ -1580,7 +1585,8 @@ static int tile_net_tx_tso(struct sk_buff *skb, struct net_device *dev)
local_irq_save(irqflags);
/* Try to acquire a completion entry and an egress slot. */
- slot = tile_net_equeue_try_reserve(dev, comps, equeue, num_edescs);
+ slot = tile_net_equeue_try_reserve(dev, skb->queue_mapping, comps,
+ equeue, num_edescs);
if (slot < 0) {
local_irq_restore(irqflags);
return NETDEV_TX_BUSY;
@@ -1674,7 +1680,8 @@ static int tile_net_tx(struct sk_buff *skb, struct net_device *dev)
local_irq_save(irqflags);
/* Try to acquire a completion entry and an egress slot. */
- slot = tile_net_equeue_try_reserve(dev, comps, equeue, num_edescs);
+ slot = tile_net_equeue_try_reserve(dev, skb->queue_mapping, comps,
+ equeue, num_edescs);
if (slot < 0) {
local_irq_restore(irqflags);
return NETDEV_TX_BUSY;
--
1.7.10.3
^ permalink raw reply related
* Re: [PATCH net-next 09/10] net/mlx4_en: Manage flow steering rules with ethtool
From: Joe Perches @ 2012-07-01 17:31 UTC (permalink / raw)
To: Ben Hutchings
Cc: Or Gerlitz, davem, roland, yevgenyp, oren, netdev, Hadar Hen Zion
In-Reply-To: <1341160685.2032.15.camel@joe2Laptop>
On Sun, 2012-07-01 at 09:38 -0700, Joe Perches wrote:
> I think this macro is suboptimal because
> negated names are easy to misuse.
>
> I think type is also unnecessary and too
> easy to mismatch or keep up to date with
> field type changes.
>
> Perhaps it's better as:
>
> #define all_zeros_or_all_ones(field) \
> ({ \
> field && (typeof(field))~field; \
> })
Umm, or not.
It helps when I actually test the code not just type
it into an email client.
!(field && (typeof(field))~field)
^ permalink raw reply
* Re: [PATCH 00/12] Swap-over-NFS without deadlocking V8
From: Eric B Munson @ 2012-07-01 17:22 UTC (permalink / raw)
To: Mel Gorman
Cc: Andrew Morton, Linux-MM, Linux-Netdev, Linux-NFS, LKML,
David Miller, Trond Myklebust, Neil Brown, Christoph Hellwig,
Peter Zijlstra, Mike Christie, Sebastian Andrzej Siewior
In-Reply-To: <1340976805-5799-1-git-send-email-mgorman@suse.de>
[-- Attachment #1: Type: text/plain, Size: 4245 bytes --]
On Fri, 29 Jun 2012, Mel Gorman wrote:
> Changelog since V7
> o Rebase to linux-next 20120629
> o bi->page_dma instead of bi->page in intel driver
> o Build fix for !CONFIG_NET (sebastian)
> o Restore PF_MEMALLOC flags correctly in all cases (jlayton)
>
> Changelog since V6
> o Rebase to linux-next 20120622
>
> Changelog since V5
> o Rebase to v3.5-rc3
>
> Changelog since V4
> o Catch if SOCK_MEMALLOC flag is cleared with rmem tokens (davem)
>
> Changelog since V3
> o Rebase to 3.4-rc5
> o kmap pages for writing to swap (akpm)
> o Move forward declaration to reduce chance of duplication (akpm)
>
> Changelog since V2
> o Nothing significant, just rebases. A radix tree lookup is replaced with
> a linear search would be the biggest rebase artifact
>
> This patch series is based on top of "Swap-over-NBD without deadlocking v14"
> as it depends on the same reservation of PF_MEMALLOC reserves logic.
>
> When a user or administrator requires swap for their application, they
> create a swap partition and file, format it with mkswap and activate it with
> swapon. In diskless systems this is not an option so if swap if required
> then swapping over the network is considered. The two likely scenarios
> are when blade servers are used as part of a cluster where the form factor
> or maintenance costs do not allow the use of disks and thin clients.
>
> The Linux Terminal Server Project recommends the use of the Network
> Block Device (NBD) for swap but this is not always an option. There is
> no guarantee that the network attached storage (NAS) device is running
> Linux or supports NBD. However, it is likely that it supports NFS so there
> are users that want support for swapping over NFS despite any performance
> concern. Some distributions currently carry patches that support swapping
> over NFS but it would be preferable to support it in the mainline kernel.
>
> Patch 1 avoids a stream-specific deadlock that potentially affects TCP.
>
> Patch 2 is a small modification to SELinux to avoid using PFMEMALLOC
> reserves.
>
> Patch 3 adds three helpers for filesystems to handle swap cache pages.
> For example, page_file_mapping() returns page->mapping for
> file-backed pages and the address_space of the underlying
> swap file for swap cache pages.
>
> Patch 4 adds two address_space_operations to allow a filesystem
> to pin all metadata relevant to a swapfile in memory. Upon
> successful activation, the swapfile is marked SWP_FILE and
> the address space operation ->direct_IO is used for writing
> and ->readpage for reading in swap pages.
>
> Patch 5 notes that patch 3 is bolting
> filesystem-specific-swapfile-support onto the side and that
> the default handlers have different information to what
> is available to the filesystem. This patch refactors the
> code so that there are generic handlers for each of the new
> address_space operations.
>
> Patch 6 adds an API to allow a vector of kernel addresses to be
> translated to struct pages and pinned for IO.
>
> Patch 7 adds support for using highmem pages for swap by kmapping
> the pages before calling the direct_IO handler.
>
> Patch 8 updates NFS to use the helpers from patch 3 where necessary.
>
> Patch 9 avoids setting PF_private on PG_swapcache pages within NFS.
>
> Patch 10 implements the new swapfile-related address_space operations
> for NFS and teaches the direct IO handler how to manage
> kernel addresses.
>
> Patch 11 prevents page allocator recursions in NFS by using GFP_NOIO
> where appropriate.
>
> Patch 12 fixes a NULL pointer dereference that occurs when using
> swap-over-NFS.
>
> With the patches applied, it is possible to mount a swapfile that is on an
> NFS filesystem. Swap performance is not great with a swap stress test taking
> roughly twice as long to complete than if the swap device was backed by NBD.
To test this set I am using memory cgroups to force swap usage. I am seeing
the cgroup controller killing my processes instead of using the nfs swapfile.
I am not yet sure if I am making a silly mistake or if something else is wrong.
Eric
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply
* Re: [PATCH net-next 09/10] net/mlx4_en: Manage flow steering rules with ethtool
From: Joe Perches @ 2012-07-01 16:38 UTC (permalink / raw)
To: Ben Hutchings
Cc: Or Gerlitz, davem, roland, yevgenyp, oren, netdev, Hadar Hen Zion
In-Reply-To: <1341158452.4852.107.camel@deadeye.wl.decadent.org.uk>
On Sun, 2012-07-01 at 17:00 +0100, Ben Hutchings wrote:
> On Sun, 2012-07-01 at 12:43 +0300, Or Gerlitz wrote:
> > From: Hadar Hen Zion <hadarh@mellanox.co.il>
[]
> > diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
[]
> > @@ -599,16 +603,360 @@ static int mlx4_en_set_rxfh_indir(struct net_device *dev,
> > return err;
> > }
> >
> > +#define not_all_zeros_or_all_ones(field, type) \
> > + (field && (type)~field)
I think this macro is suboptimal because
negated names are easy to misuse.
I think type is also unnecessary and too
easy to mismatch or keep up to date with
field type changes.
Perhaps it's better as:
#define all_zeros_or_all_ones(field) \
({ \
field && (typeof(field))~field; \
})
> > +
> > +static int mlx4_en_validate_flow(struct net_device *dev,
> > + struct ethtool_rxnfc *cmd)
> > +{
[]
> > + /* don't allow mask which isn't all 0 or 1 */
> > + if (not_all_zeros_or_all_ones(l4_mask->ip4src, __be32) ||
> > + not_all_zeros_or_all_ones(l4_mask->ip4dst, __be32) ||
> > + not_all_zeros_or_all_ones(l4_mask->psrc, __be16) ||
> > + not_all_zeros_or_all_ones(l4_mask->pdst, __be16))
> > + return -EOPNOTSUPP;
if (!all_zeros_or_all_ones(l4_mask->ip4src) ||
!all_zeros_or_all_ones(l4_mask->ip4dst) ||
!all_zeros_or_all_ones(l4_mask->psrc) ||
!all_zeros_or_all_ones(l4_mask->pdst))
^ permalink raw reply
* Re: [PATCH net-next 09/10] net/mlx4_en: Manage flow steering rules with ethtool
From: Ben Hutchings @ 2012-07-01 16:00 UTC (permalink / raw)
To: Or Gerlitz; +Cc: davem, roland, yevgenyp, oren, netdev, Hadar Hen Zion
In-Reply-To: <1341135823-29039-10-git-send-email-ogerlitz@mellanox.com>
On Sun, 2012-07-01 at 12:43 +0300, Or Gerlitz wrote:
> From: Hadar Hen Zion <hadarh@mellanox.co.il>
>
> Implement the ethtool APIs for attaching L2/L3/L4 based flow steering
> rules to the netdevice RX rings. Added set_rxnfc callback and enhanced
> the existing get_rxnfc callback.
>
> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.co.il>
> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
> ---
> drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 373 +++++++++++++++++++++++
> drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 7 +
> 2 files changed, 380 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
> index 72901ce..30de264 100644
> --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
> +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
> @@ -38,6 +38,10 @@
> #include "mlx4_en.h"
> #include "en_port.h"
>
> +#define EN_ETHTOOL_QP_ATTACH (1ull << 63)
> +#define EN_ETHTOOL_MAC_MASK 0xffffffffffffULL
> +#define EN_ETHTOOL_SHORT_MASK cpu_to_be16(0xffff)
> +#define EN_ETHTOOL_WORD_MASK cpu_to_be32(0xffffffff)
>
> static void
> mlx4_en_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo)
> @@ -599,16 +603,360 @@ static int mlx4_en_set_rxfh_indir(struct net_device *dev,
> return err;
> }
>
> +#define not_all_zeros_or_all_ones(field, type) \
> + (field && (type)~field)
> +
> +static int mlx4_en_validate_flow(struct net_device *dev,
> + struct ethtool_rxnfc *cmd)
> +{
> + struct ethtool_usrip4_spec *l3_mask;
> + struct ethtool_tcpip4_spec *l4_mask;
> + struct ethhdr *eth_mask;
> + u64 full_mac = ~0ull;
> + u64 zero_mac = 0;
> +
> + if (cmd->fs.location >= MAX_NUM_OF_FS_RULES)
> + return -EINVAL;
> +
> + switch (cmd->fs.flow_type & ~FLOW_EXT) {
> + case TCP_V4_FLOW:
> + case UDP_V4_FLOW:
> + if (cmd->fs.h_u.tcp_ip4_spec.tos)
> + return -EOPNOTSUPP;
I suspect that your filter ignores TOS, rather than only matching TOS ==
0, so you should actually be checking the corresponding field in the
mask (fs.m_u). The error code should be EINVAL.
> + l4_mask = &cmd->fs.m_u.tcp_ip4_spec;
> + /* don't allow mask which isn't all 0 or 1 */
> + if (not_all_zeros_or_all_ones(l4_mask->ip4src, __be32) ||
> + not_all_zeros_or_all_ones(l4_mask->ip4dst, __be32) ||
> + not_all_zeros_or_all_ones(l4_mask->psrc, __be16) ||
> + not_all_zeros_or_all_ones(l4_mask->pdst, __be16))
> + return -EOPNOTSUPP;
Again, here and in many further instances, the error code should be
EINVAL.
> + break;
> + case IP_USER_FLOW:
> + l3_mask = &cmd->fs.m_u.usr_ip4_spec;
> + if (cmd->fs.h_u.usr_ip4_spec.l4_4_bytes ||
> + cmd->fs.h_u.usr_ip4_spec.tos ||
I think this should be checking l4_4_bytes and tos in the mask.
> + cmd->fs.h_u.usr_ip4_spec.proto ||
> + cmd->fs.h_u.usr_ip4_spec.ip_ver != ETH_RX_NFC_IP4 ||
> + (!cmd->fs.h_u.usr_ip4_spec.ip4src &&
> + !cmd->fs.h_u.usr_ip4_spec.ip4dst) ||
> + not_all_zeros_or_all_ones(l3_mask->ip4src, __be32) ||
> + not_all_zeros_or_all_ones(l3_mask->ip4dst, __be32))
> + return -EOPNOTSUPP;
> + break;
> + case ETHER_FLOW:
> + eth_mask = &cmd->fs.m_u.ether_spec;
> + if (memcmp(eth_mask->h_source, &zero_mac, ETH_ALEN))
> + return -EOPNOTSUPP;
> + if (!memcmp(eth_mask->h_dest, &zero_mac, ETH_ALEN))
> + return -EOPNOTSUPP;
But in the next statement you test whether eth_mask->h_dest is either
all-zeroes or all-ones. Is all-zeroes valid or not? I suspect you
actually intend to reject the case where both h_dest and h_proto are
masked out.
> + if (not_all_zeros_or_all_ones(eth_mask->h_proto, __be16) ||
> + (memcmp(eth_mask->h_dest, &zero_mac, ETH_ALEN) &&
> + memcmp(eth_mask->h_dest, &full_mac, ETH_ALEN)))
> + return -EOPNOTSUPP;
> + break;
> + default:
> + return -EOPNOTSUPP;
> + }
> +
> + if ((cmd->fs.flow_type & FLOW_EXT)) {
> + if (cmd->fs.m_ext.vlan_etype ||
> + not_all_zeros_or_all_ones(cmd->fs.m_ext.vlan_tci,
> + __be16)) {
> + return -EOPNOTSUPP;
> + }
> + }
> +
> + return 0;
> +}
> +
> +static void add_ip_rule(struct mlx4_en_priv *priv,
> + struct ethtool_rxnfc *cmd,
> + struct list_head *list_h)
> +{
> + struct mlx4_spec_list *spec_l3;
> +
> + spec_l3 = kzalloc(sizeof *spec_l3, GFP_KERNEL);
> + if (!spec_l3) {
> + en_err(priv, "Fail to alloc ethtool rule.\n");
> + return;
> + }
This should return an error code as well - logging is not a substitue.
> + spec_l3->id = MLX4_NET_TRANS_RULE_ID_IPV4;
> + spec_l3->ipv4.src_ip = cmd->fs.h_u.usr_ip4_spec.ip4src;
> + if (spec_l3->ipv4.src_ip)
> + spec_l3->ipv4.src_ip_msk = EN_ETHTOOL_WORD_MASK;
> + spec_l3->ipv4.dst_ip = cmd->fs.h_u.usr_ip4_spec.ip4dst;
> + if (spec_l3->ipv4.dst_ip)
> + spec_l3->ipv4.dst_ip_msk = EN_ETHTOOL_WORD_MASK;
The conditions should be using the mask (cmd->fs.m_u) not the value.
> + list_add_tail(&spec_l3->list, list_h);
> +}
> +
> +static void add_tcp_udp_rule(struct mlx4_en_priv *priv,
> + struct ethtool_rxnfc *cmd,
> + struct list_head *list_h, int proto)
> +{
> + struct mlx4_spec_list *spec_l3;
> + struct mlx4_spec_list *spec_l4;
> +
> + spec_l3 = kzalloc(sizeof *spec_l3, GFP_KERNEL);
> + spec_l4 = kzalloc(sizeof *spec_l4, GFP_KERNEL);
> + if (!spec_l4 || !spec_l3) {
> + en_err(priv, "Fail to alloc ethtool rule.\n");
If one of them was successfully allocated, it will now be leaked.
> + return;
> + }
> +
> + spec_l3->id = MLX4_NET_TRANS_RULE_ID_IPV4;
> +
> + if (proto == TCP_V4_FLOW) {
> + spec_l4->id = MLX4_NET_TRANS_RULE_ID_TCP;
> + spec_l3->ipv4.src_ip = cmd->fs.h_u.tcp_ip4_spec.ip4src;
> + spec_l3->ipv4.dst_ip = cmd->fs.h_u.tcp_ip4_spec.ip4dst;
> + spec_l4->tcp_udp.src_port = cmd->fs.h_u.tcp_ip4_spec.psrc;
> + spec_l4->tcp_udp.dst_port = cmd->fs.h_u.tcp_ip4_spec.pdst;
> + } else {
> + spec_l4->id = MLX4_NET_TRANS_RULE_ID_UDP;
> + spec_l3->ipv4.src_ip = cmd->fs.h_u.udp_ip4_spec.ip4src;
> + spec_l3->ipv4.dst_ip = cmd->fs.h_u.udp_ip4_spec.ip4dst;
> + spec_l4->tcp_udp.src_port = cmd->fs.h_u.udp_ip4_spec.psrc;
> + spec_l4->tcp_udp.dst_port = cmd->fs.h_u.udp_ip4_spec.pdst;
> + }
> +
> + if (spec_l3->ipv4.src_ip)
> + spec_l3->ipv4.src_ip_msk = EN_ETHTOOL_WORD_MASK;
> + if (spec_l3->ipv4.dst_ip)
> + spec_l3->ipv4.dst_ip_msk = EN_ETHTOOL_WORD_MASK;
> +
> + if (spec_l4->tcp_udp.src_port)
> + spec_l4->tcp_udp.src_port_msk = EN_ETHTOOL_SHORT_MASK;
> + if (spec_l4->tcp_udp.dst_port)
> + spec_l4->tcp_udp.dst_port_msk = EN_ETHTOOL_SHORT_MASK;
All these conditions should be using the mask, not the value.
> + list_add_tail(&spec_l3->list, list_h);
> + list_add_tail(&spec_l4->list, list_h);
> +}
> +
> +static int mlx4_en_ethtool_to_net_trans_rule(struct net_device *dev,
> + struct ethtool_rxnfc *cmd,
> + struct list_head *rule_list_h)
> +{
> + int err;
> + u64 mac;
> + __be64 be_mac;
> + struct ethhdr *eth_spec;
> + struct mlx4_en_priv *priv = netdev_priv(dev);
> + struct mlx4_spec_list *spec_l2;
> + __be64 mac_msk = cpu_to_be64(EN_ETHTOOL_MAC_MASK << 16);
> +
> + err = mlx4_en_validate_flow(dev, cmd);
> + if (err)
> + return err;
> +
> + spec_l2 = kzalloc(sizeof *spec_l2, GFP_KERNEL);
> + if (!spec_l2)
> + return -ENOMEM;
> +
> + mac = priv->mac & EN_ETHTOOL_MAC_MASK;
> + be_mac = cpu_to_be64(mac << 16);
> +
> + spec_l2->id = MLX4_NET_TRANS_RULE_ID_ETH;
> + memcpy(spec_l2->eth.dst_mac_msk, &mac_msk, ETH_ALEN);
> + if ((cmd->fs.flow_type & ~FLOW_EXT) != ETHER_FLOW)
> + memcpy(spec_l2->eth.dst_mac, &be_mac, ETH_ALEN);
Does the hardware require filtering by MAC address and not only by layer
3/4 addresses?
> + if ((cmd->fs.flow_type & FLOW_EXT) && cmd->fs.m_ext.vlan_tci) {
> + spec_l2->eth.vlan_id = cmd->fs.h_ext.vlan_tci;
> + spec_l2->eth.vlan_id_msk = cpu_to_be16(0xfff);
This doesn't match mlx4_en_validate_flow(); you are replacing a mask of
0xffff with 0xfff.
> + }
> +
> + list_add_tail(&spec_l2->list, rule_list_h);
> +
> + switch (cmd->fs.flow_type & ~FLOW_EXT) {
> + case ETHER_FLOW:
> + eth_spec = &cmd->fs.h_u.ether_spec;
> + memcpy(&spec_l2->eth.dst_mac, eth_spec->h_dest, ETH_ALEN);
> + spec_l2->eth.ether_type = eth_spec->h_proto;
> + if (eth_spec->h_proto)
> + spec_l2->eth.ether_type_enable = 1;
> + break;
> + case IP_USER_FLOW:
> + add_ip_rule(priv, cmd, rule_list_h);
> + break;
> + case TCP_V4_FLOW:
> + add_tcp_udp_rule(priv, cmd, rule_list_h, TCP_V4_FLOW);
> + break;
> + case UDP_V4_FLOW:
> + add_tcp_udp_rule(priv, cmd, rule_list_h, UDP_V4_FLOW);
> + break;
All those functions need to be able to return errors.
> + }
> + return 0;
> +}
> +
> +static int mlx4_en_flow_replace(struct net_device *dev,
> + struct ethtool_rxnfc *cmd)
> +{
> + int err;
> + struct mlx4_en_priv *priv = netdev_priv(dev);
> + struct ethtool_flow_id *loc_rule;
> + struct mlx4_spec_list *spec, *tmp_spec;
> + u32 qpn;
> + u64 reg_id;
> +
> + struct mlx4_net_trans_rule rule = {
> + .queue_mode = MLX4_NET_TRANS_Q_FIFO,
> + .exclusive = 0,
> + .allow_loopback = 1,
> + .promisc_mode = MLX4_FS_PROMISC_NONE,
> + };
> +
> + rule.port = priv->port;
> + rule.priority = MLX4_DOMAIN_ETHTOOL | cmd->fs.location;
> + INIT_LIST_HEAD(&rule.list);
> +
> + /* Allow direct QP attaches if the EN_ETHTOOL_QP_ATTACH flag is set */
> + if (cmd->fs.ring_cookie == RX_CLS_FLOW_DISC)
> + return -EOPNOTSUPP;
EINVAL.
[...]
> static int mlx4_en_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
> u32 *rule_locs)
> {
> struct mlx4_en_priv *priv = netdev_priv(dev);
> + struct mlx4_en_dev *mdev = priv->mdev;
> int err = 0;
> + int i = 0, priority = 0;
> +
> + if (mdev->dev->caps.steering_mode != MLX4_STEERING_MODE_DEVICE_MANAGED)
> + return -EOPNOTSUPP;
>
> switch (cmd->cmd) {
> case ETHTOOL_GRXRINGS:
> cmd->data = priv->rx_ring_num;
> break;
> + case ETHTOOL_GRXCLSRLCNT:
> + cmd->rule_cnt = mlx4_en_get_num_flows(priv);
> + break;
> + case ETHTOOL_GRXCLSRULE:
> + err = mlx4_en_get_flow(dev, cmd, cmd->fs.location);
> + break;
> + case ETHTOOL_GRXCLSRLALL:
> + while (!err || err == -ENOENT) {
> + err = mlx4_en_get_flow(dev, cmd, i);
> + if (!err)
> + ((u32 *)(rule_locs))[priority++] = i;
I don't see any range check against cmd->rule_cnt.
Why are you casting rule_locs?
Also, are the rules really prioritised by location, so that if rule 0
and 1 match a packet then only rule 0 is applied? If they are actually
prioritised by the match type then you need to assign locations on the
driver side that reflect the actual priorities.
> + i++;
> + }
> + if (priority)
> + err = 0;
[...]
But if there are no rules defined, this is an error? That's not right.
I think you should unconditionally set err = 0 here.
Ben.
--
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 net-next 00/11] default maximal number of RSS queues in mq drivers
From: Or Gerlitz @ 2012-07-01 15:01 UTC (permalink / raw)
To: Yuval Mintz
Cc: davem, netdev, eilong, Divy Le Ray, Jon Mason,
Anirban Chakraborty, Jitendra Kalsaria, Ron Mercer, Jeff Kirsher,
Jon Mason, Andrew Gallatin, Sathya Perla, Subbu Seetharaman,
Ajit Khaparde, Matt Carlson, Michael Chan, Eric Dumazet,
Ben Hutchings
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
On 7/1/2012 4:18 PM, Yuval Mintz wrote:
> Different vendors support different number of RSS queues by default. Today,
> there exists an ethtool API through which users can change the number of
> channels their driver supports; This enables us to pursue the goal of using
> a default number of RSS queues in various multi-queue drivers.
Yuval,
I assume this posting is actually v3 of "[RFC net-next (v2) 00/14]
default maximal number of RSS queues in mq drivers" see
http://marc.info/?l=linux-netdev&m=134062509310404&w=2?
correct? so what has been changed along v0/v1/v2 --> v3? It would be
very helpful to follow
if you make sure to mark the patches that makes Vn with "PATCH net-next
Vn" and specify the changes
from Vn-1, Vn-2...V1, V0.
Or.
^ permalink raw reply
* Re: AF_BUS socket address family
From: Alan Cox @ 2012-07-01 14:16 UTC (permalink / raw)
To: David Miller; +Cc: vincent.sanders, netdev, linux-kernel
In-Reply-To: <20120630.173346.670086962166528527.davem@davemloft.net>
> The issue is that what to do when a receiver goes deaf is a policy
> issue.
Something all these protocols alredy recognize and have done for years.
The real issue for AF_BUS versus the current slow AF_UNIX approach is
that you really need to know *who* is blocked up.
That's no different to UDP multicast and needing to know how errored the
frame.
Alan
^ permalink raw reply
* Re: REGRESSION: 3.4.0->3.5.0-rc2 kernel WARNING on cable plug on Acer Aspire One, no network
From: Alex Villacís Lasso @ 2012-07-01 13:50 UTC (permalink / raw)
To: Francois Romieu; +Cc: netdev
In-Reply-To: <20120611213815.GA8517@electric-eye.fr.zoreil.com>
El 11/06/12 16:38, Francois Romieu escribió:
> Alex Villacís Lasso <a_villacis@palosanto.com> :
> [...]
>> $ grep XID dmesg-3.5.0-rc2.txt
>> [ 15.873858] r8169 0000:02:00.0: eth0: RTL8102e at 0xf7c0e000,
>> 00:1e:68:e5:5d:b1, XID 04a00000 IRQ 44
> The 8102e has not been touched by that many suspect patches but I do
> not see where the problem is :o(
>
> Can you peel off the r8169 patches between 3.4.0 and 3.5-rc ?
>
Still present in 3.5-rc5. Bisection still in progress.
^ permalink raw reply
* Re: [BUG, regression, bisected] Marvell 88E8055 NIC (sky2) fails to detect link after resume from S3
From: Michal Zatloukal @ 2012-07-01 13:30 UTC (permalink / raw)
To: Francois Romieu; +Cc: Stephen Hemminger, netdev
In-Reply-To: <20120630200236.GA30058@electric-eye.fr.zoreil.com>
Breaking news:
Fiddling with mmiotrace, I was able to get the "phy I/O error" on
previously-thought-good kernel (non-reliably; it took several
suspend-resume cycles), and also vice versa - in vty, I was able to
suspend and resume the "bad" kernel with the NIC still working (also
not reliably - in GUI, it still flaked out after first
suspend-resume).
Anyway, I've gathered traces of both good and bad behaviour, each from
the respective kernel (not that it seems to matter much). Now, what's
the recommended way of posting them? Each is over 3MB in size and
pastebin won't take them. :-(
MZ
On Sat, Jun 30, 2012 at 10:02 PM, Francois Romieu <romieu@fr.zoreil.com> wrote:
> Michal Zatloukal <myxal.mxl@gmail.com> :
> [...]
>> Is there something I can try?
>
> I have not used it for quite some time but comparing mmiotrace output
> (see Documentation/trace/mmiotrace.txt) before and after the regression
> commit may give some hint.
>
> Otherwise I would ask for help on linux-pm@vger.kernel.org
>
> --
> Ueimor
^ permalink raw reply
* [PATCH net-next 11/11] be2net: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:19 UTC (permalink / raw)
To: davem, netdev
Cc: eilong, Yuval Mintz, Sathya Perla, Subbu Seetharaman,
Ajit Khaparde
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Sathya Perla <sathya.perla@emulex.com>
Cc: Subbu Seetharaman <subbu.seetharaman@emulex.com>
Cc: Ajit Khaparde <ajit.khaparde@emulex.com>
---
drivers/net/ethernet/emulex/benet/be_main.c | 10 ++++++----
1 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index edce7af..2141bd7 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -2172,12 +2172,14 @@ static void be_msix_disable(struct be_adapter *adapter)
static uint be_num_rss_want(struct be_adapter *adapter)
{
+ u32 num = 0;
if ((adapter->function_caps & BE_FUNCTION_CAPS_RSS) &&
!sriov_want(adapter) && be_physfn(adapter) &&
- !be_is_mc(adapter))
- return (adapter->be3_native) ? BE3_MAX_RSS_QS : BE2_MAX_RSS_QS;
- else
- return 0;
+ !be_is_mc(adapter)) {
+ num = (adapter->be3_native) ? BE3_MAX_RSS_QS : BE2_MAX_RSS_QS;
+ num = min_t(u32, num, (u32)netif_get_num_default_rss_queues());
+ }
+ return num;
}
static void be_msix_enable(struct be_adapter *adapter)
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 10/11] bnx2x: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index daa894b..53659f3 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -822,7 +822,8 @@ static inline int bnx2x_calc_num_queues(struct bnx2x *bp)
{
return num_queues ?
min_t(int, num_queues, BNX2X_MAX_QUEUES(bp)) :
- min_t(int, num_online_cpus(), BNX2X_MAX_QUEUES(bp));
+ min_t(int, netif_get_num_default_rss_queues(),
+ BNX2X_MAX_QUEUES(bp));
}
static inline void bnx2x_clear_sge_mask_next_elems(struct bnx2x_fastpath *fp)
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 07/11] myri10ge: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz, Jon Mason
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Jon Mason <mason@myri.com>
---
drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
index 90153fc..fa85cf1 100644
--- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
+++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c
@@ -3775,7 +3775,7 @@ static void myri10ge_probe_slices(struct myri10ge_priv *mgp)
mgp->num_slices = 1;
msix_cap = pci_find_capability(pdev, PCI_CAP_ID_MSIX);
- ncpus = num_online_cpus();
+ ncpus = netif_get_num_default_rss_queues();
if (myri10ge_max_slices == 1 || msix_cap == 0 ||
(myri10ge_max_slices == -1 && ncpus < 2))
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 08/11] tg3: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz, Matt Carlson
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Matt Carlson <mcarlson@broadcom.com>
---
drivers/net/ethernet/broadcom/tg3.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index e47ff8b..6cbab03 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -9908,7 +9908,7 @@ static bool tg3_enable_msix(struct tg3 *tp)
int i, rc;
struct msix_entry msix_ent[tp->irq_max];
- tp->irq_cnt = num_online_cpus();
+ tp->irq_cnt = netif_get_num_default_rss_queues();
if (tp->irq_cnt > 1) {
/* We want as many rx rings enabled as there are cpus.
* In multiqueue MSI-X mode, the first MSI-X vector
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 09/11] bnx2: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz, Michael Chan
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Michael Chan <mchan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c
index 9eb7624..1901da1 100644
--- a/drivers/net/ethernet/broadcom/bnx2.c
+++ b/drivers/net/ethernet/broadcom/bnx2.c
@@ -6250,7 +6250,7 @@ bnx2_enable_msix(struct bnx2 *bp, int msix_vecs)
static int
bnx2_setup_int_mode(struct bnx2 *bp, int dis_msi)
{
- int cpus = num_online_cpus();
+ int cpus = netif_get_num_default_rss_queues();
int msix_vecs;
if (!bp->num_req_rx_rings)
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 04/11] qlge: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev
Cc: eilong, Yuval Mintz, Anirban Chakraborty, Jitendra Kalsaria,
Ron Mercer
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Anirban Chakraborty <anirban.chakraborty@qlogic.com>
Cc: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Cc: Ron Mercer <ron.mercer@qlogic.com>
---
drivers/net/ethernet/qlogic/qlge/qlge_main.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_main.c b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
index 09d8d33..3c3499d 100644
--- a/drivers/net/ethernet/qlogic/qlge/qlge_main.c
+++ b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
@@ -4649,7 +4649,7 @@ static int __devinit qlge_probe(struct pci_dev *pdev,
int err = 0;
ndev = alloc_etherdev_mq(sizeof(struct ql_adapter),
- min(MAX_CPUS, (int)num_online_cpus()));
+ min(MAX_CPUS, netif_get_num_default_rss_queues()));
if (!ndev)
return -ENOMEM;
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 05/11] cxgb3: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz, Divy Le Ray
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Divy Le Ray <divy@chelsio.com>
---
drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
index abb6ce7..9b08749 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c
@@ -3050,7 +3050,7 @@ static struct pci_error_handlers t3_err_handler = {
static void set_nqsets(struct adapter *adap)
{
int i, j = 0;
- int num_cpus = num_online_cpus();
+ int num_cpus = netif_get_num_default_rss_queues();
int hwports = adap->params.nports;
int nqsets = adap->msix_nvectors - 1;
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 03/11] vxge: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz, Jon Mason
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Jon Mason <jdmason@kudzu.us>
---
drivers/net/ethernet/neterion/vxge/vxge-main.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/neterion/vxge/vxge-main.c b/drivers/net/ethernet/neterion/vxge/vxge-main.c
index 2578eb1..2fd1edb 100644
--- a/drivers/net/ethernet/neterion/vxge/vxge-main.c
+++ b/drivers/net/ethernet/neterion/vxge/vxge-main.c
@@ -3687,7 +3687,8 @@ static int __devinit vxge_config_vpaths(
return 0;
if (!driver_config->g_no_cpus)
- driver_config->g_no_cpus = num_online_cpus();
+ driver_config->g_no_cpus =
+ netif_get_num_default_rss_queues();
driver_config->vpath_per_dev = driver_config->g_no_cpus >> 1;
if (!driver_config->vpath_per_dev)
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 06/11] cxgb4: set maximal number of default RSS queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz, Divy Le Ray
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Cc: Divy Le Ray <divy@chelsio.com>
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index e1f96fb..5ed49af 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -3493,8 +3493,8 @@ static void __devinit cfg_queues(struct adapter *adap)
*/
if (n10g)
q10g = (MAX_ETH_QSETS - (adap->params.nports - n10g)) / n10g;
- if (q10g > num_online_cpus())
- q10g = num_online_cpus();
+ if (q10g > netif_get_num_default_rss_queues())
+ q10g = netif_get_num_default_rss_queues();
for_each_port(adap, i) {
struct port_info *pi = adap2pinfo(adap, i);
--
1.7.9.rc2
^ permalink raw reply related
* [PATCH net-next 01/11] net-next: Add netif_get_num_default_rss_queues
From: Yuval Mintz @ 2012-07-01 13:18 UTC (permalink / raw)
To: davem, netdev; +Cc: eilong, Yuval Mintz
In-Reply-To: <1341148740-7375-1-git-send-email-yuvalmin@broadcom.com>
Most multi-queue networking driver consider the number of online cpus when
configuring RSS queues.
This patch adds a wrapper to the number of cpus, setting an upper limit on the
number of cpus a driver should consider (by default) when allocating resources
for his queues.
Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
include/linux/netdevice.h | 3 +++
net/core/dev.c | 11 +++++++++++
2 files changed, 14 insertions(+), 0 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 2c2ecea..ab0251d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2119,6 +2119,9 @@ static inline int netif_copy_real_num_queues(struct net_device *to_dev,
#endif
}
+#define DEFAULT_MAX_NUM_RSS_QUEUES (8)
+extern int netif_get_num_default_rss_queues(void);
+
/* Use this variant when it is known for sure that it
* is executing from hardware interrupt context or with hardware interrupts
* disabled.
diff --git a/net/core/dev.c b/net/core/dev.c
index ed674e2..69f7a1a 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1793,6 +1793,17 @@ int netif_set_real_num_rx_queues(struct net_device *dev, unsigned int rxq)
EXPORT_SYMBOL(netif_set_real_num_rx_queues);
#endif
+/* netif_get_num_default_rss_queues - default number of RSS queues
+ *
+ * This routine should set an upper limit on the number of RSS queues
+ * used by default by multiqueue devices.
+ */
+int netif_get_num_default_rss_queues()
+{
+ return min_t(int, DEFAULT_MAX_NUM_RSS_QUEUES, num_online_cpus());
+}
+EXPORT_SYMBOL(netif_get_num_default_rss_queues);
+
static inline void __netif_reschedule(struct Qdisc *q)
{
struct softnet_data *sd;
--
1.7.9.rc2
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox