On Tue, Apr 13, 2021 at 9:53 AM David Laight wrote:
>
> From: Matteo Croce
> > Sent: 12 April 2021 01:38
> >
> > Introduce skb_for_each_frag, an helper macro to iterate over the SKB frags.
>
> The real question is why, the change is:
>
> - for (i =
From: Matteo Croce
Coccinelle failed with the following error:
EXN: Failure("no position information") in net/core/dev.c
Apply it by hand as it's trivial.
Signed-off-by: Matteo Croce
---
net/core/dev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/c
From: Matteo Croce
use the new helper macro skb_for_each_frag() which allows to iterate
through all the SKB fragments.
The patch was created with Coccinelle, this was the semantic patch:
@@
struct sk_buff *skb;
identifier i;
statement S;
iterator name skb_for_each_frag;
@@
-for (i = 0; i
From: Matteo Croce
Introduce skb_for_each_frag, an helper macro to iterate over the SKB frags.
First patch introduces the helper, the second one is generated with
coccinelle and uses the macro where possible.
Last one is a chunk which have to be applied by hand.
The second patch raises some
From: Matteo Croce
Add an skb_for_each_frag() macro to iterate on SKB fragments.
Signed-off-by: Matteo Croce
---
include/linux/skbuff.h | 4
1 file changed, 4 insertions(+)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index dbf820a50a39..a8d4ccacdda5 100644
--- a/include
On Fri, Apr 9, 2021 at 11:28 PM Jakub Kicinski wrote:
>
> On Fri, 9 Apr 2021 22:44:50 +0200 Matteo Croce wrote:
> > > What pops to mind (although quite nit picky) is the question if the
> > > assembly changes much between driver which used to cache nr_frags and
> >
On Sat, Apr 10, 2021 at 2:11 AM Ilias Apalodimas
wrote:
>
> Hi Matteo,
>
> [...]
> > +bool page_pool_return_skb_page(void *data);
> > +
> > struct page_pool *page_pool_create(const struct page_pool_params *params);
> >
> > #ifdef CONFIG_PAGE_POOL
> > @@ -243,4 +247,13 @@ static inline void page_
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate increased di 10%,
from 269 Kpps to 296 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
21.78% [kernel] [k] __pi___inval_dcache_area
21.66% [mvneta
information we need for recycling in page->private and
enabling the recycling bit, or page_pool_store_mem_info() for a fragment
Since we added an extra argument on __skb_frag_unref() to handle
recycling, update the current users of the function with that.
Co-developed-by: Jesper Dangaard Bro
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate is more than doubled,
from 962 Kpps to 2047 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
30.67% [kernel] [k] page_pool_release_page
8.37% [kernel
From: Matteo Croce
This is needed by the page_pool to avoid recycling a page not allocated
via page_pool.
Signed-off-by: Matteo Croce
---
include/linux/mm_types.h | 1 +
include/net/page_pool.h | 2 ++
net/core/page_pool.c | 4
3 files changed, 7 insertions(+)
diff --git a/include
From: Jesper Dangaard Brouer
It is possible to compress/reduce the size of struct xdp_mem_info.
This change reduce struct xdp_mem_info from 8 bytes to 4 bytes.
The member xdp_mem_info.id can be reduced to u16, as the mem_id_ht
rhashtable in net/core/xdp.c is already limited by MEM_ID_MAX=0xFFFE
From: Matteo Croce
This is a respin of [1]
This patchset shows the plans for allowing page_pool to handle and
maintain DMA map/unmap of the pages it serves to the driver. For this
to work a return hook in the network core is introduced.
The overall purpose is to simplify drivers, by
On Fri, Apr 9, 2021 at 8:54 PM Jakub Kicinski wrote:
>
> On Fri, 9 Apr 2021 20:06:04 +0200 Matteo Croce wrote:
> > From: Matteo Croce
> >
> > use the new helper macro skb_for_each_frag() which allows to iterate
> > through all the SKB fragments.
> >
>
From: Matteo Croce
Coccinelle failed with the following error:
EXN: Failure("no position information") in net/core/dev.c
Apply it by hand as it's trivial.
Signed-off-by: Matteo Croce
---
net/core/dev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/c
From: Matteo Croce
use the new helper macro skb_for_each_frag() which allows to iterate
through all the SKB fragments.
The patch was created with Coccinelle, this was the semantic patch:
@@
struct sk_buff *skb;
identifier i;
statement S;
iterator name skb_for_each_frag;
@@
-for (i = 0; i
From: Matteo Croce
Add an skb_for_each_frag() macro to iterate on SKB fragments.
Signed-off-by: Matteo Croce
---
include/linux/skbuff.h | 4
1 file changed, 4 insertions(+)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index dbf820a50a39..a8d4ccacdda5 100644
--- a/include
From: Matteo Croce
Introduce skb_for_each_frag, an helper macro to iterate over the SKB frags.
First patch introduces the helper, the second one is generated with
coccinelle and uses the macro where possible.
Last one is a chunk which have to be applied by hand.
The second patch raises some
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate increased di 10%,
from 269 Kpps to 296 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
21.78% [kernel] [k] __pi___inval_dcache_area
21.66% [mvneta
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate is more than doubled,
from 962 Kpps to 2047 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
30.67% [kernel] [k] page_pool_release_page
8.37% [kernel
information we need for recycling in page->private and
enabling the recycling bit, or page_pool_store_mem_info() for a fragment
Since we added an extra argument on __skb_frag_unref() to handle
recycling, update the current users of the function with that.
Co-developed-by: Jesper Dangaard Br
From: Matteo Croce
This is needed by the page_pool to avoid recycling a page not allocated
via page_pool.
Signed-off-by: Matteo Croce
---
include/linux/mm_types.h | 1 +
include/net/page_pool.h | 2 ++
net/core/page_pool.c | 4
3 files changed, 7 insertions(+)
diff --git a/include
From: Jesper Dangaard Brouer
It is possible to compress/reduce the size of struct xdp_mem_info.
This change reduce struct xdp_mem_info from 8 bytes to 4 bytes.
The member xdp_mem_info.id can be reduced to u16, as the mem_id_ht
rhashtable in net/core/xdp.c is already limited by MEM_ID_MAX=0xFFFE
From: Matteo Croce
This is a respin of [1]
This patchset shows the plans for allowing page_pool to handle and
maintain DMA map/unmap of the pages it serves to the driver. For this
to work a return hook in the network core is introduced.
The overall purpose is to simplify drivers, by
bakin wrote:
> > > > From: Matteo Croce
> > > > Date: Mon, 22 Mar 2021 18:02:55 +0100
> > > >
> > > > > From: Matteo Croce
> > > > >
> > > > > This series enables recycling of the buffers allocated with the
>
On Mon, Mar 22, 2021 at 6:03 PM Matteo Croce wrote:
>
> From: Ilias Apalodimas
>
> During skb_release_data() intercept the packet and if it's a buffer
> coming from our page_pool API recycle it back to the pool for further
> usage.
> To achieve that we introdu
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate increased di 10%,
from 269 Kpps to 296 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
21.78% [kernel] [k] __pi___inval_dcache_area
21.66% [mvneta
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate is more than doubled,
from 962 Kpps to 2047 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
30.67% [kernel] [k] page_pool_release_page
8.37% [kernel
Dangaard Brouer
Signed-off-by: Matteo Croce
---
include/linux/skbuff.h | 33 +++
include/net/page_pool.h | 13 +
include/net/xdp.h | 1 +
net/core/page_pool.c| 43 +
net/core/skbuff.c | 20 +
From: Ilias Apalodimas
On a previous patch we added an extra argument on __skb_frag_unref() to
handle recycling. Update the current users of the function with that.
Signed-off-by: Ilias Apalodimas
Signed-off-by: Matteo Croce
---
drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c
From: Matteo Croce
This is needed by the page_pool to avoid recycling a page not allocated
via page_pool.
Signed-off-by: Matteo Croce
---
include/linux/mm_types.h | 1 +
include/net/page_pool.h | 2 ++
net/core/page_pool.c | 4
3 files changed, 7 insertions(+)
diff --git a/include
From: Jesper Dangaard Brouer
It is possible to compress/reduce the size of struct xdp_mem_info.
This change reduce struct xdp_mem_info from 8 bytes to 4 bytes.
The member xdp_mem_info.id can be reduced to u16, as the mem_id_ht
rhashtable in net/core/xdp.c is already limited by MEM_ID_MAX=0xFFFE
From: Matteo Croce
This series enables recycling of the buffers allocated with the page_pool API.
The first two patches are just prerequisite to save space in a struct and
avoid recycling pages allocated with other API.
Patch 2 was based on a previous idea from Jonathan Lemon.
The third one is
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate increased di 10%,
from 269 Kpps to 296 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
21.78% [kernel] [k] __pi___inval_dcache_area
21.66% [mvneta
From: Matteo Croce
Use the new recycling API for page_pool.
In a drop rate test, the packet rate is more than doubled,
from 962 Kpps to 2047 Kpps.
perf top on a stock system shows:
Overhead Shared Object Symbol
30.67% [kernel] [k] page_pool_release_page
8.37% [kernel
From: Ilias Apalodimas
On a previous patch we added an extra argument on __skb_frag_unref() to
handle recycling. Update the current users of the function with that.
Signed-off-by: Ilias Apalodimas
Signed-off-by: Matteo Croce
---
drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c
Dangaard Brouer
Signed-off-by: Matteo Croce
---
include/linux/skbuff.h | 33 +++
include/net/page_pool.h | 13 +
include/net/xdp.h | 1 +
net/core/page_pool.c| 43 +
net/core/skbuff.c | 20 +
From: Matteo Croce
This is needed by the page_pool to avoid recycling a page not allocated
via page_pool.
Signed-off-by: Matteo Croce
---
include/linux/mm_types.h | 1 +
include/net/page_pool.h | 2 ++
net/core/page_pool.c | 4
3 files changed, 7 insertions(+)
diff --git a/include
From: Jesper Dangaard Brouer
It is possible to compress/reduce the size of struct xdp_mem_info.
This change reduce struct xdp_mem_info from 8 bytes to 4 bytes.
The member xdp_mem_info.id can be reduced to u16, as the mem_id_ht
rhashtable in net/core/xdp.c is already limited by MEM_ID_MAX=0xFFFE
From: Matteo Croce
This series enables recycling of the buffers allocated with the page_pool API.
The first two patches are just prerequisite to save space in a struct and
avoid recycling pages allocated with other API.
Patch 2 was based on a previous idea from Jonathan Lemon.
The third one is
From: Matteo Croce
The ieee80211 class registers a callback which actually does nothing.
Given that the callback is optional, and all its accesses are protected
by a NULL check, remove it entirely.
Signed-off-by: Matteo Croce
---
net/wireless/sysfs.c | 7 ---
1 file changed, 7 deletions
On Sat, Feb 6, 2021 at 1:29 PM Matteo Croce wrote:
>
> From: Matteo Croce
>
> The ieee80211 class registers a callback which actually does nothing.
> Given that the callback is optional, and all its accesses are protected
> by a NULL check, remove it entirely.
>
> Sig
From: Matteo Croce
The ieee80211 class registers a callback which actually does nothing.
Given that the callback is optional, and all its accesses are protected
by a NULL check, remove it entirely.
Signed-off-by: Matteo Croce
---
net/wireless/sysfs.c | 7 ---
1 file changed, 7 deletions
PRIO'
>and use 'TCPOLEN_MPTCP_PRIO_ALIGN' instead, thanks to Matthieu Baerts.
> Changes since v1:
> - refactor patch to avoid using 'TCPOLEN_MPTCP_PRIO' with its old value,
>thanks to Geliang Tang.
>
> Fixes: 067065422fcd ("mptcp: add the outgoing MP_PRIO support")
> Reviewed-by: Mat Martineau
> Reviewed-by: Matthieu Baerts
> Signed-off-by: Davide Caratti
Reviewed-by: Matteo Croce
--
per aspera ad upstream
On Sat, Jan 16, 2021 at 5:41 AM David Ahern wrote:
>
> On 1/15/21 4:12 PM, Matteo Croce wrote:
> > On Fri, Jan 15, 2021 at 11:50 PM Jakub Kicinski wrote:
> >>
> >> On Fri, 15 Jan 2021 19:42:07 +0100 Matteo Croce wrote:
> >>> From: Matteo Croce
&
On Sat, Jan 16, 2021 at 5:36 AM David Ahern wrote:
>
> On 1/15/21 11:42 AM, Matteo Croce wrote:
> > From: Matteo Croce
> >
> > The ff00::/8 multicast route is created without specifying the fc_protocol
> > field, so the default RTPROT_BOOT value is used:
> >
On Fri, Jan 15, 2021 at 11:50 PM Jakub Kicinski wrote:
>
> On Fri, 15 Jan 2021 19:42:07 +0100 Matteo Croce wrote:
> > From: Matteo Croce
> >
> > Fix two wrong flags in the IPv6 multicast routes created
> > by the autoconf code.
>
> Any chance for Fixes tag
From: Matteo Croce
The multicast route ff00::/8 is created with type RTN_UNICAST:
$ ip -6 -d route
unicast ::1 dev lo proto kernel scope global metric 256 pref medium
unicast fe80::/64 dev eth0 proto kernel scope global metric 256 pref medium
unicast ff00::/8 dev eth0 proto kernel scope
From: Matteo Croce
Fix two wrong flags in the IPv6 multicast routes created
by the autoconf code.
Matteo Croce (2):
ipv6: create multicast route with RTPROT_KERNEL
ipv6: set multicast flag on the multicast route
net/ipv6/addrconf.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion
From: Matteo Croce
The ff00::/8 multicast route is created without specifying the fc_protocol
field, so the default RTPROT_BOOT value is used:
$ ip -6 -d route
unicast ::1 dev lo proto kernel scope global metric 256 pref medium
unicast fe80::/64 dev eth0 proto kernel scope global metric
From: Matteo Croce
ip has an ancient behaviour of looking at its program name to determine
the command to run. If the name is longer than 2 characters, the first two
letters are stripped and the others are interpreted as the command name:
$ ln -s /sbin/ip iproute
$ ln -s /sbin/ip ipa
On Fri, Nov 20, 2020 at 10:12 PM Jakub Kicinski wrote:
>
> On Thu, 19 Nov 2020 04:04:04 -0500 Eelco Chaudron wrote:
> > Currently, the openvswitch module is not accepting the correctly formated
> > netlink message for the TTL decrement action. For both setting and getting
> > the dec_ttl action, t
> From: Andrew Lunn
> Subject: [PATCH net-next] net: marvell: mvpp2: Fix W=1 warning with
> !CONFIG_ACPI
>
> Wrap the definition inside #ifdef/#endif.
>
> Compile tested only.
>
> Signed-off-by: Andrew Lunn
Looks good, ACPI_PTR() should be NULL if !CONFIG_ACPI
--
per aspera ad upstream
d apply too.
Acked-by: Matteo Croce
From: Matteo Croce
priv->page_pool is an array, so comparing against it will always return true.
Do a meaningful check by checking priv->page_pool[0] instead.
While at it, clear the page_pool pointers on deallocation, or when an
allocation error happens during init.
Reported-by: Colin Ia
On Mon, 6 Jul 2020 14:59:22 +0100
Colin Ian King wrote:
> Hi,
>
> Static analysis with Coverity has found a potential issue in the
> following commit:
>
> commit c2d6fe6163de80d7f7cf400ee351f56d6cdb7a5a
> Author: Matteo Croce
> Date: Thu Jul 2 16:12:43 2020 +020
On Thu, Jul 2, 2020 at 4:12 PM Matteo Croce wrote:
>
> From: Matteo Croce
>
> From: Matteo Croce
>
DUP! Sorry.
--
per aspera ad upstream
From: Sven Auhagen
Add ethtool statistics for XDP.
Signed-off-by: Sven Auhagen
---
drivers/net/ethernet/marvell/mvpp2/mvpp2.h| 8 +
.../net/ethernet/marvell/mvpp2/mvpp2_main.c | 158 --
2 files changed, 148 insertions(+), 18 deletions(-)
diff --git a/drivers/net/ethern
From: Matteo Croce
Add the transmit part of XDP support, which includes:
- support for XDP_TX in mvpp2_xdp()
- .ndo_xdp_xmit hook for AF_XDP and XDP_REDIRECT with mvpp2 as destination
mvpp2_xdp_submit_frame() is a generic function which is called by
mvpp2_xdp_xmit_back() when doing XDP_TX, and
From: Matteo Croce
Use the page_pool API for memory management.
This is a prerequisite for native XDP support.
Tested-by: Sven Auhagen
Signed-off-by: Matteo Croce
---
drivers/net/ethernet/marvell/Kconfig | 1 +
drivers/net/ethernet/marvell/mvpp2/mvpp2.h| 8 +
.../net
From: Matteo Croce
Add XDP native support.
By now only XDP_DROP, XDP_PASS and XDP_REDIRECT
verdicts are supported.
Co-developed-by: Sven Auhagen
Signed-off-by: Sven Auhagen
Signed-off-by: Matteo Croce
---
drivers/net/ethernet/marvell/mvpp2/mvpp2.h| 28 ++-
.../net/ethernet/marvell
From: Matteo Croce
From: Matteo Croce
Add XDP support to mvpp2. This series converts the driver to the
page_pool API for RX buffer management, and adds native XDP support.
XDP support comes with extack error reporting and statistics as well.
These are the performance numbers, as measured by
From: Matteo Croce
In mvpp2_swf_bm_pool_init_percpu(), a reference to a struct
mvpp2_bm_pool is obtained traversing multiple structs, when a
local variable already points to the same object.
Fix it and, while at it, give the variable a meaningful name.
Signed-off-by: Matteo Croce
---
.../net
Hi all,
Sven agreed to add the v2 of this patch to the series it depends on,
please drop this one.
I'm sending a v2 of the series soon.
Thanks,
On Wed, Jul 1, 2020 at 9:48 PM David Miller wrote:
>
> From: Sven Auhagen
> Date: Wed, 1 Jul 2020 17:30:44 +0200
>
> > static void mvpp2_read_stats(s
On Thu, Jul 2, 2020 at 11:14 AM Maciej Fijalkowski
wrote:
> > > + if (port->dev->mtu > ETH_DATA_LEN) {
> > > + netdev_err(port->dev, "Jumbo frames are not supported by XDP,
> > > current MTU %d.\n",
> > > + port->dev->mtu);
> >
> > ditto
>
> Here I agree and for e
On Thu, Jul 2, 2020 at 9:31 AM wrote:
>
> Hi Matteo,
>
> Thanks for working on this!
>
:)
> On Tue, Jun 30, 2020 at 08:09:28PM +0200, Matteo Croce wrote:
> > From: Matteo Croce
> > -static void *mvpp2_frag_alloc(const struct mvpp2_bm_pool *pool)
> > +/* R
From: Matteo Croce
Add XDP support to mvpp2. This series converts the driver to the
page_pool API for RX buffer management, and adds native XDP support.
These are the performance numbers, as measured by Sven:
SKB fwd page pool:
Rx bps 390.38 Mbps
Rx pps 762.46 Kpps
XDP fwd:
Rx bps
From: Matteo Croce
Add the transmit part of XDP support, which includes:
- support for XDP_TX in mvpp2_xdp()
- .ndo_xdp_xmit hook for AF_XDP and XDP_REDIRECT with mvpp2 as destination
mvpp2_xdp_submit_frame() is a generic function which is called by
mvpp2_xdp_xmit_back() when doing XDP_TX, and
From: Matteo Croce
Add XDP native support.
By now only XDP_DROP, XDP_PASS and XDP_REDIRECT
verdicts are supported.
Co-developed-by: Sven Auhagen
Signed-off-by: Sven Auhagen
Signed-off-by: Matteo Croce
---
drivers/net/ethernet/marvell/mvpp2/mvpp2.h| 28 ++-
.../net/ethernet/marvell
From: Matteo Croce
Use the page_pool API for memory management. This is a prerequisite for
native XDP support.
Tested-by: Sven Auhagen
Signed-off-by: Matteo Croce
---
drivers/net/ethernet/marvell/Kconfig | 1 +
drivers/net/ethernet/marvell/mvpp2/mvpp2.h| 8 +
.../net
From: Matteo Croce
In mvpp2_swf_bm_pool_init_percpu(), a reference to a struct
mvpp2_bm_pool is obtained traversing multiple structs, when a
local variable already points to the same object.
Fix it and, while at it, give the variable a meaningful name.
Signed-off-by: Matteo Croce
---
.../net
_early_drops: 0
rxq_2_packets_bm_drops: 0
rxq_3_desc_enqueue: 4
rxq_3_queue_full_drops: 0
rxq_3_packets_early_drops: 0
rxq_3_packets_bm_drops: 0
If you manage to find the commit which introduced this tag, please add
a Fixes tag.
Thanks,
--
Matteo Croce
perl -e 'for($t=0;;$t
.
>
> This eventually leads to a kernel crash.
>
> Signed-off-by: Sven Auhagen
Hi Sven,
Nice fix, if you think that I introduced it in
7d04b0b13b1175ce0c4bdc77f1278c1f120f874f,
please add a Fixes tag.
--
Matteo Croce
perl -e 'for($t=0;;$t++){print chr($t*($t>>8|$t>>13)&255)}' |aplay
On Wed, May 20, 2020 at 1:11 PM Russell King - ARM Linux admin
wrote:
>
> On Tue, May 19, 2020 at 07:05:34PM +0200, Matteo Croce wrote:
> > On Tue, 19 May 2020 12:05:20 +0200
> > Matteo Croce wrote:
> >
> > Hi,
> >
> > The patch seems to work. I&#
On Tue, 19 May 2020 12:05:20 +0200
Matteo Croce wrote:
> On Tue, May 19, 2020 at 11:54 AM Russell King - ARM Linux admin
> wrote:
> >
> > On Sat, May 09, 2020 at 09:20:50PM +0100, Russell King - ARM Linux
> > admin wrote:
> > > On Sat, May 09, 2020 at 08:
ort that feature, and the kernel restricts the packet size,
> > so we should never violate the MTU validator and end up with such a
> > packet. In any case, _if_ we were to attempt to transmit an oversized
> > packet, we have no support in the kernel to deal with that appearing
>
exit status
bpf_log_buf here seems to be a leftover, so removing it.
Signed-off-by: Matteo Croce
---
samples/bpf/lwt_len_hist_user.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/samples/bpf/lwt_len_hist_user.c b/samples/bpf/lwt_len_hist_user.c
index 587b68b1f8dd..430a4b7e353e 100644
On Sat, May 9, 2020 at 4:49 PM Russell King - ARM Linux admin
wrote:
>
> On Sat, May 09, 2020 at 02:51:05PM +0100, Russell King - ARM Linux admin
> wrote:
> > On Sat, May 09, 2020 at 03:14:05PM +0200, Matteo Croce wrote:
> > > On Sat, May 9, 2020 at 1:45 PM Russel
("net: mvpp2: cls: Use RSS contexts to handle RSS tables")
Signed-off-by: Matteo Croce
---
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
b/drivers/net/ethernet/marvell/mvp
On Sat, May 9, 2020 at 1:45 PM Russell King - ARM Linux admin
wrote:
>
> On Sat, May 09, 2020 at 11:15:58AM +, Stefan Chulski wrote:
> >
> >
> > > -Original Message-----
> > > From: Matteo Croce
> > > Sent: Saturday, May 9, 2020 3:13 AM
very effective, the throughput
goes 4x when enabled, so it would be a big drawback to disable it on
all interfaces.
Honestly I don't have any 2.5 gbit hardware to test it on eth3, so I
don't know if rxhash actually only works on the first interface of a
unit (so eth0 and eth1),
or if it just doesn't work on the gigabit one.
If someone could test it on the 2.5 gbit port, this will be helpful.
Regards,
--
Matteo Croce
per aspera ad upstream
On Thu, Apr 23, 2020 at 7:00 PM Russell King - ARM Linux admin
wrote:
>
> On Tue, Apr 14, 2020 at 01:43:02AM +0200, Matteo Croce wrote:
> > On Tue, Apr 14, 2020 at 1:21 AM Maxime Chevallier
> > wrote:
> > >
> > > The PPv2 controller has 8 RSS tables that ar
On Wed, Apr 15, 2020 at 1:48 AM David Miller wrote:
>
> From: Russell King - ARM Linux admin
> Date: Tue, 14 Apr 2020 20:47:53 +0100
>
> > This series fixes a problem with the 88x3310 PHY on Macchiatobin
> > coming out of powersave mode noticed by Matteo Croce. It se
Documents two piece of code which can't be understood at a glance.
Signed-off-by: Matteo Croce
---
include/net/flow_dissector.h | 1 +
net/core/flow_dissector.c| 6 ++
2 files changed, 7 insertions(+)
diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
On Tue, Jun 25, 2019 at 4:39 PM Matteo Croce wrote:
>
> On Thu, Jun 13, 2019 at 7:15 PM Matteo Croce wrote:
> >
> > On Wed, Jun 12, 2019 at 8:20 PM Stephen Hemminger
> > wrote:
> >
> > To me any path could work, both /tmp or in the current dir, I have no
>
);
only stop and restart if already running
> +
> + old_prog = xchg(&pp->xdp_prog, prog);
> + if (old_prog)
> + bpf_prog_put(old_prog);
> +
> + mvneta_open(dev);
^^
--
Matteo Croce
per aspera ad upstream
On Sat, Jul 27, 2019 at 12:00 AM Stephen Hemminger
wrote:
>
> Add the command alias "change" to man page.
> Don't show it on usage, since it is not commonly used.
>
> Reported-off-by: Matteo Croce
> Signed-off-by: Stephen Hemminger
> ---
> man/man8/ip-l
On July 26, 2019 8:25:14 PM GMT+02:00, Stephen Hemminger
wrote:
> On Wed, 24 Jul 2019 21:12:18 +0200
> Matteo Croce wrote:
>
> > ip link can set parameters both via the 'set' and 'change' keyword.
> > In fact, 'change' is an alias f
ip link can set parameters both via the 'set' and 'change' keyword.
In fact, 'change' is an alias for 'set'.
Document this in the help and manpage.
Fixes: 1d93483985f0 ("iplink: use netlink for link configuration")
Signed-off-by: Matteo Croce
---
ee times: the actual implementation
does 2 strlen and a memcpy to accomplish the same task.
Signed-off-by: Matteo Croce
---
include/utils.h | 2 +-
lib/utils.c | 15 ++-
2 files changed, 11 insertions(+), 6 deletions(-)
diff --git a/include/utils.h b/include/utils.h
index 1d9c1127..794d
On Tue, Jun 11, 2019 at 9:28 PM Stephen Hemminger
wrote:
>
> On Tue, 11 Jun 2019 20:05:13 +0200
> Matteo Croce wrote:
>
> > Pass the -pipe option to GCC, to use pipes instead of temp files.
> > On a slow AMD G-T40E CPU we get a non negligible 6% improvement
> &g
On Wed, Jul 10, 2019 at 1:18 AM Matteo Croce wrote:
>
> On Tue, Jul 9, 2019 at 11:38 PM Stephen Hemminger
> wrote:
> >
> > On Tue, 9 Jul 2019 22:40:40 +0200
> > Matteo Croce wrote:
> >
> > > iproute has an utility function which checks if a string i
On Tue, Jul 9, 2019 at 11:38 PM Stephen Hemminger
wrote:
>
> On Tue, 9 Jul 2019 22:40:40 +0200
> Matteo Croce wrote:
>
> > iproute has an utility function which checks if a string is a prefix for
> > another one, to allow use of abbreviated commands, e.g. 'addr&#x
ee times: the actual implementation
does 2 strlen and a memcpy to accomplish the same task.
Signed-off-by: Matteo Croce
---
include/utils.h | 2 +-
lib/utils.c | 14 +-
2 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/include/utils.h b/include/utils.h
index 927fdc17..f4d1
dadfailed home
tentative mngtmpaddr noprefixroute stable-privacy dummy0
valid_lft forever preferred_lft forever
Signed-off-by: Matteo Croce
---
net/ipv4/devinet.c | 8
1 file changed, 8 insertions(+)
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index c6bd0f7a020a
On Mon, Jul 1, 2019 at 2:38 PM Nicolas Dichtel
wrote:
>
> Le 30/06/2019 à 21:29, Matteo Croce a écrit :
> > When ip creates a netns, there is a small time interval between the
> > placeholder file creation in NETNS_RUN_DIR and the bind mount from /proc.
> >
> &
watchers only after the bind mount has been done.
Signed-off-by: Matteo Croce
---
ip/ipnetns.c | 37 ++---
1 file changed, 30 insertions(+), 7 deletions(-)
diff --git a/ip/ipnetns.c b/ip/ipnetns.c
index a883f210..23b95173 100644
--- a/ip/ipnetns.c
+++ b/ip
On Fri, Jun 28, 2019 at 7:06 PM Matteo Croce wrote:
>
> On Fri, Jun 28, 2019 at 6:27 PM David Howells wrote:
> >
> > Nicolas Dichtel wrote:
> >
> > > David Howells was working on a mount notification mechanism:
> > > https://lwn.net/Articles/760714/
this road, please either:
- disallow netns creation with names ending with .mounted
- or properly document it in the manpage
Regards,
--
Matteo Croce
per aspera ad upstream
On Thu, Jun 13, 2019 at 7:15 PM Matteo Croce wrote:
>
> On Wed, Jun 12, 2019 at 8:20 PM Stephen Hemminger
> wrote:
> >
> > On Wed, 12 Jun 2019 19:32:29 +0200
> > Matteo Croce wrote:
> >
> > > On Wed, Jun 12, 2019 at 6:04 PM Matteo Croce wrote:
>
1 - 100 of 231 matches
Mail list logo
