<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/linux.git/net/core/gro.c, branch linux-6.0.y</title>
<subtitle>Linux kernel stable tree (mirror)</subtitle>
<id>https://git.radix-linux.su/kernel/linux.git/atom?h=linux-6.0.y</id>
<link rel='self' href='https://git.radix-linux.su/kernel/linux.git/atom?h=linux-6.0.y'/>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/'/>
<updated>2022-05-16T09:18:56+00:00</updated>
<entry>
<title>net: allow gro_max_size to exceed 65536</title>
<updated>2022-05-16T09:18:56+00:00</updated>
<author>
<name>Alexander Duyck</name>
<email>alexanderduyck@fb.com</email>
</author>
<published>2022-05-13T18:34:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=0fe79f28bfaf73b66b7b1562d2468f94aa03bd12'/>
<id>urn:sha1:0fe79f28bfaf73b66b7b1562d2468f94aa03bd12</id>
<content type='text'>
Allow the gro_max_size to exceed a value larger than 65536.

There weren't really any external limitations that prevented this other
than the fact that IPv4 only supports a 16 bit length field. Since we have
the option of adding a hop-by-hop header for IPv6 we can allow IPv6 to
exceed this value and for IPv4 and non-TCP flows we can cap things at 65536
via a constant rather than relying on gro_max_size.

[edumazet] limit GRO_MAX_SIZE to (8 * 65535) to avoid overflows.

Signed-off-by: Alexander Duyck &lt;alexanderduyck@fb.com&gt;
Signed-off-by: Eric Dumazet &lt;edumazet@google.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net</title>
<updated>2022-03-11T01:16:56+00:00</updated>
<author>
<name>Jakub Kicinski</name>
<email>kuba@kernel.org</email>
</author>
<published>2022-03-11T01:16:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=1e8a3f0d2a1ef544611a7ea4a7c1512c732e0e43'/>
<id>urn:sha1:1e8a3f0d2a1ef544611a7ea4a7c1512c732e0e43</id>
<content type='text'>
net/dsa/dsa2.c
  commit afb3cc1a397d ("net: dsa: unlock the rtnl_mutex when dsa_master_setup() fails")
  commit e83d56537859 ("net: dsa: replay master state events in dsa_tree_{setup,teardown}_master")
https://lore.kernel.org/all/20220307101436.7ae87da0@canb.auug.org.au/

drivers/net/ethernet/intel/ice/ice.h
  commit 97b0129146b1 ("ice: Fix error with handling of bonding MTU")
  commit 43113ff73453 ("ice: add TTY for GNSS module for E810T device")
https://lore.kernel.org/all/20220310112843.3233bcf1@canb.auug.org.au/

drivers/staging/gdm724x/gdm_lte.c
  commit fc7f750dc9d1 ("staging: gdm724x: fix use after free in gdm_lte_rx()")
  commit 4bcc4249b4cf ("staging: Use netif_rx().")
https://lore.kernel.org/all/20220308111043.1018a59d@canb.auug.org.au/

Signed-off-by: Jakub Kicinski &lt;kuba@kernel.org&gt;
</content>
</entry>
<entry>
<title>net: Fix esp GSO on inter address family tunnels.</title>
<updated>2022-03-07T12:14:04+00:00</updated>
<author>
<name>Steffen Klassert</name>
<email>steffen.klassert@secunet.com</email>
</author>
<published>2022-03-07T12:11:41+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=23c7f8d7989e1646aac82f75761b7648c355cb8a'/>
<id>urn:sha1:23c7f8d7989e1646aac82f75761b7648c355cb8a</id>
<content type='text'>
The esp tunnel GSO handlers use skb_mac_gso_segment to
push the inner packet to the segmentation handlers.
However, skb_mac_gso_segment takes the Ethernet Protocol
ID from 'skb-&gt;protocol' which is wrong for inter address
family tunnels. We fix this by introducing a new
skb_eth_gso_segment function.

This function can be used if it is necessary to pass the
Ethernet Protocol ID directly to the segmentation handler.
First users of this function will be the esp4 and esp6
tunnel segmentation handlers.

Fixes: c35fe4106b92 ("xfrm: Add mode handlers for IPsec on layer 2")
Signed-off-by: Steffen Klassert &lt;steffen.klassert@secunet.com&gt;
</content>
</entry>
<entry>
<title>net: gro: minor optimization for dev_gro_receive()</title>
<updated>2022-02-05T15:13:52+00:00</updated>
<author>
<name>Paolo Abeni</name>
<email>pabeni@redhat.com</email>
</author>
<published>2022-02-04T11:28:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=de5a1f3ce4c862150dc442530dba19e1d1dc6bc2'/>
<id>urn:sha1:de5a1f3ce4c862150dc442530dba19e1d1dc6bc2</id>
<content type='text'>
While inspecting some perf report, I noticed that the compiler
emits suboptimal code for the napi CB initialization, fetching
and storing multiple times the memory for flags bitfield.
This is with gcc 10.3.1, but I observed the same with older compiler
versions.

We can help the compiler to do a nicer work clearing several
fields at once using an u32 alias. The generated code is quite
smaller, with the same number of conditional.

Before:
objdump -t net/core/gro.o | grep " F .text"
0000000000000bb0 l     F .text	0000000000000357 dev_gro_receive

After:
0000000000000bb0 l     F .text	000000000000033c dev_gro_receive

v1  -&gt; v2:
 - use struct_group (Alexander and Alex)

RFC -&gt; v1:
 - use __struct_group to delimit the zeroed area (Alexander)

Signed-off-by: Paolo Abeni &lt;pabeni@redhat.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>net: gro: avoid re-computing truesize twice on recycle</title>
<updated>2022-02-05T15:13:52+00:00</updated>
<author>
<name>Paolo Abeni</name>
<email>pabeni@redhat.com</email>
</author>
<published>2022-02-04T11:28:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=7881453e4adf497cf9109c84fa21eedda9ac6164'/>
<id>urn:sha1:7881453e4adf497cf9109c84fa21eedda9ac6164</id>
<content type='text'>
After commit 5e10da5385d2 ("skbuff: allow 'slow_gro' for skb
carring sock reference") and commit af352460b465 ("net: fix GRO
skb truesize update") the truesize of the skb with stolen head is
properly updated by the GRO engine, we don't need anymore resetting
it at recycle time.

v1 -&gt; v2:
 - clarify the commit message (Alexander)

Signed-off-by: Paolo Abeni &lt;pabeni@redhat.com&gt;
Reviewed-by: Eric Dumazet &lt;edumazet@google.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>gro: add ability to control gro max packet size</title>
<updated>2022-01-06T12:27:05+00:00</updated>
<author>
<name>Coco Li</name>
<email>lixiaoyan@google.com</email>
</author>
<published>2022-01-05T10:48:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=eac1b93c14d645ef147b049ace0d5230df755548'/>
<id>urn:sha1:eac1b93c14d645ef147b049ace0d5230df755548</id>
<content type='text'>
Eric Dumazet suggested to allow users to modify max GRO packet size.

We have seen GRO being disabled by users of appliances (such as
wifi access points) because of claimed bufferbloat issues,
or some work arounds in sch_cake, to split GRO/GSO packets.

Instead of disabling GRO completely, one can chose to limit
the maximum packet size of GRO packets, depending on their
latency constraints.

This patch adds a per device gro_max_size attribute
that can be changed with ip link command.

ip link set dev eth0 gro_max_size 16000

Suggested-by: Eric Dumazet &lt;edumazet@google.com&gt;
Signed-off-by: Coco Li &lt;lixiaoyan@google.com&gt;
Signed-off-by: Eric Dumazet &lt;edumazet@google.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>net: gro: populate net/core/gro.c</title>
<updated>2021-11-16T13:16:54+00:00</updated>
<author>
<name>Eric Dumazet</name>
<email>edumazet@google.com</email>
</author>
<published>2021-11-15T17:05:54+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=587652bbdd06ab38a4c1b85e40f933d2cf4a1147'/>
<id>urn:sha1:587652bbdd06ab38a4c1b85e40f933d2cf4a1147</id>
<content type='text'>
Move gro code and data from net/core/dev.c to net/core/gro.c
to ease maintenance.

gro_normal_list() and gro_normal_one() are inlined
because they are called from both files.

Signed-off-by: Eric Dumazet &lt;edumazet@google.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
<entry>
<title>net: gro: move skb_gro_receive into net/core/gro.c</title>
<updated>2021-11-16T13:16:54+00:00</updated>
<author>
<name>Eric Dumazet</name>
<email>edumazet@google.com</email>
</author>
<published>2021-11-15T17:05:53+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=e456a18a390b96f22b0de2acd4d0f49c72ed2280'/>
<id>urn:sha1:e456a18a390b96f22b0de2acd4d0f49c72ed2280</id>
<content type='text'>
net/core/gro.c will contain all core gro functions,
to shrink net/core/skbuff.c and net/core/dev.c

Signed-off-by: Eric Dumazet &lt;edumazet@google.com&gt;
Signed-off-by: David S. Miller &lt;davem@davemloft.net&gt;
</content>
</entry>
</feed>
