On Wed, Oct 21, 2015 at 03:38:51PM +0100, Alan Burlison wrote:
> On 21/10/2015 04:49, Al Viro wrote:
> >BTW, for real fun, consider this:
> >7)
> >// fd is a socket
> >fd2 = dup(fd);
> >in thread A: accept(fd);
> >in thread B: accept(fd);
> >in thread C: accept(fd2);
> >in thread D:
From: Eric Dumazet
poll(POLLOUT) on a listener should not report fd is ready for
a write().
This would break some applications using poll() and pfd.events = -1,
as they would not block in poll()
Signed-off-by: Eric Dumazet
Reported-by: Alan Burlison
On Fri, Oct 16, 2015 at 2:36 PM, Florian Fainelli wrote:
> On 16/10/15 13:35, Iyappan Subramanian wrote:
>> Add RGMII TX/RX delay configuration support. RGMII standard requires 2ns
>> delay to help the RGMII bridge receiver to sample data correctly. If the
>> default value
Mainline commit 89c22d8c3b27 ("net: Fix skb csum races when peeking")
backport into pre-3.19 stable kernels introduces a regression causing
null pointer dererefence in skb_copy_and_csum_datagram_iovec().
This commit only sets CHECKSUM_UNNECESSARY for non-shared skb, allowing
udp_recvmsg() to take
> -Original Message-
> From: netdev-ow...@vger.kernel.org [mailto:netdev-
> ow...@vger.kernel.org] On Behalf Of Joe Jin
> Sent: 23 October 2015 08:54
> To: Wei Liu; Ian Campbell; Boris Ostrovsky; Konrad Rzeszutek Wilk; David S.
> Miller
> Cc: netdev@vger.kernel.org;
On 10/23/2015 04:47 PM, Paul Durrant wrote:
>> -Original Message-
>> From: netdev-ow...@vger.kernel.org [mailto:netdev-
>> ow...@vger.kernel.org] On Behalf Of Joe Jin
>> Sent: 23 October 2015 08:54
>> To: Wei Liu; Ian Campbell; Boris Ostrovsky; Konrad Rzeszutek Wilk; David S.
>> Miller
>>
On Fri, 23 Oct 2015 10:30:24 -0700, Pravin Shelar wrote:
> I see lot of refactoring scope for vxlan code even without this patch.
> I am planing to address it in net-next.
Great, thanks a lot!
> > What about IPv6? There's IPv6 support for metadata based vxlan in
> > net.git, thus this should
Add support for the FDB add, delete, and dump operations. The add and
delete operations are implemented using directed ARL operations using
the specified MAC address and consist in a read operation, write and
readback operation.
The dump operation consists in using the ARL search and software
This patchset adds support for VFs of Netronome's NFP-4000 and NFP-6000
based NICs. We are currently also preparing the submission for the PF
driver, but it is not quite ready yet. The PF driver can be found on
GitHub:
https://github.com/Netronome/nfp-drv-kmods
Jakub Kicinski (2):
pci_ids: add
Add PCI vendor id for Netronome Systems.
Signed-off-by: Jakub Kicinski
Signed-off-by: Rolf Neugebauer
---
include/linux/pci_ids.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
struct bnxt_pf_info needs to be always defined. Move bnxt_update_vf_mac()
to bnxt_sriov.c and add some missing #ifdef CONFIG_BNXT_SRIOV.
Reported-by: Jim Hull
Tested-by: Jim Hull
Signed-off-by: Michael Chan
---
On Fri, 2015-10-23 at 11:36 -0700, Alexander Duyck wrote:
> On 10/21/2015 09:37 AM, Lan Tianyu wrote:
> > This patchset is to propose a new solution to add live migration support
> > for 82599
> > SRIOV network card.
> >
> > Im our solution, we prefer to put all device specific operation into VF
>>> On 23.10.15 at 11:05, wrote:
> On 10/23/2015 04:47 PM, Paul Durrant wrote:
>>> -Original Message-
>>> From: netdev-ow...@vger.kernel.org [mailto:netdev-
>>> ow...@vger.kernel.org] On Behalf Of Joe Jin
>>> Sent: 23 October 2015 08:54
>>> To: Wei Liu; Ian Campbell;
On Wed, Oct 21, 2015 at 11:48:25AM -0400, Sowmini Varadhan wrote:
>
> On sparc, deleting established SAs (e.g., by restarting ipsec)
> results in unaligned access messages via xfrm_del_sa ->
> km_state_notify -> xfrm_send_state_notify().
>
> Even though struct xfrm_usersa_info is aligned on
Should not allocate vnic queues number more than online cpus.
Signed-off-by: Joe Jin
Cc: Jan Beulich
Cc: Wei Liu
Cc: Ian Campbell
Cc: Boris Ostrovsky
Cc: Konrad Rzeszutek Wilk
Should not allocate xen vif queues number more than online cpus.
Signed-off-by: Joe Jin
Cc: Jan Beulich
Cc: Wei Liu
Cc: Ian Campbell
Cc: Boris Ostrovsky
Cc: Konrad Rzeszutek Wilk
From: Jason Wang
Date: Fri, 23 Oct 2015 00:57:05 -0400
> We don't have fraglist support in TAP_FEATURES. This will lead
> software segmentation of gro skb with frag list. Fixes by having
> frag list support in TAP_FEATURES.
>
> With this patch single session of netperf
From: Andrew Shewmaker
Date: Sun, 18 Oct 2015 21:59:08 -0700
> If alpha is strictly reduced by alpha >> dctcp_shift_g and if alpha is less
> than 1 << dctcp_shift_g, then alpha may never reach zero. For example,
> given shift_g=4 and alpha=15, alpha >> dctcp_shift_g yields 0
> -Original Message-
> From: Joe Jin [mailto:joe@oracle.com]
> Sent: 23 October 2015 10:05
> To: Paul Durrant; Wei Liu; Ian Campbell; Boris Ostrovsky; Konrad Rzeszutek
> Wilk; David S. Miller
> Cc: netdev@vger.kernel.org; xen-de...@lists.xenproject.org
> Subject: Re: [PATCH 0/2] limit
On Thu, 2015-10-22 at 20:56 +0200, Thomas Graf wrote:
> On 10/22/15 at 07:21pm, Hannes Frederic Sowa wrote:
> > Hi Thomas,
> >
> > On Thu, Oct 22, 2015, at 18:45, Thomas Graf wrote:
> > > I understand the race but when does it occur? Whoever creates
> > > the original interface owns it and is
vnet_hdr_len cannot be negative and is use in operation/function that
wait for unsigned value.
This patch set vnet_hdr_len as size_t.
Signed-off-by: LABBE Corentin
---
net/packet/af_packet.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
length is used in operation/function that wait for unsigned value.
Furthermore the only one call of rawv6_send_hdrinc() give a size_t as
length arguments.
So the parameter need to be set as size_t.
Signed-off-by: LABBE Corentin
---
net/ipv6/raw.c | 2 +-
1 file
The hlen member of raw6_frag_vec is used in operation/function that
wait for unsigned value. So it need to be set as size_t.
This patch do the same for the hlen variable.
Signed-off-by: LABBE Corentin
---
net/ipv6/raw.c | 4 ++--
1 file changed, 2 insertions(+), 2
chunk cannot be negative and is use in operation/function that
wait for unsigned value. This patch set it as size_t.
The patch do the same for the size variable.
Signed-off-by: LABBE Corentin
---
net/caif/caif_socket.c | 5 +++--
1 file changed, 3 insertions(+), 2
len is used in operation/function that wait for unsigned value.
Furthermore the only one call of sco_send_frame give a size_t as argument.
So the parameter need to be set as size_t.
Signed-off-by: LABBE Corentin
---
net/bluetooth/sco.c | 2 +-
1 file changed, 1
From: roopa
Date: Fri, 23 Oct 2015 05:14:32 -0700
> David, please ignore. I will re-spin the series. I see a checkpatch error
Ok.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
More majordomo
On 10/23/15 5:03 AM, Peter Zijlstra wrote:
So the bpf_perf_event_read() returns the count value, does this not also
mean that returning -EINVAL here is also 'wrong'?
I mean, sure an actual count value that high is unlikely, but its still
a broken interface.
Agree. that's not pretty interface.
On 23/10/2015 17:00, Eric Dumazet wrote:
Ermm, you *really* want me to submit a patch removing 'Conforms to
POSIX.1-2001' from *every* Linux manpage?
Only on the pages you think there is an error that matters.
If there's consensus that the current shutdown(), dup2(), close() and
accept()
On Fri, 2015-10-23 at 09:00 -0700, Eric Dumazet wrote:
> on the pages you think there is an error that matters.
>
> >
> > > Have you tested the patch I sent ?
> >
> > The AF_UNIX poll one? No, I don't have the means to do so, and in any
> > case that's not a POSIX issue, just a plain bug. I'm
On 23/10/2015 15:21, Eric Dumazet wrote:
I claim nothing. If you believe a man page should be fixed, please send
a patch to man page maintainer.
Ermm, you *really* want me to submit a patch removing 'Conforms to
POSIX.1-2001' from *every* Linux manpage?
Have you tested the patch I sent ?
On Fri, 2015-10-23 at 16:46 +0100, Alan Burlison wrote:
> On 23/10/2015 15:21, Eric Dumazet wrote:
>
> > I claim nothing. If you believe a man page should be fixed, please send
> > a patch to man page maintainer.
>
> Ermm, you *really* want me to submit a patch removing 'Conforms to
>
>>> On 23.10.15 at 11:19, wrote:
>
> @@ -164,6 +166,19 @@ struct netfront_rx_info {
> struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1];
> };
>
> +static int xennet_set_max_queues(const char *val, struct kernel_param *kp)
> +{
> + unsigned int
Currently xen vnic allowed to create lots of queues by set module param
max_queues(both netback and netfront), when queues number larger than
cpus number, it does not help for performance but need more cpu time.
This patchset limit netback and netfront max queues number to online
cpus number.
From: Nathan Sullivan
Date: Wed, 21 Oct 2015 14:17:04 -0500
> Very rarely, the KSZ9031 will appear to complete autonegotiation, but
> will drop all traffic afterwards. When this happens, the idle error
> count will read 0xFF after autonegotiation completes. Reset the
On Thu 22-10-15 00:21:31, Johannes Weiner wrote:
> The tcp memory controller has extensive provisions for future memory
> accounting interfaces that won't materialize after all. Cut the code
> base down to what's actually used, now and in the likely future.
>
> - There won't be any different
The network stack defers SKBs free, in-case free happens in IRQ or
when IRQs are disabled. This happens in __dev_kfree_skb_irq() that
writes SKBs that were free'ed during IRQ to the softirq completion
queue (softnet_data.completion_queue).
These SKBs are naturally delayed, and cleaned up during
Think twice before applying
- This patch can potentially introduce added latency in some workloads
This patch introduce bulk alloc of SKBs and allow reuse of SKBs
free'ed in same softirq cycle. SKBs are normally free'ed during TX
completion, but most high speed drivers also cleanup TX ring
It have been a long road. Back in July 2014 I realized that network
stack were hitting the kmem_cache/SLUB slowpath when freeing SKBs, but
had no solution. In Dec 2014 I had implemented a solution called
qmempool[1], that showed it was possible to improve this, but got
rejected due to being a
From: LABBE Corentin
Date: Fri, 23 Oct 2015 14:10:35 +0200
> The hlen member of raw_frag_vec is used in operation/function that wait
> for unsigned value. So it need to be set as size_t.
>
> Signed-off-by: LABBE Corentin
You really need to
On Thu, Oct 22, 2015 at 06:28:22PM +0800, Wangnan (F) wrote:
> information to analysis when glitch happen. Another way we are trying to
> implement
> now is to dynamically turn events on and off, or at least enable/disable
> sampling dynamically because the overhead of copying those samples
> is a
From: sfel...@gmail.com
Date: Thu, 22 Oct 2015 20:25:17 -0700
> From: Scott Feldman
>
> This was already done for switchdev_objs. Changing switchdev_attrs to new
> style makes switchdev API consistent for both attrs and objs.
>
> No functional changes here.
>
>
From: sfel...@gmail.com
Date: Thu, 22 Oct 2015 20:25:16 -0700
> From: Scott Feldman
>
> Fixes: 4d429c5dd ("switchdev: introduce possibility to defer obj_add/del")
> Signed-off-by: Scott Feldman
> Acked-by: Jiri Pirko
Applied.
--
To
From: Jean Sacren
The commit 3092e5e4cc79 ("i40e: add little endian conversion for
checksum") fixed the checksum bug on big-endian architecture.
But we should not execute cpu_to_le16() unconditionally. Thus, put
cpu_to_le16() under certain condition.
Cc: Jesse Brandeburg
From: Carolyn Wyborny
This patch fixes a problem where the driver output of the OEM
version string varied from the other tools. The mask value
and the order of operations were incorrect, per the original
change request. Without this patch, the version string will
On Thu 22-10-15 00:21:33, Johannes Weiner wrote:
> Socket memory can be a significant share of overall memory consumed by
> common workloads. In order to provide reasonable resource isolation
> out-of-the-box in the unified hierarchy, this type of memory needs to
> be accounted and tracked per
From: Jean Sacren
The following kernel-doc arguments for their respective functions are
missing:
1) @cd_type_cmd_tso_mss for i40e_tso();
2) @cd_type_cmd_tso_mss for i40e_tsyn();
3) @tx_ring for i40e_tx_enable_csum().
Add them all for the kernel-doc requirement.
From: Hiroshi Shimamoto
Add netlink directives and ndo entry to trust VF user.
This controls the special permission of VF user.
The administrator will dedicatedly trust VF user to use some features
which impacts security and/or performance.
The administrator never
On Fri, Oct 23, 2015 at 03:13:07PM +0800, Jason Wang wrote:
>
>
> On 10/22/2015 05:33 PM, Michael S. Tsirkin wrote:
> > On Thu, Oct 22, 2015 at 01:27:29AM -0400, Jason Wang wrote:
> >> This patch tries to poll for new added tx buffer for a while at the
> >> end of tx processing. The maximum time
On Thu, Oct 22, 2015 at 03:09:43PM -0400, Johannes Weiner wrote:
> On Thu, Oct 22, 2015 at 09:46:12PM +0300, Vladimir Davydov wrote:
> > On Thu, Oct 22, 2015 at 12:21:31AM -0400, Johannes Weiner wrote:
> > > The tcp memory controller has extensive provisions for future memory
> > > accounting
On Fri, 2015-10-23 at 15:20 +0200, casper@oracle.com wrote:
> This point was not about a 10M sockets server but in general...
>
Hey, we do not design the OS only for handling desktop workloads.
If netstat does not work on this typical server workload, then it is a
joke.
> Solaris has had
On 10/22/2015 09:35 PM, Eric W. Biederman wrote:
Daniel Borkmann writes:
On 10/20/2015 08:56 PM, Eric W. Biederman wrote:
...
Just FYI: Using a device for this kind of interface is pretty
much a non-starter as that quickly gets you into situations where
things do not
On Fri, 2015-10-23 at 15:20 +0200, casper@oracle.com wrote:
>
> >Yet another POSIX deficiency.
> >
> >When a server deals with 10,000,000+ socks, we absolutely do not care of
> >this requirement.
> >
> >O(log(n)) is still crazy if it involves O(log(n)) cache misses.
>
> You miss the fire
There is an opportunity to bulk free SKBs during reclaiming of
resources after DMA transmit completes in ixgbe_clean_tx_irq. Thus,
bulk freeing at this point does not introduce any added latency.
Simply use napi_consume_skb() which were recently introduced. The
napi_budget parameter is needed
Discovered that network stack were hitting the kmem_cache/SLUB
slowpath when freeing SKBs. Doing bulk free with kmem_cache_free_bulk
can speedup this slowpath.
NAPI context is a bit special, lets take advantage of that for bulk
free'ing SKBs.
In NAPI context we are running in softirq, which
From: Roopa Prabhu
This patch adds support for MPLS multipath routes.
Includes following changes to support multipath:
- splits struct mpls_route into 'struct mpls_route + struct mpls_nh'
- 'struct mpls_nh' represents a mpls nexthop label forwarding entry
- moves
From: Hiroshi Shimamoto
The limitation of the number of multicast address for VF is not enough
for the large scale server with SR-IOV feature. IPv6 requires the multicast
MAC address for each IP address to handle the Neighbor Solicitation
message. We couldn't assign
From: Akeem G Abodunrin
If a call to enable SR-IOV in the kernel failed, we need to disable
I40E_FLAG_VEB_MODE_ENABLED, so that bridge mode could fall back to
VEPA, which is a default.
Change-ID: I12b6f776769506db85b29bea94b9c88d0b5ee65e
Signed-off-by: Akeem G
From: Catherine Sullivan
Bump up the version...
Change-ID: Ib8d501021671ba20250115ed54330e2c182255b7
Signed-off-by: Catherine Sullivan
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
From: Mitch Williams
Not sure how this compiles at all. Use the correct struct for
manipulating the VLAN filter list. Without this, the VLAN filter
list doesn't get processed correctly, and VLAN filters will not
be re-enabled after any kind of reset.
Change-ID:
This series contains updates to i40e, i40evf, if_link, ixgbe and ixgbevf.
Anjali adds a workaround to drop any flow control frames from being
transmitted from any VSI, so that a malicious VF cannot send flow control
or PFC packets out on the wire. Also fixed a bug in debugfs by grabbing
the
From: Helin Zhang
This patch fixes a problem of possibly getting inconsistent flow control
statuses after a PF reset. Requested_mode was being set with a default
value during probing, but the initial HW state could be different from
this mode.
Change-ID:
From: Jean Sacren
In both i40e_calc_nvm_checksum() and i40e_update_nvm_checksum(), the
local variables designated by 'ret_code' are overwritten immediately. As
such, they should merely be declared.
Signed-off-by: Jean Sacren
Tested-by: Andrew Bowers
From: Jean Sacren
@flush has been missing since the inception of i40evf_irq_enable(). Add
it for the kernel doc.
Signed-off-by: Jean Sacren
Tested-by: Andrew Bowers
Acked-by: Jesse Brandeburg
From: Hiroshi Shimamoto
Implements the new netdev op to trust VF in ixgbe.
The administrator can turn on and off VF trusted by ip command which
supports trust message.
# ip link set dev eth0 vf 1 trust on
or
# ip link set dev eth0 vf 1 trust off
Send a ping to
From: Anjali Singhai Jain
This patch adds a workaround to drop any flow control frames from being
transmitted from any VSI. FW can still send flow control frames if flow
control is enabled.
With this patch in place a malicious VF cannot send flow control or PFC
packets
From: Andy Shevchenko
Instead of using a custom approach change the code to use %*ph format
specifier.
Signed-off-by: Andy Shevchenko
Tested-by: Andrew Bowers
Acked-by: Jesse Brandeburg
From: Jean Sacren
In i40evf_msix_aq(), the first two lines of rd32() are mainly to clear
the registers. If we initialize 'val' at this point, it will be
overwritten immediately. We shall simply discard the return value here.
When we initialize 'val', we might as well include
From: Akeem G Abodunrin
Fix i40e_is_vsi_uplink_mode_veb to check if bridge is actually
in VEB mode before allowing LB in the add VSI routine, instead of
unconditionally returning VEB bridge mode.
Change-ID: I162397b1bdd02367735fe9baaeb51465be2a3ce9
Signed-off-by:
From: Anjali Singhai Jain
The new code flow requires us to grab the filter list lock before
adding/deleting the filter.
Change-ID: I4eaef508ab4da2d1b2e23f20f2a78d931d5b6aeb
Signed-off-by: Anjali Singhai Jain
Tested-by: Andrew Bowers
On Fri, Oct 23, 2015 at 12:57:05AM -0400, Jason Wang wrote:
> We don't have fraglist support in TAP_FEATURES. This will lead
> software segmentation of gro skb with frag list. Fixes by having
> frag list support in TAP_FEATURES.
>
> With this patch single session of netperf receiving were
From: David Miller
Date: Fri, 23 Oct 2015 06:32:44 -0700 (PDT)
> From: sfel...@gmail.com
> Date: Thu, 22 Oct 2015 20:25:17 -0700
>
>> From: Scott Feldman
>>
>> This was already done for switchdev_objs. Changing switchdev_attrs to new
>> style makes
From: Jeff Kirsher
Date: Fri, 23 Oct 2015 06:17:42 -0700
> This series contains updates to i40e, i40evf, if_link, ixgbe and ixgbevf.
Looks good, pulled, thanks a lot Jeff.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message
From: Tom Lendacky
Date: Fri, 23 Oct 2015 08:35:52 -0500
> Can you queue this up for stable? It is applicable to 4.1 and 4.2.
Ok, done.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
More
Hi LABBE,
[auto build test WARNING on net/master -- if it's inappropriate base, please
suggest rules for selecting the more suitable base]
url:
https://github.com/0day-ci/linux/commits/LABBE-Corentin/net-change-len-parameter-type-for-memcpy_-to-from-_msg/20151023-201642
config: tile
From: Li RongQing
the returned buffer of register_sysctl() is stored into net_header
variable, but net_header is not used after, and compiler maybe
optimise the variable out, and lead kmemleak reported the below warning
comm "swapper/0", pid 1, jiffies 4294937448
On Fri, Oct 23, 2015 at 06:08:52AM -0700, David Miller wrote:
> From: LABBE Corentin
> Date: Fri, 23 Oct 2015 14:10:35 +0200
>
> > The hlen member of raw_frag_vec is used in operation/function that wait
> > for unsigned value. So it need to be set as size_t.
> >
> >
From: LABBE Corentin
Date: Fri, 23 Oct 2015 15:21:00 +0200
> I have added all mailing list given by get_maintainer but this was done per
> patch with --to-cmd/--cc-cmd
> Does it is better to put all mailing list globally ?
>
> For the (0/N) mail, I forgot to add a
From: Michal Hocko
Date: Fri, 23 Oct 2015 15:19:56 +0200
> On Thu 22-10-15 00:21:33, Johannes Weiner wrote:
>> Socket memory can be a significant share of overall memory consumed by
>> common workloads. In order to provide reasonable resource isolation
>> out-of-the-box in the
On Thu 22-10-15 00:21:35, Johannes Weiner wrote:
> The vmpressure metric is based on reclaim efficiency, which in turn is
> an attribute of the LRU. However, vmpressure events are currently
> reported at the source of pressure rather than at the reclaim level.
>
> Switch the reporting to the
Hi LABBE,
[auto build test WARNING on net/master -- if it's inappropriate base, please
suggest rules for selecting the more suitable base]
url:
https://github.com/0day-ci/linux/commits/LABBE-Corentin/net-change-len-parameter-type-for-memcpy_-to-from-_msg/20151023-201642
config: xtensa
Hi LABBE,
[auto build test WARNING on net/master -- if it's inappropriate base, please
suggest rules for selecting the more suitable base]
url:
https://github.com/0day-ci/linux/commits/LABBE-Corentin/net-change-len-parameter-type-for-memcpy_-to-from-_msg/20151023-201642
config: tile
From: Neil Armstrong
Date: Thu, 22 Oct 2015 10:37:45 +0200
> In order to avoid locked signal false positive for nested mdiobus
> read/write calls, nested code was introduced in mv88e6xxx and
> mdio-mux.
> But mv88e6060 also needs such nested mdiobus read/write calls.
>
From: Roopa Prabhu
Date: Fri, 23 Oct 2015 06:03:26 -0700
> This patch adds support for MPLS multipath routes.
Series applied, thanks a lot Roopa.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
From: sfel...@gmail.com
Date: Thu, 22 Oct 2015 20:25:15 -0700
> From: Scott Feldman
>
> When adding vlans with multiple IFLA_BRIDGE_VLAN_INFO attrs set in AFSPEC,
> we would wipe the vlan obj struct after the first IFLA_BRIDGE_VLAN_INFO.
> Fix this by only clearing what's
From: roy.qing...@gmail.com
Date: Fri, 23 Oct 2015 20:59:49 +0800
> From: Li RongQing
>
> the returned buffer of register_sysctl() is stored into net_header
> variable, but net_header is not used after, and compiler maybe
> optimise the variable out, and lead kmemleak
Hi LABBE,
[auto build test WARNING on net/master -- if it's inappropriate base, please
suggest rules for selecting the more suitable base]
url:
https://github.com/0day-ci/linux/commits/LABBE-Corentin/net-change-len-parameter-type-for-memcpy_-to-from-_msg/20151023-201642
config: x86_64
On Fri, 2015-10-23 at 11:52 +0200, casper@oracle.com wrote:
>
> >Ho-hum... It could even be made lockless in fast path; the problems I see
> >are
> > * descriptor-to-file lookup becomes unsafe in a lot of locking
> >conditions. Sure, most of that happens on the entry to some syscall,
From: Roopa Prabhu
This patch adds support for MPLS multipath routes.
Includes following changes to support multipath:
- splits struct mpls_route into 'struct mpls_route + struct mpls_nh'.
- struct mpls_nh represents a mpls nexthop label forwarding entry
- Adds
From: Robert Shearman
Change the selection of a multipath route to use a flow-based
hash. This more suitable for traffic sensitive to reordering within a
flow (e.g. TCP, L2VPN) and whilst still allowing a good distribution
of traffic given enough flows.
Selection of the
CHECKSUM_PARTIAL should only be used on plain vanilla IPv6 + UDP packets
in ip6_append_data. Some drivers don't correctly handle extension headers,
especially not ipv6 fragmentation which could result in broken checksums.
1) This patch improves the test for fragmentation and extension headers
in
On Thu 22-10-15 00:21:34, Johannes Weiner wrote:
> Letting shrink_slab() handle the root_mem_cgroup, and implicitely the
> !CONFIG_MEMCG case, allows shrink_zone() to invoke the shrinkers
> unconditionally from within the memcg iteration loop.
>
> Signed-off-by: Johannes Weiner
>Yet another POSIX deficiency.
>
>When a server deals with 10,000,000+ socks, we absolutely do not care of
>this requirement.
>
>O(log(n)) is still crazy if it involves O(log(n)) cache misses.
You miss the fire point of the algorithm; you *always* find an available
file descriptor in O(log(N))
On 10/23/2015 04:59 AM, David Miller wrote:
From: Tom Lendacky
Date: Wed, 21 Oct 2015 15:37:05 -0500
The code currently uses the lightweight dma_wmb barrier before updating
the current descriptor count. Under heavy load, the Tx cleanup routine
was seeing the updated
On Fri, Oct 23, 2015 at 02:10:29PM +0200, LABBE Corentin wrote:
> vnet_hdr_len cannot be negative and is use in operation/function that
> wait for unsigned value.
> This patch set vnet_hdr_len as size_t.
>
> Signed-off-by: LABBE Corentin
These things need to be done
On 23/10/2015 14:02, Eric Dumazet wrote:
Other data is used to make sure that it only takes O(log(n)) to find the
lowest available file descriptor entry. (Where n, I think, is the returned
descriptor)
Yet another POSIX deficiency.
When a server deals with 10,000,000+ socks, we absolutely do
On Fri, Oct 23, 2015 at 01:48:49PM +0900, YOSHIFUJI Hideaki wrote:
> Hi,
>
> John W. Linville wrote:
> > NOTE: Link-local IPv6 addresses for remote endpoints are not supported,
> > since the driver currently has no capacity for binding a geneve
> > interface to a specific link.
> >
> >
Should not allocate vnic queues number more than online cpus.
Signed-off-by: Joe Jin
Cc: Jan Beulich
Cc: Wei Liu
Cc: Ian Campbell
Cc: Boris Ostrovsky
Cc: Konrad Rzeszutek Wilk
On 10/23/2015 12:17 AM, Sergei Shtylyov wrote:
[...]
If memory is not yet available
when the controller is trying to use the invalid descriptor, the
controller will see it and DMA will stop.
That means leaving RACT=0 and that's what the driver is even doing...
Hm, then I don't
On Thu 22-10-15 00:21:30, Johannes Weiner wrote:
> A later patch will need this symbol in files other than memcontrol.c,
> so export it now and replace mem_cgroup_root_css at the same time.
>
> Signed-off-by: Johannes Weiner
Acked-by: Michal Hocko
> ---
>
The variable copied in llc_ui_sendmsg() cannot be negative and is used
in functions that wait for unsigned value, so set it as size_t (like it
is in llc_ui_recvmsg())
Signed-off-by: LABBE Corentin
---
net/llc/af_llc.c | 3 ++-
1 file changed, 2 insertions(+), 1
1 - 100 of 248 matches
Mail list logo