- 03 Dec, 2015 1 commit
-
-
KY Srinivasan authored
As part of reducing the size of the hv_netvsc_packet, resize some of the variables based on their usage. Signed-off-by:
K. Y. Srinivasan <kys@microsoft.com> Reviewed-by:
Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
- 02 Dec, 2015 18 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== Intel Wired LAN Driver Updates 2015-12-01 This series contains updates to i40e and i40evf only. Helin adds new fields to i40e_vsi to store user configured RSS config data and the code to use it. Also renamed RSS items to clarify functionality and scope to users. Fixed a confusing kernel message of enabling RSS size by reporting it together with the hardware maximum RSS size. Anjali fixes the issue of forcing writeback too often causing us to not benefit from NAPI. Jesse adds a prefetch for data early in the transmit path to help immensely for pktgen and forwarding workloads. Fixed the i40e driver that was possibly sleeping inside critical section of code. Carolyn fixes an issue where adminq init failures always provided a message that NVM was newer than expected, when this is not always the case for init_adminq failures. Fixed by adding a check for that specific error condition and a different helpful message otherwise. Mitch fixes error message by telling the user which VF is being naughty, rather than making them guess. Updated the queue_vector array from a statically-sized member of the adapter structure, to a dynamically-allocated and -sized array. This reduces the size of the adapter structure and allows us to support any number of queue vectors in the future without changing the code. ==================== Signed-off-by: -
Jesse Brandeburg authored
With the final edition of the patches to remove sleeps from the driver's entry points, the grab_rtnl argument is no longer needed, so partially revert the commit that added it. Change-ID: Ib9778476242586cc9e58b670f5f48d415cb59003 Signed-off-by:
Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by:
Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Jesse Brandeburg authored
The driver was being called by VLAN, bonding, teaming operations that expected to be able to hold locks like rcu_read_lock(). This causes the driver to be held to the requirement to not sleep, and was found by the kernel debug options for checking sleep inside critical section, and the locking validator. Change-ID: Ibc68c835f5ffa8ffe0638ffe910a66fc5649a7f7 Signed-off-by:
-
Catherine Sullivan authored
Bump. Change-ID: I00ebbb2e5e5572f947502b8f6db4d94f666d6b14 Signed-off-by:
Catherine Sullivan <catherine.sullivan@intel.com> Tested-by:
-
Mitch Williams authored
Instead of awkwardly keeping a fixed array of pointers in the adapter struct and then allocating ring structs individually, just keep a single pointer and allocate a single blob for the arrays. This simplifies code, shrinks the adapter structure, and future-proofs the driver by not limiting the number of rings we can handle. Change-ID: I31334ff911a6474954232cfe4bc98ccca3c769ff Signed-off-by:
Mitch Williams <mitch.a.williams@intel.com> Tested-by:
-
Mitch Williams authored
Change the queue_vector array from a statically-sized member of the adapter structure to a dynamically-allocated and -sized array. This reduces the size of the adapter structure, and allows us to support any number of queue vectors in the future without changing the code. Change-ID: I08dc622cb2f2ad01e832e51c1ad9b86524730693 Signed-off-by:
-
Mitch Williams authored
If, upon a midnight dreary, the PF returns ERR_PARAM when the VF is requesting resources, that's fatal. Either the firmware or NVM is badly, badly misconfigured, or this VF has been disabled due to a previous VF driver sending a bunch of bogus messages. Either way, there is no recovery from this. Don't ponder weak and weary, just quit. Change-ID: I09d9f16cc4ee7fec3b57646a289d33838c1c5bf5 Signed-off-by:
-
Mitch Williams authored
If we get an invalid message from a VF, we should tell the user which VF is being naughty, rather than making them guess. Change-ID: I9252cef7baea3d8584043ed6ff12619a94e2f99c Signed-off-by:
-
Helin Zhang authored
This patch fixes the confusing kernel message of enabled RSS size, by reporting it together with the hardware maximum RSS size. Change-ID: I64864dbfbc13beccc180a7871680def1f3d5a339 Signed-off-by:
Helin Zhang <helin.zhang@intel.com> Tested-by:
-
Carolyn Wyborny authored
This patch fixes an issue where adminq init failures always provided a message that NVM was newer than expected. This is not always the case for init_adminq failures. Without this patch, if adminq init fails for any reason, newer NVM message would be given. This problem is fixed by adding a check for that specific error condition and a different hopefully helpful message otherwise. Change-ID: Iaeaebee4e398989eae40bb70f943ab66a3a521a5 Signed-off-by:
Carolyn Wyborny <carolyn.wyborny@intel.com> Tested-by:
-
Helin Zhang authored
This patch adds new fields to i40e_vsi to store user configured RSS config data and code to use it. Change-ID: Ic5d3db8d9df52182b560248f8cdca9c5c7546879 Signed-off-by:
-
Helin Zhang authored
There are two ways to get RSS, this patch implements two functions with the same input parameters, and creates a more generic function for getting RSS configuration. Change-ID: I12d3b712c21455d47dd0a5aae58fc9b7c680db59 Signed-off-by:
-
Helin Zhang authored
There are two ways to configure RSS, this patch adjusts those two functions with the same input parameters, and creates a more generic function for configuring RSS. Change-ID: Iace73bdeba4831909979bef221011060ab327f71 Signed-off-by:
-
Helin Zhang authored
This patch renames old VF adapter specific RSS function to clarify its scope. Change-ID: Ie5253083a44c677ebb7709a8a3a18402ad2dc6a6 Signed-off-by:
-
Jesse Brandeburg authored
Issue a prefetch for data early in the transmit path. This should not be generally needed for Tx traffic, but it helps immensely for pktgen workloads and should help for forwarding workloads as well. Change-ID: Iefee870c20599e0c4240e1d8637e4f16b625f83a Signed-off-by:
-
Anjali Singhai Jain authored
This patch fixes the issue of forcing WB too often causing us to not benefit from NAPI. Without this patch we were forcing WB/arming interrupt too often taking away the benefits of NAPI and causing a performance impact. With this patch we disable force WB in the clean routine for X710 and XL710 adapters. X722 adapters do not enable interrupt to force a WB and benefit from WB_ON_ITR and hence force WB is left enabled for those adapters. For XL710 and X710 adapters if we have less than 4 packets pending a software Interrupt triggered from service task will force a WB. This patch also changes the conditions for setting RS bit as described in code comments. This optimizes when the HW does a tail bump amd when it does a WB. It also optimizes when we do a wmb. Change-ID: Id831e1ae7d3e2ec3f52cd0917b41ce1d22d75d9d Signed-off-by:
Anjali Singhai Jain <anjali.singhai@intel.com> Tested-by:
-
Helin Zhang authored
This patch renames rss_size to alloc_rss_size in i40e_pf, which is clearer and avoids confusion. It also adds comments to the other related structure members to help clarify usage. Change-ID: Ia90090609d006ab589cb639975bb8a0af795d16f Signed-off-by:
-
Helin Zhang authored
This patch adds new fields to i40e_vsi to store user configured RSS config data and code to use it. Signed-off-by:
-
- 01 Dec, 2015 15 commits
-
-
Arnd Bergmann authored
The 'sc' ISDN driver relies on using readl() to access ISA I/O memory. This has been deprecated and produced warnings since linux-2.3.23, disabled by default since 2.4.10 and finally removed in 2.6.5. I found this because the compiling the driver for ARM produces a warning: In file included from ../drivers/isdn/sc/includes.h:8:0, from ../drivers/isdn/sc/init.c:13: ../arch/arm/include/asm/io.h:115:21: note: expected 'const volatile void *' but argument is of type 'long unsigned int' It is pretty clear that this driver has not been used for a long time and there is no point fixing it now, so let's remove it. Signed-off-by:Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Giuseppe CAVALLARO authored
For GMAC newer than 3.40a there is a new register (Reg_9) that provides the status of all modules of the transmit and receive paths and FIFO status. These can be exposed via ethtool. Signed-off-by:
Giuseppe Cavallaro <peppe.cavallaro@st.com> Signed-off-by:
-
David S. Miller authored
Yuval Mintz says: ==================== qede/qed: Implement various ethtool operations This series adds several new ethtool operations to qede: - {get, set}_channels - {get, set}_ringparam - set_phys_id - nway_reset - {get, set}_pauseparam As well as extending the qed APIs to support these commands. ==================== Signed-off-by: -
Sudarsana Kalluru authored
Signed-off-by:
Sudarsana Kalluru <Sudarsana.Kalluru@qlogic.com> Signed-off-by:
Yuval Mintz <Yuval.Mintz@qlogic.com> Signed-off-by:
-
Sudarsana Kalluru authored
Signed-off-by:
-
Sudarsana Kalluru authored
Signed-off-by:
-
Sudarsana Kalluru authored
Physical LEDs are being controlled by the management FW. This adds the qed functionality required to request management FW to change the LED configuration, as well as the necessary APIs for this functionality to later be used by the protocol drivers. Signed-off-by:
-
Sudarsana Kalluru authored
Signed-off-by:
-
Sudarsana Kalluru authored
Signed-off-by:
-
David S. Miller authored
Bert Kenward says: ==================== Basic support for Solarflare 8000 series NICs The upcoming Solarflare 8000 series 10G/40G network card supports a similar interface to the current 7000 series cards. This patch series provides basic support for these cards, making no use of any new functionality. v2: fix indenting in ef10.c in patch 1/2. ==================== Signed-off-by: -
Bert Kenward authored
Also add support for 7000 series 40G NIC VF. Signed-off-by:
Bert Kenward <bkenward@solarflare.com> Signed-off-by:
-
Bert Kenward authored
The Solarflare 8000 series NIC will use a new TSO scheme. The current driver refuses to load if the current TSO scheme is not found. Remove that check and instead make the TSO version a per-queue parameter. Signed-off-by:
-
Nikolay Aleksandrov authored
This module allows to insert errors in some of netdevice's notifier events. All network drivers use these notifiers to signal various events and to check if they are allowed, e.g. PRECHANGEMTU and CHANGEMTU afterwards. Until recently I had to run failure tests by injecting a custom module, but now this infrastructure makes it trivial to test these failure paths. Some of the recent bugs I fixed were found using this module. Here's an example: $ cd /sys/kernel/debug/notifier-error-inject/netdev $ echo -22 > actions/NETDEV_CHANGEMTU/error $ ip link set eth0 mtu 1024 RTNETLINK answers: Invalid argument CC: Akinobu Mita <akinobu.mita@gmail.com> CC: "David S. Miller" <davem@davemloft.net> CC: netdev <netdev@vger.kernel.org> Signed-off-by:
Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Signed-off-by:
-
Vitaly Kuznetsov authored
There are several issues in hv_netvsc driver with regards to link status change handling: - RNDIS_STATUS_NETWORK_CHANGE results in calling userspace helper doing '/etc/init.d/network restart' and this is inappropriate and broken for many reasons. - link_watch infrastructure only sends one notification per second and in case of e.g. paired disconnect/connect events we get only one notification with last status. This makes it impossible to handle such situations in userspace. Redo link status changes handling in the following way: - Create a list of reconfig events in network device context. - On a reconfig event add it to the list of events and schedule netvsc_link_change(). - In netvsc_link_change() ensure 2-second delay between link status changes. - Handle RNDIS_STATUS_NETWORK_CHANGE as a paired disconnect/connect event. Signed-off-by:
Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by:
-
Rainer Weikusat authored
The current unix_dgram_recvmsg does a wake up for every received datagram. This seems wasteful as only SOCK_DGRAM client sockets in an n:1 association with a server socket will ever wait because of the associated condition. The patch below changes the function such that the wake up only happens if wq_has_sleeper indicates that someone actually wants to be notified. Testing with SOCK_SEQPACKET and SOCK_DGRAM socket seems to confirm that this is an improvment. Signed-Off-By:
Rainer Weikusat <rweikusat@mobileactivedefense.com> Signed-off-by:
-
- 30 Nov, 2015 6 commits
-
-
David S. Miller authored
Nikolay Aleksandrov says: ==================== net: ipmr: more cleanups and mfc netlink support This set continues with the minor cleanups in the first 6 patches and patch 7 adds the first new feature - MFC manipulation via netlink. It registers NEWROUTE/DELROUTE for that purpose and uses the same semantics as the already present netlink dump. The only new attribute that is used is RTA_PREFSRC to denote an MFC_PROXY entry. Currently the table must exist before adding an entry, and new tables can be created only via setsockopt, but that will be changed in the future. This set was tested with modified iproute2 which supports NEWROUTE/DELROUTE for RTNL_FAMILY_IPMR. ==================== Signed-off-by: -
Nikolay Aleksandrov authored
This patch adds support to add and remove MFC entries. It uses the same attributes like the already present dump support in order to be consistent. There's one new entry - RTA_PREFSRC, it's used to denote an MFC_PROXY entry (see MRT_ADD_MFC vs MRT_ADD_MFC_PROXY). The already existing infrastructure is used to create and delete the entries, the netlink message gets converted internally to a struct mfcctl which is used with ipmr_mfc_add/delete. The other used attributes are: RTA_IIF - used for mfcc_parent (when adding it's required to be valid) RTA_SRC - used for mfcc_origin RTA_DST - used for mfcc_mcastgrp RTA_TABLE - the MRT table id RTA_MULTIPATH - the "oifs" ttl array Signed-off-by:
-
Nikolay Aleksandrov authored
We can have both errors and we'll return the second one, fix it to return an error at a time as it's normal. I've overlooked this in my previous set. Signed-off-by:
-
Nikolay Aleksandrov authored
Move the inline pimsm_enabled() to pim.h and rename it to ipmr_pimsm_enabled to show it's for the ipv4 ipmr code since pim.h is used by IPv6 too. Signed-off-by:
-
Nikolay Aleksandrov authored
Move the definitions of VIF_EXISTS() and struct mr_table to mroute.h Signed-off-by:
-
Nikolay Aleksandrov authored
Remove extra spaces and tabs, adjust function definitions, remove an unnecessary ifdef (already used below, just move code) and drop extern from the functions. Signed-off-by:
-
