aboutsummaryrefslogtreecommitdiff
path: root/drivers/net/ethernet
AgeCommit message (Collapse)AuthorFilesLines
2019-03-22net/mlx5: Fix compilation warning in eq.cSaeed Mahameed1-4/+2
mlx5_eq_table_get_rmap is being used only when CONFIG_RFS_ACCEL is enabled, this patch fixes the below warning when CONFIG_RFS_ACCEL is disabled. drivers/.../mlx5/core/eq.c:903:18: [-Werror=missing-prototypes] error: no previous prototype for ‘mlx5_eq_table_get_rmap’ Fixes: f2f3df550139 ("net/mlx5: EQ, Privatize eq_table and friends") Signed-off-by: Saeed Mahameed <[email protected]>
2019-03-22net/mlx5: Simplify mlx5_sriov_is_enabled() by using pci core APIParav Pandit2-8/+5
It is desired to get rid of num_vfs stored inside mlx5_core_sriov to safely support vports more than vfs. To reduce dependency on mlx5_core_sriov num_vfs, start using pci_num_vf() from pci core. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Bodong Wang <[email protected]> Signed-off-by: Saeed Mahameed <[email protected]>
2019-03-22net/mlx5: Rename total_vfs to total_vportsParav Pandit1-2/+2
Macro MLX5_TOTAL_VPORTS() returns total number of vports. Therefore, rename variable total_vfs to total_vports to improve code readability. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Bodong Wang <[email protected]> Signed-off-by: Saeed Mahameed <[email protected]>
2019-03-22net/mlx5: Simplify sriov enable/disable flowParav Pandit1-36/+8
Simplify sriov enable/disable flow for below two checks. 1. PCI core driver allows sriov configuration only on a PF. This is done in drivers/pci/pci-sysfs.c sriov_attrs_are_visible(). 2. PCI core driver allow sriov enablement if the sriov is currently disabled for for a PF. This is done in drivers/pci/pci-sysfs.c sriov_numvfs_store(). Hence there is no need for mlx5 driver to duplicate such checks. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Bodong Wang <[email protected]> Signed-off-by: Saeed Mahameed <[email protected]>
2019-03-22ice: Get VF VSI instances directly via PFAkeem G Abodunrin1-8/+27
This patch changes how we get VF VSIs instances. Instead of relying on mailbox virtual channel message to retrieve VSI, it is more reliable getting it directly via VF object in PF data structure. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Don't let VF know that it is untrustedAkeem G Abodunrin1-4/+8
Don't let the VF know it's not trusted when it tries to add more than permitted additional MAC addresses. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Set LAN_EN for all directional rulesYashaswini Raghuram Prathivadi Bhayankaram1-3/+9
The LAN_EN bit for a switch rule determines if the packet can go out on the wire or not. Set the LAN_EN flag in the switch action for all directional rules. Signed-off-by: Yashaswini Raghuram Prathivadi Bhayankaram <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Do not set LB_EN for prune switch rulesChristopher N Bednarz1-1/+6
LB_EN for prune switch rules was causing all TX traffic to loopback to the internal switch and dropped. When running bi-directional stress workloads with RDMA the RDPU would hang blocking tx and rx traffic. Signed-off-by: Christopher N Bednarz <[email protected]> Reviewed-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Enable LAN_EN for the right recipesYashaswini Raghuram Prathivadi Bhayankaram1-9/+20
In VEB mode, enable LAN_EN bit in the action fields for filter rules corresponding to the right recipes. Signed-off-by: Yashaswini Raghuram Prathivadi Bhayankaram <[email protected]> Reviewed-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Add support for PF/VF promiscuous modeAkeem G Abodunrin7-31/+527
Implement support for VF promiscuous mode, MAC/VLAN/MAC_VLAN and PF multicast MAC/VLAN/MAC_VLAN promiscuous mode. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: code cleanup in ice_sched.cVictor Raj1-8/+9
This patch does some clean up in the Tx scheduler code: 1. Adjust the stack variable usage 2. Modify the debug prints to display the FW error 3. Add additional debug prints while adding/removing VSIs Signed-off-by: Victor Raj <[email protected]> Reviewed-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Remove unused vsi_id fieldAnirudh Venkataramanan1-1/+0
Remove unused vsi_id field from struct ice_sched_vsi_info. Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: fix some function prototype and signature style issuesBruce Allan11-87/+108
Put the return type on a separate line for function prototypes and signatures that would exceed the 80-character limit if both were on the same line. Signed-off-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: fix the divide by zero issueKiran Patil1-1/+12
Static analysis flagged a potential divide by zero error because vsi->num_rxq can become zero in certain condition and it is used as divisor. Signed-off-by: Kiran Patil <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Fix issue reconfiguring VF queuesAkeem G Abodunrin3-16/+76
When VF requested for queues changes, we need to update LAN Tx queue with correct number of VF queue pairs and re-allocate VF resources based on this new requested number of queues, which is constraint within maximum queue supported per VF. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Remove unused function prototypeAnirudh Venkataramanan1-10/+0
Commit 7c710869d64e ("ice: Add handlers for VF netdevice operations") seems to have inadvertently introduced a function prototype for ice_set_vf_bw that isn't implemented. Remove it. Fixes: 7c710869d64e ("ice: Add handlers for VF netdevice operations") Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: fix static analysis warningsBruce Allan4-4/+4
cppcheck warns "Identical condition '<var>', second condition is always false". Fix them. Signed-off-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Fix issue reclaiming resources back to the pool after resetAkeem G Abodunrin1-3/+23
This patch fixes issue reclaiming VF resources back to the pool after reset - Since we only allocate HW vector for all VFs and track together with resources allocation for PF with ice_search_res, we need to free VFs resources separately, using first VF vector index to traverse the list. Otherwise tracker starts from the last assigned vectors list and causes maximum supported number of HW vectors, 1024 to be exhausted, depending on the number of VFs enabled, which causes a lot of unwanted issues, and failed to reassign vectors for VFs. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22ice: Enable MAC anti-spoof by defaultAkeem G Abodunrin1-0/+9
This patch enables MAC anti-spoof by default, with creation of VF VSIs or when the VF VSIs are being re-initialized. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-22r8169: use netif_start_queue instead of netif_wake_qeueue in rtl8169_start_xmitHeiner Kallweit1-1/+1
Replace the call to netif_wake_queue in rtl8169_start_xmit with netif_start_queue as we don't need to actually wake up the queue since we are still in mid transmit so we just need to reset the bit so it doesn't prevent the next transmit. (Description shamelessly copied from a mail sent by Alex.) Suggested-by: Alexander Duyck <[email protected]> Signed-off-by: Heiner Kallweit <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-22r8169: don't read interrupt mask register in interrupt handlerHeiner Kallweit1-2/+4
After the original patch network starts to crash on heavy load. It's not fully clear why this additional register read has such side effects, but removing it fixes the issue. Thanks also to Alex for his contribution and hints. [0] https://marc.info/?t=155268170400002&r=1&w=2 Fixes: e782410ed237 ("r8169: improve spurious interrupt detection") Reported-by: VDR User <[email protected]> Tested-by: VDR User <[email protected]> Signed-off-by: Heiner Kallweit <[email protected]> Reviewed-by: Alexander Duyck <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-21nfp: remove defines for unused control bitsJakub Kicinski2-5/+1
NFP driver ABI contains bits for L2 switching which were never implemented in initially envisioned form. Remove the defines, and open up the possibility of reclaiming the bits for other uses. Signed-off-by: Jakub Kicinski <[email protected]> Reviewed-by: Dirk van der Merwe <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-21net: hns3: fix for not calculating tx bd num correctlyYunsheng Lin2-8/+6
When there is only one byte in a frag, the current calculation using "(size + HNS3_MAX_BD_SIZE - 1) >> HNS3_MAX_BD_SIZE_OFFSET" will return zero, because HNS3_MAX_BD_SIZE is 65535 and HNS3_MAX_BD_SIZE_OFFSET is 16. So it will cause tx error when a frag's size is one byte. This patch fixes it by using DIV_ROUND_UP. Fixes: 3fe13ed95dd3 ("net: hns3: avoid mult + div op in critical data path") Signed-off-by: Yunsheng Lin <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-21net: systemport: Remove print of base addressFlorian Fainelli1-3/+3
Since commit ad67b74d2469 ("printk: hash addresses printed with %p") pointers are being hashed when printed. Displaying the virtual memory at bootup time is not helpful, especially given we use a dev_info() which already displays the platform device's address. Signed-off-by: Florian Fainelli <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-21net: macb: Add null check for PCLK and HCLKHarini Katakam1-2/+8
Both PCLK and HCLK are "required" clocks according to macb devicetree documentation. There is a chance that devm_clk_get doesn't return a negative error but just a NULL clock structure instead. In such a case the driver proceeds as usual and uses pclk value 0 to calculate MDC divisor which is incorrect. Hence fix the same in clock initialization. Signed-off-by: Harini Katakam <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20dpaa2-eth: Fix possible access beyond end of arrayIoana Ciocoi Radulescu1-1/+1
Make sure we don't try to enqueue XDP_REDIRECT frames to an inexistent FQ. While it is guaranteed not to have more than one queue per core, having fewer queues than CPUs on an interface is a valid configuration. Fixes: d678be1dc1ec ("dpaa2-eth: add XDP_REDIRECT support") Reported-by: Jesper Dangaard Brouer <[email protected]> Signed-off-by: Ioana Radulescu <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: ks8851: Deduplicate register macrosLukas Wunner3-330/+34
The ks8851 chip is sold either with an SPI interface (KSZ8851SNL) or with a so-called non-PCI interface (KSZ8851-16MLL). When the driver for the latter was introduced with commit a55c0a0ed415 ("drivers/net: ks8851_mll ethernet network driver"), it duplicated the register macros introduced by the driver for the former with commit 3ba81f3ece3c ("net: Micrel KS8851 SPI network driver"). The chips are almost identical, so the duplication seems unwarranted. There are a handful of bits which are in use on the KSZ8851-16MLL but reserved on the KSZ8851SNL, and vice-versa, but there are no actual collisions. Thus, remove the duplicate definitions from the KSZ8851-16MLL driver. Mark all bits which differ between the two chips. Move the SPI frame opcodes, which are specific to KSZ8851SNL, to its driver. The KSZ8851-16MLL driver added a RXFCTR_THRESHOLD_MASK macro which is a duplication of the RXFCTR_RXFCT_MASK macro, rename it where it's used. Same for P1MBCR_FORCE_FDX, which duplicates the BMCR_FULLDPLX macro and OBCR_ODS_16MA, which duplicates OBCR_ODS_16mA. Signed-off-by: Lukas Wunner <[email protected]> Cc: Frank Pavlic <[email protected]> Cc: Ben Dooks <[email protected]> Cc: Tristram Ha <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: ks8851: Fix register macro misnomersLukas Wunner1-25/+27
In the header file accompanying the ks8851 driver, the P1SCLMD register macros are misnamed, they actually pertain to the P1CR register. The P1CR macros in turn pertain to the P1SR register, see pages 65 to 68 of the spec: http://www.hqchip.com/uploads/pdf/201703/47c98946d6c97a4766e14db3f24955f2.pdf The misnomers have no negative consequences so far because the macros aren't used by ks8851.c, but that's about to change. Signed-off-by: Lukas Wunner <[email protected]> Cc: Frank Pavlic <[email protected]> Cc: Ben Dooks <[email protected]> Cc: Tristram Ha <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: ks8851: Set initial carrier state to downLukas Wunner1-0/+2
The ks8851 chip's initial carrier state is down. A Link Change Interrupt is signaled once interrupts are enabled if the carrier is up. The ks8851 driver has it backwards by assuming that the initial carrier state is up. The state is therefore misrepresented if the interface is opened with no cable attached. Fix it. The Link Change interrupt is sometimes not signaled unless the P1MBSR register (which contains the Link Status bit) is read on ->ndo_open(). This might be a hardware erratum. Read the register by calling mii_check_link(), which has the desirable side effect of setting the carrier state to down if the cable was detached while the interface was closed. Signed-off-by: Lukas Wunner <[email protected]> Cc: Frank Pavlic <[email protected]> Cc: Ben Dooks <[email protected]> Cc: Tristram Ha <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: ks8851: Delay requesting IRQ until openedLukas Wunner1-13/+11
The ks8851 driver currently requests the IRQ before registering the net_device. Because the net_device name is used as IRQ name and is still "eth%d" when the IRQ is requested, it's impossibe to tell IRQs apart if multiple ks8851 chips are present. Most other drivers delay requesting the IRQ until the net_device is opened. Do the same. The driver doesn't enable interrupts on the chip before opening the net_device and disables them when closing it, so there doesn't seem to be a need to request the IRQ already on probe. Signed-off-by: Lukas Wunner <[email protected]> Cc: Frank Pavlic <[email protected]> Cc: Ben Dooks <[email protected]> Cc: Tristram Ha <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: ks8851: Reassert reset pin if chip ID check failsLukas Wunner1-1/+1
Commit 73fdeb82e963 ("net: ks8851: Add optional vdd_io regulator and reset gpio") amended the ks8851 driver to briefly assert the chip's reset pin on probe. It also amended the probe routine's error path to reassert the reset pin if a subsequent initialization step fails. However the commit misplaced reassertion of the reset pin in the error path such that it is not performed if the check of the Chip ID and Enable Register (CIDER) fails. The error path is therefore slightly asymmetrical to the probe routine's body. Fix it. Signed-off-by: Lukas Wunner <[email protected]> Cc: Frank Pavlic <[email protected]> Cc: Stephen Boyd <[email protected]> Cc: Nishanth Menon <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: ks8851: Dequeue RX packets explicitlyLukas Wunner1-4/+4
The ks8851 driver lets the chip auto-dequeue received packets once they have been read in full. It achieves that by setting the ADRFE flag in the RXQCR register ("Auto-Dequeue RXQ Frame Enable"). However if allocation of a packet's socket buffer or retrieval of the packet over the SPI bus fails, the packet will not have been read in full and is not auto-dequeued. Such partial retrieval of a packet confuses the chip's RX queue management: On the next RX interrupt, the first packet read from the queue will be the one left there previously and this one can be retrieved without issues. But for any newly received packets, the frame header status and byte count registers (RXFHSR and RXFHBCR) contain bogus values, preventing their retrieval. The chip allows explicitly dequeueing a packet from the RX queue by setting the RRXEF flag in the RXQCR register ("Release RX Error Frame"). This could be used to dequeue the packet in case of an error, but if that error is a failed SPI transfer, it is unknown if the packet was transferred in full and was auto-dequeued or if it was only transferred in part and requires an explicit dequeue. The safest approach is thus to always dequeue packets explicitly and forgo auto-dequeueing. Without this change, I've witnessed packet retrieval break completely when an SPI DMA transfer fails, requiring a chip reset. Explicit dequeueing magically fixes this and makes packet retrieval absolutely robust for me. The chip's documentation suggests auto-dequeuing and uses the RRXEF flag only to dequeue error frames which the driver doesn't want to retrieve. But that seems to be a fair-weather approach. Signed-off-by: Lukas Wunner <[email protected]> Cc: Frank Pavlic <[email protected]> Cc: Ben Dooks <[email protected]> Cc: Tristram Ha <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: hns3: Make hclge_destroy_cmd_queue staticYueHaibing1-1/+1
Fix sparse warning: drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c:414:6: warning: symbol 'hclge_destroy_cmd_queue' was not declared. Should it be static? Signed-off-by: YueHaibing <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20net: remove 'fallback' argument from dev->ndo_select_queue()Paolo Abeni16-43/+27
After the previous patch, all the callers of ndo_select_queue() provide as a 'fallback' argument netdev_pick_tx. The only exceptions are nested calls to ndo_select_queue(), which pass down the 'fallback' available in the current scope - still netdev_pick_tx. We can drop such argument and replace fallback() invocation with netdev_pick_tx(). This avoids an indirect call per xmit packet in some scenarios (TCP syn, UDP unconnected, XDP generic, pktgen) with device drivers implementing such ndo. It also clean the code a bit. Tested with ixgbe and CONFIG_FCOE=m With pktgen using queue xmit: threads vanilla patched (kpps) (kpps) 1 2334 2428 2 4166 4278 4 7895 8100 v1 -> v2: - rebased after helper's name change Signed-off-by: Paolo Abeni <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20qed: Define new MF bit for no_vlan configSudarsana Reddy Kalluru3-5/+8
The patch introduces a new Multi-Function bit for cases where firmware shouldn't perform the insertion of vlan-0 tag. The new bit is defined to abstract the implementation from the actual MF mode. Signed-off-by: Sudarsana Reddy Kalluru <[email protected]> Signed-off-by: Ariel Elior <[email protected]> Signed-off-by: Michal Kalderon <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20qede: Populate mbi version in ethtool driver query data.Sudarsana Reddy Kalluru1-5/+19
The patch adds support to display MBI image version in 'ethtool -i' output. Signed-off-by: Sudarsana Reddy Kalluru <[email protected]> Signed-off-by: Ariel Elior <[email protected]> Signed-off-by: Michal Kalderon <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2019-03-20Merge branch '100GbE' of ↵David S. Miller11-116/+197
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue Jeff Kirsher says: ==================== 100GbE Intel Wired LAN Driver Updates 2019-03-19 This series contains updates to ice driver only. Michal adds support for the pruning enable flag to avoid seeing broadcast packets on different VLANs. Akeem fixes an issue with VF queues being disabled and the VF netdev network carrier being lost after reset. Fixed an issue issue when doing PFR and CORER resets, where all VF VSIs need to be reset and rebuilt with the main VSIs before replaying all VSIs. Resolved an issue to properly initialize VFs in the guest OS via PCI passthrough. Bruce adds a local variable to avoid unnecessary de-references throughout ice_probe(). Brett cleans up the code a bit by removing the need for a local variable and re-designs the loop to simply return when get a successful result. Cleans up the code to replace loop calls with a predefined macro to make the code more consistent. Updated the driver to ensure ITR granularity is always 2 usecs. Refactors the calculation of VSIs per PF into a general function that can calculate per PF allocations for not just VSIs but across multiple resource types. Improve the driver performance of the driver when using the default settings by determining the ring size and the number of descriptors for transmit and receive based on a calculation with the PAGE_SIZE, ICE_MAX_NUM_DESC, and ICE_REQ_DESC_MULTIPLE. Chinh fixes an issue, where a reserved bit was possibly being set when it should never be set. ==================== Signed-off-by: David S. Miller <[email protected]>
2019-03-20Merge branch '1GbE' of ↵David S. Miller12-18/+1392
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue Jeff Kirsher says: ==================== 1GbE Intel Wired LAN Driver Updates 2019-03-19 This series contains updates to e100, e1000, e1000e, igb, igc and ixgbe. Serhey Popovych fixes the return value for several of our older drivers for netdev_update_features() to notify of changes applied. Kai-Heng Feng fixes the WoL setting for system suspend, which should not set to runtime suspend settings for igb. Then fixes a power management issue with e1000e for CNP+ devices. Colin Ian King fixes whitespace issue (indentation), which helps with readability. Sasha provides the remaining changes for igc, including the enabling of multi-queues to receive. Added support for displaying and configuring network flow classification (NFC) via ethtool. Added additional statistics and basic counters for igc. Fixed a typo, so it aligns with our other drivers. ==================== Signed-off-by: David S. Miller <[email protected]>
2019-03-19ice: Determine descriptor count and ring size based on PAGE_SIZEBrett Creeley3-11/+43
Currently we set the default number of Tx and Rx descriptors to 128 by default. For Rx this amounts to a full page (assuming 4K pages) because each Rx descriptor is 32 Bytes, but for Tx it only amounts to a half page because each Tx descriptor is 16 Bytes (assuming 4K pages). Instead of assuming 4K pages, determine the ring size and the number of descriptors for Tx and Rx based on a calculation using the PAGE_SIZE, ICE_MAX_NUM_DESC, and ICE_REQ_DESC_MULTIPLE. This change is being made to improve the performance of the driver when using the default settings. Signed-off-by: Brett Creeley <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: Reset all VFs with VFLR during SR-IOV init flowAkeem G Abodunrin1-1/+1
During SR-IOV initialization, we allocate and setup VFs with reset, and since we were going to inform Firmware about our intention to do VFLR by disabling LAN TX Queue, then we really have to complete VF reset flow with VFLR using appropriate registers - Otherwise, reset status bit for VF in the Guest OS might returns DEADBEEF. This resolves issue to properly initialize VFs in the Guest OS via PCI passthrough. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: Get resources per functionBrett Creeley1-5/+8
ice_get_guar_num_vsi currently calculates the number of VSIs per PF. Rework this into a general function ice_get_num_per_func, that can calculate per PF allocations for not just VSIs but across multiple resource types. Signed-off-by: Brett Creeley <[email protected]> Reviewed-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: Implement flow to reset VFs with PFR and other resetsAkeem G Abodunrin1-6/+2
All VF VSIs need to be reset and rebuild with the main VSIs before replaying all VSIs, so that all existing switch filters, scheduler tree and other configuration could be replayed at once. This fixes issues when doing PFR and CORER reset. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: configure GLINT_ITR to always have an ITR gran of 2Brett Creeley4-18/+44
Instead of hoping that our ITR granularity will be 2 usec program the GLINT_CTL register to make sure the ITR granularity is always 2 usecs. Now that we know what the ITR granularity will be get rid of the check in ice_probe() to verify our previous assumption. Signed-off-by: Brett Creeley <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: use ice_for_each_vsi macro when possibleBrett Creeley3-9/+8
Replace all instances of: for (i = 0; i < pf->num_alloc_vsi; i++) with the following macro: ice_for_each_vsi(pf, i) This will allow the code to be consistent since there are currently cases of using both. Signed-off-by: Brett Creeley <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice : Ensure only valid bits are set in ice_aq_set_phy_cfgChinh T Cao3-2/+15
In the ice_aq_set_phy_cfg AQ command, the 16.4 bit is reserved. This patch will make sure that this bit will never be set to 1. Signed-off-by: Chinh T Cao <[email protected]> Reviewed-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: remove redundant variable and if conditionBrett Creeley1-7/+4
In ice_pf_rxq_wait we are using an unnecessary local variable and also we are checking if the timeout time was reached after the loop. Get rid of the local variable and return 0 right when we get a successful result. This makes it so we can return -ETIMEDOUT if we ever exit the loop because we know the timeout time has been hit. Signed-off-by: Brett Creeley <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: avoid multiple unnecessary de-references in probeBruce Allan1-18/+15
Add a local variable struct device *dev to avoid unnecessary de-references throughout ice_probe(). Signed-off-by: Bruce Allan <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: Fix issue with VF reset and multiple VFs support on PFsAkeem G Abodunrin1-7/+13
This patch fixes issues with VF queues being disabled, and VF netdev network carrier being lost after reset. Basically, we need to check if VF is enabled, and queue configured in reset_all_vfs flow, and disable/enable those queues appropriately whenever the function is called after Global/CORER/PFR reset/rebuild/replay. Signed-off-by: Akeem G Abodunrin <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19ice: Fix broadcast traffic in port VLAN modeMichal Swiatkowski1-32/+44
Set egress (Rx) pruning enable flag for VF VSI in VSI ctxt to enable prune action. To avoid seeing broadcast packet in different VLAN, pruning enable flag in VSI ctxt should be set. Write new functions (fill VSI ctx) to not repeat send ctxt code. Signed-off-by: Michal Swiatkowski <[email protected]> Signed-off-by: Anirudh Venkataramanan <[email protected]> Tested-by: Andrew Bowers <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2019-03-19igc: Remove unneeded hw_dbg printsSasha Neftin1-4/+0
Remove unneeded hw_dbg prints from igc_ethtool.c file. Clean up code. Signed-off-by: Sasha Neftin <[email protected]> Tested-by: Aaron Brown <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.