- 27 Jul, 2018 4 commits
-
-
Jia-Ju Bai authored
init_PStack() is never called in atomic context. It calls kmalloc() with GFP_ATOMIC, which is not necessary. GFP_ATOMIC can be replaced with GFP_KERNEL. This is found by a static analysis tool named DCNS written by myself. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jia-Ju Bai authored
nj_probe() is never called in atomic context. It calls kzalloc() with GFP_ATOMIC, which is not necessary. GFP_ATOMIC can be replaced with GFP_KERNEL. This is found by a static analysis tool named DCNS written by myself. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jia-Ju Bai authored
hfc_probe() is never called in atomic context. It calls kzalloc() with GFP_ATOMIC, which is not necessary. GFP_ATOMIC can be replaced with GFP_KERNEL. This is found by a static analysis tool named DCNS written by myself. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
YueHaibing authored
hns_dsaf_roce_reset is exported and used in hns_roce_hw_v1.c In commit 336a443b ("net: hns: Make many functions static") I make it static wrongly. drivers/infiniband/hw/hns/hns_roce_hw_v1.o: In function `hns_roce_v1_reset': hns_roce_hw_v1.c:(.text+0x37ac): undefined reference to `hns_dsaf_roce_reset' hns_roce_hw_v1.c:(.text+0x37cc): undefined reference to `hns_dsaf_roce_reset' Fixes: 336a443b ("net: hns: Make many functions static") Signed-off-by: YueHaibing <yuehaibing@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
- 26 Jul, 2018 30 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== 10GbE Intel Wired LAN Driver Updates 2018-07-26 This series contains updates to ixgbe and igb. Tony fixes ixgbe to add checks to ensure jumbo frames or LRO get enabled after an XDP program is loaded. Shannon Nelson adds the missing security configuration registers to the ixgbe register dump, which will help in debugging. Christian Grönke fixes an issue in igb that occurs on SGMII based SPF mdoules, by reverting changes from 2 previous patches. The issue was that initialization would fail on the fore mentioned modules because the driver would try to reset the PHY before obtaining the PHY address of the SGMII attached PHY. Venkatesh Srinivas replaces wmb() with dma_wmb() for doorbell writes, which avoids SFENCEs before the doorbell writes. Alex cleans up and refactors ixgbe Tx/Rx shutdown to reduce time needed to stop the device. The code refactor allows us to take the completion time into account when disabling queues, so that on some platforms with higher completion times, would not result in receive queues disabled messages. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jiri Pirko authored
Once user manually deletes the chain using "chain del", the chain cannot be marked as explicitly created anymore. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Fixes: 32a4f5ec ("net: sched: introduce chain object to uapi") Signed-off-by: David S. Miller <davem@davemloft.net>
-
Doron Roberts-Kedes authored
The zerocopy path ultimately calls iov_iter_get_pages, which defines the step function for ITER_KVECs as simply, return -EFAULT. Taking the non-zerocopy path for ITER_KVECs avoids the unnecessary fallback. See https://lore.kernel.org/lkml/20150401023311.GL29656@ZenIV.linux.org.uk/T/#u for a discussion of why zerocopy for vmalloc data is not a good idea. Discovered while testing NBD traffic encrypted with ktls. Fixes: c46234eb ("tls: RX path for ktls") Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Gustavo A. R. Silva authored
Code at line 1850 is unreachable. Fix this by removing the break statement above it, so the code for case RTM_GETCHAIN can be properly executed. Addresses-Coverity-ID: 1472050 ("Structurally dead code") Fixes: 32a4f5ec ("net: sched: introduce chain object to uapi") Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Guillaume Nault authored
The tunnel reception hook is only used by l2tp_ppp for skipping PPP framing bytes. This is a session specific operation, but once a PPP session sets ->recv_payload_hook on its tunnel, all frames received by the tunnel will enter pppol2tp_recv_payload_hook(), including those targeted at Ethernet sessions (an L2TPv3 tunnel can multiplex PPP and Ethernet sessions). So this mechanism is wrong, and uselessly complex. Let's just move this functionality to the pppol2tp rx handler and drop ->recv_payload_hook. Signed-off-by: Guillaume Nault <g.nault@alphalink.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
-
YueHaibing authored
when tipc_own_id failed to obtain node identity,dev_put should be call before return -EINVAL. Fixes: 682cd3cf ("tipc: confgiure and apply UDP bearer MTU on running links") Signed-off-by: YueHaibing <yuehaibing@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Vakul Garg authored
Removed checks against non-NULL before calling kfree_skb() and crypto_free_aead(). These functions are safe to be called with NULL as an argument. Signed-off-by: Vakul Garg <vakul.garg@nxp.com> Acked-by: Dave Watson <davejwatson@fb.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Vinicius Costa Gomes authored
This will allow to install a child qdisc under cbs. The main use case is to install ETF (Earliest TxTime First) qdisc under cbs, so there's another level of control for time-sensitive traffic. Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
YueHaibing authored
Fixes the following sparse warning: drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:73:20: warning: symbol 'hns_ae_get_handle' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:332:6: warning: symbol 'hns_ae_stop' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:360:6: warning: symbol 'hns_ae_toggle_ring_irq' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:580:6: warning: symbol 'hns_ae_update_stats' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:663:6: warning: symbol 'hns_ae_get_stats' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:695:6: warning: symbol 'hns_ae_get_strings' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:728:5: warning: symbol 'hns_ae_get_sset_count' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:774:6: warning: symbol 'hns_ae_update_led_status' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:786:5: warning: symbol 'hns_ae_cpld_set_led_id' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:798:6: warning: symbol 'hns_ae_get_regs' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c:823:5: warning: symbol 'hns_ae_get_regs_len' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c:342:6: warning: symbol 'hns_gmac_update_stats' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c:934:12: warning: symbol 'hns_mac_get_vaddr' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c:953:5: warning: symbol 'hns_mac_get_cfg' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c:343:6: warning: symbol 'hns_dsaf_srst_chns' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c:366:1: warning: symbol 'hns_dsaf_srst_chns_acpi' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c:373:6: warning: symbol 'hns_dsaf_roce_srst' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c:387:6: warning: symbol 'hns_dsaf_roce_srst_acpi' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c:571:5: warning: symbol 'hns_mac_get_sfp_prsnt' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c:589:5: warning: symbol 'hns_mac_get_sfp_prsnt_acpi' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c:31:12: warning: symbol 'g_dsaf_mode_match' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c:45:5: warning: symbol 'hns_dsaf_get_cfg' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c:962:6: warning: symbol 'hns_dsaf_tcam_addr_get' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c:2087:6: warning: symbol 'hns_dsaf_port_work_rate_cfg' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c:2837:5: warning: symbol 'hns_dsaf_roce_reset' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c:76:5: warning: symbol 'hns_ppe_common_get_cfg' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c:107:6: warning: symbol 'hns_ppe_common_free_cfg' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c:340:6: warning: symbol 'hns_ppe_uninit_ex' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c:708:5: warning: symbol 'hns_rcb_get_ring_num' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c:744:14: warning: symbol 'hns_rcb_common_get_vaddr' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c:314:6: warning: symbol 'hns_xgmac_update_stats' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_enet.c:1303:6: warning: symbol 'hns_nic_update_stats' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_enet.c:1585:6: warning: symbol 'hns_nic_poll_controller' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_enet.c:1938:6: warning: symbol 'hns_set_multicast_list' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_enet.c:1960:6: warning: symbol 'hns_nic_set_rx_mode' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:661:6: warning: symbol 'hns_get_ringparam' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:811:6: warning: symbol 'hns_get_channels' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:828:6: warning: symbol 'hns_get_ethtool_stats' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:886:6: warning: symbol 'hns_get_strings' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:976:5: warning: symbol 'hns_get_sset_count' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:1010:5: warning: symbol 'hns_phy_led_set' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:1032:5: warning: symbol 'hns_set_phys_id' was not declared. Should it be static? drivers/net/ethernet/hisilicon/hns/hns_ethtool.c:1106:6: warning: symbol 'hns_get_regs' was not declared. Should it be static? Signed-off-by: YueHaibing <yuehaibing@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Anders Roxell authored
Add the tls binary to .gitignore Fixes: 7f657d5b ("selftests: tls: add selftests for TLS sockets") Signed-off-by: Anders Roxell <anders.roxell@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jiri Pirko authored
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Alexander Duyck authored
This change is meant to allow us to take completion time into account when disabling queues. Previously we were just working with hard coded values for how long we should wait. This worked fine for the standard case where completion timeout was operating in the 50us to 50ms range, however on platforms that have higher completion timeout times this was resulting in Rx queues disable messages being displayed as we weren't waiting long enough for outstanding Rx DMA completions. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Don Buchholz <donald.buchholz@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Alexander Duyck authored
This change is meant to help reduce the time needed to shutdown the transmit and receive paths for the device. Specifically what we now do after this patch is disable the transmit path first at the netdev level, and then work on disabling the Rx. This way while we are waiting on the Rx queues to be disabled the Tx queues have an opportunity to drain out. In addition I have dropped the 10ms timeout that was left in the ixgbe_down function that seems to have been carried through from back in e1000 as far as I can tell. We shouldn't need it since we don't actually disable the Tx until much later and we have additional logic in place for verifying the Tx queues have been disabled. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Don Buchholz <donald.buchholz@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Venkatesh Srinivas authored
igb writes to doorbells to post transmit and receive descriptors; after writing descriptors to memory but before writing to doorbells, use dma_wmb() rather than wmb(). wmb() is more heavyweight than necessary before doorbell writes. On x86, this avoids SFENCEs before doorbell writes in both the tx and rx refill paths. Signed-off-by: Venkatesh Srinivas <venkateshs@google.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Christian Grönke authored
This patch reverts two previous applied patches to fix an issue that appeared when using SGMII based SFP modules. In the current state the driver will try to reset the PHY before obtaining the phy_addr of the SGMII attached PHY. That leads to an error in e1000_write_phy_reg_sgmii_82575. Causing the initialization to fail: igb: Intel(R) Gigabit Ethernet Network Driver - version 5.4.0-k igb: Copyright (c) 2007-2014 Intel Corporation. igb: probe of ????:??:??.? failed with error -3 The patches being reverted are: commit 18278533 Author: Aaron Sierra <asierra@xes-inc.com> Date: Tue Nov 29 10:03:56 2016 -0600 igb: reset the PHY before reading the PHY ID commit 440aeca4 Author: Matwey V Kornilov <matwey@sai.msu.ru> Date: Thu Nov 24 13:32:48 2016 +0300 igb: Explicitly select page 0 at initialization The first reverted patch directly causes the problem mentioned above. In case of SGMII the phy_addr is not known at this point and will only be obtained by 'igb_get_phy_id_82575' further down in the code. The second removed patch selects forces selection of page 0 in the PHY. Something that the reset tries to address as well. As pointed out by Alexander Duzck, the patch below fixes the same issue but in the proper location: commit 4e684f59 Author: Chris J Arges <christopherarges@gmail.com> Date: Wed Nov 2 09:13:42 2016 -0500 igb: Workaround for igb i210 firmware issue Reverts: 440aeca4. Reverts: 18278533. Signed-off-by: Christian Grönke <c.groenke@infodas.de> Reviewed-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Shannon Nelson authored
Add the ixgbe's security configuration registers into the register dump. Signed-off-by: Shannon Nelson <shannon.nelson@oracle.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Tony Nguyen authored
XDP does not support jumbo frames or LRO. These checks are being made outside the driver when an XDP program is loaded, however, there is nothing preventing these from changing after an XDP program is loaded. Add the checks so that while an XDP program is loaded, do not allow MTU to be changed or LRO to be enabled. Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Gustavo A. R. Silva authored
Currently, code at label *out* is unreachable. Fix this by updating variable *ret* with -EINVAL, so the jump to *out* can be properly executed instead of directly returning from function. Addresses-Coverity-ID: 1472059 ("Structurally dead code") Fixes: 1e2b44e7 ("rds: Enable RDS IPv6 support") Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Acked-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Anders Roxell authored
Build error, implicit declaration of function __inet6_ehashfn shows up When RDS is enabled but not IPV6. net/rds/connection.c: In function ‘rds_conn_bucket’: net/rds/connection.c:67:9: error: implicit declaration of function ‘__inet6_ehashfn’; did you mean ‘__inet_ehashfn’? [-Werror=implicit-function-declaration] hash = __inet6_ehashfn(lhash, 0, fhash, 0, rds_hash_secret); ^~~~~~~~~~~~~~~ __inet_ehashfn Current code adds IPV6 as a depends on in config RDS. Fixes: eee2fa6a ("rds: Changing IP address internal representation to struct in6_addr") Signed-off-by: Anders Roxell <anders.roxell@linaro.org> Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Ursula Braun says: ==================== net/smc: patches 2018-07-25 here are 4 more patches for SMC: The first one is just a small code cleanup in preparation for patch 2. Patch 2 switches to the use of the vlan-gid for VLAN traffic. Patch 3 improves diagnosis when creating SMC connections. Patch 4 improves synchronization between local and remote link groups. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Karsten Graul authored
Send an orderly DELETE LINK request before termination of a link group, add support for client triggered DELETE LINK processing. And send a disorderly DELETE LINK before module is unloaded. Signed-off-by: Karsten Graul <kgraul@linux.ibm.com> Signed-off-by: Ursula Braun <ubraun@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Karsten Graul authored
Remember the fallback reason code and the peer diagnosis code for smc sockets, and provide them in smc_diag.c to the netlink interface. And add more detailed reason codes. Signed-off-by: Karsten Graul <kgraul@linux.ibm.com> Signed-off-by: Ursula Braun <ubraun@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Ursula Braun authored
SMC code uses the base gid for VLAN traffic. The gids exchanged in the CLC handshake and the gid index used for the QP have to switch from the base gid to the appropriate vlan gid. When searching for a matching IB device port for a certain vlan device, it does not make sense to return an IB device port, which is not enabled for the used vlan_id. Add another check whether a vlan gid exists for a certain IB device port. Signed-off-by: Ursula Braun <ubraun@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Ursula Braun authored
Link confirmation will always be sent across the new link being confirmed. This allows to shrink the parameter list. No functional change. Signed-off-by: Ursula Braun <ubraun@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
David S. Miller authored
Jakub Kicinski says: ==================== nfp: protect from theoretical size overflows and SR-IOV errors This small set changes the handling of pci_sriov_set_totalvfs() errors. nfp is the only driver which fails probe on pci_sriov_set_totalvfs() errors. It turns out some BIOS configurations may break SR-IOV and users who don't use that feature should not suffer. Remaining patches makes sure we use overflow-safe function for ring allocation, even though ring sizes are limited. It won't hurt and we can also enable fallback to vmalloc() if memory is tight while at it. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jakub Kicinski authored
Use array_size() and store the size as full size_t to protect from theoretical size overflow when handling HW descriptor rings. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jakub Kicinski authored
Commit 7f1c684a ("nfp: setup xdp_rxq_info") mixed the cache cold and cache hot data in the nfp_net_rx_ring structure (ignoring the feedback), to try to fit the structure into 2 cache lines after struct xdp_rxq_info was added. Now that we are about to add a new field the structure will grow back to 3 cache lines, so order the members correctly. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jakub Kicinski authored
Use kvcalloc() instead of tmp variable + kzalloc() when allocating SW buffer information to allow falling back to vmalloc and to protect from theoretical integer overflow. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Jakub Kicinski authored
On machines with buggy ACPI tables or when SR-IOV is already enabled we may not be able to set the SR-IOV VF limit in sysfs, it's not fatal because the limit is imposed by the driver anyway. Only the sysfs 'sriov_totalvfs' attribute will be too high. Print an error to inform user about the failure but allow probe to continue. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
YueHaibing authored
The dma_mapping_error() returns true or false, but we want to return -ENOMEM if there was an error. Fixes: 174fd259 ("amd-xgbe: Implement split header receive support") Signed-off-by: YueHaibing <yuehaibing@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
- 25 Jul, 2018 6 commits
-
-
David S. Miller authored
Ido Schimmel says: ==================== mlxsw: Introduce algorithmic TCAM support The Spectrum-2 ASIC uses an algorithmic TCAM (A-TCAM) where multiple exact matches lookups are performed instead of a single lookup as with standard circuit TCAM (C-TCAM) memory. This allows for higher scale and reduced power consumption. The lookups are performed by masking a packet using different masks (e.g., {dst_ip/24, ethtype}) defined for the region and looking for an exact match. Eventually, the rule with the highest priority will be picked. Since the number of masks per-region is limited, the ASIC includes a C-TCAM that can be used as a spill area for rules that do not fit into the A-TCAM. The driver currently uses a C-TCAM only mode which is similar to Spectrum-1. However, this mode severely limits both the number of supported ACL rules and the performance of the ACL lookup. This patch set introduces initial support for the A-TCAM mode where the C-TCAM is only used for rule spillage. The first five patches add the registers and ASIC resources needed in order to make use of the A-TCAM. Next three patches are the "meat" and add the eRP core which is used to manage the masks used by each ACL region. The individual commit messages are lengthy and aim to thoroughly explain the subject. The next seven patches perform small adjustments in the code and the related data structures and are meant to prepare the code base to the introduction of the A-TCAM in the last two patches. Various A-TCAM optimization will be the focus of follow-up patch sets: * Pruning - Used to reduce the number of lookups. Each rule will include a prune vector that indicates which masks should not be considered for further lookups as they cannot result in a higher priority match * Bloom filter - Used to reduce the number of lookups. Before performing a lookup with a given mask the ASIC will consult a bloom filter (managed by the driver) that indicates whether a match might exist using the considered mask * Masks aggregation - Used to increase scale and reduce lookups. Masks that only differ by up to eight consecutive bits (delta bits) can be aggregated into a single mask. The delta bits then become a part of the rule's key. For example, dst_ip/16 and dst_ip/17 can be represented as dst_ip/16 with a delta bit of one. Rules using the aggregated mask then specify whether the 17-th bit should be masked or not and its value ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
-
Ido Schimmel authored
Now that all the pieces are in place we can start using the A-TCAM instead of only using the C-TCAM. This allows for much higher scale and better performance (to be improved further by follow-up patch sets). Perform the integration with the A-TCAM and the eRP core by reverting the changes introduced by "mlxsw: spectrum_acl: Enable C-TCAM only mode in eRP core" and add calls from the C-TCAM code into the eRP core. Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Ido Schimmel authored
Implement rule insertion and deletion into the A-TCAM before we flip the driver to start using the A-TCAM. Rule insertion into the A-TCAM is very similar to C-TCAM, but there are subtle differences between regions of different sizes (i.e., different number of key blocks). Specifically, as explained in "mlxsw: spectrum_acl: Allow encoding a partial key", in 12 key blocks regions a rule is split into two and the two halves of the rule are linked using a "large entry key ID". Such differences are abstracted away by using different region operations per region type. Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Ido Schimmel authored
When A-TCAM will be used together with C-TCAM, the C-TCAM code will need to call into the eRP core in order to get an eRP for an inserted entry. The eRP core takes an A-TCAM region as one of its arguments, so pass the C-TCAM region to the insertion function which will later allow us to derive the A-TCAM region, given it contains the C-TCAM one. Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Ido Schimmel authored
Before we start using the A-TCAM we need to make sure the region is properly initialized. This includes the setting of its type (which affects the size of its eRP table, for example) and its registration with the eRP core. Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-
Ido Schimmel authored
Each TCAM region currently uses its own resources and there is no sharing between the different regions. This is going to change with A-TCAM as each region will need to allocate an eRP table from the global eRP tables array. Make the global TCAM resources available to each region by passing the TCAM private data to the region initialization routine. Signed-off-by: Ido Schimmel <idosch@mellanox.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-