- 20 May, 2021 9 commits
-
-
Andre Guedes authored
Add support for transmitting packets via AF_XDP zero-copy mechanism. The packet transmission itself is implemented by igc_xdp_xmit_zc() which is called from igc_clean_tx_irq() when the ring has AF_XDP zero-copy enabled. Likewise i40e and ice drivers, the transmission budget used is the number of descriptors available on the ring. A new tx buffer type is introduced to 'enum igc_tx_buffer_type' to indicate the tx buffer uses memory from xsk pool so it can be properly cleaned after transmission or when the ring is cleaned. The I225 controller has only 4 Tx hardware queues so the main difference between igc and other Intel drivers that support AF_XDP zero-copy is that there is no tx ring dedicated exclusively to XDP. Instead, tx rings are shared between the network stack and XDP, and netdev queue lock is used to ensure mutual exclusion. This is the same approach implemented to support XDP_TX and XDP_REDIRECT actions. Signed-off-by:
Andre Guedes <andre.guedes@intel.com> Signed-off-by:
Vedang Patel <vedang.patel@intel.com> Signed-off-by:
Jithu Joseph <jithu.joseph@intel.com> Reviewed-by:
Maciej Fijalkowski <maciej.fijalkowski@intel.com> Tested-by:
Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com>
-
Andre Guedes authored
Add support for receiving packets via AF_XDP zero-copy mechanism. Add a new flag to 'enum igc_ring_flags_t' to indicate the ring has AF_XDP zero-copy enabled so proper ring setup is carried out during ring configuration in igc_configure_rx_ring(). RX buffers can now be allocated via the shared pages mechanism (default behavior of the driver) or via xsk pool (when AF_XDP zero-copy is enabled) so a union is added to the 'struct igc_rx_buffer' to cover both cases. When AF_XDP zero-copy is enabled, rx buffers are allocated from the xsk pool using the new helper igc_alloc_rx_buffers_zc() which is the counterpart of igc_alloc_rx_buffers(). Likewise other Intel drivers that support AF_XDP zero-copy, in igc we have a dedicated path for cleaning up rx irqs when zero-copy is enabled. This avoids adding too many checks within igc_clean_rx_irq(), resulting in a more readable and efficient code since this function is called from the hot-path of the driver. Signed-off-by:
-
Andre Guedes authored
Up to this point, Tx buffers are associated with either a skb or a xdpf, and the IGC_TX_FLAGS_XDP flag was enough to distinguish between these two case. However, with upcoming patches that will add AF_XDP zero-copy support, a third case will be introduced so this flag-based approach won't fit well. In preparation to land AF_XDP zero-copy support, replace the IGC_TX_FLAGS_XDP flag by an enum which will be extended once zero-copy support is introduced to the driver. Signed-off-by:
-
Andre Guedes authored
In preparation for AF_XDP zero-copy support, encapsulate the code that unmaps Tx buffers into its own local helper so we can reuse it, avoiding code duplication. Signed-off-by:
-
Andre Guedes authored
In preparation for AF_XDP zero-copy support, encapsulate the code that updates the driver RX stats in its own local helper so it can be reused in the zero-copy path. Likewise, encapsulate TX stats code as well. Signed-off-by:
-
Andre Guedes authored
Refactor XDP rxq info registration code, preparing the driver for AF_XDP zero-copy support which is added by upcoming patches. Currently, xdp_rxq and memory model are both registered during RX resource setup time by igc_xdp_register_rxq_info() helper. With AF_XDP, we want to register the memory model later on while configuring the ring because we will know which memory model type to register (MEM_TYPE_PAGE_SHARED or MEM_TYPE_XSK_BUFF_POOL). The helpers igc_xdp_register_rxq_info() and igc_xdp_unregister_rxq_ info() are not useful anymore so they are removed. Signed-off-by:
-
Andre Guedes authored
Refactor igc_clean_rx_ring() helper, preparing the code for AF_XDP zero-copy support which is added by upcoming patches. The refactor consists of encapsulating page-shared specific code into its own helper, leaving common code that will be shared by both page-shared and xsk pool in igc_clean_rx_ring(). Signed-off-by:
-
Andre Guedes authored
Refactor __igc_xdp_run_prog() helper from igc_xdp_run_prog(), preparing the code for AF_XDP zero-copy support which is added by upcoming patches. The existing igc_xdp_run_prog() caters to regular XDP rx path which has to verify if bpf_prog is not NULL. Zero-copy path assumes that bpf_prog is not NULL and hence this check is not required. Therefore it makes sense to refactor the common code into a helper function, to avoid code duplication. Signed-off-by:
-
Andre Guedes authored
Move the helper igc_xdp_is_enabled() to igc_xdp.h so it can be reused in igc_xdp.c by upcoming patches that will introduce AF_XDP zero-copy support to the driver. Signed-off-by:
-
- 19 May, 2021 31 commits
-
-
Wei Yongjun authored
Fix to return a negative error code -ENOMEM from the error handling case instead of 0, as done elsewhere in this function. Fixes: c6e08d62 ("net: qrtr: Allocate workqueue before kernel_bind") Reported-by:
Hulk Robot <hulkci@huawei.com> Signed-off-by:
Wei Yongjun <weiyongjun1@huawei.com> Reviewed-by:
Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Wei Yongjun authored
In case of error, the function mdiobus_get_phy() returns NULL pointer not ERR_PTR(). The IS_ERR() test in the return value check should be replaced with NULL test. Reported-by:
-
Davide Caratti authored
modern userspace applications, like OVN, can configure the TC datapath to "recirculate" packets several times. If more than 4 "recirculation" rules are configured, packets can be dropped by __tcf_classify(). Changing the maximum number of reclassifications (from 4 to 16) should be sufficient to prevent drops in most use cases, and guard against loops at the same time. Signed-off-by:
Davide Caratti <dcaratti@redhat.com> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-nextDavid S. Miller authored
Alexei Starovoitov says: ==================== pull-request: bpf-next 2021-05-19 The following pull-request contains BPF updates for your *net-next* tree. We've added 43 non-merge commits during the last 11 day(s) which contain a total of 74 files changed, 3717 insertions(+), 578 deletions(-). The main changes are: 1) syscall program type, fd array, and light skeleton, from Alexei. 2) Stop emitting static variables in skeleton, from Andrii. 3) Low level tc-bpf api, from Kumar. 4) Reduce verifier kmalloc/kfree churn, from Lorenz. ====================
-
David S. Miller authored
Ido Schimmel says: ==================== mlxsw: Add support for new multipath hash policies This patchset adds support for two new multipath hash policies in mlxsw. Patch #1 emits net events whenever the net.ipv{4,6}.fib_multipath_hash_fields sysctls are changed. This allows listeners to react to changes in the packet fields used for the computation of the multipath hash. Patches #2-#3 refactor the code in mlxsw that is responsible for the configuration of the multipath hash, so that it will be easier to extend for the two new policies. Patch #4 adds the register fields required to support the new policies. Patch #5-#7 add support for inner layer 3 and custom multipath hash policies. Tested using following forwarding selftests: * custom_multipath_hash.sh * gre_custom_multipath_hash.sh * gre_inner_v4_multipath.sh * gre_inner_v6_multipath.sh ==================== -
Ido Schimmel authored
When this policy is set, only enable the packet fields that were enabled by user space for multipath hash computation. Signed-off-by:
Ido Schimmel <idosch@nvidia.com> Reviewed-by:
Petr Machata <petrm@nvidia.com> Signed-off-by:
-
Ido Schimmel authored
When this policy is set, the kernel uses the inner layer 3 fields for multipath hash computation and falls back to the outer fields if no encapsulation was encountered. This behavior is most likely influenced by the behavior of the flow dissector, which is used for the packet dissection. The Spectrum ASIC, however, cannot fallback to outer fields if inner fields are not available. This should not result in a discrepancy from the software data path because if several flows have matching inner fields, they will tend to have matching outer fields as well. Therefore, implement this policy by enabling both outer and inner layer 3 fields for the multipath hash computation. Signed-off-by:
-
Ido Schimmel authored
Outer IPv4 and IPv6 addresses are used by multiple multipath hash policies. Factor out helpers that set these fields to increase code sharing between different policies. Signed-off-by:
-
Ido Schimmel authored
The RECRv2 register is used for setting up the router's ECMP hash configuration. Extend it with inner packet fields to allow the ECMP hash to be calculated based on inner flow information. Signed-off-by:
-
Ido Schimmel authored
Currently, the multipath hash configuration is written directly to the register payload. While this is OK for the two currently supported policies, it is going to be hard to follow when more policies and more packet fields are added. Instead, set the required headers and fields in a bitmap and then dump it to the register payload. Signed-off-by:
-
Ido Schimmel authored
The code was written when only two multipath hash policies were present, so the if statement was sufficient. The next patch and future patches are going to add support for more policies, so move to a switch statement. Signed-off-by:
-
Ido Schimmel authored
In-kernel notifications are already sent when the multipath hash policy itself changes, but not when the multipath hash fields change. Add these notifications, so that interested listeners (e.g., switch ASIC drivers) could perform the necessary configuration. Signed-off-by:
-
Stephan Gerhold authored
S3FWRN5 depends on a clock input ("XI" pin) to function properly. Depending on the hardware configuration this could be an always-on oscillator or some external clock that must be explicitly enabled. So far we assumed that the clock is always-on. Make the driver request an (optional) clock from the device tree and make sure the clock is running before starting S3FWRN5. Note: S3FWRN5 asserts "GPIO2" whenever it needs the clock input to function correctly. On some hardware configurations, GPIO2 is connected directly to an input pin of the external clock provider (e.g. the main PMIC of the SoC). In that case, it can automatically AND the clock enable bit and clock request from S3FWRN5 so that the clock is actually only enabled when needed. It is also conceivable that on some other hardware configuration S3FWRN5's GPIO2 might be connected as a regular GPIO input of the SoC. In that case, follow-up patches could extend the driver to request the GPIO, set up an interrupt and only enable the clock when requested by S3FWRN5. Signed-off-by:Stephan Gerhold <stephan@gerhold.net> Reviewed-by:
Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com> Signed-off-by:
-
Stephan Gerhold authored
On some systems, S3FWRN5 depends on having an external clock enabled to function correctly. Allow declaring that clock in the device tree. Signed-off-by:
-
Zheng Yejian authored
loginuid/sessionid/secid have been read from 'current' instead of struct netlink_skb_parms, the parameter 'skb' seems no longer needed. Fixes: c53fa1ed ("netlink: kill loginuid/sessionid/sid members from struct netlink_skb_parms") Signed-off-by:
Zheng Yejian <zhengyejian1@huawei.com> Signed-off-by:
-
David S. Miller authored
Guangbin Huang says: ==================== net: intel: some cleanups This patchset adds some cleanups for intel e1000/e1000e ethernet driver. ==================== Reviewed-by:
-
Hao Chen authored
There is a misspell word "retreived" in comment, so fix it to "retrieved". Signed-off-by:
Hao Chen <chenhao288@hisilicon.com> Signed-off-by:
Guangbin Huang <huangguangbin2@huawei.com> Signed-off-by:
-
Hao Chen authored
There are double "slot" in comment, so remove the redundant one. Signed-off-by:
-
Hao Chen authored
There are double "the" in comment, so remove the redundant one. Signed-off-by:
-
Hao Chen authored
There are double "in" and "to" in comments, so remove the redundant one. Signed-off-by:
-
Hao Chen authored
There are double "slot" in comment, so remove the redundant one. Signed-off-by:
-
David S. Miller authored
Hui Tang says: ==================== net: ethernet: remove leading spaces before tabs There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' ==================== Signed-off-by: -
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: Masahiro Yamada <masahiroy@kernel.org> Signed-off-by:
Hui Tang <tanghui20@huawei.com> Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: Armin Wolf <W_Armin@gmx.de> Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: Masahiro Yamada <masahiroy@kernel.org> Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: Vaibhav Gupta <vaibhavgupta40@gmail.com> Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org> Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: Masahiro Yamada <masahiroy@kernel.org> Signed-off-by:
-
Hui Tang authored
There are a few leading spaces before tabs and remove it by running the following commard: $ find . -name '*.c' | xargs sed -r -i 's/^[ ]+\t/\t/' $ find . -name '*.h' | xargs sed -r -i 's/^[ ]+\t/\t/' Cc: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by:
-
