- 17 Oct, 2013 6 commits
-
-
Eyal Perry authored
Mellanox ConnectX architecture is: mlx4_core is the lower level PCI driver which register on the PCI id, and protocol specific drivers are depended on it: mlx4_en - for Ethernet and mlx4_ib for Infiniband. NIC could have multiple ports which can change their type dynamically. We use the request_module() call to load the relevant protocol driver when needed: on loading time or at port type change event. Signed-off-by:
Eyal Perry <eyalpe@mellanox.com> Signed-off-by:
Amir Vadai <amirv@mellanox.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Amir Vadai authored
Clean up warning added by commit fe6f700d "net/mlx4_core: Respond to operation request by firmware". Signed-off-by:
-
Or Gerlitz authored
Small code cleanup: 1. change MLX4_DEV_CAP_FLAGS2_REASSIGN_MAC_EN to MLX4_DEV_CAP_FLAG2_REASSIGN_MAC_EN 2. put MLX4_SET_PORT_PRIO2TC and MLX4_SET_PORT_SCHEDULER in the same union with the other MLX4_SET_PORT_yyy Signed-off-by:
Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by:
-
Or Gerlitz authored
Remove code that triggers trivial build warnings. drivers/net/ethernet/mellanox/mlx4/cmd.c: In function ‘mlx4_set_vf_vlan’: drivers/net/ethernet/mellanox/mlx4/cmd.c:2256: warning: variable ‘vf_oper’ set but not used drivers/net/ethernet/mellanox/mlx4/mcg.c: In function ‘mlx4_map_sw_to_hw_steering_mode’: drivers/net/ethernet/mellanox/mlx4/mcg.c:648: warning: comparison of unsigned expression < 0 is always false drivers/net/ethernet/mellanox/mlx4/mcg.c: In function ‘mlx4_map_sw_to_hw_steering_id’: drivers/net/ethernet/mellanox/mlx4/mcg.c:685: warning: comparison of unsigned expression < 0 is always false drivers/net/ethernet/mellanox/mlx4/mcg.c: In function ‘mlx4_hw_rule_sz’: drivers/net/ethernet/mellanox/mlx4/mcg.c:712: warning: comparison of unsigned expression < 0 is always false drivers/net/ethernet/mellanox/mlx4/fw.c: In function ‘mlx4_opreq_action’: drivers/net/ethernet/mellanox/mlx4/fw.c:1732: warning: variable ‘type_m’ set but not used drivers/net/ethernet/mellanox/mlx4/srq.c:302: warning: no previous prototype for ‘mlx4_srq_lookup’ Signed-off-by:
-
Eric Dumazet authored
TCP listener refactoring, part 6 : Use sock_gen_put() from inet_diag_dump_one_icsk() for future SYN_RECV support. Signed-off-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
-
git://git.open-mesh.org/linux-mergeDavid S. Miller authored
Included changes: - ensure RecordRoute information is added to BAT_ICMP echo_request/reply only - use VLAN_ETH_HLEN when possible - use htons when possible - substitute old fragmentation code with a new improved implementation by Martin Hundebøll - create common header for BAT_ICMP packets to improve extendibility - consider the network coding overhead when computing the overall room needed by batman headers - add dummy soft-interface rx mode handler - minor code refactoring and cleanups Signed-off-by:
-
- 12 Oct, 2013 16 commits
-
-
Linus Lüssing authored
We do not actually need to set any rx filters for the virtual batman soft interface. However a dummy handler enables a user to set static multicast listeners for instance. Signed-off-by:
Linus Lüssing <linus.luessing@web.de> Signed-off-by:
Marek Lindner <lindner_marek@yahoo.de> Signed-off-by:
Antonio Quartulli <antonio@meshcoding.com>
-
Antonio Quartulli authored
This is a simple batadv_tt_save_orig_buffer() refactoring aiming to make it more generic and avoid useless casts. Signed-off-by:
Antonio Quartulli <antonio@open-mesh.com> Signed-off-by:
-
Antonio Quartulli authored
Implement batadv_tt_entries() to get the number of entries fitting in a given amount of bytes. This computation is done several times in the code and therefore it is useful to have an helper function. Signed-off-by:
-
Simon Wunderlich authored
This is not used anymore with the new fragmentation, and it might actually mess up the bonding code because find_router() assumes it is only called once per packet. Signed-off-by:
Simon Wunderlich <simon@open-mesh.com> Signed-off-by:
-
Marek Lindner authored
The module prints a warning when the MTU on the hard interface is too small to transfer payload traffic without fragmentation. The required MTU is calculated based on the encapsulation header size. If network coding is compild into the module its header size is taken into account as well. Signed-off-by:
-
Antonio Quartulli authored
the icmp and the icmp_rr packets share the same initial fields since they use the same code to be processed and forwarded. Extract the common fields and put them into a separate struct so that future ICMP packets can be easily added without bloating the packet definition. However, keep the seqno field outside of the newly created common header because future ICMP types may require a bigger sequence number space. This change breaks compatibility due to fields reordering in the ICMP headers. Signed-off-by:
-
Antonio Quartulli authored
When comparing a network ordered value with a constant, it is better to convert the constant at compile time by means of htons() instead of converting the value at runtime using ntohs(). This refactoring may slightly improve the code performance. Moreover substitute __constant_htons() with htons() since the latter increase readability and it is smart enough to be as efficient as the former Signed-off-by:
Antonio Quartulli <ordex@autistici.org> Signed-off-by:
Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
-
Martin Hundebøll authored
Non-broadcast packets larger than MTU are fragmented and sent with an encapsulating header. Up to 16 fragments are supported, which are sent in reverse order on the wire to allow minimal memory copying when creating fragments. Signed-off-by:
Martin Hundebøll <martin@hundeboll.net> Signed-off-by:
-
Martin Hundebøll authored
Fragments arriving at their destination are buffered for later merge. Merged packets are passed to the main receive function as had they never been fragmented. Fragments are forwarded without merging if the MTU of the outgoing interface is smaller than the size of the merged packet. Signed-off-by:
-
Martin Hundebøll authored
Remove the existing fragmentation code before adding the new version and delete unicast.{h,c}. batadv_unicast_send_skb() is moved to send.c and renamed to batadv_send_skb_unicast(). fragmentation entry in sysfs (bat_priv->fragmentation) is kept for use in the new fragmentation code. BATADV_UNICAST_FRAG packet type is renamed to BATADV_FRAG for use in the new fragmentation code. Signed-off-by: -
Antonio Quartulli authored
Signed-off-by:
-
Antonio Quartulli authored
In case of a VLAN tagged frame the ethhdr pointer is moved forward by 4 bytes so that the offset of h_proto in struct ethhdr matches the real h_vlan_encapsulated_proto address in the skb. While this trickery is correct it makes the code harder to understand and may lead to bugs in case of re-use of ethhdr for other purposes. This patch introduces a proto variable to make things cleaner and easier to understand. Signed-off-by:
-
Antonio Quartulli authored
batadv_tt_global_entry_free_ref uses call_rcu to schedule a function which will only free the global entry itself. For this reason call_rcu is useless and kfree_rcu can be used to simplify the code. Signed-off-by:
-
Antonio Quartulli authored
batadv_tt_global_add_orig is neither used nor implemented anymore, therefore it is possible to remove its declaration Signed-off-by:
-
Antonio Quartulli authored
batadv_tt_global_add is not used anymore outside of the TT code thanks to the TVLV implementation. It can therefore be declared as static Last user has been removed by 3de4e64df0f1326db7cc0ef25f5af8522850252d ("batman-adv: tvlv - convert roaming adv packet to use tvlv unicast packets") Moreover make it return bool since its result can be either 0 or 1. Reported-by: -
Simon Wunderlich authored
Adding host information for record route is only required for ICMP requests and replys, and should not be added to just any (future?) packet type. Signed-off-by:
-
- 11 Oct, 2013 2 commits
-
-
Eric Dumazet authored
1) We need to take a timestamp only for skb that should be cloned. Other skbs are not in write queue and no rtt estimation is done on them. 2) the unlikely() hint is wrong for receivers (they send pure ACK) Signed-off-by:
Yuchung Cheng <ycheng@google.com> Signed-off-by:
-
git://git.open-mesh.org/linux-mergeDavid S. Miller authored
Included changes: - update emails for A. Quartulli and M. Lindner in MAINTAINERS - switch to the next on-the-wire protocol version - introduce the T(ype) V(ersion) L(ength) V(alue) framework - adjust the existing components to make them use the new TVLV code - make the TT component use CRC32 instead of CRC16 - totally remove the VIS functionality (has been moved to userspace) - reorder packet types and flags - add static checks on packet format - remove __packed from batadv_ogm_packet
-
- 10 Oct, 2013 16 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-nextDavid S. Miller authored
Jeff Kirsher says: ==================== This series contains updates to i40e only. Alex provides the majority of the patches against i40e, where he does cleanup of the Tx and RX queues and to align the code with the known good Tx/Rx queue code in the ixgbe driver. Anjali provides an i40e patch to update link events to not print to the log until the device is administratively up. Catherine provides a patch to update the driver version. ==================== Signed-off-by: -
Eric Dumazet authored
In commit 634fb979 ("inet: includes a sock_common in request_sock") I forgot that the two ports in sock_common do not have same byte order : skc_dport is __be16 (network order), but skc_num is __u16 (host order) So sparse complains because ir_loc_port (mapped into skc_num) is considered as __u16 while it should be __be16 Let rename ir_loc_port to ireq->ir_num (analogy with inet->inet_num), and perform appropriate htons/ntohs conversions. Signed-off-by:
Wu Fengguang <fengguang.wu@intel.com> Signed-off-by:
-
Catherine Sullivan authored
Update the version number of the driver. Signed-off-by:
Catherine Sullivan <catherine.sullivan@intel.com> Signed-off-by:
Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by:
Kavindya Deegala <kavindya.s.deegala@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Alexander Duyck authored
This change brings support for 64 bit netstats to the driver. Previously the stats were 64 bit but highly racy due to the fact that 64 bit transactions are not atomic on 32 bit systems. This change makes is so that the 64 bit byte and packet stats are reliable on all architectures. Signed-off-by:
Alexander Duyck <alexander.h.duyck@intel.com> Signed-off-by:
-
Alexander Duyck authored
Allocate the queue pairs individually instead of as a group. This allows for much easier queue management as it is possible to dynamically resize the queues without having to free and allocate the entire block. Ease statistic collection by treating Tx/Rx queue pairs as a single unit. Each pair is allocated together and starts with a Tx queue and ends with an Rx queue. By ordering them this way it is possible to know the Rx offset based on a pointer to the Tx queue. Signed-off-by:
-
Alexander Duyck authored
This replaces the ring container array with a linked list. The idea is to make the logic much easier to deal with since this will allow us to call a simple helper function from the q_vectors to go through the entire list. Signed-off-by:
-
Alexander Duyck authored
Allocate the q_vectors individually. The advantage to this is that it allows for easier freeing and allocation. In addition it makes it so that we could do node specific allocations at some point in the future. Signed-off-by:
-
Alexander Duyck authored
This makes it so that the Tx and Rx byte and packet counts are separated from the rest of the statistics. This allows for better isolation of these stats when we move them into the 64 bit statistics. Simplify things by re-ordering how the stats display in ethtool. Instead of displaying all of the Tx queues as a block, followed by all the Rx queues, the new order is Tx[0], Rx[0], Tx[1], Rx[1], ..., Tx[n], Rx[n]. This reduces the loops and cleans up the display for testing purposes since it is very easy to verify if flow director is doing the right thing as the Tx and Rx queue pair are shown in pairs. Signed-off-by:
-
Alexander Duyck authored
Implement BQL (byte queue limit) support in i40e. Signed-off-by:
-
Alexander Duyck authored
Drop Tx flag and TXSW which is tested but never set. As a result of this change we can drop a complicated check that always resulted in the final result of i40e_tx_csum being equal to the CHECKSUM_PARTIAL value. As such we can replace the entire function call with just a check for skb->summed == CHECKSUM_PARTIAL. Signed-off-by:
-
Alexander Duyck authored
Sync the fast path for i40e_tx_map and i40e_clean_tx_irq so that they are similar to igb and ixgbe. - Only update the Tx descriptor ring in tx_map - Make skb mapping always on the first buffer in the chain - Drop the use of MAPPED_AS_PAGE Tx flag - Only store flags on the first buffer_info structure Signed-off-by:
-
Alexander Duyck authored
Avoid directly incrementing next_to_use for multiple reasons. The main reason being that if we directly increment it then it can attain a state where it is equal to the ring count. Technically this is a state it should not be able to reach but the way this is written it now can. This patch pulls the value off into a register and then increments it and writes back either the value or 0 depending on if the value is equal to the ring count. Signed-off-by:
-
Alexander Duyck authored
- drop the mapped_as_page u8 from the Tx buffer info as it was unused - use the DMA unmap accessors for Tx DMA - replace checks of DMA with checks of the unmap length to verify if an unmap is needed - update the Tx buffer layout to make it consistent with igb, ixgbe Signed-off-by:
-
Eric Dumazet authored
sk_pacing_rate is read by sch_fq packet scheduler at any time, with no synchronization, so make sure we update it in a sensible way. ACCESS_ONCE() is how we instruct compiler to not do stupid things, like using the memory location as a temporary variable. Signed-off-by:
-
Eric Dumazet authored
TCP listener refactoring, part 5 : We want to be able to insert request sockets (SYN_RECV) into main ehash table instead of the per listener hash table to allow RCU lookups and remove listener lock contention. This patch includes the needed struct sock_common in front of struct request_sock This means there is no more inet6_request_sock IPv6 specific structure. Following inet_request_sock fields were renamed as they became macros to reference fields from struct sock_common. Prefix ir_ was chosen to avoid name collisions. loc_port -> ir_loc_port loc_addr -> ir_loc_addr rmt_addr -> ir_rmt_addr rmt_port -> ir_rmt_port iif -> ir_iif Signed-off-by:
-
Eric Dumazet authored
skb_gro_receive() is currently limited to 16 or 17 MSS per GRO skb, typically 24616 bytes, because it fills up to MAX_SKB_FRAGS frags. It's relatively easy to extend the skb using frag_list to allow more frags to be appended into the last sk_buff. This still builds very efficient skbs, and allows reaching 45 MSS per skb. (45 MSS GRO packet uses one skb plus a frag_list containing 2 additional sk_buff) High speed TCP flows benefit from this extension by lowering TCP stack cpu usage (less packets stored in receive queue, less ACK packets processed) Forwarding setups could be hurt, as such skbs will need to be linearized, although its not a new problem, as GRO could already provide skbs with a frag_list. We could make the 65536 bytes threshold a tunable to mitigate this. (First time we need to linearize skb in skb_needs_linearize(), we could lower the tunable to ~16*1460 so that following skb_gro_receive() calls build smaller skbs) Signed-off-by:
-
