Skip to content

K
Kernel

openeuler / Kernel
接近 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 16 6月, 2017 5 次提交
    • T
      net/mlx4_en: Increase default TX ring size · 77788b5b
      由 Tariq Toukan 提交于 
      Increase the default TX ring size (from 512 to 1024) to match
the RX ring size.
This gives the XDP TX ring a better chance to keep up with the
rate of its RX ring in case of a high load of XDP_TX actions.

Tested:
Ethtool counter rx_xdp_tx_full used to increase, after applying this
patch it stopped.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Reviewed-by: NSaeed Mahameed <saeedm@mellanox.com>
Cc: kernel-team@fb.com
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      77788b5b
    • T
      net/mlx4_en: Poll XDP TX completion queue in RX NAPI · 6c78511b
      由 Tariq Toukan 提交于 
      Instead of having their own NAPIs, XDP TX completion queues get
polled within the corresponding RX NAPI.
This prevents any possible race on TX ring prod/cons indices,
between the context that issues the transmits (RX NAPI) and the
context that handles the completions (was previously done in
a separate NAPI).

This also improves performance, as it decreases the number
of NAPIs running on a CPU, saving the overhead of syncing
and switching between the contexts.

Performance tests:
Tested on ConnectX3Pro, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
Single queue no-RSS optimization ON.

XDP_TX packet rate:
-------------------------------------
     | Before    | After     | Gain |
IPv4 | 12.0 Mpps | 13.8 Mpps |  15% |
IPv6 | 12.0 Mpps | 13.8 Mpps |  15% |
-------------------------------------
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Reviewed-by: NSaeed Mahameed <saeedm@mellanox.com>
Cc: kernel-team@fb.com
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      6c78511b
    • T
      net/mlx4_en: Improve XDP xmit function · 36ea7964
      由 Tariq Toukan 提交于 
      Several performance improvements in XDP TX datapath,
including:
- Ring a single doorbell for XDP TX ring per NAPI budget,
  instead of doing it per a lower threshold (was 8).
  This includes removing the flow of immediate doorbell ringing
  in case of a full TX ring.
- Compiler branch predictor hints.
- Calculate values in compile time rather than in runtime.

Performance tests:
Tested on ConnectX3Pro, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
Single queue no-RSS optimization ON.

XDP_TX packet rate:
-------------------------------------
     | Before    | After     | Gain |
IPv4 | 10.3 Mpps | 12.0 Mpps |  17% |
IPv6 | 10.3 Mpps | 12.0 Mpps |  17% |
-------------------------------------
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Reviewed-by: NSaeed Mahameed <saeedm@mellanox.com>
Cc: kernel-team@fb.com
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      36ea7964
    • S
      net/mlx4_en: Optimized single ring steering · 4931c6ef
      由 Saeed Mahameed 提交于 
      Avoid touching RX QP RSS context when loading with only
one RX ring, to allow optimized A0 RX steering.

Enable by:
- loading mlx4_core with module param: log_num_mgm_entry_size = -6.
- then: ethtool -L <interface> rx 1

Performance tests:
Tested on ConnectX3Pro, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz

XDP_DROP packet rate:
-------------------------------------
     | Before    | After     | Gain |
IPv4 | 20.5 Mpps | 28.1 Mpps |  37% |
IPv6 | 18.4 Mpps | 28.1 Mpps |  53% |
-------------------------------------
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Cc: kernel-team@fb.com
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      4931c6ef
    • T
      net/mlx4_en: Remove unused argument in TX datapath function · cf97050d
      由 Tariq Toukan 提交于 
      Remove owner argument, as it is obsolete and unused.
This also saves the overhead of calculating its value in data-path.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Reviewed-by: NSaeed Mahameed <saeedm@mellanox.com>
Cc: kernel-team@fb.com
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      cf97050d
  3. 08 6月, 2017 1 次提交
  5. 10 3月, 2017 9 次提交
  7. 27 2月, 2017 1 次提交
    • E
      net/mlx4_en: fix overflow in mlx4_en_init_timestamp() · 47d3a075
      由 Eric Dumazet 提交于 
      The cited commit makes a great job of finding optimal shift/multiplier
values assuming a 10 seconds wrap around, but forgot to change the
overflow_period computation.

It overflows in cyclecounter_cyc2ns(), and the final result is 804 ms,
which is silly.

Lets simply use 5 seconds, no need to recompute this, given how it is
supposed to work.

Later, we will use a timer instead of a work queue, since the new RX
allocation schem will no longer need mlx4_en_recover_from_oom() and the
service_task firing every 250 ms.

Fixes: 31c128b6 ("net/mlx4_en: Choose time-stamping shift value according to HW frequency")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Cc: Eugenia Emantayev <eugenia@mellanox.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      47d3a075
  9. 20 2月, 2017 1 次提交
  11. 16 2月, 2017 1 次提交
  13. 03 2月, 2017 1 次提交
  15. 09 12月, 2016 1 次提交
  17. 30 11月, 2016 1 次提交
    • E
      mlx4: give precise rx/tx bytes/packets counters · 40931b85
      由 Eric Dumazet 提交于 
      mlx4 stats are chaotic because a deferred work queue is responsible
to update them every 250 ms.

Even sampling stats every one second with "sar -n DEV 1" gives
variations like the following :

lpaa23:~# sar -n DEV 1 10 | grep eth0 | cut -c1-65
07:39:22         eth0 146877.00 3265554.00   9467.15 4828168.50
07:39:23         eth0 146587.00 3260329.00   9448.15 4820445.98
07:39:24         eth0 146894.00 3259989.00   9468.55 4819943.26
07:39:25         eth0 110368.00 2454497.00   7113.95 3629012.17  <<>>
07:39:26         eth0 146563.00 3257502.00   9447.25 4816266.23
07:39:27         eth0 145678.00 3258292.00   9389.79 4817414.39
07:39:28         eth0 145268.00 3253171.00   9363.85 4809852.46
07:39:29         eth0 146439.00 3262185.00   9438.97 4823172.48
07:39:30         eth0 146758.00 3264175.00   9459.94 4826124.13
07:39:31         eth0 146843.00 3256903.00   9465.44 4815381.97
Average:         eth0 142827.50 3179259.70   9206.30 4700578.16

This patch allows rx/tx bytes/packets counters being folded at the
time we need stats.

We now can fetch stats every 1 ms if we want to check NIC behavior
on a small time window. It is also easier to detect anomalies.

lpaa23:~# sar -n DEV 1 10 | grep eth0 | cut -c1-65
07:42:50         eth0 142915.00 3177696.00   9212.06 4698270.42
07:42:51         eth0 143741.00 3200232.00   9265.15 4731593.02
07:42:52         eth0 142781.00 3171600.00   9202.92 4689260.16
07:42:53         eth0 143835.00 3192932.00   9271.80 4720761.39
07:42:54         eth0 141922.00 3165174.00   9147.64 4679759.21
07:42:55         eth0 142993.00 3207038.00   9216.78 4741653.05
07:42:56         eth0 141394.06 3154335.64   9113.85 4663731.73
07:42:57         eth0 141850.00 3161202.00   9144.48 4673866.07
07:42:58         eth0 143439.00 3180736.00   9246.05 4702755.35
07:42:59         eth0 143501.00 3210992.00   9249.99 4747501.84
Average:         eth0 142835.66 3182165.93   9206.98 4704874.08
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      40931b85
  19. 25 11月, 2016 1 次提交
    • E
      mlx4: reorganize struct mlx4_en_tx_ring · e3f42f84
      由 Eric Dumazet 提交于 
      Goal is to reorganize this critical structure to increase performance.

ndo_start_xmit() should only dirty one cache line, and access as few
cache lines as possible.

Add sp_ (Slow Path) prefix to fields that are not used in fast path,
to make clear what is going on.

After this patch pahole reports something much better, as all
ndo_start_xmit() needed fields are packed into two cache lines instead
of seven or eight

struct mlx4_en_tx_ring {
	u32                        last_nr_txbb;         /*     0   0x4 */
	u32                        cons;                 /*   0x4   0x4 */
	long unsigned int          wake_queue;           /*   0x8   0x8 */
	struct netdev_queue *      tx_queue;             /*  0x10   0x8 */
	u32                        (*free_tx_desc)(struct mlx4_en_priv *, struct mlx4_en_tx_ring *, int, u8, u64, int); /*  0x18   0x8 */
	struct mlx4_en_rx_ring *   recycle_ring;         /*  0x20   0x8 */

	/* XXX 24 bytes hole, try to pack */

	/* --- cacheline 1 boundary (64 bytes) --- */
	u32                        prod;                 /*  0x40   0x4 */
	unsigned int               tx_dropped;           /*  0x44   0x4 */
	long unsigned int          bytes;                /*  0x48   0x8 */
	long unsigned int          packets;              /*  0x50   0x8 */
	long unsigned int          tx_csum;              /*  0x58   0x8 */
	long unsigned int          tso_packets;          /*  0x60   0x8 */
	long unsigned int          xmit_more;            /*  0x68   0x8 */
	struct mlx4_bf             bf;                   /*  0x70  0x18 */
	/* --- cacheline 2 boundary (128 bytes) was 8 bytes ago --- */
	__be32                     doorbell_qpn;         /*  0x88   0x4 */
	__be32                     mr_key;               /*  0x8c   0x4 */
	u32                        size;                 /*  0x90   0x4 */
	u32                        size_mask;            /*  0x94   0x4 */
	u32                        full_size;            /*  0x98   0x4 */
	u32                        buf_size;             /*  0x9c   0x4 */
	void *                     buf;                  /*  0xa0   0x8 */
	struct mlx4_en_tx_info *   tx_info;              /*  0xa8   0x8 */
	int                        qpn;                  /*  0xb0   0x4 */
	u8                         queue_index;          /*  0xb4   0x1 */
	bool                       bf_enabled;           /*  0xb5   0x1 */
	bool                       bf_alloced;           /*  0xb6   0x1 */
	u8                         hwtstamp_tx_type;     /*  0xb7   0x1 */
	u8 *                       bounce_buf;           /*  0xb8   0x8 */
	/* --- cacheline 3 boundary (192 bytes) --- */
	long unsigned int          queue_stopped;        /*  0xc0   0x8 */
	struct mlx4_hwq_resources  sp_wqres;             /*  0xc8  0x58 */
	/* --- cacheline 4 boundary (256 bytes) was 32 bytes ago --- */
	struct mlx4_qp             sp_qp;                /* 0x120  0x30 */
	/* --- cacheline 5 boundary (320 bytes) was 16 bytes ago --- */
	struct mlx4_qp_context     sp_context;           /* 0x150  0xf8 */
	/* --- cacheline 9 boundary (576 bytes) was 8 bytes ago --- */
	cpumask_t                  sp_affinity_mask;     /* 0x248  0x20 */
	enum mlx4_qp_state         sp_qp_state;          /* 0x268   0x4 */
	u16                        sp_stride;            /* 0x26c   0x2 */
	u16                        sp_cqn;               /* 0x26e   0x2 */

	/* size: 640, cachelines: 10, members: 36 */
	/* sum members: 600, holes: 1, sum holes: 24 */
	/* padding: 16 */
};

Instead of this silly placement :

struct mlx4_en_tx_ring {
	u32                        last_nr_txbb;         /*     0   0x4 */
	u32                        cons;                 /*   0x4   0x4 */
	long unsigned int          wake_queue;           /*   0x8   0x8 */

	/* XXX 48 bytes hole, try to pack */

	/* --- cacheline 1 boundary (64 bytes) --- */
	u32                        prod;                 /*  0x40   0x4 */

	/* XXX 4 bytes hole, try to pack */

	long unsigned int          bytes;                /*  0x48   0x8 */
	long unsigned int          packets;              /*  0x50   0x8 */
	long unsigned int          tx_csum;              /*  0x58   0x8 */
	long unsigned int          tso_packets;          /*  0x60   0x8 */
	long unsigned int          xmit_more;            /*  0x68   0x8 */
	unsigned int               tx_dropped;           /*  0x70   0x4 */

	/* XXX 4 bytes hole, try to pack */

	struct mlx4_bf             bf;                   /*  0x78  0x18 */
	/* --- cacheline 2 boundary (128 bytes) was 16 bytes ago --- */
	long unsigned int          queue_stopped;        /*  0x90   0x8 */
	cpumask_t                  affinity_mask;        /*  0x98  0x10 */
	struct mlx4_qp             qp;                   /*  0xa8  0x30 */
	/* --- cacheline 3 boundary (192 bytes) was 24 bytes ago --- */
	struct mlx4_hwq_resources  wqres;                /*  0xd8  0x58 */
	/* --- cacheline 4 boundary (256 bytes) was 48 bytes ago --- */
	u32                        size;                 /* 0x130   0x4 */
	u32                        size_mask;            /* 0x134   0x4 */
	u16                        stride;               /* 0x138   0x2 */

	/* XXX 2 bytes hole, try to pack */

	u32                        full_size;            /* 0x13c   0x4 */
	/* --- cacheline 5 boundary (320 bytes) --- */
	u16                        cqn;                  /* 0x140   0x2 */

	/* XXX 2 bytes hole, try to pack */

	u32                        buf_size;             /* 0x144   0x4 */
	__be32                     doorbell_qpn;         /* 0x148   0x4 */
	__be32                     mr_key;               /* 0x14c   0x4 */
	void *                     buf;                  /* 0x150   0x8 */
	struct mlx4_en_tx_info *   tx_info;              /* 0x158   0x8 */
	struct mlx4_en_rx_ring *   recycle_ring;         /* 0x160   0x8 */
	u32                        (*free_tx_desc)(struct mlx4_en_priv *, struct mlx4_en_tx_ring *, int, u8, u64, int); /* 0x168   0x8 */
	u8 *                       bounce_buf;           /* 0x170   0x8 */
	struct mlx4_qp_context     context;              /* 0x178  0xf8 */
	/* --- cacheline 9 boundary (576 bytes) was 48 bytes ago --- */
	int                        qpn;                  /* 0x270   0x4 */
	enum mlx4_qp_state         qp_state;             /* 0x274   0x4 */
	u8                         queue_index;          /* 0x278   0x1 */
	bool                       bf_enabled;           /* 0x279   0x1 */
	bool                       bf_alloced;           /* 0x27a   0x1 */

	/* XXX 5 bytes hole, try to pack */

	/* --- cacheline 10 boundary (640 bytes) --- */
	struct netdev_queue *      tx_queue;             /* 0x280   0x8 */
	int                        hwtstamp_tx_type;     /* 0x288   0x4 */

	/* size: 704, cachelines: 11, members: 36 */
	/* sum members: 587, holes: 6, sum holes: 65 */
	/* padding: 52 */
};
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      e3f42f84
  21. 03 11月, 2016 3 次提交
  23. 12 9月, 2016 1 次提交
  25. 07 9月, 2016 1 次提交
    • B
      net/mlx4_en: protect ring->xdp_prog with rcu_read_lock · 326fe02d
      由 Brenden Blanco 提交于 
      Depending on the preempt mode, the bpf_prog stored in xdp_prog may be
freed despite the use of call_rcu inside bpf_prog_put. The situation is
possible when running in PREEMPT_RCU=y mode, for instance, since the rcu
callback for destroying the bpf prog can run even during the bh handling
in the mlx4 rx path.

Several options were considered before this patch was settled on:

Add a napi_synchronize loop in mlx4_xdp_set, which would occur after all
of the rings are updated with the new program.
This approach has the disadvantage that as the number of rings
increases, the speed of update will slow down significantly due to
napi_synchronize's msleep(1).

Add a new rcu_head in bpf_prog_aux, to be used by a new bpf_prog_put_bh.
The action of the bpf_prog_put_bh would be to then call bpf_prog_put
later. Those drivers that consume a bpf prog in a bh context (like mlx4)
would then use the bpf_prog_put_bh instead when the ring is up. This has
the problem of complexity, in maintaining proper refcnts and rcu lists,
and would likely be harder to review. In addition, this approach to
freeing must be exclusive with other frees of the bpf prog, for instance
a _bh prog must not be referenced from a prog array that is consumed by
a non-_bh prog.

The placement of rcu_read_lock in this patch is functionally the same as
putting an rcu_read_lock in napi_poll. Actually doing so could be a
potentially controversial change, but would bring the implementation in
line with sk_busy_loop (though of course the nature of those two paths
is substantially different), and would also avoid future copy/paste
problems with future supporters of XDP. Still, this patch does not take
that opinionated option.

Testing was done with kernels in either PREEMPT_RCU=y or
CONFIG_PREEMPT_VOLUNTARY=y+PREEMPT_RCU=n modes, with neither exhibiting
any drawback. With PREEMPT_RCU=n, the extra call to rcu_read_lock did
not show up in the perf report whatsoever, and with PREEMPT_RCU=y the
overhead of rcu_read_lock (according to perf) was the same before/after.
In the rx path, rcu_read_lock is eventually called for every packet
from netif_receive_skb_internal, so the napi poll call's rcu_read_lock
is easily amortized.

v2:
Remove extra rcu_read_lock in mlx4_en_process_rx_cq body
Annotate xdp_prog with __rcu, and convert all usages to rcu_assign or
rcu_dereference[_protected] as appropriate.
Add explicit mutex lock around rcu_assign instead of xchg loop.

Fixes: d576acf0 ("net/mlx4_en: add page recycle to prepare rx ring for tx support")
Acked-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <alexei.starovoitov@gmail.com>
Signed-off-by: NBrenden Blanco <bblanco@plumgrid.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      326fe02d
  27. 20 7月, 2016 4 次提交
    • B
      net/mlx4_en: add xdp forwarding and data write support · 9ecc2d86
      由 Brenden Blanco 提交于 
      A user will now be able to loop packets back out of the same port using
a bpf program attached to xdp hook. Updates to the packet contents from
the bpf program is also supported.

For the packet write feature to work, the rx buffers are now mapped as
bidirectional when the page is allocated. This occurs only when the xdp
hook is active.

When the program returns a TX action, enqueue the packet directly to a
dedicated tx ring, so as to avoid completely any locking. This requires
the tx ring to be allocated 1:1 for each rx ring, as well as the tx
completion running in the same softirq.

Upon tx completion, this dedicated tx ring recycles pages without
unmapping directly back to the original rx ring. In steady state tx/drop
workload, effectively 0 page allocs/frees will occur.

In order to separate out the paths between free and recycle, a
free_tx_desc func pointer is introduced that is optionally updated
whenever recycle_ring is activated. By default the original free
function is always initialized.
Signed-off-by: NBrenden Blanco <bblanco@plumgrid.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      9ecc2d86
    • B
      net/mlx4_en: add page recycle to prepare rx ring for tx support · d576acf0
      由 Brenden Blanco 提交于 
      The mlx4 driver by default allocates order-3 pages for the ring to
consume in multiple fragments. When the device has an xdp program, this
behavior will prevent tx actions since the page must be re-mapped in
TODEVICE mode, which cannot be done if the page is still shared.

Start by making the allocator configurable based on whether xdp is
running, such that order-0 pages are always used and never shared.

Since this will stress the page allocator, add a simple page cache to
each rx ring. Pages in the cache are left dma-mapped, and in drop-only
stress tests the page allocator is eliminated from the perf report.

Note that setting an xdp program will now require the rings to be
reconfigured.

Before:
 26.91%  ksoftirqd/0  [mlx4_en]         [k] mlx4_en_process_rx_cq
 17.88%  ksoftirqd/0  [mlx4_en]         [k] mlx4_en_alloc_frags
  6.00%  ksoftirqd/0  [mlx4_en]         [k] mlx4_en_free_frag
  4.49%  ksoftirqd/0  [kernel.vmlinux]  [k] get_page_from_freelist
  3.21%  swapper      [kernel.vmlinux]  [k] intel_idle
  2.73%  ksoftirqd/0  [kernel.vmlinux]  [k] bpf_map_lookup_elem
  2.57%  swapper      [mlx4_en]         [k] mlx4_en_process_rx_cq

After:
 31.72%  swapper      [kernel.vmlinux]       [k] intel_idle
  8.79%  swapper      [mlx4_en]              [k] mlx4_en_process_rx_cq
  7.54%  swapper      [kernel.vmlinux]       [k] poll_idle
  6.36%  swapper      [mlx4_core]            [k] mlx4_eq_int
  4.21%  swapper      [kernel.vmlinux]       [k] tasklet_action
  4.03%  swapper      [kernel.vmlinux]       [k] cpuidle_enter_state
  3.43%  swapper      [mlx4_en]              [k] mlx4_en_prepare_rx_desc
  2.18%  swapper      [kernel.vmlinux]       [k] native_irq_return_iret
  1.37%  swapper      [kernel.vmlinux]       [k] menu_select
  1.09%  swapper      [kernel.vmlinux]       [k] bpf_map_lookup_elem
Signed-off-by: NBrenden Blanco <bblanco@plumgrid.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d576acf0
    • B
      net/mlx4_en: add support for fast rx drop bpf program · 47a38e15
      由 Brenden Blanco 提交于 
      Add support for the BPF_PROG_TYPE_XDP hook in mlx4 driver.

In tc/socket bpf programs, helpers linearize skb fragments as needed
when the program touches the packet data. However, in the pursuit of
speed, XDP programs will not be allowed to use these slower functions,
especially if it involves allocating an skb.

Therefore, disallow MTU settings that would produce a multi-fragment
packet that XDP programs would fail to access. Future enhancements could
be done to increase the allowable MTU.

The xdp program is present as a per-ring data structure, but as of yet
it is not possible to set at that granularity through any ndo.
Signed-off-by: NBrenden Blanco <bblanco@plumgrid.com>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      47a38e15
    • E
      net/mlx4_en: Add resilience in low memory systems · ec25bc04
      由 Eugenia Emantayev 提交于 
      This patch fixes the lost of Ethernet port on low memory system,
when driver frees its resources and fails to allocate new resources.
Issue could happen while changing number of channels, rings size or
changing the timestamp configuration.
This fix is necessary because of removing vmap use in the code.
When vmap was in use driver could allocate non-contiguous memory
and make it contiguous with vmap. Now it could fail to allocate
a large chunk of contiguous memory and lose the port.
Current code tries to allocate new resources and then upon success
frees the old resources.

Fixes: 73898db0 ('net/mlx4: Avoid wrong virtual mappings')
Signed-off-by: NEugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      ec25bc04
  29. 24 6月, 2016 1 次提交
  31. 18 6月, 2016 1 次提交
  33. 26 5月, 2016 3 次提交
  35. 06 5月, 2016 1 次提交
    • H
      net/mlx4: Avoid wrong virtual mappings · 73898db0
      由 Haggai Abramovsky 提交于 
      The dma_alloc_coherent() function returns a virtual address which can
be used for coherent access to the underlying memory.  On some
architectures, like arm64, undefined behavior results if this memory is
also accessed via virtual mappings that are not coherent.  Because of
their undefined nature, operations like virt_to_page() return garbage
when passed virtual addresses obtained from dma_alloc_coherent().  Any
subsequent mappings via vmap() of the garbage page values are unusable
and result in bad things like bus errors (synchronous aborts in ARM64
speak).

The mlx4 driver contains code that does the equivalent of:
vmap(virt_to_page(dma_alloc_coherent)), this results in an OOPs when the
device is opened.

Prevent Ethernet driver to run this problematic code by forcing it to
allocate contiguous memory. As for the Infiniband driver, at first we
are trying to allocate contiguous memory, but in case of failure roll
back to work with fragmented memory.
Signed-off-by: NHaggai Abramovsky <hagaya@mellanox.com>
Signed-off-by: NYishai Hadas <yishaih@mellanox.com>
Reported-by: NDavid Daney <david.daney@cavium.com>
Tested-by: NSinan Kaya <okaya@codeaurora.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      73898db0
  37. 22 4月, 2016 1 次提交
  39. 26 2月, 2016 1 次提交
  41. 19 11月, 2015 1 次提交
    • E
      mlx4: remove mlx4_en_low_latency_recv() · 868fdb06
      由 Eric Dumazet 提交于 
      Busy polling can now be handled in generic NAPI poll infrastructure.
This removes complexity and fast path overhead :

mlx4 used two spin_lock()/spin_unlock() pair per napi->poll() call
in mlx4_en_cq_lock_napi()/mlx4_en_cq_unlock_napi()

Tested:

Without busy polling :

lpaa23:~# echo 0 >/proc/sys/net/core/busy_read
lpaa24:~# echo 0 >/proc/sys/net/core/busy_read
lpaa23:~# ./netperf -H lpaa24 -t TCP_RR
MIGRATED TCP REQUEST/RESPONSE TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lpaa24.prod.google.com () port 0 AF_INET : first burst 0
Local /Remote
Socket Size   Request  Resp.   Elapsed  Trans.
Send   Recv   Size     Size    Time     Rate
bytes  Bytes  bytes    bytes   secs.    per sec

16384  87380  1        1       10.00    47330.78

With busy polling :

lpaa23:~# echo 70 >/proc/sys/net/core/busy_read
lpaa24:~# echo 70 >/proc/sys/net/core/busy_read
lpaa23:~# ./netperf -H lpaa24 -t TCP_RR
MIGRATED TCP REQUEST/RESPONSE TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lpaa24.prod.google.com () port 0 AF_INET : first burst 0
Local /Remote
Socket Size   Request  Resp.   Elapsed  Trans.
Send   Recv   Size     Size    Time     Rate
bytes  Bytes  bytes    bytes   secs.    per sec

16384  87380  1        1       10.00    97643.55
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      868fdb06