Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
1 年多 前同步成功

通知 160
Star 36
Fork 7
cloud-kernel
cloud-kernel
切换分支/标签
  1. 03 6月, 2018 25 次提交
  3. 02 6月, 2018 15 次提交
    • A
      Revert "vfio/type1: Improve memory pinning process for raw PFN mapping" · 89c29def
      由 Alex Williamson 提交于 
      Bisection by Amadeusz Sławiński implicates this commit leading to bad
page state issues after VM shutdown, likely due to unbalanced page
references.  The original commit was intended only as a performance
improvement, therefore revert for offline rework.

Link: https://lkml.org/lkml/2018/6/2/97
Fixes: 356e88eb ("vfio/type1: Improve memory pinning process for raw PFN mapping")
Cc: Jason Cai (Xiang Feng) <jason.cai@linux.alibaba.com>
Reported-by: NAmadeusz Sławiński <amade@asmblr.net>
Signed-off-by: NAlex Williamson <alex.williamson@redhat.com>
      89c29def
    • D
      Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next · 1ffdd8e1
      由 David S. Miller 提交于 
      Pablo Neira Ayuso says:

====================
Netfilter/IPVS updates for net-next

The following patchset contains Netfilter/IPVS updates for your net-next
tree, the most relevant things in this batch are:

1) Compile masquerade infrastructure into NAT module, from Florian Westphal.
   Same thing with the redirection support.

2) Abort transaction if early initialization of the commit phase fails.
   Also from Florian.

3) Get rid of synchronize_rcu() by using rule array in nf_tables, from
   Florian.

4) Abort nf_tables batch if fatal signal is pending, from Florian.

5) Use .call_rcu nfnetlink from nf_tables to make dumps fully lockless.
   From Florian Westphal.

6) Support to match transparent sockets from nf_tables, from Máté Eckl.

7) Audit support for nf_tables, from Phil Sutter.

8) Validate chain dependencies from commit phase, fall back to fine grain
   validation only in case of errors.

9) Attach dst to skbuff from netfilter flowtable packet path, from
   Jason A. Donenfeld.

10) Use artificial maximum attribute cap to remove VLA from nfnetlink.
    Patch from Kees Cook.

11) Add extension to allow to forward packets through neighbour layer.

12) Add IPv6 conntrack helper support to IPVS, from Julian Anastasov.

13) Add IPv6 FTP conntrack support to IPVS, from Julian Anastasov.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      1ffdd8e1
    • D
      Merge tag 'mlx5e-updates-2018-06-01' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux · f39c6b29
      由 David S. Miller 提交于 
      Saeed Mahameed says:

====================
mlx5e-updates-2018-06-01

1) From Tariq, Two patches to Fix IPoIB issues introduced in
   "net/mlx5e: TX, Use actual WQE size for SQ edge fill"

2) From Eran, Additional improvements to mlx5e statistics reporting

3) From Maor, Increase aRFS flow tables size

4) From Adi, Support MTU change for ethernet representors

5) From Ilan and Adi, Handle QP error events in FPGA

6) From Tariq, last 10 patches mainly deals with RX buffer scheme improvements for legacy RQ
   to use only order-0 pages and fragmented SKBs for large MTUs.

-  Tariq starts with some refactoring and removing HW LRO support from traditional
   (legacy) RQ, since it complicates the buffer scheme and removing it makes it smoother
   to move to cyclic descriptor buffer for traditional RQ.

- Use cyclic WQ in legacy RQ, which has many benefits and paves the way for fragmented SKBs
  for large MTUs.

- Enhance legacy Receive Queue memory scheme, such that only order-0 pages are used.
  Whenever possible, prefer using a linear SKB, and build it wrapping the WQE buffer.
  Otherwise (for example, jumbo frames on x86), use non-linear SKB, with as many frags
  as needed. In this case, multiple WQE scatter entries are used, up to a maximum of 4
  frags and 10KB of MTU.

- TX statistics access improvements.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      f39c6b29
    • D
      Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf · cd075ce4
      由 David S. Miller 提交于 
      Daniel Borkmann says:

====================
pull-request: bpf 2018-06-02

The following pull-request contains BPF updates for your *net* tree.

The main changes are:

1) BPF uapi fix in struct bpf_prog_info and struct bpf_map_info in
   order to fix offsets on 32 bit archs.

This will have a minor merge conflict with net-next which has the
__u32 gpl_compatible:1 bitfield in struct bpf_prog_info at this
location. Resolution is to use the gpl_compatible member.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      cd075ce4
    • D
      bpf: fix uapi hole for 32 bit compat applications · 36f9814a
      由 Daniel Borkmann 提交于 
      In 64 bit, we have a 4 byte hole between ifindex and netns_dev in the
case of struct bpf_map_info but also struct bpf_prog_info. In net-next
commit b85fab0e ("bpf: Add gpl_compatible flag to struct bpf_prog_info")
added a bitfield into it to expose some flags related to programs. Thus,
add an unnamed __u32 bitfield for both so that alignment keeps the same
in both 32 and 64 bit cases, and can be naturally extended from there
as in b85fab0e.

Before:

  # file test.o
  test.o: ELF 32-bit LSB relocatable, Intel 80386, version 1 (SYSV), not stripped
  # pahole test.o
  struct bpf_map_info {
	__u32                      type;                 /*     0     4 */
	__u32                      id;                   /*     4     4 */
	__u32                      key_size;             /*     8     4 */
	__u32                      value_size;           /*    12     4 */
	__u32                      max_entries;          /*    16     4 */
	__u32                      map_flags;            /*    20     4 */
	char                       name[16];             /*    24    16 */
	__u32                      ifindex;              /*    40     4 */
	__u64                      netns_dev;            /*    44     8 */
	__u64                      netns_ino;            /*    52     8 */

	/* size: 64, cachelines: 1, members: 10 */
	/* padding: 4 */
  };

After (same as on 64 bit):

  # file test.o
  test.o: ELF 32-bit LSB relocatable, Intel 80386, version 1 (SYSV), not stripped
  # pahole test.o
  struct bpf_map_info {
	__u32                      type;                 /*     0     4 */
	__u32                      id;                   /*     4     4 */
	__u32                      key_size;             /*     8     4 */
	__u32                      value_size;           /*    12     4 */
	__u32                      max_entries;          /*    16     4 */
	__u32                      map_flags;            /*    20     4 */
	char                       name[16];             /*    24    16 */
	__u32                      ifindex;              /*    40     4 */

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

	__u64                      netns_dev;            /*    48     8 */
	__u64                      netns_ino;            /*    56     8 */
	/* --- cacheline 1 boundary (64 bytes) --- */

	/* size: 64, cachelines: 1, members: 10 */
	/* sum members: 60, holes: 1, sum holes: 4 */
  };
Reported-by: NDmitry V. Levin <ldv@altlinux.org>
Reported-by: NEugene Syromiatnikov <esyr@redhat.com>
Fixes: 52775b33 ("bpf: offload: report device information about offloaded maps")
Fixes: 675fc275 ("bpf: offload: report device information for offloaded programs")
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NAlexei Starovoitov <ast@kernel.org>
      36f9814a
    • T
      net/mlx5e: TX, Separate cachelines of xmit and completion stats · f65a59ff
      由 Tariq Toukan 提交于 
      Avoid false sharing of cachelines by separating the cachelines of
TX stats that are dertied in xmit flow and in completion flow.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      f65a59ff
    • T
      net/mlx5e: RX, Always prefer Linear SKB configuration · 5ffd8194
      由 Tariq Toukan 提交于 
      Prefer the linear SKB configuration of Legacy RQ over the
non-linear one of Striding RQ.

This implies that ConnectX-4 LX now uses legacy RQ by default,
as it does not support the linear configuration of Striding RQ.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      5ffd8194
    • T
      net/mlx5e: RX, Enhance legacy Receive Queue memory scheme · 069d1146
      由 Tariq Toukan 提交于 
      Enhance the memory scheme of the legacy RQ, such that
only order-0 pages are used.

Whenever possible, prefer using a linear SKB, and build it
wrapping the WQE buffer.

Otherwise (for example, jumbo frames on x86), use non-linear SKB,
with as many frags as needed. In this case, multiple WQE
scatter entries are used, up to a maximum of 4 frags and 10KB of MTU.

This implied to remove support of HW LRO in legacy RQ, as it would
require large number of page allocations and scatter entries per WQE
on archs with PAGE_SIZE = 4KB, yielding bad performance.

In earlier patches, we guaranteed that all completions are in-order,
and that we use a cyclic WQ.
This creates an oppurtunity for a performance optimization:
The mapping between a "struct mlx5e_dma_info", and the
WQEs (struct mlx5e_wqe_frag_info) pointing to it, is constant
across different cycles of a WQ. This allows initializing
the mapping in the time of RQ creation, and not handle it
in datapath.

A struct mlx5e_dma_info that is shared between different WQEs
is allocated by the first WQE, and freed by the last one.
This implies an important requirement: WQEs that share the same
struct mlx5e_dma_info must be posted within the same NAPI.
Otherwise, upon completion, struct mlx5e_wqe_frag_info would mistakenly
point to the new struct mlx5e_dma_info, not the one that was posted
(and the HW wrote to).
This bulking requirement is actually good also for performance reasons,
hence we extend the bulk beyong the minimal requirement above.

With this memory scheme, the RQs memory footprint is reduce by a
factor of 2 on x86, and by a factor of 32 on PowerPC.
Same factors apply for the number of pages in a GRO session.

Performance tests:
ConnectX-4, single core, single RX ring, default MTU.

x86:
CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz

Packet rate (early drop in TC): no degradation
TCP streams: ~5% improvement

PowerPC:
CPU: POWER8 (raw), altivec supported

Packet rate (early drop in TC): 20% gain
TCP streams: 25% gain
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      069d1146
    • T
      net/mlx5e: RX, Use cyclic WQ in legacy RQ · 99cbfa93
      由 Tariq Toukan 提交于 
      Now that LRO is not supported for Legacy RQ, there is no source of
out-of-order completions in the WQ, and we can use a cyclic one.
This has multiple advantages:
- reduces the WQE size (smaller PCI transactions).
- lower overhead in datapath (no handling of 'next' pointers).
- no reserved WQE for the WQ head (was need in linked-list).
- allows using a constant map between frag and dma_info struct, in downstream patch.

Performance tests:
ConnectX-4, single core, single RX ring.
Major gain in packet rate of single ring XDP drop.
Bottleneck is shifted form HW (at 16Mpps) to SW (at 20Mpps).
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      99cbfa93
    • T
      net/mlx5e: RX, Split WQ objects for different RQ types · 422d4c40
      由 Tariq Toukan 提交于 
      Replace the common RQ WQ object with two separate ones for the
different RQ types.
This is in preparation for switching to using a cyclic WQ type
in Legacy RQ.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      422d4c40
    • T
      net/mlx5e: RX, Remove HW LRO support in legacy RQ · 6c3a823e
      由 Tariq Toukan 提交于 
      Current LRO implementation in Legacy RQ uses high-order pages.
In downstream patches of this series we complete the transition
to using only order-0 pages in RX datapath (which was already done
in Striding RQ).

Unlike the more advanced Striding RQ, Legacy RQ does not make reuse
of any non-consumed buffers of non-full LRO sessions, and combining
it with order-0 pages has many performance drawbacks.

Hence, here we totally remove LRO support in Legacy RQ.
This guarantees having no out-of-order completions, which allows using
a cyclic work queue (instead of a linked-list) in a downstream patch.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      6c3a823e
    • T
      net/mlx5e: RX, Dedicate a function for copying SKB header · 386471f1
      由 Tariq Toukan 提交于 
      Get the logic of copying the packet header into the SKB linear part
into a generic function. Function does copy length alignment
and dma buffer sync.

It is currently called only within the MPWQE flow.
In a downstream patch, it will be called within the legacy RQ flow
as well.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      386471f1
    • T
      net/mlx5e: RX, Generalise function of SKB frag addition · fa698366
      由 Tariq Toukan 提交于 
      Rename it and pass truesize as an extra argument, as it will be used also
in Legacy RQ in a downstream patch.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      fa698366
    • T
      net/mlx5e: RX, Generalise name of non-linear SKB head size · 75aa889f
      由 Tariq Toukan 提交于 
      Make name more generic by dropping MPWRQ from it, as it will be
used also in Legacy RQ in a downstream patch.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      75aa889f
    • T
      net/mlx5e: TX, Obsolete maintaining local copies of skb->len/data · 5e7d77a9
      由 Tariq Toukan 提交于 
      Instead of maintaining a local copy of skb->len/data and updating
it upon every copy to the WQE inline part, just calculate it once
when needed, using the ihs.

This obsoletes the function mlx5e_tx_skb_pull_inline.
Signed-off-by: NTariq Toukan <tariqt@mellanox.com>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
      5e7d77a9