DBGvpp# show classify tables index 32352
TableIdx Sessions NextTbl NextNode
32352 0 -1 -1
Heap: base 0x7ffed77e1000, size 4k, locked, unmap-on-destroy, name
'classify'
page stats: page-size 4K, total 1, mapped 1, not-mapped 0
numa 0: 1 pages, 4k bytes
total: 3.95K, used: 960, free: 3.02K, trimmabl
If it is multi-producer/multi-consumer from my experience it is still much more
costly than using simple per-thread cache scheme.
VPP buffer pools are using that architecture. Each thread maintains own cache,
and lock happens only when bulk transfer is needed from cache to the global
freelist o
There are few additional modes added to the ring library (a year back) in DPDK
that improve the performance when there are threads on control plane and data
plane doing enqueue/dequeue from the same ring. Are you talking about these or
just the ring in general?
Thanks,
Honnappa
> -Original
I believe it was just ring that they cared about.
-Original Message-
From: Damjan Marion
Sent: Monday, October 25, 2021 11:08 AM
To: bjerem...@gmail.com
Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] Linking DPDK libs to plugins
Ok, i’m affraid that to implement this you will need to i
Hi Neale,
I've tried with simplified version first (I know there's a
_vec_resize_will_expand macro), something like:
always_inline u8
clib_bitmap_will_expand(uword * ai, uword i)
{
uword i0 = i / BITS (ai[0]);
return vec_len(ai) < i0;
}
I've put a check in load_balance_destroy just before pool_put
> On 25.10.2021., at 19:02, Mrityunjay Kumar wrote:
>
> Damjon Hi,
>
> I am users of DPDK since August 2011, and VPP guess it was Jan 2017,
VPP is just open-sourced in 2017. It started back in 2004:
https://patents.google.com/patent/US7961636B1/en
>
> Not sure, about contributor but we
Damjon Hi,
I am users of DPDK since August 2011, and VPP guess it was Jan 2017,
Not sure, about contributor but we should think to take dpdk as main frame
in VPP,
For this, we need to deprecate vlib_buffer_t, marry entire VPP with
rte_mbuf,
This can lead to avoid extra overhead to translate t
Hi,
Yes. SiFive HiFive boards are available. But they do not support Vector
Extension yet.
Also Qemu is ready for RiscV. Ubuntu images are available for RIscV.
Thanks,
Hrishikesh
On Mon, Oct 25, 2021 at 9:56 PM Damjan Marion wrote:
>
>
>
> > On 14.10.2021., at 15:43, Hrishikesh Karanjikar <
>
> On 14.10.2021., at 15:43, Hrishikesh Karanjikar
> wrote:
>
>
> Hi,
>
> Is VPP ported for the Risc-V processor?
> Is there any project going for the same?
>
I was looking at that a year ago but I was not able to find any suitable dev
board.
Is there anything new on the market?
—
Damj
Ok, i’m affraid that to implement this you will need to introduce lot of mess.
At the end probably will be easier to implement that functionality natively.
Which exact implementation of the dpdk mempool you are looking to use (ring,
stack, bucket, ...)?
—
Damjan
> On 25.10.2021., at 17:39,
Hi Damjan,
Thanks for the reply
Here are the details:
1. We want to use only the rte_mempool infrastructure for lockless global
memory pools. We will not be using any mbuf infrastructure from dpdk
2. We want to use this infra across our multiple plugins
3. We want to be able to include rte_mem
tried increasing the memory size of the classify table as per the
suggestion in https://lists.fd.io/g/vpp-dev/topic/72538712#15861 but still
facing the crash
On Mon, Oct 25, 2021 at 3:31 PM Akash S R via lists.fd.io wrote:
> Hey Stanislav,
>
> We have configured heap-size on .conf as 7GB and it
> On 25.10.2021., at 01:13, bjerem...@gmail.com wrote:
>
> Greetings,
>
> Let me preface this by saying that I really do not know much about the CMake
> utility. But I am trying to see if there is a way to make the DPDK libs
> accessible to other plugins (aside from the dpdk plugin) that ar
Hey Stanislav,
We have configured heap-size on .conf as 7GB and it is free.
Below are the prints:
ram memory:
root@5495e5b86e58:/home/Immaculate/Oct12/vpp# free -h
totalusedfree shared buff/cache
available
Mem: 7.6G2.1G139M 21
Hi Akash,
Looks like you've run out of memory. Did you try to increase heap-size in
the configuration file?
On Mon, 25 Oct 2021 at 10:36, Akash S R wrote:
> Hi Mates,
>
> A fix if done to resolve the crash for classify table in vnet_classify.c
>
>
>
> *vnet_classify_new_table*
>
>
>
> vec_valid
Hi Mates,
A fix if done to resolve the crash for classify table in vnet_classify.c
*vnet_classify_new_table*
vec_validate_aligned (t->buckets, nbuckets - 1, CLIB_CACHE_LINE_BYTES);
// oldheap = clib_mem_set_heap (t->mheap);
clib_spinlock_init (&t->writer_lock);
oldheap = clib_m
16 matches
Mail list logo
