VPP typically works best if configured with a relatively small number of
buffers. On the other hand, nsim can induce a large amount of delay, so the
buffers cannot be used to temporarily store the data.
Florin
> On Jan 29, 2019, at 3:19 AM, Raj <rajlistu...@gmail.com> wrote:
>
> Hello all,
>
> While I was browsing through the nsim code (plugins/nsim),
>
> in nsim_inline() we are copying the packet data to ep->data like below
>
> clib_memcpy (ep->data, vlib_buffer_get_current (b[0]),
> ep->current_length);
>
> and in nsim_input.c the same packet is transferring to TX
>
> clib_memcpy (b0->data, ep->data, ep->current_length)
>
> The data is defined in the below structure
> typedef struct
> {
> f64 tx_time;
> u32 tx_sw_if_index;
> u32 current_length;
> CLIB_CACHE_LINE_ALIGN_MARK (pad);
> u8 data[WHEEL_ENTRY_DATA_SIZE];
> } nsim_wheel_entry_t;
>
> I am just wondering if there is any special reason why 'data' as such
> is copied here from the vlib buffer to the wheel buffer, instead of
> using the pointer(copying just the address), and thus avoiding a
> packet copy?
>
> Is it not possible to keep the packet in vlib buffer itself till it
> is transmitted NIC/DPDK TX after a delay?
>
> Thanks and Regards,
>
> Raj
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
>
> View/Reply Online (#12040): https://lists.fd.io/g/vpp-dev/message/12040
> Mute This Topic: https://lists.fd.io/mt/29581029/675152
> Group Owner: vpp-dev+ow...@lists.fd.io
> Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [fcoras.li...@gmail.com]
> -=-=-=-=-=-=-=-=-=-=-=-
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#12048): https://lists.fd.io/g/vpp-dev/message/12048
Mute This Topic: https://lists.fd.io/mt/29581029/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
