blob: a194a88c65008e51c640b48f8ab435cb0f5915b1 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
|
VRLP and VITA Fields
====================
All VRT (VITA 49.0) packets will be carried in a VRL (VITA 49.1) frame when outside the FPGA core.
Every VRL frame will hold exactly one VRT packet.
VRT packets must have stream ID
VRT packets must NOT have integer seconds timestamps
VRT packets must NOT have class ID
VRT packets may have fractional seconds timestamps
VRT packets may have trailer
Only VITA IF Data and VITA Extension Context packets are allowed
VITA Compressed Headers
=======================
Structure of a normal VRL-encapsulated VRT packet, in 64-bit wide format:
VRLP { VRLP_text[31:0], frame_count[11:0], frame_size[19:0] }
VRT0 { VRT_Header[31:0], Stream ID[31:0] }
VRT1 { Fractional timestamp[63:0] } OPTIONAL
... Packet contents ...
Optional Trailer
"VEND"
Since packet contents can have variable length, 64-bit alignment of trailer and VEND will vary
Inside the FPGA switching fabric, the first line of "compressed vita packets" will be called the "CHDR".
The "CHDR" compresses the contents of the VRLP header and VRT header and streamID, and has the following format:
BIT Meaning
63 is_extension_context (0 for data, 1 for control packets)
62 reserved
61 has_time (0 for no, 1 for time field on next line)
60 EOB (end of burst indicator)
59:48 12-bit sequence number (from VRLP frame count, bottom 4 bits should match VRT seq number)
47:32 16-bit length in bytes (from VRT length in words32)
31:0 stream ID
Inside the fabric, the "VEND" from VRLP is dropped.
|
