[openib-general] [PATCH] [MAD] changes to ib_create_send_mad

[Hal Rosenstock]

On Thu, 2005-05-05 at 14:48, Sean Hefty wrote:
> Sean Hefty wrote:
> > Hal Rosenstock wrote:
> > 
> >> On PayloadLength, single segment sends are fine. It is multisegment
> >> sends which seem wrong to me.
> >>
> >> Case 1: ib_create_send_mad with hdr_len 0x38 data_len 0x278
> >>         paylen_newwin stored in header is 0x28C which seems correct
> >>         this creates 4 segments
> >>         1-3 segments paylen_newwin is 0x6E0, 4th segment is 0x34
> >>
> >> Case 2: ib_create_send_mad with hdr_len 0x38 data_len 0x620
> >>         paylen_newwin stored in header is 0x634 which seems correct
> >>         this creates 8 segments
> >>         1-7 segments paylen_newwin is 0x6E0, 4th segment is 0xBC
> >>
> >> Last paylen_newwin in both cases appears to me to be correct but the 
> >> paylen_newwin in the 1-n segments (0x6E0) seems wrong to me.
> 
> Err... I spoke too soon.  Here's what I see on both the send and receive 
> side:  Using SA class,
> 
> hdr_len = 56, data_len = 632 (x278)
> 1st payload: 0x370, last: 0x34
> 
> hdr_len = 56, data_len = 1568 (x620)
> 1st payload: 0x6e0, last: 0xbc
> 
> The middle segments in both cases equal the 1st payload, which is what I 
> would expect.  

Me too. (They should be ignored anyhow).

> The payload values in both cases are correct.

Let me look again. I don't think I was hallucinating...

-- Hal