[srslte-users] what unit will "get_required_prb_dl" return PRB or RBG

[Francisco Paisana]

Hey,

Thanks for noticing this. We confirm the existing bug in the conversions
between RBGs and PRBs. We will try to fix this for the next release.

Regards,
Francisco

From: Kewin Rausch <krausch at fbk.eu>
> Date: mar., 24 abr. 2018 9:06
> Subject: Re: [srslte-users] what unit will "get_required_prb_dl" return
> PRB or RBG
> To: Zhouyou Gu <guzhouyou at gmail.com>
> Cc: srslte-users <srslte-users at lists.softwareradiosystems.com>
>
> Hey
>

what will "sched_ue::get_required_prb_dl" return, the number of physical
> RB(PRB) or the number of RB group(RBG).
> From what I see in the code, the PRBs, and not the groups.
>
>
> Yes, It seems like that. Even worse, such value (range 0 - 100) is used in
> new_allocation(pending_rb, &newtx_mask) to allocate a mask of PRG.
> Now, inside this procedure a loop iterate on an array of maximum MAX_RBG
> (25) elements. Luckyly enough the loop has 2 checks, and the i < total_rb
> saves the day and avoid to have stack corruption, but still the returned
> mask is inappropriate, probably.
> In the normal RR scheduler of that metric interface probably the problem
> does not arise since the whole subrame is given to a single UE.
> Cheers,
> Kewin R.
> ​
> On Sat, Apr 21, 2018 at 7:09 AM, Zhouyou Gu <guzhouyou at gmail.com> wrote:
> Hello all
> what will "sched_ue::get_required_prb_dl" return, the number of physical
> RB(PRB) or the number of RB group(RBG).
> If it is PRB, there is no conversion between PRB and RBG in the
> dl_metric_rr::get_required_rbg, and it looks like a bug.
> If it is RBG, please give some hit on where the PRB to RBG
> conversion happens in "sched_ue::get_required_prb_dl".
> Thanks a lot
> Yours
> Charles
>
>
>
> _______________________________________________
> srslte-users mailing list
> srslte-users at lists.softwareradiosystems.com
> http://www.softwareradiosystems.com/mailman/listinfo/srslte-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.srsran.com/pipermail/srsran-users/attachments/20180424/7d1f5922/attachment.htm>