> -----Original Message----- > From: Stanislaw Gruszka [mailto:sgruszka@xxxxxxxxxx] > Sent: Monday, October 08, 2018 5:19 PM > To: Kalle Valo > Cc: Tony Chuang; Larry.Finger@xxxxxxxxxxxx; Pkshih; Andy Huang; > linux-wireless@xxxxxxxxxxxxxxx > Subject: Re: [RFC v2 04/12] rtw88: trx files > > On Sat, Oct 06, 2018 at 03:32:11PM +0300, Kalle Valo wrote: > > Tony Chuang <yhchuang@xxxxxxxxxxx> writes: > > > > >> > + pkt_info->bmc = bmc; > > >> > + pkt_info->sec_type = sec_type; > > >> > + pkt_info->tx_pkt_size = skb->len; > > >> > + pkt_info->offset = chip->tx_pkt_desc_sz; > > >> > + pkt_info->qsel = skb->priority; > > >> > > >> Shouldn't be qsel somehow mapped from skb->priority ? > > > > > > Firmware handles it. > > > > What if meaning of skb->priority changes in the future? I don't think > > it's safe to provide kernel internal values to firmware, you should do > > some kind of mapping in the driver. > > We already have some functions that do mapping: > rtw_hw_queue_mapping(struct sk_buff *skb) > and > rtw_pci_get_tx_qsel(struct sk_buff *skb, u8 queue) > Preasumly they should be used here as well. > > Stanislaw > Stanislaw is right, PCI will get the qsel again and overwrite it. Maybe we should remove the extra qsel code in tx.c BTW, skb->priority is mapped to TID 0-15 in mac80211, and that's what we want to provide to the firmware. Yan-Hsuan Chuang
