> > Hi Igor > > I think all other implementations return -EINVAL or -E2BIG or similar > when the value is not supported. > > Also, given that a u8 is being passed, is cfg->downshift_counter > 255 > possible? I'm not even sure 255 makes any sense. Autoneg takes around > 1.5s, maybe longer. Do you really want to wait 255 * 1.5 seconds > before downshifting? Even 15*1.5 seems a long time. Hi Andrew, here I'm just blindly follow the value limits of firmware interface (two device revisions have different counter field width here). To make behavior consistent you are right, we probably can leave 15 max and return EINVAL otherwise. Igor
- [PATCH net-next 0/3] net: atlantic: phy tunables from mac d... Igor Russkikh
- [PATCH net-next 1/3] ethtool: allow netdev driver to d... Igor Russkikh
- [PATCH net-next 2/3] net: atlantic: implement phy down... Igor Russkikh
- Re: [PATCH net-next 2/3] net: atlantic: implement ... Andrew Lunn
- Re: [PATCH net-next 2/3] net: atlantic: implem... Igor Russkikh
- [PATCH net-next 3/3] net: atlantic: implement media de... Igor Russkikh
- Re: [PATCH net-next 3/3] net: atlantic: implement ... Andrew Lunn
- Re: [PATCH net-next 3/3] net: atlantic: implem... Igor Russkikh
- Re: [PATCH net-next 3/3] net: atlantic: im... Andrew Lunn
- Re: [EXT] Re: [PATCH net-next 3/3] ne... Igor Russkikh
- Re: [EXT] Re: [PATCH net-next 3/3... Andrew Lunn
- Re: [PATCH net-next 0/3] net: atlantic: phy tunables f... Andrew Lunn
- Re: [PATCH net-next 0/3] net: atlantic: phy tunabl... Jakub Kicinski
- Re: [PATCH net-next 0/3] net: atlantic: phy tu... Andrew Lunn
- Re: [PATCH net-next 0/3] net: atlantic: ph... Jakub Kicinski