[net,v2,0/2] Fix PHY handle no longer parsing

Message

Sit, Michael Wei Hong March 14, 2023, 7:02 a.m. UTC

After the fixed link support was introduced, it is observed that PHY
no longer attach to the MAC properly.

This patch series fixes the issue and maintains fixed-link capability.

Michael Sit Wei Hong (2):
  net: stmmac: fix PHY handle parsing
  net: stmmac: move fixed-link support fixup code

 .../net/ethernet/stmicro/stmmac/dwmac-intel.c | 11 ---------
 .../net/ethernet/stmicro/stmmac/stmmac_main.c | 24 +++++++++++++++++--
 2 files changed, 22 insertions(+), 13 deletions(-)

v2: Initialize fwnode before using the variable

Comments

Russell King (Oracle) March 17, 2023, 8:54 p.m. UTC | #1

On Tue, Mar 14, 2023 at 03:02:06PM +0800, Michael Sit Wei Hong wrote:
> After the fixed link support was introduced, it is observed that PHY
> no longer attach to the MAC properly.

You are aware, I hope, that fixed-link and having a PHY are mutually
exclusive?

In other words, you can't use fixed-link to specify parameters for a
PHY.

Sit, Michael Wei Hong March 21, 2023, 8:37 a.m. UTC | #2

> -----Original Message-----
> From: Russell King <[e-mail]>
> Sent: Saturday, March 18, 2023 4:54 AM
> To: Sit, Michael Wei Hong <[e-mail]>
> Cc: Giuseppe Cavallaro <[e-mail]>; Alexandre
> Torgue <[e-mail]>; Jose Abreu
> <[e-mail]>; David S . Miller
> <[e-mail]>; Eric Dumazet
> <[e-mail]>; Jakub Kicinski <[e-mail]>;
> Paolo Abeni <[e-mail]>; Maxime Coquelin
> <[e-mail]>; Ong, Boon Leong
> <[e-mail]>; [e-mail]; linux-
> [e-mail]; linux-arm-
> [e-mail]; [e-mail]; Looi,
> Hong Aun <[e-mail]>; Voon, Weifeng
> <[e-mail]>; Lai, Peter Jun Ann
> <[e-mail]>
> Subject: Re: [PATCH net v2 0/2] Fix PHY handle no longer parsing
> 
> On Tue, Mar 14, 2023 at 03:02:06PM +0800, Michael Sit Wei Hong
> wrote:
> > After the fixed link support was introduced, it is observed that
> PHY
> > no longer attach to the MAC properly.
> 
> You are aware, I hope, that fixed-link and having a PHY are
> mutually exclusive?
> 
> In other words, you can't use fixed-link to specify parameters for
> a PHY.
> 
Yes but when the fixed-link support code was added, all our non
fixed-link devices are not attaching the PHYs to the MACs, hence
the reason for this patch series, to ensure both fixed-link and
non fixed-link scenarios work properly.
> --
> RMK's Patch system:
> https://www.armlinux.org.uk/developer/patches/
> FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at
> last!

Russell King (Oracle) March 21, 2023, 10:22 a.m. UTC | #3

On Tue, Mar 21, 2023 at 08:37:48AM +0000, Sit, Michael Wei Hong wrote:
> 
> 
> > -----Original Message-----
> > From: Russell King <[e-mail]>
> > Sent: Saturday, March 18, 2023 4:54 AM
> > To: Sit, Michael Wei Hong <[e-mail]>
> > Cc: Giuseppe Cavallaro <[e-mail]>; Alexandre
> > Torgue <[e-mail]>; Jose Abreu
> > <[e-mail]>; David S . Miller
> > <[e-mail]>; Eric Dumazet
> > <[e-mail]>; Jakub Kicinski <[e-mail]>;
> > Paolo Abeni <[e-mail]>; Maxime Coquelin
> > <[e-mail]>; Ong, Boon Leong
> > <[e-mail]>; [e-mail]; linux-
> > [e-mail]; linux-arm-
> > [e-mail]; [e-mail]; Looi,
> > Hong Aun <[e-mail]>; Voon, Weifeng
> > <[e-mail]>; Lai, Peter Jun Ann
> > <[e-mail]>
> > Subject: Re: [PATCH net v2 0/2] Fix PHY handle no longer parsing
> > 
> > On Tue, Mar 14, 2023 at 03:02:06PM +0800, Michael Sit Wei Hong
> > wrote:
> > > After the fixed link support was introduced, it is observed that
> > PHY
> > > no longer attach to the MAC properly.
> > 
> > You are aware, I hope, that fixed-link and having a PHY are
> > mutually exclusive?
> > 
> > In other words, you can't use fixed-link to specify parameters for
> > a PHY.
> > 
> Yes but when the fixed-link support code was added, all our non
> fixed-link devices are not attaching the PHYs to the MACs, hence
> the reason for this patch series, to ensure both fixed-link and
> non fixed-link scenarios work properly.

Ah, it's the "there should be a PHY here" case you're trying to cater
for, rather than actually a fixed-link.

So, how about adding a helper that is basically:

bool phylink_expects_phy(struct phylink *phylink)
{
	return pl->cfg_link_an_mode == MLO_AN_PHY;
}
EXPORT_SYMBOL_GPL(phylink_expects_phy);

and use that to test whether you need to manually find a PHY to attach,
rather than trying to detect whether a fixed-link stanza appears in
firmware?

Thanks.