[PATCH net] net: stmmac: dwmac-socfpga: Don't access SGMII adapter when not available

Maxime Chevallier maxime.chevallier at bootlin.com
Tue Nov 28 02:20:37 PST 2023


Hello Thomas,

On Tue, 28 Nov 2023 10:18:41 +0100
Thomas Petazzoni <thomas.petazzoni at bootlin.com> wrote:

> On Tue, 28 Nov 2023 10:45:37 +0100
> Maxime Chevallier <maxime.chevallier at bootlin.com> wrote:
> 
> > The SGMII adapter isn't present on all dwmac-socfpga implementations.
> > Make sure we don't try to configure it if we don't have this adapter.
> > 
> > Fixes: 5d1f3fe7d2d5 ("net: stmmac: dwmac-sogfpga: use the lynx pcs driver")
> > Signed-off-by: Maxime Chevallier <maxime.chevallier at bootlin.com>
> > ---
> >  drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> > index ba2ce776bd4d..ae120792e1b6 100644
> > --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> > +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> > @@ -243,7 +243,8 @@ static void socfpga_sgmii_config(struct socfpga_dwmac *dwmac, bool enable)
> >  {
> >  	u16 val = enable ? SGMII_ADAPTER_ENABLE : SGMII_ADAPTER_DISABLE;
> >  
> > -	writew(val, dwmac->sgmii_adapter_base + SGMII_ADAPTER_CTRL_REG);
> > +	if (dwmac->sgmii_adapter_base)
> > +		writew(val, dwmac->sgmii_adapter_base + SGMII_ADAPTER_CTRL_REG);
> >  }
> >  
> >  static int socfpga_set_phy_mode_common(int phymode, u32 *val)  
> 
> Reviewed-by: Thomas Petazzoni <thomas.petazzoni at bootlin.com>
> 
> As a follow-up improvement, there's an open-coded version of
> socfpga_sgmii_config() in socfpga_dwmac_fix_mac_speed(), which could be
> rewritten as such:
> 
> 	socfpga_sgmii_config(dwmac, false);
>
> 	if (splitter_base) {
> 		val = readl(splitter_base + EMAC_SPLITTER_CTRL_REG);
> 		val &= ~EMAC_SPLITTER_CTRL_SPEED_MASK;
 [...]

I did saw this, but as this is merely a non-functional rework, I'd
like to target this to net-next, so I'll wait for the fix to land and
follow-up with this rework indeed.

Thanks for the review,

Maxime



More information about the linux-arm-kernel mailing list