[dpdk-dev] [PATCH 27/29] net/ixgbe/base: add write flush required by Inphi

[Dai, Wei]

> -----Original Message-----
> From: Yigit, Ferruh
> Sent: Tuesday, December 6, 2016 3:41 AM
> To: Dai, Wei <wei.dai at intel.com>; Zhang, Helin <helin.zhang at intel.com>;
> Ananyev, Konstantin <konstantin.ananyev at intel.com>
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 27/29] net/ixgbe/base: add write flush required
> by Inphi
> 
> On 12/4/2016 6:31 AM, Wei Dai wrote:
> > This patch updates Inphi configuration to flush the register write
> > with
> 
> Do we really need to mention from Inphi here? If so, can you please explain
> what it is?
Inphi (www.inphi.com) is a company which provides PHYs.
So I will use "Inphi PHY" instead of "Inphi" in v2 patch set.

> 
> > a reg read. Inphi is configured in ixgbe_setup_mac_link_sfp_x550a.
> > The Inphy setup flow has been updated to read configuration reg, write
> > only linear/non-linear, and then read (write flush).
> 
> Also patch does [1] seems not mentioned in the commit log, can you please add
> information for it?
Yes, following statement is redundant, but in order to simplify the process to 
keep up with the shared code provided by another team (Intel Network Division), 
I'd like to keep it here. Anyway it is harmless.

> 
> [1]
> > +		reg_phy_ext &= ~((IXGBE_CS4227_EDC_MODE_CX1 << 1) |
> > +				 (IXGBE_CS4227_EDC_MODE_SR << 1));
> 
> >
> > Signed-off-by: Wei Dai <wei.dai at intel.com>
> > ---
> >  drivers/net/ixgbe/base/ixgbe_x550.c | 14 ++++++++++++++
> >  1 file changed, 14 insertions(+)
> >
> > diff --git a/drivers/net/ixgbe/base/ixgbe_x550.c
> > b/drivers/net/ixgbe/base/ixgbe_x550.c
> > index 4a98530..a57ba74 100644
> > --- a/drivers/net/ixgbe/base/ixgbe_x550.c
> > +++ b/drivers/net/ixgbe/base/ixgbe_x550.c
> > @@ -2834,12 +2834,26 @@ s32 ixgbe_setup_mac_link_sfp_x550a(struct
> > ixgbe_hw *hw,
> >
> >  		/* Configure CS4227/CS4223 LINE side to proper mode. */
> >  		reg_slice = IXGBE_CS4227_LINE_SPARE24_LSB + slice_offset;
> > +
> > +		ret_val = hw->phy.ops.read_reg(hw, reg_slice,
> > +					IXGBE_MDIO_ZERO_DEV_TYPE, &reg_phy_ext);
> > +
> > +		if (ret_val != IXGBE_SUCCESS)
> > +			return ret_val;
> > +
> > +		reg_phy_ext &= ~((IXGBE_CS4227_EDC_MODE_CX1 << 1) |
> > +				 (IXGBE_CS4227_EDC_MODE_SR << 1));
> > +
> >  		if (setup_linear)
> >  			reg_phy_ext = (IXGBE_CS4227_EDC_MODE_CX1 << 1) | 0x1;
> >  		else
> >  			reg_phy_ext = (IXGBE_CS4227_EDC_MODE_SR << 1) | 0x1;
> >  		ret_val = hw->phy.ops.write_reg(hw, reg_slice,
> >  					 IXGBE_MDIO_ZERO_DEV_TYPE, reg_phy_ext);
> > +
> > +		/* Flush previous write with a read */
> > +		ret_val = hw->phy.ops.read_reg(hw, reg_slice,
> > +					IXGBE_MDIO_ZERO_DEV_TYPE, &reg_phy_ext);
> >  	}
> >  	return ret_val;
> >  }
> >