| Message ID | 1374238894-24906-1-git-send-email-festevam@gmail.com |
|---|---|
| State | Accepted |
| Delegated to: | Tom Rini |
| Headers | show |
On Fri, Jul 19, 2013 at 9:01 AM, Fabio Estevam <festevam@gmail.com> wrote: > From: Fabio Estevam <fabio.estevam@freescale.com> > > Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the > checking for ESTATUS_1000_XHALF, but it incorrectly sets the > SUPPORTED_1000baseX_Full flag in this case. Signed-off-by: Charles Coldwell <coldwell@gmail.com> -- Charles M. Coldwell, W1CMC Belmont, Massachusetts, New England "Turn on, log in, tune out"
Hi Charles, On Fri, Jul 19, 2013 at 10:59 AM, Charles Coldwell <coldwell@gmail.com> wrote: > On Fri, Jul 19, 2013 at 9:01 AM, Fabio Estevam <festevam@gmail.com> wrote: >> From: Fabio Estevam <fabio.estevam@freescale.com> >> >> Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the >> checking for ESTATUS_1000_XHALF, but it incorrectly sets the >> SUPPORTED_1000baseX_Full flag in this case. > > Signed-off-by: Charles Coldwell <coldwell@gmail.com> I guess you meant Acked-by instead?
On Fri, Jul 19, 2013 at 10:01 AM, Fabio Estevam <festevam@gmail.com> wrote: > Hi Charles, > > On Fri, Jul 19, 2013 at 10:59 AM, Charles Coldwell <coldwell@gmail.com> wrote: >> On Fri, Jul 19, 2013 at 9:01 AM, Fabio Estevam <festevam@gmail.com> wrote: >>> From: Fabio Estevam <fabio.estevam@freescale.com> >>> >>> Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the >>> checking for ESTATUS_1000_XHALF, but it incorrectly sets the >>> SUPPORTED_1000baseX_Full flag in this case. >> >> Signed-off-by: Charles Coldwell <coldwell@gmail.com> > > I guess you meant Acked-by instead? Sigh. Indeed, acked. My LKML skills are rusty. -- Charles M. Coldwell, W1CMC Belmont, Massachusetts, New England "Turn on, log in, tune out"
Fabio Estevam <festevam@gmail.com> writes: > Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the > checking for ESTATUS_1000_XHALF, but it incorrectly sets the > SUPPORTED_1000baseX_Full flag in this case. > > Set the SUPPORTED_1000baseX_Half flag instead. Nice catch. Reviewed-By: Sascha Silbe <t-uboot@infra-silbe.de> No test done as my current test set-up is 100Base-T only. Sascha
On Fri, Jul 19, 2013 at 10:01:34AM -0300, Fabio Estevam wrote: > From: Fabio Estevam <fabio.estevam@freescale.com> > > Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the > checking for ESTATUS_1000_XHALF, but it incorrectly sets the > SUPPORTED_1000baseX_Full flag in this case. > > Set the SUPPORTED_1000baseX_Half flag instead. > > Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com> So, do we need both patches to fix the problem?
On Fri, Jul 19, 2013 at 12:55 PM, Tom Rini <trini@ti.com> wrote: > On Fri, Jul 19, 2013 at 10:01:34AM -0300, Fabio Estevam wrote: > >> From: Fabio Estevam <fabio.estevam@freescale.com> >> >> Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the >> checking for ESTATUS_1000_XHALF, but it incorrectly sets the >> SUPPORTED_1000baseX_Full flag in this case. >> >> Set the SUPPORTED_1000baseX_Half flag instead. >> >> Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com> > > So, do we need both patches to fix the problem? One patch fixes "the" problem; the other fixes another problem we haven't seen yet (but will). -- Charles M. Coldwell, W1CMC Belmont, Massachusetts, New England "Turn on, log in, tune out"
On Fri, Jul 19, 2013 at 8:01 AM, Fabio Estevam <festevam@gmail.com> wrote: > From: Fabio Estevam <fabio.estevam@freescale.com> > > Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the > checking for ESTATUS_1000_XHALF, but it incorrectly sets the > SUPPORTED_1000baseX_Full flag in this case. > > Set the SUPPORTED_1000baseX_Half flag instead. > > Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com> Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
Hi Tom, On Fri, Jul 19, 2013 at 1:56 PM, Charles Coldwell <coldwell@gmail.com> wrote: > On Fri, Jul 19, 2013 at 12:55 PM, Tom Rini <trini@ti.com> wrote: >> On Fri, Jul 19, 2013 at 10:01:34AM -0300, Fabio Estevam wrote: >> >>> From: Fabio Estevam <fabio.estevam@freescale.com> >>> >>> Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the >>> checking for ESTATUS_1000_XHALF, but it incorrectly sets the >>> SUPPORTED_1000baseX_Full flag in this case. >>> >>> Set the SUPPORTED_1000baseX_Half flag instead. >>> >>> Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com> >> >> So, do we need both patches to fix the problem? > > One patch fixes "the" problem; the other fixes another problem we > haven't seen yet (but will). I agree with Charles. Sascha Silbe's patch fixes the regression we noticed on some mx6 boards. This one is about a different issue and it is not related to the previous problem. IMHO we should apply both of them. Thanks, Fabio Estevam
On Fri, Jul 19, 2013 at 10:01:34AM -0300, Fabio Estevam wrote: > From: Fabio Estevam <fabio.estevam@freescale.com> > > Commit de1d786e (add support for Xilinx 1000BASE-X phy (GTX)) introduced the > checking for ESTATUS_1000_XHALF, but it incorrectly sets the > SUPPORTED_1000baseX_Full flag in this case. > > Set the SUPPORTED_1000baseX_Half flag instead. > > Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com> Applied to u-boot/master, thanks!
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 5ddb926..5692b1c 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -404,7 +404,7 @@ int genphy_config(struct phy_device *phydev) if (val & ESTATUS_1000_XFULL) features |= SUPPORTED_1000baseX_Full; if (val & ESTATUS_1000_XHALF) - features |= SUPPORTED_1000baseX_Full; + features |= SUPPORTED_1000baseX_Half; } phydev->supported = features;