| Message ID | 1462257612-28746-8-git-send-email-sjg@chromium.org |
|---|---|
| State | Accepted |
| Commit | f1dcc19b213df97127df9c6f4bec296ae2f91f38 |
| Delegated to: | Andreas Bießmann |
| Headers | show |
Hello Simon, Am 03.05.2016 um 08:40 schrieb Simon Glass: > Add driver-model support to this driver. The old code remains for now so > that we can convert boards one at a time. > > Signed-off-by: Simon Glass <sjg@chromium.org> > --- > > drivers/net/macb.c | 119 +++++++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 119 insertions(+) Thanks! Reviewed-by: Heiko Schocher <hs@denx.de> tested on the smartweb, corvus, taurus and axm board Tested-by: Heiko Schocher <hs@denx.de> bye, Heiko
On Tue, May 3, 2016 at 1:40 AM, Simon Glass <sjg@chromium.org> wrote: > Add driver-model support to this driver. The old code remains for now so > that we can convert boards one at a time. > > Signed-off-by: Simon Glass <sjg@chromium.org> Acked-by: Joe Hershberger <joe.hershberger@ni.com> But with a nit... > --- <snip> > +static const struct eth_ops macb_eth_ops = { > + .start = macb_start, > + .send = macb_send, > + .recv = macb_recv, > + .stop = macb_stop, > + .free_pkt = macb_free_pkt, > + .write_hwaddr = macb_write_hwaddr, > +}; > + > +static int macb_eth_probe(struct udevice *dev) > +{ > + struct eth_pdata *pdata = dev_get_platdata(dev); > + struct macb_device *macb = dev_get_priv(dev); > + > + macb->regs = (void *)pdata->iobase; > + > + _macb_eth_initialize(macb); > +#if defined(CONFIG_CMD_MII) || defined(CONFIG_PHYLIB) > + miiphy_register(dev->name, macb_miiphy_read, macb_miiphy_write); It's unfortunate that you're proliferating the oldest phy API. My semantic patch can come along and clean it up later as long as this is in before I run it again. > + macb->bus = miiphy_get_dev_by_name(dev->name); > +#endif > + > + return 0; > +}
Hi, On 2016/5/3 15:53, Heiko Schocher wrote: > Hello Simon, > > Am 03.05.2016 um 08:40 schrieb Simon Glass: >> Add driver-model support to this driver. The old code remains for now so >> that we can convert boards one at a time. >> >> Signed-off-by: Simon Glass <sjg@chromium.org> >> --- >> >> drivers/net/macb.c | 119 >> +++++++++++++++++++++++++++++++++++++++++++++++++++++ >> 1 file changed, 119 insertions(+) > > Thanks! > > Reviewed-by: Heiko Schocher <hs@denx.de> > > tested on the smartweb, corvus, taurus and axm board > > Tested-by: Heiko Schocher <hs@denx.de> I tried to test this patch series on SAMA5D2 Xplained board, but I have the following compile warning , did you experience it. ----8<----------- drivers/net/macb.c: In function 'macb_phy_init': drivers/net/macb.c:487:9: warning: passing argument 3 of 'phy_connect' from incompatible pointer type [enabled by default] In file included from include/miiphy.h:22:0, from drivers/net/macb.c:36: include/phy.h:226:20: note: expected 'struct udevice *' but argument is of type 'const struct device **' --->8---------------- > > bye, > Heiko Best Regards, Wenyou Yang
Hi > -----Original Message----- > From: U-Boot [mailto:u-boot-bounces@lists.denx.de] On Behalf Of Heiko > Schocher > Sent: 2016年5月3日 15:54 > To: Simon Glass <sjg@chromium.org> > Cc: U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > <joe.hershberger@ni.com> > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver model > > Hello Simon, > > Am 03.05.2016 um 08:40 schrieb Simon Glass: > > Add driver-model support to this driver. The old code remains for now > > so that we can convert boards one at a time. > > > > Signed-off-by: Simon Glass <sjg@chromium.org> > > --- > > > > drivers/net/macb.c | 119 > +++++++++++++++++++++++++++++++++++++++++++++++++++++ > > 1 file changed, 119 insertions(+) > > Thanks! > > Reviewed-by: Heiko Schocher <hs@denx.de> > > tested on the smartweb, corvus, taurus and axm board > > Tested-by: Heiko Schocher <hs@denx.de> I tried to test this patch series on SAMA5D2 Xplained board, but I have the compile warning below. Did you experience it? ---8<--------- drivers/net/macb.c: In function 'macb_phy_init': drivers/net/macb.c:487:9: warning: passing argument 3 of 'phy_connect' from incompatible pointer type [enabled by default] In file included from include/miiphy.h:22:0, from drivers/net/macb.c:36: include/phy.h:226:20: note: expected 'struct udevice *' but argument is of type 'const struct device **' --->8-------- Thanks. > > bye, > Heiko > -- > DENX Software Engineering GmbH, Managing Director: Wolfgang Denk > HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany > _______________________________________________ > U-Boot mailing list > U-Boot@lists.denx.de > http://lists.denx.de/mailman/listinfo/u-boot Best Regards, Wenyou Yang
Hi Joe, On 3 May 2016 at 14:54, Joe Hershberger <joe.hershberger@gmail.com> wrote: > On Tue, May 3, 2016 at 1:40 AM, Simon Glass <sjg@chromium.org> wrote: >> Add driver-model support to this driver. The old code remains for now so >> that we can convert boards one at a time. >> >> Signed-off-by: Simon Glass <sjg@chromium.org> > > Acked-by: Joe Hershberger <joe.hershberger@ni.com> > > But with a nit... > >> --- > > <snip> > >> +static const struct eth_ops macb_eth_ops = { >> + .start = macb_start, >> + .send = macb_send, >> + .recv = macb_recv, >> + .stop = macb_stop, >> + .free_pkt = macb_free_pkt, >> + .write_hwaddr = macb_write_hwaddr, >> +}; >> + >> +static int macb_eth_probe(struct udevice *dev) >> +{ >> + struct eth_pdata *pdata = dev_get_platdata(dev); >> + struct macb_device *macb = dev_get_priv(dev); >> + >> + macb->regs = (void *)pdata->iobase; >> + >> + _macb_eth_initialize(macb); >> +#if defined(CONFIG_CMD_MII) || defined(CONFIG_PHYLIB) >> + miiphy_register(dev->name, macb_miiphy_read, macb_miiphy_write); > > It's unfortunate that you're proliferating the oldest phy API. My > semantic patch can come along and clean it up later as long as this is > in before I run it again. OK, hopefully that will work. > >> + macb->bus = miiphy_get_dev_by_name(dev->name); >> +#endif >> + >> + return 0; >> +} Regards, Simon
Hi, On 4 May 2016 at 01:32, Yang, Wenyou <Wenyou.Yang@atmel.com> wrote: > > Hi > > > -----Original Message----- > > From: U-Boot [mailto:u-boot-bounces@lists.denx.de] On Behalf Of Heiko > > Schocher > > Sent: 2016年5月3日 15:54 > > To: Simon Glass <sjg@chromium.org> > > Cc: U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > > <joe.hershberger@ni.com> > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver model > > > > Hello Simon, > > > > Am 03.05.2016 um 08:40 schrieb Simon Glass: > > > Add driver-model support to this driver. The old code remains for now > > > so that we can convert boards one at a time. > > > > > > Signed-off-by: Simon Glass <sjg@chromium.org> > > > --- > > > > > > drivers/net/macb.c | 119 > > +++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > 1 file changed, 119 insertions(+) > > > > Thanks! > > > > Reviewed-by: Heiko Schocher <hs@denx.de> > > > > tested on the smartweb, corvus, taurus and axm board > > > > Tested-by: Heiko Schocher <hs@denx.de> > > I tried to test this patch series on SAMA5D2 Xplained board, but I have the compile warning below. Did you experience it? > > ---8<--------- > drivers/net/macb.c: In function 'macb_phy_init': > drivers/net/macb.c:487:9: warning: passing argument 3 of 'phy_connect' from incompatible pointer type [enabled by default] > In file included from include/miiphy.h:22:0, > from drivers/net/macb.c:36: > include/phy.h:226:20: note: expected 'struct udevice *' but argument is of type 'const struct device **' > --->8-------- No I don't see that problem. I did a full build test. What is the board config name you are using? > > Thanks. > > > > > bye, > > Heiko > > -- > > DENX Software Engineering GmbH, Managing Director: Wolfgang Denk > > HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany > > _______________________________________________ > > U-Boot mailing list > > U-Boot@lists.denx.de > > http://lists.denx.de/mailman/listinfo/u-boot > > > Best Regards, > Wenyou Yang Regards, Simon
Hi Simon, > -----Original Message----- > From: sjg@google.com [mailto:sjg@google.com] On Behalf Of Simon Glass > Sent: 2016年5月5日 10:38 > To: Yang, Wenyou <Wenyou.Yang@atmel.com> > Cc: hs@denx.de; U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > <joe.hershberger@ni.com> > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver model > > Hi, > > On 4 May 2016 at 01:32, Yang, Wenyou <Wenyou.Yang@atmel.com> wrote: > > > > Hi > > > > > -----Original Message----- > > > From: U-Boot [mailto:u-boot-bounces@lists.denx.de] On Behalf Of > > > Heiko Schocher > > > Sent: 2016年5月3日 15:54 > > > To: Simon Glass <sjg@chromium.org> > > > Cc: U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > > > <joe.hershberger@ni.com> > > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver > > > model > > > > > > Hello Simon, > > > > > > Am 03.05.2016 um 08:40 schrieb Simon Glass: > > > > Add driver-model support to this driver. The old code remains for > > > > now so that we can convert boards one at a time. > > > > > > > > Signed-off-by: Simon Glass <sjg@chromium.org> > > > > --- > > > > > > > > drivers/net/macb.c | 119 > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > > 1 file changed, 119 insertions(+) > > > > > > Thanks! > > > > > > Reviewed-by: Heiko Schocher <hs@denx.de> > > > > > > tested on the smartweb, corvus, taurus and axm board > > > > > > Tested-by: Heiko Schocher <hs@denx.de> > > > > I tried to test this patch series on SAMA5D2 Xplained board, but I have the > compile warning below. Did you experience it? > > > > ---8<--------- > > drivers/net/macb.c: In function 'macb_phy_init': > > drivers/net/macb.c:487:9: warning: passing argument 3 of 'phy_connect' > > from incompatible pointer type [enabled by default] In file included from > include/miiphy.h:22:0, > > from drivers/net/macb.c:36: > > include/phy.h:226:20: note: expected 'struct udevice *' but argument is of type > 'const struct device **' > > --->8-------- > > No I don't see that problem. I did a full build test. What is the board config name > you are using? The board is SAMA5D2 Xplained board, the .config file is attached. I noticed that in include/phy.h file, phy_connect() has different prototype for enabling CONFIG_DM_ETH or not. So, I think this issue should be exist. > > > > Thanks. > > > > > > > > bye, > > > Heiko > > > -- > > > DENX Software Engineering GmbH, Managing Director: Wolfgang Denk > > > HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, > > > Germany _______________________________________________ > > > U-Boot mailing list > > > U-Boot@lists.denx.de > > > http://lists.denx.de/mailman/listinfo/u-boot > > > > > > Best Regards, > > Wenyou Yang > > Regards, > Simon Best Regards, Wenyou Yang
Hi, On May 4, 2016 21:15, "Yang, Wenyou" <Wenyou.Yang@atmel.com> wrote: > > Hi Simon, > > > -----Original Message----- > > From: sjg@google.com [mailto:sjg@google.com] On Behalf Of Simon Glass > > Sent: 2016年5月5日 10:38 > > To: Yang, Wenyou <Wenyou.Yang@atmel.com> > > Cc: hs@denx.de; U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > > <joe.hershberger@ni.com> > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver model > > > > Hi, > > > > On 4 May 2016 at 01:32, Yang, Wenyou <Wenyou.Yang@atmel.com> wrote: > > > > > > Hi > > > > > > > -----Original Message----- > > > > From: U-Boot [mailto:u-boot-bounces@lists.denx.de] On Behalf Of > > > > Heiko Schocher > > > > Sent: 2016年5月3日 15:54 > > > > To: Simon Glass <sjg@chromium.org> > > > > Cc: U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > > > > <joe.hershberger@ni.com> > > > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver > > > > model > > > > > > > > Hello Simon, > > > > > > > > Am 03.05.2016 um 08:40 schrieb Simon Glass: > > > > > Add driver-model support to this driver. The old code remains for > > > > > now so that we can convert boards one at a time. > > > > > > > > > > Signed-off-by: Simon Glass <sjg@chromium.org> > > > > > --- > > > > > > > > > > drivers/net/macb.c | 119 > > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > > > 1 file changed, 119 insertions(+) > > > > > > > > Thanks! > > > > > > > > Reviewed-by: Heiko Schocher <hs@denx.de> > > > > > > > > tested on the smartweb, corvus, taurus and axm board > > > > > > > > Tested-by: Heiko Schocher <hs@denx.de> > > > > > > I tried to test this patch series on SAMA5D2 Xplained board, but I have the > > compile warning below. Did you experience it? > > > > > > ---8<--------- > > > drivers/net/macb.c: In function 'macb_phy_init': > > > drivers/net/macb.c:487:9: warning: passing argument 3 of 'phy_connect' > > > from incompatible pointer type [enabled by default] In file included from > > include/miiphy.h:22:0, > > > from drivers/net/macb.c:36: > > > include/phy.h:226:20: note: expected 'struct udevice *' but argument is of type > > 'const struct device **' > > > --->8-------- > > > > No I don't see that problem. I did a full build test. What is the board config name > > you are using? > > The board is SAMA5D2 Xplained board, the .config file is attached. > > I noticed that in include/phy.h file, phy_connect() has different prototype for enabling CONFIG_DM_ETH or not. > > So, I think this issue should be exist. Is that board in mainline? > > > > > > > Thanks. > > > > > > > > > > > bye, > > > > Heiko > > > > -- > > > > DENX Software Engineering GmbH, Managing Director: Wolfgang Denk > > > > HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, > > > > Germany _______________________________________________ > > > > U-Boot mailing list > > > > U-Boot@lists.denx.de > > > > http://lists.denx.de/mailman/listinfo/u-boot Regards, Simon
Hi, On 2016/5/5 11:18, Simon Glass wrote: > > Hi, > > On May 4, 2016 21:15, "Yang, Wenyou" <Wenyou.Yang@atmel.com > <mailto:Wenyou.Yang@atmel.com>> wrote: > > > > Hi Simon, > > > > > -----Original Message----- > > > From: sjg@google.com <mailto:sjg@google.com> > [mailto:sjg@google.com <mailto:sjg@google.com>] On Behalf Of Simon Glass > > > Sent: 2016年5月5日 10:38 > > > To: Yang, Wenyou <Wenyou.Yang@atmel.com > <mailto:Wenyou.Yang@atmel.com>> > > > Cc: hs@denx.de <mailto:hs@denx.de>; U-Boot Mailing List > <u-boot@lists.denx.de <mailto:u-boot@lists.denx.de>>; Joe Hershberger > > > <joe.hershberger@ni.com <mailto:joe.hershberger@ni.com>> > > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver model > > > > > > Hi, > > > > > > On 4 May 2016 at 01:32, Yang, Wenyou <Wenyou.Yang@atmel.com > <mailto:Wenyou.Yang@atmel.com>> wrote: > > > > > > > > Hi > > > > > > > > > -----Original Message----- > > > > > From: U-Boot [mailto:u-boot-bounces@lists.denx.de > <mailto:u-boot-bounces@lists.denx.de>] On Behalf Of > > > > > Heiko Schocher > > > > > Sent: 2016年5月3日 15:54 > > > > > To: Simon Glass <sjg@chromium.org <mailto:sjg@chromium.org>> > > > > > Cc: U-Boot Mailing List <u-boot@lists.denx.de > <mailto:u-boot@lists.denx.de>>; Joe Hershberger > > > > > <joe.hershberger@ni.com <mailto:joe.hershberger@ni.com>> > > > > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver > > > > > model > > > > > > > > > > Hello Simon, > > > > > > > > > > Am 03.05.2016 um 08:40 schrieb Simon Glass: > > > > > > Add driver-model support to this driver. The old code > remains for > > > > > > now so that we can convert boards one at a time. > > > > > > > > > > > > Signed-off-by: Simon Glass <sjg@chromium.org > <mailto:sjg@chromium.org>> > > > > > > --- > > > > > > > > > > > > drivers/net/macb.c | 119 > > > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > > > > 1 file changed, 119 insertions(+) > > > > > > > > > > Thanks! > > > > > > > > > > Reviewed-by: Heiko Schocher <hs@denx.de <mailto:hs@denx.de>> > > > > > > > > > > tested on the smartweb, corvus, taurus and axm board > > > > > > > > > > Tested-by: Heiko Schocher <hs@denx.de <mailto:hs@denx.de>> > > > > > > > > I tried to test this patch series on SAMA5D2 Xplained board, but > I have the > > > compile warning below. Did you experience it? > > > > > > > > ---8<--------- > > > > drivers/net/macb.c: In function 'macb_phy_init': > > > > drivers/net/macb.c:487:9: warning: passing argument 3 of > 'phy_connect' > > > > from incompatible pointer type [enabled by default] In file > included from > > > include/miiphy.h:22:0, > > > > from drivers/net/macb.c:36: > > > > include/phy.h:226:20: note: expected 'struct udevice *' but > argument is of type > > > 'const struct device **' > > > > --->8-------- > > > > > > No I don't see that problem. I did a full build test. What is the > board config name > > > you are using? > > > > The board is SAMA5D2 Xplained board, the .config file is attached. > > > > I noticed that in include/phy.h file, phy_connect() has different > prototype for enabling CONFIG_DM_ETH or not. > > > > So, I think this issue should be exist. > > Is that board in mainline? > Yes, in mainline. The type of 'dev' parameter of phy_connect() is struct udevice *, instead of struct eth_device * when enabling CONFIG_DM_ETH. > > > > > > > > > > Thanks. > > > > > > > > > > > > > > bye, > > > > > Heiko > > > > > -- > > > > > DENX Software Engineering GmbH, Managing Director: Wolfgang Denk > > > > > HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, > > > > > Germany _______________________________________________ > > > > > U-Boot mailing list > > > > > U-Boot@lists.denx.de <mailto:U-Boot@lists.denx.de> > > > > > http://lists.denx.de/mailman/listinfo/u-boot > > Regards, > Simon > Best Regards, Wenyou Yang
Hi, On 4 May 2016 at 21:37, Yang, Wenyou <wenyou.yang@atmel.com> wrote: > Hi, > > > On 2016/5/5 11:18, Simon Glass wrote: > > Hi, > > On May 4, 2016 21:15, "Yang, Wenyou" <Wenyou.Yang@atmel.com> wrote: >> >> Hi Simon, >> >> > -----Original Message----- >> > From: sjg@google.com [mailto:sjg@google.com] On Behalf Of Simon Glass >> > Sent: 2016年5月5日 10:38 >> > To: Yang, Wenyou <Wenyou.Yang@atmel.com> >> > Cc: hs@denx.de; U-Boot Mailing List <u-boot@lists.denx.de>; Joe >> > Hershberger >> > <joe.hershberger@ni.com> >> > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver model >> > >> > Hi, >> > >> > On 4 May 2016 at 01:32, Yang, Wenyou <Wenyou.Yang@atmel.com> wrote: >> > > >> > > Hi >> > > >> > > > -----Original Message----- >> > > > From: U-Boot [mailto:u-boot-bounces@lists.denx.de] On Behalf Of >> > > > Heiko Schocher >> > > > Sent: 2016年5月3日 15:54 >> > > > To: Simon Glass <sjg@chromium.org> >> > > > Cc: U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger >> > > > <joe.hershberger@ni.com> >> > > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver >> > > > model >> > > > >> > > > Hello Simon, >> > > > >> > > > Am 03.05.2016 um 08:40 schrieb Simon Glass: >> > > > > Add driver-model support to this driver. The old code remains for >> > > > > now so that we can convert boards one at a time. >> > > > > >> > > > > Signed-off-by: Simon Glass <sjg@chromium.org> >> > > > > --- >> > > > > >> > > > > drivers/net/macb.c | 119 >> > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++ >> > > > > 1 file changed, 119 insertions(+) >> > > > >> > > > Thanks! >> > > > >> > > > Reviewed-by: Heiko Schocher <hs@denx.de> >> > > > >> > > > tested on the smartweb, corvus, taurus and axm board >> > > > >> > > > Tested-by: Heiko Schocher <hs@denx.de> >> > > >> > > I tried to test this patch series on SAMA5D2 Xplained board, but I >> > > have the >> > compile warning below. Did you experience it? >> > > >> > > ---8<--------- >> > > drivers/net/macb.c: In function 'macb_phy_init': >> > > drivers/net/macb.c:487:9: warning: passing argument 3 of 'phy_connect' >> > > from incompatible pointer type [enabled by default] In file included >> > > from >> > include/miiphy.h:22:0, >> > > from drivers/net/macb.c:36: >> > > include/phy.h:226:20: note: expected 'struct udevice *' but argument >> > > is of type >> > 'const struct device **' >> > > --->8-------- >> > >> > No I don't see that problem. I did a full build test. What is the board >> > config name >> > you are using? >> >> The board is SAMA5D2 Xplained board, the .config file is attached. >> >> I noticed that in include/phy.h file, phy_connect() has different >> prototype for enabling CONFIG_DM_ETH or not. >> >> So, I think this issue should be exist. > > Is that board in mainline? > > Yes, in mainline. > > The type of 'dev' parameter of phy_connect() is struct udevice *, instead of > struct eth_device * when enabling CONFIG_DM_ETH. OK, sounds like it needs a patch. Can you fix it when you enable DM_ETH for that board? Regards, Simon
Hi Simon, > -----Original Message----- > From: sjg@google.com [mailto:sjg@google.com] On Behalf Of Simon Glass > Sent: 2016年5月6日 1:19 > To: Yang, Wenyou <Wenyou.Yang@atmel.com> > Cc: U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > <joe.hershberger@ni.com>; hs@denx.de > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver model > > Hi, > > On 4 May 2016 at 21:37, Yang, Wenyou <wenyou.yang@atmel.com> wrote: > > Hi, > > > > > > On 2016/5/5 11:18, Simon Glass wrote: > > > > Hi, > > > > On May 4, 2016 21:15, "Yang, Wenyou" <Wenyou.Yang@atmel.com> wrote: > >> > >> Hi Simon, > >> > >> > -----Original Message----- > >> > From: sjg@google.com [mailto:sjg@google.com] On Behalf Of Simon > >> > Glass > >> > Sent: 2016年5月5日 10:38 > >> > To: Yang, Wenyou <Wenyou.Yang@atmel.com> > >> > Cc: hs@denx.de; U-Boot Mailing List <u-boot@lists.denx.de>; Joe > >> > Hershberger <joe.hershberger@ni.com> > >> > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to driver > >> > model > >> > > >> > Hi, > >> > > >> > On 4 May 2016 at 01:32, Yang, Wenyou <Wenyou.Yang@atmel.com> wrote: > >> > > > >> > > Hi > >> > > > >> > > > -----Original Message----- > >> > > > From: U-Boot [mailto:u-boot-bounces@lists.denx.de] On Behalf Of > >> > > > Heiko Schocher > >> > > > Sent: 2016年5月3日 15:54 > >> > > > To: Simon Glass <sjg@chromium.org> > >> > > > Cc: U-Boot Mailing List <u-boot@lists.denx.de>; Joe Hershberger > >> > > > <joe.hershberger@ni.com> > >> > > > Subject: Re: [U-Boot] [PATCH 07/18] net: macb: Convert to > >> > > > driver model > >> > > > > >> > > > Hello Simon, > >> > > > > >> > > > Am 03.05.2016 um 08:40 schrieb Simon Glass: > >> > > > > Add driver-model support to this driver. The old code remains > >> > > > > for now so that we can convert boards one at a time. > >> > > > > > >> > > > > Signed-off-by: Simon Glass <sjg@chromium.org> > >> > > > > --- > >> > > > > > >> > > > > drivers/net/macb.c | 119 > >> > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++ > >> > > > > 1 file changed, 119 insertions(+) > >> > > > > >> > > > Thanks! > >> > > > > >> > > > Reviewed-by: Heiko Schocher <hs@denx.de> > >> > > > > >> > > > tested on the smartweb, corvus, taurus and axm board > >> > > > > >> > > > Tested-by: Heiko Schocher <hs@denx.de> > >> > > > >> > > I tried to test this patch series on SAMA5D2 Xplained board, but > >> > > I have the > >> > compile warning below. Did you experience it? > >> > > > >> > > ---8<--------- > >> > > drivers/net/macb.c: In function 'macb_phy_init': > >> > > drivers/net/macb.c:487:9: warning: passing argument 3 of 'phy_connect' > >> > > from incompatible pointer type [enabled by default] In file > >> > > included from > >> > include/miiphy.h:22:0, > >> > > from drivers/net/macb.c:36: > >> > > include/phy.h:226:20: note: expected 'struct udevice *' but > >> > > argument is of type > >> > 'const struct device **' > >> > > --->8-------- > >> > > >> > No I don't see that problem. I did a full build test. What is the > >> > board config name you are using? > >> > >> The board is SAMA5D2 Xplained board, the .config file is attached. > >> > >> I noticed that in include/phy.h file, phy_connect() has different > >> prototype for enabling CONFIG_DM_ETH or not. > >> > >> So, I think this issue should be exist. > > > > Is that board in mainline? > > > > Yes, in mainline. > > > > The type of 'dev' parameter of phy_connect() is struct udevice *, > > instead of struct eth_device * when enabling CONFIG_DM_ETH. > > OK, sounds like it needs a patch. Can you fix it when you enable DM_ETH for that > board? Okay, I will try to fix it. Best Regards, Wenyou Yang
diff --git a/drivers/net/macb.c b/drivers/net/macb.c index ecdfc1b..63fb466 100644 --- a/drivers/net/macb.c +++ b/drivers/net/macb.c @@ -4,6 +4,7 @@ * SPDX-License-Identifier: GPL-2.0+ */ #include <common.h> +#include <dm.h> /* * The u-boot networking stack is a little weird. It seems like the @@ -28,7 +29,9 @@ */ #include <net.h> +#ifndef CONFIG_DM_ETH #include <netdev.h> +#endif #include <malloc.h> #include <miiphy.h> @@ -100,11 +103,15 @@ struct macb_device { unsigned long dummy_desc_dma; const struct device *dev; +#ifndef CONFIG_DM_ETH struct eth_device netdev; +#endif unsigned short phy_addr; struct mii_dev *bus; }; +#ifndef CONFIG_DM_ETH #define to_macb(_nd) container_of(_nd, struct macb_device, netdev) +#endif static int macb_is_gem(struct macb_device *macb) { @@ -194,8 +201,13 @@ void __weak arch_get_mdio_control(const char *name) int macb_miiphy_read(const char *devname, u8 phy_adr, u8 reg, u16 *value) { +#ifdef CONFIG_DM_ETH + struct udevice *dev = eth_get_dev_by_name(devname); + struct macb_device *macb = dev_get_priv(dev); +#else struct eth_device *dev = eth_get_dev_by_name(devname); struct macb_device *macb = to_macb(dev); +#endif if (macb->phy_addr != phy_adr) return -1; @@ -208,8 +220,13 @@ int macb_miiphy_read(const char *devname, u8 phy_adr, u8 reg, u16 *value) int macb_miiphy_write(const char *devname, u8 phy_adr, u8 reg, u16 value) { +#ifdef CONFIG_DM_ETH + struct udevice *dev = eth_get_dev_by_name(devname); + struct macb_device *macb = dev_get_priv(dev); +#else struct eth_device *dev = eth_get_dev_by_name(devname); struct macb_device *macb = to_macb(dev); +#endif if (macb->phy_addr != phy_adr) return -1; @@ -764,6 +781,7 @@ static void _macb_eth_initialize(struct macb_device *macb) macb_writel(macb, NCFGR, ncfgr); } +#ifndef CONFIG_DM_ETH static int macb_send(struct eth_device *netdev, void *packet, int length) { struct macb_device *macb = to_macb(netdev); @@ -849,5 +867,106 @@ int macb_eth_initialize(int id, void *regs, unsigned int phy_addr) #endif return 0; } +#endif /* !CONFIG_DM_ETH */ + +#ifdef CONFIG_DM_ETH + +static int macb_start(struct udevice *dev) +{ + struct macb_device *macb = dev_get_priv(dev); + + return _macb_init(macb, dev->name); +} + +static int macb_send(struct udevice *dev, void *packet, int length) +{ + struct macb_device *macb = dev_get_priv(dev); + + return _macb_send(macb, dev->name, packet, length); +} + +static int macb_recv(struct udevice *dev, int flags, uchar **packetp) +{ + struct macb_device *macb = dev_get_priv(dev); + + macb->next_rx_tail = macb->rx_tail; + macb->wrapped = false; + + return _macb_recv(macb, packetp); +} + +static int macb_free_pkt(struct udevice *dev, uchar *packet, int length) +{ + struct macb_device *macb = dev_get_priv(dev); + + reclaim_rx_buffers(macb, macb->next_rx_tail); + + return 0; +} + +static void macb_stop(struct udevice *dev) +{ + struct macb_device *macb = dev_get_priv(dev); + + _macb_halt(macb); +} + +static int macb_write_hwaddr(struct udevice *dev) +{ + struct eth_pdata *plat = dev_get_platdata(dev); + struct macb_device *macb = dev_get_priv(dev); + + return _macb_write_hwaddr(macb, plat->enetaddr); +} + +static const struct eth_ops macb_eth_ops = { + .start = macb_start, + .send = macb_send, + .recv = macb_recv, + .stop = macb_stop, + .free_pkt = macb_free_pkt, + .write_hwaddr = macb_write_hwaddr, +}; + +static int macb_eth_probe(struct udevice *dev) +{ + struct eth_pdata *pdata = dev_get_platdata(dev); + struct macb_device *macb = dev_get_priv(dev); + + macb->regs = (void *)pdata->iobase; + + _macb_eth_initialize(macb); +#if defined(CONFIG_CMD_MII) || defined(CONFIG_PHYLIB) + miiphy_register(dev->name, macb_miiphy_read, macb_miiphy_write); + macb->bus = miiphy_get_dev_by_name(dev->name); +#endif + + return 0; +} + +static int macb_eth_ofdata_to_platdata(struct udevice *dev) +{ + struct eth_pdata *pdata = dev_get_platdata(dev); + + pdata->iobase = dev_get_addr(dev); + return 0; +} + +static const struct udevice_id macb_eth_ids[] = { + { .compatible = "cdns,macb" }, + { } +}; + +U_BOOT_DRIVER(eth_macb) = { + .name = "eth_macb", + .id = UCLASS_ETH, + .of_match = macb_eth_ids, + .ofdata_to_platdata = macb_eth_ofdata_to_platdata, + .probe = macb_eth_probe, + .ops = &macb_eth_ops, + .priv_auto_alloc_size = sizeof(struct macb_device), + .platdata_auto_alloc_size = sizeof(struct eth_pdata), +}; +#endif #endif
Add driver-model support to this driver. The old code remains for now so that we can convert boards one at a time. Signed-off-by: Simon Glass <sjg@chromium.org> --- drivers/net/macb.c | 119 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 119 insertions(+)