Message ID | 1472452180.5372631.1348777689505.JavaMail.root@advansee.com |
---|---|
State | Awaiting Upstream |
Delegated to: | Stefano Babic |
Headers | show |
Hi Benoît, On Thu, Sep 27, 2012 at 5:28 PM, Benoît Thébaudeau <benoit.thebaudeau@advansee.com> wrote: > The FEC is the only SoC Ethernet support available on i.MX25, so define > cpu_eth_init() only for it instead of returning a misleading success code. Yes, but someone may want to use mx25 with a external LAN device (such as LAN92xx) instead of FEC. Regards, Fabio Estevam
Hi Fabio, On Thursday, September 27, 2012 10:55:22 PM, Fabio Estevam wrote: > Hi Benoît, > > On Thu, Sep 27, 2012 at 5:28 PM, Benoît Thébaudeau > <benoit.thebaudeau@advansee.com> wrote: > > The FEC is the only SoC Ethernet support available on i.MX25, so > > define > > cpu_eth_init() only for it instead of returning a misleading > > success code. > > Yes, but someone may want to use mx25 with a external LAN device > (such > as LAN92xx) instead of FEC. Yes, but in that case I think that there should be a board_eth_init() function defined, so this patch won't cause any trouble here. See in net/eth.c:eth_initialize() how these functions and their default implementations are handled. Best regards, Benoît
On 27/09/2012 23:43, Benoît Thébaudeau wrote: > Hi Fabio, > > On Thursday, September 27, 2012 10:55:22 PM, Fabio Estevam wrote: >> Hi Benoît, >> >> On Thu, Sep 27, 2012 at 5:28 PM, Benoît Thébaudeau >> <benoit.thebaudeau@advansee.com> wrote: >>> The FEC is the only SoC Ethernet support available on i.MX25, so >>> define >>> cpu_eth_init() only for it instead of returning a misleading >>> success code. >> >> Yes, but someone may want to use mx25 with a external LAN device >> (such >> as LAN92xx) instead of FEC. > > Yes, but in that case I think that there should be a board_eth_init() function > defined, so this patch won't cause any trouble here. See in > net/eth.c:eth_initialize() how these functions and their default implementations > are handled. I think also that in this case a board_eth_init() must be used - the #ifdef in cpu_eth_init() is then superflous. Best regards, Stefano
diff --git u-boot-imx-e1eb75b.orig/arch/arm/cpu/arm926ejs/mx25/generic.c u-boot-imx-e1eb75b/arch/arm/cpu/arm926ejs/mx25/generic.c index 5da2582..6374248 100644 --- u-boot-imx-e1eb75b.orig/arch/arm/cpu/arm926ejs/mx25/generic.c +++ u-boot-imx-e1eb75b/arch/arm/cpu/arm926ejs/mx25/generic.c @@ -206,9 +206,13 @@ void enable_caches(void) #endif } +#if defined(CONFIG_FEC_MXC) +/* + * Initializes on-chip ethernet controllers. + * to override, implement board_eth_init() + */ int cpu_eth_init(bd_t *bis) { -#if defined(CONFIG_FEC_MXC) struct ccm_regs *ccm = (struct ccm_regs *)IMX_CCM_BASE; ulong val; @@ -216,10 +220,8 @@ int cpu_eth_init(bd_t *bis) val |= (1 << 23); writel(val, &ccm->cgr0); return fecmxc_initialize(bis); -#else - return 0; -#endif } +#endif int get_clocks(void) {
The FEC is the only SoC Ethernet support available on i.MX25, so define cpu_eth_init() only for it instead of returning a misleading success code. Signed-off-by: Benoît Thébaudeau <benoit.thebaudeau@advansee.com> Cc: Stefano Babic <sbabic@denx.de> --- .../arch/arm/cpu/arm926ejs/mx25/generic.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)