* [U-Boot] net: fec_mxc: multiple calls to fec_mii_setspeed()
@ 2018-03-14 8:15 Yasushi SHOJI
2018-03-15 19:49 ` Joe Hershberger
0 siblings, 1 reply; 3+ messages in thread
From: Yasushi SHOJI @ 2018-03-14 8:15 UTC (permalink / raw)
To: u-boot
Hello,
I just noticed that in fec_mxc.c, fec_mii_setspeed() is called twice if
you are calling fecmxc_initialize_multi(), once from
fec_get_miibus(), and the other from fec_probe().
Is this intended? If so, would you mind to elaborate a bit?
Thanks,
--
yashi
^ permalink raw reply [flat|nested] 3+ messages in thread
* [U-Boot] net: fec_mxc: multiple calls to fec_mii_setspeed()
2018-03-14 8:15 [U-Boot] net: fec_mxc: multiple calls to fec_mii_setspeed() Yasushi SHOJI
@ 2018-03-15 19:49 ` Joe Hershberger
2018-03-16 0:01 ` Troy Kisky
0 siblings, 1 reply; 3+ messages in thread
From: Joe Hershberger @ 2018-03-15 19:49 UTC (permalink / raw)
To: u-boot
On Wed, Mar 14, 2018 at 3:15 AM, Yasushi SHOJI <yasushi.shoji@gmail.com> wrote:
> Hello,
>
> I just noticed that in fec_mxc.c, fec_mii_setspeed() is called twice if
> you are calling fecmxc_initialize_multi(), once from
> fec_get_miibus(), and the other from fec_probe().
>
> Is this intended? If so, would you mind to elaborate a bit?
I think this is a question for Troy since it seems he made the change
in commit fe428b909b83f0ef83d1fbc7a446bfb60bc4fe01 (net: fec_mxc: get
phydev before fec_probe). I'm not sure why that would be intended, but
also don't know what it hurts other than execution time.
-Joe
^ permalink raw reply [flat|nested] 3+ messages in thread
* [U-Boot] net: fec_mxc: multiple calls to fec_mii_setspeed()
2018-03-15 19:49 ` Joe Hershberger
@ 2018-03-16 0:01 ` Troy Kisky
0 siblings, 0 replies; 3+ messages in thread
From: Troy Kisky @ 2018-03-16 0:01 UTC (permalink / raw)
To: u-boot
On 3/15/2018 12:49 PM, Joe Hershberger wrote:
> On Wed, Mar 14, 2018 at 3:15 AM, Yasushi SHOJI <yasushi.shoji@gmail.com> wrote:
>> Hello,
>>
>> I just noticed that in fec_mxc.c, fec_mii_setspeed() is called twice if
>> you are calling fecmxc_initialize_multi(), once from
>> fec_get_miibus(), and the other from fec_probe().
>>
>> Is this intended? If so, would you mind to elaborate a bit?
>
> I think this is a question for Troy since it seems he made the change
> in commit fe428b909b83f0ef83d1fbc7a446bfb60bc4fe01 (net: fec_mxc: get
> phydev before fec_probe). I'm not sure why that would be intended, but
> also don't know what it hurts other than execution time.
>
> -Joe
>
Probably playing it safe that
/* Reset chip. */
writel(readl(&fec->eth->ecntrl) | FEC_ECNTRL_RESET, &fec->eth->ecntrl);
might mess up that register on some chip. A lot of manuals to check before removing
from fec_probe
Troy
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-03-16 0:01 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-03-14 8:15 [U-Boot] net: fec_mxc: multiple calls to fec_mii_setspeed() Yasushi SHOJI
2018-03-15 19:49 ` Joe Hershberger
2018-03-16 0:01 ` Troy Kisky
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.