On Mon, Sep 03, 2018 at 10:14:34PM +0200, Sam Ravnborg wrote:
> Hi Roland.
>
> > >
> > > In this code snip we only see that share_conf_val is used (line 149),
> > > it is not assigned.
> > > So we do not really see the context of your message in the code snip.
> > >
> > > Sam
> >
> > Thank you for your feedback. I took the opportunity and had a closer
> > look at the code. Here is the full context of the file from before the
> > patch:
> >
> > 83 static int imx_iomux_v3_set_state(struct pinctrl_device *pdev, struct device_node *np)
> > 84 {
> ...
> > 90 u32 share_conf_val;
> > 91
> ...
> > 94 if (share_conf) {
> ...
> > 110 share_conf_val =
> > 111 FIELD_PREP(SHARE_CONF_PAD_CTL_DSE, drive_strength) |
> > 112 FIELD_PREP(SHARE_CONF_PAD_CTL_SRE, slew_rate);
> ...
> > 142 for (i = 0; i < npins; i++) {
> ...
> > 148 u32 conf_val = share_conf ?
> > 149 share_conf_val : be32_to_cpu(*list++);
>
> The comment was only that despite your effort the changelog
> did not provide enough context.
> Above I have provided enough context to your otherwise nice explanation.
>
> > So if you feel that the (old) compiler is wrong here about the warning,
> > and the code itself is correct enough, feel free to leave out that patch
> > from the queue.
> The patch is IMO fine, but the changelog could be better.
Changed that to:
we have the pattern:
if (share_conf)
share_conf_val = ...;
...
if (share_conf)
use(share_conf_val);
GCC 5.4.0 doesn't recognize this so explicitly initialize
share_conf_val.
And applied this series.
Sascha
--
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
