On Thu, Sep 30, 2010 at 10:00:37AM -0700, jason wrote:
> Henry Ptasinski wrote:
> > Jason,
> >
> > It looks like tabs have been converted to spaces in this patch, even in the
> > unchanged code. git am doesn't like it, but it applies fine with patch -l.
> > None of the other patches in this series seem to have an issue (but patch
> > 8 in v2 has the same problem).
> >
>
> Do you have the line numbers of a few examples? I looked at my version of wl_iw.c and nothing jumped out at me...
As far as I can tell, all tabs have been converted in the patch. First
diff in your patch, for example, has spaces at the beginning of each line of
code:
--- a/drivers/staging/brcm80211/brcmfmac/wl_iw.c
+++ b/drivers/staging/brcm80211/brcmfmac/wl_iw.c
@@ -342,7 +342,8 @@ wl_iw_config_commit(struct net_device *dev,
WL_TRACE(("%s: SIOCSIWCOMMIT\n", dev->name));
- if ((error = dev_wlc_ioctl(dev, WLC_GET_SSID, &ssid, sizeof(ssid))))
+ error = dev_wlc_ioctl(dev, WLC_GET_SSID, &ssid, sizeof(ssid));
+ if (error)
return error;
ssid.SSID_len = dtoh32(ssid.SSID_len);
vs. the original code from wl_iw.c:
WL_TRACE(("%s: SIOCSIWCOMMIT\n", dev->name));
if ((error = dev_wlc_ioctl(dev, WLC_GET_SSID, &ssid, sizeof(ssid))))
return error;
ssid.SSID_len = dtoh32(ssid.SSID_len);
(Hmm. "Take a look at the whitespace differences" he says ...)
> Postfix and Mutt aren't known for doing that though. It's more likely that
> the original code had some bad white space voodoo. If so, those are pretty
> far down there on the list of cleanups. Unless it makes merging patches a
> pia, then we can move it to the top of the list...
No complaints on deferring whitespace cleanup. I just noticed that git didn't
seem to like the patch, apparently because whitespace changes cause it to have
trouble finding the context (patch also fails, but "patch -l" works).
- Henry
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
