Messages in this thread | | | Date | Mon, 8 Jul 2013 15:28:12 +0800 | From | Fengguang Wu <> | Subject | Re: [v9fs:for-linus 7/7] ERROR: code indent should use tabs where possible |
| |
Hi Joe,
On Sun, Jul 07, 2013 at 09:54:27PM -0700, Joe Perches wrote: > On Mon, 2013-07-08 at 12:45 +0800, Fengguang Wu wrote: > > On Sun, Jul 07, 2013 at 08:54:04PM -0700, Joe Perches wrote: > > > On Mon, 2013-07-08 at 11:30 +0800, Fengguang Wu wrote: > > > > In long run I'd like to disable > > > > the check types that tend to have false positives. As for now, these > > > > checks are disabled in my checkpatch robot: > > > > > > Are you using a .checkpatch.conf file? > > > > Nope. > > > > > > PATCH_PREFIX > > > > LONG_LINE > > > > CAMELCASE > > > > > > That's a --strict test > > > > Yes. The CAMELCASE checks are false positives for the USB subsystem. > > Where they followed the USB spec and use names like bInterfaceClass > > all over the places. > > The latest version of checkpatch should avoid this.
That'd be great! So I'll use the latest version and take the CAMELCASE check back.
> > > > MISSING_SIGN_OFF > > > > > > Why would you even consider this one? > > > > It produces > > > > ERROR: Missing Signed-off-by: line(s) > > > > on > > commit: 8bb495e3f02401ee6f76d1b1d77f3ac9f079e376 Linux 3.10 > > Well, clearly Linus should sign off on his own commits...
Heh.
> > tree: git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git master > > commit: 97966e7ec944580bc3f46712f34ecb3c854fdd4b [37/81] Revert "ALSA: hda - Fix wrong power setup for HP paths of VIA codecs" > > > > tree: git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git master > > head: 120b1922175789a8a9bdafb7ea755fc63de2f392 > > commit: 10296279acc5d3caeca7b2cfde329c8280fe67e5 [32/81] Merge branch 'for-linus' > > > > and many other commits > > Not in -next as of today.
Since I test bleeding edge trees and many of them are not formal enough, I'm afraid I have to ignore the missing Signed-off-by lines..
10071 F Jul 05 Cc fengguang.wu ( 37:0) [asoc:topic/ak4554 0/2] ERROR: Missing Signed-off-by: line(s) 10087 F Jul 05 Cc fengguang.wu ( 36:0) [sound:master 32/81] ERROR: Missing Signed-off-by: line(s) 10088 F Jul 05 Cc fengguang.wu ( 36:0) [sound:master 37/81] ERROR: Missing Signed-off-by: line(s) 10116 F Jul 06 Cc fengguang.wu ( 37:0) [x0148406:tmp 0/45] ERROR: Missing Signed-off-by: line(s) 10117 F Jul 06 Cc fengguang.wu ( 37:0) [x0148406:tmp 1/45] ERROR: Missing Signed-off-by: line(s) 10118 F Jul 06 Cc fengguang.wu ( 37:0) [x0148406:tmp 2/45] ERROR: Missing Signed-off-by: line(s) 10119 F Jul 06 Cc fengguang.wu ( 37:0) [x0148406:tmp 3/45] ERROR: Missing Signed-off-by: line(s) 10120 F Jul 06 Cc fengguang.wu ( 37:0) [mmc:mmc-next 0/88] ERROR: Missing Signed-off-by: line(s) 10183 F Jul 07 Cc fengguang.wu ( 37:0) [target:target-per-cpu-ida 0/58] ERROR: Missing Signed-off-by: line(s)
> > > > TRAILING_STATEMENTS > > > False positives? Got an example? > > tree: git://git.freedesktop.org/git/nouveau/linux-2.6 drm-nouveau-next > > head: d2989b534ef6834ebf2425aecc040b894b567c91 > > commit: 01672ef454307bf63e93defb3599399b678ff58b [3/68] drm/nve0/fifo: copy engine context stored in ramfc, not externally > > > > ERROR: trailing statements should be on next line > > #28: FILE: drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c:146: > > + case NVDEV_ENGINE_GR : addr = 0x0210; break; > > Not false positives.
Yeah, however it does represent a common practice that I feel like respecting the developers' own choice.
Thanks, Fengguang
| |