Re: [PATCH v4 00/13] staging: lustre: lnet: code cleanups

2015-05-23 Thread Michael Shuey
Hm, that's unfortunate - but my own fault for lack of proper
etiquette.  I'll give this a week or two to settle, and build up
patches against other parts of lustre in the meantime.

BTW, you keep mentioning a v5 that I sent.  Where is that, exactly?
The last round of patches I sent I've kept labeled as "PATCH v4", and
I only hit git send-email once.  Could you forward me something from
this v5 series, so I could see if anything is amiss on my end?
--
Mike Shuey


On Sat, May 23, 2015 at 6:14 AM, Dan Carpenter  wrote:
> We would have applied the v3 patchset but now I don't know because we're
> up to v5.  We can't apply v5 because there are problems with it.  No
> one responded to v3 so Greg still might apply it or he might find these
> email threads too scrambled and delete everything and ask for a resend.
>
> It's pretty messed up so just wait for Greg to get to it before sending
> more patches?
>
> Basically you should only send patches which you assume will be applied.
> If no one responds after 3 days then probably that means everyone from
> the peanut gallery (Me, Sudip, Joe, the lustre devs), we don't have an
> issue.  Then Greg does the last review (2-3 weeks later perhaps).  But
> if it makes it past all the other reviews then generally Greg also will
> be ok with it.
>
> Greg applies patches in first come, first applied order.  If they don't
> apply then you have to redo it.  He doesn't invest a lot of time into
> figuring out why.  So you have to coordinate with the other devs, it's
> up to you how you do that.
>
> regards,
> dan carpenter
>
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel


Re: [PATCH v4 00/13] staging: lustre: lnet: code cleanups

2015-05-23 Thread Michael Shuey
Ah - that explains it.  I added additional files to the series, but
simply re-applied the original round of patches.  They would've not
been tagged as v4, while the patches pertaining to the newly-modified
files had v4 in the subject.

I'll be more thorough in future patch revisions.  Thanks for the explanation.
--
Mike Shuey


On Sat, May 23, 2015 at 8:39 AM, Sudip Mukherjee
 wrote:
> On Sat, May 23, 2015 at 08:09:54AM -0400, Michael Shuey wrote:
>> BTW, you keep mentioning a v5 that I sent.  Where is that, exactly?
>> The last round of patches I sent I've kept labeled as "PATCH v4", and
>> I only hit git send-email once.  Could you forward me something from
>> this v5 series, so I could see if anything is amiss on my end?
>
> I think it was not a v5. But what happened is in your series some
> of the patches were marked as v4 and some were not having any version
> so it appeared like a fresh series again. Like 1/13,2/13,3/13 has v4
> but the others donot have a version.
>
> regards
> sudip
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel


Re: [lustre-devel] [PATCH 1/3] staging: lustre: checkpatch cleanups for nidstring.c

2015-11-01 Thread Michael Shuey
I suspect you're over-thinking it.  The maintainers appear to be
reacting to the different types of style changes - "checkpatch
cleanups" is an awfully broad commit message.  I'd suggest breaking
this patch (and any others like it) into two pieces; one with
whitespace cleanups, and one with the "== NULL" fixes (and mentioning
both by kind in the commit message, rather than just attributing to
checkpatch).  Then issue a v2 of the series, and see where you land.

Of course, YMMV. :-)

--
Mike Shuey


On Sun, Nov 1, 2015 at 6:07 PM, Simmons, James A.  wrote:
>>On Thu, Oct 29, 2015 at 07:28:21PM -0400, James Simmons wrote:
>>> With nidstring now having the latest fixes we can
>>> now clean up all the remaining checkpatch errors
>>> for nidstring.c.
>>
>>Please be specific as to exactly what you changed, and break it up into
>>one-patch-per-thing.  And no, "fix all checkpatch errors" is not "one
>>thing"
>
> Hmm. This makes me think I might be going about this wrong.  Instead of
> doing style changes per file I should be doing one style change per subsystem
> instead. Unless you prefer doing these style changes on per file base. Perhaps
> for now I should focus on pushing the fixes that have cumulated and once
> caught up then finished off the style issues.
> ___
> lustre-devel mailing list
> lustre-de...@lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-devel-lustre.org
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel