Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754612Ab0HPPoR (ORCPT ); Mon, 16 Aug 2010 11:44:17 -0400 Received: from esgaroth.petrovitsch.at ([78.47.184.11]:4862 "EHLO esgaroth.petrovitsch.priv.at" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754233Ab0HPPoQ (ORCPT ); Mon, 16 Aug 2010 11:44:16 -0400 X-DKIM: Sendmail DKIM Filter v2.8.3 unknown-host o7GFhm6w016524 Subject: Re: [PATCH] perf tools: Don't use brace expansion. From: Bernd Petrovitsch To: Arnaldo Carvalho de Melo Cc: Peter Zijlstra , Kusanagi Kouichi , Paul Mackerras , Ingo Molnar , linux-kernel@vger.kernel.org In-Reply-To: <20100816152914.GC13095@ghostprotocols.net> References: <20100816124101.1829415C039@msa104.auone-net.jp> <1281967761.17267.6.camel@thorin> <1281969043.1926.1469.camel@laptop> <1281970478.17267.12.camel@thorin> <20100816152914.GC13095@ghostprotocols.net> Content-Type: text/plain; charset="UTF-8" Date: Mon, 16 Aug 2010 17:43:47 +0200 Message-ID: <1281973427.17267.16.camel@thorin> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 (2.28.3-1.fc12) Content-Transfer-Encoding: 7bit X-DCC-dcc1-Metrics: esgaroth.petrovitsch.priv.at; whitelist Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1910 Lines: 50 On Mon, 2010-08-16 at 12:29 -0300, Arnaldo Carvalho de Melo wrote: > Em Mon, Aug 16, 2010 at 04:54:38PM +0200, Bernd Petrovitsch escreveu: > > On Mon, 2010-08-16 at 16:30 +0200, Peter Zijlstra wrote: > > > On Mon, 2010-08-16 at 16:09 +0200, Bernd Petrovitsch wrote: > > > > > -$(shell sh -c 'mkdir -p $(OUTPUT)scripts/{perl,python}/Perf-Trace-Util/' 2> /dev/null) > > > > > -$(shell sh -c 'mkdir -p $(OUTPUT)util/{ui/browsers,scripting-engines}/' 2> /dev/null) > > > > > The other solution is to use standard-make features like in > > > > mkdir -p $(foreach d,ui/browsers scripting-engines,$(OUTPUT)util/$(d)/) 2> /dev/null > > > > > Is there actually a specific reason for the > > > > $(shell sh -c '...') > > > > around? > > > > It looks superflous. > > > > I think the reason is is that nobody who touched that file really knew > > > make all that well. Your version looks fine to me. > > > Ah, the reason is that they are not part of a rule but on the top-level > > (and thus always executed). > > So it worked by luck! /me runs :-P IMHO it's not really luck with GNU-make. > More seriously, so there is a reason for that to be like that and you're > not aware of any other shorter or more convenient way of achieving that One (obvious) alternative is to have rules triggering on the non-existence of these directories. > goal, right? Hmm, I'm not a "perf person". Which are the sufficient use-cases/tests that one can do to play around with the Makefile? `make -C tools/perf` is probably not enough. Any hints anyone? Bernd -- Bernd Petrovitsch Email : bernd@petrovitsch.priv.at LUGA : http://www.luga.at -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/