From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp2200-217.mail.aliyun.com (smtp2200-217.mail.aliyun.com [121.197.200.217]) by sourceware.org (Postfix) with ESMTPS id A27423857C56 for ; Wed, 4 Nov 2020 02:25:21 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org A27423857C56 Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=c-sky.com Authentication-Results: sourceware.org; spf=none smtp.mailfrom=jiejie_rong@c-sky.com X-Alimail-AntiSpam: AC=CONTINUE; BC=0.0418741|-1; CH=green; DM=|CONTINUE|false|; DS=CONTINUE|ham_enroll_verification|0.261113-0.00671839-0.732168; FP=0|0|0|0|0|-1|-1|-1; HT=ay29a033018047212; MF=jiejie_rong@c-sky.com; NM=1; PH=DS; RN=5; RT=5; SR=0; TI=SMTPD_---.IsHnKfD_1604456717; Received: from 30.225.212.51(mailfrom:jiejie_rong@c-sky.com fp:SMTPD_---.IsHnKfD_1604456717) by smtp.aliyun-inc.com(10.194.98.226); Wed, 04 Nov 2020 10:25:17 +0800 Date: Wed, 4 Nov 2020 10:25:11 +0800 From: Jojo R To: Richard Sandiford Cc: segher@kernel.crashing.org, yunhai.syh@alibaba-inc.com, gcc-patches@gcc.gnu.org, joseph@codesourcery.com Message-ID: <8bff8454-9577-4527-8f93-4fbb502404cf@Spark> In-Reply-To: References: <20200915091550.7424-1-jiejie_rong@c-sky.com> <84a137c5-13e5-48c1-a5fb-eab4ccdf505e@Spark> Subject: Re: [PATCH v7] genemit.c (main): split insn-emit.c for compiling parallelly X-Readdle-Message-ID: 8bff8454-9577-4527-8f93-4fbb502404cf@Spark MIME-Version: 1.0 X-Spam-Status: No, score=-10.2 required=5.0 tests=BAYES_00, BODY_8BITS, GIT_PATCH_0, HTML_MESSAGE, KAM_DMARC_STATUS, KAM_LAZY_DOMAIN_SECURITY, KAM_SHORT, SPF_HELO_NONE, SPF_NONE, TXREP, UNPARSEABLE_RELAY autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline X-Content-Filtered-By: Mailman/MimeDel 2.1.29 X-BeenThere: gcc-patches@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 04 Nov 2020 02:25:25 -0000 Jojo =E5=9C=A8 2020=E5=B9=B410=E6=9C=8830=E6=97=A5 +0800 PM11:29=EF=BC=8CRicha= rd Sandiford =EF=BC=8C=E5=86=99=E9=81=93=EF=BC= =9A > Jojo R writes: > > Jojo > > =E5=9C=A8 2020=E5=B9=B410=E6=9C=8827=E6=97=A5 +0800 PM10:14=EF=BC=8CR= ichard Sandiford =EF=BC=8C=E5=86=99=E9=81=93= =EF=BC=9A > > > Jojo R writes: > > > > gcc/ChangeLog: > > > > > > > > * genemit.c (main): Print 'split line'. > > > > * Makefile.in (insn-emit.c): Define split count and file > > > > > > > > --- > > > > gcc/Makefile.in =7C 19 +++++++++ > > > > gcc/genemit.c =7C 104 +++++++++++++++++++++++++++++--------------= ----- > > > > 2 files changed, 83 insertions(+), 40 deletions(-) > > > > > > > > diff --git a/gcc/Makefile.in b/gcc/Makefile.in > > > > index 79e854aa938..a7fcc7d5949 100644 > > > > --- a/gcc/Makefile.in > > > > +++ b/gcc/Makefile.in > > > > =40=40 -1258,6 +1258,21 =40=40 ANALYZER=5FOBJS =3D =5C > > > > =23 We put the *-match.o and insn-*.o files first so that a paral= lel make > > > > =23 will build them sooner, because they are large and otherwise = tend to be > > > > =23 the last objects to finish building. > > > > + > > > > +=23 target overrides > > > > +-include =24(tmake=5Ffile) > > > > + > > > > +INSN-GENERATED-SPLIT-NUM =3F=3D 0 > > > > + > > > > +insn-generated-split-num =3D =24(shell i=3D1; j=3D=60expr =24(IN= SN-GENERATED-SPLIT-NUM) + 1=60; =5C > > > > + while test =24=24i -le =24=24j; do =5C > > > > + echo =24=24i; i=3D=60expr =24=24i + 1=60; =5C > > > > + done) > > > > + > > > > +insn-emit-split-c :=3D =24(foreach o, =24(shell for i in =24(ins= n-generated-split-num); do echo =24=24i; done), insn-emit=24(o).c) > > > > +insn-emit-split-obj =3D =24(patsubst %.c,%.o, =24(insn-emit-spli= t-c)) > > > > +=24(insn-emit-split-c): insn-emit.c > > > > > > Sorry for the slow reply. I stand by what I said in > > > https://gcc.gnu.org/pipermail/gcc-patches/2020-August/552863.html: > > > > > > I think we should use the same wordlist technique as check=5Fp=5Fnu= mbers=5B0-6=5D. > > > So I guess the first step would be to rename check=5Fp=5Fnumbers=5B= 0-6=5D to > > > something more general and use it both here and in check=5Fp=5Fnumb= ers. > > > > > > I think that would be better than having two different ways of > > > generating lists of numbers, one directly in make and one calling > > > out to the shell. But I didn't want to reassert that comment in > > > case anyone was prepared to approve the patch in its current form. > > > > > > > Ok & Thanks. > > > > It=E2=80=99s fixed in patch v8. > > Thanks. Like I say, I think we should rename check=5Fp=5Fnumbers* at th= e > same time, since it's now used for more than just parallel check. Maybe= > s/check=5Fp=5Fnumbers/number=5Fseries/. Ok & It=E2=80=99s fixed in patch v9. > > But otherwise it looks good. > > > > BTW, do you have a copyright assignment on file=3F > > > > I email the patch without copyright, and I think it is same with othe= r gcc community patch. > > Some changes can be so small and mechanical that they're not in practic= e > copyrightable, but all other changes need a copyright assignment. > Unfortunately this patch is too complex to fall into the first category= . > See: > > https://gcc.gnu.org/contribute.html=23legal > > for more details about the requirement and process. > =46rom the patch of=C2=A0https://gcc.gnu.org/legacy-ml/gcc-patches/2018-0= 7/msg01289.html we have supported c-sky port in GCC backend, I think we also need to subm= it a copyright for it and have submitted that. I am contacting=C2=A0assign=40gnu.org=C2=A0to check it. Thanks. > Thanks, > Richard