From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=unavailable autolearn_force=no version=3.4.4 X-Received: by 2002:a37:a942:: with SMTP id s63-v6mr4284079qke.50.1527180261925; Thu, 24 May 2018 09:44:21 -0700 (PDT) X-Received: by 2002:a1f:aa12:: with SMTP id t18-v6mr209901vke.12.1527180261730; Thu, 24 May 2018 09:44:21 -0700 (PDT) Path: eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!feeder.eternal-september.org!news.gegeweb.eu!gegeweb.org!news.redatomik.org!newsfeed.xs4all.nl!newsfeed8.news.xs4all.nl!85.12.16.70.MISMATCH!peer03.ams1!peer.ams1.xlned.com!news.xlned.com!peer03.am4!peer.am4.highwinds-media.com!peer01.iad!feed-me.highwinds-media.com!news.highwinds-media.com!x25-v6no929949qto.0!news-out.google.com!p41-v6ni279qtp.1!nntp.google.com!x25-v6no929937qto.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Thu, 24 May 2018 09:44:21 -0700 (PDT) In-Reply-To: <294fa0cd-ec72-4f0f-8065-0a3d5e1087fa@googlegroups.com> Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=47.185.233.194; posting-account=zwxLlwoAAAChLBU7oraRzNDnqQYkYbpo NNTP-Posting-Host: 47.185.233.194 References: <5c2523c1-9ea5-453c-b80e-9cb0dcd16de0@googlegroups.com> <293cf892-1320-49e6-a25f-a36ea098cd34@googlegroups.com> <294fa0cd-ec72-4f0f-8065-0a3d5e1087fa@googlegroups.com> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: Subject: Re: DragonEgg has been revived From: "Dan'l Miller" Injection-Date: Thu, 24 May 2018 16:44:21 +0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Received-Bytes: 6431 X-Received-Body-CRC: 1071313035 Xref: reader02.eternal-september.org comp.lang.ada:52642 Date: 2018-05-24T09:44:21-07:00 List-Id: On Thursday, May 24, 2018 at 10:38:10 AM UTC-5, Dan'l Miller wrote: > Here is where the fallacious theorem was postulated. Shark8 years for = =E2=80=9Cgenerat[ing] code for =E2=80=A6 targets > without =E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2= =E2=80=A2=E2=80=A2=E2=80=A2any=E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2= =E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2=E2=80=A2 GPL restrictions on the gener= ated code=E2=80=9D. =20 Btw, I meant: Shark8 yearns for, not years for (although I am pretty sure that he has bee= n yearning for that for years). > Btw, while we are on the topic of re-compiling Target Code, it seems that= another act on the Target Code > written by FSF GCC as licensed by GPLv3-with-RLEv3.1 might be dodgy/sketc= hy/perhaps-not-permitted > without-providing-original-source-code-as-per-GPLv3's-terms:=20 > > 4) Imagine a (closed-source EULA-licensed paid-software) machine-code-to-= machine-code > translator/re-compiler that coverts from one ISA-and-ABI to another drast= ically-different ISA-and-ABI. > The input Target Code written by FSF GCC could be considered IR to this r= e-compiler. The resulting > output machine-code on the drastically-different ISA and ABI could* rever= t under the RLEv3.1 to > full-fledged GPLv3.=20 > > * if someone were to consider the input Target Code as IR, much the same = as the McSema line of > reasoning.=20 > > And then the Installation Information portion of the GPLv3 could come int= o play, unless the re-compiled > drastically-different-ISA-&-ABI machine-code was installed on something = =E2=80=9Cindustrial=E2=80=9D sufficiently > resembling a fighter-jet cockpit or telephone switch in a central office = or metal-stamping machine on a > factory floor. So that close-source re-compiler that someone paid money = to EULA-license its usage > must now be provided along with the source code for free to any possessor= on any hardware whose > =E2=80=9Csignificant mode of use=E2=80=9D is at least partially non-=E2= =80=9Cindustrial=E2=80=9D.=20 Oh great. Now I have convinced myself that there is enough sketchiness/dod= giness in this vicinity to question whether Apple's forthcoming App-Store L= LVM-bitcode-IR recompiler wouldn't satisfy any one or more required terms o= f any GPLv3-with-RLEv3.1-licensed compiler (FSF or otherwise) in such as wa= y as to cause both: 1) the app to need to provide its Source Code; and 2) Apple to need to provide the Installation Information and Corresponding = Source Code on a =E2=80=9Cconsumer=E2=80=9D device that is not-=E2=80=9Csuf= ficiently=E2=80=9D =E2=80=9Cindustrial=E2=80=9D, where that consumer device= (e.g., an Apple iDevice) might be a User Product in the terms of GPLv3. and thus 3) merely compiling the app with GPLv3-with-RLEv3.1-licensed compiler (and = the app invoking at least one subroutine in the RLEv3.1-licensed runtime) c= ould conceivably be enough for Apple to reject such apps from the App Store= , so that Apple need not satisfy #2 above. https://thenextweb.com/apple/2015/06/17/apples-biggest-developer-news-at-ww= dc-that-nobodys-talking-about-bitcode Apparently Apple is either working on a McSema-esque technology or would re= quire the app to be (also?) provided in the form of LLVM-bitcode-IR (in add= ition? to ARM-processor machine-code) at the time of submission to the App = Store for approval. Dang it. The possibility of sketchiness/dodginess in this vicinity of FSF = GCC licensing (regarding IR tools outside of my control, downstream from me= complicating my life upstream) erodes part of my motivation for GELI as pr= oposed earlier in a branch of this thread GELI project: (i.e., the GPLv3-with-RLEv3.1-licensed GNAT-with-LLVM-&-Polly-backend), whe= re GELI would be an LSP peer of the GIGI tree-transducer in FSF GNAT. When= an =E2=80=9Cllvm-=E2=80=9D-prefixed target name is on the GNAT command lin= e, GELI would shunt* off to LLVM-world from the point of the Ada iterator/c= ursor walking GNAT's fully-semantically-adorned Ada-IR tree. (Non-=E2=80= =9Cllvm-=E2=80=9D-prefixed targets on the GNAT command line would utilize t= he traditional GIGI-GENERIC-highGIMPLE-lowGIMPLE-RTL backend.) * all within one GCC executable without writing out any IR to files, so as = to comply with my stricter interpretation of the terms of RLEv3.1 intertwin= ed with the terms of GPLv3. One of my prime motivators for GELI and its GPLv3-with-RLEv3.1-licensed LLV= M-backended FSF GNAT was to utilize Ada2012 to write iDevice apps for submi= ssion to the App Store (since on-nonjailbroken-iDevice debugging by gdb is = disallowed nowadays due to gdb's lack of codesigning after Apple's switchov= er to LLVM-world toolchain). Dang it. Hmmmmm. Must rethink how to help Byron's generation of LLVM bi= tcode IR arrive sooner.