Hi Everyone, LLVM is a growing project, and many of us are very fond of it. :) LLVM is continuing to grow, both in maturity in specific areas and in scope of areas that it is applicable to. When we first started the project, we focused on the design of the intermediate representation. It is a strong design goal that the IR be a self-contained virtual instruction set, which fully describes the program. Because of this, we named the compiler LLVM, which reflects well on the design of the IR. However, the scope of the LLVM project is outgrowing this name. Today LLVM does many "non-VMy" tasks, such as serving as a great static compiler. It also has components that overlap with traditional low-level tool chain components like assemblers and linkers. Further, LLVM's scope is about to grow significantly with new front-end technologies (e.g. HLVM, new SoC work on a python front-end, etc). For all of these reasons, I think that "LLVM" is an increasingly poor name for the project as a whole, and it causes a large amount of confusion, particularly with people who do not know much about it yet. For what it is worth, this is not a new thought. I have been kicking around the idea of renaming the project for several years now, but have been stymied by not being able to come up with a better name! The problem is hard: how do you concisely describe a modern, modular, component based compiler and tool-chain system, which can be used for many different things, hopefully many of which we haven't even thought of yet? How do you pick a name that both memberable, relatively unique (searchable), has an open domain name, etc? How do you come up with a name that is amenable to making a logo? So far, I haven't! :) Note that the name need not capture every aspect of the project. Just having a distinguished name with no specific connotation is probably good enough. 20 years ago, "google" and "yahoo" had very different meanings, and "mozilla" or "firefox" were pretty meaningless. Today, there is very strong awareness of what they are. As such, I'd like to open up a forum for naming ideas. I think we need to continue to use the name 'LLVM' to refer to the IR (i.e. that which is described in http://llvm.org/docs/LangRef.html), and we should plan to attach LLVM as a suffix to the project name for several years to come: For example: "Use the Foo/LLVM Compiler System, it runs infinite loops faster than the competition!". To make this more fun, "success" is extremely subjective, and I have no idea how we will declare a victor (we can figure it out as we go, right? :). I propose that people add ideas to the wiki that Vikram and Reid are getting set up, and we kick some potential names around at the developer mtg in May. That gives us 5 or 6 weeks to come up with a name and/or Logo. The wiki isn't up yet, so consider this to be a head start :). If we find a good name, we can honor the one who came up with it with a small amount of booty. For example, we're prepared to award a compiler textbook of your choice (MSRP < $100) to the winner, optionally signed by people at the dev meeting if you so desire and if they agree. Alternatively, we could do a $100 gift card to Amazon or some other place. What say you? Are you up for the challenge? -Chris p.s., no I don't think "Foo Compiler System" is a good choice. ;-) -- http://nondot.org/sabre/ http://llvm.org/
Chris Lattner wrote:> As such, I'd like to open up a forum for naming ideas. I think we need to > continue to use the name 'LLVM' to refer to the IR (i.e. that which is > described in http://llvm.org/docs/LangRef.html), and we should plan to > attach LLVM as a suffix to the project name for several years to come: For > example: "Use the Foo/LLVM Compiler System, it runs infinite loops faster > than the competition!". >How about Illuvium? It's pronounceable, it pays homage to "LLVM", and the domain's already taken by someone who might be persuaded to give it up :)
On Wed, 11 Apr 2007, Jeff Cohen wrote:> How about Illuvium? It's pronounceable, it pays homage to "LLVM", and > the domain's already taken by someone who might be persuaded to give it > up :)a non goal: sounding like the old name. :) -Chris -- http://nondot.org/sabre/ http://llvm.org/
On Wed, 11 Apr 2007, Chris Lattner wrote:> However, the scope of the LLVM project is outgrowing this name. > Today LLVM does many "non-VMy" tasks, such as serving as a great > static compiler. It also has components that overlap with > traditional low-level tool chain components like assemblers and > linkers. Further, LLVM's scope is about to grow significantly > with new front-end technologies (e.g. HLVM, new SoC work on a > python front-end, etc). For all of these reasons, I think that > "LLVM" is an increasingly poor name for the project as a whole, > and it causes a large amount of confusion, particularly with > people who do not know much about it yet.Oh yeah, please avoid 'VM' in the name. :) -Chris -- http://nondot.org/sabre/ http://llvm.org/
My suggestion: "Warlock" or "Warloc" It can be thought of as standing for "We aren't like other compilers," but that doesn't have to be pointed out, as it's also a word on its own. It has a sense of being something powerful and/or magical, and should be feasible to make a logo to go along with. --Owen
lcs for language compiling system? It's a suggestive of the old name, easy to remember, easy to say, and seems to be easy to google. Advanced could be prepended to it to make it alcs to make it even more unique. -e
OmniC or Omnipiler or Omnicom -- stands for "Omnipotent Compiler" I prefer the first. On 4/12/07, Erick Tryzelaar <erickt at dslextreme.com> wrote:> > lcs for language compiling system? It's a suggestive of the old name, > easy to remember, easy to say, and seems to be easy to google. Advanced > could be prepended to it to make it alcs to make it even more unique. > > -e > _______________________________________________ > LLVM Developers mailing list > LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20070412/9e6f5693/attachment.html>
BTW, In case I didn't mention it before, the names are being kept track of here: http://llvm.org/DevMtgMay2007.html#names There's also some ideas for name generation there. Thanks for all the recent submissions! Reid. On Wed, 2007-04-11 at 21:44 -0700, Chris Lattner wrote:> Hi Everyone, > > LLVM is a growing project, and many of us are very fond of it. :) LLVM > is continuing to grow, both in maturity in specific areas and in scope of > areas that it is applicable to. > > When we first started the project, we focused on the design of the > intermediate representation. It is a strong design goal that the IR be a > self-contained virtual instruction set, which fully describes the program. > Because of this, we named the compiler LLVM, which reflects well on the > design of the IR. > > However, the scope of the LLVM project is outgrowing this name. Today > LLVM does many "non-VMy" tasks, such as serving as a great static > compiler. It also has components that overlap with traditional low-level > tool chain components like assemblers and linkers. Further, LLVM's scope > is about to grow significantly with new front-end technologies (e.g. HLVM, > new SoC work on a python front-end, etc). For all of these reasons, I > think that "LLVM" is an increasingly poor name for the project as a whole, > and it causes a large amount of confusion, particularly with people who > do not know much about it yet. > > For what it is worth, this is not a new thought. I have been kicking > around the idea of renaming the project for several years now, but have > been stymied by not being able to come up with a better name! The problem > is hard: how do you concisely describe a modern, modular, component based > compiler and tool-chain system, which can be used for many different > things, hopefully many of which we haven't even thought of yet? How do > you pick a name that both memberable, relatively unique (searchable), has > an open domain name, etc? How do you come up with a name that is amenable > to making a logo? So far, I haven't! :) > > Note that the name need not capture every aspect of the project. Just > having a distinguished name with no specific connotation is probably good > enough. 20 years ago, "google" and "yahoo" had very different meanings, > and "mozilla" or "firefox" were pretty meaningless. Today, there is very > strong awareness of what they are. > > > As such, I'd like to open up a forum for naming ideas. I think we need to > continue to use the name 'LLVM' to refer to the IR (i.e. that which is > described in http://llvm.org/docs/LangRef.html), and we should plan to > attach LLVM as a suffix to the project name for several years to come: For > example: "Use the Foo/LLVM Compiler System, it runs infinite loops faster > than the competition!". > > To make this more fun, "success" is extremely subjective, and I have no > idea how we will declare a victor (we can figure it out as we go, right? > :). I propose that people add ideas to the wiki that Vikram and Reid are > getting set up, and we kick some potential names around at the developer > mtg in May. That gives us 5 or 6 weeks to come up with a name and/or > Logo. The wiki isn't up yet, so consider this to be a head start :). > > If we find a good name, we can honor the one who came up with it with a > small amount of booty. For example, we're prepared to award a compiler > textbook of your choice (MSRP < $100) to the winner, optionally signed by > people at the dev meeting if you so desire and if they agree. > Alternatively, we could do a $100 gift card to Amazon or some other place. > > What say you? Are you up for the challenge? > > -Chris > > p.s., no I don't think "Foo Compiler System" is a good choice. ;-) >
Mithril The fictional metal from JRR Tolkien The Lord of the Rings. Gandalf says: "Mithril! All folk desired it. It could be beaten like copper, and polished like glass; and the Dwarves could make of it a metal, light and yet harder than tempered steel. Its beauty was like to that of common silver, but the beauty of mithril did not tarnish or grow dim." Paolo Invernizzi On Apr 12, 2007, at 6:44 AM, Chris Lattner wrote:> Hi Everyone, > > LLVM is a growing project, and many of us are very fond of it. :) > LLVM > is continuing to grow, both in maturity in specific areas and in > scope of > areas that it is applicable to. > > When we first started the project, we focused on the design of the > intermediate representation. It is a strong design goal that the > IR be a > self-contained virtual instruction set, which fully describes the > program. > Because of this, we named the compiler LLVM, which reflects well on > the > design of the IR. > > However, the scope of the LLVM project is outgrowing this name. Today > LLVM does many "non-VMy" tasks, such as serving as a great static > compiler. It also has components that overlap with traditional low- > level > tool chain components like assemblers and linkers. Further, LLVM's > scope > is about to grow significantly with new front-end technologies > (e.g. HLVM, > new SoC work on a python front-end, etc). For all of these reasons, I > think that "LLVM" is an increasingly poor name for the project as a > whole, > and it causes a large amount of confusion, particularly with people > who > do not know much about it yet. > > For what it is worth, this is not a new thought. I have been kicking > around the idea of renaming the project for several years now, but > have > been stymied by not being able to come up with a better name! The > problem > is hard: how do you concisely describe a modern, modular, component > based > compiler and tool-chain system, which can be used for many different > things, hopefully many of which we haven't even thought of yet? > How do > you pick a name that both memberable, relatively unique > (searchable), has > an open domain name, etc? How do you come up with a name that is > amenable > to making a logo? So far, I haven't! :) > > Note that the name need not capture every aspect of the project. Just > having a distinguished name with no specific connotation is > probably good > enough. 20 years ago, "google" and "yahoo" had very different > meanings, > and "mozilla" or "firefox" were pretty meaningless. Today, there > is very > strong awareness of what they are. > > > As such, I'd like to open up a forum for naming ideas. I think we > need to > continue to use the name 'LLVM' to refer to the IR (i.e. that which is > described in http://llvm.org/docs/LangRef.html), and we should plan to > attach LLVM as a suffix to the project name for several years to > come: For > example: "Use the Foo/LLVM Compiler System, it runs infinite loops > faster > than the competition!". > > To make this more fun, "success" is extremely subjective, and I > have no > idea how we will declare a victor (we can figure it out as we go, > right? > :). I propose that people add ideas to the wiki that Vikram and > Reid are > getting set up, and we kick some potential names around at the > developer > mtg in May. That gives us 5 or 6 weeks to come up with a name and/or > Logo. The wiki isn't up yet, so consider this to be a head start :). > > If we find a good name, we can honor the one who came up with it > with a > small amount of booty. For example, we're prepared to award a > compiler > textbook of your choice (MSRP < $100) to the winner, optionally > signed by > people at the dev meeting if you so desire and if they agree. > Alternatively, we could do a $100 gift card to Amazon or some other > place. > > What say you? Are you up for the challenge? > > -Chris > > p.s., no I don't think "Foo Compiler System" is a good choice. ;-) > > -- > http://nondot.org/sabre/ > http://llvm.org/ > _______________________________________________ > LLVM Developers mailing list > LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20070412/ae8fc379/attachment.html>
Hi The Dragon book, led my thoughts to "Here be dragons" http://en.wikipedia.org/wiki/Here_be_dragons "In another context, software programmers sometimes use it to indicate especially difficult or obscure sections of code in a program so that others do not tamper with them." Why not some dragon name ? or maybe someone can use this idea to come up with something else. /f Paolo Invernizzi wrote:> Mithril > > The fictional metal from JRR Tolkien The Lord of the Rings. > > Gandalf says: > > "Mithril! All folk desired it. It could be beaten like copper, and > polished like glass; and the Dwarves could make of it a metal, light and > yet harder than tempered steel. Its beauty was like to that of common > silver, but the beauty of mithril did not tarnish or grow dim." > > Paolo Invernizzi
On Apr 11, 2007, at 9:44 PM, Chris Lattner wrote:> I think we need to > continue to use the name 'LLVM' to refer to the IR (i.e. that which is > described in http://llvm.org/docs/LangRef.html),Makes sense.> and we should plan to > attach LLVM as a suffix to the project name for several years to > come: For > example: "Use the Foo/LLVM Compiler System, it runs infinite loops > faster > than the competition!".I think, it is a good idea to have clean separation and avoid adding suffix. In today's internet age, dots between 'Foo' and LLVM are connected immediately. People will know that Foo was formerly known as LLVM. Plus as soon as they read/learn about 'Foo' they will realize that it uses IR named LLVM :) - Devang
On Wed, 11 Apr 2007 21:53:15 -0700 Jeff Cohen <jeffc at jolt-lang.org> wrote:> > How about Illuvium? It's pronounceable, it pays homage to "LLVM", and > the domain's already taken by someone who might be persuaded to give it > up :)I like it :) Simon.
On 4/12/07, Chris Lattner <sabre at nondot.org> wrote:> Note that the name need not capture every aspect of the project. Just > having a distinguished name with no specific connotation is probably good > enough. 20 years ago, "google" and "yahoo" had very different meanings, > and "mozilla" or "firefox" were pretty meaningless. Today, there is very > strong awareness of what they are. >That works great for something that gets massive exposure, but GNOME has made me wary of (what seems to be) completely unrelated names. I too often get lost in the soup of pango, cairo, epiphany, evolution, metacity, festival, evince, and such, so I'm not a fan of clever, but unrelated, words. A portmanteau could be nice, though. One of the nicer project names I've seen recently is Alexandria, for a book database program ( http://alexandria.rubyforge.org/ ). It unfortunately fails the searchability test, but does brilliantly at reminding you what it is. What about heading off into Greek mythology? Perhaps Dædalus, the greatly skilled Athenian architect that built the palace of Cnossus. There's also Hephæstus, God of smiths, builder of Helios's chariot, maker of Talos, Crete and Europa's guardian robot, and fabricator of the weapons of the Gods; or Lemnos, the island with his workshop. Unfortunately, they're not the easiest words to spell, and Vulcan, Hephæstus's roman name, is somewhat used already. ~ Scott McMurray
On Apr 12, 2007, at 2:11 PM, me22 wrote:> On 4/12/07, Chris Lattner <sabre at nondot.org> wrote: >> Note that the name need not capture every aspect of the project. >> Just >> having a distinguished name with no specific connotation is >> probably good >> enough. 20 years ago, "google" and "yahoo" had very different >> meanings, >> and "mozilla" or "firefox" were pretty meaningless. Today, there >> is very >> strong awareness of what they are. >> > That works great for something that gets massive exposure, but GNOME > has made me wary of (what seems to be) completely unrelated names. I > too often get lost in the soup of pango, cairo, epiphany, evolution, > metacity, festival, evince, and such, so I'm not a fan of clever, but > unrelated, words. A portmanteau could be nice, though. > > One of the nicer project names I've seen recently is Alexandria, for a > book database program ( http://alexandria.rubyforge.org/ ). It > unfortunately fails the searchability test, but does brilliantly at > reminding you what it is. > > What about heading off into Greek mythology? Perhaps Dædalus, the > greatly skilled Athenian architect that built the palace of Cnossus. > There's also Hephæstus, God of smiths, builder of Helios's chariot, > maker of Talos, Crete and Europa's guardian robot, and fabricator of > the weapons of the Gods; or Lemnos, the island with his workshop.We could combine dragons and mythology (but in this case Norse) and get Jörmungandr (the Midgard Serpent). This has the advantage of non- ascii characters AND the fact that I only have a vague concept of how to pronounce it :)> > Unfortunately, they're not the easiest words to spell, and Vulcan, > Hephæstus's roman name, is somewhat used already. > > ~ Scott McMurray > > _______________________________________________ > LLVM Developers mailing list > LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
All, We're now keeping track of the naming process, ideas, etc. on this page: http://llvm.org/Name.html (would someone PLEASE get us a wiki!!!) ;) Reid. On Wed, 2007-04-11 at 21:44 -0700, Chris Lattner wrote:> Hi Everyone, > > LLVM is a growing project, and many of us are very fond of it. :) LLVM > is continuing to grow, both in maturity in specific areas and in scope of > areas that it is applicable to. > > When we first started the project, we focused on the design of the > intermediate representation. It is a strong design goal that the IR be a > self-contained virtual instruction set, which fully describes the program. > Because of this, we named the compiler LLVM, which reflects well on the > design of the IR. > > However, the scope of the LLVM project is outgrowing this name. Today > LLVM does many "non-VMy" tasks, such as serving as a great static > compiler. It also has components that overlap with traditional low-level > tool chain components like assemblers and linkers. Further, LLVM's scope > is about to grow significantly with new front-end technologies (e.g. HLVM, > new SoC work on a python front-end, etc). For all of these reasons, I > think that "LLVM" is an increasingly poor name for the project as a whole, > and it causes a large amount of confusion, particularly with people who > do not know much about it yet. > > For what it is worth, this is not a new thought. I have been kicking > around the idea of renaming the project for several years now, but have > been stymied by not being able to come up with a better name! The problem > is hard: how do you concisely describe a modern, modular, component based > compiler and tool-chain system, which can be used for many different > things, hopefully many of which we haven't even thought of yet? How do > you pick a name that both memberable, relatively unique (searchable), has > an open domain name, etc? How do you come up with a name that is amenable > to making a logo? So far, I haven't! :) > > Note that the name need not capture every aspect of the project. Just > having a distinguished name with no specific connotation is probably good > enough. 20 years ago, "google" and "yahoo" had very different meanings, > and "mozilla" or "firefox" were pretty meaningless. Today, there is very > strong awareness of what they are. > > > As such, I'd like to open up a forum for naming ideas. I think we need to > continue to use the name 'LLVM' to refer to the IR (i.e. that which is > described in http://llvm.org/docs/LangRef.html), and we should plan to > attach LLVM as a suffix to the project name for several years to come: For > example: "Use the Foo/LLVM Compiler System, it runs infinite loops faster > than the competition!". > > To make this more fun, "success" is extremely subjective, and I have no > idea how we will declare a victor (we can figure it out as we go, right? > :). I propose that people add ideas to the wiki that Vikram and Reid are > getting set up, and we kick some potential names around at the developer > mtg in May. That gives us 5 or 6 weeks to come up with a name and/or > Logo. The wiki isn't up yet, so consider this to be a head start :). > > If we find a good name, we can honor the one who came up with it with a > small amount of booty. For example, we're prepared to award a compiler > textbook of your choice (MSRP < $100) to the winner, optionally signed by > people at the dev meeting if you so desire and if they agree. > Alternatively, we could do a $100 gift card to Amazon or some other place. > > What say you? Are you up for the challenge? > > -Chris > > p.s., no I don't think "Foo Compiler System" is a good choice. ;-) >
me22 wrote:> One of the nicer project names I've seen recently is Alexandria, for a > book database program ( http://alexandria.rubyforge.org/ ). It > unfortunately fails the searchability test, but does brilliantly at > reminding you what it is.Along these lines, is there any mythical characters or historical persons which are associated with translation (which is the primary role of a compiler framework)? I can only think of 'Babel', but I'm sure there must be some other possibilities... m.
> We're now keeping track of the naming process, ideas, etc. on this page: > > http://llvm.org/Name.html > > (would someone PLEASE get us a wiki!!!) ;)I can do a Media Wiki on our datacomms.net server for LLVM thats if :- a) You can not host one at UIUC b) Someone with prior Wiki experience cannot host one. Otherwise I would be willing to do the learning to host one. Aaron
> On Apr 11, 2007, at 9:44 PM, Chris Lattner wrote: > >> I think we need to >> continue to use the name 'LLVM' to refer to the IR (i.e. that >> which is >> described in http://llvm.org/docs/LangRef.html),On the wackier side of "things that don't mean what they seem": CosaNostra ("our thing"). All seriousness aside, I liked Illiacc (+1 for that one from me.) I'd also thought "Blastula" might also be descriptive of the current and future developing state of llvm. -scooter
[Apologies in advance for the train of thought prose, but it is brainstorming after all…] I'm going to focus on self-descriptive names rather than literary or fantasy references… Advanced Compiler Kit, affectionately known as “ACK!”? It has an ill- deserved nod to NeXT, even. (Completely the wrong language, after all.) Core Compiler? Heh, I don't think that'd get past certain influential marketing departments. Nor is this a C API, but that didn't stop Core Data or Core Image, so y'never know. Veloce, schnell, rapido, rápide. Various translations of fast. Schnell is fun, but to an English speaker it has an imperative sense to it that makes it useless as an adjective, IMO. Veloce has a nice ring to it, though; that's Italian. Veloce Code? Velocicode? I think the kit-core-foundation-framework concept is at least as important to the project as is anything else. In fact, if the name can communicate 'framework for fast', then the 'compiler' or 'code' concept falls out naturally (within the target audience). Here are some translations of several related phrases… Kit rapide. Base rapide. Noyau rapide. Code rapide. Compilateur rapide. Cadre rapide. (French) Schneller Installationssatz. Schneller Kern. Schnelle Grundlage. Schneller Code. Schneller Compiler. Schneller Rahmen. (German) Corredo veloce. Fondamento veloce. Nucleo veloce. Codice veloce. Compilatore veloce. Struttura veloce. (Italian) Kit rápido. Fundación rápida. Base rápida. Código rápido. Recopilador rápido. Marco rápido. (Spanish) The only one that strikes me is the Italian nucleo veloce (fast core); the others aren't comprehensible to an English speaker. But maybe these can give others some ideas. corecompiler, velocecode, velocicode, and nucleoveloce dot org are all available. ack.org is of course not. — Gordon -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20070416/3f490837/attachment.html>
Gordon Henriksen wrote:> Veloce, schnell, rapido, rápide. Various translations of fast."Nopsa", Finnish for rapid. -- Pertti
Hi Gordon, Good ideas here. If you want these included, please provide a patch against the Name.html page in the llvm-www repository. Thanks. Reid. On Mon, 2007-04-16 at 15:13 -0400, Gordon Henriksen wrote:> [Apologies in advance for the train of thought prose, but it is > brainstorming after all…] > > > I'm going to focus on self-descriptive names rather than literary or > fantasy references… > > > > > Advanced Compiler Kit, affectionately known as “ACK!”? It has an > ill-deserved nod to NeXT, even. (Completely the wrong language, after > all.) > > > Core Compiler? Heh, I don't think that'd get past certain influential > marketing departments. Nor is this a C API, but that didn't stop Core > Data or Core Image, so y'never know. > > > > > Veloce, schnell, rapido, rápide. Various translations of fast. Schnell > is fun, but to an English speaker it has an imperative sense to it > that makes it useless as an adjective, IMO. Veloce has a nice ring to > it, though; that's Italian. Veloce Code? Velocicode? > > > > > I think the kit-core-foundation-framework concept is at least as > important to the project as is anything else. In fact, if the name can > communicate 'framework for fast', then the 'compiler' or 'code' > concept falls out naturally (within the target audience). Here are > some translations of several related phrases… > > > Kit rapide. Base rapide. Noyau rapide. Code rapide. Compilateur > rapide. Cadre rapide. (French) > Schneller Installationssatz. Schneller Kern. Schnelle Grundlage. > Schneller Code. Schneller Compiler. Schneller Rahmen. (German) > Corredo veloce. Fondamento veloce. Nucleo veloce. Codice veloce. > Compilatore veloce. Struttura veloce. (Italian) > Kit rápido. Fundación rápida. Base rápida. Código rápido. Recopilador > rápido. Marco rápido. (Spanish) > > > The only one that strikes me is the Italian nucleo veloce (fast core); > the others aren't comprehensible to an English speaker. But maybe > these can give others some ideas. > > > > > corecompiler, velocecode, velocicode, and nucleoveloce dot org are all > available. ack.org is of course not. > > > — Gordon > > > > > > _______________________________________________ > LLVM Developers mailing list > LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev