Hi,
In http://reviews.llvm.org/D10969, Eric asked me to explain the wider context of
the TargetTuple object that was replacing Triple on llvmdev so here it is.
Before I start, I'm sure I don't know the full extent of GNU triple
ambiguity and lack of canonicity. Additional examples are welcome.
The Problem
As you know, LLVM uses a GNU Triple is as a target description that can be
relied upon to make decisions. It's used for various decisions such as the
default cpu, the alignment of types, the object format, the names for libcalls,
and a wide variety of others.
In using it like this, LLVM assumes that triples are unambiguous and have a
specific defined meaning. Unfortunately, this assumption fails for a number of
reasons.
The first reason is that compiler options can overrule the triple but leave it
unchanged. For example, in GCC mips-linux-gnu-gcc normally produces 32-bit
MIPS-I output using the O32 ABI, but 'mips-linux-gnu-gcc -mips64'
normally produces 64-bit MIPS-III output using the N32 ABI. Like GCC, compiler
options to mips-linux-gnu-clang should (and mostly do but MIPS has a few
crashing cases caused by triple misuse) overrule the triple. However, we
don't mutate the triple to reflect this so any decisions based on the
overridable state cannot rely on the triple to accurately reflect the desired
behaviour.
It's worth mentioning here that some targets have hacks to partially mutate
the triple in clang to work around issues they would otherwise have in the
backend but this is done on an ad-hoc basis for specific details (e.g. mips
<-> mipsel for -EL and -EB).
The second reason is that there is no canonical meaning for a given GNU Triple,
it varies between vendors and over time. There is also no requirement for
vendors to have a unique GNU Triple for their toolchain. For GCC, it's
fairly common for distributors to change the meanings of triples using options
like --with-arch, --with-cpu, --with-abi, etc. There are also some
target-specific options such as --with-mode to select ARM/Thumb by default and
--with-nan for MIPS NAN encoding selection. Different vendors use different
configure options and may change them at will. When they do change them, the
vendors often desire to keep the same triple to be able to drop in the new
version without causing wider impact on their environment. For example, assuming
I'm reading debian/rules2 for Debian's gcc-4.9 package correctly then
the i386-linux-gnu means i486 on Debian Etch and Lenny but means i586 on more
recent versions. On a similar note, on Debian, mips-linux-gnu targets MIPS-II
(optimised for typical MIPS32 implementations) rather than the usual MIPS-I. The
last example of this ambiguity I'd like to reference is that mentioned by
https://wiki.debian.org/Multiarch/Tuples#Why_not_use_GNU_triplets.3F. In that
example, hard-float and soft-float on ARM both used arm-linux-gnueabi but were
mutually incompatible. The Multiarch tuples described on that page are an
attempt to resolve the ambiguity but I'm told that they aren't likely to
be universally adopted.
The third reason, is that different triples can mean the same thing. Jim
Grosbach has mentioned that the prefixes of the GNU Triple are different between
Linux and Darwin for ARM despite sharing the same meaning (presumably subject to
the issues above). As a result decisions based on the string have to take care
of multiple possible values. Mips has a similar issue too since a host triple
(and therefore default target triple) of mips64-linux-gnu needs to behave like
mips-linux-gnu on a 32-bit Mips port of Debian.
Although not included in the description of the assumption above, one additional
flaw in the use of GNU Triples is that they are sometimes inadequate as a
description of the target. One example affecting MIPS in particular is that the
ABI is not represented in the GNU Triple we require significant API changes to
get this information where we need it. It would be helpful to be able to pass
such information through the existing plumbing.
The Planned Solution
The plan is to split the GNU Triple represented by the llvm::Triple object into
two pieces. The first piece is the existing llvm::Triple and is responsible for
parsing the GNU triple and canonicalizing it. The second piece is a mutable
target description named llvm::TargetTuple. TargetTuple is responsible for
interpreting the triple according to the vendor's rules, providing an
interface to allow mutation by tools, and authoritatively defining the target
being targeted without the ambiguity of GNU Triples. As an example,
'mips-linux-gnu-clang -EL ...' would:
// Parse the GNU Triple
llvm::Triple GnuTriple("mips-linux-gnu");
// Convert it to a TargetTuple according to the (possibly customized) meanings
in
// use by the vendor.
llvm::TargetTuple TT(GnuTriple);
// Then mutate the TargetTuple according to the compiler options (or equivalent
depending
// on the tool, for example disassemblers would mutate it according to the
object headers).
if (hasOption("-EL"))
TT.setLittleEndian()
...
At this point, TT would be
"+mipsel-unknown-linux-gnu-elf32-some-other-stuff" (exact
serialization is t.b.d and may end up target dependent) which we can then rely
on in the rest of LLVM. This split resolves the issue of llvm::Triple objects
not being reliable when used as a target description since TargetTuple will
reflect the result of interpreting the triple as well as applying appropriate
options. It also provides a suitable place for vendors to define the meanings of
their GNU Triples.
One significant detail is the way vendors customize the meaning of their
Triples. Currently, the plan is to nominate a constructor
(TargetTuple::TargetTuple(const Triple &)) a vendor can patch to redefine
their triples with the default implementation being the 'usual' meaning
(the meaning that should be used in the absence of customization). One nice
benefit of this configure-by-source-patch approach is that vendors can customize
multiple triples as easily as their native triple or intended target triple. To
use Debian as an example again, they would be able to customize all their
supported triples such that 'clang -target arm-linux-gnueabihf' on the
amd64 port targets their armhf port using the same customization that makes
'clang' on the armhf port do the right thing natively. Android, and
toolchains for heterogenous platform would likely benefit from this too. This
configure-by-source-patch approach seems to make some people uncomfortable so we
may have to find another way to configure the triples (tablegen?).
To reach this result the plan is to do the following:
1. Replace any remaining std::string's and StringRef's containing
GNU triples with Triple objects.
2. Split the llvm::Triple class into llvm::Triple and llvm::TargetTuple
classes. Both are identical in implementation and almost identical in interface
at this stage.
3. Gradually replace Triples with TargetTuples until the C APIs and the
LLVM-IR are the only place inside LLVM where Triples are still used.
4. Change the implementation of TargetTuple to whatever is convenient for
LLVM's internals and decide on a serialization.
5. Replace serialized Triples with serialized TargetTuples in LLVM-IR.
a. Maintain backwards compatibility with IR using triples, at least for a
while.
6. Add TargetTuple support to the C API. Exact API is t.b.d.
7. Have the API users mutate the TargetTuple appropriately.
Renato: This has been revised slightly from the last one we discussed due to
public C++ API's being used internally as well as externally.
Where we are now
I've just started posting patches for step 2 and 3 of the plan. My working
copy is nearly at step 4.
What's next
Upstream step 2 and 3 and then begin replacing the TargetTuple implementation as
per step 4.
Previous Discussions
http://thread.gmane.org/gmane.comp.compilers.llvm.devel/86020/focus=86073. I
should mention that I've since been made aware that the original topic of
private label prefixes could be solved in a much simpler way than previously
thought. The triple related discussion is still relevant though.
I understand from Renato that there are more threads over the last few years but
I haven't looked for them.
Daniel Sanders
Leading Software Design Engineer, MIPS Processor IP
Imagination Technologies Limited
www.imgtec.com<http://www.imgtec.com/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20150708/abf1fac8/attachment.html>
On 8 July 2015 at 15:31, Daniel Sanders <Daniel.Sanders at imgtec.com> wrote:> The first reason is that compiler options can overrule the triple but leave > it unchanged. For example, in GCC mips-linux-gnu-gcc normally produces > 32-bit MIPS-I output using the O32 ABI, but 'mips-linux-gnu-gcc –mips64' > normally produces 64-bit MIPS-III output using the N32 ABI. Like GCC, > compiler options to mips-linux-gnu-clang should (and mostly do but MIPS has > a few crashing cases caused by triple misuse) overrule the triple. However, > we don't mutate the triple to reflect this so any decisions based on the > overridable state cannot rely on the triple to accurately reflect the > desired behaviour.Another very annoying fact is that the Clang driver re-parses triples many times, and sometimes they change the triple based on a CPU, and then end up with a different CPU. There was a bug that if you passed "thumbv7", it would not recognise, pick "ARM7TDMI" CPU, and later change the triple to "armv4t" because of that, and pass *that* to the internal processes (GAS, IAS, linker). This bug has been fixed by adding "thumb" to it, but the underlying reason it happened means there are plenty of other similar bugs waiting to happen. We need to fix the mechanism in which we understand targets, and having an unambiguous description that spans across *all* LLVM projects (including Clang, LLD, LLDB) and tools (llc, lli, llvm-mc, etc) is the ultimate goal. Most targets do not have those problems, but Mips and ARM are a big mess. That's why we're so interested in making that happen.> This configure-by-source-patch approach seems to make > some people uncomfortable so we may have to find another way to configure > the triples (tablegen?).Another option that would make *all* distributions happy would be to adopt the same approach as GCC and have CMake options for default ABI choices. This would be harder to implement, but we can hide the mess under a separate class (TargetABI?). I actually prefer this solution to either tablegen or patch-sets.> a. Maintain backwards compatibility with IR using triples, at least > for a while.Probably forever... :( Again, this can be isolated in TargetABI or some other place.> Renato: This has been revised slightly from the last one we discussed due to > public C++ API's being used internally as well as externally.Roger. I'd also wanted to add the TargetParser changes and how this all fit into the plan. TargetParser is a target-specific class that knows how to parse strings and convert to platform choices (and back) in a way that is both unambiguous and ubiquitous. Meaning, all tools and projects should use the TargetParser to understand what are the platform specific options, what do they mean and how are they related to each other. The TargetTuple is the perfect companion, since we can use the TargetParser to understand triples, cpu/fpu names, extensions, etc, and related them inside the Tuple in an ABI-specific way (via the ABI modifiers). Right now, both Clang and LLVM ARM assembler are using the ARMTargetParser, but the idea is to expand this to the Triple when TargetTuple beings Daniel's plan step #4 onwards. We'll probably have a MIPSTargetParser, too. And probably refactor the hierarchy of that, to make sure we can get polymorphism and so on. But I wanted to keep-it-simple-and-stupid before we had a reason not to.> I understand from Renato that there are more threads over the last few years > but I haven't looked for them.There were numerous discussions about the driver complicated structure, never-ending bugs, parsing mismatches and triple shenanigans over the last 5 years. I couldn't possibly link all of them here. :) Specifically between Mips and ARM, I think me and Reed had a few specific discussions a few years ago, but I can't seem to find them. That was probably 2010/2011. cheers, --renato
> -----Original Message----- > From: Renato Golin [mailto:renato.golin at linaro.org] > Sent: 08 July 2015 16:09 > To: Daniel Sanders > Cc: LLVM Developers Mailing List (llvmdev at cs.uiuc.edu); Eric Christopher > (echristo at gmail.com); Jim Grosbach (grosbach at apple.com) > Subject: Re: The Trouble with Triples > > On 8 July 2015 at 15:31, Daniel Sanders <Daniel.Sanders at imgtec.com> wrote: > > The first reason is that compiler options can overrule the triple but leave > > it unchanged. For example, in GCC mips-linux-gnu-gcc normally produces > > 32-bit MIPS-I output using the O32 ABI, but 'mips-linux-gnu-gcc –mips64' > > normally produces 64-bit MIPS-III output using the N32 ABI. Like GCC, > > compiler options to mips-linux-gnu-clang should (and mostly do but MIPS > has > > a few crashing cases caused by triple misuse) overrule the triple. However, > > we don't mutate the triple to reflect this so any decisions based on the > > overridable state cannot rely on the triple to accurately reflect the > > desired behaviour. > > Another very annoying fact is that the Clang driver re-parses triples > many times, and sometimes they change the triple based on a CPU, and > then end up with a different CPU. > > There was a bug that if you passed "thumbv7", it would not recognise, > pick "ARM7TDMI" CPU, and later change the triple to "armv4t" because > of that, and pass *that* to the internal processes (GAS, IAS, linker). > > This bug has been fixed by adding "thumb" to it, but the underlying > reason it happened means there are plenty of other similar bugs > waiting to happen. We need to fix the mechanism in which we understand > targets, and having an unambiguous description that spans across *all* > LLVM projects (including Clang, LLD, LLDB) and tools (llc, lli, > llvm-mc, etc) is the ultimate goal. > > Most targets do not have those problems, but Mips and ARM are a big > mess. That's why we're so interested in making that happen.This reminded me of something I noticed in passing and haven't investigated yet. I think TargetRegistry::lookupTarget(const std::string &ArchName, Triple&, std::string&) can do this kind of thing when ArchName is a registered architecture but is not understood by Triple::getArchTypeForLLVMName(). I haven't had chance to try it but it looks like you'd get the intended target architecture without updating the triple to match (so x86_64-linux-gnu could target a completely different architecture).> > This configure-by-source-patch approach seems to make > > some people uncomfortable so we may have to find another way to > configure > > the triples (tablegen?). > > Another option that would make *all* distributions happy would be to > adopt the same approach as GCC and have CMake options for default ABI > choices. > > This would be harder to implement, but we can hide the mess under a > separate class (TargetABI?). I actually prefer this solution to either > tablegen or patch-sets.I can see a way to make the CMake option approach work nicely for native. The constructor can check for the default triple and apply the effects of the CMake options to it. I don't think there's a good way to support the multiple triples or heterogenous use cases via CMake options but support for that was more a happy coincidence rather than intentional design. If we take this route, I'm hoping I don't need to do autoconf since I don't know it very well.> > a. Maintain backwards compatibility with IR using triples, at least > > for a while. > > Probably forever... :( > > Again, this can be isolated in TargetABI or some other place.As far as I know there is no backwards compatibility promise, but equally it doesn't seem reasonable to give no notice before removing it. I'm therefore thinking that we can deprecate it in one release (3.7 or 3.8), then remove it in the next.
Hi Daniel, I'm not sure I agree with the basic idea of using the target triple as a way of encoding all of the pieces of target data as a string. I think in a number of cases what we need to do is either open up API to the back end to specify things, or encode the information into the IR when it's different from the generic triple. Ideally the triple will have enough information to do basic layout and anything else can be either gotten from the IR or passed via option. My suggestion on a route forward here is that we should look at the particular API and areas of the backend that you're having an issue with and figure out how to best communicate the data you'd like to the appropriate area. I realize this probably seems a little vague and handwavy, but I don't know what areas you've been having problems with lately. I'll absolutely help with this effort if you need assistance or guidance in any way. Thanks! -eric On Wed, Jul 8, 2015 at 7:31 AM Daniel Sanders <Daniel.Sanders at imgtec.com> wrote:> Hi, > > > > In http://reviews.llvm.org/D10969, Eric asked me to explain the wider > context of the TargetTuple object that was replacing Triple on llvmdev so > here it is. > > > > Before I start, I'm sure I don't know the full extent of GNU triple > ambiguity and lack of canonicity. Additional examples are welcome. > > > > *The Problem* > > > > As you know, LLVM uses a GNU Triple is as a target description that can be > relied upon to make decisions. It's used for various decisions such as the > default cpu, the alignment of types, the object format, the names for > libcalls, and a wide variety of others. > > In using it like this, LLVM assumes that triples are unambiguous and have > a specific defined meaning. Unfortunately, this assumption fails for a > number of reasons. > > > > The first reason is that compiler options can overrule the triple but > leave it unchanged. For example, in GCC mips-linux-gnu-gcc normally > produces 32-bit MIPS-I output using the O32 ABI, but 'mips-linux-gnu-gcc > –mips64' normally produces 64-bit MIPS-III output using the N32 ABI. Like > GCC, compiler options to mips-linux-gnu-clang should (and mostly do but > MIPS has a few crashing cases caused by triple misuse) overrule the triple. > However, we don't mutate the triple to reflect this so any decisions based > on the overridable state cannot rely on the triple to accurately reflect > the desired behaviour. > > It's worth mentioning here that some targets have hacks to partially > mutate the triple in clang to work around issues they would otherwise have > in the backend but this is done on an ad-hoc basis for specific details > (e.g. mips <-> mipsel for –EL and -EB). > > > > The second reason is that there is no canonical meaning for a given GNU > Triple, it varies between vendors and over time. There is also no > requirement for vendors to have a unique GNU Triple for their toolchain. > For GCC, it's fairly common for distributors to change the meanings of > triples using options like --with-arch, --with-cpu, --with-abi, etc. There > are also some target-specific options such as --with-mode to select > ARM/Thumb by default and --with-nan for MIPS NAN encoding selection. > Different vendors use different configure options and may change them at > will. When they do change them, the vendors often desire to keep the same > triple to be able to drop in the new version without causing wider impact > on their environment. For example, assuming I'm reading debian/rules2 for > Debian's gcc-4.9 package correctly then the i386-linux-gnu means i486 on > Debian Etch and Lenny but means i586 on more recent versions. On a similar > note, on Debian, mips-linux-gnu targets MIPS-II (optimised for typical > MIPS32 implementations) rather than the usual MIPS-I. The last example of > this ambiguity I'd like to reference is that mentioned by > https://wiki.debian.org/Multiarch/Tuples#Why_not_use_GNU_triplets.3F. In > that example, hard-float and soft-float on ARM both used arm-linux-gnueabi > but were mutually incompatible. The Multiarch tuples described on that page > are an attempt to resolve the ambiguity but I'm told that they aren't > likely to be universally adopted. > > > > The third reason, is that different triples can mean the same thing. Jim > Grosbach has mentioned that the prefixes of the GNU Triple are different > between Linux and Darwin for ARM despite sharing the same meaning > (presumably subject to the issues above). As a result decisions based on > the string have to take care of multiple possible values. Mips has a > similar issue too since a host triple (and therefore default target triple) > of mips64-linux-gnu needs to behave like mips-linux-gnu on a 32-bit Mips > port of Debian. > > > > Although not included in the description of the assumption above, one > additional flaw in the use of GNU Triples is that they are sometimes > inadequate as a description of the target. One example affecting MIPS in > particular is that the ABI is not represented in the GNU Triple we require > significant API changes to get this information where we need it. It would > be helpful to be able to pass such information through the existing > plumbing. > > > > *The Planned Solution* > > > > The plan is to split the GNU Triple represented by the llvm::Triple object > into two pieces. The first piece is the existing llvm::Triple and is > responsible for parsing the GNU triple and canonicalizing it. The second > piece is a mutable target description named llvm::TargetTuple. TargetTuple > is responsible for interpreting the triple according to the vendor's rules, > providing an interface to allow mutation by tools, and authoritatively > defining the target being targeted without the ambiguity of GNU Triples. As > an example, 'mips-linux-gnu-clang –EL …' would: > > // Parse the GNU Triple > > llvm::Triple GnuTriple("mips-linux-gnu"); > > // Convert it to a TargetTuple according to the (possibly customized) > meanings in > > // use by the vendor. > > llvm::TargetTuple TT(GnuTriple); > // Then mutate the TargetTuple according to the compiler options (or > equivalent depending > > // on the tool, for example disassemblers would mutate it according to the > object headers). > if (hasOption("-EL")) > TT.setLittleEndian() > > ... > > At this point, TT would be > "+mipsel-unknown-linux-gnu-elf32-some-other-stuff" (exact serialization is > t.b.d and may end up target dependent) which we can then rely on in the > rest of LLVM. This split resolves the issue of llvm::Triple objects not > being reliable when used as a target description since TargetTuple will > reflect the result of interpreting the triple as well as applying > appropriate options. It also provides a suitable place for vendors to > define the meanings of their GNU Triples. > > > > One significant detail is the way vendors customize the meaning of their > Triples. Currently, the plan is to nominate a constructor > (TargetTuple::TargetTuple(const Triple &)) a vendor can patch to redefine > their triples with the default implementation being the 'usual' meaning > (the meaning that should be used in the absence of customization). One nice > benefit of this configure-by-source-patch approach is that vendors can > customize multiple triples as easily as their native triple or intended > target triple. To use Debian as an example again, they would be able to > customize all their supported triples such that 'clang –target > arm-linux-gnueabihf' on the amd64 port targets their armhf port using the > same customization that makes 'clang' on the armhf port do the right thing > natively. Android, and toolchains for heterogenous platform would likely > benefit from this too. This configure-by-source-patch approach seems to > make some people uncomfortable so we may have to find another way to > configure the triples (tablegen?). > > > > To reach this result the plan is to do the following: > > 1. Replace any remaining std::string's and StringRef's containing > GNU triples with Triple objects. > > 2. Split the llvm::Triple class into llvm::Triple and > llvm::TargetTuple classes. Both are identical in implementation and almost > identical in interface at this stage. > > 3. Gradually replace Triples with TargetTuples until the C APIs and > the LLVM-IR are the only place inside LLVM where Triples are still used. > > 4. Change the implementation of TargetTuple to whatever is > convenient for LLVM's internals and decide on a serialization. > > 5. Replace serialized Triples with serialized TargetTuples in > LLVM-IR. > > a. Maintain backwards compatibility with IR using triples, at least > for a while. > > 6. Add TargetTuple support to the C API. Exact API is t.b.d. > > 7. Have the API users mutate the TargetTuple appropriately. > > Renato: This has been revised slightly from the last one we discussed due > to public C++ API's being used internally as well as externally. > > > > *Where we are now* > > > > I've just started posting patches for step 2 and 3 of the plan. My working > copy is nearly at step 4. > > > > *What's next* > > > > Upstream step 2 and 3 and then begin replacing the TargetTuple > implementation as per step 4. > > > > *Previous Discussions* > > > > http://thread.gmane.org/gmane.comp.compilers.llvm.devel/86020/focus=86073. > I should mention that I've since been made aware that the original topic of > private label prefixes could be solved in a much simpler way than > previously thought. The triple related discussion is still relevant though. > > I understand from Renato that there are more threads over the last few > years but I haven't looked for them. > > > > > > *Daniel Sanders* > > Leading Software Design Engineer, MIPS Processor IP > > Imagination Technologies Limited > > www.imgtec.com > > >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20150729/fa2d58f4/attachment.html>
On 29 July 2015 at 21:44, Eric Christopher <echristo at gmail.com> wrote:> I'm not sure I agree with the basic idea of using the target triple as a way > of encoding all of the pieces of target data as a string.Hi Eric, That's not the idea at all. The Triple object will remain unchanged. The Tuple will be the API to handle getting/setting parameters depending on the Triple, compiler flags, attributes, etc. There will be no string representation of all options, as that would be impossible, or at least, highly undesirable, especially in the IR. The Tuple is for the sole use of front-ends, middle-ends and back-ends to communicate and understand the *same* meaning regarding the *same* input. Having a Tuple class that encodes details of the targets go a long way to ensure that, since you can directly pass the Tuple when you build the Target objects, and the information it provides will be identical, no matter where it is. Right now, we have multiple representations of the targets' information because the Triple object cannot encode every aspect of them, especially problematic between Clang and LLVM. The decision to create a new class (Tuple) is because Triple already has a legacy-heavy meaning, which should not change. This is not about the serialization of the target information, but rather the consistent manipulation of it inside the compiler.> My suggestion on a route forward here is that we should look at the > particular API and areas of the backend that you're having an issue with and > figure out how to best communicate the data you'd like to the appropriate > area.That is precisely what he's doing. :) cheers, --renato
Hi Eric, Thanks for getting back to me on this.> I'm not sure I agree with the basic idea of using the target triple as a way of > encoding all of the pieces of target data as a string. I think in a number of > cases what we need to do is either open up API to the back end to specify things, > or encode the information into the IR when it's different from the generic triple. > Ideally the triple will have enough information to do basic layout and anything > else can be either gotten from the IR or passed via option.(from the context, you might have meant 'tuple' where you've written 'triple'. I'm answering based on the assumption you meant 'triple') The GNU triple is already used as a way of encoding a large amount of the target data in a string but unfortunately, while this data is passed throughout LLVM, it isn't reliable because GNU triples are ambiguous and inconsistent. For example, in GCC toolchains mips-linux-gnu probably means a MIPS target on Gnu/Linux but anything beyond that (ISA revision, default ABI, multilib layout, etc.) is up to the person who built the toolchain and may change over time. Another example is that Debian's definition for i386-linux-gnu has been i486 and i586 at various points in time. The proposed TargetTuple is a direct replacement for the GNU triple and is intended to resolve this ambiguity and move away from a string-based implementation (we need to keep a string serialization though, see below). Essentially, I'm trying to push the ambiguity out of the internals and give the distributor control of how the ambiguity is resolved for their environment. Once that is done, we'll be able to rely on the TargetTuple for information about the target such as ABI's, architecture revisions, endianness, etc. I agree that we should open up the API to specify the appropriate data and that is something that TargetTuple will acquire during step 4 and 7 of the plan (mostly step 7 where compiler/tool options begin mutating the target tuple). I don't agree with keeping the GNU triple around though for two main reasons. The first is that most people believe that GNU triples accurately describe the target and there will be a strong temptation to inappropriately base logic on them. The second is that the meaning of the triple varies between toolchain builds and over time and there is a significant potential for bugs where different parts of the toolchain use different meanings for the same GNU triple (due to rebuilding or switching toolchains, or moving objects from system to system). We ought to resolve the ambiguity once and then stick to that interpretation. The string serialization I mentioned above is useful for LLVM-IR as part of a direct replacement for the 'target triple' statement. We could split this statement up into smaller pieces but the migration to target tuples is already difficult so I think it would be best to do a direct replacement first and redesign the IR statements later if we want to. The serialization is also useful for command line options on internal tools such as llc to give us precise control over our tests that the GNU triple can't deliver. This will be particularly important when distributors can apply their own disambiguations to GNU triples. The serialization may also be useful as part of a C API but I haven't given the C API much thought beyond preserving the current API. Hopefully, that helps clear up your concerns. Let me know if there's anything that still seems strange.> My suggestion on a route forward here is that we should look at the particular > API and areas of the backend that you're having an issue with and figure out > how to best communicate the data you'd like to the appropriate area. I realize > this probably seems a little vague and handwavy, but I don't know what areas > you've been having problems with lately. I'll absolutely help with this effort if > you need assistance or guidance in any way.The MIPS specific problems are broad and varied. Some of the bigger ones are: * Building clang on a 32-bit Debian and a 64-bit MIPS processor produces a compiler that cannot target the native system. The release packages work around this by 'cross-compiling' from the host triple to the target triple which are different strings (mips-linux-gnu vs mips64-linux-gnu) but have the same meaning. * It's not possible to produce a clang that can generate code for both 32-bit and 64-bit MIPS without one of them needing a -target option to change the GNU triple. This is because we based the logic on the triple and lack anything else to use. * Various details (ELF headers, label prefixes, exception personality, JIT target, etc.) depend on the ABI and OS Distribution rather than just 32-bit vs 64-bit * It's not possible to implement clang in a way that can support all of mips-linux-gnu's possible meanings. mips-mti-linux-gnu, and mips-img-linux-gnu have the same problem to a lesser degree. ________________________________________ From: Eric Christopher [echristo at gmail.com] Sent: 29 July 2015 21:44 To: Daniel Sanders; LLVM Developers Mailing List (llvmdev at cs.uiuc.edu) Cc: Renato Golin (renato.golin at linaro.org); Jim Grosbach (grosbach at apple.com) Subject: Re: The Trouble with Triples Hi Daniel, I'm not sure I agree with the basic idea of using the target triple as a way of encoding all of the pieces of target data as a string. I think in a number of cases what we need to do is either open up API to the back end to specify things, or encode the information into the IR when it's different from the generic triple. Ideally the triple will have enough information to do basic layout and anything else can be either gotten from the IR or passed via option. My suggestion on a route forward here is that we should look at the particular API and areas of the backend that you're having an issue with and figure out how to best communicate the data you'd like to the appropriate area. I realize this probably seems a little vague and handwavy, but I don't know what areas you've been having problems with lately. I'll absolutely help with this effort if you need assistance or guidance in any way. Thanks! -eric On Wed, Jul 8, 2015 at 7:31 AM Daniel Sanders <Daniel.Sanders at imgtec.com<mailto:Daniel.Sanders at imgtec.com>> wrote: Hi, In http://reviews.llvm.org/D10969, Eric asked me to explain the wider context of the TargetTuple object that was replacing Triple on llvmdev so here it is. Before I start, I'm sure I don't know the full extent of GNU triple ambiguity and lack of canonicity. Additional examples are welcome. The Problem As you know, LLVM uses a GNU Triple is as a target description that can be relied upon to make decisions. It's used for various decisions such as the default cpu, the alignment of types, the object format, the names for libcalls, and a wide variety of others. In using it like this, LLVM assumes that triples are unambiguous and have a specific defined meaning. Unfortunately, this assumption fails for a number of reasons. The first reason is that compiler options can overrule the triple but leave it unchanged. For example, in GCC mips-linux-gnu-gcc normally produces 32-bit MIPS-I output using the O32 ABI, but 'mips-linux-gnu-gcc –mips64' normally produces 64-bit MIPS-III output using the N32 ABI. Like GCC, compiler options to mips-linux-gnu-clang should (and mostly do but MIPS has a few crashing cases caused by triple misuse) overrule the triple. However, we don't mutate the triple to reflect this so any decisions based on the overridable state cannot rely on the triple to accurately reflect the desired behaviour. It's worth mentioning here that some targets have hacks to partially mutate the triple in clang to work around issues they would otherwise have in the backend but this is done on an ad-hoc basis for specific details (e.g. mips <-> mipsel for –EL and -EB). The second reason is that there is no canonical meaning for a given GNU Triple, it varies between vendors and over time. There is also no requirement for vendors to have a unique GNU Triple for their toolchain. For GCC, it's fairly common for distributors to change the meanings of triples using options like --with-arch, --with-cpu, --with-abi, etc. There are also some target-specific options such as --with-mode to select ARM/Thumb by default and --with-nan for MIPS NAN encoding selection. Different vendors use different configure options and may change them at will. When they do change them, the vendors often desire to keep the same triple to be able to drop in the new version without causing wider impact on their environment. For example, assuming I'm reading debian/rules2 for Debian's gcc-4.9 package correctly then the i386-linux-gnu means i486 on Debian Etch and Lenny but means i586 on more recent versions. On a similar note, on Debian, mips-linux-gnu targets MIPS-II (optimised for typical MIPS32 implementations) rather than the usual MIPS-I. The last example of this ambiguity I'd like to reference is that mentioned by https://wiki.debian.org/Multiarch/Tuples#Why_not_use_GNU_triplets.3F. In that example, hard-float and soft-float on ARM both used arm-linux-gnueabi but were mutually incompatible. The Multiarch tuples described on that page are an attempt to resolve the ambiguity but I'm told that they aren't likely to be universally adopted. The third reason, is that different triples can mean the same thing. Jim Grosbach has mentioned that the prefixes of the GNU Triple are different between Linux and Darwin for ARM despite sharing the same meaning (presumably subject to the issues above). As a result decisions based on the string have to take care of multiple possible values. Mips has a similar issue too since a host triple (and therefore default target triple) of mips64-linux-gnu needs to behave like mips-linux-gnu on a 32-bit Mips port of Debian. Although not included in the description of the assumption above, one additional flaw in the use of GNU Triples is that they are sometimes inadequate as a description of the target. One example affecting MIPS in particular is that the ABI is not represented in the GNU Triple we require significant API changes to get this information where we need it. It would be helpful to be able to pass such information through the existing plumbing. The Planned Solution The plan is to split the GNU Triple represented by the llvm::Triple object into two pieces. The first piece is the existing llvm::Triple and is responsible for parsing the GNU triple and canonicalizing it. The second piece is a mutable target description named llvm::TargetTuple. TargetTuple is responsible for interpreting the triple according to the vendor's rules, providing an interface to allow mutation by tools, and authoritatively defining the target being targeted without the ambiguity of GNU Triples. As an example, 'mips-linux-gnu-clang –EL …' would: // Parse the GNU Triple llvm::Triple GnuTriple("mips-linux-gnu"); // Convert it to a TargetTuple according to the (possibly customized) meanings in // use by the vendor. llvm::TargetTuple TT(GnuTriple); // Then mutate the TargetTuple according to the compiler options (or equivalent depending // on the tool, for example disassemblers would mutate it according to the object headers). if (hasOption("-EL")) TT.setLittleEndian() ... At this point, TT would be "+mipsel-unknown-linux-gnu-elf32-some-other-stuff" (exact serialization is t.b.d and may end up target dependent) which we can then rely on in the rest of LLVM. This split resolves the issue of llvm::Triple objects not being reliable when used as a target description since TargetTuple will reflect the result of interpreting the triple as well as applying appropriate options. It also provides a suitable place for vendors to define the meanings of their GNU Triples. One significant detail is the way vendors customize the meaning of their Triples. Currently, the plan is to nominate a constructor (TargetTuple::TargetTuple(const Triple &)) a vendor can patch to redefine their triples with the default implementation being the 'usual' meaning (the meaning that should be used in the absence of customization). One nice benefit of this configure-by-source-patch approach is that vendors can customize multiple triples as easily as their native triple or intended target triple. To use Debian as an example again, they would be able to customize all their supported triples such that 'clang –target arm-linux-gnueabihf' on the amd64 port targets their armhf port using the same customization that makes 'clang' on the armhf port do the right thing natively. Android, and toolchains for heterogenous platform would likely benefit from this too. This configure-by-source-patch approach seems to make some people uncomfortable so we may have to find another way to configure the triples (tablegen?). To reach this result the plan is to do the following: 1. Replace any remaining std::string's and StringRef's containing GNU triples with Triple objects. 2. Split the llvm::Triple class into llvm::Triple and llvm::TargetTuple classes. Both are identical in implementation and almost identical in interface at this stage. 3. Gradually replace Triples with TargetTuples until the C APIs and the LLVM-IR are the only place inside LLVM where Triples are still used. 4. Change the implementation of TargetTuple to whatever is convenient for LLVM's internals and decide on a serialization. 5. Replace serialized Triples with serialized TargetTuples in LLVM-IR. a. Maintain backwards compatibility with IR using triples, at least for a while. 6. Add TargetTuple support to the C API. Exact API is t.b.d. 7. Have the API users mutate the TargetTuple appropriately. Renato: This has been revised slightly from the last one we discussed due to public C++ API's being used internally as well as externally. Where we are now I've just started posting patches for step 2 and 3 of the plan. My working copy is nearly at step 4. What's next Upstream step 2 and 3 and then begin replacing the TargetTuple implementation as per step 4. Previous Discussions http://thread.gmane.org/gmane.comp.compilers.llvm.devel/86020/focus=86073. I should mention that I've since been made aware that the original topic of private label prefixes could be solved in a much simpler way than previously thought. The triple related discussion is still relevant though. I understand from Renato that there are more threads over the last few years but I haven't looked for them. Daniel Sanders Leading Software Design Engineer, MIPS Processor IP Imagination Technologies Limited www.imgtec.com<http://www.imgtec.com/>