James Molloy via llvm-dev
2016-Jan-05 08:49 UTC
[llvm-dev] Diff to add ARMv6L to Target parser
Hi, IMO we should support this, even though if given the option I'd have asked the linux guys not to invent a new triple. It's in linux now, and `uname -a` is a very standard way of obtaining the host's triple. James On Tue, 5 Jan 2016 at 08:34 Tim Northover via llvm-dev < llvm-dev at lists.llvm.org> wrote:> On 4 January 2016 at 14:42, Artyom Skrobov via llvm-dev > <llvm-dev at lists.llvm.org> wrote: > > Ah, I see: ARMv7L is now an alias for ARMv7A. > > The "armv7l" comes directly from the Linux kernel I believe: "armv7" + > "little-endian" (arch/arm/kernel/setup.c:638 ish). I'm not entirely > convinced the "l" belongs in any triple seen by LLVM, but if we do > support it we should probably handle it analogously to the "el"/"eb" > suffixes. > > Cheers. > > Tim. > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20160105/0c17289d/attachment.html>
Renato Golin via llvm-dev
2016-Jan-05 09:13 UTC
[llvm-dev] Diff to add ARMv6L to Target parser
On 5 January 2016 at 08:49, James Molloy via llvm-dev <llvm-dev at lists.llvm.org> wrote:> IMO we should support this, even though if given the option I'd have asked > the linux guys not to invent a new triple. It's in linux now, and `uname -a` > is a very standard way of obtaining the host's triple.We do support v7l and we should support v6l. I though we did that already, since RaspberyPi has been supported for years. Anyhow, Artyom's proposal is the best, IMO, to treat it like an alias and handle like v7A internally. If we end up needing specific decisions in the driver, it should stay in the driver. cheers, --renato
Tim Northover via llvm-dev
2016-Jan-05 10:28 UTC
[llvm-dev] Diff to add ARMv6L to Target parser
On 5 January 2016 at 09:13, Renato Golin <renato.golin at linaro.org> wrote:> Anyhow, Artyom's proposal is the best, IMO, to treat it like an alias > and handle like v7A internally. If we end up needing specific > decisions in the driver, it should stay in the driver.That's rather a hack, given that the 'l' actually has semantic meaning, but I suppose I could live with it. Tim.