| Message ID | CABn4x4Nw++17Gt406P9kXTmFbnYsmtVikAY=tvkWxsKJLaJjEA@mail.gmail.com |
|---|---|
| State | Accepted |
| Headers | show |
Hi Andy, On 31/10/13 16:30, Andrew McDermott wrote: > Attached is a patch for building your copy[1] of binutils on AArch64. I > realize that this is already present in the upstream version of binutils > but is here for convenience. > > [1] http://hg.code.sourceforge.net/p/binutilsaarch64/code I'm not quite sure what you are intending here. Are you asking me to apply this patch to the binutilsaarch64 repo up on sourceforge? I don't really understand what the point of that would be. The above sourceforge repo is used when building an hsdis library for an OpenJDK that runs on Intel using our Intel-based AArch64 simulator. Is there a reason why you want to use this version of binutils when building on AArch64 itself? (as opposed to using the native binutils, that is). regards, Andrew Dinn ----------- Principal Software Engineer Red Hat UK Ltd Registered in UK and Wales under Company Registration No. 3798903 Directors: Michael Cunningham (USA), Matt Parson (USA), Charlie Peters (USA), Paul Hickey (Ireland)
On 31 Oct 2013, at 09:53, Andrew Dinn <adinn@redhat.com> wrote: > Hi Andy, > > On 31/10/13 16:30, Andrew McDermott wrote: >> Attached is a patch for building your copy[1] of binutils on AArch64. I >> realize that this is already present in the upstream version of binutils >> but is here for convenience. >> >> [1] http://hg.code.sourceforge.net/p/binutilsaarch64/code > > I'm not quite sure what you are intending here. Are you asking me to > apply this patch to the binutilsaarch64 repo up on sourceforge? > > I don't really understand what the point of that would be. The above > sourceforge repo is used when building an hsdis library for an OpenJDK > that runs on Intel using our Intel-based AArch64 simulator. Is there a > reason why you want to use this version of binutils when building on > AArch64 itself? (as opposed to using the native binutils, that is). Apologies for the lack of clarity. I was building self-hosted and rather than switch out to the native binutils I just tried to use the version that come from my checkout of the OpenJDK-8 aarch64-port. If you don’t want to apply it to your repo that’s fine - I just copied you for completeness rather than just making it work locally for me, and then moving on. > > regards, > > > Andrew Dinn > ----------- > Principal Software Engineer > Red Hat UK Ltd > Registered in UK and Wales under Company Registration No. 3798903 > Directors: Michael Cunningham (USA), Matt Parson (USA), Charlie Peters > (USA), Paul Hickey (Ireland)
Hi Andy, On 31/10/13 17:13, Andrew McDermott wrote: > Apologies for the lack of clarity. No problem. Just wanted to be sure what your note was about. > I was building self-hosted and rather than switch out to the native > binutils I just tried to use the version that come from my checkout of > the OpenJDK-8 aarch64-port. If you don’t want to apply it to your repo > that’s fine - I just copied you for completeness rather than just making > it work locally for me, and then moving on. I am actually very happy to add the patch if it would help you or others. I realised after posting that one advantage of using it on AArch64 is that disassembly listings would use our special register names like rscratch1, rthread etc (although that may be a disadvantage from certain perspectives). I have applied your patch and pushed it up to sourceforge. regards, Andrew Dinn -----------
From cb1c0efe8019e0f73bd169f47053e405aae3c857 Mon Sep 17 00:00:00 2001 From: Andrew McDermott <andrew.mcdermott@linaro.org> Date: Thu, 31 Oct 2013 09:07:55 -0700 Subject: [PATCH] config.guess: add support for AArch64 Signed-off-by: Andrew McDermott <andrew.mcdermott@linaro.org> --- config.guess | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/config.guess b/config.guess index b02565c..3fbec45 100755 --- a/config.guess +++ b/config.guess @@ -875,6 +875,13 @@ EOF if test "$?" = 0 ; then LIBC="libc1" ; else LIBC="" ; fi echo ${UNAME_MACHINE}-unknown-linux-gnu${LIBC} exit ;; + aarch64:Linux:*:*) + echo ${UNAME_MACHINE}-unknown-linux-gnu + exit ;; + aarch64_be:Linux:*:*) + UNAME_MACHINE=aarch64_be + echo ${UNAME_MACHINE}-unknown-linux-gnu + exit ;; arm*:Linux:*:*) eval $set_cc_for_build if echo __ARM_EABI__ | $CC_FOR_BUILD -E - 2>/dev/null \ -- 1.7.9.5