public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Ian Lance Taylor <iant@golang.org>
To: Andreas Krebbel <krebbel@linux.vnet.ibm.com>
Cc: gcc-patches <gcc-patches@gcc.gnu.org>
Subject: Re: libgo patch committed: Change build procedure to use build tags
Date: Sat, 13 Aug 2016 00:23:00 -0000	[thread overview]
Message-ID: <CAOyqgcVYS4D5pcJMsvGHgMdWrT-83ZwqjzwmHuaU44OkEpr-AA@mail.gmail.com> (raw)
In-Reply-To: <6e128c9e-8f4d-8222-f3cc-2425b88275e3@linux.vnet.ibm.com>

On Fri, Aug 12, 2016 at 4:19 AM, Andreas Krebbel
<krebbel@linux.vnet.ibm.com> wrote:
> On 08/06/2016 02:36 AM, Ian Lance Taylor wrote:
>> Go packages use build tags (see the section on Build Constraints at
>> https://golang.org/pkg/go/build/) to select which files to build on
>> specific systems.
>>
>> Previously the libgo Makefile explicitly listed the set of files to
>> compile for each package.  For packages that use build tags, this
>> required a lot of awkward automake conditionals in the Makefile.
>>
>> This patch changes the build to look at the build tags in the files.
>> The new shell script libgo/match.sh does the matching.  This required
>> adjusting a lot of build tags, and removing some files that are never
>> used.  I verified that the exact same sets of files are compiled on
>> x86_64-pc-linux-gnu.  I also tested the build on i386-sun-solaris
>> (building for both 32-bit and 64-bit).
>>
>> Writing match.sh revealed some bugs in the build tag handling that
>> already exists, in a slightly different form, in the gotest shell
>> script.  This patch fixes those problems as well.
>>
>> The old code used automake conditionals to handle systems that were
>> missing strerror_r and wait4.  Rather than deal with those in Go,
>> those functions are now implemented in runtime/go-nosys.c when
>> necessary, so the Go code can simply assume that they exist.
>>
>> The os testsuite looked for dir_unix.go, which was never built for
>> gccgo and has now been removed.  I changed the testsuite to look for
>> dir.go instead.
>>
>> Note that if you have an existing build directory, you will have to
>> remove all the .dep files in TARGET/libgo after updating to this
>> patch.  There isn't anything that will force them to update
>> automatically.
>>
>> Bootstrapped on x86_64-pc-linux-gnu and i386-sun-solaris.  Ran Go
>> testsuite on x86_64-pc-linux-gnu.  Committed to mainline.
>>
>> Ian
>
> This appears to break libgo build on s390x:
>
> libtool: compile:  /home3/andreas/clean/gcc-7.0.0-64-clean-build/./gcc/gccgo
> -B/home3/andreas/clean/gcc-7.0.0-64-clean-build/./gcc/
> -B/home3/andreas/clean/gcc-7.0.0-64-clean-install/s390x-ibm-linux-gnu/bin/ -B/h
> ome3/andreas/clean/gcc-7.0.0-64-clean-install/s390x-ibm-linux-gnu/lib/ -isystem
> /home3/andreas/clean/gcc-7.0.0-64-clean-install/s390x-ibm-linux-gnu/include -isystem
> /home3/andreas/clean/gcc-7.0.0-64-clean-instal
> l/s390x-ibm-linux-gnu/sys-include -O2 -g -I . -c -fgo-pkgpath=hash/crc32
> /home3/andreas/clean/../gcc/libgo/go/hash/crc32/crc32.go
> /home3/andreas/clean/../gcc/libgo/go/hash/crc32/crc32_generic.go /home3/andreas/c
> lean/../gcc/libgo/go/hash/crc32/crc32_s390x.go  -fPIC -o hash/.libs/crc32.o
> /home3/andreas/clean/../gcc/libgo/go/hash/crc32/crc32_s390x.go:56:1: error: redefinition of
> ‘updateCastagnoli’
>  func updateCastagnoli(crc uint32, p []byte) uint32 {
>  ^
> /home3/andreas/clean/../gcc/libgo/go/hash/crc32/crc32_generic.go:12:1: note: previous definition of
> ‘updateCastagnoli’ was here
>  func updateCastagnoli(crc uint32, p []byte) uint32 {
>  ^
> /home3/andreas/clean/../gcc/libgo/go/hash/crc32/crc32_s390x.go:81:1: error: redefinition of ‘updateIEEE’
>  func updateIEEE(crc uint32, p []byte) uint32 {
>  ^
> /home3/andreas/clean/../gcc/libgo/go/hash/crc32/crc32_generic.go:20:1: note: previous definition of
> ‘updateIEEE’ was here
>  func updateIEEE(crc uint32, p []byte) uint32 {
>  ^
> make[4]: *** [hash/crc32.lo] Error 1
> make[4]: *** Waiting for unfinished jobs....
>
>
> Adding more "build ignore"s does fix the build for me.

Thanks for the patch.  Committed.   (FYI, you shouldn't make changes
directly to libgo in the GCC repository--the files there are copied
from a separate repository.  See libgo/README.)


> However, I'm wondering why we have to disable the asm tuned files for gccgo? For s390 I really would
> like to enable them since they make a huge difference.

I would like to enable them also.  They are not enabled today because
the assembler code is written in the syntax accepted by
https://golang.org/cmd/asm.  The files in question are

https://tip.golang.org/src/crypto/aes/asm_s390x.s
https://tip.golang.org/src/hash/crc32/crc32_s390x.s

If someone could rewrite those into gas syntax, we could use them in gccgo.


> The same is probably also required in ./hash/crc32/crc32_amd64p32.go
> I could add this as well.

It is not needed there, because GOARCH will never match amd64p32.
That is only used for NaCL, which gccgo does not support.

Ian

      reply	other threads:[~2016-08-13  0:23 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-06  0:36 Ian Lance Taylor
2016-08-07 12:14 ` Andreas Schwab
2016-08-07 15:18   ` Matthias Klose
2016-08-07 22:33     ` Ian Lance Taylor
2016-09-04 16:24     ` Matthias Klose
2016-09-23 21:53       ` Ian Lance Taylor
2016-08-08 18:15 ` Lynn A. Boger
2016-08-08 18:26   ` Ian Lance Taylor
2016-08-08 19:07     ` Lynn A. Boger
2016-08-08 20:34   ` Ian Lance Taylor
2016-08-09 15:46     ` Lynn A. Boger
2016-08-11 15:16 ` Rainer Orth
2016-08-11 21:36   ` Ian Lance Taylor
2016-08-12  9:15     ` Rainer Orth
2016-08-12 13:56       ` Rainer Orth
2016-08-13  2:53         ` Ian Lance Taylor
2016-08-13  0:14       ` Ian Lance Taylor
2016-08-12 11:20 ` Andreas Krebbel
2016-08-13  0:23   ` Ian Lance Taylor [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAOyqgcVYS4D5pcJMsvGHgMdWrT-83ZwqjzwmHuaU44OkEpr-AA@mail.gmail.com \
    --to=iant@golang.org \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=krebbel@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).