bug-gnulib@gnu.org mirror (unofficial)
 help / color / mirror / Atom feed
From: "Martin Storsjö" <martin@martin.st>
To: bug-gnulib@gnu.org
Subject: Re: Issues with posix functions on modern macOS/Xcode
Date: Tue, 1 Dec 2020 15:50:42 +0200 (EET)	[thread overview]
Message-ID: <alpine.DEB.2.23.453.2012011549280.4295@cone.martin.st> (raw)
In-Reply-To: <alpine.DEB.2.23.453.2012011509530.4295@cone.martin.st>

[-- Attachment #1: Type: text/plain, Size: 1777 bytes --]

On Tue, 1 Dec 2020, Martin Storsjö wrote:

> On Xcode 12.2 on macOS 11.0 on arm64, the following sequence fails to build:
>
> $ ./gnulib-tool --create-testdir --dir=/tmp/testdir-posix \
>               --single-configure --avoid=havelib-tests \
>               `./posix-modules`
> $ cd /tmp/testdir-posix
> $ ./configure CPPFLAGS=-Wall 2>&1 | tee log1
> $ make 2>&1 | tee log2
> $ make check 2>&1 | tee log3
>
>
> The "make" step fails with the following error:
>
> gcc  -g -O2   -o test-dprintf-posix2 test-dprintf-posix2.o libtests.a 
> ../gllib/libgnu.a libtests.a ../gllib/libgnu.a libtests.a
> Undefined symbols for architecture arm64:
>  "_vm_region", referenced from:
>      _vma_iterate in libtests.a(vma-iter.o)
> ld: symbol(s) not found for architecture arm64
>
>
> This same bit does succeed on x86_64 though - it looks like 
> gnulib/lib/vma-iter.c has a macos specific arch check like this:
>
> # if defined __ppc64__ || defined __x86_64__
>
> This probably needs to be amended with "|| defined __aarch64__".
>
> On macOS 10.15.7 (Catalina) on x86_64, both with Xcode 11.7 and 12.2, the 
> following tests fail:
>
> FAIL: test-dprintf-posix2.sh
> FAIL: test-fflush2.sh
> FAIL: test-fprintf-posix3.sh
> FAIL: test-fpurge
> FAIL: test-ftell.sh
> FAIL: test-ftell2.sh
> FAIL: test-ftello.sh
> FAIL: test-ftello2.sh
> FAIL: test-futimens
> FAIL: test-renameat
> FAIL: test-renameatu
> FAIL: test-strsignal
> FAIL: test-utime
> FAIL: test-utimens
> FAIL: test-utimensat

FWIW, after fixing the vma-iter ifdef, the same tests fail in the same way 
on arm64 as on x86_64 (except that test-dprintf-posix2.sh doesn't seem to 
fail there).

So in any case, if the tests are made to pass on existing x86_64, they'll 
most probably work just the same way on arm64.

// Martin

  reply	other threads:[~2020-12-01 13:50 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-01 13:14 Issues with posix functions on modern macOS/Xcode Martin Storsjö
2020-12-01 13:50 ` Martin Storsjö [this message]
2020-12-01 18:58 ` Bruno Haible
2020-12-02 18:56 ` Bruno Haible
2021-01-02 20:43   ` Bruno Haible
2021-03-21  2:57     ` Bruno Haible
2021-03-21 13:58       ` Bruno Haible
2020-12-02 21:52 ` strsignal on macOS Bruno Haible
2020-12-03 20:59 ` fprintf-posix " Bruno Haible
2020-12-04 23:24 ` utime " Bruno Haible
2021-01-02 16:03 ` renameatu: Fix test failures " Bruno Haible
2021-01-02 18:06 ` utimens: " Bruno Haible

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

  List information: https://lists.gnu.org/mailman/listinfo/bug-gnulib

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

  git send-email \
    --in-reply-to=alpine.DEB.2.23.453.2012011549280.4295@cone.martin.st \
    --to=martin@martin.st \
    --cc=bug-gnulib@gnu.org \
    /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).