GNU bug report logs -
#54915
[PATCH] Guile rejects empty vendor in GNU triplets, as used by NetBSD
Previous Next
Full log
View this message in rfc822 format
[Message part 1 (text/plain, inline)]
Your message dated Thu, 16 Jun 2022 09:43:12 +0200
with message-id <878rpxoxyn.fsf <at> gnu.org>
and subject line Re: bug#54915: [PATCH] Guile rejects empty vendor in GNU triplets, as used by NetBSD
has caused the debbugs.gnu.org bug report #54915,
regarding [PATCH] Guile rejects empty vendor in GNU triplets, as used by NetBSD
to be marked as done.
(If you believe you have received this mail in error, please contact
help-debbugs <at> gnu.org.)
--
54915: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=54915
GNU Bug Tracking System
Contact help-debbugs <at> gnu.org with problems
[Message part 2 (message/rfc822, inline)]
[Message part 3 (text/plain, inline)]
NetBSD uses an empty vendor string in a GNU target triplet, like
`x86_64--netbsd' or `armv7--netbsdelf-eabihf'; it's been like this
since the mid-'90s. (Before that, NetBSD omitted the vendor
altogether, so if x86_64 existed then it might have been
`x86_64-netbsd'.)
This runs afoul of a rule in guile's validate-target in
module/system/base/target.scm that requires all components to be
nonempty, so pkgsrc's guile build has been carrying around a
workaround -- just for guile, not for any other packages I know of --
for ages. Without the workaround in pkgsrc:
> BOOTSTRAP(stage0) GUILEC ice-9/eval.go
> Backtrace:
> 7 (apply-smob/1 #<boot-closure 736f6474ff80 (_ _)> #<exc?> ?)
> 6 (apply-smob/1 #<boot-closure 736f2e2fbd80 ()>)
> 5 (apply-smob/0 #<thunk 736f2e2fbe40>)
> 4 (primitive-eval ((@ (ice-9 control) %) (begin (# #) ?)))
> 3 (apply-smob/1 #<boot-closure 736f63786a80 ()>)
> 2 (apply-smob/1 #<boot-closure 736f265a9500 ()>)
> 1 (scm-error misc-error #f "~A ~S" ("invalid target" "?") #)
> 0 (apply-smob/1 #<boot-closure 736f6474ff40 (_ . _)> # #f ?)
>
> ERROR: In procedure apply-smob/1:
> invalid target "x86_64--netbsd"
The attached patch relaxes validate-target so it accepts an empty
vendor string (but still requires all other components to be nonempty
as before). With the patch and without the workaround in pkgsrc,
guile builds happily.
[emptyvendor.patch (text/plain, attachment)]
[Message part 5 (message/rfc822, inline)]
Hi,
Taylor R Campbell <campbell+guile <at> mumble.net> skribis:
>>From 12440a85559c3de5e6bced9c9377f3d5d7f5948e Mon Sep 17 00:00:00 2001
> From: Taylor R Campbell <campbell+guile <at> mumble.net>
> Date: Wed, 13 Apr 2022 09:51:08 +0000
> Subject: [PATCH] Allow empty vendor string in GNU target triplets.
>
> NetBSD and pkgsrc have been using an empty vendor string since the
> mid-'90s, such as x86_64--netbsd. pkgsrc has been carrying around a
> workaround just the guile build for a long time. (Before that,
> NetBSD omitted the vendor altogether, so if x86_64 existed then it
> might have been `x86_64-netbsd', but that caused more problems.)
> This change makes Guile accept an empty vendor string so workarounds
> are no longer necessary.
>
> * module/system/base/target.scm (validate-target): Allow empty vendor
> string in GNU target triplets.
I added tests in cross-compilation.test and committed.
Thanks!
Ludo’.
This bug report was last modified 3 years and 34 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.