GNU bug report logs - #39883
28.0.50; macOS blank frame

Previous Next

Package: emacs;

Reported by: Philippe Spiesser <spiesser.philippe <at> orange.fr>

Date: Tue, 3 Mar 2020 15:33:01 UTC

Severity: normal

Found in version 28.0.50

Done: Alan Third <alan <at> idiocy.org>

Bug is archived. No further changes may be made.

Full log


Message #46 received at 39883 <at> debbugs.gnu.org (full text, mbox):

From: Ashish SHUKLA <ashish.is <at> lostca.se>
To: Alan Third <alan <at> idiocy.org>, 39883 <at> debbugs.gnu.org
Subject: Re: bug#39883: 28.0.50; macOS blank frame
Date: Thu, 2 Apr 2020 11:49:45 +0530
[Message part 1 (text/plain, inline)]
On 4/2/20 12:23 AM, Alan Third wrote:
> On Wed, Apr 01, 2020 at 09:51:37AM +0530, Ashish SHUKLA wrote:
>> On 3/31/20 10:22 PM, Alan Third wrote:
>>> On Mon, Mar 30, 2020 at 01:24:59AM +0530, Ashish SHUKLA wrote:
>>>>
>>>> It seems like building with same configure flags, but outside nix causes
>>>> issue to disappear. So, maybe it's something to do with the nix environment.
>>>
>>> Nix used to use headers from macOS 10.10 or thereabouts, even on newer
>>> versions of macOS. I don’t know if this is still the case. The problem
>>> is that our compile‐time feature detection code detects the wrong
>>> features as a result.
>>>
>>> Could you perhaps try changing src/nsterm.h so that this:
>>>
>>> #define NS_DRAW_TO_BUFFER 1
>>>
>>> is always defined? You should be able to just delete or comment out
>>> the #if and #endif.
>>>
>>> I don’t know how easy that is for you with Nix, hopefully it’s pretty
>>> straight forward.
>>>
>>
>> Hi Alan,
>>
>> It turns out after deleting those lines the problem is resolved. I will
>> investigate on Nix side, what went wrong.
> 
> It’s quite possible we’ll need to provide some sort of option to force
> this, or to do run‐time feature detection. I don’t much fancy trying
> to code up the runtime detection for this feature, but if we need it
> we need it.
> 
> For my own future reference: perhaps we should do a test in
> ./configure rather than rely on the contents of the headers?
> 

That definitely seems more foolproof. Let me know if you like me to test
your diff.

Thanks!
-- 
Ashish SHUKLA

[signature.asc (application/pgp-signature, attachment)]

This bug report was last modified 5 years and 32 days ago.

Previous Next


GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.