[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Windows 9X crash

From: Eli Zaretskii
Subject: Re: Windows 9X crash
Date: Sat, 16 Jan 2016 11:15:09 +0200

> Cc: address@hidden, address@hidden
> From: Jussi Lahdenniemi <address@hidden>
> Date: Fri, 15 Jan 2016 12:33:23 +0200
> On 15.1.2016 12.18, Eli Zaretskii wrote:
> > I think the 9X-specific code should be #ifdef'ed away in 64-bit
> > builds, is that right?  The alignment code and the "p - 1" stuff is
> > wrong for 64-bit code anyway, I think.
> It can be #ifdef'd away if wanted, but it's not used on 64-bit OSes 
> anyway (due to the "if (os_subtype == OS_9X)" in init_heap), and it's 
> not causing any warnings or errors in the 64-bit build process.
> Of course, should Microsoft release a 64-bit version of the Windows 9X 
> family appear on the market, then the code would be broken :)
> The "p-1" stuff would not be broken for a 64-bit Windows 98, though; the 
> "+8"s would (they should be "+12" to ensure enough room for 64-bit 
> void*s).  But then again, the theoretical 64-bit Windows 98 would most 
> probably HeapAllocate on 8-byte boundaries - and in that case, the code 
> would work perfectly as-is.

OK, I pushed the patch to the emacs-25 branch.  Thanks.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]