Mark H Weaver writes:
>> I've attached a patch to (hopefully) fix this problem. Unfortunately,
>> David tells me that this doesn't solve his reload issue, so unless he
>> made a mistake in testing it, [...]
>
> It turned out that David made a mistake in testing, and this patch
> _does_ fix his p
Replying to myself...
> David Pirotte has been experiencing a problem where (reload-module ...)
> was failing to trigger a recompilation even after the source file has
> been modified. [...]
[...]
> I've attached a patch to (hopefully) fix this problem. Unfortunately,
> David tells me that this
> David Pirotte has been experiencing a problem where (reload-module ...)
> was failing to trigger a recompilation even after the source file has
> been modified.
Don't know if this is related or not, but it occured to me or my
coworkers several times that a given scm file was not recompiled to a
On Wed 01 Feb 2012 22:47, Mark H Weaver writes:
> The problem is that there's a case when 'search_path' leaves the
> 'stat_buf' uninitialized. If the provided 'filename' is an absolute
> pathname, then it simply returns this pathname unchanged without
> touching the 'stat_buf'. This is bad, bec
uess there's another bug still lurking,
or perhaps this fix is incorrect.
Reviews solicited.
Thanks,
Mark
>From b3e8ae048d87ab28c90b1c1c635a5116ee6f080c Mon Sep 17 00:00:00 2001
From: Mark H Weaver
Date: Wed, 1 Feb 2012 16:35:32 -0500
Subject: [PATCH] Fix search_path to fi
