[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Issue 1890 in lilypond: Compiler warnings in make
From: |
Keith OHara |
Subject: |
Re: Issue 1890 in lilypond: Compiler warnings in make |
Date: |
Sat, 17 Sep 2011 04:07:39 +0000 (UTC) |
User-agent: |
Loom/3.14 (http://gmane.org/) |
<lilypond <at> googlecode.com> writes:
> Our discussions about GOP PROP 5 (make) led to the conclusion that compiler
> warnings are not desirable and can point to potential bugs. My recent
> builds have shown about 195 warnings, so I'm opening this issue to allow
> these to be considered. This is on 64 bit ubuntu - not sure if that's what
> is causing some of these.
These warnings look like they come from a system with a 64-bit address space
(size_t) but a 32-bit natural word size (int). Most of them say the same
thing: the memory could hold an array so large (more than two thousand
million elements) that the processor cannot fit an element count in a
normal-size (int) register.
We could explicitly promise to never have an array with over two thousand
million elements by defining vsize differently. I would ignore the warnings
until you start doing a lot of compiling on this system, and even then I
would try to suppress this type of warning before changing the source code.