[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 6/8] vl.c: Use parse_uint_full() for NUMA nodeid
From: |
Eduardo Habkost |
Subject: |
Re: [Qemu-devel] [PATCH 6/8] vl.c: Use parse_uint_full() for NUMA nodeid |
Date: |
Wed, 16 Jan 2013 16:18:35 -0200 |
User-agent: |
Mutt/1.5.21 (2010-09-15) |
On Wed, Jan 16, 2013 at 10:54:34AM -0700, Eric Blake wrote:
> On 01/16/2013 10:42 AM, Eduardo Habkost wrote:
> >
> > Or, we could simply change the meaning of "nodeid=010" and document it
> > in the commit message and QEMU ChageLog. Which option do the QEMU
> > maintainers prefer?
>
> I'm not a maintainer, so anyone is free to overrule me; but I'm
> personally fine with just documenting that the semantic change is
> intentional. It's enough of a corner case that probably no one was
> relying on leading 0 still giving decimal parsing (libvirt certainly
> wasn't relying on it], and the flexibility of allowing hex parsing is
> useful. But do document it in the commit message; a silent change is
> not fair.
I will add a 'base' parameter to parse_uint*() anyway, as it may be
useful somewhere else. Then keeping the existing behavior will be
trivial, so I will keep it by now.
Later, when changing the code to use QemuOpts, we may decide to change
the behavior (to keep the code simpler) and document the semantic
change.
--
Eduardo
[Qemu-devel] [PATCH 2/8] vl.c: Fix off-by-one bug when handling "-numa node" argument, Eduardo Habkost, 2013/01/16
[Qemu-devel] [PATCH 7/8] vl.c: Extract -numa "cpus" parsing to separate function, Eduardo Habkost, 2013/01/16
[Qemu-devel] [PATCH 3/8] vl.c: Abort on unknown -numa option type, Eduardo Habkost, 2013/01/16
[Qemu-devel] [PATCH 8/8] vl.c: validate -numa "cpus" parameter properly, Eduardo Habkost, 2013/01/16
[Qemu-devel] [PATCH 1/8] cutils: unsigned int parsing functions, Eduardo Habkost, 2013/01/16