This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH] Drop PER_THREAD conditionals.
- From: OndÅej BÃlka <neleai at seznam dot cz>
- To: Siddhesh Poyarekar <siddhesh at redhat dot com>
- Cc: libc-alpha at sourceware dot org
- Date: Sun, 8 Dec 2013 01:15:34 +0100
- Subject: Re: [PATCH] Drop PER_THREAD conditionals.
- Authentication-results: sourceware.org; auth=none
- References: <20131206000402 dot GA17751 at domone dot podge> <20131206034209 dot GD14845 at spoyarek dot pnq dot redhat dot com>
On Fri, Dec 06, 2013 at 09:12:09AM +0530, Siddhesh Poyarekar wrote:
> On Fri, Dec 06, 2013 at 01:04:02AM +0100, OndÅej BÃlka wrote:
> > Hi, when I looked at malloc code I wondered on what conditions
> > we compile malloc with PER_THREAD. As in Makefile we have PER_THREAD
> > enabled unconditionally. There may be reason to keep a PER_THREAD but
> > for simplicity of implementation I would just remove it.
> > Could somebody shed light on this.
> As a Fedora maintainer I agree with removing [!PER_THREAD] bits since
> we don't change that define and it makes the code easier to read. It
> made sense to keep it when the PER_THREAD bits were new, to test bugs
> against the older implementation for comparison, but I believe the
> PER_THREAD bits are mature enough now to get rid of the alternative.
> Other distribution maintainers should also pitch in with an opinion on
> this since I don't know if anyone actually overrides that define in
> their package.
> I don't think the ChangeLog is correct, but the patch itself looks OK.
So how should changelog look like?