This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH] Add generic HAVE_RM_CTX implementation
- From: Siddhesh Poyarekar <siddhesh dot poyarekar at gmail dot com>
- To: Wilco <wdijkstr at arm dot com>
- Cc: Siddhesh Poyarekar <siddhesh at redhat dot com>, GNU C Library <libc-alpha at sourceware dot org>, Marcus Shawcroft <marcus dot shawcroft at gmail dot com>
- Date: Mon, 12 May 2014 18:56:52 +0530
- Subject: Re: [PATCH] Add generic HAVE_RM_CTX implementation
- Authentication-results: sourceware.org; auth=none
- References: <000101cf58af$83900a90$8ab01fb0$ at com> <20140430113915 dot GZ10922 at spoyarek dot pnq dot redhat dot com> <000601cf6479$a0d81590$e28840b0$ at com> <20140512060235 dot GD1224 at spoyarek dot pnq dot redhat dot com> <001e01cf6dda$1a976020$4fc62060$ at com>
On 12 May 2014 17:32, Wilco <email@example.com> wrote:
> Marcus (AArch64) already reviewed it.
I don't see his review on this thread; is it on another thread?
> I agree we should avoid PLTs for internal calls. However I'm certain
> most targets wouldn't build if I used __fegetenv etc. libm_hidden_ver
> is used inconsistently, so this is yet another area that needs a cleanup.
Fair enough. We can look at it as a separate cleanup then.
> Solving that wasn't the purpose of my patch - in fact it is consistent
> with the rest of math_private.h.
> Btw can we agree on what the rules are? Eg.
> * Every exported symbol should have a libm_hidden_ver.
> * All internal calls must use the hidden symbol.
An exported symbol should have a hidden alias if it is called
internally (and internal calls should use that hidden symbol), to
avoid the extra PLT dereference in the internal call. So it is not
necessary to define a hidden alias for *every* exported symbol.