On 2016-09-05 12:31, Rob Browning wrote: > Aurelien Jarno <aure...@debian.org> writes: > > > The patch i have attached is actually a backport of the above patch. It > > was present in various branches, so I might have backported one with a > > different commit number, but in practice it's the same. > > Assuming you mean the gmalloc patch, do you know if there's a very > similar patch upstream? The one you've attached is quite a bit > different than the one I quoted.
Yes, the 0020-Always-define-gmalloc-etc.-in-src-gmalloc.c.patch actually comes from upstream commit 4b1436b702d56eedd27a0777fc7232cdfb7ac4f6. It looks different given that the emacs 24 branch do not include the commits to add hybrid malloc [2] and thus some #undef #define are needed instead of changing the HYBRID_MALLOC conditional. OTOH, you might want to consider my patch as a mix of [1] and [2]. Aurelien [1] http://git.savannah.gnu.org/cgit/emacs.git/commit/?id=4b1436b702d56eedd27a0777fc7232cdfb7ac4f6 [2] http://git.savannah.gnu.org/cgit/emacs.git/commit/?id=ea652500776aacbb8cb64f9ecca16a2d2c7add80 http://git.savannah.gnu.org/cgit/emacs.git/commit/?id=a4817d834e7d125d31049dfb6fd0a0df4782bad0 -- Aurelien Jarno GPG: 4096R/1DDD8C9B aurel...@aurel32.net http://www.aurel32.net
