Re: fix ltdldatafiles corruption bug [libtool--gary--1.0--patch-14]

2005-04-03 Thread Ralf Wildenhues
* Gary V. Vaughan wrote on Fri, Apr 01, 2005 at 07:40:37PM CEST: Okay to commit to HEAD, and backport to branch-2-0? Makes new tests 1 and 2 fail silently (i.e., they don't work because they use -I, but report success). Fixes my `libtoolize' corruption. Getting the value of

Re: fix ltdldatafiles corruption bug [libtool--gary--1.0--patch-14]

2005-04-03 Thread Ralf Wildenhues
Sorry.. * Ralf Wildenhues wrote on Sun, Apr 03, 2005 at 10:10:43AM CEST: * Gary V. Vaughan wrote on Fri, Apr 01, 2005 at 07:40:37PM CEST: With this applied, the following test shows some problems: current `serial' computation is flawed (sorry, I should've checked more carefully before),

fix ltdldatafiles corruption bug [libtool--gary--1.0--patch-14]

2005-04-01 Thread Gary V. Vaughan
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Okay to commit to HEAD, and backport to branch-2-0? Getting the value of ltdldatafiles from a submake was not portable, and had problems with GNU make-3.80 at least. This changeset pulls the installation rules for