On Sat, 10 Aug 2013 11:31:19 +0200 Gary Jennejohn <gljennj...@googlemail.com> wrote:
> On Sat, 10 Aug 2013 11:10:40 +0200 > Rainer Hurling <rhur...@gwdg.de> wrote: > > > Yes, I can confirm, that it builds, installs and runs fine for me. > > > > The patch should be placed as > > x11/nvidia-driver/files/patch-src__nvidia_subr.c, shoudn't it? > > > > Many thanks for this work. > > > > Thanks for testing. > > Yes, putting the patch into files/ with that name works also and > is much simpler than the steps I outlined. > Placing the patch in files as recommended here doesn't play well with the obvious intention of the REINPLACE command: the patch only applies to 319.25. I use the cutting edge 325.15. The patch doesn't apply since some lines shifted - here comes the tricky REINPLACE part of the Makefile in place. I simply adapted your patches discussed and introduced here and "adapted" the REINPLACE statements/pattern around line 160 in the toplevel Makefile of port x11/nvidia-driver. Find the patch attached - I forgot to raise PORTREVISION=1. I'm now sending this email from the prior crashing box with the patch discussed applied via the Makefile to 325.15. Thanks a lot for the fast help. Regards, Oliver
diff -Nur nvidia-driver.orig/Makefile nvidia-driver/Makefile
--- nvidia-driver.orig/Makefile 2013-08-10 14:46:08.000000000 +0200
+++ nvidia-driver/Makefile 2013-08-10 14:41:52.000000000 +0200
@@ -158,8 +158,8 @@
.endif
# Catch up with KVA space allocation API changes in recent -CURRENT
.if ${OSVERSION} > 1000040
- ${REINPLACE_CMD} -e 's/kmem_free(kernel_map,/kva_free(/ ; \
- /kmem_alloc_contig/s/map/arena/' ${WRKSRC}/src/nvidia_subr.c
+ ${REINPLACE_CMD} -e 's/kmem_free(kernel_map,/kmem_free(kmem_arena,/ ; \
+ /kmem_alloc_contig/s/kernel_map/kmem_arena/'
${WRKSRC}/src/nvidia_subr.c
.endif
# Process OPTIONS
.if ${PORT_OPTIONS:MFREEBSD_AGP}
signature.asc
Description: PGP signature
