This is the mail archive of the
mailing list for the newlib project.
Re: [Patch, microblaze]: Add config/microblaze.mt for target_makefile_frag
- From: Michael Eager <eager at eagerm dot com>
- To: David Holsgrove <david dot holsgrove at xilinx dot com>
- Cc: newlib at sourceware dot org, edgari at xilinx dot com, jwilliams at xilinx dot com, nmekala at xilinx dot com
- Date: Sat, 27 Jul 2013 08:43:49 -0700
- Subject: Re: [Patch, microblaze]: Add config/microblaze.mt for target_makefile_frag
- References: <CAM=EW8aKiBZehGSUXE7PCO9mD7nRfFWjHYV=ZemJSf6CUSiH_A at mail dot gmail dot com> <5dc33039-a0eb-4b2e-a769-eca7578fed6b at DB9EHSMHS003 dot ehs dot local> <51F2960D dot 8060805 at eagerm dot com>
On 07/26/13 08:30, Michael Eager wrote:
On 07/26/13 07:58, David Holsgrove wrote:
Mirror MIPS method of creating copy of default.mt which drops the
compilation of generic sbrk.c to instead continue using the microblaze
I'll apply this and let you know if it corrects the gcc regressions.
This patch does correct the gcc test suite regressions.
I think that this is not the right way to fix this problem. This
patch hides the underlying problem caused by the previous patch.
There should be no need to reference generic functions in a makefile
fragment for target-specific functions. No other target, except for
mips (and now microblaze) does this. Every other target builds the
target-specific functions correctly instead of the generic functions,
and, as far as I am aware, so did microblaze before the patch was
applied which built Newlib incorrectly.
Michael Eager email@example.com
1960 Park Blvd., Palo Alto, CA 94306 650-325-8077