On 11/20/19 10:12 PM, David Malcolm wrote:
On Wed, 2019-11-20 at 16:07 +0100, Martin Liška wrote:
On 11/20/19 4:02 PM, David Malcolm wrote:
On Wed, 2019-11-20 at 15:51 +0100, Martin Liška wrote:
On 11/20/19 3:49 PM, David Malcolm wrote:
On Wed, 2019-11-06 at 11:30 +0100, Martin Liska wrote:
gcc/ChangeLog:
2019-11-06 Martin Liska <mli...@suse.cz>
* Makefile.in: Include params.opt.
* flag-types.h (enum parloops_schedule_type): Add
parloops_schedule_type used in params.opt.
* params.opt: New file.
---
gcc/Makefile.in | 2 +-
gcc/flag-types.h | 11 +
gcc/params.opt | 967
+++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 979 insertions(+), 1 deletion(-)
create mode 100644 gcc/params.opt
Is your params.def -> params.opt script available
somewhere? (sorry if
I missed it, I didn't see it looking over the patch kit)
Hi David.
(I'm rebasing my static analyzer patch kit and am about to
convert
my
params.def additions to be params.opt additions, though I'm
only
adding
4 params, so hopefully doing it by hand will be trivial)
No, it's not public. Please convert it by hand, it will be faster
;)
Fair enough.
Does this new machinery mean we can have per-frontend params (by
putting the Param options in the pertinent .opt file) ? (not sure
what
that will do to LTO)
I guess so. Note that now parameters are first class citizens same as
options.
So that, having a per-FE should work fine.
Thanks.
FWIW, given my branch adds support for in-tree plugins (which can add
options), I've added them to the plugin.opt file:
Which seams reasonable to me! We already have a target-specific parameter
that lives in the corresponding target .opt file:
gcc/config/aarch64/aarch64.opt:-param=aarch64-sve-compare-costs=
Martin
https://gcc.gnu.org/ml/gcc-patches/2019-11/msg02030.html
Dave
