diff options
| author | Kyrylo Tkachov <kyrylo.tkachov@arm.com> | 2018-12-07 17:05:21 +0000 |
|---|---|---|
| committer | Kyrylo Tkachov <ktkachov@gcc.gnu.org> | 2018-12-07 17:05:21 +0000 |
| commit | 886f092f59d8a903fb5b99e757df5e4523dae825 (patch) | |
| tree | f915a21954bebd42a5981532532d5398d1a2354a /gcc/tree-nested.c | |
| parent | c844c4028f5e8e1370f889db18042caeb83518fc (diff) | |
| download | gcc-886f092f59d8a903fb5b99e757df5e4523dae825.zip gcc-886f092f59d8a903fb5b99e757df5e4523dae825.tar.gz gcc-886f092f59d8a903fb5b99e757df5e4523dae825.tar.bz2 | |
[AArch64][2/2] Add sve_width -moverride tunable
On top of the previous patch that implements TARGET_ESTIMATED_POLY_VALUE
and adds an sve_width tuning field to the CPU structs, this patch implements
an -moverride knob to adjust this sve_width field to allow for experimentation.
Again, reminder that this only has an effect when compiling for VLA-SVE that is,
without msve-vector-bits=<foo>. This just adjusts tuning heuristics in the compiler,,
like profitability thresholds for vectorised versioned loops, and others.
It can be used, for example like -moverride=sve_width=256 to set the sve_width
tuning field to 256. Widths outside of the accepted SVE widths [128 - 2048] are rejected
as you'd expect.
* config/aarch64/aarch64.c (aarch64_tuning_override_functions): Add
sve_width entry.
(aarch64_parse_sve_width_string): Define.
* gcc.target/aarch64/sve/override_sve_width_1.c: New test.
From-SVN: r266898
Diffstat (limited to 'gcc/tree-nested.c')
0 files changed, 0 insertions, 0 deletions