On 8/25/21 7:49 PM, Martin Sebor wrote:
On 8/25/21 1:34 PM, Andrew MacLeod wrote:
FWIW, I see duplicate output in the EVRP1 dump that looks just like
it could be the output under Non-varying global ranges. I wonder if
it's the result of the block in ranger_cache::fill_block_cache guarded
by if (DEBUG_RANGE_CACHE).
Martin
The cache debug mechanism hasnt been fully reworked yet... i moved
GORI and the ranger to the trace class.. I haven't decided yet what I
want to change in the cache. its a bit dated.
regardless, if your sources are up to date, DEBUG_RANGE_CACHE is now
guarded by:
#define DEBUG_RANGE_CACHE (dump_file && (param_evrp_mode &
EVRP_MODE_CACHE) \
== EVRP_MODE_CACHE)
so you'd only see that output if you specify --param=evrp-mode=cache
or debug
It's not a big deal but the output I noticed is actually coming
from gimple_ranger::dump_bb (). In the test I happened to look
at it shows up just before Non-varying global ranges: It's
attached just to show you what I mean.
Martin
If you ask for details, you get a ranger IL dump with ranges after the
pass... Its difficult to show any contextual range calculations otherwise.
