Hi,
the frequency of self recursive call can never be 1 or more or the
function would never finish. Yet we could do that based on static
guesses and that may confuse IPA passes expondentially increasing
frequency of functions.
This patch fixes it by simply capping the BB containing self recursive
call. This is not very nice (profile will be inconsistent) but I do not
see how to do it better short of repeating the frequency propagation and
slowly decreasing probabilities of the control dependent edges. That
would probably buy little compared to this hack.
Bootstrapped/regtested x86_64-linux, comitted.
Honza
* predict.c (estimate_bb_frequencies): Cap recursive calls by 90%.
diff --git a/gcc/predict.c b/gcc/predict.c
index a7ae977c866..0a317a7a4ac 100644
--- a/gcc/predict.c
+++ b/gcc/predict.c
@@ -3892,7 +3892,30 @@ estimate_bb_frequencies (bool force)
cfun->cfg->count_max = profile_count::uninitialized ();
FOR_BB_BETWEEN (bb, ENTRY_BLOCK_PTR_FOR_FN (cfun), NULL, next_bb)
{
- sreal tmp = BLOCK_INFO (bb)->frequency * freq_max + sreal (1, -1);
+ sreal tmp = BLOCK_INFO (bb)->frequency;
+ if (tmp >= 1)
+ {
+ gimple_stmt_iterator gsi;
+ tree decl;
+
+ /* Self recursive calls can not have frequency greater than 1
+ or program will never terminate. This will result in an
+ inconsistent bb profile but it is better than greatly confusing
+ IPA cost metrics. */
+ for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi))
+ if (is_gimple_call (gsi_stmt (gsi))
+ && (decl = gimple_call_fndecl (gsi_stmt (gsi))) != NULL
+ && recursive_call_p (current_function_decl, decl))
+ {
+ if (dump_file)
+ fprintf (dump_file, "Dropping frequency of recursive call"
+ " in bb %i from %f\n", bb->index,
+ tmp.to_double ());
+ tmp = (sreal)9 / (sreal)10;
+ break;
+ }
+ }
+ tmp = tmp * freq_max + sreal (1, -1);
profile_count count = profile_count::from_gcov_type (tmp.to_int ());
/* If we have profile feedback in which this function was never
