procyberian/gcc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

tree-switch-conversion: don't apply switch size limit on jump tables

Browse source 
commit 56946c801a ("gimple: Add limit after which slower switchlower
algs are used [PR117091] [PR117352]") introduced a limit on the number
of cases of a switch. It also bails out on finding jump tables if the
switch is too large. This introduces a compile time regression during
bootstrap. A riscv bootstrap takes hours longer. Particularly
insn-attrtab.cc will take hours instead of minutes. Fix this by not
applying the switch size limit on jump tables.

An alternative would be to implement greedy switch clustering for jump
tables as is done for switch bitmap clustering.

gcc/ChangeLog:

	PR tree-optimization/118032
	* tree-switch-conversion.cc (jump_table_cluster::find_jump_tables):
	Remove param_switch_lower_slow_alg_max_cases check.
This commit is contained in:
Mark Wielaard 2025-01-05 18:00:36 +01:00
parent 830bead485
commit 668cad04b1
1 changed files with 0 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
gcc/tree-switch-conversion.cc
View file

@ -1643,10 +1643,6 @@ jump_table_cluster::find_jump_tables (vec<cluster *> &clusters)
unsigned l = clusters.length ();
/* Note: l + 1 is the number of cases of the switch. */
if (l + 1 > (unsigned) param_switch_lower_slow_alg_max_cases)
return clusters.copy ();
auto_vec<min_cluster_item> min;
min.reserve (l + 1);