On 11/26/25 1:11 PM, Dimitar Dimitrov wrote:
Starting with r16-4438-ga93f80feeef744, the edges started to be sorted
in ascending order. Thus the most likely branches were deprioritized
instead of optimized to fallthroughs.
Fix by restoring the sorting order prior to r16-4438-ga93f80feeef744.
There are no regressions for C and C++ on x86_64-pc-linux-gnu.
The new tests fail for the respective targets without this patch, and
pass with it.
PR rtl-optimization/122675
gcc/ChangeLog:
* bb-reorder.cc (edge_order): Fix BB edge ordering to be
descending.
gcc/testsuite/ChangeLog:
* gcc.target/aarch64/pr122675-1.c: New test.
* gcc.target/i386/pr122675-1.c: New test.
* gcc.target/riscv/pr122675-1.c: New test.
So the comment for the function says we want to order with the highest
execution frequency first, this patch would seem to do the opposite.
Similarly at the call site from within reorder_basic_blocks_simple, the
expectation is the most desirable (highest priority?) edges should be first.
At the least I think updates to the comments are advisable. Explaining
why descending order is preferred seems highly advisable too.
jeff
