gcc: merge upstream fix for PR 83496 (fixes FS#814)

This bug was causing miscompiled code related to branch delay slots

Signed-off-by: Felix Fietkau <nbd@nbd.name>
This commit is contained in:
Felix Fietkau 2018-02-25 20:25:20 +01:00
parent 38dc678166
commit 25aaff9100

View file

@ -0,0 +1,136 @@
From: ebotcazou <ebotcazou@138bc75d-0d04-0410-961f-82ee72b054a4>
Date: Mon, 26 Feb 2018 16:29:30 +0000
Subject: [PATCH] PR rtl-optimization/83496 * reorg.c
(steal_delay_list_from_target): Change REDUNDANT array from booleans to
RTXes. Call fix_reg_dead_note on every non-null element.
(steal_delay_list_from_fallthrough): Call fix_reg_dead_note on a
redundant insn, if any. (relax_delay_slots): Likewise.
(update_reg_unused_notes): Rename REDUNDANT_INSN to OTHER_INSN.
git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@257996 138bc75d-0d04-0410-961f-82ee72b054a4
---
create mode 120000 gcc/testsuite/gcc.c-torture/execute/20180226-1.c
--- a/gcc/reorg.c
+++ b/gcc/reorg.c
@@ -1035,7 +1035,8 @@ check_annul_list_true_false (int annul_t
static void
steal_delay_list_from_target (rtx_insn *insn, rtx condition, rtx_sequence *seq,
- vec<rtx_insn *> *delay_list, resources *sets,
+ vec<rtx_insn *> *delay_list,
+ struct resources *sets,
struct resources *needed,
struct resources *other_needed,
int slots_to_fill, int *pslots_filled,
@@ -1048,7 +1049,7 @@ steal_delay_list_from_target (rtx_insn *
int used_annul = 0;
int i;
struct resources cc_set;
- bool *redundant;
+ rtx_insn **redundant;
/* We can't do anything if there are more delay slots in SEQ than we
can handle, or if we don't know that it will be a taken branch.
@@ -1087,7 +1088,7 @@ steal_delay_list_from_target (rtx_insn *
if (! targetm.can_follow_jump (insn, seq->insn (0)))
return;
- redundant = XALLOCAVEC (bool, XVECLEN (seq, 0));
+ redundant = XALLOCAVEC (rtx_insn *, XVECLEN (seq, 0));
for (i = 1; i < seq->len (); i++)
{
rtx_insn *trial = seq->insn (i);
@@ -1151,7 +1152,10 @@ steal_delay_list_from_target (rtx_insn *
we therefore decided not to copy. */
for (i = 1; i < seq->len (); i++)
if (redundant[i])
- update_block (seq->insn (i), insn);
+ {
+ fix_reg_dead_note (redundant[i], insn);
+ update_block (seq->insn (i), insn);
+ }
/* Show the place to which we will be branching. */
*pnew_thread = first_active_target_insn (JUMP_LABEL (seq->insn (0)));
@@ -1198,6 +1202,7 @@ steal_delay_list_from_fallthrough (rtx_i
for (i = 1; i < seq->len (); i++)
{
rtx_insn *trial = seq->insn (i);
+ rtx_insn *prior_insn;
/* If TRIAL sets CC0, stealing it will move it too far from the use
of CC0. */
@@ -1209,8 +1214,9 @@ steal_delay_list_from_fallthrough (rtx_i
break;
/* If this insn was already done, we don't need it. */
- if (redundant_insn (trial, insn, *delay_list))
+ if ((prior_insn = redundant_insn (trial, insn, *delay_list)))
{
+ fix_reg_dead_note (prior_insn, insn);
update_block (trial, insn);
delete_from_delay_slot (trial);
continue;
@@ -1790,15 +1796,14 @@ fix_reg_dead_note (rtx_insn *start_insn,
}
}
-/* Delete any REG_UNUSED notes that exist on INSN but not on REDUNDANT_INSN.
+/* Delete any REG_UNUSED notes that exist on INSN but not on OTHER_INSN.
This handles the case of udivmodXi4 instructions which optimize their
- output depending on whether any REG_UNUSED notes are present.
- we must make sure that INSN calculates as many results as REDUNDANT_INSN
- does. */
+ output depending on whether any REG_UNUSED notes are present. We must
+ make sure that INSN calculates as many results as OTHER_INSN does. */
static void
-update_reg_unused_notes (rtx_insn *insn, rtx redundant_insn)
+update_reg_unused_notes (rtx_insn *insn, rtx other_insn)
{
rtx link, next;
@@ -1810,8 +1815,7 @@ update_reg_unused_notes (rtx_insn *insn,
|| !REG_P (XEXP (link, 0)))
continue;
- if (! find_regno_note (redundant_insn, REG_UNUSED,
- REGNO (XEXP (link, 0))))
+ if (!find_regno_note (other_insn, REG_UNUSED, REGNO (XEXP (link, 0))))
remove_note (insn, link);
}
}
@@ -2324,9 +2328,8 @@ follow_jumps (rtx label, rtx_insn *jump,
taken and THREAD_IF_TRUE is set. This is used for the branch at the
end of a loop back up to the top.
- OWN_THREAD and OWN_OPPOSITE_THREAD are true if we are the only user of the
- thread. I.e., it is the fallthrough code of our jump or the target of the
- jump when we are the only jump going there.
+ OWN_THREAD is true if we are the only user of the thread, i.e. it is
+ the target of the jump when we are the only jump going there.
If OWN_THREAD is false, it must be the "true" thread of a jump. In that
case, we can only take insns from the head of the thread for our delay
@@ -3117,7 +3120,7 @@ relax_delay_slots (rtx_insn *first)
/* Look at every JUMP_INSN and see if we can improve it. */
for (insn = first; insn; insn = next)
{
- rtx_insn *other;
+ rtx_insn *other, *prior_insn;
bool crossing;
next = next_active_insn (insn);
@@ -3223,8 +3226,9 @@ relax_delay_slots (rtx_insn *first)
/* See if the first insn in the delay slot is redundant with some
previous insn. Remove it from the delay slot if so; then set up
to reprocess this insn. */
- if (redundant_insn (pat->insn (1), delay_insn, vNULL))
+ if ((prior_insn = redundant_insn (pat->insn (1), delay_insn, vNULL)))
{
+ fix_reg_dead_note (prior_insn, insn);
update_block (pat->insn (1), insn);
delete_from_delay_slot (pat->insn (1));
next = prev_active_insn (next);