https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96770
Bug ID: 96770
Summary: -mpure-code produces suboptimal code for relocations
with small offset for thumb-1
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: clyon at gcc dot gnu.org
Target Milestone: ---
As discussed in PR94538, -mpure-code produces suboptimal code for relocations
with small offset for thumb-1.
int arr[10];
int *f4 (void) { return &arr[1]; }
Compiled with -O2 -mpure-code,
-mcpu=cortex-m0:
f4:
movs r3, #:upper8_15:#.LC0
lsls r3, #8
adds r3, #:upper0_7:#.LC0
lsls r3, #8
adds r3, #:lower8_15:#.LC0
lsls r3, #8
adds r3, #:lower0_7:#.LC0
@ sp needed
ldr r0, [r3]
adds r0, r0, #4
bx lr
We should avoid the extra load from the literal pool (related to PR96767), and
the 'adds r0,r0,4'.
-mcpu=cortex-m23:
f4:
movw r0, #:lower16:.LANCHOR0
@ sp needed
movt r0, #:upper16:.LANCHOR0
adds r0, r0, #4
bx lr
For reference, -mcpu=cortex-m3 produces:
f4:
movw r0, #:lower16:.LANCHOR0+4
movt r0, #:upper16:.LANCHOR0+4
bx lr
We should generate the same code for cortex-m23.
