Fix unified syntax in ARM inline assembly

GCC 4.9 always emits assembly with divided syntax. Setting unified
syntax in inline assembly causes the assembler to complain about
GCC's generated code, because the directive extends past the scope
of the inline asm. Fix this by setting divided mode at the end of
the inline assembly block.

The assembler directives are hidden behind macros because later
versions of GCC won't need this workaround: they can be told to
use the unified syntax with -masm-syntax-unified.

Change-Id: Ic09e729e5bbb6fd44d08dac348daf6f55c75d7d8
This commit is contained in:
Aidan MacDonald 2023-03-23 18:16:15 +00:00
parent 86429dbf1e
commit 58b2e45782
7 changed files with 28 additions and 9 deletions

View file

@ -1005,6 +1005,16 @@ Lyre prototype 1 */
#define ROCKBOX_STRICT_ALIGN 1
#endif
/*
* These macros are for switching on unified syntax in inline assembly.
* Older versions of GCC emit assembly in divided syntax with no option
* to enable unified syntax.
*
* FIXME: This needs to be looked at after the toolchain bump
*/
#define BEGIN_ARM_ASM_SYNTAX_UNIFIED ".syntax unified\n"
#define END_ARM_ASM_SYNTAX_UNIFIED ".syntax divided\n"
#if defined(CPU_ARM) && defined(__ASSEMBLER__)
.syntax unified
/* ARMv4T doesn't switch the T bit when popping pc directly, we must use BX */