Hi Guys,
I am applying the patch below to add support for G10 versions of the
RL78 assembler multiply and divide functions to the libgcc library.
Cheers
Nick
libgcc/ChangeLog
2014-11-13 Nick Clifton <ni...@redhat.com>
* config/rl78/divmodhi.S: Add support for the G10 architecture.
Use START_FUNC and END_FUNC macros to enable linker garbage
collection.
* config/rl78/divmodqi.S: Likewise.
* config/rl78/divmodsi.S: Likewise.
* config/rl78/mulsi3.S: Likewise.
* config/rl78/lib2div.c: Remove G10 functions.
* config/rl78/lib2muls.c: Likewise.
* config/rl78/t-rl8 (HOST_LIBGCC2_CFLAGS): Define.
* config/rl78/vregs.h (START_FUNC): New macro.
(END_FUNC): New macro.
Index: libgcc/config/rl78/divmodhi.S
===================================================================
--- libgcc/config/rl78/divmodhi.S (revision 217462)
+++ libgcc/config/rl78/divmodhi.S (working copy)
@@ -23,11 +23,9 @@
see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
<http://www.gnu.org/licenses/>. */
-#ifndef __RL78_G10__
-
#include "vregs.h"
- .macro make_generic which,need_result
+.macro MAKE_GENERIC which,need_result
.if \need_result
quot = r8
@@ -53,12 +51,12 @@
bitB0 = bit
bitB1 = bit+1
-#if 1
#define bit bc
#define bitB0 c
#define bitB1 b
-#endif
+ START_FUNC __generic_hidivmod\which
+
num_lt_den\which:
.if \need_result
movw r8, #0
@@ -72,12 +70,13 @@
;; to store sign information that must remain intact
.if \need_result
+ .global __generic_hidiv
+__generic_hidiv:
-generic_div:
-
.else
-generic_mod:
+ .global __generic_himod
+__generic_himod:
.endif
@@ -193,42 +192,40 @@
main_loop_done\which:
ret
- .endm
+ END_FUNC __generic_hidivmod\which
+.endm
+;----------------------------------------------------------------------
- make_generic _d 1
- make_generic _m 0
+ MAKE_GENERIC _d 1
+ MAKE_GENERIC _m 0
;----------------------------------------------------------------------
- .global ___udivhi3
- .type ___udivhi3,@function
-___udivhi3:
+START_FUNC ___udivhi3
;; r8 = 4[sp] / 6[sp]
- call $!generic_div
+ call $!__generic_hidiv
ret
- .size ___udivhi3, . - ___udivhi3
+END_FUNC ___udivhi3
- .global ___umodhi3
- .type ___umodhi3,@function
-___umodhi3:
+START_FUNC ___umodhi3
;; r8 = 4[sp] % 6[sp]
- call $!generic_mod
+ call $!__generic_himod
ret
- .size ___umodhi3, . - ___umodhi3
+END_FUNC ___umodhi3
;----------------------------------------------------------------------
- .macro neg_ax
+.macro NEG_AX
movw hl, ax
movw ax, #0
subw ax, [hl]
movw [hl], ax
- .endm
+.endm
- .global ___divhi3
- .type ___divhi3,@function
-___divhi3:
+;----------------------------------------------------------------------
+
+START_FUNC ___divhi3
;; r8 = 4[sp] / 6[sp]
movw de, #0
mov a, [sp+5]
@@ -237,7 +234,7 @@
mov a, [sp+7]
mov1 cy, a.7
bc $div_signed_den
- call $!generic_div
+ call $!__generic_hidiv
ret
div_signed_num:
@@ -244,7 +241,7 @@
;; neg [sp+4]
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mov d, #1
mov a, [sp+7]
mov1 cy, a.7
@@ -253,10 +250,10 @@
;; neg [sp+6]
movw ax, sp
addw ax, #6
- neg_ax
+ NEG_AX
mov e, #1
div_unsigned_den:
- call $!generic_div
+ call $!__generic_hidiv
mov a, d
cmp0 a
@@ -264,13 +261,13 @@
;; We have to restore the numerator [sp+4]
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mov a, d
div_skip_restore_num:
xor a, e
bz $div_no_neg
movw ax, #r8
- neg_ax
+ NEG_AX
div_no_neg:
mov a, e
cmp0 a
@@ -277,15 +274,13 @@
bz $div_skip_restore_den
movw ax, sp
addw ax, #6
- neg_ax
+ NEG_AX
div_skip_restore_den:
ret
- .size ___divhi3, . - ___divhi3
+END_FUNC ___divhi3
- .global ___modhi3
- .type ___modhi3,@function
-___modhi3:
+START_FUNC ___modhi3
;; r8 = 4[sp] % 6[sp]
movw de, #0
mov a, [sp+5]
@@ -294,7 +289,7 @@
mov a, [sp+7]
mov1 cy, a.7
bc $mod_signed_den
- call $!generic_mod
+ call $!__generic_himod
ret
mod_signed_num:
@@ -301,7 +296,7 @@
;; neg [sp+4]
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mov d, #1
mov a, [sp+7]
mov1 cy, a.7
@@ -310,19 +305,19 @@
;; neg [sp+6]
movw ax, sp
addw ax, #6
- neg_ax
+ NEG_AX
mod_unsigned_den:
- call $!generic_mod
+ call $!__generic_himod
mov a, d
cmp0 a
bz $mod_no_neg
movw ax, #r8
- neg_ax
+ NEG_AX
;; Also restore numerator
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mod_no_neg:
mov a, e
cmp0 a
@@ -329,9 +324,7 @@
bz $mod_skip_restore_den
movw ax, sp
addw ax, #6
- neg_ax
+ NEG_AX
mod_skip_restore_den:
ret
- .size ___modhi3, . - ___modhi3
-
-#endif
+END_FUNC ___modhi3
Index: libgcc/config/rl78/divmodqi.S
===================================================================
--- libgcc/config/rl78/divmodqi.S (revision 217462)
+++ libgcc/config/rl78/divmodqi.S (working copy)
@@ -23,11 +23,9 @@
see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
<http://www.gnu.org/licenses/>. */
-#ifndef __RL78_G10__
-
#include "vregs.h"
- .macro make_generic which,need_result
+.macro MAKE_GENERIC which,need_result
.if \need_result
quot = r8
@@ -41,12 +39,12 @@
bit = r14
.endif
-#if 1
#define bit b
#define den c
#define bitden bc
-#endif
+ START_FUNC __generic_qidivmod\which
+
num_lt_den\which:
.if \need_result
mov r8, #0
@@ -72,12 +70,13 @@
;; to store sign information that must remain intact
.if \need_result
+ .global __generic_qidiv
+__generic_qidiv:
-generic_div:
-
.else
-generic_mod:
+ .global __generic_qimod
+__generic_qimod:
.endif
@@ -111,7 +110,8 @@
; while (den < num && !(den & (1L << BITS_MINUS_1)))
shift_den_bit\which:
- .macro sdb_one\which
+
+.macro SDB_ONE\which
mov a, den
mov1 cy,a.7
bc $enter_main_loop\which
@@ -125,10 +125,10 @@
;; bit <<= 1
shl bit, 1
- .endm
+.endm
- sdb_one\which
- sdb_one\which
+ SDB_ONE\which
+ SDB_ONE\which
br $shift_den_bit\which
@@ -164,42 +164,41 @@
main_loop_done\which:
ret
- .endm
+ END_FUNC __generic_qidivmod\which
+.endm
- make_generic _d 1
- make_generic _m 0
+;----------------------------------------------------------------------
+ MAKE_GENERIC _d 1
+ MAKE_GENERIC _m 0
+
;----------------------------------------------------------------------
- .global ___udivqi3
- .type ___udivqi3,@function
-___udivqi3:
+START_FUNC ___udivqi3
;; r8 = 4[sp] / 6[sp]
movw hl, sp
- br $!generic_div
- .size ___udivqi3, . - ___udivqi3
+ br $!__generic_qidiv
+END_FUNC ___udivqi3
- .global ___umodqi3
- .type ___umodqi3,@function
-___umodqi3:
+START_FUNC ___umodqi3
;; r8 = 4[sp] % 6[sp]
movw hl, sp
- br $!generic_mod
- .size ___umodqi3, . - ___umodqi3
+ br $!__generic_qimod
+END_FUNC ___umodqi3
;----------------------------------------------------------------------
- .macro neg_ax
+.macro NEG_AX
movw hl, ax
mov a, #0
sub a, [hl]
mov [hl], a
- .endm
+.endm
- .global ___divqi3
- .type ___divqi3,@function
-___divqi3:
+;----------------------------------------------------------------------
+
+START_FUNC ___divqi3
;; r8 = 4[sp] / 6[sp]
movw hl, sp
movw de, #0
@@ -209,7 +208,7 @@
mov a, [sp+6]
mov1 cy, a.7
bc $div_signed_den
- br $!generic_div
+ br $!__generic_qidiv
div_signed_num:
;; neg [sp+4]
@@ -227,7 +226,7 @@
mov [hl+6], a
mov e, #1
div_unsigned_den:
- call $!generic_div
+ call $!__generic_qidiv
mov a, d
cmp0 a
@@ -235,13 +234,13 @@
;; We have to restore the numerator [sp+4]
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mov a, d
div_skip_restore_num:
xor a, e
bz $div_no_neg
movw ax, #r8
- neg_ax
+ NEG_AX
div_no_neg:
mov a, e
cmp0 a
@@ -248,15 +247,13 @@
bz $div_skip_restore_den
movw ax, sp
addw ax, #6
- neg_ax
+ NEG_AX
div_skip_restore_den:
ret
- .size ___divqi3, . - ___divqi3
+END_FUNC ___divqi3
- .global ___modqi3
- .type ___modqi3,@function
-___modqi3:
+START_FUNC ___modqi3
;; r8 = 4[sp] % 6[sp]
movw hl, sp
movw de, #0
@@ -266,7 +263,7 @@
mov a, [hl+6]
mov1 cy, a.7
bc $mod_signed_den
- br $!generic_mod
+ br $!__generic_qimod
mod_signed_num:
;; neg [sp+4]
@@ -284,7 +281,7 @@
mov [hl+6], a
mov e, #1
mod_unsigned_den:
- call $!generic_mod
+ call $!__generic_qimod
mov a, d
cmp0 a
@@ -295,7 +292,7 @@
;; Also restore numerator
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mod_no_neg:
mov a, e
cmp0 a
@@ -302,9 +299,7 @@
bz $mod_skip_restore_den
movw ax, sp
addw ax, #6
- neg_ax
+ NEG_AX
mod_skip_restore_den:
ret
- .size ___modqi3, . - ___modqi3
-
-#endif
+END_FUNC ___modqi3
Index: libgcc/config/rl78/divmodsi.S
===================================================================
--- libgcc/config/rl78/divmodsi.S (revision 217462)
+++ libgcc/config/rl78/divmodsi.S (working copy)
@@ -23,11 +23,9 @@
see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
<http://www.gnu.org/licenses/>. */
-#ifndef __RL78_G10__
-
#include "vregs.h"
- .macro make_generic which,need_result
+.macro MAKE_GENERIC which,need_result
.if \need_result
quot = r8
@@ -69,6 +67,8 @@
bitB2 = bit+2
bitB3 = bit+3
+START_FUNC __generic_sidivmod\which
+
num_lt_den\which:
.if \need_result
movw r8, #0
@@ -100,12 +100,13 @@
;; to store sign information that must remain intact
.if \need_result
+ .global __generic_sidiv
+__generic_sidiv:
-generic_div:
-
.else
-generic_mod:
+ .global __generic_simod
+__generic_simod:
.endif
@@ -119,6 +120,14 @@
cmpw ax, [hl+8] ; numL
bh $num_lt_den\which
+#ifdef __RL78_G10__
+ movw ax, denL
+ push ax
+ movw ax, bitL
+ push ax
+ movw ax, bitH
+ push ax
+#else
sel rb2
push ax ; denL
; push bc ; denH
@@ -125,6 +134,7 @@
push de ; bitL
push hl ; bitH - stored in BC
sel rb0
+#endif
;; (quot,rem) = 16[sp] /% 20[sp]
@@ -360,6 +370,14 @@
bnz $main_loop_himode\which
main_loop_done_himode\which:
+#ifdef __RL78_G10__
+ pop ax
+ movw bitH, ax
+ pop ax
+ movw bitL, ax
+ pop ax
+ movw denL, ax
+#else
sel rb2
pop hl ; bitH - stored in BC
pop de ; bitL
@@ -366,35 +384,35 @@
; pop bc ; denH
pop ax ; denL
sel rb0
+#endif
ret
- .endm
+END_FUNC __generic_sidivmod\which
+.endm
- make_generic _d 1
- make_generic _m 0
+;----------------------------------------------------------------------
+ MAKE_GENERIC _d 1
+ MAKE_GENERIC _m 0
+
;----------------------------------------------------------------------
- .global ___udivsi3
- .type ___udivsi3,@function
-___udivsi3:
+START_FUNC ___udivsi3
;; r8 = 4[sp] / 8[sp]
- call $!generic_div
+ call $!__generic_sidiv
ret
- .size ___udivsi3, . - ___udivsi3
+END_FUNC ___udivsi3
- .global ___umodsi3
- .type ___umodsi3,@function
-___umodsi3:
+START_FUNC ___umodsi3
;; r8 = 4[sp] % 8[sp]
- call $!generic_mod
+ call $!__generic_simod
ret
- .size ___umodsi3, . - ___umodsi3
+END_FUNC ___umodsi3
;----------------------------------------------------------------------
- .macro neg_ax
+.macro NEG_AX
movw hl, ax
movw ax, #0
subw ax, [hl]
@@ -404,11 +422,11 @@
decw ax
subw ax, [hl+2]
movw [hl+2], ax
- .endm
+.endm
- .global ___divsi3
- .type ___divsi3,@function
-___divsi3:
+;----------------------------------------------------------------------
+
+START_FUNC ___divsi3
;; r8 = 4[sp] / 8[sp]
movw de, #0
mov a, [sp+7]
@@ -417,14 +435,14 @@
mov a, [sp+11]
mov1 cy, a.7
bc $div_signed_den
- call $!generic_div
+ call $!__generic_sidiv
ret
-
+
div_signed_num:
;; neg [sp+4]
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mov d, #1
mov a, [sp+11]
mov1 cy, a.7
@@ -433,10 +451,10 @@
;; neg [sp+8]
movw ax, sp
addw ax, #8
- neg_ax
+ NEG_AX
mov e, #1
div_unsigned_den:
- call $!generic_div
+ call $!__generic_sidiv
mov a, d
cmp0 a
@@ -444,13 +462,13 @@
;; We have to restore the numerator [sp+4]
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mov a, d
div_skip_restore_num:
xor a, e
bz $div_no_neg
movw ax, #r8
- neg_ax
+ NEG_AX
div_no_neg:
mov a, e
cmp0 a
@@ -458,15 +476,13 @@
;; We have to restore the denominator [sp+8]
movw ax, sp
addw ax, #8
- neg_ax
+ NEG_AX
div_skip_restore_den:
ret
- .size ___divsi3, . - ___divsi3
+END_FUNC ___divsi3
- .global ___modsi3
- .type ___modsi3,@function
-___modsi3:
+START_FUNC ___modsi3
;; r8 = 4[sp] % 8[sp]
movw de, #0
mov a, [sp+7]
@@ -475,14 +491,14 @@
mov a, [sp+11]
mov1 cy, a.7
bc $mod_signed_den
- call $!generic_mod
+ call $!__generic_simod
ret
-
+
mod_signed_num:
;; neg [sp+4]
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mov d, #1
mov a, [sp+11]
mov1 cy, a.7
@@ -491,20 +507,20 @@
;; neg [sp+8]
movw ax, sp
addw ax, #8
- neg_ax
+ NEG_AX
mov e, #1
mod_unsigned_den:
- call $!generic_mod
+ call $!__generic_simod
mov a, d
cmp0 a
bz $mod_no_neg
movw ax, #r8
- neg_ax
+ NEG_AX
;; We have to restore [sp+4] as well.
movw ax, sp
addw ax, #4
- neg_ax
+ NEG_AX
mod_no_neg:
.if 1
mov a, e
@@ -512,10 +528,8 @@
bz $mod_skip_restore_den
movw ax, sp
addw ax, #8
- neg_ax
+ NEG_AX
mod_skip_restore_den:
.endif
ret
- .size ___modsi3, . - ___modsi3
-
-#endif
+END_FUNC ___modsi3
Index: libgcc/config/rl78/lib2div.c
===================================================================
--- libgcc/config/rl78/lib2div.c (revision 217462)
+++ libgcc/config/rl78/lib2div.c (working copy)
@@ -34,41 +34,6 @@
#define C3B(a,b,c) a##b##c
#define C3(a,b,c) C3B(a,b,c)
-#ifdef __RL78_G10__
-
-#define UINT_TYPE uint32_type
-#define SINT_TYPE sint32_type
-#define BITS_MINUS_1 31
-#define NAME_MODE si
-
-#include "rl78-divmod.h"
-
-#undef UINT_TYPE
-#undef SINT_TYPE
-#undef BITS_MINUS_1
-#undef NAME_MODE
-
-#define UINT_TYPE uint16_type
-#define SINT_TYPE sint16_type
-#define BITS_MINUS_1 15
-#define NAME_MODE hi
-
-#include "rl78-divmod.h"
-
-#undef UINT_TYPE
-#undef SINT_TYPE
-#undef BITS_MINUS_1
-#undef NAME_MODE
-
-#define UINT_TYPE uint08_type
-#define SINT_TYPE sint08_type
-#define BITS_MINUS_1 7
-#define NAME_MODE qi
-
-#include "rl78-divmod.h"
-
-#endif
-
/* See the comment by the definition of LIBGCC2_UNITS_PER_WORD in
m32c.h for why we are creating extra versions of some of the
functions defined in libgcc2.c. */
Index: libgcc/config/rl78/lib2mul.c
===================================================================
--- libgcc/config/rl78/lib2mul.c (revision 217462)
+++ libgcc/config/rl78/lib2mul.c (working copy)
@@ -30,30 +30,10 @@
#define C3B(a,b,c) a##b##c
#define C3(a,b,c) C3B(a,b,c)
-#ifdef __RL78_G10__
-
-#define UINT_TYPE uint32_type
-#define BITS_MINUS_1 31
-#define NAME_MODE si
-
-#include "rl78-mul.h"
-
#undef UINT_TYPE
#undef BITS_MINUS_1
#undef NAME_MODE
-#define UINT_TYPE uint16_type
-#define BITS_MINUS_1 15
-#define NAME_MODE hi
-
-#include "rl78-mul.h"
-
-#endif
-
-#undef UINT_TYPE
-#undef BITS_MINUS_1
-#undef NAME_MODE
-
#define UINT_TYPE uint08_type
#define BITS_MINUS_1 7
#define NAME_MODE qi
Index: libgcc/config/rl78/mulsi3.S
===================================================================
--- libgcc/config/rl78/mulsi3.S (revision 217462)
+++ libgcc/config/rl78/mulsi3.S (working copy)
@@ -24,12 +24,6 @@
#include "vregs.h"
-; the G10 only has one register bank, so cannot use these optimized
-; versions. Use the C version instead.
-
-#ifndef __RL78_G10__
-
-
;----------------------------------------------------------------------
; Register use:
@@ -39,19 +33,22 @@
; DE count (resL-tmp)
; HL [sp+4]
- .text
- nop
- .global ___mulsi3 ; (USI a, USI b)
-___mulsi3:
+START_FUNC ___mulsi3
;; A is at [sp+4]
;; B is at [sp+8]
;; result is in R8..R11
+#ifdef __RL78_G10__
+ movw ax, r16
+ push ax
+ movw ax, r18
+ push ax
+#else
sel rb2
push ax
push bc
sel rb0
-
+#endif
clrw ax
movw r8, ax
movw r16, ax
@@ -62,9 +59,19 @@
cmpw ax, #0xffff
bnz $2f
movw ax, [sp+8]
+#ifdef __RL78_G10__
+ push bc
+ movw bc, r8
+ xchw ax, bc
+ subw ax, bc
+ movw r8, ax
+ movw ax, bc
+ pop bc
+#else
sel rb1
subw ax, r_0
sel rb0
+#endif
br $1f
2:
movw bc, ax
@@ -80,9 +87,19 @@
cmpw ax, #0xffff
bnz $2f
movw ax, [sp+12]
+#ifdef __RL78_G10__
+ push bc
+ movw bc, r8
+ xchw ax, bc
+ subw ax, bc
+ movw r8, ax
+ movw ax, bc
+ pop bc
+#else
sel rb1
subw ax, r_0
sel rb0
+#endif
br $1f
2:
movw bc, ax
@@ -113,12 +130,28 @@
movw r10, ax
movw ax, bc
-
.Lmul_hisi_top:
movw bc, #0
.Lmul_hisi_loop:
shrw ax, 1
+#ifdef __RL78_G10__
+ push ax
+ bnc $.Lmul_hisi_no_add_g10
+ movw ax, r8
+ addw ax, r10
+ movw r8, ax
+ sknc
+ incw r16
+ movw ax, r16
+ addw ax, r_2
+ movw r16, ax
+.Lmul_hisi_no_add_g10:
+ movw ax, r10
+ shlw ax, 1
+ movw r10, ax
+ pop ax
+#else
bnc $.Lmul_hisi_no_add
sel rb1
addw ax, bc
@@ -130,11 +163,29 @@
sel rb1
shlw bc, 1
sel rb0
+#endif
rolwc bc, 1
cmpw ax, #0
bz $.Lmul_hisi_done
shrw ax, 1
+#ifdef __RL78_G10__
+ push ax
+ bnc $.Lmul_hisi_no_add2_g10
+ movw ax, r8
+ addw ax, r10
+ movw r8, ax
+ movw ax, r16
+ sknc
+ incw ax
+ addw ax, r_2
+ movw r16, ax
+.Lmul_hisi_no_add2_g10:
+ movw ax, r10
+ shlw ax, 1
+ movw r10, ax
+ pop ax
+#else
bnc $.Lmul_hisi_no_add2
sel rb1
addw ax, bc
@@ -146,6 +197,7 @@
sel rb1
shlw bc, 1
sel rb0
+#endif
rolwc bc, 1
cmpw ax, #0
bnz $.Lmul_hisi_loop
@@ -154,18 +206,25 @@
movw ax, r16
movw r10, ax
-
+
+#ifdef __RL78_G10__
+ pop ax
+ movw r18, ax
+ pop ax
+ movw r16, ax
+#else
sel rb2
pop bc
pop ax
sel rb0
+#endif
ret
+END_FUNC ___mulsi3
;----------------------------------------------------------------------
- .global ___mulhi3
-___mulhi3:
+START_FUNC ___mulhi3
movw r8, #0
movw ax, [sp+6]
movw bc, ax
@@ -179,9 +238,18 @@
br $.Lmul_hi_loop
.Lmul_hi_top:
+#ifdef __RL78_G10__
+ push ax
+ movw ax, r8
+ addw ax, r_2
+ movw r8, ax
+ pop ax
+#else
sel rb1
addw ax, r_2
sel rb0
+#endif
+
.Lmul_hi_no_add:
shlw bc, 1
.Lmul_hi_loop:
@@ -198,5 +266,4 @@
.Lmul_hi_done:
ret
-
-#endif
+END_FUNC ___mulhi3
Index: libgcc/config/rl78/t-rl78
===================================================================
--- libgcc/config/rl78/t-rl78 (revision 217462)
+++ libgcc/config/rl78/t-rl78 (working copy)
@@ -30,3 +30,5 @@
$(srcdir)/config/rl78/divmodqi.S \
$(srcdir)/config/rl78/signbit.S \
$(srcdir)/config/rl78/cmpsi2.S
+
+HOST_LIBGCC2_CFLAGS += -Os -ffunction-sections -fdata-sections
Index: libgcc/config/rl78/vregs.h
===================================================================
--- libgcc/config/rl78/vregs.h (revision 217462)
+++ libgcc/config/rl78/vregs.h (working copy)
@@ -54,3 +54,18 @@
r23 = 0xffeef
#endif
+
+ /* Start a function in its own section, so that it
+ can be subject to linker garbage collection. */
+.macro START_FUNC name
+ .pushsection .text.\name,"ax",@progbits
+ .global \name
+ .type \name , @function
+\name:
+.endm
+
+ /* End the function. Set the size. */
+.macro END_FUNC name
+ .size \name , . - \name
+ .popsection
+.endm
: LocalWords: RL rl divmodhi divmodqi divmodsi mulsi muls CFLAGS vregs
