Skip to content

Commit

Permalink
Fix asm errors when using PGO
Browse files Browse the repository at this point in the history
  • Loading branch information
tpietrus committed Oct 7, 2024
1 parent 4caa8a7 commit a5cd907
Showing 1 changed file with 156 additions and 154 deletions.
310 changes: 156 additions & 154 deletions frontend/plat_suniv.c
View file
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
/*
/*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
Expand All @@ -9,7 +9,7 @@
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
Expand Down Expand Up @@ -220,51 +220,51 @@ static inline uint16_t mask_filter(uint16_t *s)

static inline void blit256_x15(uint16_t* dst, const void* src)
{
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
"a: .word 0x80008000 ;\n"
" mov r0, %0 ;\n"
" mov r1, %1 ;\n"
" mov r10, #16 ;\n"
" ldr r9, a ;\n"
"0: ldmia r1!, {r2-r8,lr} ;\n"
" orr r2, r9 ;\n"
" orr r3, r9 ;\n"
" orr r4, r9 ;\n"
" orr r5, r9 ;\n"
" orr r6, r9 ;\n"
" orr r7, r9 ;\n"
" orr r8, r9 ;\n"
" orr lr, r9 ;\n"
" str r2, [r0], #4 ;\n"
" lsr r2, #16 ;\n"
" orr r2, r3, lsl #16 ;\n"
" str r2, [r0], #4 ;\n"
" lsr r3, #16 ;\n"
" orr r3, r4, lsl #16 ;\n"
" str r3, [r0], #4 ;\n"
" lsr r4, #16 ;\n"
" orr r4, r4, lsl #16 ;\n"
" str r4, [r0], #4 ;\n"
" str r5, [r0], #4 ;\n"
" str r6, [r0], #4 ;\n"
" lsr r6, #16 ;\n"
" orr r6, r7, lsl #16 ;\n"
" str r6, [r0], #4 ;\n"
" lsr r7, #16 ;\n"
" orr r7, r8, lsl #16 ;\n"
" str r7, [r0], #4 ;\n"
" lsr r8, #16 ;\n"
" orr r8, r8, lsl #16 ;\n"
" str r8, [r0], #4 ;\n"
" subs r10, #1 ;\n"
" str lr, [r0], #4 ;\n"
" bgt 0b ;\n"
:: "r"(dst), "r"(src)
: "cc", "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
"a: .word 0x80008000 ;\n"
" mov r10, #16 ;\n"
" ldr r9, a ;\n"
"0: ldmia %[src]!, {r2-r8,lr} ;\n"
" orr r2, r9 ;\n"
" orr r3, r9 ;\n"
" orr r4, r9 ;\n"
" orr r5, r9 ;\n"
" orr r6, r9 ;\n"
" orr r7, r9 ;\n"
" orr r8, r9 ;\n"
" orr lr, r9 ;\n"
" str r2, [%[dst]], #4 ;\n"
" lsr r2, #16 ;\n"
" orr r2, r3, lsl #16 ;\n"
" str r2, [%[dst]], #4 ;\n"
" lsr r3, #16 ;\n"
" orr r3, r4, lsl #16 ;\n"
" str r3, [%[dst]], #4 ;\n"
" lsr r4, #16 ;\n"
" orr r4, r4, lsl #16 ;\n"
" str r4, [%[dst]], #4 ;\n"
" str r5, [%[dst]], #4 ;\n"
" str r6, [%[dst]], #4 ;\n"
" lsr r6, #16 ;\n"
" orr r6, r7, lsl #16 ;\n"
" str r6, [%[dst]], #4 ;\n"
" lsr r7, #16 ;\n"
" orr r7, r8, lsl #16 ;\n"
" str r7, [%[dst]], #4 ;\n"
" lsr r8, #16 ;\n"
" orr r8, r8, lsl #16 ;\n"
" str r8, [%[dst]], #4 ;\n"
" subs r10, #1 ;\n"
" str lr, [%[dst]], #4 ;\n"
" bgt 0b ;\n"
: [dst] "+r" (dst), [src] "+r" (src)
:
: "cc", "memory", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
}


static inline void blit256_x24(uint16_t* dst, const void* src)
{
uint32_t uCount=32;
Expand All @@ -288,30 +288,31 @@ static inline void blit256_x24(uint16_t* dst, const void* src)

static inline void blit320_x15(uint16_t* dst, const void* src)
{
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
"c: .word 0x80008000 ;\n"
" mov r0, %0 ;\n"
" mov r1, %1 ;\n"
" mov r10, #20 ;\n"
" ldr r9, c ;\n"
"0: ldmia r1!, {r2-r8,lr} ;\n"
" orr r2, r9 ;\n"
" orr r3, r9 ;\n"
" orr r4, r9 ;\n"
" orr r5, r9 ;\n"
" orr r6, r9 ;\n"
" orr r7, r9 ;\n"
" orr r8, r9 ;\n"
" orr lr, r9 ;\n"
" subs r10, #1 ;\n"
" stmia r0!, {r2-r8,lr} ;\n"
" bgt 0b ;\n"
:: "r"(dst), "r"(src)
: "cc", "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
" mov r9, #0x8000 ;\n" // Load upper part
" lsl r9, r9, #16 ;\n" // Shift left to position
" orr r9, r9, #0x8000 ;\n" // OR with lower part
" mov r10, #20 ;\n"
"0: ldmia %[src]!, {r2-r8,lr} ;\n"
" orr r2, r9 ;\n"
" orr r3, r9 ;\n"
" orr r4, r9 ;\n"
" orr r5, r9 ;\n"
" orr r6, r9 ;\n"
" orr r7, r9 ;\n"
" orr r8, r9 ;\n"
" orr lr, r9 ;\n"
" subs r10, #1 ;\n"
" stmia %[dst]!, {r2-r8,lr} ;\n"
" bgt 0b ;\n"
: [dst] "+r" (dst), [src] "+r" (src)
:
: "cc", "memory", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
}


static inline void blit320_x24(uint16_t* dst, const void* src)
{
uint32_t uCount=20;
Expand Down Expand Up @@ -341,35 +342,35 @@ static inline void blit320_x24(uint16_t* dst, const void* src)

static inline void blit368_x15(uint16_t* dst, const void* src)
{
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
" .macro ustr rl rt ;\n"
" lsr \\rl, #16 ;\n"
" orr \\rl, \\rt, lsl #16 ;\n"
" orr \\rl, r9 ;\n"
" str \\rl, [r0], #4 ;\n"
" .endm ;\n"
"e: .word 0x80008000 ;\n"
" mov r0, %0 ;\n"
" mov r1, %1 ;\n"
" mov r10, #23 ;\n"
" ldr r9, e ;\n"
"0: ldmia r1!, {r2-r8,lr} ;\n"
" ustr r2, r3 ;\n"
" ustr r3, r4 ;\n"
" ustr r4, r5 ;\n"
" subs r10, #1 ;\n"
" orr r6, r9 ;\n"
" orr r7, r9 ;\n"
" orr r8, r9 ;\n"
" orr lr, r9 ;\n"
" stmia r0!, {r6-r8,lr} ;\n"
" bgt 0b ;\n"
:: "r"(dst), "r"(src)
: "cc", "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
" .macro ustr rl, rt ;\n"
" lsr \\rl, #16 ;\n"
" orr \\rl, \\rt, lsl #16 ;\n"
" orr \\rl, r9 ;\n"
" str \\rl, [%[dst]], #4 ;\n"
" .endm ;\n"
"e: .word 0x80008000 ;\n"
" mov r10, #23 ;\n"
" ldr r9, e ;\n"
"0: ldmia %[src]!, {r2-r8,lr} ;\n"
" ustr r2, r3 ;\n"
" ustr r3, r4 ;\n"
" ustr r4, r5 ;\n"
" subs r10, #1 ;\n"
" orr r6, r9 ;\n"
" orr r7, r9 ;\n"
" orr r8, r9 ;\n"
" orr lr, r9 ;\n"
" stmia %[dst]!, {r6-r8,lr} ;\n"
" bgt 0b ;\n"
: [dst] "+r" (dst), [src] "+r" (src)
:
: "cc", "memory", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
}


static inline void blit368_x24(uint16_t* dst, const void* src)
{
uint32_t uCount=20;
Expand Down Expand Up @@ -439,42 +440,43 @@ static inline void blit384_x24(uint16_t* dst, const void* src)

static inline void blit512_x15(uint16_t* dst, const void* src)
{
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
"i: .word 0x80008000 ;\n"
" mov r0, %0 ;\n"
" mov r1, %1 ;\n"
" mov r10, #32 ;\n"
" ldr r9, i ;\n"
"0: ldmia r1!, {r2-r8,lr} ;\n"
" lsl r2, #16 ;\n"
" lsr r2, #16 ;\n"
" orr r2, r3, lsl #16 ;\n"
" orr r2, r9 ;\n"
" str r2, [r0], #4 ;\n"
" lsr r4, #16 ;\n"
" lsr r3, #16 ;\n"
" orr r3, r4, lsl #16 ;\n"
" orr r3, r9 ;\n"
" str r3, [r0], #4 ;\n"
" lsr r5, #16 ;\n"
" orr r5, r6, lsl #16 ;\n"
" orr r5, r9 ;\n"
" str r5, [r0], #4 ;\n"
" lsr r8, #16 ;\n"
" lsr lr, #16 ;\n"
" orr r7, r9 ;\n"
" str r7, [r0], #4 ;\n"
" orr r8, lr, lsl #16 ;\n"
" subs r10, #1 ;\n"
" orr r8, r9 ;\n"
" str r8, [r0], #4 ;\n"
" bgt 0b ;\n"
:: "r"(dst), "r"(src)
: "cc", "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
" mov r9, #0x8000 ;\n" // Load upper part
" lsl r9, r9, #16 ;\n" // Shift left to position
" orr r9, r9, #0x8000 ;\n" // OR with lower part
" mov r10, #32 ;\n"
"0: ldmia %[src]!, {r2-r8,lr} ;\n"
" lsl r2, #16 ;\n"
" lsr r2, #16 ;\n"
" orr r2, r3, lsl #16 ;\n"
" orr r2, r9 ;\n"
" str r2, [%[dst]], #4 ;\n"
" lsr r4, #16 ;\n"
" lsr r3, #16 ;\n"
" orr r3, r4, lsl #16 ;\n"
" orr r3, r9 ;\n"
" str r3, [%[dst]], #4 ;\n"
" lsr r5, #16 ;\n"
" orr r5, r6, lsl #16 ;\n"
" orr r5, r9 ;\n"
" str r5, [%[dst]], #4 ;\n"
" lsr r8, #16 ;\n"
" lsr lr, #16 ;\n"
" orr r7, r9 ;\n"
" str r7, [%[dst]], #4 ;\n"
" orr r8, lr, lsl #16 ;\n"
" subs r10, #1 ;\n"
" orr r8, r9 ;\n"
" str r8, [%[dst]], #4 ;\n"
" bgt 0b ;\n"
: [dst] "+r" (dst), [src] "+r" (src)
:
: "cc", "memory", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
}


static inline void blit512_x24(uint16_t* dst, const void* src)
{
uint32_t uCount=32;
Expand All @@ -499,32 +501,32 @@ static inline void blit512_x24(uint16_t* dst, const void* src)

static inline void blit640_x15(uint16_t* dst, const void* src)
{
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
" .macro lhw_str rl rt ;\n"
" lsl \\rl, #16 ;\n"
" lsr \\rl, #16 ;\n"
" orr \\rl, \\rt, lsl #16 ;\n"
" orr \\rl, r9 ;\n"
" str \\rl, [r0], #4 ;\n"
" .endm ;\n"
"k: .word 0x80008000 ;\n"
" mov r0, %0 ;\n"
" mov r1, %1 ;\n"
" mov r10, #40 ;\n"
" ldr r9, k ;\n"
"0: ldmia r1!, {r2-r8,lr} ;\n"
" lhw_str r2, r3 ;\n"
" lhw_str r4, r5 ;\n"
" lhw_str r6, r7 ;\n"
" subs r10, #1 ;\n"
" lhw_str r8, lr ;\n"
" bgt 0b ;\n"
:: "r"(dst), "r"(src)
: "cc", "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
asm volatile (
" .arm ;\n"
" .align 2 ;\n"
" .macro lhw_str rl, rt ;\n"
" lsl \\rl, #16 ;\n"
" lsr \\rl, #16 ;\n"
" orr \\rl, \\rt, lsl #16 ;\n"
" orr \\rl, r9 ;\n"
" str \\rl, [%[dst]], #4 ;\n"
" .endm ;\n"
"k: .word 0x80008000 ;\n"
" mov r10, #40 ;\n"
" ldr r9, k ;\n"
"0: ldmia %[src]!, {r2-r8,lr} ;\n"
" lhw_str r2, r3 ;\n"
" lhw_str r4, r5 ;\n"
" lhw_str r6, r7 ;\n"
" subs r10, #1 ;\n"
" lhw_str r8, lr ;\n"
" bgt 0b ;\n"
: [dst] "+r" (dst), [src] "+r" (src)
:
: "cc", "memory", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "lr");
}


static inline void blit640_x24(uint16_t* dst, const void* src)
{
uint32_t uCount=20;
Expand Down Expand Up @@ -714,11 +716,11 @@ void plat_init(void)
screen = SDL_SetVideoMode(320, 240, 16, SDL_HWSURFACE | SDL_DOUBLEBUF);
if (screen == NULL) {
printf("%s, failed to set video mode\n", __func__);
exit(-1);
exit(-1);
}
SDL_ShowCursor(0);
system("echo 10 > /proc/sys/vm/swappiness");

memdev = open("/dev/mem", O_RDWR);
if(memdev < 0){
printf("failed to open /dev/mem\n");
Expand Down

0 comments on commit a5cd907

Please sign in to comment.