forked from len0rd/rockbox
5d05b9d3e9
This ports id Software's Quake to run on the SDL plugin runtime. The source code originated from id under the GPLv2 license. I used https://github.com/ahefner/sdlquake as the base of my port. Performance is, unsurprisingly, not on par with what you're probably used to on PC. I average about 10FPS on ipod6g, but it's still playable. Sound works well enough, but in-game music is not supported. I've written ARM assembly routines for the inner sound loop. Make sure you turn the "brightness" all the way down, or colors will look funky. To run, extract Quake's data files to /.rockbox/quake. Have fun! Change-Id: I4285036e967d7f0722802d43cf2096c808ca5799
191 lines
7 KiB
ArmAsm
191 lines
7 KiB
ArmAsm
.cpu arm926ej-s
|
|
.fpu softvfp
|
|
.eabi_attribute 20, 1
|
|
.eabi_attribute 21, 1
|
|
.eabi_attribute 23, 3
|
|
.eabi_attribute 24, 1
|
|
.eabi_attribute 25, 1
|
|
.eabi_attribute 26, 1
|
|
.eabi_attribute 30, 2
|
|
.eabi_attribute 18, 4
|
|
.file "snd_mix_generic.c"
|
|
@ GNU C (GCC) version 4.4.4 (arm-elf-eabi)
|
|
@ compiled by GNU C version 4.8.2 20140206 (prerelease), GMP version 4.3.2, MPFR version 2.4.2.
|
|
@ GGC heuristics: --param ggc-min-expand=100 --param ggc-min-heapsize=131072
|
|
@ options passed: -imultilib arm926ej-s -D__USES_INITFINI__
|
|
@ ./snd_mix_generic.c -mcpu=arm926ej-s -O3 -fverbose-asm
|
|
@ options enabled: -falign-loops -fargument-alias -fauto-inc-dec
|
|
@ -fbranch-count-reg -fcaller-saves -fcommon -fcprop-registers
|
|
@ -fcrossjumping -fcse-follow-jumps -fdefer-pop
|
|
@ -fdelete-null-pointer-checks -fearly-inlining
|
|
@ -feliminate-unused-debug-types -fexpensive-optimizations
|
|
@ -fforward-propagate -ffunction-cse -fgcse -fgcse-after-reload -fgcse-lm
|
|
@ -fguess-branch-probability -fident -fif-conversion -fif-conversion2
|
|
@ -findirect-inlining -finline -finline-functions
|
|
@ -finline-functions-called-once -finline-small-functions -fipa-cp
|
|
@ -fipa-cp-clone -fipa-pure-const -fipa-reference -fira-share-save-slots
|
|
@ -fira-share-spill-slots -fivopts -fkeep-static-consts
|
|
@ -fleading-underscore -fmath-errno -fmerge-constants -fmerge-debug-strings
|
|
@ -fmove-loop-invariants -fomit-frame-pointer -foptimize-register-move
|
|
@ -foptimize-sibling-calls -fpeephole -fpeephole2 -fpredictive-commoning
|
|
@ -freg-struct-return -fregmove -freorder-blocks -freorder-functions
|
|
@ -frerun-cse-after-loop -fsched-interblock -fsched-spec
|
|
@ -fsched-stalled-insns-dep -fschedule-insns -fschedule-insns2
|
|
@ -fsection-anchors -fsigned-zeros -fsplit-ivs-in-unroller
|
|
@ -fsplit-wide-types -fstrict-aliasing -fstrict-overflow -fthread-jumps
|
|
@ -ftoplevel-reorder -ftrapping-math -ftree-builtin-call-dce -ftree-ccp
|
|
@ -ftree-ch -ftree-copy-prop -ftree-copyrename -ftree-cselim -ftree-dce
|
|
@ -ftree-dominator-opts -ftree-dse -ftree-fre -ftree-loop-im
|
|
@ -ftree-loop-ivcanon -ftree-loop-optimize -ftree-parallelize-loops=
|
|
@ -ftree-pre -ftree-reassoc -ftree-scev-cprop -ftree-sink -ftree-sra
|
|
@ -ftree-switch-conversion -ftree-ter -ftree-vect-loop-version
|
|
@ -ftree-vectorize -ftree-vrp -funit-at-a-time -funswitch-loops
|
|
@ -fverbose-asm -fzero-initialized-in-bss -mlittle-endian -msched-prolog
|
|
|
|
@ Compiler executable checksum: 6bc9eb36ed7f27ae4e8601da7e401eb4
|
|
|
|
.text
|
|
.align 2
|
|
.global SND_PaintChannelFrom8
|
|
.type SND_PaintChannelFrom8, %function
|
|
SND_PaintChannelFrom8:
|
|
// r0: true_lvol
|
|
// r1: true_rvol
|
|
// r2: *sfx
|
|
// r3: count
|
|
|
|
@ args = 0, pretend = 0, frame = 0
|
|
@ frame_needed = 0, uses_anonymous_args = 0
|
|
@ link register save eliminated.
|
|
cmp r3, #0 @ count
|
|
stmfd sp!, {r4, r5, r6, r7, r8, sl} @,
|
|
ble .L10 @, // exit early if count <= 0
|
|
ldr ip, .L13 @ tmp207, // ip = paintbuffer
|
|
ldr r5, .L13+4 @ tmp224, // r5 = 32767
|
|
ldr ip, [ip, #0] @ ivtmp.61, paintbuffer // ip = &paintbuffer
|
|
|
|
// this does {rvol,lvol} &= 0xffff in a really convoluted way
|
|
mov r1, r1, asl #16 @ tmp206, true_rvol, // r1 = rvol << 16
|
|
mov r0, r0, asl #16 @ tmp205, true_lvol, // r0 = lvol << 16
|
|
mov r6, r1, lsr #16 @ pretmp.46, tmp206, // rvol_16 = r1 >> 16
|
|
mov r7, r0, lsr #16 @ pretmp.44, tmp205, // lvol_16 = r0 >> 16
|
|
mov r1, #0 @ i, // i = 0
|
|
// loop?
|
|
.L9:
|
|
ldrsb r0, [r2, r1] @ tmp208,* i // r0 = (byte) sfx[i]
|
|
ldrsh r8, [ip, #0] @ tmp213,* ivtmp.61 // r8 = (int16) paintbuffer[0]
|
|
|
|
// r0 &= 0xffff
|
|
mov r0, r0, asl #16 @ tmp209, tmp208, // r0 <<= 16
|
|
mov r0, r0, lsr #16 @ D.1289, tmp209, // r0 >>= 16
|
|
|
|
|
|
mul r4, r0, r7 @ tmp210, D.1289, pretmp.44 // r4 = sfx[i] * (lvol_16)
|
|
|
|
mov r4, r4, asl #16 @ tmp212, tmp210, // r4 <<= 16
|
|
|
|
add r4, r8, r4, asr #16 @, val, tmp213, tmp212, // r8 += r4 & 0xffff
|
|
|
|
// handle saturation
|
|
cmp r4, r5 @ val, tmp224 // compare r4, 32767
|
|
ldrgt r8, .L13+4 @ D.1341, // r8 = 32767 if saturated
|
|
bgt .L4 @, // jump to .L4 if saturated
|
|
cmn r4, #32768 @ val, // compare r4, -32768
|
|
movge r4, r4, asl #16 @ tmp216, val, // if not saturated, &= 0xffff
|
|
movlt r8, #32768 @ D.1341, // r8 = -32768 if saturated
|
|
movge r8, r4, lsr #16 @ D.1341, tmp216, // r8 = val
|
|
|
|
// r8 is final value of paintbuffer[2*i+0]
|
|
|
|
// now handle right channel
|
|
.L4:
|
|
// sl is scaled right channel
|
|
|
|
// this does paintbuffer[1] += rvol * sfx[i]
|
|
mul sl, r0, r6 @ tmp217, D.1289, pretmp.46 // sl = sfx[i] * rvol_16
|
|
ldrsh r4, [ip, #2] @ tmp220, // r4 = paintbuffer[1]
|
|
mov r0, sl, asl #16 @ tmp219, tmp217, // r0 = sl << 16
|
|
add r0, r4, r0, asr #16 @, val, tmp220, tmp219, // r0 = r4 + (r0 >> 16)
|
|
|
|
// now saturate
|
|
cmp r0, r5 @ val, tmp224 // compare final right channel to 32767
|
|
|
|
strh r8, [ip, #0] @ movhi @ D.1341,* ivtmp.61 // write left channel back to paintbuffer
|
|
ldrgt r0, .L13+4 @ D.1348, // r0 = 32767 if saturated
|
|
bgt .L7 @, // skip to .L7
|
|
cmn r0, #32768 @ val, // check < -32768
|
|
movge r0, r0, asl #16 @ tmp223, val, // <<= 16 if not saturated
|
|
movlt r0, #32768 @ D.1348, // if < -32768, saturate
|
|
movge r0, r0, lsr #16 @ D.1348, tmp223, // >>= 16 if not saturated
|
|
.L7:
|
|
add r1, r1, #1 @ i, i, // i++
|
|
cmp r1, r3 @ i, count // i =? count
|
|
strh r0, [ip, #2] @ movhi @ D.1348 , // store right channel to paintbuffer
|
|
add ip, ip, #4 @ ivtmp.61, ivtmp.61, // paintbuffer += 4 bytes
|
|
bne .L9 @, // restart if i != count
|
|
.L10:
|
|
ldmfd sp!, {r4, r5, r6, r7, r8, sl} // pop stack
|
|
bx lr // return
|
|
.L14:
|
|
.align 2
|
|
.L13:
|
|
.word paintbuffer
|
|
.word 32767
|
|
.size SND_PaintChannelFrom8, .-SND_PaintChannelFrom8
|
|
.align 2
|
|
.global SND_PaintChannelFrom16
|
|
.type SND_PaintChannelFrom16, %function
|
|
SND_PaintChannelFrom16:
|
|
@ args = 0, pretend = 0, frame = 0
|
|
@ frame_needed = 0, uses_anonymous_args = 0
|
|
@ link register save eliminated.
|
|
cmp r3, #0 @ count
|
|
stmfd sp!, {r4, r5, r6, r7, r8, sl} @,
|
|
ble .L24 @,
|
|
ldr ip, .L26 @ tmp146,
|
|
ldr r7, .L26+4 @ tmp164,
|
|
ldr ip, [ip, #0] @ ivtmp.119, paintbuffer
|
|
mov r4, #0 @ i,
|
|
.L23:
|
|
mov r5, r4, asl #1 @ tmp147, i,
|
|
ldrsh r5, [r2, r5] @ data,
|
|
ldrsh r8, [ip, #0] @ tmp152,* ivtmp.119
|
|
mul r6, r5, r0 @ tmp148, data, true_lvol
|
|
mov r6, r6, asl #8 @ tmp151, tmp148,
|
|
add r6, r8, r6, asr #16 @, val, tmp152, tmp151,
|
|
cmp r6, r7 @ val, tmp164
|
|
ldrgt r8, .L26+4 @ D.1355,
|
|
bgt .L18 @,
|
|
cmn r6, #32768 @ val,
|
|
movge r6, r6, asl #16 @ tmp155, val,
|
|
movlt r8, #32768 @ D.1355,
|
|
movge r8, r6, lsr #16 @ D.1355, tmp155,
|
|
.L18:
|
|
mul sl, r5, r1 @ tmp156, data, true_rvol
|
|
ldrsh r6, [ip, #2] @ tmp160,
|
|
mov r5, sl, asl #8 @ tmp159, tmp156,
|
|
add r5, r6, r5, asr #16 @, val, tmp160, tmp159,
|
|
cmp r5, r7 @ val, tmp164
|
|
strh r8, [ip, #0] @ movhi @ D.1355,* ivtmp.119
|
|
ldrgt r5, .L26+4 @ D.1362,
|
|
bgt .L21 @,
|
|
cmn r5, #32768 @ val,
|
|
movge r5, r5, asl #16 @ tmp163, val,
|
|
movlt r5, #32768 @ D.1362,
|
|
movge r5, r5, lsr #16 @ D.1362, tmp163,
|
|
.L21:
|
|
add r4, r4, #2 @ i, i,
|
|
cmp r3, r4 @ count, i
|
|
strh r5, [ip, #2] @ movhi @ D.1362,
|
|
add ip, ip, #8 @ ivtmp.119, ivtmp.119,
|
|
bgt .L23 @,
|
|
.L24:
|
|
ldmfd sp!, {r4, r5, r6, r7, r8, sl}
|
|
bx lr
|
|
.L27:
|
|
.align 2
|
|
.L26:
|
|
.word paintbuffer
|
|
.word 32767
|
|
.size SND_PaintChannelFrom16, .-SND_PaintChannelFrom16
|
|
.ident "GCC: (GNU) 4.4.4"
|