From 17ed0d696a7a3649cc41dcf182e162c85002b33a Mon Sep 17 00:00:00 2001 From: notaz Date: Sat, 26 Feb 2011 23:15:43 +0200 Subject: [PATCH] gte_neon: implement MVMVA, some fixes --- libpcsxcore/gte_neon.h | 7 +- libpcsxcore/gte_neon.s | 188 ++++++++++++++++++++++++++++++----------- plugins/dfsound/oss.c | 5 +- plugins/dfsound/spu.c | 7 +- 4 files changed, 154 insertions(+), 53 deletions(-) diff --git a/libpcsxcore/gte_neon.h b/libpcsxcore/gte_neon.h index 36b65328..59455f54 100644 --- a/libpcsxcore/gte_neon.h +++ b/libpcsxcore/gte_neon.h @@ -1,3 +1,4 @@ -void gteRTPS_neon(void *cp2_regs); -void gteRTPT_neon(void *cp2_regs); -void gteNCLIP_neon(void *cp2_regs); +void gteRTPS_neon(void *cp2_regs, int opcode); +void gteRTPT_neon(void *cp2_regs, int opcode); +void gteMVMVA_neon(void *cp2_regs, int opcode); +void gteNCLIP_neon(void *cp2_regs, int opcode); diff --git a/libpcsxcore/gte_neon.s b/libpcsxcore/gte_neon.s index 5b6546a3..22bde4e7 100644 --- a/libpcsxcore/gte_neon.s +++ b/libpcsxcore/gte_neon.s @@ -52,21 +52,20 @@ scratch: orrvs lr, #\pflags .endm -@ approximate signed gteIR|123 [32] flags -@ in: rr 123 as gteIR|123 +@ get gteIR|123 flags from gteMAC|123 +@ in: rr 123 as gteMAC|123 @ trash: r2,r3 .macro do_irs_flags rr1 rr2 rr3 - mov r2, #0x10000 - cmn r2, \rr1, lsl #16 @ adds ... - rsbvcs r3, r2, \rr1, lsl #16 - orrvs lr, #(1<<31)|(1<<24) @ IR1/limB1 - cmn r2, \rr2, lsl #16 - rsbvcs r3, r2, \rr2, lsl #16 - orrvs lr, #(1<<31) - orrvs lr, #(1<<23) @ IR2/limB2 - cmn r2, \rr3, lsl #16 - subvcs r3, r2, \rr3, lsl #16 - orrvs lr, #(1<<22) @ IR3/limB3 + add r2, \rr1, #0x8000 + add r3, \rr2, #0x8000 + lsrs r2, #16 + orrne lr, #(1<<31)|(1<<24) @ IR1/limB1 + lsrs r3, #16 + add r2, \rr3, #0x8000 + orrne lr, #(1<<31) + orrne lr, #(1<<23) @ IR2/limB2 + lsrs r2, #16 + orrne lr, #(1<<22) @ IR3/limB3 .endm @@ -138,7 +137,7 @@ scratch: .global gteRTPS_neon @ r0=CP2 (d,c), gteRTPS_neon: - push {r4-r7,lr} + push {r4-r6,lr} @ fmrx r4, fpscr @ vmrs? at least 40 cycle hit movw r1, #:lower16:scratch @@ -214,9 +213,7 @@ gteRTPS_neon: do_mac_flags r4, r5, r6 vshr.u32 d11, #15 @ quotient (limE) - add r3, r0, #4*9 - ldmia r3, {r4-r6} @ gteIR|123 do_irs_flags r4, r5, r6 vmlal.s32 q2, d18, d11[0]@ gteOF|XY + gteIR|12 * quotient @@ -227,15 +224,12 @@ gteRTPS_neon: vqshl.s32 d19, d18, #5 @ 11bit precision ldr r4, [r1] @ quotient - - mov r2, r1 - vst1.32 d18, [r2]! @ || writeback fS|XY2 before limG - vst1.32 d19, [r2] @ || and after 11bit saturation - movs r3, r6, lsr #16 orrne lr, #(1<<31) orrne lr, #(1<<18) @ fSZ (limD) + vst1.32 d18, [r1, :64] @ writeback fS|XY2 before limG + vshr.s32 d18, d19, #16+5@ can't vqshrn because of insn vadd.s64 d20, d7 @ | gteDQB + gteDQA * quotient vmovn.s32 d18, q9 @ fS|XY2 [s16] @@ -247,25 +241,25 @@ gteRTPS_neon: add r3, r0, #4*24 vst1.32 d20[0], [r3] @ gteMAC0 + movs r4, r4, lsr #17 + orrne lr, #(1<<31) + orrne lr, #(1<<17) @ limE + vmax.s32 d21, d20, d31 vmov.i32 d22, #0x1000 vmin.s32 d21, d22 add r3, r0, #4*8 vst1.16 d21[0], [r3] @ gteIR0 - movs r4, r4, lsr #17 + ldmia r1, {r4,r5} @ fS|XY2 before limG, after 11bit sat + add r2, r4, #0x400<<16 + add r3, r5, #0x400<<16 + lsrs r2, #16+11 + orrne lr, #(1<<14) @ limG1 + orrne lr, #(1<<31) + lsrs r3, #16+11 + orrne lr, #(1<<13) @ limG2 orrne lr, #(1<<31) - orrne lr, #(1<<17) @ limE - - ldmia r1, {r4-r7} @ fS|XY2 before limG, after 11bit sat - subs r2, r6, #1<<21 - addvcs r3, r6, #1<<21 - orrvs lr, #(1<<14) @ limG1 - orrvs lr, #(1<<31) - subs r2, r7, #1<<21 - addvcs r3, r7, #1<<21 - orrvs lr, #(1<<13) @ limG2 - orrvs lr, #(1<<31) adds r2, r4, #1 addvcs r3, r5, #1 orrvs lr, #(1<<16) @ F @@ -288,7 +282,7 @@ gteRTPS_neon: str lr, [r0, #4*(32+31)] @ gteFLAG - pop {r4-r7,pc} + pop {r4-r6,pc} .size gteRTPS_neon, .-gteRTPS_neon @@ -461,7 +455,6 @@ gteRTPT_neon: vpmax.s32 d17, d15, d31 @ || vshr.s32 q11, #16+5 @ can't vqshrn because of insn vshr.s32 d24, #16+5 @ encoding doesn't allow 21 :( - vqshl.s32 q7, #5 @ || min/max pairs shifted vsli.u64 d16, d17, #32 @ || pack in-pair min/max vadd.s64 d26, d7 @ | gteDQB + gteDQA * quotient vmovn.s32 d12, q11 @ fS|XY(v) [s16] v=0,1 @@ -494,23 +487,25 @@ gteRTPT_neon: vst1.32 d10, [r3]! @ gteIR12 vst1.32 d11[0], [r3] @ ..3 - @ ~20 cycles + @ ~23 cycles orrne lr, #(1<<31) @ limE orrne lr, #(1<<17) @ limE ldmia r1, {r4-r9} - subs r2, r4, #1<<21 @ min fSX - addvcs r3, r6, #1<<21 @ max fSX - orrvs lr, #(1<<31) @ limG1 - orrvs lr, #(1<<14) - subs r2, r5, #1<<21 @ min fSY - addvcs r3, r7, #1<<21 @ max fSY - orrvs lr, #(1<<31) @ limG2 - orrvs lr, #(1<<13) + add r2, r4, #0x400<<16 @ min fSX + add r3, r6, #0x400<<16 @ max fSX + lsrs r2, #16+11 + lsreqs r3, #16+11 + orrne lr, #(1<<31) @ limG1 + orrne lr, #(1<<14) + add r2, r5, #0x400<<16 @ min fSY + add r3, r7, #0x400<<16 @ max fSY + lsrs r2, #16+11 + lsreqs r3, #16+11 + orrne lr, #(1<<31) @ limG2 + orrne lr, #(1<<13) adds r2, r9, #1 - orrvs lr, #(1<<31) @ F - orrvs lr, #(1<<16) + orrvs lr, #(1<<16) @ F (31 already done by above) subs r3, r8, #1 - orrvs lr, #(1<<31) @ F ldr r4, [r0, #4*24] @ gteMAC0 orrvs lr, #(1<<15) @@ -531,6 +526,105 @@ gteRTPT_neon: +.global gteMVMVA_neon @ r0=CP2 (d,c), op +gteMVMVA_neon: + push {r4-r5,lr} + + add r12, r0, #4*32 + + ubfx r2, r1, #15, #2 @ v + + vmov.i32 q0, #0 @ d0,d1 + vmov.i32 q1, #0 @ d2,d3 + vmov.i32 q2, #0 @ d4,d5 + cmp r2, #3 + addeq r4, r0, #4*9 + addne r3, r0, r2, lsl #3 + ldmeqia r4, {r3-r5} + ldmneia r3, {r4,r5} + pkhbteq r4, r3, r4, lsl #16 + uxth r5, r5 + vmov.32 d8[0], r4 + vmov.32 d8[1], r5 @ VXYZ(v) + ubfx r3, r1, #17, #2 @ mx + ubfx r2, r1, #13, #2 @ cv + cmp r3, #3 + beq 0f @ very rare case + add r3, r12, r3, lsl #5 + vldmia r3, {d0-d2} @ MXxy/gteR* [16*9] +0: + cmp r2, #3 + add r3, r12, r2, lsl #5 + beq 0f + add r3, #4*5 + vldmia r3, {d4-d5} @ CVx/gteTR* + +0: + vmov.i32 q15, #0 + vext.16 d2, d1, d2, #2 @ xxx3 -> x321 + vext.16 d1, d0, d1, #3 @ xx32 -> x321 + vshll.s32 q3, d5, #12 @ gteTRZ/CV3 + vshll.s32 q2, d4, #12 @ gteTR|XY/CV12 + + vmull.s16 q8, d0, d8 + vmull.s16 q9, d1, d8 + vmull.s16 q10, d2, d8 + vpadd.s32 d16, d16, d17 + vpadd.s32 d17, d18, d19 + vpadd.s32 d18, d20, d21 + vpadal.s32 q2, q8 + vpadal.s32 q3, q9 + tst r1, #1<<19 + beq 0f + vshr.s64 q2, q2, #12 + vshr.s64 q3, q3, #12 +0: + vqmovn.s64 d8, q2 @ gteMAC|12 + vqmovn.s64 d9, q3 @ gteMAC3 + + tst r1, #1<<10 + add r3, r0, #4*25 + vqmovn.s32 d10, q4 @ gteIR|123 + vst1.32 d8, [r3]! + vst1.32 d9[0], [r3] @ wb gteMAC|123 + + beq 0f + vmax.s16 d10, d31 +0: + vmovl.s16 q9, d10 @ expand gteIR|123 + add r3, r0, #4*9 + vst1.32 d18, [r3]! + vst1.32 d19[0], [r3] + + tst r1, #1<<10 @ lm + mov r2, #0 + mov lr, #0 @ gteFLAG + mov r12, #15 + moveq r2, #0x8000 @ adj + moveq r12, #16 @ shift + + add r3, r0, #4*25 + ldmia r3, {r3-r5} @ gteMAC|123 + + do_mac_flags r3, r4, r5 + + add r3, r2 + add r4, r2 + add r5, r2 + asrs r3, r12 + orrne lr, #(1<<31)|(1<<24) @ IR1/limB1 + asrs r4, r12 + orrne lr, #(1<<31) + orrne lr, #(1<<23) @ IR2/limB2 + asrs r5, r12 + orrne lr, #(1<<22) @ IR3/limB3 + str lr, [r0, #4*(32+31)] @ gteFLAG + + pop {r4-r5,pc} + .size gteMVMVA_neon, .-gteMVMVA_neon + + + @ the name is misnormer, this doesn't use NEON but oh well.. .global gteNCLIP_neon @ r0=CP2 (d,c), gteNCLIP_neon: diff --git a/plugins/dfsound/oss.c b/plugins/dfsound/oss.c index f4dd215d..929d9213 100644 --- a/plugins/dfsound/oss.c +++ b/plugins/dfsound/oss.c @@ -65,9 +65,10 @@ void SetupSound(void) } // we use 64 fragments with 1024 bytes each + // rearmed: now using 10*4096 for better latency - fragsize=10; - myfrag=(63<<16)|fragsize; + fragsize=12; + myfrag=(10<<16)|fragsize; if(ioctl(oss_audio_fd,SNDCTL_DSP_SETFRAGMENT,&myfrag)==-1) { diff --git a/plugins/dfsound/spu.c b/plugins/dfsound/spu.c index c455d3d7..afd5fd6c 100644 --- a/plugins/dfsound/spu.c +++ b/plugins/dfsound/spu.c @@ -771,7 +771,7 @@ void CALLBACK SPUasync(unsigned long cycle) if(iSpuAsyncWait) { iSpuAsyncWait++; - if(iSpuAsyncWait<=64) return; + if(iSpuAsyncWait<=16) return; iSpuAsyncWait=0; } @@ -780,6 +780,11 @@ void CALLBACK SPUasync(unsigned long cycle) if(!bSpuInit) return; // -> no init, no call MAINThread(0); // -> linux high-compat mode + + // abuse iSpuAsyncWait mechanism to reduce calls to above function + // to make it do larger chunks + // note: doing it less often than once per frame causes skips + iSpuAsyncWait=1; } } -- 2.39.2