diff options
Diffstat (limited to 'libpcsxcore')
| -rw-r--r-- | libpcsxcore/gte_arm.S | 14 | ||||
| -rw-r--r-- | libpcsxcore/gte_neon.S | 10 |
2 files changed, 13 insertions, 11 deletions
diff --git a/libpcsxcore/gte_arm.S b/libpcsxcore/gte_arm.S index e711e82..d3f210d 100644 --- a/libpcsxcore/gte_arm.S +++ b/libpcsxcore/gte_arm.S @@ -182,7 +182,7 @@ gteRTPS_nf_arm: pop {r0, r12} .endif 1: - ldrd r6, [r0,#4*(32+24)] @ gteOFXY + ldrd r6, r7, [r0, #4*(32+24)] @ gteOFXY cmp r9, #0x20000 add r1, r0, #4*12 @ gteSXY0 movhs r9, #0x20000 @@ -194,7 +194,7 @@ gteRTPS_nf_arm: mov r3, r7, asr #31 smlal r7, r3, r11, r9 lsr r6, #16 - /* gteDQA, gteDQB */ ldrd r10,[r0, #4*(32+27)] + /* gteDQA, gteDQB */ ldrd r10,r11, [r0, #4*(32+27)] orr r6, r2, lsl #16 @ (gteOFX + gteIR1 * q) >> 16 ssatx_prep r2, 11 lsr r7, #16 @@ -224,7 +224,7 @@ gteRTPT_nf_arm: rtpt_arm_loop: add r1, r0, lr, lsl #1 - ldrd r8, [r1] @ VXYZ(v) + ldrd r8, r9, [r1] @ VXYZ(v) do_rtpx_mac ssatx_prep r6, 16 @@ -253,7 +253,7 @@ rtpt_arm_loop: 1: cmp r9, #0x20000 add r1, r0, #4*12 movhs r9, #0x20000 - ldrd r6, [r0,#4*(32+24)] @ gteOFXY + ldrd r6, r7, [r0,#4*(32+24)] @ gteOFXY /* quotient */ subhs r9, #1 mov r2, r6, asr #31 smlal r6, r2, r10, r9 @@ -272,7 +272,7 @@ rtpt_arm_loop: cmp lr, #12 blt rtpt_arm_loop - ldrd r4, [r0, #4*(32+27)] @ gteDQA, gteDQB + ldrd r4, r5, [r0, #4*(32+27)] @ gteDQA, gteDQB add r1, r0, #4*9 @ gteIR1 mla r3, r4, r9, r5 @ gteDQB + gteDQA * q stmia r1, {r10,r11,r12} @ gteIR123 save @@ -464,7 +464,7 @@ gteNCLIP_arm: .endif str r2, [r0, #4*9] #ifdef HAVE_ARMV5 - ldrd r2, [r0, #4*26] @ gteMAC23 + ldrd r2, r3, [r0, #4*26] @ gteMAC23 #else ldr r2, [r0, #4*26] ldr r3, [r0, #4*27] @@ -495,7 +495,7 @@ gteNCLIP_arm: .endif orrlt r12, #1<<22 #ifdef HAVE_ARMV5 - strd r2, [r0, #4*10] @ gteIR23 + strd r2, r3, [r0, #4*10] @ gteIR23 #else str r2, [r0, #4*10] str r3, [r0, #4*11] diff --git a/libpcsxcore/gte_neon.S b/libpcsxcore/gte_neon.S index 9fafb27..d83cf23 100644 --- a/libpcsxcore/gte_neon.S +++ b/libpcsxcore/gte_neon.S @@ -6,6 +6,8 @@ */ +.syntax unified + .bss .align 6 @ cacheline @@ -271,11 +273,11 @@ gteRTPS_neon: orrne lr, #(1<<13) @ limG2 orrne lr, #(1<<31) adds r2, r4, #1 - addvcs r3, r5, #1 + addsvc r3, r5, #1 orrvs lr, #(1<<16) @ F orrvs lr, #(1<<31) subs r2, r4, #1 - subvcs r3, r5, #1 + subsvc r3, r5, #1 orrvs lr, #(1<<31) ldr r4, [r0, #4*24] @ gteMAC0 @@ -506,13 +508,13 @@ gteRTPT_neon: add r2, r4, #0x400<<16 @ min fSX add r3, r6, #0x400<<16 @ max fSX lsrs r2, #16+11 - lsreqs r3, #16+11 + lsrseq 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 + lsrseq r3, #16+11 orrne lr, #(1<<31) @ limG2 orrne lr, #(1<<13) adds r2, r9, #1 |