~/f/rav1e/RPMS.2 ~/f/rav1e ~/f/rav1e RPMS.2/librav1e0_8-0.8.1-1.1.x86_64.rpm RPMS/librav1e0_8-0.8.1-1.1.x86_64.rpm differ: byte 225, line 1 Comparing librav1e0_8-0.8.1-1.1.x86_64.rpm to librav1e0_8-0.8.1-1.1.x86_64.rpm comparing the rpm tags of librav1e0_8 --- old-rpm-tags +++ new-rpm-tags @@ -107 +107 @@ -/usr/lib64/librav1e.so.0.8.1 ba584173aa11d47cb60a6b655c778179ec6ff7f669d65b3b48f7fcadb1378a5d 0 +/usr/lib64/librav1e.so.0.8.1 c2b20fc75555da0ce6e008fc4359bd45beccda2cb1d9afdc1fe71a29ee382789 0 comparing rpmtags comparing RELEASE comparing PROVIDES comparing scripts comparing filelist comparing file checksum creating rename script RPM file checksum differs. Extracting packages /usr/lib64/librav1e.so.0.8.1 differs in ELF sections --- old /usr/lib64/librav1e.so.0.8.1 (objdump) +++ new /usr/lib64/librav1e.so.0.8.1 (objdump) @@ -107,61 +107,61 @@ 0930 cc060000 12000000 00000000 00000000 ................ 0940 00000000 00000000 da060000 12000000 ................ 0950 00000000 00000000 00000000 00000000 ................ - 0960 f6000000 12001000 d0d11100 00000000 ................ + 0960 f6000000 12001000 90d11100 00000000 ................ 0970 da030000 00000000 30010000 12001000 ........0....... - 0980 d0d51100 00000000 1a000000 00000000 ................ - 0990 6a010000 12001000 00d61100 00000000 j............... + 0980 90d51100 00000000 1a000000 00000000 ................ + 0990 6a010000 12001000 c0d51100 00000000 j............... 09a0 45000000 00000000 ce000000 12001000 E............... - 09b0 10cf1100 00000000 94020000 00000000 ................ - 09c0 c7010000 12001000 40d81100 00000000 ........@....... + 09b0 d0ce1100 00000000 94020000 00000000 ................ + 09c0 c7010000 12001000 00d81100 00000000 ................ 09d0 0f000000 00000000 07020000 12001000 ................ - 09e0 60d81100 00000000 29000000 00000000 `.......)....... - 09f0 1a020000 12001000 90d81100 00000000 ................ + 09e0 20d81100 00000000 29000000 00000000 .......)....... + 09f0 1a020000 12001000 50d81100 00000000 ........P....... 0a00 9f010000 00000000 0d030000 12001000 ................ - 0a10 a0e01100 00000000 14000000 00000000 ................ - 0a20 4d030000 12001000 a0e31100 00000000 M............... + 0a10 60e01100 00000000 14000000 00000000 `............... + 0a20 4d030000 12001000 60e31100 00000000 M.......`....... 0a30 2a000000 00000000 4f010000 12001000 *.......O....... - 0a40 f0d51100 00000000 07000000 00000000 ................ - 0a50 ec010000 12001000 50d81100 00000000 ........P....... + 0a40 b0d51100 00000000 07000000 00000000 ................ + 0a50 ec010000 12001000 10d81100 00000000 ................ 0a60 0f000000 00000000 4c020000 12001000 ........L....... - 0a70 60db1100 00000000 29000000 00000000 `.......)....... - 0a80 71020000 12001000 c0db1100 00000000 q............... + 0a70 20db1100 00000000 29000000 00000000 .......)....... + 0a80 71020000 12001000 80db1100 00000000 q............... 0a90 c1000000 00000000 a8020000 12001000 ................ - 0aa0 20dd1100 00000000 e9000000 00000000 ............... - 0ab0 32030000 12001000 50e11100 00000000 2.......P....... + 0aa0 e0dc1100 00000000 e9000000 00000000 ................ + 0ab0 32030000 12001000 10e11100 00000000 2............... 0ac0 49020000 00000000 e3000000 12001000 I............... - 0ad0 b0d11100 00000000 13000000 00000000 ................ - 0ae0 e6020000 12001000 60e01100 00000000 ........`....... + 0ad0 70d11100 00000000 13000000 00000000 p............... + 0ae0 e6020000 12001000 20e01100 00000000 ........ ....... 0af0 06000000 00000000 04040000 12001000 ................ - 0b00 90f11100 00000000 ca000000 00000000 ................ - 0b10 3a020000 12001000 30da1100 00000000 :.......0....... + 0b00 50f11100 00000000 ca000000 00000000 P............... + 0b10 3a020000 12001000 f0d91100 00000000 :............... 0b20 2e010000 00000000 8e020000 12001000 ................ - 0b30 90dc1100 00000000 84000000 00000000 ................ - 0b40 d8030000 12001000 d0f01100 00000000 ................ + 0b30 50dc1100 00000000 84000000 00000000 P............... + 0b40 d8030000 12001000 90f01100 00000000 ................ 0b50 50000000 00000000 60020000 12001000 P.......`....... - 0b60 90db1100 00000000 28000000 00000000 ........(....... - 0b70 cf020000 12001000 f0df1100 00000000 ................ + 0b60 50db1100 00000000 28000000 00000000 P.......(....... + 0b70 cf020000 12001000 b0df1100 00000000 ................ 0b80 69000000 00000000 88030000 12001000 i............... - 0b90 b0e41100 00000000 06000000 00000000 ................ - 0ba0 c4030000 12001000 c0f01100 00000000 ................ + 0b90 70e41100 00000000 06000000 00000000 p............... + 0ba0 c4030000 12001000 80f01100 00000000 ................ 0bb0 10000000 00000000 29040000 12001000 ........)....... - 0bc0 88f21100 00000000 08000000 00000000 ................ - 0bd0 0d010000 12001000 b0d51100 00000000 ................ + 0bc0 48f21100 00000000 08000000 00000000 H............... + 0bd0 0d010000 12001000 70d51100 00000000 ........p....... 0be0 15000000 00000000 8d010000 12001000 ................ - 0bf0 50d61100 00000000 37000000 00000000 P.......7....... - 0c00 fb020000 12001000 70e01100 00000000 ........p....... + 0bf0 10d61100 00000000 37000000 00000000 ........7....... + 0c00 fb020000 12001000 30e01100 00000000 ........0....... 0c10 29000000 00000000 f3030000 12001000 )............... - 0c20 20f11100 00000000 6e000000 00000000 .......n....... - 0c30 16040000 12001000 60f21100 00000000 ........`....... + 0c20 e0f01100 00000000 6e000000 00000000 ........n....... + 0c30 16040000 12001000 20f21100 00000000 ........ ....... 0c40 28000000 00000000 ab010000 12001000 (............... - 0c50 90d61100 00000000 a4010000 00000000 ................ - 0c60 bf020000 12001000 10de1100 00000000 ................ + 0c50 50d61100 00000000 a4010000 00000000 P............... + 0c60 bf020000 12001000 d0dd1100 00000000 ................ 0c70 df010000 00000000 1f030000 12001000 ................ - 0c80 c0e01100 00000000 86000000 00000000 ................ - 0c90 70030000 12001000 d0e31100 00000000 p............... + 0c80 80e01100 00000000 86000000 00000000 ................ + 0c90 70030000 12001000 90e31100 00000000 p............... 0ca0 d9000000 00000000 9e030000 12001000 ................ - 0cb0 c0e41100 00000000 a9000000 00000000 ................ - 0cc0 b3030000 12001000 70e51100 00000000 ........p....... + 0cb0 80e41100 00000000 a9000000 00000000 ................ + 0cc0 b3030000 12001000 30e51100 00000000 ........0....... 0cd0 4b0b0000 00000000 K....... Contents of section .gnu.version: 0cd8 00000000 00000000 02000200 03000200 ................ @@ -398,8491 +398,8491 @@ 1af8 2e32006c 69627261 7631652e 736f2e30 .2.librav1e.so.0 1b08 2e3800 .8. Contents of section .rela.dyn: - 01b10 80803000 00000000 08000000 00000000 ..0............. - 01b20 e0d90800 00000000 88803000 00000000 ..........0..... - 01b30 08000000 00000000 82a12400 00000000 ..........$..... - 01b40 90803000 00000000 08000000 00000000 ..0............. - 01b50 20da0800 00000000 98803000 00000000 .........0..... + 01b10 40703000 00000000 08000000 00000000 @p0............. + 01b20 a0d90800 00000000 48703000 00000000 ........Hp0..... + 01b30 08000000 00000000 42a12400 00000000 ........B.$..... + 01b40 50703000 00000000 08000000 00000000 Pp0............. + 01b50 e0d90800 00000000 58703000 00000000 ........Xp0..... 01b60 08000000 00000000 e6ee0200 00000000 ................ - 01b70 b0803000 00000000 08000000 00000000 ..0............. - 01b80 e6ee0200 00000000 c8803000 00000000 ..........0..... + 01b70 70703000 00000000 08000000 00000000 pp0............. + 01b80 e6ee0200 00000000 88703000 00000000 .........p0..... 01b90 08000000 00000000 e6ee0200 00000000 ................ - 01ba0 e0803000 00000000 08000000 00000000 ..0............. - 01bb0 e6ee0200 00000000 f8803000 00000000 ..........0..... + 01ba0 a0703000 00000000 08000000 00000000 .p0............. + 01bb0 e6ee0200 00000000 b8703000 00000000 .........p0..... 01bc0 08000000 00000000 e6ee0200 00000000 ................ - 01bd0 10813000 00000000 08000000 00000000 ..0............. - 01be0 e6ee0200 00000000 28813000 00000000 ........(.0..... + 01bd0 d0703000 00000000 08000000 00000000 .p0............. + 01be0 e6ee0200 00000000 e8703000 00000000 .........p0..... 01bf0 08000000 00000000 e6ee0200 00000000 ................ - 01c00 40813000 00000000 08000000 00000000 @.0............. - 01c10 e6ee0200 00000000 58813000 00000000 ........X.0..... + 01c00 00713000 00000000 08000000 00000000 .q0............. + 01c10 e6ee0200 00000000 18713000 00000000 .........q0..... 01c20 08000000 00000000 e6ee0200 00000000 ................ - 01c30 70813000 00000000 08000000 00000000 p.0............. - 01c40 e6ee0200 00000000 88813000 00000000 ..........0..... + 01c30 30713000 00000000 08000000 00000000 0q0............. + 01c40 e6ee0200 00000000 48713000 00000000 ........Hq0..... 01c50 08000000 00000000 e6ee0200 00000000 ................ - 01c60 a0813000 00000000 08000000 00000000 ..0............. - 01c70 e6ee0200 00000000 b8813000 00000000 ..........0..... + 01c60 60713000 00000000 08000000 00000000 `q0............. + 01c70 e6ee0200 00000000 78713000 00000000 ........xq0..... 01c80 08000000 00000000 3aec0200 00000000 ........:....... - 01c90 d0813000 00000000 08000000 00000000 ..0............. - 01ca0 601b1000 00000000 e8813000 00000000 `.........0..... - 01cb0 08000000 00000000 a0391000 00000000 .........9...... - 01cc0 f0813000 00000000 08000000 00000000 ..0............. - 01cd0 e6ee0200 00000000 08823000 00000000 ..........0..... + 01c90 90713000 00000000 08000000 00000000 .q0............. + 01ca0 201b1000 00000000 a8713000 00000000 ........q0..... + 01cb0 08000000 00000000 60391000 00000000 ........`9...... + 01cc0 b0713000 00000000 08000000 00000000 .q0............. + 01cd0 e6ee0200 00000000 c8713000 00000000 .........q0..... 01ce0 08000000 00000000 e6ee0200 00000000 ................ - 01cf0 20823000 00000000 08000000 00000000 .0............. - 01d00 e6ee0200 00000000 38823000 00000000 ........8.0..... + 01cf0 e0713000 00000000 08000000 00000000 .q0............. + 01d00 e6ee0200 00000000 f8713000 00000000 .........q0..... 01d10 08000000 00000000 e6ee0200 00000000 ................ - 01d20 50823000 00000000 08000000 00000000 P.0............. - 01d30 e6ee0200 00000000 68823000 00000000 ........h.0..... + 01d20 10723000 00000000 08000000 00000000 .r0............. + 01d30 e6ee0200 00000000 28723000 00000000 ........(r0..... 01d40 08000000 00000000 e6ee0200 00000000 ................ - 01d50 80823000 00000000 08000000 00000000 ..0............. - 01d60 e6ee0200 00000000 98823000 00000000 ..........0..... + 01d50 40723000 00000000 08000000 00000000 @r0............. + 01d60 e6ee0200 00000000 58723000 00000000 ........Xr0..... 01d70 08000000 00000000 e6ee0200 00000000 ................ - 01d80 b0823000 00000000 08000000 00000000 ..0............. - 01d90 e6ee0200 00000000 c8823000 00000000 ..........0..... + 01d80 70723000 00000000 08000000 00000000 pr0............. + 01d90 e6ee0200 00000000 88723000 00000000 .........r0..... 01da0 08000000 00000000 e6ee0200 00000000 ................ - 01db0 e0823000 00000000 08000000 00000000 ..0............. - 01dc0 e6ee0200 00000000 f8823000 00000000 ..........0..... + 01db0 a0723000 00000000 08000000 00000000 .r0............. + 01dc0 e6ee0200 00000000 b8723000 00000000 .........r0..... 01dd0 08000000 00000000 e6ee0200 00000000 ................ - 01de0 10833000 00000000 08000000 00000000 ..0............. - 01df0 e6ee0200 00000000 28833000 00000000 ........(.0..... + 01de0 d0723000 00000000 08000000 00000000 .r0............. + 01df0 e6ee0200 00000000 e8723000 00000000 .........r0..... 01e00 08000000 00000000 e6ee0200 00000000 ................ - 01e10 40833000 00000000 08000000 00000000 @.0............. - 01e20 e6ee0200 00000000 58833000 00000000 ........X.0..... + 01e10 00733000 00000000 08000000 00000000 .s0............. + 01e20 e6ee0200 00000000 18733000 00000000 .........s0..... 01e30 08000000 00000000 e6ee0200 00000000 ................ - 01e40 70833000 00000000 08000000 00000000 p.0............. - 01e50 e6ee0200 00000000 88833000 00000000 ..........0..... + 01e40 30733000 00000000 08000000 00000000 0s0............. + 01e50 e6ee0200 00000000 48733000 00000000 ........Hs0..... 01e60 08000000 00000000 e6ee0200 00000000 ................ - 01e70 a0833000 00000000 08000000 00000000 ..0............. - 01e80 e6ee0200 00000000 b8833000 00000000 ..........0..... + 01e70 60733000 00000000 08000000 00000000 `s0............. /usr/lib64/librav1e.so.0.8.1 differs in assembler output --- old /usr/lib64/librav1e.so.0.8.1 (disasm) +++ new /usr/lib64/librav1e.so.0.8.1 (disasm) @@ -436846,4785 +436846,11235 @@ vzeroupper ret -rav1e_cdef_filter_4x4_16bpc_avx2: - push %rbx - push %rbp - mov %rsp,%rbp - and $something,%rsp +rav1e_cdef_filter_8x8_8bpc_sse4: sub $something,%rsp - mov offset(%rbp),%r10d - lea -offset(%rip),%r11 # - vmovdqu (%rdi),%xmm0 - vmovdqu (%rdi,%rsi,1),%xmm1 - lea (%rsi,%rsi,2),%rbx - vmovdqu (%rdi,%rsi,2),%xmm2 - vmovdqu (%rdi,%rbx,1),%xmm3 - vpbroadcastd offset(%r11),%ymm7 - vmovdqa %xmm0,offset(%rsp) - vmovdqa %xmm1,offset(%rsp) - vmovdqa %xmm2,offset(%rsp) - vmovdqa %xmm3,offset(%rsp) + mov offset(%rsp),%r10d + movdqa -offset(%rip),%xmm6 # + pxor %xmm7,%xmm7 + lea (%rdi,%rsi,4),%rax + lea (%rsi,%rsi,2),%r11 test $something,%r10b - je - vmovdqu (%rcx),%xmm0 - vmovdqu (%rcx,%rsi,1),%xmm1 - vmovdqa %xmm0,offset(%rsp) - vmovdqa %xmm1,offset(%rsp) + je + movdqu (%rdi),%xmm0 + movdqu (%rdi,%rsi,1),%xmm1 + movdqu (%rdi,%rsi,2),%xmm2 + movdqu (%rdi,%r11,1),%xmm3 + movdqa %xmm0,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm0 + movdqa %xmm4,offset(%rsp) + movdqa %xmm0,offset(%rsp) + movdqa %xmm1,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm1 + movdqa %xmm4,offset(%rsp) + movdqa %xmm1,offset(%rsp) + movdqa %xmm2,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm2 + movdqa %xmm4,offset(%rsp) + movdqa %xmm2,offset(%rsp) + movdqa %xmm3,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm3 + movdqa %xmm4,offset(%rsp) + movdqa %xmm3,offset(%rsp) + movdqu (%rax),%xmm0 + movdqu (%rax,%rsi,1),%xmm1 + movdqu (%rax,%rsi,2),%xmm2 + movdqu (%rax,%r11,1),%xmm3 + movdqa %xmm0,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm0 + movdqa %xmm4,offset(%rsp) + movdqa %xmm0,offset(%rsp) + movdqa %xmm1,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm1 + movdqa %xmm4,offset(%rsp) + movdqa %xmm1,offset(%rsp) + movdqa %xmm2,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm2 + movdqa %xmm4,offset(%rsp) + movdqa %xmm2,offset(%rsp) + movdqa %xmm3,%xmm4 + punpcklbw %xmm7,%xmm4 + punpckhbw %xmm7,%xmm3 + movdqa %xmm4,offset(%rsp) + movdqa %xmm3,offset(%rsp) + jmp + pmovzxbw (%rdi),%xmm0 + pmovzxbw (%rdi,%rsi,1),%xmm1 + pmovzxbw (%rdi,%rsi,2),%xmm2 + pmovzxbw (%rdi,%r11,1),%xmm3 + movdqa %xmm0,offset(%rsp) + movdqa %xmm1,offset(%rsp) + movdqa %xmm2,offset(%rsp) + movdqa %xmm3,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) + pmovzxbw (%rax),%xmm0 + pmovzxbw (%rax,%rsi,1),%xmm1 + pmovzxbw (%rax,%rsi,2),%xmm2 + pmovzxbw (%rax,%r11,1),%xmm3 + movdqa %xmm0,offset(%rsp) + movdqa %xmm1,offset(%rsp) + movdqa %xmm2,offset(%rsp) + movdqa %xmm3,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) test $something,%r10b - je - vmovd -offset(%rcx),%xmm0 - vmovd -offset(%rcx,%rsi,1),%xmm1 - vmovd %xmm0,offset(%rsp) - vmovd %xmm1,offset(%rsp) - jmp - vmovdqa %ymm7,offset(%rsp) - vmovd %xmm7,offset(%rsp) - vmovd %xmm7,offset(%rsp) + je test $something,%r10b - je - vmovdqu (%r8),%xmm0 - vmovdqu (%r8,%rsi,1),%xmm1 - vmovdqa %xmm0,offset(%rsp) - vmovdqa %xmm1,offset(%rsp) + je test $something,%r10b - je - vmovd -offset(%r8),%xmm0 - vmovd -offset(%r8,%rsi,1),%xmm1 - vmovd %xmm0,offset(%rsp) - vmovd %xmm1,offset(%rsp) - jmp - vmovdqa %ymm7,offset(%rsp) - vmovd %xmm7,offset(%rsp) - vmovd %xmm7,offset(%rsp) + je + movdqu -offset(%rcx),%xmm0 + movdqu -offset(%rcx,%rsi,1),%xmm1 + movdqa %xmm0,%xmm2 + punpckhbw %xmm7,%xmm2 + punpcklbw %xmm7,%xmm0 + movdqa %xmm1,%xmm3 + punpckhbw %xmm7,%xmm3 + punpcklbw %xmm7,%xmm1 + movdqu %xmm2,offset(%rsp) + movdqu %xmm3,offset(%rsp) + movdqu %xmm0,offset(%rsp) + movdqu %xmm1,offset(%rsp) + jmp + movdqu -offset(%rcx),%xmm0 + movdqu -offset(%rcx,%rsi,1),%xmm1 + movdqa %xmm0,%xmm2 + punpckhbw %xmm7,%xmm2 + punpcklbw %xmm7,%xmm0 + movdqa %xmm1,%xmm3 + punpckhbw %xmm7,%xmm3 + punpcklbw %xmm7,%xmm1 + movdqa %xmm0,offset(%rsp) + movdqa %xmm2,offset(%rsp) + movdqa %xmm1,offset(%rsp) + movdqa %xmm3,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) + jmp test $something,%r10b - je - vmovd (%rdx),%xmm0 - vmovd offset(%rdx),%xmm1 - vmovd offset(%rdx),%xmm2 - vmovd offset(%rdx),%xmm3 - vmovd %xmm0,offset(%rsp) - vmovd %xmm1,offset(%rsp) - vmovd %xmm2,offset(%rsp) - vmovd %xmm3,offset(%rsp) - jmp - vmovd %xmm7,offset(%rsp) - vmovd %xmm7,offset(%rsp) - vmovd %xmm7,offset(%rsp) - vmovd %xmm7,offset(%rsp) + je + movdqu (%rcx),%xmm0 + movdqu (%rcx,%rsi,1),%xmm1 + movdqa %xmm0,%xmm2 + punpckhbw %xmm7,%xmm2 + punpcklbw %xmm7,%xmm0 + movdqa %xmm1,%xmm3 + punpckhbw %xmm7,%xmm3 + punpcklbw %xmm7,%xmm1 + movd %xmm2,offset(%rsp) + movd %xmm3,offset(%rsp) + movd %xmm6,offset(%rsp) + movd %xmm6,offset(%rsp) + movdqa %xmm0,offset(%rsp) + movdqa %xmm1,offset(%rsp) + jmp + pmovzxbw (%rcx),%xmm0 + pmovzxbw (%rcx,%rsi,1),%xmm1 RPMS.2/rav1e-0.8.1-1.1.x86_64.rpm RPMS/rav1e-0.8.1-1.1.x86_64.rpm differ: byte 225, line 1 Comparing rav1e-0.8.1-1.1.x86_64.rpm to rav1e-0.8.1-1.1.x86_64.rpm comparing the rpm tags of rav1e --- old-rpm-tags +++ new-rpm-tags @@ -218 +218 @@ -/usr/bin/rav1e 1b58b2726a3a065fcf78504c1297a66429526cfec225f3eef6a10cdd5e157972 0 +/usr/bin/rav1e 65711a078e82badc648c54b7357c4a8d2e49e9c5c82a0dab9cd09569caadf623 0 comparing rpmtags comparing RELEASE comparing PROVIDES comparing scripts comparing filelist comparing file checksum creating rename script RPM file checksum differs. Extracting packages /usr/bin/rav1e differs in ELF sections --- old /usr/bin/rav1e (objdump) +++ new /usr/bin/rav1e (objdump) @@ -2281,2659 +2281,2659 @@ 090a8 08000000 00000000 e6e90200 00000000 ................ 090b8 c8a84300 00000000 08000000 00000000 ..C............. 090c8 3de50200 00000000 78b34300 00000000 =.......x.C..... - 090d8 08000000 00000000 a0b13d00 00000000 ..........=..... + 090d8 08000000 00000000 80173c00 00000000 ..........<..... 090e8 10bd4300 00000000 08000000 00000000 ..C............. - 090f8 70ac3d00 00000000 18bd4300 00000000 p.=.......C..... - 09108 08000000 00000000 b0ac3d00 00000000 ..........=..... + 090f8 50123c00 00000000 18bd4300 00000000 P.<.......C..... + 09108 08000000 00000000 90123c00 00000000 ..........<..... 09118 20bd4300 00000000 08000000 00000000 .C............. - 09128 70ae3d00 00000000 28bd4300 00000000 p.=.....(.C..... - 09138 08000000 00000000 80ae3d00 00000000 ..........=..... + 09128 50143c00 00000000 28bd4300 00000000 P.<.....(.C..... + 09138 08000000 00000000 60143c00 00000000 ........`.<..... 09148 30bd4300 00000000 08000000 00000000 0.C............. - 09158 c0ac3d00 00000000 38bd4300 00000000 ..=.....8.C..... - 09168 08000000 00000000 00b03d00 00000000 ..........=..... + 09158 a0123c00 00000000 38bd4300 00000000 ..<.....8.C..... + 09168 08000000 00000000 e0153c00 00000000 ..........<..... 09178 40bd4300 00000000 08000000 00000000 @.C............. - 09188 20b03d00 00000000 48bd4300 00000000 .=.....H.C..... - 09198 08000000 00000000 10b03d00 00000000 ..........=..... + 09188 00163c00 00000000 48bd4300 00000000 ..<.....H.C..... + 09198 08000000 00000000 f0153c00 00000000 ..........<..... 091a8 50bd4300 00000000 08000000 00000000 P.C............. - 091b8 90ae3d00 00000000 58bd4300 00000000 ..=.....X.C..... - 091c8 08000000 00000000 20b13d00 00000000 ........ .=..... + 091b8 70143c00 00000000 58bd4300 00000000 p.<.....X.C..... + 091c8 08000000 00000000 00173c00 00000000 ..........<..... 091d8 60bd4300 00000000 08000000 00000000 `.C............. - 091e8 f0b03d00 00000000 68bd4300 00000000 ..=.....h.C..... - 091f8 08000000 00000000 d0ac3d00 00000000 ..........=..... + 091e8 d0163c00 00000000 68bd4300 00000000 ..<.....h.C..... + 091f8 08000000 00000000 b0123c00 00000000 ..........<..... 09208 70bd4300 00000000 08000000 00000000 p.C............. - 09218 00b13d00 00000000 78bd4300 00000000 ..=.....x.C..... - 09228 08000000 00000000 a0ae3d00 00000000 ..........=..... + 09218 e0163c00 00000000 78bd4300 00000000 ..<.....x.C..... + 09228 08000000 00000000 80143c00 00000000 ..........<..... 09238 80bd4300 00000000 08000000 00000000 ..C............. - 09248 10b13d00 00000000 88bd4300 00000000 ..=.......C..... - 09258 08000000 00000000 30b03d00 00000000 ........0.=..... + 09248 f0163c00 00000000 88bd4300 00000000 ..<.......C..... + 09258 08000000 00000000 10163c00 00000000 ..........<..... 09268 98bd4300 00000000 08000000 00000000 ..C............. - 09278 b0c03d00 00000000 a0bd4300 00000000 ..=.......C..... - 09288 08000000 00000000 80c13d00 00000000 ..........=..... + 09278 90263c00 00000000 a0bd4300 00000000 .&<.......C..... + 09288 08000000 00000000 60273c00 00000000 ........`'<..... 09298 a8bd4300 00000000 08000000 00000000 ..C............. - 092a8 d0c63d00 00000000 b0bd4300 00000000 ..=.......C..... - 092b8 08000000 00000000 f0c63d00 00000000 ..........=..... + 092a8 b02c3c00 00000000 b0bd4300 00000000 .,<.......C..... + 092b8 08000000 00000000 d02c3c00 00000000 .........,<..... 092c8 b8bd4300 00000000 08000000 00000000 ..C............. - 092d8 a0c13d00 00000000 c0bd4300 00000000 ..=.......C..... - 092e8 08000000 00000000 10cc3d00 00000000 ..........=..... + 092d8 80273c00 00000000 c0bd4300 00000000 .'<.......C..... + 092e8 08000000 00000000 f0313c00 00000000 .........1<..... 092f8 c8bd4300 00000000 08000000 00000000 ..C............. - 09308 50cc3d00 00000000 d0bd4300 00000000 P.=.......C..... - 09318 08000000 00000000 30cc3d00 00000000 ........0.=..... + 09308 30323c00 00000000 d0bd4300 00000000 02<.......C..... + 09318 08000000 00000000 10323c00 00000000 .........2<..... 09328 d8bd4300 00000000 08000000 00000000 ..C............. - 09338 10c73d00 00000000 e0bd4300 00000000 ..=.......C..... - 09348 08000000 00000000 10ce3d00 00000000 ..........=..... + 09338 f02c3c00 00000000 e0bd4300 00000000 .,<.......C..... + 09348 08000000 00000000 f0333c00 00000000 .........3<..... 09358 e8bd4300 00000000 08000000 00000000 ..C............. - 09368 b0cd3d00 00000000 f0bd4300 00000000 ..=.......C..... - 09378 08000000 00000000 c0c13d00 00000000 ..........=..... + 09368 90333c00 00000000 f0bd4300 00000000 .3<.......C..... + 09378 08000000 00000000 a0273c00 00000000 .........'<..... 09388 f8bd4300 00000000 08000000 00000000 ..C............. - 09398 d0cd3d00 00000000 00be4300 00000000 ..=.......C..... - 093a8 08000000 00000000 30c73d00 00000000 ........0.=..... + 09398 b0333c00 00000000 00be4300 00000000 .3<.......C..... + 093a8 08000000 00000000 102d3c00 00000000 .........-<..... 093b8 08be4300 00000000 08000000 00000000 ..C............. - 093c8 f0cd3d00 00000000 10be4300 00000000 ..=.......C..... - 093d8 08000000 00000000 70cc3d00 00000000 ........p.=..... + 093c8 d0333c00 00000000 10be4300 00000000 .3<.......C..... + 093d8 08000000 00000000 50323c00 00000000 ........P2<..... 093e8 20be4300 00000000 08000000 00000000 .C............. - 093f8 700a3e00 00000000 28be4300 00000000 p.>.....(.C..... - 09408 08000000 00000000 d00a3e00 00000000 ..........>..... + 093f8 50703c00 00000000 28be4300 00000000 Pp<.....(.C..... + 09408 08000000 00000000 b0703c00 00000000 .........p<..... 09418 30be4300 00000000 08000000 00000000 0.C............. - 09428 200f3e00 00000000 38be4300 00000000 .>.....8.C..... - 09438 08000000 00000000 400f3e00 00000000 ........@.>..... + 09428 00753c00 00000000 38be4300 00000000 .u<.....8.C..... + 09438 08000000 00000000 20753c00 00000000 ........ u<..... 09448 40be4300 00000000 08000000 00000000 @.C............. - 09458 f00a3e00 00000000 48be4300 00000000 ..>.....H.C..... - 09468 08000000 00000000 a0133e00 00000000 ..........>..... + 09458 d0703c00 00000000 48be4300 00000000 .p<.....H.C..... + 09468 08000000 00000000 80793c00 00000000 .........y<..... 09478 50be4300 00000000 08000000 00000000 P.C............. - 09488 e0133e00 00000000 58be4300 00000000 ..>.....X.C..... - 09498 08000000 00000000 c0133e00 00000000 ..........>..... + 09488 c0793c00 00000000 58be4300 00000000 .y<.....X.C..... + 09498 08000000 00000000 a0793c00 00000000 .........y<..... 094a8 60be4300 00000000 08000000 00000000 `.C............. - 094b8 600f3e00 00000000 68be4300 00000000 `.>.....h.C..... - 094c8 08000000 00000000 00193e00 00000000 ..........>..... + 094b8 40753c00 00000000 68be4300 00000000 @u<.....h.C..... + 094c8 08000000 00000000 e07e3c00 00000000 .........~<..... 094d8 70be4300 00000000 08000000 00000000 p.C............. - 094e8 e0183e00 00000000 78be4300 00000000 ..>.....x.C..... - 094f8 08000000 00000000 100b3e00 00000000 ..........>..... + 094e8 c07e3c00 00000000 78be4300 00000000 .~<.....x.C..... + 094f8 08000000 00000000 f0703c00 00000000 .........p<..... 09508 a8be4300 00000000 08000000 00000000 ..C............. - 09518 f0533e00 00000000 f0be4300 00000000 .S>.......C..... - 09528 08000000 00000000 205d3e00 00000000 ........ ]>..... + 09518 d0b93c00 00000000 f0be4300 00000000 ..<.......C..... + 09528 08000000 00000000 00c33c00 00000000 ..........<..... 09538 30bf4300 00000000 08000000 00000000 0.C............. - 09548 10a33e00 00000000 b8bf4300 00000000 ..>.......C..... - 09558 08000000 00000000 b0b23d00 00000000 ..........=..... + 09548 f0083d00 00000000 b8bf4300 00000000 ..=.......C..... + 09558 08000000 00000000 90183c00 00000000 ..........<..... 09568 c0bf4300 00000000 08000000 00000000 ..C............. - 09578 00b33d00 00000000 c8bf4300 00000000 ..=.......C..... - 09588 08000000 00000000 c0b43d00 00000000 ..........=..... + 09578 e0183c00 00000000 c8bf4300 00000000 ..<.......C..... + 09588 08000000 00000000 a01a3c00 00000000 ..........<..... 09598 d0bf4300 00000000 08000000 00000000 ..C............. - 095a8 d0b43d00 00000000 d8bf4300 00000000 ..=.......C..... - 095b8 08000000 00000000 10b33d00 00000000 ..........=..... + 095a8 b01a3c00 00000000 d8bf4300 00000000 ..<.......C..... + 095b8 08000000 00000000 f0183c00 00000000 ..........<..... 095c8 e0bf4300 00000000 08000000 00000000 ..C............. - 095d8 30b83d00 00000000 e8bf4300 00000000 0.=.......C..... - 095e8 08000000 00000000 50b83d00 00000000 ........P.=..... + 095d8 101e3c00 00000000 e8bf4300 00000000 ..<.......C..... + 095e8 08000000 00000000 301e3c00 00000000 ........0.<..... 095f8 f0bf4300 00000000 08000000 00000000 ..C............. - 09608 40b83d00 00000000 f8bf4300 00000000 @.=.......C..... - 09618 08000000 00000000 e0b43d00 00000000 ..........=..... + 09608 201e3c00 00000000 f8bf4300 00000000 .<.......C..... + 09618 08000000 00000000 c01a3c00 00000000 ..........<..... 09628 00c04300 00000000 08000000 00000000 ..C............. - 09638 50b93d00 00000000 08c04300 00000000 P.=.......C..... - 09648 08000000 00000000 20b93d00 00000000 ........ .=..... + 09638 301f3c00 00000000 08c04300 00000000 0.<.......C..... + 09648 08000000 00000000 001f3c00 00000000 ..........<..... 09658 10c04300 00000000 08000000 00000000 ..C............. - 09668 20b33d00 00000000 18c04300 00000000 .=.......C..... - 09678 08000000 00000000 30b93d00 00000000 ........0.=..... + 09668 00193c00 00000000 18c04300 00000000 ..<.......C..... + 09678 08000000 00000000 101f3c00 00000000 ..........<..... 09688 20c04300 00000000 08000000 00000000 .C............. - 09698 f0b43d00 00000000 28c04300 00000000 ..=.....(.C..... - 096a8 08000000 00000000 40b93d00 00000000 ........@.=..... + 09698 d01a3c00 00000000 28c04300 00000000 ..<.....(.C..... + 096a8 08000000 00000000 201f3c00 00000000 ........ .<..... 096b8 30c04300 00000000 08000000 00000000 0.C............. - 096c8 60b83d00 00000000 40c04300 00000000 `.=.....@.C..... - 096d8 08000000 00000000 e0b93d00 00000000 ..........=..... + 096c8 401e3c00 00000000 40c04300 00000000 @.<.....@.C..... + 096d8 08000000 00000000 c01f3c00 00000000 ..........<..... 096e8 48c04300 00000000 08000000 00000000 H.C............. - 096f8 30ba3d00 00000000 50c04300 00000000 0.=.....P.C..... - 09708 08000000 00000000 e0bb3d00 00000000 ..........=..... + 096f8 10203c00 00000000 50c04300 00000000 . <.....P.C..... + 09708 08000000 00000000 c0213c00 00000000 .........!<..... 09718 58c04300 00000000 08000000 00000000 X.C............. - 09728 f0bb3d00 00000000 60c04300 00000000 ..=.....`.C..... - 09738 08000000 00000000 40ba3d00 00000000 ........@.=..... + 09728 d0213c00 00000000 60c04300 00000000 .!<.....`.C..... + 09738 08000000 00000000 20203c00 00000000 ........ <..... 09748 68c04300 00000000 08000000 00000000 h.C............. - 09758 a0be3d00 00000000 70c04300 00000000 ..=.....p.C..... - 09768 08000000 00000000 c0be3d00 00000000 ..........=..... + 09758 80243c00 00000000 70c04300 00000000 .$<.....p.C..... + 09768 08000000 00000000 a0243c00 00000000 .........$<..... 09778 78c04300 00000000 08000000 00000000 x.C............. - 09788 b0be3d00 00000000 80c04300 00000000 ..=.......C..... - 09798 08000000 00000000 00bc3d00 00000000 ..........=..... + 09788 90243c00 00000000 80c04300 00000000 .$<.......C..... + 09798 08000000 00000000 e0213c00 00000000 .........!<..... 097a8 88c04300 00000000 08000000 00000000 ..C............. - 097b8 d0bf3d00 00000000 90c04300 00000000 ..=.......C..... - 097c8 08000000 00000000 a0bf3d00 00000000 ..........=..... + 097b8 b0253c00 00000000 90c04300 00000000 .%<.......C..... + 097c8 08000000 00000000 80253c00 00000000 .........%<..... 097d8 98c04300 00000000 08000000 00000000 ..C............. - 097e8 50ba3d00 00000000 a0c04300 00000000 P.=.......C..... - 097f8 08000000 00000000 b0bf3d00 00000000 ..........=..... + 097e8 30203c00 00000000 a0c04300 00000000 0 <.......C..... + 097f8 08000000 00000000 90253c00 00000000 .........%<..... 09808 a8c04300 00000000 08000000 00000000 ..C............. - 09818 10bc3d00 00000000 b0c04300 00000000 ..=.......C..... /usr/bin/rav1e differs in assembler output --- old /usr/bin/rav1e (disasm) +++ new /usr/bin/rav1e (disasm) @@ -669478,39840 +669478,34485 @@ int3 int3 -rav1e_ipred_paeth_16bpc_avx512icl: - lea -offset(%rip),%rax # - tzcnt %ecx,%ecx - movslq (%rax,%rcx,4),%rcx - vpbroadcastw (%rdx),%zmm19 - add %rax,%rcx - jmp *%rcx - vpbroadcastq offset(%rdx),%zmm20 - vmovsldup -offset(%rax),%zmm23 - lea (%rsi,%rsi,2),%rax - vpsubw %zmm19,%zmm20,%zmm21 - vpabsw %zmm21,%zmm22 - sub $something,%rdx - vbroadcasti32x4 (%rdx),%zmm18 - vpshufb %zmm23,%zmm18,%zmm18 - vpaddw %zmm18,%zmm21,%zmm16 - vpsubw %zmm19,%zmm16,%zmm17 - vpsubw %zmm20,%zmm16,%zmm16 - vpabsw %zmm17,%zmm17 - vpabsw %zmm16,%zmm16 - vpcmpgtw %zmm17,%zmm16,%k1 - vpminsw %zmm17,%zmm16,%zmm16 - vpcmpgtw %zmm16,%zmm22,%k2 - vpblendmw %zmm19,%zmm20,%zmm16{%k1} - vpblendmw %zmm16,%zmm18,%zmm16{%k2} - vextracti32x4 $something,%zmm16,%xmm17 - vextracti32x4 $something,%ymm16,%xmm24 - vextracti32x4 $something,%zmm16,%xmm25 - vmovq %xmm16,(%rdi) - vmovq %xmm17,(%rdi,%rsi,1) - vmovq %xmm24,(%rdi,%rsi,2) - vmovq %xmm25,(%rdi,%rax,1) - sub $something,%r8d - jl +rav1e_cdef_filter_4x4_16bpc_avx2: + push %rbx + push %rbp + mov %rsp,%rbp + and $something,%rsp + sub $something,%rsp + mov offset(%rbp),%r10d + lea -offset(%rip),%r11 # + vmovdqu (%rdi),%xmm0 + vmovdqu (%rdi,%rsi,1),%xmm1 + lea (%rsi,%rsi,2),%rbx + vmovdqu (%rdi,%rsi,2),%xmm2 + vmovdqu (%rdi,%rbx,1),%xmm3 + vpbroadcastd offset(%r11),%ymm7 + vmovdqa %xmm0,offset(%rsp) + vmovdqa %xmm1,offset(%rsp) + vmovdqa %xmm2,offset(%rsp) + vmovdqa %xmm3,offset(%rsp) + test $something,%r10b + je + vmovdqu (%rcx),%xmm0 + vmovdqu (%rcx,%rsi,1),%xmm1 + vmovdqa %xmm0,offset(%rsp) + vmovdqa %xmm1,offset(%rsp) + test $something,%r10b + je + vmovd -offset(%rcx),%xmm0 + vmovd -offset(%rcx,%rsi,1),%xmm1 + vmovd %xmm0,offset(%rsp) + vmovd %xmm1,offset(%rsp) + jmp + vmovdqa %ymm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + test $something,%r10b + je + vmovdqu (%r8),%xmm0 + vmovdqu (%r8,%rsi,1),%xmm1 + vmovdqa %xmm0,offset(%rsp) + vmovdqa %xmm1,offset(%rsp) + test $something,%r10b + je + vmovd -offset(%r8),%xmm0 + vmovd -offset(%r8,%rsi,1),%xmm1 + vmovd %xmm0,offset(%rsp) + vmovd %xmm1,offset(%rsp) + jmp + vmovdqa %ymm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + test $something,%r10b + je + vmovd (%rdx),%xmm0 + vmovd offset(%rdx),%xmm1 + vmovd offset(%rdx),%xmm2 + vmovd offset(%rdx),%xmm3 + vmovd %xmm0,offset(%rsp) + vmovd %xmm1,offset(%rsp) + vmovd %xmm2,offset(%rsp) + vmovd %xmm3,offset(%rsp) + jmp + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + test $something,%r10b + jne + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + vmovd %xmm7,offset(%rsp) + mov offset(%rbp),%eax + mov offset(%rbp),%ecx + vpbroadcastd offset(%r11),%ymm8 + lea (%r11,%rcx,2),%rcx + test %r9d,%r9d + je + vmovd %r9d,%xmm6 + vpbroadcastw %xmm6,%ymm6 + lzcnt %r9d,%r8d + rorx $something,%r9d,%r10d + cmpl $something,offset(%rbp) + cmove %r10d,%r9d + mov offset(%rbp),%r10d + and $something,%r9d + sub $something,%r10d + vpbroadcastd offset(%r11,%r9,1),%ymm9 + vpbroadcastd offset(%r11,%r9,1),%ymm10 + test %eax,%eax + je + vmovd %eax,%xmm0 + vpbroadcastw %xmm0,%ymm13 + lzcnt %eax,%eax + xor %r9d,%r9d + add %r10d,%r8d + cmovs %r9d,%r8d + add %r10d,%eax + lea offset(%rsp),%r10 + mov %r8,(%rsp) + mov %rax,offset(%rsp) + call + jmp + add %r10d,%r8d + cmovs %eax,%r8d + lea offset(%rsp),%r10 + mov %r8,(%rsp) + call + mov %rbp,%rsp + pop %rbp + pop %rbx + vzeroupper + ret + mov offset(%rbp),%r10d + vmovd %eax,%xmm6 + vpbroadcastw %xmm6,%ymm6 + tzcnt %eax,%eax + sub %eax,%r10d + mov %r10,offset(%rsp) + lea offset(%rsp),%r10 + call + jmp + nopl offset(%rax,%rax,1) + movsbq offset(%rcx),%r8 + vmovdqa (%r10),%ymm1 + vpunpcklqdq offset(%r10),%ymm1,%ymm1 + vmovdqu (%r10,%r8,1),%ymm2 + vpunpcklqdq offset(%r10,%r8,1),%ymm2,%ymm2 + neg %r8 + vmovdqu (%r10,%r8,1),%ymm3 + vpunpcklqdq offset(%r10,%r8,1),%ymm3,%ymm3 + movsbq offset(%rcx),%r8 + vpsubw %ymm1,%ymm2,%ymm2 + vpsubw %ymm1,%ymm3,%ymm3 + vpabsw %ymm2,%ymm4 + vpsrlw offset(%rsp),%ymm4,%ymm5 + vpsubusw %ymm5,%ymm6,%ymm0 + vpabsw %ymm3,%ymm5 + vpminsw %ymm4,%ymm0,%ymm0 + vpsrlw offset(%rsp),%ymm5,%ymm4 + vpsignw %ymm2,%ymm0,%ymm0 + vpsubusw %ymm4,%ymm6,%ymm2 + vpminsw %ymm5,%ymm2,%ymm2 + vmovdqu (%r10,%r8,1),%ymm4 + vpunpcklqdq offset(%r10,%r8,1),%ymm4,%ymm4 + neg %r8 + vmovdqu (%r10,%r8,1),%ymm5 + vpunpcklqdq offset(%r10,%r8,1),%ymm5,%ymm5 + vpsubw %ymm1,%ymm4,%ymm4 + vpsubw %ymm1,%ymm5,%ymm5 + vpsignw %ymm3,%ymm2,%ymm2 + vpabsw %ymm4,%ymm3 + vpaddw %ymm2,%ymm0,%ymm0 + vpsrlw offset(%rsp),%ymm3,%ymm2 + vpsubusw %ymm2,%ymm6,%ymm7 + vpabsw %ymm5,%ymm2 + vpminsw %ymm3,%ymm7,%ymm7 overalldiffered=2 (number of pkgs that are not bit-by-bit identical: 0 is good) overall=1