summaryrefslogtreecommitdiffstats
path: root/media/libwebp/src/dsp/mips_macro.h
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-07 19:33:14 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-07 19:33:14 +0000
commit36d22d82aa202bb199967e9512281e9a53db42c9 (patch)
tree105e8c98ddea1c1e4784a60a5a6410fa416be2de /media/libwebp/src/dsp/mips_macro.h
parentInitial commit. (diff)
downloadfirefox-esr-36d22d82aa202bb199967e9512281e9a53db42c9.tar.xz
firefox-esr-36d22d82aa202bb199967e9512281e9a53db42c9.zip
Adding upstream version 115.7.0esr.upstream/115.7.0esr
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to '')
-rw-r--r--media/libwebp/src/dsp/mips_macro.h200
1 files changed, 200 insertions, 0 deletions
diff --git a/media/libwebp/src/dsp/mips_macro.h b/media/libwebp/src/dsp/mips_macro.h
new file mode 100644
index 0000000000..44aba9b71d
--- /dev/null
+++ b/media/libwebp/src/dsp/mips_macro.h
@@ -0,0 +1,200 @@
+// Copyright 2014 Google Inc. All Rights Reserved.
+//
+// Use of this source code is governed by a BSD-style license
+// that can be found in the COPYING file in the root of the source
+// tree. An additional intellectual property rights grant can be found
+// in the file PATENTS. All contributing project authors may
+// be found in the AUTHORS file in the root of the source tree.
+// -----------------------------------------------------------------------------
+//
+// MIPS common macros
+
+#ifndef WEBP_DSP_MIPS_MACRO_H_
+#define WEBP_DSP_MIPS_MACRO_H_
+
+#if defined(__GNUC__) && defined(__ANDROID__) && LOCAL_GCC_VERSION == 0x409
+#define WORK_AROUND_GCC
+#endif
+
+#define STR(s) #s
+#define XSTR(s) STR(s)
+
+// O0[31..16 | 15..0] = I0[31..16 | 15..0] + I1[31..16 | 15..0]
+// O1[31..16 | 15..0] = I0[31..16 | 15..0] - I1[31..16 | 15..0]
+// O - output
+// I - input (macro doesn't change it)
+#define ADD_SUB_HALVES(O0, O1, \
+ I0, I1) \
+ "addq.ph %[" #O0 "], %[" #I0 "], %[" #I1 "] \n\t" \
+ "subq.ph %[" #O1 "], %[" #I0 "], %[" #I1 "] \n\t"
+
+// O - output
+// I - input (macro doesn't change it)
+// I[0/1] - offset in bytes
+#define LOAD_IN_X2(O0, O1, \
+ I0, I1) \
+ "lh %[" #O0 "], " #I0 "(%[in]) \n\t" \
+ "lh %[" #O1 "], " #I1 "(%[in]) \n\t"
+
+// I0 - location
+// I1..I9 - offsets in bytes
+#define LOAD_WITH_OFFSET_X4(O0, O1, O2, O3, \
+ I0, I1, I2, I3, I4, I5, I6, I7, I8, I9) \
+ "ulw %[" #O0 "], " #I1 "+" XSTR(I9) "*" #I5 "(%[" #I0 "]) \n\t" \
+ "ulw %[" #O1 "], " #I2 "+" XSTR(I9) "*" #I6 "(%[" #I0 "]) \n\t" \
+ "ulw %[" #O2 "], " #I3 "+" XSTR(I9) "*" #I7 "(%[" #I0 "]) \n\t" \
+ "ulw %[" #O3 "], " #I4 "+" XSTR(I9) "*" #I8 "(%[" #I0 "]) \n\t"
+
+// O - output
+// IO - input/output
+// I - input (macro doesn't change it)
+#define MUL_SHIFT_SUM(O0, O1, O2, O3, O4, O5, O6, O7, \
+ IO0, IO1, IO2, IO3, \
+ I0, I1, I2, I3, I4, I5, I6, I7) \
+ "mul %[" #O0 "], %[" #I0 "], %[kC2] \n\t" \
+ "mul %[" #O1 "], %[" #I0 "], %[kC1] \n\t" \
+ "mul %[" #O2 "], %[" #I1 "], %[kC2] \n\t" \
+ "mul %[" #O3 "], %[" #I1 "], %[kC1] \n\t" \
+ "mul %[" #O4 "], %[" #I2 "], %[kC2] \n\t" \
+ "mul %[" #O5 "], %[" #I2 "], %[kC1] \n\t" \
+ "mul %[" #O6 "], %[" #I3 "], %[kC2] \n\t" \
+ "mul %[" #O7 "], %[" #I3 "], %[kC1] \n\t" \
+ "sra %[" #O0 "], %[" #O0 "], 16 \n\t" \
+ "sra %[" #O1 "], %[" #O1 "], 16 \n\t" \
+ "sra %[" #O2 "], %[" #O2 "], 16 \n\t" \
+ "sra %[" #O3 "], %[" #O3 "], 16 \n\t" \
+ "sra %[" #O4 "], %[" #O4 "], 16 \n\t" \
+ "sra %[" #O5 "], %[" #O5 "], 16 \n\t" \
+ "sra %[" #O6 "], %[" #O6 "], 16 \n\t" \
+ "sra %[" #O7 "], %[" #O7 "], 16 \n\t" \
+ "addu %[" #IO0 "], %[" #IO0 "], %[" #I4 "] \n\t" \
+ "addu %[" #IO1 "], %[" #IO1 "], %[" #I5 "] \n\t" \
+ "subu %[" #IO2 "], %[" #IO2 "], %[" #I6 "] \n\t" \
+ "subu %[" #IO3 "], %[" #IO3 "], %[" #I7 "] \n\t"
+
+// O - output
+// I - input (macro doesn't change it)
+#define INSERT_HALF_X2(O0, O1, \
+ I0, I1) \
+ "ins %[" #O0 "], %[" #I0 "], 16, 16 \n\t" \
+ "ins %[" #O1 "], %[" #I1 "], 16, 16 \n\t"
+
+// O - output
+// I - input (macro doesn't change it)
+#define SRA_16(O0, O1, O2, O3, \
+ I0, I1, I2, I3) \
+ "sra %[" #O0 "], %[" #I0 "], 16 \n\t" \
+ "sra %[" #O1 "], %[" #I1 "], 16 \n\t" \
+ "sra %[" #O2 "], %[" #I2 "], 16 \n\t" \
+ "sra %[" #O3 "], %[" #I3 "], 16 \n\t"
+
+// temp0[31..16 | 15..0] = temp8[31..16 | 15..0] + temp12[31..16 | 15..0]
+// temp1[31..16 | 15..0] = temp8[31..16 | 15..0] - temp12[31..16 | 15..0]
+// temp0[31..16 | 15..0] = temp0[31..16 >> 3 | 15..0 >> 3]
+// temp1[31..16 | 15..0] = temp1[31..16 >> 3 | 15..0 >> 3]
+// O - output
+// I - input (macro doesn't change it)
+#define SHIFT_R_SUM_X2(O0, O1, O2, O3, O4, O5, O6, O7, \
+ I0, I1, I2, I3, I4, I5, I6, I7) \
+ "addq.ph %[" #O0 "], %[" #I0 "], %[" #I4 "] \n\t" \
+ "subq.ph %[" #O1 "], %[" #I0 "], %[" #I4 "] \n\t" \
+ "addq.ph %[" #O2 "], %[" #I1 "], %[" #I5 "] \n\t" \
+ "subq.ph %[" #O3 "], %[" #I1 "], %[" #I5 "] \n\t" \
+ "addq.ph %[" #O4 "], %[" #I2 "], %[" #I6 "] \n\t" \
+ "subq.ph %[" #O5 "], %[" #I2 "], %[" #I6 "] \n\t" \
+ "addq.ph %[" #O6 "], %[" #I3 "], %[" #I7 "] \n\t" \
+ "subq.ph %[" #O7 "], %[" #I3 "], %[" #I7 "] \n\t" \
+ "shra.ph %[" #O0 "], %[" #O0 "], 3 \n\t" \
+ "shra.ph %[" #O1 "], %[" #O1 "], 3 \n\t" \
+ "shra.ph %[" #O2 "], %[" #O2 "], 3 \n\t" \
+ "shra.ph %[" #O3 "], %[" #O3 "], 3 \n\t" \
+ "shra.ph %[" #O4 "], %[" #O4 "], 3 \n\t" \
+ "shra.ph %[" #O5 "], %[" #O5 "], 3 \n\t" \
+ "shra.ph %[" #O6 "], %[" #O6 "], 3 \n\t" \
+ "shra.ph %[" #O7 "], %[" #O7 "], 3 \n\t"
+
+// precrq.ph.w temp0, temp8, temp2
+// temp0 = temp8[31..16] | temp2[31..16]
+// ins temp2, temp8, 16, 16
+// temp2 = temp8[31..16] | temp2[15..0]
+// O - output
+// IO - input/output
+// I - input (macro doesn't change it)
+#define PACK_2_HALVES_TO_WORD(O0, O1, O2, O3, \
+ IO0, IO1, IO2, IO3, \
+ I0, I1, I2, I3) \
+ "precrq.ph.w %[" #O0 "], %[" #I0 "], %[" #IO0 "] \n\t" \
+ "precrq.ph.w %[" #O1 "], %[" #I1 "], %[" #IO1 "] \n\t" \
+ "ins %[" #IO0 "], %[" #I0 "], 16, 16 \n\t" \
+ "ins %[" #IO1 "], %[" #I1 "], 16, 16 \n\t" \
+ "precrq.ph.w %[" #O2 "], %[" #I2 "], %[" #IO2 "] \n\t" \
+ "precrq.ph.w %[" #O3 "], %[" #I3 "], %[" #IO3 "] \n\t" \
+ "ins %[" #IO2 "], %[" #I2 "], 16, 16 \n\t" \
+ "ins %[" #IO3 "], %[" #I3 "], 16, 16 \n\t"
+
+// preceu.ph.qbr temp0, temp8
+// temp0 = 0 | 0 | temp8[23..16] | temp8[7..0]
+// preceu.ph.qbl temp1, temp8
+// temp1 = temp8[23..16] | temp8[7..0] | 0 | 0
+// O - output
+// I - input (macro doesn't change it)
+#define CONVERT_2_BYTES_TO_HALF(O0, O1, O2, O3, O4, O5, O6, O7, \
+ I0, I1, I2, I3) \
+ "preceu.ph.qbr %[" #O0 "], %[" #I0 "] \n\t" \
+ "preceu.ph.qbl %[" #O1 "], %[" #I0 "] \n\t" \
+ "preceu.ph.qbr %[" #O2 "], %[" #I1 "] \n\t" \
+ "preceu.ph.qbl %[" #O3 "], %[" #I1 "] \n\t" \
+ "preceu.ph.qbr %[" #O4 "], %[" #I2 "] \n\t" \
+ "preceu.ph.qbl %[" #O5 "], %[" #I2 "] \n\t" \
+ "preceu.ph.qbr %[" #O6 "], %[" #I3 "] \n\t" \
+ "preceu.ph.qbl %[" #O7 "], %[" #I3 "] \n\t"
+
+// temp0[31..16 | 15..0] = temp0[31..16 | 15..0] + temp8[31..16 | 15..0]
+// temp0[31..16 | 15..0] = temp0[31..16 <<(s) 7 | 15..0 <<(s) 7]
+// temp1..temp7 same as temp0
+// precrqu_s.qb.ph temp0, temp1, temp0:
+// temp0 = temp1[31..24] | temp1[15..8] | temp0[31..24] | temp0[15..8]
+// store temp0 to dst
+// IO - input/output
+// I - input (macro doesn't change it)
+#define STORE_SAT_SUM_X2(IO0, IO1, IO2, IO3, IO4, IO5, IO6, IO7, \
+ I0, I1, I2, I3, I4, I5, I6, I7, \
+ I8, I9, I10, I11, I12, I13) \
+ "addq.ph %[" #IO0 "], %[" #IO0 "], %[" #I0 "] \n\t" \
+ "addq.ph %[" #IO1 "], %[" #IO1 "], %[" #I1 "] \n\t" \
+ "addq.ph %[" #IO2 "], %[" #IO2 "], %[" #I2 "] \n\t" \
+ "addq.ph %[" #IO3 "], %[" #IO3 "], %[" #I3 "] \n\t" \
+ "addq.ph %[" #IO4 "], %[" #IO4 "], %[" #I4 "] \n\t" \
+ "addq.ph %[" #IO5 "], %[" #IO5 "], %[" #I5 "] \n\t" \
+ "addq.ph %[" #IO6 "], %[" #IO6 "], %[" #I6 "] \n\t" \
+ "addq.ph %[" #IO7 "], %[" #IO7 "], %[" #I7 "] \n\t" \
+ "shll_s.ph %[" #IO0 "], %[" #IO0 "], 7 \n\t" \
+ "shll_s.ph %[" #IO1 "], %[" #IO1 "], 7 \n\t" \
+ "shll_s.ph %[" #IO2 "], %[" #IO2 "], 7 \n\t" \
+ "shll_s.ph %[" #IO3 "], %[" #IO3 "], 7 \n\t" \
+ "shll_s.ph %[" #IO4 "], %[" #IO4 "], 7 \n\t" \
+ "shll_s.ph %[" #IO5 "], %[" #IO5 "], 7 \n\t" \
+ "shll_s.ph %[" #IO6 "], %[" #IO6 "], 7 \n\t" \
+ "shll_s.ph %[" #IO7 "], %[" #IO7 "], 7 \n\t" \
+ "precrqu_s.qb.ph %[" #IO0 "], %[" #IO1 "], %[" #IO0 "] \n\t" \
+ "precrqu_s.qb.ph %[" #IO2 "], %[" #IO3 "], %[" #IO2 "] \n\t" \
+ "precrqu_s.qb.ph %[" #IO4 "], %[" #IO5 "], %[" #IO4 "] \n\t" \
+ "precrqu_s.qb.ph %[" #IO6 "], %[" #IO7 "], %[" #IO6 "] \n\t" \
+ "usw %[" #IO0 "], " XSTR(I13) "*" #I9 "(%[" #I8 "]) \n\t" \
+ "usw %[" #IO2 "], " XSTR(I13) "*" #I10 "(%[" #I8 "]) \n\t" \
+ "usw %[" #IO4 "], " XSTR(I13) "*" #I11 "(%[" #I8 "]) \n\t" \
+ "usw %[" #IO6 "], " XSTR(I13) "*" #I12 "(%[" #I8 "]) \n\t"
+
+#define OUTPUT_EARLY_CLOBBER_REGS_10() \
+ : [temp1]"=&r"(temp1), [temp2]"=&r"(temp2), [temp3]"=&r"(temp3), \
+ [temp4]"=&r"(temp4), [temp5]"=&r"(temp5), [temp6]"=&r"(temp6), \
+ [temp7]"=&r"(temp7), [temp8]"=&r"(temp8), [temp9]"=&r"(temp9), \
+ [temp10]"=&r"(temp10)
+
+#define OUTPUT_EARLY_CLOBBER_REGS_18() \
+ OUTPUT_EARLY_CLOBBER_REGS_10(), \
+ [temp11]"=&r"(temp11), [temp12]"=&r"(temp12), [temp13]"=&r"(temp13), \
+ [temp14]"=&r"(temp14), [temp15]"=&r"(temp15), [temp16]"=&r"(temp16), \
+ [temp17]"=&r"(temp17), [temp18]"=&r"(temp18)
+
+#endif // WEBP_DSP_MIPS_MACRO_H_
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-25 16:09:35 +0000