~ [ source navigation ] ~ [ diff markup ] ~ [ identifier search ] ~

TOMOYO Linux Cross Reference
Linux/arch/xtensa/lib/mulsi3.S

Version: ~ [ linux-6.12-rc7 ] ~ [ linux-6.11.7 ] ~ [ linux-6.10.14 ] ~ [ linux-6.9.12 ] ~ [ linux-6.8.12 ] ~ [ linux-6.7.12 ] ~ [ linux-6.6.60 ] ~ [ linux-6.5.13 ] ~ [ linux-6.4.16 ] ~ [ linux-6.3.13 ] ~ [ linux-6.2.16 ] ~ [ linux-6.1.116 ] ~ [ linux-6.0.19 ] ~ [ linux-5.19.17 ] ~ [ linux-5.18.19 ] ~ [ linux-5.17.15 ] ~ [ linux-5.16.20 ] ~ [ linux-5.15.171 ] ~ [ linux-5.14.21 ] ~ [ linux-5.13.19 ] ~ [ linux-5.12.19 ] ~ [ linux-5.11.22 ] ~ [ linux-5.10.229 ] ~ [ linux-5.9.16 ] ~ [ linux-5.8.18 ] ~ [ linux-5.7.19 ] ~ [ linux-5.6.19 ] ~ [ linux-5.5.19 ] ~ [ linux-5.4.285 ] ~ [ linux-5.3.18 ] ~ [ linux-5.2.21 ] ~ [ linux-5.1.21 ] ~ [ linux-5.0.21 ] ~ [ linux-4.20.17 ] ~ [ linux-4.19.323 ] ~ [ linux-4.18.20 ] ~ [ linux-4.17.19 ] ~ [ linux-4.16.18 ] ~ [ linux-4.15.18 ] ~ [ linux-4.14.336 ] ~ [ linux-4.13.16 ] ~ [ linux-4.12.14 ] ~ [ linux-4.11.12 ] ~ [ linux-4.10.17 ] ~ [ linux-4.9.337 ] ~ [ linux-4.4.302 ] ~ [ linux-3.10.108 ] ~ [ linux-2.6.32.71 ] ~ [ linux-2.6.0 ] ~ [ linux-2.4.37.11 ] ~ [ unix-v6-master ] ~ [ ccs-tools-1.8.12 ] ~ [ policy-sample ] ~
Architecture: ~ [ i386 ] ~ [ alpha ] ~ [ m68k ] ~ [ mips ] ~ [ ppc ] ~ [ sparc ] ~ [ sparc64 ] ~

  1 /* SPDX-License-Identifier: GPL-2.0-or-later WITH GCC-exception-2.0 */
  2 #include <linux/linkage.h>
  3 #include <asm/asmmacro.h>
  4 #include <asm/core.h>
  5 
  6         .macro  do_addx2 dst, as, at, tmp
  7 #if XCHAL_HAVE_ADDX
  8         addx2   \dst, \as, \at
  9 #else
 10         slli    \tmp, \as, 1
 11         add     \dst, \tmp, \at
 12 #endif
 13         .endm
 14 
 15         .macro  do_addx4 dst, as, at, tmp
 16 #if XCHAL_HAVE_ADDX
 17         addx4   \dst, \as, \at
 18 #else
 19         slli    \tmp, \as, 2
 20         add     \dst, \tmp, \at
 21 #endif
 22         .endm
 23 
 24         .macro  do_addx8 dst, as, at, tmp
 25 #if XCHAL_HAVE_ADDX
 26         addx8   \dst, \as, \at
 27 #else
 28         slli    \tmp, \as, 3
 29         add     \dst, \tmp, \at
 30 #endif
 31         .endm
 32 
 33 ENTRY(__mulsi3)
 34 
 35         abi_entry_default
 36 
 37 #if XCHAL_HAVE_MUL32
 38         mull    a2, a2, a3
 39 
 40 #elif XCHAL_HAVE_MUL16
 41         or      a4, a2, a3
 42         srai    a4, a4, 16
 43         bnez    a4, .LMUL16
 44         mul16u  a2, a2, a3
 45         abi_ret_default
 46 .LMUL16:
 47         srai    a4, a2, 16
 48         srai    a5, a3, 16
 49         mul16u  a7, a4, a3
 50         mul16u  a6, a5, a2
 51         mul16u  a4, a2, a3
 52         add     a7, a7, a6
 53         slli    a7, a7, 16
 54         add     a2, a7, a4
 55 
 56 #elif XCHAL_HAVE_MAC16
 57         mul.aa.hl a2, a3
 58         mula.aa.lh a2, a3
 59         rsr     a5, ACCLO
 60         umul.aa.ll a2, a3
 61         rsr     a4, ACCLO
 62         slli    a5, a5, 16
 63         add     a2, a4, a5
 64 
 65 #else /* !MUL32 && !MUL16 && !MAC16 */
 66 
 67         /* Multiply one bit at a time, but unroll the loop 4x to better
 68            exploit the addx instructions and avoid overhead.
 69            Peel the first iteration to save a cycle on init.  */
 70 
 71         /* Avoid negative numbers.  */
 72         xor     a5, a2, a3      /* Top bit is 1 if one input is negative.  */
 73         do_abs  a3, a3, a6
 74         do_abs  a2, a2, a6
 75 
 76         /* Swap so the second argument is smaller.  */
 77         sub     a7, a2, a3
 78         mov     a4, a3
 79         movgez  a4, a2, a7      /* a4 = max (a2, a3) */
 80         movltz  a3, a2, a7      /* a3 = min (a2, a3) */
 81 
 82         movi    a2, 0
 83         extui   a6, a3, 0, 1
 84         movnez  a2, a4, a6
 85 
 86         do_addx2 a7, a4, a2, a7
 87         extui   a6, a3, 1, 1
 88         movnez  a2, a7, a6
 89 
 90         do_addx4 a7, a4, a2, a7
 91         extui   a6, a3, 2, 1
 92         movnez  a2, a7, a6
 93 
 94         do_addx8 a7, a4, a2, a7
 95         extui   a6, a3, 3, 1
 96         movnez  a2, a7, a6
 97 
 98         bgeui   a3, 16, .Lmult_main_loop
 99         neg     a3, a2
100         movltz  a2, a3, a5
101         abi_ret_default
102 
103         .align  4
104 .Lmult_main_loop:
105         srli    a3, a3, 4
106         slli    a4, a4, 4
107 
108         add     a7, a4, a2
109         extui   a6, a3, 0, 1
110         movnez  a2, a7, a6
111 
112         do_addx2 a7, a4, a2, a7
113         extui   a6, a3, 1, 1
114         movnez  a2, a7, a6
115 
116         do_addx4 a7, a4, a2, a7
117         extui   a6, a3, 2, 1
118         movnez  a2, a7, a6
119 
120         do_addx8 a7, a4, a2, a7
121         extui   a6, a3, 3, 1
122         movnez  a2, a7, a6
123 
124         bgeui   a3, 16, .Lmult_main_loop
125 
126         neg     a3, a2
127         movltz  a2, a3, a5
128 
129 #endif /* !MUL32 && !MUL16 && !MAC16 */
130 
131         abi_ret_default
132 
133 ENDPROC(__mulsi3)
134 EXPORT_SYMBOL(__mulsi3)

~ [ source navigation ] ~ [ diff markup ] ~ [ identifier search ] ~

kernel.org | git.kernel.org | LWN.net | Project Home | SVN repository | Mail admin

Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.

sflogo.php