mirror of
https://github.com/jart/cosmopolitan.git
synced 2025-01-31 11:37:35 +00:00
7e0a09feec
This change ports APE Loader to Linux AARCH64, so that Raspberry Pi users can run programs like redbean, without the executable needing to modify itself. Progress has also slipped into this change on the issue of making progress better conforming to user expectations and industry standards regarding which symbols we're allowed to declare
44 lines
1.6 KiB
C
44 lines
1.6 KiB
C
/* clang-format off */
|
|
/* ===-- popcountdi2.c - Implement __popcountdi2 ----------------------------===
|
|
*
|
|
* The LLVM Compiler Infrastructure
|
|
*
|
|
* This file is dual licensed under the MIT and the University of Illinois Open
|
|
* Source Licenses. See LICENSE.TXT for details.
|
|
*
|
|
* ===----------------------------------------------------------------------===
|
|
*
|
|
* This file implements __popcountdi2 for the compiler_rt library.
|
|
*
|
|
* ===----------------------------------------------------------------------===
|
|
*/
|
|
|
|
__static_yoink("huge_compiler_rt_license");
|
|
|
|
#include "third_party/compiler_rt/int_lib.h"
|
|
|
|
/* Returns: count of 1 bits */
|
|
|
|
COMPILER_RT_ABI si_int
|
|
__popcountdi2(di_int a)
|
|
{
|
|
#ifdef __POPCNT__
|
|
asm("popcnt\t%1,%0" : "=r"(a) : "r"(a) : "cc");
|
|
return a;
|
|
#else
|
|
du_int x2 = (du_int)a;
|
|
x2 = x2 - ((x2 >> 1) & 0x5555555555555555uLL);
|
|
/* Every 2 bits holds the sum of every pair of bits (32) */
|
|
x2 = ((x2 >> 2) & 0x3333333333333333uLL) + (x2 & 0x3333333333333333uLL);
|
|
/* Every 4 bits holds the sum of every 4-set of bits (3 significant bits) (16) */
|
|
x2 = (x2 + (x2 >> 4)) & 0x0F0F0F0F0F0F0F0FuLL;
|
|
/* Every 8 bits holds the sum of every 8-set of bits (4 significant bits) (8) */
|
|
su_int x = (su_int)(x2 + (x2 >> 32));
|
|
/* The lower 32 bits hold four 16 bit sums (5 significant bits). */
|
|
/* Upper 32 bits are garbage */
|
|
x = x + (x >> 16);
|
|
/* The lower 16 bits hold two 32 bit sums (6 significant bits). */
|
|
/* Upper 16 bits are garbage */
|
|
return (x + (x >> 8)) & 0x0000007F; /* (7 significant bits) */
|
|
#endif
|
|
}
|