prettier

Makefile

@@ -3,8 +3,7 @@ SRCS  = $(filter %.c,$(FILES))
 BINS  = $(SRCS:%.c=%.elf)
 
 MUSLFILES = $(wildcard ./musl/*)
-MUSLSRCS  = $(filter %.s,$(MUSLFILES))
-MUSLOBJS  = $(MUSLSRCS:%.s=%.o)
+MUSLOBJS  = $(filter %.o,$(MUSLFILES))
 
 CFLAGS = -Wall -Werror -pedantic -Wno-unused -std=gnu2y\
 		 -O3 -c -g -fno-stack-protector -fno-pie\

day5.c

@@ -1,4 +1,5 @@
 #include "lib.h"
+#include <stdlib.h>
 
 #define DBG 0
 #define PRETTY 1
@@ -38,6 +39,15 @@ struct line {
 alignas(0x40) static
 struct line ranges[256];
 
+static
+int linecmp(const void *av, const void *bv) {
+    const struct line *a = av;
+    const struct line *b = bv;
+    if (a->sv < b->sv) return -1;
+    else if (a->sv > b->sv) return 1;
+    else return 0;
+}
+
 static
 unsigned long do_part1(size_t file_len, unsigned char file[file_len]) {
     unsigned char *s = file;
@@ -66,13 +76,13 @@ unsigned long do_part1(size_t file_len, unsigned char file[file_len]) {
         s++;
         d++;
     } while (*s != '\n');
-    d->sv = 0;
+
 #if PRETTY
-    d->pad0[0] = 'F';
-    d->pad0[1] = 'I';
-    d->pad0[2] = 'N';
-    d->pad0[3] = 'I';
+    size_t range_cnt = d - ranges;
+    qsort(ranges, range_cnt, sizeof(*ranges), linecmp);
 #endif
+
+    d->sv = 0;
     s++;
 
     do {
@@ -108,32 +118,32 @@ unsigned long do_part2(size_t file_len, unsigned char file[file_len]) {
         unsigned char *tmp;
         d->sv = grabbcd(s, &tmp);
 #if PRETTY
-        snprintd(d->sv, sizeof(d->s), d->s);
+        snprinth(d->sv, sizeof(d->s), d->s);
 #endif
         s = tmp;
         s++;
         d->ev = grabbcd(s, &tmp);
 #if PRETTY
-        snprintd(d->ev, sizeof(d->e), d->e);
+        snprinth(d->ev, sizeof(d->e), d->e);
 #endif
         s = tmp;
         s++;
         d++;
     } while (*s != '\n');
-    d->sv = 0;
+
 #if PRETTY
-    d->pad0[0] = 'F';
-    d->pad0[1] = 'I';
-    d->pad0[2] = 'N';
-    d->pad0[3] = 'I';
+    size_t range_cnt = d - ranges;
+    qsort(ranges, range_cnt, sizeof(*ranges), linecmp);
 #endif
 
+    d->sv = 0;
+
+    outer:
     for (struct line *di = ranges; di->sv; di++) {
         bcdint dis = di->sv;
         bcdint die = di->ev;
         if (dis == -1) continue;
-        for (struct line *dj = ranges; dj->sv; dj++) {
-            if (dj == di) continue;
+        for (struct line *dj = di+1; dj->sv; dj++) {
             bcdint djs = dj->sv;
             bcdint dje = dj->ev;
             if (djs == -1) continue;

lib.h

@@ -140,7 +140,6 @@ void snprinth(unsigned long v, size_t n, unsigned char d[n]) {
         v >>= 4;
         if (p == d) return;
     }
-    while (p >= d) *p-- = 0;
 }
 
 static

musl/atomic.h

@@ -0,0 +1,333 @@
+#ifndef _ATOMIC_H
+#define _ATOMIC_H
+
+#include <stdint.h>
+
+#include "atomic_arch.h"
+
+#ifdef a_ll
+
+#ifndef a_pre_llsc
+#define a_pre_llsc()
+#endif
+
+#ifndef a_post_llsc
+#define a_post_llsc()
+#endif
+
+#ifndef a_cas
+#define a_cas a_cas
+static inline int a_cas(volatile int *p, int t, int s)
+{
+	int old;
+	a_pre_llsc();
+	do old = a_ll(p);
+	while (old==t && !a_sc(p, s));
+	a_post_llsc();
+	return old;
+}
+#endif
+
+#ifndef a_swap
+#define a_swap a_swap
+static inline int a_swap(volatile int *p, int v)
+{
+	int old;
+	a_pre_llsc();
+	do old = a_ll(p);
+	while (!a_sc(p, v));
+	a_post_llsc();
+	return old;
+}
+#endif
+
+#ifndef a_fetch_add
+#define a_fetch_add a_fetch_add
+static inline int a_fetch_add(volatile int *p, int v)
+{
+	int old;
+	a_pre_llsc();
+	do old = a_ll(p);
+	while (!a_sc(p, (unsigned)old + v));
+	a_post_llsc();
+	return old;
+}
+#endif
+
+#ifndef a_fetch_and
+#define a_fetch_and a_fetch_and
+static inline int a_fetch_and(volatile int *p, int v)
+{
+	int old;
+	a_pre_llsc();
+	do old = a_ll(p);
+	while (!a_sc(p, old & v));
+	a_post_llsc();
+	return old;
+}
+#endif
+
+#ifndef a_fetch_or
+#define a_fetch_or a_fetch_or
+static inline int a_fetch_or(volatile int *p, int v)
+{
+	int old;
+	a_pre_llsc();
+	do old = a_ll(p);
+	while (!a_sc(p, old | v));
+	a_post_llsc();
+	return old;
+}
+#endif
+
+#endif
+
+#ifdef a_ll_p
+
+#ifndef a_cas_p
+#define a_cas_p a_cas_p
+static inline void *a_cas_p(volatile void *p, void *t, void *s)
+{
+	void *old;
+	a_pre_llsc();
+	do old = a_ll_p(p);
+	while (old==t && !a_sc_p(p, s));
+	a_post_llsc();
+	return old;
+}
+#endif
+
+#endif
+
+#ifndef a_cas
+#error missing definition of a_cas
+#endif
+
+#ifndef a_swap
+#define a_swap a_swap
+static inline int a_swap(volatile int *p, int v)
+{
+	int old;
+	do old = *p;
+	while (a_cas(p, old, v) != old);
+	return old;
+}
+#endif
+
+#ifndef a_fetch_add
+#define a_fetch_add a_fetch_add
+static inline int a_fetch_add(volatile int *p, int v)
+{
+	int old;
+	do old = *p;
+	while (a_cas(p, old, (unsigned)old+v) != old);
+	return old;
+}
+#endif
+
+#ifndef a_fetch_and
+#define a_fetch_and a_fetch_and
+static inline int a_fetch_and(volatile int *p, int v)
+{
+	int old;
+	do old = *p;
+	while (a_cas(p, old, old&v) != old);
+	return old;
+}
+#endif
+#ifndef a_fetch_or
+#define a_fetch_or a_fetch_or
+static inline int a_fetch_or(volatile int *p, int v)
+{
+	int old;
+	do old = *p;
+	while (a_cas(p, old, old|v) != old);
+	return old;
+}
+#endif
+
+#ifndef a_and
+#define a_and a_and
+static inline void a_and(volatile int *p, int v)
+{
+	a_fetch_and(p, v);
+}
+#endif
+
+#ifndef a_or
+#define a_or a_or
+static inline void a_or(volatile int *p, int v)
+{
+	a_fetch_or(p, v);
+}
+#endif
+
+#ifndef a_inc
+#define a_inc a_inc
+static inline void a_inc(volatile int *p)
+{
+	a_fetch_add(p, 1);
+}
+#endif
+
+#ifndef a_dec
+#define a_dec a_dec
+static inline void a_dec(volatile int *p)
+{
+	a_fetch_add(p, -1);
+}
+#endif
+
+#ifndef a_store
+#define a_store a_store
+static inline void a_store(volatile int *p, int v)
+{
+#ifdef a_barrier
+	a_barrier();
+	*p = v;
+	a_barrier();
+#else
+	a_swap(p, v);
+#endif
+}
+#endif
+
+#ifndef a_barrier
+#define a_barrier a_barrier
+static inline void a_barrier()
+{
+	volatile int tmp = 0;
+	a_cas(&tmp, 0, 0);
+}
+#endif
+
+#ifndef a_spin
+#define a_spin a_barrier
+#endif
+
+#ifndef a_and_64
+#define a_and_64 a_and_64
+static inline void a_and_64(volatile uint64_t *p, uint64_t v)
+{
+	union { uint64_t v; uint32_t r[2]; } u = { v };
+	if (u.r[0]+1) a_and((int *)p, u.r[0]);
+	if (u.r[1]+1) a_and((int *)p+1, u.r[1]);
+}
+#endif
+
+#ifndef a_or_64
+#define a_or_64 a_or_64
+static inline void a_or_64(volatile uint64_t *p, uint64_t v)
+{
+	union { uint64_t v; uint32_t r[2]; } u = { v };
+	if (u.r[0]) a_or((int *)p, u.r[0]);
+	if (u.r[1]) a_or((int *)p+1, u.r[1]);
+}
+#endif
+
+#ifndef a_cas_p
+typedef char a_cas_p_undefined_but_pointer_not_32bit[-sizeof(char) == 0xffffffff ? 1 : -1];
+#define a_cas_p a_cas_p
+static inline void *a_cas_p(volatile void *p, void *t, void *s)
+{
+	return (void *)a_cas((volatile int *)p, (int)t, (int)s);
+}
+#endif
+
+#ifndef a_or_l
+#define a_or_l a_or_l
+static inline void a_or_l(volatile void *p, long v)
+{
+	if (sizeof(long) == sizeof(int)) a_or(p, v);
+	else a_or_64(p, v);
+}
+#endif
+
+#ifndef a_crash
+#define a_crash a_crash
+static inline void a_crash()
+{
+	*(volatile char *)0=0;
+}
+#endif
+
+#ifndef a_ctz_32
+#define a_ctz_32 a_ctz_32
+static inline int a_ctz_32(uint32_t x)
+{
+#ifdef a_clz_32
+	return 31-a_clz_32(x&-x);
+#else
+	static const char debruijn32[32] = {
+		0, 1, 23, 2, 29, 24, 19, 3, 30, 27, 25, 11, 20, 8, 4, 13,
+		31, 22, 28, 18, 26, 10, 7, 12, 21, 17, 9, 6, 16, 5, 15, 14
+	};
+	return debruijn32[(x&-x)*0x076be629 >> 27];
+#endif
+}
+#endif
+
+#ifndef a_ctz_64
+#define a_ctz_64 a_ctz_64
+static inline int a_ctz_64(uint64_t x)
+{
+	static const char debruijn64[64] = {
+		0, 1, 2, 53, 3, 7, 54, 27, 4, 38, 41, 8, 34, 55, 48, 28,
+		62, 5, 39, 46, 44, 42, 22, 9, 24, 35, 59, 56, 49, 18, 29, 11,
+		63, 52, 6, 26, 37, 40, 33, 47, 61, 45, 43, 21, 23, 58, 17, 10,
+		51, 25, 36, 32, 60, 20, 57, 16, 50, 31, 19, 15, 30, 14, 13, 12
+	};
+	if (sizeof(long) < 8) {
+		uint32_t y = x;
+		if (!y) {
+			y = x>>32;
+			return 32 + a_ctz_32(y);
+		}
+		return a_ctz_32(y);
+	}
+	return debruijn64[(x&-x)*0x022fdd63cc95386dull >> 58];
+}
+#endif
+
+static inline int a_ctz_l(unsigned long x)
+{
+	return (sizeof(long) < 8) ? a_ctz_32(x) : a_ctz_64(x);
+}
+
+#ifndef a_clz_64
+#define a_clz_64 a_clz_64
+static inline int a_clz_64(uint64_t x)
+{
+#ifdef a_clz_32
+	if (x>>32)
+		return a_clz_32(x>>32);
+	return a_clz_32(x) + 32;
+#else
+	uint32_t y;
+	int r;
+	if (x>>32) y=x>>32, r=0; else y=x, r=32;
+	if (y>>16) y>>=16; else r |= 16;
+	if (y>>8) y>>=8; else r |= 8;
+	if (y>>4) y>>=4; else r |= 4;
+	if (y>>2) y>>=2; else r |= 2;
+	return r | !(y>>1);
+#endif
+}
+#endif
+
+#ifndef a_clz_32
+#define a_clz_32 a_clz_32
+static inline int a_clz_32(uint32_t x)
+{
+	x >>= 1;
+	x |= x >> 1;
+	x |= x >> 2;
+	x |= x >> 4;
+	x |= x >> 8;
+	x |= x >> 16;
+	x++;
+	return 31-a_ctz_32(x);
+}
+#endif
+
+#endif

musl/atomic_arch.h

@@ -0,0 +1,123 @@
+#define a_cas a_cas
+static inline int a_cas(volatile int *p, int t, int s)
+{
+	__asm__ __volatile__ (
+		"lock ; cmpxchg %3, %1"
+		: "=a"(t), "=m"(*p) : "a"(t), "r"(s) : "memory" );
+	return t;
+}
+
+#define a_cas_p a_cas_p
+static inline void *a_cas_p(volatile void *p, void *t, void *s)
+{
+	__asm__( "lock ; cmpxchg %3, %1"
+		: "=a"(t), "=m"(*(void *volatile *)p)
+		: "a"(t), "r"(s) : "memory" );
+	return t;
+}
+
+#define a_swap a_swap
+static inline int a_swap(volatile int *p, int v)
+{
+	__asm__ __volatile__(
+		"xchg %0, %1"
+		: "=r"(v), "=m"(*p) : "0"(v) : "memory" );
+	return v;
+}
+
+#define a_fetch_add a_fetch_add
+static inline int a_fetch_add(volatile int *p, int v)
+{
+	__asm__ __volatile__(
+		"lock ; xadd %0, %1"
+		: "=r"(v), "=m"(*p) : "0"(v) : "memory" );
+	return v;
+}
+
+#define a_and a_and
+static inline void a_and(volatile int *p, int v)
+{
+	__asm__ __volatile__(
+		"lock ; and %1, %0"
+		: "=m"(*p) : "r"(v) : "memory" );
+}
+
+#define a_or a_or
+static inline void a_or(volatile int *p, int v)
+{
+	__asm__ __volatile__(
+		"lock ; or %1, %0"
+		: "=m"(*p) : "r"(v) : "memory" );
+}
+
+#define a_and_64 a_and_64
+static inline void a_and_64(volatile uint64_t *p, uint64_t v)
+{
+	__asm__ __volatile(
+		"lock ; and %1, %0"
+		 : "=m"(*p) : "r"(v) : "memory" );
+}
+
+#define a_or_64 a_or_64
+static inline void a_or_64(volatile uint64_t *p, uint64_t v)
+{
+	__asm__ __volatile__(
+		"lock ; or %1, %0"
+		 : "=m"(*p) : "r"(v) : "memory" );
+}
+
+#define a_inc a_inc
+static inline void a_inc(volatile int *p)
+{
+	__asm__ __volatile__(
+		"lock ; incl %0"
+		: "=m"(*p) : "m"(*p) : "memory" );
+}
+
+#define a_dec a_dec
+static inline void a_dec(volatile int *p)
+{
+	__asm__ __volatile__(
+		"lock ; decl %0"
+		: "=m"(*p) : "m"(*p) : "memory" );
+}
+
+#define a_store a_store
+static inline void a_store(volatile int *p, int x)
+{
+	__asm__ __volatile__(
+		"mov %1, %0 ; lock ; orl $0,(%%rsp)"
+		: "=m"(*p) : "r"(x) : "memory" );
+}
+
+#define a_barrier a_barrier
+static inline void a_barrier()
+{
+	__asm__ __volatile__( "" : : : "memory" );
+}
+
+#define a_spin a_spin
+static inline void a_spin()
+{
+	__asm__ __volatile__( "pause" : : : "memory" );
+}
+
+#define a_crash a_crash
+static inline void a_crash()
+{
+	__asm__ __volatile__( "hlt" : : : "memory" );
+}
+
+#define a_ctz_64 a_ctz_64
+static inline int a_ctz_64(uint64_t x)
+{
+	__asm__( "bsf %1,%0" : "=r"(x) : "r"(x) );
+	return x;
+}
+
+#define a_clz_64 a_clz_64
+static inline int a_clz_64(uint64_t x)
+{
+	__asm__( "bsr %1,%0 ; xor $63,%0" : "=r"(x) : "r"(x) );
+	return x;
+}

musl/qsort.c

@@ -0,0 +1,219 @@
+/* Copyright (C) 2011 by Lynn Ochs
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+/* Minor changes by Rich Felker for integration in musl, 2011-04-27. */
+
+/* Smoothsort, an adaptive variant of Heapsort.  Memory usage: O(1).
+   Run time: Worst case O(n log n), close to O(n) in the mostly-sorted case. */
+
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "atomic.h"
+#define ntz(x) a_ctz_l((x))
+
+typedef int (*cmpfun)(const void *, const void *, void *);
+
+static inline int pntz(size_t p[2]) {
+	int r = ntz(p[0] - 1);
+	if(r != 0 || (r = 8*sizeof(size_t) + ntz(p[1])) != 8*sizeof(size_t)) {
+		return r;
+	}
+	return 0;
+}
+
+static void cycle(size_t width, unsigned char* ar[], int n)
+{
+	unsigned char tmp[256];
+	size_t l;
+	int i;
+
+	if(n < 2) {
+		return;
+	}
+
+	ar[n] = tmp;
+	while(width) {
+		l = sizeof(tmp) < width ? sizeof(tmp) : width;
+		memcpy(ar[n], ar[0], l);
+		for(i = 0; i < n; i++) {
+			memcpy(ar[i], ar[i + 1], l);
+			ar[i] += l;
+		}
+		width -= l;
+	}
+    ar[n] = 0;
+}
+
+/* shl() and shr() need n > 0 */
+static inline void shl(size_t p[2], int n)
+{
+	if(n >= 8 * sizeof(size_t)) {
+		n -= 8 * sizeof(size_t);
+		p[1] = p[0];
+		p[0] = 0;
+	}
+	p[1] <<= n;
+	p[1] |= p[0] >> (sizeof(size_t) * 8 - n);
+	p[0] <<= n;
+}
+
+static inline void shr(size_t p[2], int n)
+{
+	if(n >= 8 * sizeof(size_t)) {
+		n -= 8 * sizeof(size_t);
+		p[0] = p[1];
+		p[1] = 0;
+	}
+	p[0] >>= n;
+	p[0] |= p[1] << (sizeof(size_t) * 8 - n);
+	p[1] >>= n;
+}
+
+static void sift(unsigned char *head, size_t width, cmpfun cmp, void *arg, int pshift, size_t lp[])
+{
+	unsigned char *rt, *lf;
+	unsigned char *ar[14 * sizeof(size_t) + 1];
+	int i = 1;
+
+	ar[0] = head;
+	while(pshift > 1) {
+		rt = head - width;
+		lf = head - width - lp[pshift - 2];
+
+		if(cmp(ar[0], lf, arg) >= 0 && cmp(ar[0], rt, arg) >= 0) {
+			break;
+		}
+		if(cmp(lf, rt, arg) >= 0) {
+			ar[i++] = lf;
+			head = lf;
+			pshift -= 1;
+		} else {
+			ar[i++] = rt;
+			head = rt;
+			pshift -= 2;
+		}
+	}
+	cycle(width, ar, i);
+}
+
+static void trinkle(unsigned char *head, size_t width, cmpfun cmp, void *arg, size_t pp[2], int pshift, int trusty, size_t lp[])
+{
+	unsigned char *stepson,
+	              *rt, *lf;
+	size_t p[2];
+	unsigned char *ar[14 * sizeof(size_t) + 1];
+	int i = 1;
+	int trail;
+
+	p[0] = pp[0];
+	p[1] = pp[1];
+
+	ar[0] = head;
+	while(p[0] != 1 || p[1] != 0) {
+		stepson = head - lp[pshift];
+		if(cmp(stepson, ar[0], arg) <= 0) {
+			break;
+		}
+		if(!trusty && pshift > 1) {
+			rt = head - width;
+			lf = head - width - lp[pshift - 2];
+			if(cmp(rt, stepson, arg) >= 0 || cmp(lf, stepson, arg) >= 0) {
+				break;
+			}
+		}
+
+		ar[i++] = stepson;
+		head = stepson;
+		trail = pntz(p);
+		shr(p, trail);
+		pshift += trail;
+		trusty = 0;
+	}
+	if(!trusty) {
+		cycle(width, ar, i);
+		sift(head, width, cmp, arg, pshift, lp);
+	}
+}
+
+void qsort_r(void *base, size_t nel, size_t width, cmpfun cmp, void *arg)
+{
+	size_t lp[12*sizeof(size_t)];
+	size_t i, size = width * nel;
+	unsigned char *head, *high;
+	size_t p[2] = {1, 0};
+	int pshift = 1;
+	int trail;
+
+	if (!size) return;
+
+	head = base;
+	high = head + size - width;
+
+	/* Precompute Leonardo numbers, scaled by element width */
+	for(lp[0]=lp[1]=width, i=2; (lp[i]=lp[i-2]+lp[i-1]+width) < size; i++);
+
+	while(head < high) {
+		if((p[0] & 3) == 3) {
+			sift(head, width, cmp, arg, pshift, lp);
+			shr(p, 2);
+			pshift += 2;
+		} else {
+			if(lp[pshift - 1] >= high - head) {
+				trinkle(head, width, cmp, arg, p, pshift, 0, lp);
+			} else {
+				sift(head, width, cmp, arg, pshift, lp);
+			}
+
+			if(pshift == 1) {
+				shl(p, 1);
+				pshift = 0;
+			} else {
+				shl(p, pshift - 1);
+				pshift = 1;
+			}
+		}
+
+		p[0] |= 1;
+		head += width;
+	}
+
+	trinkle(head, width, cmp, arg, p, pshift, 0, lp);
+
+	while(pshift != 1 || p[0] != 1 || p[1] != 0) {
+		if(pshift <= 1) {
+			trail = pntz(p);
+			shr(p, trail);
+			pshift += trail;
+		} else {
+			shl(p, 2);
+			pshift -= 2;
+			p[0] ^= 7;
+			shr(p, 1);
+			trinkle(head - lp[pshift] - width, width, cmp, arg, p, pshift + 1, 1, lp);
+			shl(p, 1);
+			p[0] |= 1;
+			trinkle(head - width, width, cmp, arg, p, pshift, 1, lp);
+		}
+		head -= width;
+	}
+}

musl/qsort_nr.c

@@ -0,0 +1,16 @@
+#include <stdlib.h>
+
+typedef int (*cmpfun)(const void *, const void *);
+typedef int (*cmpfun_r)(const void *, const void *, cmpfun);
+void qsort_r(void *base, size_t nel, size_t width, cmpfun_r cmp, cmpfun arg);
+
+
+static int wrapper_cmp(const void *v1, const void *v2, cmpfun cmp)
+{
+	return cmp(v1, v2);
+}
+
+void qsort(void *base, size_t nel, size_t width, cmpfun cmp)
+{
+	qsort_r(base, nel, width, wrapper_cmp, cmp);
+}