#include "lib/lib.h"
#include "lib/fastbuf.h"
#include "lib/mempool.h"
+#include "lib/stkstring.h"
#include "lib/sorter/common.h"
-#include <fcntl.h>
+#include <string.h>
+#include <sys/time.h>
+#include <time.h>
-void *
-sorter_alloc(struct sort_context *ctx, uns size)
-{
- return mp_alloc_zero(ctx->pool, size);
-}
-
-struct sort_bucket *
-sbuck_new(struct sort_context *ctx)
-{
- struct sort_bucket *b = sorter_alloc(ctx, sizeof(struct sort_bucket));
- b->ctx = ctx;
- return b;
-}
-
-void
-sbuck_drop(struct sort_bucket *b)
-{
- if (b)
- {
- ASSERT(!(b->flags & SBF_DESTROYED));
- if (b->n.prev)
- clist_remove(&b->n);
- bclose(b->fb);
- bzero(b, sizeof(*b));
- b->flags = SBF_DESTROYED;
- }
-}
-
-sh_off_t
-sbuck_size(struct sort_bucket *b)
-{
- if ((b->flags & SBF_OPEN_WRITE) && !(b->flags & SBF_SWAPPED_OUT))
- return btell(b->fb);
- else
- return b->size;
-}
-
-int
-sbuck_have(struct sort_bucket *b)
-{
- return b && sbuck_size(b);
-}
-
-static int
-sbuck_has_file(struct sort_bucket *b)
-{
- return (b->fb || (b->flags & SBF_SWAPPED_OUT));
-}
+#define F_BSIZE(b) stk_fsize(sbuck_size(b))
static void
-sbuck_swap_in(struct sort_bucket *b)
-{
- if (b->flags & SBF_SWAPPED_OUT)
- {
- b->fb = bopen(b->filename, O_RDWR, sorter_stream_bufsize);
- if (b->flags & SBF_OPEN_WRITE)
- bseek(b->fb, 0, SEEK_END);
- bconfig(b->fb, BCONFIG_IS_TEMP_FILE, 1);
- b->flags &= ~SBF_SWAPPED_OUT;
- SORT_XTRACE("Swapped in %s", b->filename);
- }
-}
-
-struct fastbuf *
-sbuck_read(struct sort_bucket *b)
-{
- sbuck_swap_in(b);
- if (b->flags & SBF_OPEN_READ)
- return b->fb;
- else if (b->flags & SBF_OPEN_WRITE)
- {
- b->size = btell(b->fb);
- b->flags = (b->flags & ~SBF_OPEN_WRITE) | SBF_OPEN_READ;
- brewind(b->fb);
- return b->fb;
- }
- else
- ASSERT(0);
-}
-
-struct fastbuf *
-sbuck_write(struct sort_bucket *b)
-{
- sbuck_swap_in(b);
- if (b->flags & SBF_OPEN_WRITE)
- ASSERT(b->fb);
- else
- {
- ASSERT(!(b->flags & (SBF_OPEN_READ | SBF_DESTROYED)));
- b->fb = bopen_tmp(sorter_stream_bufsize);
- b->flags |= SBF_OPEN_WRITE;
- b->filename = mp_strdup(b->ctx->pool, b->fb->name);
- }
- return b->fb;
-}
-
-void
-sbuck_swap_out(struct sort_bucket *b)
+sorter_start_timer(struct sort_context *ctx)
{
- if ((b->flags & (SBF_OPEN_READ | SBF_OPEN_WRITE)) && b->fb)
- {
- if (b->flags & SBF_OPEN_WRITE)
- b->size = btell(b->fb);
- bconfig(b->fb, BCONFIG_IS_TEMP_FILE, 0);
- bclose(b->fb);
- b->fb = NULL;
- b->flags |= SBF_SWAPPED_OUT;
- SORT_XTRACE("Swapped out %s", b->filename);
- }
+ init_timer(&ctx->start_time);
}
-void
-sorter_alloc_buf(struct sort_context *ctx)
+static void
+sorter_stop_timer(struct sort_context *ctx, uns *account_to)
{
- if (ctx->big_buf)
- return;
- u64 bs = MAX(sorter_bufsize/2, 1);
- bs = ALIGN_TO(bs, (u64)CPU_PAGE_SIZE);
- ctx->big_buf = big_alloc(2*bs);
- ctx->big_buf_size = 2*bs;
- ctx->big_buf_half = ((byte*) ctx->big_buf) + bs;
- ctx->big_buf_half_size = bs;
- SORT_XTRACE("Allocated sorting buffer (%jd bytes)", (uintmax_t) bs);
+ ctx->last_pass_time = get_timer(&ctx->start_time);
+ *account_to += ctx->last_pass_time;
}
-void
-sorter_free_buf(struct sort_context *ctx)
+static uns
+sorter_speed(struct sort_context *ctx, u64 size)
{
- if (!ctx->big_buf)
- return;
- big_free(ctx->big_buf, ctx->big_buf_size);
- ctx->big_buf = NULL;
- SORT_XTRACE("Freed sorting buffer");
+ if (!size)
+ return 0;
+ if (!ctx->last_pass_time)
+ return 0;
+ return (uns)((double)size / (1<<20) * 1000 / ctx->last_pass_time);
}
-static int sorter_presort(struct sort_context *ctx, struct sort_bucket *in, struct sort_bucket *out, struct sort_bucket *out_only)
+static int
+sorter_presort(struct sort_context *ctx, struct sort_bucket *in, struct sort_bucket *out, struct sort_bucket *out_only)
{
sorter_alloc_buf(ctx);
if (in->flags & SBF_CUSTOM_PRESORT)
{
+ /*
+ * The trick with automatic joining, which we use for the normal presorter,
+ * is not necessary with the custom presorter, because the custom presorter
+ * is never called in the middle of the sorted data.
+ */
struct fastbuf *f = sbuck_write(out);
- return ctx->custom_presort(f, ctx->big_buf, ctx->big_buf_size); // FIXME: out_only optimization?
+ out->runs++;
+ return ctx->custom_presort(f, ctx->big_buf, ctx->big_buf_size);
}
return ctx->internal_sort(ctx, in, out, out_only);
}
-static inline struct sort_bucket *
-sbuck_join_to(struct sort_bucket *b)
+static struct sort_bucket *
+sbuck_join_to(struct sort_bucket *b, sh_off_t *sizep)
{
if (sorter_debug & SORT_DEBUG_NO_JOIN)
return NULL;
struct sort_bucket *out = (struct sort_bucket *) b->n.prev; // Such bucket is guaranteed to exist
- return (out->flags & SBF_FINAL) ? out : NULL;
+ if (!(out->flags & SBF_FINAL))
+ return NULL;
+ ASSERT(out->runs == 1);
+ *sizep = sbuck_size(out);
+ return out;
+}
+
+static sh_off_t
+sbuck_ins_or_join(struct sort_bucket *b, cnode *list_pos, struct sort_bucket *join, sh_off_t join_size)
+{
+ if (join && join->runs >= 2)
+ {
+ if (b)
+ sbuck_drop(b);
+ ASSERT(join->runs == 2);
+ join->runs--;
+ return sbuck_size(join) - join_size;
+ }
+ else if (b)
+ {
+ clist_insert_after(&b->n, list_pos);
+ return sbuck_size(b);
+ }
+ else
+ return 0;
}
static void
{
// The final bucket doesn't have any file associated yet, so replace
// it with the new bucket.
- SORT_XTRACE("Replaced final bucket");
+ SORT_XTRACE(3, "Replaced final bucket");
b->flags |= SBF_FINAL;
sbuck_drop(join);
}
else
{
- SORT_TRACE("Copying %jd bytes to output file", (uintmax_t) sbuck_size(b));
+ SORT_TRACE("Copying to output file: %s", F_BSIZE(b));
struct fastbuf *src = sbuck_read(b);
struct fastbuf *dest = sbuck_write(join);
bbcopy(src, dest, ~0U);
{
struct sort_bucket *ins[3] = { NULL }, *outs[3] = { NULL };
cnode *list_pos = b->n.prev;
- struct sort_bucket *join = sbuck_join_to(b);
+ sh_off_t join_size;
+ struct sort_bucket *join = sbuck_join_to(b, &join_size);
if (!(sorter_debug & SORT_DEBUG_NO_PRESORT) || (b->flags & SBF_CUSTOM_PRESORT))
{
- SORT_TRACE("Presorting");
+ SORT_XTRACE(3, "%s", ((b->flags & SBF_CUSTOM_PRESORT) ? "Custom presorting" : "Presorting"));
+ sorter_start_timer(ctx);
ins[0] = sbuck_new(ctx);
if (!sorter_presort(ctx, b, ins[0], join ? : ins[0]))
{
- SORT_XTRACE("Sorted in memory");
- if (join)
- sbuck_drop(ins[0]);
- else
- clist_insert_after(&ins[0]->n, list_pos);
+ sorter_stop_timer(ctx, &ctx->total_pre_time);
+ sh_off_t size = sbuck_ins_or_join(ins[0], list_pos, join, join_size);
+ SORT_XTRACE(((b->flags & SBF_SOURCE) ? 1 : 3), "Sorted in memory (%s, %dMB/s)", stk_fsize(size), sorter_speed(ctx, size));
sbuck_drop(b);
return;
}
while (sorter_presort(ctx, b, ins[i], ins[i]))
i = 1-i;
sbuck_drop(b);
+ sorter_stop_timer(ctx, &ctx->total_pre_time);
+ SORT_TRACE("Presorting pass (%d+%d runs, %s+%s, %dMB/s)",
+ ins[0]->runs, ins[1]->runs,
+ F_BSIZE(ins[0]), F_BSIZE(ins[1]),
+ sorter_speed(ctx, sbuck_size(ins[0]) + sbuck_size(ins[1])));
}
else
{
- SORT_TRACE("Skipped presorting");
+ SORT_XTRACE(2, "Presorting disabled");
ins[0] = b;
}
- SORT_TRACE("Main sorting");
+ SORT_XTRACE(3, "Main sorting");
+ uns pass = 0;
do {
- if (ins[0]->runs == 1 && ins[1]->runs == 1 && join)
+ ++pass;
+ sorter_start_timer(ctx);
+ if (ins[0]->runs <= 1 && ins[1]->runs <= 1 && join)
{
// This is guaranteed to produce a single run, so join if possible
outs[0] = join;
outs[1] = NULL;
ctx->twoway_merge(ctx, ins, outs);
- ASSERT(outs[0]->runs == 2);
- outs[0]->runs--;
- SORT_TRACE("Pass done (joined final run)");
+ sh_off_t size = sbuck_ins_or_join(NULL, NULL, join, join_size);
+ sorter_stop_timer(ctx, &ctx->total_ext_time);
+ SORT_TRACE("Mergesort pass %d (final run, %s, %dMB/s)", pass, stk_fsize(size), sorter_speed(ctx, size));
sbuck_drop(ins[0]);
sbuck_drop(ins[1]);
return;
outs[1] = sbuck_new(ctx);
outs[2] = NULL;
ctx->twoway_merge(ctx, ins, outs);
- SORT_TRACE("Pass done (%d+%d runs, %jd+%jd bytes)", outs[0]->runs, outs[1]->runs, (uintmax_t) sbuck_size(outs[0]), (uintmax_t) sbuck_size(outs[1]));
+ sorter_stop_timer(ctx, &ctx->total_ext_time);
+ SORT_TRACE("Mergesort pass %d (%d+%d runs, %s+%s, %dMB/s)", pass,
+ outs[0]->runs, outs[1]->runs,
+ F_BSIZE(outs[0]), F_BSIZE(outs[1]),
+ sorter_speed(ctx, sbuck_size(outs[0]) + sbuck_size(outs[1])));
sbuck_drop(ins[0]);
sbuck_drop(ins[1]);
memcpy(ins, outs, 3*sizeof(struct sort_bucket *));
clist_insert_after(&ins[0]->n, list_pos);
}
+static void
+sorter_multiway(struct sort_context *ctx, struct sort_bucket *b)
+{
+ clist parts;
+ cnode *list_pos = b->n.prev;
+ sh_off_t join_size;
+ struct sort_bucket *join = sbuck_join_to(b, &join_size);
+ uns trace_level = (b->flags & SBF_SOURCE) ? 1 : 3;
+
+ clist_init(&parts);
+ ASSERT(!(sorter_debug & SORT_DEBUG_NO_PRESORT));
+ SORT_XTRACE(3, "%s", ((b->flags & SBF_CUSTOM_PRESORT) ? "Custom presorting" : "Presorting"));
+ uns cont;
+ uns part_cnt = 0;
+ u64 total_size = 0;
+ sorter_start_timer(ctx);
+ do
+ {
+ struct sort_bucket *p = sbuck_new(ctx);
+ cont = sorter_presort(ctx, b, p, (!part_cnt && join) ? join : p);
+ if (sbuck_have(p))
+ {
+ part_cnt++;
+ clist_add_tail(&parts, &p->n);
+ total_size += sbuck_size(p);
+ sbuck_swap_out(p);
+ }
+ else
+ sbuck_drop(p);
+ }
+ while (cont);
+ sorter_stop_timer(ctx, &ctx->total_pre_time);
+ sorter_free_buf(ctx);
+ sbuck_drop(b);
+
+ if (part_cnt <= 1)
+ {
+ sh_off_t size = sbuck_ins_or_join(clist_head(&parts), list_pos, (part_cnt ? NULL : join), join_size);
+ SORT_XTRACE(trace_level, "Sorted in memory (%s, %dMB/s)", stk_fsize(size), sorter_speed(ctx, size));
+ return;
+ }
+
+ SORT_TRACE("Multi-way presorting pass (%d parts, %s, %dMB/s)", part_cnt, stk_fsize(total_size), sorter_speed(ctx, total_size));
+
+ uns max_ways = 1 << sorter_max_multiway_bits;
+ struct sort_bucket *ways[max_ways+1];
+ SORT_XTRACE(3, "Starting up to %d-way merge", max_ways);
+ for (;;)
+ {
+ uns n = 0;
+ struct sort_bucket *p;
+ while (n < max_ways && (p = clist_head(&parts)))
+ {
+ clist_remove(&p->n);
+ ways[n++] = p;
+ }
+ ways[n] = NULL;
+ ASSERT(n > 1);
+
+ struct sort_bucket *out;
+ if (clist_empty(&parts) && join)
+ out = join;
+ else
+ out = sbuck_new(ctx);
+ sorter_start_timer(ctx);
+ ctx->multiway_merge(ctx, ways, out);
+ sorter_stop_timer(ctx, &ctx->total_ext_time);
+
+ for (uns i=0; i<n; i++)
+ sbuck_drop(ways[i]);
+
+ if (clist_empty(&parts))
+ {
+ sh_off_t size = sbuck_ins_or_join((join ? NULL : out), list_pos, join, join_size);
+ SORT_TRACE("Multi-way merge completed (%d ways, %s, %dMB/s)", n, stk_fsize(size), sorter_speed(ctx, size));
+ return;
+ }
+ else
+ {
+ sbuck_swap_out(out);
+ clist_add_tail(&parts, &out->n);
+ SORT_TRACE("Multi-way merge pass (%d ways, %s, %dMB/s)", n, F_BSIZE(out), sorter_speed(ctx, sbuck_size(out)));
+ }
+ }
+}
+
+static void
+sorter_radix(struct sort_context *ctx, struct sort_bucket *b, uns bits)
+{
+ // Add more bits if requested and allowed.
+ bits = MIN(bits + sorter_add_radix_bits, sorter_max_radix_bits);
+
+ uns nbuck = 1 << bits;
+ SORT_XTRACE(3, "Running radix split on %s with hash %d bits of %d (expecting %s buckets)",
+ F_BSIZE(b), bits, b->hash_bits, stk_fsize(sbuck_size(b) / nbuck));
+ sorter_free_buf(ctx);
+ sorter_start_timer(ctx);
+
+ struct sort_bucket **outs = alloca(nbuck * sizeof(struct sort_bucket *));
+ for (uns i=nbuck; i--; )
+ {
+ outs[i] = sbuck_new(ctx);
+ outs[i]->hash_bits = b->hash_bits - bits;
+ clist_insert_after(&outs[i]->n, &b->n);
+ }
+
+ ctx->radix_split(ctx, b, outs, b->hash_bits - bits, bits);
+
+ u64 min = ~(u64)0, max = 0, sum = 0;
+ for (uns i=0; i<nbuck; i++)
+ {
+ u64 s = sbuck_size(outs[i]);
+ min = MIN(min, s);
+ max = MAX(max, s);
+ sum += s;
+ if (nbuck > 4)
+ sbuck_swap_out(outs[i]);
+ }
+
+ sorter_stop_timer(ctx, &ctx->total_ext_time);
+ SORT_TRACE("Radix split (%d buckets, %s min, %s max, %s avg, %dMB/s)", nbuck,
+ stk_fsize(min), stk_fsize(max), stk_fsize(sum / nbuck), sorter_speed(ctx, sum));
+ sbuck_drop(b);
+}
+
+static void
+sorter_decide(struct sort_context *ctx, struct sort_bucket *b)
+{
+ // Drop empty buckets
+ if (!sbuck_have(b))
+ {
+ SORT_XTRACE(4, "Dropping empty bucket");
+ sbuck_drop(b);
+ return;
+ }
+
+ // How many bits of bucket size we have to reduce before it fits in the RAM?
+ // (this is insanely large if the input size is unknown, but it serves our purpose)
+ u64 insize = sbuck_size(b);
+ u64 mem = ctx->internal_estimate(ctx, b) * 0.8; // Magical factor accounting for various non-uniformities
+ uns bits = 0;
+ while ((insize >> bits) > mem)
+ bits++;
+
+ // Calculate the possibilities of radix splits
+ uns radix_bits;
+ if (!ctx->radix_split ||
+ (b->flags & SBF_CUSTOM_PRESORT) ||
+ (sorter_debug & SORT_DEBUG_NO_RADIX))
+ radix_bits = 0;
+ else
+ {
+ radix_bits = MIN(bits, b->hash_bits);
+ radix_bits = MIN(radix_bits, sorter_max_radix_bits);
+ if (radix_bits < sorter_min_radix_bits)
+ radix_bits = 0;
+ }
+
+ // The same for multi-way merges
+ uns multiway_bits;
+ if (!ctx->multiway_merge ||
+ (sorter_debug & SORT_DEBUG_NO_MULTIWAY) ||
+ (sorter_debug & SORT_DEBUG_NO_PRESORT))
+ multiway_bits = 0;
+ else
+ {
+ multiway_bits = MIN(bits, sorter_max_multiway_bits);
+ if (multiway_bits < sorter_min_multiway_bits)
+ multiway_bits = 0;
+ }
+
+ SORT_XTRACE(3, "Decisions: size=%s max=%s runs=%d bits=%d hash=%d -> radix=%d multi=%d",
+ stk_fsize(insize), stk_fsize(mem), b->runs, bits, b->hash_bits,
+ radix_bits, multiway_bits);
+
+ // If the input already consists of a single run, just join it
+ if (b->runs)
+ return sorter_join(b);
+
+ // If everything fits in memory, the 2-way strategy will sort it in memory
+ if (!bits)
+ return sorter_twoway(ctx, b);
+
+ // If we can reduce everything in one pass, do so and prefer radix splits
+ if (radix_bits == bits)
+ return sorter_radix(ctx, b, radix_bits);
+ if (multiway_bits == bits)
+ return sorter_multiway(ctx, b);
+
+ // Otherwise, reduce as much as possible and again prefer radix splits
+ if (radix_bits)
+ return sorter_radix(ctx, b, radix_bits);
+ if (multiway_bits)
+ return sorter_multiway(ctx, b);
+
+ // Fall back to 2-way strategy if nothing else applies
+ return sorter_twoway(ctx, b);
+}
+
void
sorter_run(struct sort_context *ctx)
{
ctx->pool = mp_new(4096);
clist_init(&ctx->bucket_list);
-
- /* FIXME: There should be a way how to detect size of the input file */
- /* FIXME: Remember to test sorting of empty files */
+ sorter_prepare_buf(ctx);
+ asort_start_threads(0);
// Create bucket containing the source
struct sort_bucket *bin = sbuck_new(ctx);
else
bin->fb = ctx->in_fb;
bin->ident = "in";
- bin->size = ~(u64)0;
+ bin->size = ctx->in_size;
bin->hash_bits = ctx->hash_bits;
clist_add_tail(&ctx->bucket_list, &bin->n);
+ SORT_XTRACE(2, "Input size: %s, %d hash bits", F_BSIZE(bin), bin->hash_bits);
+ ctx->fb_params = (bin->size < sorter_small_input) ? &sorter_small_fb_params : &sorter_fb_params;
// Create bucket for the output
struct sort_bucket *bout = sbuck_new(ctx);
bout->flags = SBF_FINAL;
- bout->fb = ctx->out_fb;
+ if (bout->fb = ctx->out_fb)
+ bout->flags |= SBF_OPEN_WRITE;
bout->ident = "out";
bout->runs = 1;
clist_add_head(&ctx->bucket_list, &bout->n);
+ // Repeatedly sort buckets
struct sort_bucket *b;
while (bout = clist_head(&ctx->bucket_list), b = clist_next(&ctx->bucket_list, &bout->n))
- {
- if (!sbuck_have(b))
- sbuck_drop(b);
- else if (b->runs == 1)
- sorter_join(b);
- else
- sorter_twoway(ctx, b);
- }
+ sorter_decide(ctx, b);
+ asort_stop_threads();
sorter_free_buf(ctx);
sbuck_write(bout); // Force empty bucket to a file
- SORT_XTRACE("Final size: %jd", (uintmax_t) sbuck_size(bout));
+ SORT_XTRACE(2, "Final size: %s", F_BSIZE(bout));
+ SORT_XTRACE(2, "Final timings: %.3fs external sorting, %.3fs presorting, %.3fs internal sorting",
+ ctx->total_ext_time/1000., ctx->total_pre_time/1000., ctx->total_int_time/1000.);
ctx->out_fb = sbuck_read(bout);
+ mp_delete(ctx->pool);
}
projects / libucw.git / blobdiff