fast-import: use writev(3p) to send cat-blob responses

When answering a `cat-blob` command, `cat_blob()` issues three separate
calls to write(3p) on the cat-blob fd: one for the header line, one for
the full blob payload, and one for the trailing newline. Frontends like
git-filter-repo issue these commands in bulk, once per rewritten blob,
so the syscall overhead adds up.

Use `writev_in_full()` to send all three parts with a single syscall.

This can be benchmarked with the following setup:

    $ git cat-file --unordered --filter=object:type=blob
        --batch-check='cat-blob %(objectname)' --batch-all-objects >request
    $ git fast-import --cat-blob-fd=3 <request

Executing this with 100,000 objects in linux.git:

  Benchmark 1: HEAD~
    Time (mean ± σ):      1.320 s ±  0.003 s    [User: 1.154 s, System: 0.161 s]
    Range (min … max):    1.314 s …  1.324 s    10 runs

  Benchmark 2: HEAD
    Time (mean ± σ):      1.270 s ±  0.022 s    [User: 1.133 s, System: 0.132 s]
    Range (min … max):    1.209 s …  1.282 s    10 runs

  Summary
    HEAD ran
      1.04 ± 0.02 times faster than HEAD~

Signed-off-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

diff --git a/builtin/fast-import.c b/builtin/fast-import.c
index aa656c5195d3669943d973b226b727e35135c56e..48fda01c94359c0641bc0c3edde707fc6145d21e 100644 (file)
--- a/builtin/fast-import.c
+++ b/builtin/fast-import.c
@@ -3332,6 +3332,7 @@ static void cat_blob_write(const char *buf, unsigned long size)
 static void cat_blob(struct object_entry *oe, struct object_id *oid)
 {
        struct strbuf line = STRBUF_INIT;
+       struct iovec iov[3];
        unsigned long size;
        enum object_type type = 0;
        char *buf;
@@ -3365,10 +3366,21 @@ static void cat_blob(struct object_entry *oe, struct object_id *oid)
        strbuf_reset(&line);
        strbuf_addf(&line, "%s %s %"PRIuMAX"\n", oid_to_hex(oid),
                    type_name(type), (uintmax_t)size);
-       cat_blob_write(line.buf, line.len);
+
+       /*
+        * Write the header, the payload and the trailing newline with a
+        * single writev(3p) call instead of three separate write(3p) calls.
+        */
+       iov[0].iov_base = line.buf;
+       iov[0].iov_len = line.len;
+       iov[1].iov_base = buf;
+       iov[1].iov_len = size;
+       iov[2].iov_base = (void *) "\n";
+       iov[2].iov_len = 1;
+
+       if (writev_in_full(cat_blob_fd, iov, ARRAY_SIZE(iov)) < 0)
+               die_errno(_("write to frontend failed"));
        strbuf_release(&line);
-       cat_blob_write(buf, size);
-       cat_blob_write("\n", 1);
        if (oe && oe->pack_id == pack_id) {
                last_blob.offset = oe->idx.offset;
                strbuf_attach(&last_blob.data, buf, size, size + 1);