From bb33663b402060b6664d3a1829daad0e44fa9af0 Mon Sep 17 00:00:00 2001
From: Ori Bernstein <ori@eigenstate.org>
Date: Sat, 11 Jun 2022 16:36:45 +0000
Subject: [PATCH] git/get: keep sending what we have until we get an ack

Git9 was sloppy about telling git what commits we have.

We would list the commits at the tip of the branch, but not
walk down it, which means we would request too much data if
our local branches were ahead of the remote.

This patch changes that, sending the tips *and* the first
256 commits after them, so that git can produce a better
pack for us, with fewer redundant commits.
---
 sys/src/cmd/git/get.c   | 75 +++++++++++++++++++++++++++++++++--------
 sys/src/cmd/git/git.h   |  1 +
 sys/src/cmd/git/proto.c | 14 ++++++++
 sys/src/cmd/git/serve.c | 14 --------
 4 files changed, 76 insertions(+), 28 deletions(-)

diff --git a/sys/src/cmd/git/get.c b/sys/src/cmd/git/get.c
index 95dfe106e..77547c0c0 100644
--- a/sys/src/cmd/git/get.c
+++ b/sys/src/cmd/git/get.c
@@ -177,17 +177,35 @@ fail(char *pack, char *idx, char *msg, ...)
 	exits(buf);
 }
 
+void
+enqueueparent(Objq *q, Object *o)
+{
+	Object *p;
+	int i;
+
+	if(o->type != GCommit)
+		return;
+	for(i = 0; i < o->commit->nparent; i++){
+		if((p = readobject(o->commit->parent[i])) == nil)
+			continue;
+		qput(q, p, 0);
+		unref(p);
+	}
+}
+
 int
 fetchpack(Conn *c)
 {
 	char buf[Pktmax], *sp[3], *ep;
-	char *packtmp, *idxtmp, **ref;
+	char *packtmp, *idxtmp, **ref, *caps;
 	Hash h, *have, *want;
-	int nref, refsz, first;
-	int i, n, l, req, pfd;
+	int nref, refsz, first, nsent;
+	int i, l, n, req, pfd;
 	vlong packsz;
 	Objset hadobj;
 	Object *o;
+	Objq haveq;
+	Qelt e;
 
 	nref = 0;
 	refsz = 16;
@@ -234,6 +252,7 @@ fetchpack(Conn *c)
 	if(writephase(c) == -1)
 		sysfatal("write: %r");
 	req = 0;
+	caps = " multi_ack";
 	for(i = 0; i < nref; i++){
 		if(hasheq(&have[i], &want[i]))
 			continue;
@@ -241,30 +260,58 @@ fetchpack(Conn *c)
 			unref(o);
 			continue;
 		}
-		n = snprint(buf, sizeof(buf), "want %H\n", want[i]);
-		if(writepkt(c, buf, n) == -1)
+		if(fmtpkt(c, "want %H%s\n", want[i], caps) == -1)
 			sysfatal("could not send want for %H", want[i]);
+		caps = "";
 		req = 1;
 	}
 	flushpkt(c);
+
+	nsent = 0;
+	qinit(&haveq);
 	osinit(&hadobj);
+	/*
+	 * We know we have these objects, and we want to make sure that
+	 * they end up at the front of the queue. Send the 'have lines'
+	 * first, and then enqueue their parents for a second round of
+	 * sends.
+	 */
 	for(i = 0; i < nref; i++){
 		if(hasheq(&have[i], &Zhash) || oshas(&hadobj, have[i]))
 			continue;
 		if((o = readobject(have[i])) == nil)
 			sysfatal("missing object we should have: %H", have[i]);
+		if(fmtpkt(c, "have %H", o->hash) == -1)
+			sysfatal("write: %r");
+		enqueueparent(&haveq, o);
 		osadd(&hadobj, o);
-		unref(o);	
-		n = snprint(buf, sizeof(buf), "have %H\n", have[i]);
-		if(writepkt(c, buf, n + 1) == -1)
-			sysfatal("could not send have for %H", have[i]);
+		unref(o);
+	}
+	/*
+	 * While we could short circuit this and check if upstream has
+	 * acked our objects, for the first 256 haves, this is simple
+	 * enough.
+	 *
+	 * Also, doing multiple rounds of reference discovery breaks
+	 * when using smart http.
+	 */
+	while(req && qpop(&haveq, &e) && nsent < 256){
+		if(oshas(&hadobj, e.o->hash))
+			continue;
+		if((o = readobject(e.o->hash)) == nil)
+			sysfatal("missing object we should have: %H", have[i]);
+		if(fmtpkt(c, "have %H", o->hash) == -1)
+			sysfatal("write: %r");
+		enqueueparent(&haveq, o);
+		osadd(&hadobj, o);
+		unref(o);
+		nsent++;
 	}
 	osclear(&hadobj);
+	qclear(&haveq);
 	if(!req)
 		flushpkt(c);
-
-	n = snprint(buf, sizeof(buf), "done\n");
-	if(writepkt(c, buf, n) == -1)
+	if(fmtpkt(c, "done\n") == -1)
 		sysfatal("write: %r");
 	if(!req)
 		goto showrefs;
@@ -298,9 +345,9 @@ fetchpack(Conn *c)
 		if(strncmp(buf, "PACK", 4) == 0)
 			break;
 		l = strtol(buf, &ep, 16);
-		if(l == 0 || ep != buf + 4)
+		if(ep != buf + 4)
 			sysfatal("fetch packfile: junk pktline");
-		if(readn(c->rfd, buf, l) != l)
+		if(readn(c->rfd, buf, l-4) != l-4)
 			sysfatal("fetch packfile: short read");
 	}
 	if(write(pfd, "PACK", 4) != 4)
diff --git a/sys/src/cmd/git/git.h b/sys/src/cmd/git/git.h
index 88afcc1d7..c3ba7b7a9 100644
--- a/sys/src/cmd/git/git.h
+++ b/sys/src/cmd/git/git.h
@@ -313,6 +313,7 @@ Delta*	deltify(Object*, Dtab*, int*);
 /* proto handling */
 int	readpkt(Conn*, char*, int);
 int	writepkt(Conn*, char*, int);
+int	fmtpkt(Conn*, char*, ...);
 int	flushpkt(Conn*);
 void	initconn(Conn*, int, int);
 int	gitconnect(Conn *, char *, char *);
diff --git a/sys/src/cmd/git/proto.c b/sys/src/cmd/git/proto.c
index ddc2258f3..9af09ec19 100644
--- a/sys/src/cmd/git/proto.c
+++ b/sys/src/cmd/git/proto.c
@@ -93,6 +93,20 @@ writepkt(Conn *c, char *buf, int nbuf)
 	return 0;
 }
 
+int
+fmtpkt(Conn *c, char *fmt, ...)
+{
+	char pkt[Pktmax];
+	va_list ap;
+	int n;
+
+	va_start(ap, fmt);
+	n = vsnprint(pkt, sizeof(pkt), fmt, ap);
+	n = writepkt(c, pkt, n);
+	va_end(ap);
+	return n;
+}
+
 int
 flushpkt(Conn *c)
 {
diff --git a/sys/src/cmd/git/serve.c b/sys/src/cmd/git/serve.c
index 895c6f5e0..da7cd56b0 100644
--- a/sys/src/cmd/git/serve.c
+++ b/sys/src/cmd/git/serve.c
@@ -8,20 +8,6 @@
 char	*pathpfx = nil;
 int	allowwrite;
 
-int
-fmtpkt(Conn *c, char *fmt, ...)
-{
-	char pkt[Pktmax];
-	va_list ap;
-	int n;
-
-	va_start(ap, fmt);
-	n = vsnprint(pkt, sizeof(pkt), fmt, ap);
-	n = writepkt(c, pkt, n);
-	va_end(ap);
-	return n;
-}
-
 int
 showrefs(Conn *c)
 {