> If you have trouble convincing people we need this for some new
> reason, then maybe you could review the existing callers to see if
> some of the existing call sites make it any more convincing.
>
> A very quick review, I found:
>
> send_message_to_frontend() initStringInfo(&buf) 1024 is probably too big
> HandleParallelMessages() seems to know exactly how many bytes are
> needed. Can this use a read-only StringInfo?
> send_feedback() seems to know what size of the buffer it needs
I looked into send_feedback(). Maybe with the new API we could do
something like attached?
> buf_init() knows the exact size.
>
> I didn't review the patch in detail, but I think "initsize" would be a
> better parameter name than "size".
>
> David
>
>
diff --git a/src/backend/replication/logical/worker.c b/src/backend/replication/logical/worker.c
index 334bf3e7aff..66cc4ec49a1 100644
--- a/src/backend/replication/logical/worker.c
+++ b/src/backend/replication/logical/worker.c
@@ -3885,8 +3885,10 @@ send_feedback(XLogRecPtr recvpos, bool force, bool requestReply)
if (!reply_message)
{
MemoryContext oldctx = MemoryContextSwitchTo(ApplyContext);
-
- reply_message = makeStringInfo();
+ int initsize =
+ sizeof(char) + sizeof(int64) * 3 +
+ sizeof(TimestampTz) + sizeof(bool) + 1;
+ reply_message = makeStringInfoExtended(initsize);
MemoryContextSwitchTo(oldctx);
}
else
