v2-0002-Clean-up-constants-in-pg_dump-s-compression-API.patch

From 4aa7603d891c62bf9d95af9910b8fb4b0fe2fb10 Mon Sep 17 00:00:00 2001
From: Georgios Kokolatos <gkokolatos@pm.me>
Date: Thu, 16 Mar 2023 16:30:09 +0000
Subject: [PATCH v2 2/3] Clean up constants in pg_dump's compression API.

Prior to the introduction of the API, pg_dump would use the ZLIB_[IN|OUT]_SIZE
constants to handle buffer sizes throughout. This behaviour is confusing after
the introduction of the API. Ammend it by introducing a DEFAULT_IO_BUFFER_SIZE
constant to use when appropriate while giving the opportunity to specific
compression implementations to use their own.

With the help and guidance of Tomas Vondra.
---
 src/bin/pg_dump/compress_gzip.c       | 22 +++++++++++-----------
 src/bin/pg_dump/compress_io.h         |  5 ++---
 src/bin/pg_dump/compress_lz4.c        | 17 +++++++++--------
 src/bin/pg_dump/compress_none.c       |  4 ++--
 src/bin/pg_dump/pg_backup_directory.c |  4 ++--
 5 files changed, 26 insertions(+), 26 deletions(-)

diff --git a/src/bin/pg_dump/compress_gzip.c b/src/bin/pg_dump/compress_gzip.c
index 29e2fd8d50..4106d4c866 100644
--- a/src/bin/pg_dump/compress_gzip.c
+++ b/src/bin/pg_dump/compress_gzip.c
@@ -120,8 +120,8 @@ WriteDataToArchiveGzip(ArchiveHandle *AH, CompressorState *cs,
 		 * actually allocate one extra byte because some routines want to
 		 * append a trailing zero byte to the zlib output.
 		 */
-		gzipcs->outbuf = pg_malloc(ZLIB_OUT_SIZE + 1);
-		gzipcs->outsize = ZLIB_OUT_SIZE;
+		gzipcs->outsize = DEFAULT_IO_BUFFER_SIZE;
+		gzipcs->outbuf = pg_malloc(gzipcs->outsize + 1);
 
 		/*
 		 * A level of zero simply copies the input one block at the time. This
@@ -158,10 +158,10 @@ ReadDataFromArchiveGzip(ArchiveHandle *AH, CompressorState *cs)
 	zp->zfree = Z_NULL;
 	zp->opaque = Z_NULL;
 
-	buf = pg_malloc(ZLIB_IN_SIZE);
-	buflen = ZLIB_IN_SIZE;
+	buflen = DEFAULT_IO_BUFFER_SIZE;
+	buf = pg_malloc(buflen);
 
-	out = pg_malloc(ZLIB_OUT_SIZE + 1);
+	out = pg_malloc(DEFAULT_IO_BUFFER_SIZE + 1);
 
 	if (inflateInit(zp) != Z_OK)
 		pg_fatal("could not initialize compression library: %s",
@@ -176,14 +176,14 @@ ReadDataFromArchiveGzip(ArchiveHandle *AH, CompressorState *cs)
 		while (zp->avail_in > 0)
 		{
 			zp->next_out = (void *) out;
-			zp->avail_out = ZLIB_OUT_SIZE;
+			zp->avail_out = DEFAULT_IO_BUFFER_SIZE;
 
 			res = inflate(zp, 0);
 			if (res != Z_OK && res != Z_STREAM_END)
 				pg_fatal("could not uncompress data: %s", zp->msg);
 
-			out[ZLIB_OUT_SIZE - zp->avail_out] = '\0';
-			ahwrite(out, 1, ZLIB_OUT_SIZE - zp->avail_out, AH);
+			out[DEFAULT_IO_BUFFER_SIZE - zp->avail_out] = '\0';
+			ahwrite(out, 1, DEFAULT_IO_BUFFER_SIZE - zp->avail_out, AH);
 		}
 	}
 
@@ -192,13 +192,13 @@ ReadDataFromArchiveGzip(ArchiveHandle *AH, CompressorState *cs)
 	while (res != Z_STREAM_END)
 	{
 		zp->next_out = (void *) out;
-		zp->avail_out = ZLIB_OUT_SIZE;
+		zp->avail_out = DEFAULT_IO_BUFFER_SIZE;
 		res = inflate(zp, 0);
 		if (res != Z_OK && res != Z_STREAM_END)
 			pg_fatal("could not uncompress data: %s", zp->msg);
 
-		out[ZLIB_OUT_SIZE - zp->avail_out] = '\0';
-		ahwrite(out, 1, ZLIB_OUT_SIZE - zp->avail_out, AH);
+		out[DEFAULT_IO_BUFFER_SIZE - zp->avail_out] = '\0';
+		ahwrite(out, 1, DEFAULT_IO_BUFFER_SIZE - zp->avail_out, AH);
 	}
 
 	if (inflateEnd(zp) != Z_OK)
diff --git a/src/bin/pg_dump/compress_io.h b/src/bin/pg_dump/compress_io.h
index b03d5b325b..60e1735834 100644
--- a/src/bin/pg_dump/compress_io.h
+++ b/src/bin/pg_dump/compress_io.h
@@ -17,9 +17,8 @@
 
 #include "pg_backup_archiver.h"
 
-/* Initial buffer sizes used in zlib compression. */
-#define ZLIB_OUT_SIZE	4096
-#define ZLIB_IN_SIZE	4096
+/* Default size used for IO buffers */
+#define DEFAULT_IO_BUFFER_SIZE	4096
 
 extern char *supports_compression(const pg_compress_specification compression_spec);
 
diff --git a/src/bin/pg_dump/compress_lz4.c b/src/bin/pg_dump/compress_lz4.c
index df2b4c9546..0c0eb09d68 100644
--- a/src/bin/pg_dump/compress_lz4.c
+++ b/src/bin/pg_dump/compress_lz4.c
@@ -20,9 +20,6 @@
 #include <lz4.h>
 #include <lz4frame.h>
 
-#define LZ4_OUT_SIZE	(4 * 1024)
-#define LZ4_IN_SIZE		(16 * 1024)
-
 /*
  * LZ4F_HEADER_SIZE_MAX first appeared in v1.7.5 of the library.
  * Redefine it for installations with a lesser version.
@@ -57,7 +54,7 @@ ReadDataFromArchiveLZ4(ArchiveHandle *AH, CompressorState *cs)
 	size_t		buflen;
 	size_t		cnt;
 
-	buflen = LZ4_IN_SIZE;
+	buflen = DEFAULT_IO_BUFFER_SIZE;
 	buf = pg_malloc(buflen);
 	decbuf = pg_malloc(buflen);
 
@@ -206,7 +203,7 @@ LZ4File_init(LZ4File *fs, int size, bool compressing)
 
 	if (fs->compressing)
 	{
-		fs->buflen = LZ4F_compressBound(LZ4_IN_SIZE, &fs->prefs);
+		fs->buflen = LZ4F_compressBound(DEFAULT_IO_BUFFER_SIZE, &fs->prefs);
 		if (fs->buflen < LZ4F_HEADER_SIZE_MAX)
 			fs->buflen = LZ4F_HEADER_SIZE_MAX;
 
@@ -242,9 +239,11 @@ LZ4File_init(LZ4File *fs, int size, bool compressing)
 			return 1;
 		}
 
-		fs->buflen = size > LZ4_OUT_SIZE ? size : LZ4_OUT_SIZE;
+		if (size > DEFAULT_IO_BUFFER_SIZE)
+			fs->buflen = size;
+		else
+			fs->buflen = DEFAULT_IO_BUFFER_SIZE;
 		fs->buffer = pg_malloc(fs->buflen);
-
 		fs->overflowalloclen = fs->buflen;
 		fs->overflowbuf = pg_malloc(fs->overflowalloclen);
 		fs->overflowlen = 0;
@@ -421,8 +420,10 @@ LZ4File_write(const void *ptr, size_t size, CompressFileHandle *CFH)
 
 	while (remaining > 0)
 	{
-		int			chunk = remaining < LZ4_IN_SIZE ? remaining : LZ4_IN_SIZE;
+		int			chunk = DEFAULT_IO_BUFFER_SIZE;
 
+		if (remaining < DEFAULT_IO_BUFFER_SIZE)
+			chunk = remaining;
 		remaining -= chunk;
 
 		status = LZ4F_compressUpdate(fs->ctx, fs->buffer, fs->buflen,
diff --git a/src/bin/pg_dump/compress_none.c b/src/bin/pg_dump/compress_none.c
index bd479fde59..f4a7d7c193 100644
--- a/src/bin/pg_dump/compress_none.c
+++ b/src/bin/pg_dump/compress_none.c
@@ -33,8 +33,8 @@ ReadDataFromArchiveNone(ArchiveHandle *AH, CompressorState *cs)
 	char	   *buf;
 	size_t		buflen;
 
-	buf = pg_malloc(ZLIB_OUT_SIZE);
-	buflen = ZLIB_OUT_SIZE;
+	buflen = DEFAULT_IO_BUFFER_SIZE;
+	buf = pg_malloc(buflen);
 
 	while ((cnt = cs->readF(AH, &buf, &buflen)))
 	{
diff --git a/src/bin/pg_dump/pg_backup_directory.c b/src/bin/pg_dump/pg_backup_directory.c
index 1cd9805ef7..8b92f42ac5 100644
--- a/src/bin/pg_dump/pg_backup_directory.c
+++ b/src/bin/pg_dump/pg_backup_directory.c
@@ -394,8 +394,8 @@ _PrintFileData(ArchiveHandle *AH, char *filename)
 	if (!CFH)
 		pg_fatal("could not open input file \"%s\": %m", filename);
 
-	buf = pg_malloc(ZLIB_OUT_SIZE);
-	buflen = ZLIB_OUT_SIZE;
+	buflen = DEFAULT_IO_BUFFER_SIZE;
+	buf = pg_malloc(buflen);
 
 	while (CFH->read_func(buf, buflen, &cnt, CFH) == 0 && cnt > 0)
 	{
-- 
2.34.1