zlib: slim down zlib_deflate() workspace when possible

Instead of always creating a huge (268K) deflate_workspace with the
maximum compression parameters (windowBits=15, memLevel=8), allow the
caller to obtain a smaller workspace by specifying smaller parameter
values.

For example, when capturing oops and panic reports to a medium with
limited capacity, such as NVRAM, compression may be the only way to
capture the whole report. In this case, a small workspace (24K works
fine) is a win, whether you allocate the workspace when you need it (i.e.,
during an oops or panic) or at boot time.

I've verified that this patch works with all accepted values of windowBits
(positive and negative), memLevel, and compression level.

Signed-off-by: Jim Keniston <[e-mail]>
Cc: Herbert Xu <[e-mail]>
Cc: David Miller <[e-mail]>
Cc: Chris Mason <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>
Signed-off-by: Linus Torvalds <[e-mail]>

crypto/deflate.c

···
 	int ret = 0;
 	struct z_stream_s *stream = &ctx->comp_stream;
 
-	stream->workspace = vzalloc(zlib_deflate_workspacesize());
+	stream->workspace = vzalloc(zlib_deflate_workspacesize(
+				-DEFLATE_DEF_WINBITS, DEFLATE_DEF_MEMLEVEL));
 	if (!stream->workspace) {
 		ret = -ENOMEM;
 		goto out;

crypto/zlib.c

···
 	struct zlib_ctx *ctx = crypto_tfm_ctx(crypto_pcomp_tfm(tfm));
 	struct z_stream_s *stream = &ctx->comp_stream;
 	struct nlattr *tb[ZLIB_COMP_MAX + 1];
+	int window_bits, mem_level;
 	size_t workspacesize;
 	int ret;
 
···
 
 	zlib_comp_exit(ctx);
 
-	workspacesize = zlib_deflate_workspacesize();
+	window_bits = tb[ZLIB_COMP_WINDOWBITS]
+					? nla_get_u32(tb[ZLIB_COMP_WINDOWBITS])
+					: MAX_WBITS;
+	mem_level = tb[ZLIB_COMP_MEMLEVEL]
+					? nla_get_u32(tb[ZLIB_COMP_MEMLEVEL])
+					: DEF_MEM_LEVEL;
+
+	workspacesize = zlib_deflate_workspacesize(window_bits, mem_level);
 	stream->workspace = vzalloc(workspacesize);
 	if (!stream->workspace)
 		return -ENOMEM;
···
 				tb[ZLIB_COMP_METHOD]
 					? nla_get_u32(tb[ZLIB_COMP_METHOD])
 					: Z_DEFLATED,
-				tb[ZLIB_COMP_WINDOWBITS]
-					? nla_get_u32(tb[ZLIB_COMP_WINDOWBITS])
-					: MAX_WBITS,
-				tb[ZLIB_COMP_MEMLEVEL]
-					? nla_get_u32(tb[ZLIB_COMP_MEMLEVEL])
-					: DEF_MEM_LEVEL,
+				window_bits,
+				mem_level,
 				tb[ZLIB_COMP_STRATEGY]
 					? nla_get_u32(tb[ZLIB_COMP_STRATEGY])
 					: Z_DEFAULT_STRATEGY);

drivers/net/ppp_deflate.c

···
 
 	state->strm.next_in   = NULL;
 	state->w_size         = w_size;
-	state->strm.workspace = vmalloc(zlib_deflate_workspacesize());
+	state->strm.workspace = vmalloc(zlib_deflate_workspacesize(-w_size, 8));
 	if (state->strm.workspace == NULL)
 		goto out_free;
 

fs/btrfs/zlib.c

···
 	if (!workspace)
 		return ERR_PTR(-ENOMEM);
 
-	workspace->def_strm.workspace = vmalloc(zlib_deflate_workspacesize());
+	workspace->def_strm.workspace = vmalloc(zlib_deflate_workspacesize(
+						MAX_WBITS, MAX_MEM_LEVEL));
 	workspace->inf_strm.workspace = vmalloc(zlib_inflate_workspacesize());
 	workspace->buf = kmalloc(PAGE_CACHE_SIZE, GFP_NOFS);
 	if (!workspace->def_strm.workspace ||

fs/jffs2/compr_zlib.c

···
 
 static int __init alloc_workspaces(void)
 {
-	def_strm.workspace = vmalloc(zlib_deflate_workspacesize());
+	def_strm.workspace = vmalloc(zlib_deflate_workspacesize(MAX_WBITS,
+							MAX_MEM_LEVEL));
 	if (!def_strm.workspace) {
-		printk(KERN_WARNING "Failed to allocate %d bytes for deflate workspace\n", zlib_deflate_workspacesize());
+		printk(KERN_WARNING "Failed to allocate %d bytes for deflate workspace\n", zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL));
 		return -ENOMEM;
 	}
-	D1(printk(KERN_DEBUG "Allocated %d bytes for deflate workspace\n", zlib_deflate_workspacesize()));
+	D1(printk(KERN_DEBUG "Allocated %d bytes for deflate workspace\n", zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL)));
 	inf_strm.workspace = vmalloc(zlib_inflate_workspacesize());
 	if (!inf_strm.workspace) {
 		printk(KERN_WARNING "Failed to allocate %d bytes for inflate workspace\n", zlib_inflate_workspacesize());

fs/logfs/compr.c

···
 
 int __init logfs_compr_init(void)
 {
-	size_t size = max(zlib_deflate_workspacesize(),
+	size_t size = max(zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL),
 			zlib_inflate_workspacesize());
 	stream.workspace = vmalloc(size);
 	if (!stream.workspace)

include/linux/zlib.h

···
 
                         /* basic functions */
 
-extern int zlib_deflate_workspacesize (void);
+extern int zlib_deflate_workspacesize (int windowBits, int memLevel);
 /*
    Returns the number of bytes that needs to be allocated for a per-
-   stream workspace.  A pointer to this number of bytes should be
-   returned in stream->workspace before calling zlib_deflateInit().
+   stream workspace with the specified parameters.  A pointer to this
+   number of bytes should be returned in stream->workspace before
+   you call zlib_deflateInit() or zlib_deflateInit2().  If you call
+   zlib_deflateInit(), specify windowBits = MAX_WBITS and memLevel =
+   MAX_MEM_LEVEL here.  If you call zlib_deflateInit2(), the windowBits
+   and memLevel parameters passed to zlib_deflateInit2() must not
+   exceed those passed here.
 */
 
 /* 

lib/zlib_deflate/deflate.c

···
     deflate_state *s;
     int noheader = 0;
     deflate_workspace *mem;
+    char *next;
 
     ush *overlay;
     /* We overlay pending_buf and d_buf+l_buf. This works since the average
···
 	strategy < 0 || strategy > Z_HUFFMAN_ONLY) {
         return Z_STREAM_ERROR;
     }
+
+    /*
+     * Direct the workspace's pointers to the chunks that were allocated
+     * along with the deflate_workspace struct.
+     */
+    next = (char *) mem;
+    next += sizeof(*mem);
+    mem->window_memory = (Byte *) next;
+    next += zlib_deflate_window_memsize(windowBits);
+    mem->prev_memory = (Pos *) next;
+    next += zlib_deflate_prev_memsize(windowBits);
+    mem->head_memory = (Pos *) next;
+    next += zlib_deflate_head_memsize(memLevel);
+    mem->overlay_memory = next;
+
     s = (deflate_state *) &(mem->deflate_memory);
     strm->state = (struct internal_state *)s;
     s->strm = strm;
···
     return flush == Z_FINISH ? finish_done : block_done;
 }
 
-int zlib_deflate_workspacesize(void)
+int zlib_deflate_workspacesize(int windowBits, int memLevel)
 {
-    return sizeof(deflate_workspace);
+    if (windowBits < 0) /* undocumented feature: suppress zlib header */
+        windowBits = -windowBits;
+
+    /* Since the return value is typically passed to vmalloc() unchecked... */
+    BUG_ON(memLevel < 1 || memLevel > MAX_MEM_LEVEL || windowBits < 9 ||
+							windowBits > 15);
+
+    return sizeof(deflate_workspace)
+        + zlib_deflate_window_memsize(windowBits)
+        + zlib_deflate_prev_memsize(windowBits)
+        + zlib_deflate_head_memsize(memLevel)
+        + zlib_deflate_overlay_memsize(memLevel);
 }

lib/zlib_deflate/defutil.h

···
 typedef struct deflate_workspace {
     /* State memory for the deflator */
     deflate_state deflate_memory;
-    Byte window_memory[2 * (1 << MAX_WBITS)];
-    Pos prev_memory[1 << MAX_WBITS];
-    Pos head_memory[1 << (MAX_MEM_LEVEL + 7)];
-    char overlay_memory[(1 << (MAX_MEM_LEVEL + 6)) * (sizeof(ush)+2)];
+    Byte *window_memory;
+    Pos *prev_memory;
+    Pos *head_memory;
+    char *overlay_memory;
 } deflate_workspace;
+
+#define zlib_deflate_window_memsize(windowBits) \
+	(2 * (1 << (windowBits)) * sizeof(Byte))
+#define zlib_deflate_prev_memsize(windowBits) \
+	((1 << (windowBits)) * sizeof(Pos))
+#define zlib_deflate_head_memsize(memLevel) \
+	((1 << ((memLevel)+7)) * sizeof(Pos))
+#define zlib_deflate_overlay_memsize(memLevel) \
+	((1 << ((memLevel)+6)) * (sizeof(ush)+2))
 
 /* Output a byte on the stream.
  * IN assertion: there is enough room in pending_buf.