incbin: if we have to fread(), try to do it only once...

If we can't mmap a file and instead have to fread(), if the data is
small enough that we can reasonably accomodate it in a memory buffer,
then just read it once.

It seems rather unlikely that very large files would be used with
TIMES anyway.

Also note: the previous comment about nasm_file_size[_by_path]() being
invoked twice was spurious; it does not actually happen.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>

3 files changed, 17 insertions, 4 deletions

diff --git a/asm/assemble.c b/asm/assemble.c
index 463d7378..b09f7cbb 100644
--- a/asm/assemble.c
+++ b/asm/assemble.c
@@ -544,6 +544,9 @@ static bool jmp_match(int32_t segment, int64_t offset, int bits,
     return is_byte;
 }
 
+/* This is totally just a wild guess what is reasonable... */
+#define INCBIN_MAX_BUF (ZERO_BUF_SIZE * 16)
+
 int64_t assemble(int32_t segment, int64_t start, int bits, iflag_t cp,
                  insn * instruction)
 {
@@ -610,11 +613,13 @@ int64_t assemble(int32_t segment, int64_t start, int bits, iflag_t cp,
     } else if (instruction->opcode == I_INCBIN) {
         const char *fname = instruction->eops->stringval;
         FILE *fp;
-        static char buf[BUFSIZ];
         size_t t = instruction->times;
         off_t base = 0;
         off_t len;
         const void *map = NULL;
+        char *buf = NULL;
+        size_t blk = 0;         /* Buffered I/O block size */
+        size_t m = 0;           /* Bytes last read */
 
         fp = nasm_open_read(fname, NF_BINARY|NF_FORMAP);
         if (!fp) {
@@ -651,6 +656,10 @@ int64_t assemble(int32_t segment, int64_t start, int bits, iflag_t cp,
 
         /* Try to map file data */
         map = nasm_map_file(fp, base, len);
+        if (!map) {
+            blk = len < (off_t)INCBIN_MAX_BUF ? (size_t)len : INCBIN_MAX_BUF;
+            buf = nasm_malloc(blk);
+        }
 
         while (t--) {
             data.insoffs = 0;
@@ -658,6 +667,8 @@ int64_t assemble(int32_t segment, int64_t start, int bits, iflag_t cp,
 
             if (map) {
                 out_rawdata(&data, map, len);
+            } else if ((off_t)m == len) {
+                out_rawdata(&data, buf, len);
             } else {
                 off_t l = len;
 
@@ -668,8 +679,7 @@ int64_t assemble(int32_t segment, int64_t start, int bits, iflag_t cp,
                     goto end_incbin;
                 }
                 while (l > 0) {
-                    size_t m = l < (off_t)sizeof(buf) ? (size_t)l : sizeof(buf);
-                    m = fread(buf, 1, m, fp);
+                    m = fread(buf, 1, l < (off_t)blk ? (size_t)l : blk, fp);
                     if (!m || feof(fp)) {
                         /*
                          * This shouldn't happen unless the file
@@ -699,6 +709,8 @@ int64_t assemble(int32_t segment, int64_t start, int bits, iflag_t cp,
                        " reading file `%s'", fname);
         }
     close_done:
+        if (buf)
+            nasm_free(buf);
         if (map)
             nasm_unmap_file(map, len);
         fclose(fp);
diff --git a/include/nasmlib.h b/include/nasmlib.h
index e24ac1d5..b1c490ca 100644
--- a/include/nasmlib.h
+++ b/include/nasmlib.h
@@ -455,7 +455,7 @@ FILE *nasm_open_write(const char *filename, enum file_flags flags);
 /* Probe for existence of a file */
 bool nasm_file_exists(const char *filename);
 
-#define ZERO_BUF_SIZE 4096      /* Default value */
+#define ZERO_BUF_SIZE 65536     /* Default value */
 #if defined(BUFSIZ) && (BUFSIZ > ZERO_BUF_SIZE)
 # undef ZERO_BUF_SIZE
 # define ZERO_BUF_SIZE BUFSIZ
diff --git a/test/incbin.asm b/test/incbin.asm
index 634302ea..20aa4d9f 100644
--- a/test/incbin.asm
+++ b/test/incbin.asm
@@ -1,5 +1,6 @@
 	db '*** ONCE ***', 0Ah
 	incbin "incbin.data",32
 
+	section more start=0x1000000
 	db '*** TWELVE ***', 0Ah
 	times 12 incbin "incbin.data",32