[Qemu-devel] [PATCH] fix for treating "long" as "int64_t" in block-vmdk.c

[Qemu-devel] [PATCH] fix for treating "long" as "int64_t" in block-vmdk.c

[Zeev Tarantov]

On 32bit x86 I get these warnings:

block-vmdk.c: In function 'vmdk_create':
block-vmdk.c:796: warning: format '%lu' expects type 'long unsigned
int', but argument 8 has type 'int64_t'
block-vmdk.c:796: warning: format '%lu' expects type 'long unsigned
int', but argument 8 has type 'int64_t'

It seems the code was developed on a 64bit machine because the author
treats "long in" as "int64_t".
This patch fixes the warnings and the downcast.
There are no more uses of "long"in the code.
Please consider committing this, as well as my previous patch.

--- block-vmdk.c        (revision 5208)
+++ block-vmdk.c        (working copy)
@@ -710,13 +710,13 @@
         "createType=\"monolithicSparse\"\n"
         "\n"
         "# Extent description\n"
-        "RW %lu SPARSE \"%s\"\n"
+        "RW %llu SPARSE \"%s\"\n"
         "\n"
         "# The Disk Data Base \n"
         "#DDB\n"
         "\n"
         "ddb.virtualHWVersion = \"%d\"\n"
-        "ddb.geometry.cylinders = \"%lu\"\n"
+        "ddb.geometry.cylinders = \"%llu\"\n"
         "ddb.geometry.heads = \"16\"\n"
         "ddb.geometry.sectors = \"63\"\n"
         "ddb.adapterType = \"ide\"\n";
@@ -792,7 +792,7 @@
     if ((temp_str = strrchr(real_filename, ':')) != NULL)
         real_filename = temp_str + 1;
     snprintf(desc, sizeof(desc), desc_template, (unsigned int)time(NULL),
-             (unsigned long)total_size, real_filename,
+             (uint64_t)total_size, real_filename,
              (flags & BLOCK_FLAG_COMPAT6 ? 6 : 4), total_size / (63 * 16));

     /* write the descriptor */

-Zeev