[PATCH 2/2] cache: fix integer conversion error in parsing cache_pool_max_chunks

The allocation/cache_pool_max_chunks config is read as an int32_t,
then converted to uint64_t. If the value in lvm.conf exceeds INT32_MAX,
conversion to uint64_t results in truncation followed by sign extension,
causing the original unsigned representation to be lost. For example, a
config value of 6442450944 is converted to 18446744071562067968.

This fix preserves the original u64 value from the parser. Further
adjustments should be applied to fit dm-cache's metadata limits.

Signed-off-by: Ming-Hung Tsai <[email protected]>
---
 lib/config/config.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/config/config.c b/lib/config/config.c
index 9df9c925c152..fc7d16279f31 100644
--- a/lib/config/config.c
+++ b/lib/config/config.c
@@ -2555,7 +2555,7 @@ uint64_t get_default_allocation_cache_pool_max_chunks_CFG(struct cmd_context *cm
 	 *       newer targets may scale better.
 	 */
 	uint64_t default_max_chunks = DEFAULT_CACHE_POOL_MAX_CHUNKS;
-	uint64_t max_chunks = find_config_tree_int(cmd, allocation_cache_pool_max_chunks_CFG, profile);
+	uint64_t max_chunks = find_config_tree_int64(cmd, allocation_cache_pool_max_chunks_CFG, profile);
 
 	if (!max_chunks)
 		max_chunks = default_max_chunks;
-- 
2.47.0