revert mkfs change; might run out of inodes, but large number of inodes
makes mkfs not want to mkfs large filesystems. find optimal formula (e.g. dynamic capping of no. of inodes) later.
This commit is contained in:
@@ -266,9 +266,12 @@ char *argv[];
|
||||
if (blocks == 0) pexit("Can't open prototype file");
|
||||
}
|
||||
if (i == 0) {
|
||||
u32_t kb = div64u(mul64u(blocks, block_size), 1024);
|
||||
i = kb / 2;
|
||||
if (kb >= 100000) i = kb / 4;
|
||||
i = blocks / 2;
|
||||
if (blocks >= 20000) i = blocks / 3;
|
||||
if (blocks >= 40000) i = blocks / 4;
|
||||
if (blocks >= 60000) i = blocks / 5;
|
||||
if (blocks >= 80000) i = blocks / 6;
|
||||
if (blocks >= 100000) i = blocks / 7;
|
||||
|
||||
/* round up to fill inode block */
|
||||
i += inodes_per_block - 1;
|
||||
|
||||
Reference in New Issue
Block a user