Re: [PATCH] fstests: fix blksize_t printf format warnings across architectures

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]





在 2024/11/21 08:51, Darrick J. Wong 写道:
On Thu, Nov 21, 2024 at 08:36:58AM +1030, Qu Wenruo wrote:


在 2024/11/20 22:10, Anand Jain 写道:
Fix format string warnings when printing blksize_t values that vary
across architectures. The warning occurs because blksize_t is defined
differently between architectures: aarch64 architectures blksize_t is
int, on x86-64 it's long-int.  Cast the values to long. Fixes warnings
as below.

   seek_sanity_test.c:110:45: warning: format '%ld' expects argument of type
   'long int', but argument 3 has type 'blksize_t' {aka 'int'}

   attr_replace_test.c:70:22: warning: format '%ld' expects argument of type
   'long int', but argument 3 has type '__blksize_t' {aka 'int'}

Why not just use %zu instead?

 From printf(3):

        z      A  following  integer conversion corresponds to a size_t
               or ssize_t argument, or a following n conversion  corre‐
               sponds to a pointer to a size_t argument.

blksize_t is not a ssize_t.

You're right, it's indeed different and it's more complex than I thought.

blksize_t is __SYSCALL_SLONG_TYPE, which has extra handling on x86_64 with its x32 mode handling. Only for x86_64 x32 mode it is __SQUAD_TYPE (depending on wordsize again) otherwise it's __SLONGWORD_TYPE (fixed to long int).

Meanwhile ssize_t is __SWORD_TYPE, which is only dependent on word size.
For 32bit word size it's __int64_t, and for 64bit it's long int.

So blksize_t is more weird than ssize_t.

Then force converting to long (int) is fine.


--D

Thanks,
Qu


Signed-off-by: Anand Jain <anand.jain@xxxxxxxxxx>
---
   src/attr_replace_test.c | 2 +-
   src/seek_sanity_test.c  | 2 +-
   2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/attr_replace_test.c b/src/attr_replace_test.c
index 1218e7264c8f..5d560a633361 100644
--- a/src/attr_replace_test.c
+++ b/src/attr_replace_test.c
@@ -67,7 +67,7 @@ int main(int argc, char *argv[])
   	if (ret < 0) die();
   	size = sbuf.st_blksize * 3 / 4;
   	if (!size)
-		fail("Invalid st_blksize(%ld)\n", sbuf.st_blksize);
+		fail("Invalid st_blksize(%ld)\n", (long)sbuf.st_blksize);

Although for this case, I'd prefer to use "long int" just for the sake of consistency.

Otherwise looks good to me.

Reviewed-by: Qu Wenruo <wqu@xxxxxxxx>

Thanks,
Qu

   	size = MIN(size, maxsize);
   	value = malloc(size);
   	if (!value)
diff --git a/src/seek_sanity_test.c b/src/seek_sanity_test.c
index a61ed3da9a8f..c5930357911f 100644
--- a/src/seek_sanity_test.c
+++ b/src/seek_sanity_test.c
@@ -107,7 +107,7 @@ static int get_io_sizes(int fd)
   		offset += pos ? 0 : 1;
   	alloc_size = offset;
   done:
-	fprintf(stdout, "Allocation size: %ld\n", alloc_size);
+	fprintf(stdout, "Allocation size: %ld\n", (long)alloc_size);
   	return 0;

   fail:








[Index of Archives]     [Linux Filesystems Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux