We get a warning from 'make headers_check' about a newly introduced
usage of integer types in the scsi/scsi_bsg_ufs.h uapi header:
usr/include/scsi/scsi_bsg_ufs.h:18: found __[us]{8,16,32,64} type without #include <linux/types.h>
Aside from the missing linux/types.h inclusion, I also noticed that it
uses the wrong types: 'u32' is not available at all in user space,
and 'uint32_t' depends on the inclusion of a standard header that
we should not include from kernel headers.
Change the all to __u32 and similar types here.
I also note the usage of '__be32' and '__be16' that seems unfortunate
for a user space API. I wonder if it would be better to define the
interface in terms of a CPU-endian structure and convert it in kernel
space.
Fixes: e77044c5a842 ("scsi: ufs-bsg: Add support for uic commands in ufs_bsg_request()")
Fixes: df032bf27a41 ("scsi: ufs: Add a bsg endpoint that supports UPIUs")
Signed-off-by: Arnd Bergmann <[email protected]>
---
include/uapi/scsi/scsi_bsg_ufs.h | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/include/uapi/scsi/scsi_bsg_ufs.h b/include/uapi/scsi/scsi_bsg_ufs.h
index 1b25930688bc..17c7abd0803a 100644
--- a/include/uapi/scsi/scsi_bsg_ufs.h
+++ b/include/uapi/scsi/scsi_bsg_ufs.h
@@ -8,6 +8,7 @@
#ifndef SCSI_BSG_UFS_H
#define SCSI_BSG_UFS_H
+#include <linux/types.h>
/*
* This file intended to be included by both kernel and user space
*/
@@ -15,7 +16,7 @@
#define UFS_CDB_SIZE 16
#define UPIU_TRANSACTION_UIC_CMD 0x1F
/* uic commands are 4DW long, per UFSHCI V2.1 paragraph 5.6.1 */
-#define UIC_CMD_SIZE (sizeof(u32) * 4)
+#define UIC_CMD_SIZE (sizeof(__u32) * 4)
/**
* struct utp_upiu_header - UPIU header structure
@@ -59,7 +60,7 @@ struct utp_upiu_query {
*/
struct utp_upiu_cmd {
__be32 exp_data_transfer_len;
- u8 cdb[UFS_CDB_SIZE];
+ __u8 cdb[UFS_CDB_SIZE];
};
/**
@@ -81,7 +82,7 @@ struct utp_upiu_req {
/* request (CDB) structure of the sg_io_v4 */
struct ufs_bsg_request {
- uint32_t msgcode;
+ __u32 msgcode;
struct utp_upiu_req upiu_req;
};
@@ -95,10 +96,10 @@ struct ufs_bsg_reply {
* msg and status fields. The per-msgcode reply structure
* will contain valid data.
*/
- uint32_t result;
+ __u32 result;
/* If there was reply_payload, how much was received? */
- uint32_t reply_payload_rcv_len;
+ __u32 reply_payload_rcv_len;
struct utp_upiu_req upiu_rsp;
};
--
2.18.0
Hi Arnd,
> We get a warning from 'make headers_check' about a newly introduced
> usage of integer types in the scsi/scsi_bsg_ufs.h uapi header:
>
> usr/include/scsi/scsi_bsg_ufs.h:18: found __[us]{8,16,32,64} type without
> #include <linux/types.h>
>
> Aside from the missing linux/types.h inclusion, I also noticed that it
> uses the wrong types: 'u32' is not available at all in user space,
> and 'uint32_t' depends on the inclusion of a standard header that
> we should not include from kernel headers.
>
> Change the all to __u32 and similar types here.
Ok, Thank you for fixing this.
Reviewed-by: Avri Altman <[email protected]>
>
> I also note the usage of '__be32' and '__be16' that seems unfortunate
> for a user space API. I wonder if it would be better to define the
> interface in terms of a CPU-endian structure and convert it in kernel
> space.
I think its fine to leave this as it is.
The user-space app that will use this API, is the mmc-utils sibling for ufs devices.
It uses the ufs transport-layer protocol, so whomever uses it,
Is aware of, and will account for the hw endianness whenever needed,
Just like it's done in mmc-utils.
Thanks a lot,
Avri
Arnd,
> We get a warning from 'make headers_check' about a newly introduced
> usage of integer types in the scsi/scsi_bsg_ufs.h uapi header:
>
> usr/include/scsi/scsi_bsg_ufs.h:18: found __[us]{8,16,32,64} type without #include <linux/types.h>
>
> Aside from the missing linux/types.h inclusion, I also noticed that it
> uses the wrong types: 'u32' is not available at all in user space,
> and 'uint32_t' depends on the inclusion of a standard header that
> we should not include from kernel headers.
>
> Change the all to __u32 and similar types here.
>
> I also note the usage of '__be32' and '__be16' that seems unfortunate
> for a user space API. I wonder if it would be better to define the
> interface in terms of a CPU-endian structure and convert it in kernel
> space.
Applied to 4.20/scsi-queue, thank you!
--
Martin K. Petersen Oracle Linux Engineering