Skip to contentSkip to conversation

scsi: message: fusion: remove unused timeleft variable

PATCH In review Tom Rix 2 participants 3 messages Mar 31, 2023

Subsystems (2)
lore
Share & tools
lore

Tom Rix

author
#0 lore raw

PATCH 1/1 scsi: message: fusion: remove unused timeleft variable

clang with W=1 reports drivers/message/fusion/mptsas.c:4796:17: error: variable
'timeleft' set but not used [-Werror,-Wunused-but-set-variable]

        unsigned long    timeleft;
                         ^

This variable is not used so remove it.

---
 drivers/message/fusion/mptsas.c | 3 ---
 1 file changed, 3 deletions(-)
drivers/message/fusion/mptsas.c +0 -3
index 86f16f3ea478..d458665e2fc9 100644
@@ -4793,7 +4793,6 @@ mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun,
 	MPT_FRAME_HDR	*mf;
 	SCSITaskMgmt_t	*pScsiTm;
 	int		 retval;
-	unsigned long	 timeleft;
 
 	*issue_reset = 0;
 	mf = mpt_get_msg_frame(mptsasDeviceResetCtx, ioc);
@@ -4829,8 +4828,6 @@ mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun,
 	mpt_put_msg_frame_hi_pri(mptsasDeviceResetCtx, ioc, mf);
 
 	/* Now wait for the command to complete */
-	timeleft = wait_for_completion_timeout(&ioc->taskmgmt_cmds.done,
-	    timeout*HZ);
 	if (!(ioc->taskmgmt_cmds.status & MPT_MGMT_STATUS_COMMAND_GOOD)) {
 		retval = -1; /* return failure */
 		dtmprintk(ioc, printk(MYIOC_s_ERR_FMT
Signature · 1 line
2.27.0
#1 lore raw
Quoted 26 lines · Tom Rix

clang with W=1 reports drivers/message/fusion/mptsas.c:4796:17: error: variable
'timeleft' set but not used [-Werror,-Wunused-but-set-variable]

        unsigned long    timeleft;
                         ^

This variable is not used so remove it.

---
 drivers/message/fusion/mptsas.c | 3 ---
 1 file changed, 3 deletions(-)
drivers/message/fusion/mptsas.c +0 -1
index 86f16f3ea478..d458665e2fc9 100644
@@ -4793,7 +4793,6 @@ mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun,
        MPT_FRAME_HDR   *mf;
        SCSITaskMgmt_t  *pScsiTm;
        int              retval;
-       unsigned long    timeleft;
 
        *issue_reset = 0;
        mf = mpt_get_msg_frame(mptsasDeviceResetCtx, ioc);

drivers/message/fusion/mptsas.c mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun, · line 4833

Quoted 5 lines
drivers/message/fusion/mptsas.c +0 -2
@@ -4829,8 +4828,6 @@ mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun,
        mpt_put_msg_frame_hi_pri(mptsasDeviceResetCtx, ioc, mf);
 
        /* Now wait for the command to complete */
-       timeleft = wait_for_completion_timeout(&ioc->taskmgmt_cmds.done,
-           timeout*HZ);

It looks bad to remove the call to wait_for_completion_timeout(). Is it truly not needed? If it's needed, the "timeleft" should be checked or a comment left to explain why it's not checked.

-bw

Quoted 7 lines
drivers/message/fusion/mptsas.c +0 -1
        if (!(ioc->taskmgmt_cmds.status & MPT_MGMT_STATUS_COMMAND_GOOD)) {
                retval = -1; /* return failure */
                dtmprintk(ioc, printk(MYIOC_s_ERR_FMT
--

2.27.0

author
#2 lore raw
Quoted 28 lines · Bill Wendling
Quoted 27 lines · Tom Rix

clang with W=1 reports drivers/message/fusion/mptsas.c:4796:17: error: variable
'timeleft' set but not used [-Werror,-Wunused-but-set-variable]

         unsigned long    timeleft;
                          ^

This variable is not used so remove it.

---
  drivers/message/fusion/mptsas.c | 3 ---
  1 file changed, 3 deletions(-)
drivers/message/fusion/mptsas.c +0 -1
index 86f16f3ea478..d458665e2fc9 100644
@@ -4793,7 +4793,6 @@ mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun,
         MPT_FRAME_HDR   *mf;
         SCSITaskMgmt_t  *pScsiTm;
         int              retval;
-       unsigned long    timeleft;
 
         *issue_reset = 0;
         mf = mpt_get_msg_frame(mptsasDeviceResetCtx, ioc);
@@ -4829,8 +4828,6 @@ mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun,
         mpt_put_msg_frame_hi_pri(mptsasDeviceResetCtx, ioc, mf);
 
Quoted 6 lines

drivers/message/fusion/mptsas.c mptsas_issue_tm(MPT_ADAPTER *ioc, u8 type, u8 channel, u8 id, u64 lun, · line 4833

Quoted 3 lines
drivers/message/fusion/mptsas.c +0 -2
         /* Now wait for the command to complete */
-       timeleft = wait_for_completion_timeout(&ioc->taskmgmt_cmds.done,
-           timeout*HZ);

It looks bad to remove the call to wait_for_completion_timeout(). Is it truly not needed? If it's needed, the "timeleft" should be checked or a comment left to explain why it's not checked.

Yeah. this is a screw up on my part, sorry.

T

Quoted 10 lines

-bw

Quoted 7 lines
drivers/message/fusion/mptsas.c +0 -1
         if (!(ioc->taskmgmt_cmds.status & MPT_MGMT_STATUS_COMMAND_GOOD)) {
                 retval = -1; /* return failure */
                 dtmprintk(ioc, printk(MYIOC_s_ERR_FMT
--

2.27.0