[PATCH] cciss: fix code style in drivers/block/cciss_scsi.c

[PATCH] cciss: fix code style in drivers/block/cciss_scsi.c

[Steven Liu]

Hi, Mike Miller,

if kmalloc memory for ld_buff failed,it will free currentsd  and
ld_buff once more
so i think this patch may be better

fix code style in drivers/block/cciss_scsi.c

Signed-off-by: Liu Qi <lingjiujianke@gmail.com>
---
 drivers/block/cciss_scsi.c |   20 +++++++++++++++++---
 1 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/drivers/block/cciss_scsi.c b/drivers/block/cciss_scsi.c
index e1d0e2c..ad2417f 100644
--- a/drivers/block/cciss_scsi.c
+++ b/drivers/block/cciss_scsi.c
@@ -1198,13 +1198,24 @@ cciss_update_non_disk_devices(int cntl_num, int hostno)

        c = (ctlr_info_t *) hba[cntl_num];
        ld_buff = kzalloc(reportlunsize, GFP_KERNEL);
+       if(ld_buff == NULL){
+               printk(KERN_ERR "cciss: out of memory\n");
+               goto out1;
+       }
+
        inq_buff = kmalloc(OBDR_TAPE_INQ_SIZE, GFP_KERNEL);
+       if(inq_buff == NULL){
+               printk(KERN_ERR "cciss: out of memory\n");
+               goto out2;
+       }
+
        currentsd = kzalloc(sizeof(*currentsd) *
                        (CCISS_MAX_SCSI_DEVS_PER_HBA+1), GFP_KERNEL);
-       if (ld_buff == NULL || inq_buff == NULL || currentsd == NULL) {
+       if(currentsd == NULL){
                printk(KERN_ERR "cciss: out of memory\n");
-               goto out;
+               goto out3;
        }
+
        this_device = &currentsd[CCISS_MAX_SCSI_DEVS_PER_HBA];
        if (cciss_scsi_do_report_phys_luns(c, ld_buff, reportlunsize) == 0) {
                ch = &ld_buff->LUNListLength[0];
@@ -1289,9 +1300,12 @@ cciss_update_non_disk_devices(int cntl_num, int hostno)

        adjust_cciss_scsi_table(cntl_num, hostno, currentsd, ncurrent);
 out:
+       kfree(currentsd);
+out3:
        kfree(inq_buff);
+out2:
        kfree(ld_buff);
-       kfree(currentsd);
+out1:
        return;
 }

--
1.7.1

Re: [PATCH] cciss: fix code style in drivers/block/cciss_scsi.c

[Li Zefan]

Steven Liu wrote:
> Hi, Mike Miller,
> 
> if kmalloc memory for ld_buff failed,it will free currentsd  and
> ld_buff once more

No, it won't.

> so i think this patch may be better
> 
> fix code style in drivers/block/cciss_scsi.c
> 

There is no bug or coding style issue, so just leave the code as it is.