scsi: myrs: avoid stack overflow warning
authorArnd Bergmann <arnd@arndb.de>
Fri, 2 Nov 2018 15:44:13 +0000 (16:44 +0100)
committerMartin K. Petersen <martin.petersen@oracle.com>
Tue, 6 Nov 2018 03:35:01 +0000 (22:35 -0500)
Putting a 1024 byte data structure on the stack is generally a bad idea.
On 32-bit systems, it also triggers a compile-time warning when building
with -Og:

drivers/scsi/myrs.c: In function 'myrs_get_ctlr_info':
drivers/scsi/myrs.c:212:1: error: the frame size of 1028 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]

We only really need three members of the structure, so just read them
manually here instead of copying the entire structure.

Fixes: 77266186397c ("scsi: myrs: Add Mylex RAID controller (SCSI interface)")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
drivers/scsi/myrs.c

index 0264a2e2bc190e068832a89452215919b9b22c30..b8d54ef8cf6dfb84c6d4cb3ce2714071cc2de1d6 100644 (file)
@@ -163,9 +163,12 @@ static unsigned char myrs_get_ctlr_info(struct myrs_hba *cs)
        dma_addr_t ctlr_info_addr;
        union myrs_sgl *sgl;
        unsigned char status;
-       struct myrs_ctlr_info old;
+       unsigned short ldev_present, ldev_critical, ldev_offline;
+
+       ldev_present = cs->ctlr_info->ldev_present;
+       ldev_critical = cs->ctlr_info->ldev_critical;
+       ldev_offline = cs->ctlr_info->ldev_offline;
 
-       memcpy(&old, cs->ctlr_info, sizeof(struct myrs_ctlr_info));
        ctlr_info_addr = dma_map_single(&cs->pdev->dev, cs->ctlr_info,
                                        sizeof(struct myrs_ctlr_info),
                                        DMA_FROM_DEVICE);
@@ -198,9 +201,9 @@ static unsigned char myrs_get_ctlr_info(struct myrs_hba *cs)
                    cs->ctlr_info->rbld_active +
                    cs->ctlr_info->exp_active != 0)
                        cs->needs_update = true;
-               if (cs->ctlr_info->ldev_present != old.ldev_present ||
-                   cs->ctlr_info->ldev_critical != old.ldev_critical ||
-                   cs->ctlr_info->ldev_offline != old.ldev_offline)
+               if (cs->ctlr_info->ldev_present != ldev_present ||
+                   cs->ctlr_info->ldev_critical != ldev_critical ||
+                   cs->ctlr_info->ldev_offline != ldev_offline)
                        shost_printk(KERN_INFO, cs->host,
                                     "Logical drive count changes (%d/%d/%d)\n",
                                     cs->ctlr_info->ldev_critical,