Re: [ANNOUNCE]: version 2.00.3 megaraid driver for 2.4.x and 2.5.67 kernels

From: Christoph Hellwig (hch@infradead.org)
Date: Thu Apr 17 2003 - 07:38:20 EST


On Wed, Apr 16, 2003 at 04:34:22PM -0400, Mukker, Atul wrote:
> New megaraid driver 2.00.3 is now available at
> ftp://ftp.lsil.com/pub/linux-megaraid For this driver, a patch is also
> available for 2.5.67 kernel.

Hmm, why are you patching and announcing 2.00.3 here if 2.00.4beta is
already out?

Anyway, here's some comments for the driver:

static int hba_count;
static adapter_t *hba_soft_state[MAX_CONTROLLERS];

    a) please avoid typedefs for structs. Also adapter_t is a bit
    too generic. What about struct mega_adapter instead?
    b) please don't use such global arrays but always use the
    ->hostdata field.

static struct proc_dir_entry *mega_proc_dir_entry;

static struct notifier_block mega_notifier = {
        .notifier_call = megaraid_reboot_notify
};

    Do you really need this? Why can you use
    struct device_driver->shutdown?

/*
 * megaraid_validate_parms()
 *
 * Validate that any module parms passed in
 * have proper values.
 */
static void
megaraid_validate_parms(void)
{
        if( (max_cmd_per_lun <= 0) || (max_cmd_per_lun > MAX_CMD_PER_LUN) )
                max_cmd_per_lun = MAX_CMD_PER_LUN;
        if( max_mbox_busy_wait > MBOX_BUSY_WAIT )
                max_mbox_busy_wait = MBOX_BUSY_WAIT;
}

    Please run the whole driver through scripts/Lindent

/**
 * megaraid_detect()
 * @host_template - Our soft state maintained by mid-layer
 *
 * the detect entry point for the mid-layer.
 * We scan the PCI bus for our controllers and start them.
 *
 * Note: PCI_DEVICE_ID_PERC4_DI below represents the PERC4/Di class of
 * products. All of them share the same vendor id, device id, and subsystem
 * vendor id but different subsystem ids. As of now, driver does not use the
 * subsystem id.
 */
static int
megaraid_detect(Scsi_Host_Template *host_template)

    Please get rid of ->detect and use the new-style pci API +
    scsi_add_host instead.

        host_template->proc_name = "megaraid";

    CAn be initialized at compile-time.

        printk(KERN_NOTICE "megaraid: " MEGARAID_VERSION);

        megaraid_validate_parms();

        memset(mega_hbas, 0, sizeof (mega_hbas));

        hba_count = 0;

    These are in .bss

        }

        return;

}
    This is superflous.

/*
 * megaraid_proc_info()
 *
 * Returns data to be displayed in /proc/scsi/megaraid/X
 */
static int
megaraid_proc_info(char *buffer, char **start, off_t offset, int length,
                int host_no, int inout)
{
        *start = buffer;
        return 0;
}

    Just don't implement this method instead?

volatile static int internal_done_flag = 0;
volatile static int internal_done_errcode = 0;

static DECLARE_WAIT_QUEUE_HEAD (internal_wait);

static void internal_done (Scsi_Cmnd *cmd)
{
        internal_done_errcode = cmd->result;
        internal_done_flag++;
        wake_up (&internal_wait);
}

/* shouldn't be used, but included for completeness */

static int
megaraid_command (Scsi_Cmnd *cmd)
{
        internal_done_flag = 0;

        /* Queue command, and wait until it has completed */
        megaraid_queue (cmd, internal_done);

        while (!internal_done_flag)
                interruptible_sleep_on (&internal_wait);

        return internal_done_errcode;
}

    This looks horribly racy.

                                return TRUE;
                        }
                }
        }

        return FALSE;
}

    Please don't use TRUE/FALSE but 1/0 directly.

static int
megadev_open (struct inode *inode, struct file *filep)
{
        /*
         * Only allow superuser to access private ioctl interface
         */
        if( !capable(CAP_SYS_ADMIN) ) return -EACCES;

        if (!try_module_get(THIS_MODULE)) {
                return -ENXIO;
        }

   This is broken as hell (and I fixed it in the old megraid driver
   ages ago!) Just set ->owner in the file_operation.

        /*
         * Wait till all the issued commands are complete and there are no
         * commands in the pending queue
         */
        while( atomic_read(&adapter->pend_cmds) > 0 ||
                        !list_empty(&adapter->pending_list) ) {

                sleep_on_timeout( &wq, 1*HZ ); /* sleep for 1s */
        }

  Again racy. You need to opencode this (similar to wait_event, maybe
  just add wait_even_timeout).

        if (adapter->read_ldidmap) {
                struct list_head *pos;
                list_for_each(pos, &adapter->pending_list) {
                        scb = list_entry(pos, scb_t, list);

  list_for_each_entry()?

/**
 * mega_allocate_inquiry()
 * @dma_handle - handle returned for dma address
 * @pdev - handle to pci device
 *
 * allocates memory for inquiry structure
 */
static inline caddr_t
mega_allocate_inquiry(dma_addr_t *dma_handle, struct pci_dev *pdev)
{
        return pci_alloc_consistent(pdev, sizeof(mega_inquiry3), dma_handle);
}

static inline void
mega_free_inquiry(caddr_t inquiry, dma_addr_t dma_handle, struct pci_dev *pdev)
{
        pci_free_consistent(pdev, sizeof(mega_inquiry3), inquiry, dma_handle);
}

Do you really need a wrapper for those?

        /*
         * For all internal commands, the buffer must be allocated in <4GB
         * address range
         */
        if( make_local_pdev(adapter, &pdev) != 0 ) return -1;

    So use dma_alloc_coherent with a GFP_KERNEL arg instead of such ugly hacks.

        /*
         * Wait till this command finishes. Do not use
         * wait_event_interruptible(). It causes panic if CTRL-C is hit when
         * dumping e.g., physical disk information through /proc interface.
         */
#if 0
        wait_event_interruptible(adapter->int_waitq, scmd->state);
#endif
        wait_event(adapter->int_waitq, scmd->state);

    Maybe you need to actually look at wait_event_interruptible and
    handle it's return value? :)

static Scsi_Host_Template driver_template = MEGARAID;

    Get rid of the ugly macro and add the members right here.

While we're at it: Please also get rid of all those prototypes in
megaraid.h
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Apr 23 2003 - 22:00:21 EST