On 2021/12/09 16:17, Hannes Reinecke wrote:
> On 12/9/21 1:38 AM, Damien Le Moal wrote:
>> On 2021/12/09 1:31, Hannes Reinecke wrote:
>>> Implement module parameter 'pci_dump' and move the DPRINTK calls
>>> over to dev_printk().
>>>
>>> Signed-off-by: Hannes Reinecke <hare@xxxxxxx>
>>> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@xxxxxxxxxxx>
>>> ---
>>> drivers/ata/sata_mv.c | 88 ++++++++++++++++++++++++-------------------
>>> 1 file changed, 49 insertions(+), 39 deletions(-)
>>>
>>> diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
>>> index cae4c1eab102..f0257685495f 100644
>>> --- a/drivers/ata/sata_mv.c
>>> +++ b/drivers/ata/sata_mv.c
>>> @@ -83,6 +83,10 @@ module_param(irq_coalescing_usecs, int, S_IRUGO);
>>> MODULE_PARM_DESC(irq_coalescing_usecs,
>>> "IRQ coalescing time threshold in usecs");
>>>
>>> +static int pci_dump;
>>> +module_param(pci_dump, int, S_IRUGO);
>>> +MODULE_PARM_DESC(pci_dump, "Enable dumping of PCI registers on error");
>>> +
>>> enum {
>>> /* BAR's are enumerated in terms of pci_resource_start() terms */
>>> MV_PRIMARY_BAR = 0, /* offset 0x10: memory space */
>>> @@ -1248,42 +1252,43 @@ static int mv_stop_edma(struct ata_port *ap)
>>> return err;
>>> }
>>>
>>> -#ifdef ATA_DEBUG
>>> -static void mv_dump_mem(void __iomem *start, unsigned bytes)
>>> +static void mv_dump_mem(struct device *dev, void __iomem *start, unsigned bytes)
>>> {
>>> - int b, w;
>>> + int b, w, o;
>>> + unsigned char linebuf[38];
>>> +
>>> for (b = 0; b < bytes; ) {
>>> - DPRINTK("%p: ", start + b);
>>> - for (w = 0; b < bytes && w < 4; w++) {
>>> - printk("%08x ", readl(start + b));
>>> + for (w = 0, o = 0; b < bytes && w < 4; w++) {
>>> + o += snprintf(linebuf + o, 38 - o,
>>> + "%08x ", readl(start + b));
>>> b += sizeof(u32);
>>> }
>>> - printk("\n");
>>> + dev_printk(KERN_DEBUG, dev, "%s: %p: %s\n",
>>> + __func__, start + b, linebuf);
>>
>> Why not dev_dbg() ? Same comment for all the prints below.
>>
> Because the entire thing is encapsulated by the 'pci_dump' parameter, ie
> it'll be only ever called when 'pci_dump' is set.
> So it feels redundant to require both, 'pci_dump' _and_ dynamic debugging.
>
> But I can move it over to dynamic debugging and kill the pci_dump parameter.
That sounds simpler to me.
>
> Cheers,
>
> Hannes
--
Damien Le Moal
Western Digital Research
