On 12/14/18 4:27 PM, Sergei Shtylyov wrote:
On 12/13/2018 01:46 PM, Hannes Reinecke wrote:
Replace all DPRINTK calls with the ata_XXX_dbg functions.
Signed-off-by: Hannes Reinecke <hare@xxxxxxxx>
---
drivers/ata/sata_nv.c | 22 ++++++++++------------
1 file changed, 10 insertions(+), 12 deletions(-)
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 72c9b922a77b..aa2611d638ea 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -1451,7 +1451,7 @@ static unsigned int nv_adma_qc_issue(struct ata_queued_cmd *qc)
writew(qc->hw_tag, mmio + NV_ADMA_APPEND);
- DPRINTK("Issued tag %u\n", qc->hw_tag);
+ ata_dev_dbg(qc->dev, "Issued tag %u\n", qc->hw_tag);
Don't we lose printing out __func__ this way?
Yes, but given that this message is pretty unique (for this driver) I
thought the omission wasn't too bad.
I can re-add it if you insist...
return 0;
}
@@ -2029,8 +2029,6 @@ static unsigned int nv_swncq_issue_atacmd(struct ata_port *ap,
if (qc == NULL)
return 0;
- DPRINTK("Enter\n");
-
You said "replace all", not "remove some". :-)
Though w/o __func__ this is pretty useless indeed...
Which is why I removed it.
I'll be updating the description.
[...]
@@ -2053,7 +2051,7 @@ static unsigned int nv_swncq_qc_issue(struct ata_queued_cmd *qc)
if (qc->tf.protocol != ATA_PROT_NCQ)
return ata_bmdma_qc_issue(qc);
- DPRINTK("Enter\n");
+ ata_dev_dbg(qc->dev, "Enter\n");
Same here, do we print out __func__ now? Else this is quite pointless.
From what I can see this is primarily so that you can trace the control
flow, but I wonder if there are not better ways nowadays (one thinks of
ftrace ...).
I guess I'll just drop the pointless "ENTER" messages.
if (!pp->qc_active)
nv_swncq_issue_atacmd(ap, qc);
[...]
@@ -2136,10 +2134,10 @@ static int nv_swncq_sdbfis(struct ata_port *ap)
*/
lack_dhfis = 1;
- DPRINTK("id 0x%x QC: qc_active 0x%x,"
+ ata_port_dbg(ap, "QC: qc_active 0x%llx,"
Why silently change "%x" to "%llx"?
Because the compiler complained?
Do I need to update the description for this change, too?
Cheers,
Hannes