- 05 Jan, 2022 28 commits
-
-
Hannes Reinecke authored
Signed-off-by:
Hannes Reinecke <hare@suse.de> Signed-off-by:
Damien Le Moal <damien.lemoal@opensource.wdc.com>
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Quite some information from the VPRINTK() is already covered by tracepoints, so remove the pointless calls and convert the remaining ones to structured logging. Signed-off-by:
-
Hannes Reinecke authored
Drop pointless VPRINTK() call and convert the remaining one to dev_dbg(). Signed-off-by:
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Drop pointless VPRINTK() calls for entering and existing interrupt routines and convert the remaining calls to ata_port_dbg(). Signed-off-by:
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Drop pointless VPRINTK() calls and convert the remaining calls to the existing bmdma tracepoint. Signed-off-by:
-
Hannes Reinecke authored
Drop pointless VPRINTK() calls for entering routines and setting up sg tables. And while we're at it, remove the disabled debugging messages. Signed-off-by:
-
Hannes Reinecke authored
Drop pointless VPRINTK() calls for entering and existing interrupt routines and convert the remaining calls to dev_dbg(). Signed-off-by:
-
Hannes Reinecke authored
Most of the information is already covered by tracepoints (if not downright pointless), so remove the VPRINTK() calls. And while we're at it, remove ata_scsi_dump_cdb(), too, as this information can be retrieved from scsi tracing. Signed-off-by:
-
Hannes Reinecke authored
Use standard ata logging macros instead of the hand-crafted PDPRINTK and remove duplicate logging messages. Signed-off-by:
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Replace all DPRINTK calls with the ata_XXX_dbg functions. Signed-off-by:
-
Hannes Reinecke authored
The DPRINTK call doesn't print information which isn't already covered by tracepoints later on. Remove it. Signed-off-by:
-
Hannes Reinecke authored
There is only so much information to be glanced when the interrupt routine is called and exited, so remove these DPRINTK() calls. Signed-off-by:
-
Hannes Reinecke authored
Move the DPRINTK calls over to dynamic debugging. Signed-off-by:
-
Hannes Reinecke authored
Always '-1', so drop it and simplify the function. Signed-off-by:
-
Hannes Reinecke authored
Replace all DPRINTK calls with ata_dev_dbg(). Signed-off-by:
-
Hannes Reinecke authored
Rename ata_get_cmd_descrip() to ata_get_cmd_name() and simplify it to return "unknown" instead of NULL. Signed-off-by:
-
Hannes Reinecke authored
Use standard pr_{debug,info,notice,warn,err} macros instead of the hand-crafted printk helpers. Signed-off-by: -
Hannes Reinecke authored
Add tracepoints for ATA error handling. Signed-off-by:
-
Hannes Reinecke authored
Drop DPRINTK calls for cdb translation as they are already covered by other traces, and also drop the DPRINTK calls in ata_scsi_hotplug(). Signed-off-by:
-
Hannes Reinecke authored
Add tracepoints for the HSM state machine and drop DPRINTK calls Signed-off-by:
-
Hannes Reinecke authored
Add tracepoints for bus-master DMA and taskfile related functions. That allows us to drop the relevant DPRINTK() calls. Signed-off-by:
-
- 04 Jan, 2022 12 commits
-
-
Hannes Reinecke authored
Convert the existing ata_qc_issue() tracepoint into a template, and add tracepoints for ata_qc_prep() and ata_qc_issue() based on that template. Signed-off-by:
-
Hannes Reinecke authored
To follow the flow of control we should be using tracepoints, as they will tie in with the actual I/O flow and deliver a better overview about what it happening. This patch adds tracepoints for hard reset, soft reset, and postreset and adds them in the libata-eh control flow. With that we can drop the reset DPRINTK calls in the various drivers. Signed-off-by:
-
Hannes Reinecke authored
With moving ata_dev_dbg() over to dynamic debugging ATA_HORKAGE_DUMP_ID will now print out the raw IDENTIFY data without a header unless explicitly enable via dyndebug. So move the logging level up to INFO and have the header printed always. Signed-off-by:
-
Hannes Reinecke authored
Use ata_dev_dbg() to print out the information in ata_dump_id() and remove the ata_msg_probe() conditional. Signed-off-by:
-
Hannes Reinecke authored
Add an ata_port_classify() helper to print out the results from the device classification and remove the debugging statements from ata_dev_classify(). Signed-off-by:
-
Hannes Reinecke authored
Signed-off-by:
-
Hannes Reinecke authored
Debugging messages in pci init functions or sg setup are pretty much pointless, as the workflow pretty much decides what happened. So drop them. Signed-off-by:
-
Changcheng Deng authored
Use min() in order to make code cleaner. Reported-by:
Zeal Robot <zealci@zte.com.cn> Signed-off-by:
Changcheng Deng <deng.changcheng@zte.com.cn> Signed-off-by:
-
Andy Shevchenko authored
platform_get_irq() will print a message when it fails. No need to repeat this. Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by:
-
Andy Shevchenko authored
If 64-bit mask attempt fails, the 32-bit will fail by the very same reason. Don't even try the latter. It's a continuation of the changes that contains, e.g. dcc02c19 ("sata_sil24: use dma_set_mask_and_coherent"). Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
-
Andy Shevchenko authored
sata_dwc_port_stop() is not used before being defined, remove redundant forward declaration. Signed-off-by:
-
Andy Shevchenko authored
Use temporary variable for struct device to make code neater. Signed-off-by:
-
