On 10/8/24 11:29 AM, Mark Brown wrote: > On Tue, Oct 08, 2024 at 10:34:39AM +0100, Tudor Ambarus wrote: >> On 10/8/24 9:32 AM, Mihai Sain wrote: > >>> Add support to print the controller version and used irq >>> similar to other at91 drivers (spi, twi, usart). > >>> + dev_info(&pdev->dev, "AT91 QSPI Controller version %#x (irq %d)\n", >>> + atmel_qspi_read(aq, QSPI_VERSION), irq); > >> This pollutes the console. Better to add a dev_dbg if you care. >> And irq number doesn't bring too much value as you can see it in dt, >> isn't it? > > The objective of bringing the various AT91 drivers into consistency does > seem useful so if this isn't OK for this driver we should probably right, consistency is good. > update the other drivers as well. Ensuring that people can get at the Can be a goal. My point was that printing too much info in the boot log may hide other more important information. Printing IP versions, irqs, dma channels acquired (or worse, that a driver probed successful) shall be part of debug prints if someone cares. > IP version does feel useful, I guess it could also be a sysfs thing? I'd also consider debugfs for IP version, it has less restrictions. Mihai, do as you find best, it's just a suggestion.