[...] > Be careful printing errors per-IO; you could get thousands of them if things > go bad. > The block layer print_req_error() uses printk_ratelimited(KERN_ERR) for that > reason, > and the SCSI layer scsi_io_completion_action() maintains a ratelimit on its > own. > > The dev_err_ratelimited() macro might be a good fit here.
Thanks for the tip. I updated the patch and send a V2 for that. I adjusted also to commit message to make it more explicit. Thanks for the review. Erwan,
