On Mon, Nov 08, 2010 at 09:27:22AM -0800, Luben Tuikov wrote: > --- On Mon, 11/8/10, Matthew Wilcox <willy@xxxxxxxxxxxxxxx> wrote: > > -0700, Luben Tuikov wrote: > > > Eliminate an infinite loop whereby the SCSI layer > > > would reissue a command (which would be failed by > > > the driver) ad infinitum. (Invariably due to the > > > driver's profuse use of the > > > SCSI_MLQUEUE_DEVICE_BUSY returned result in its > > > queuecommand() method.) > > > > > > Also add a debug option and a few debug prints. > > > > Why have you added your own debug scheme instead of using > > dev_dbg? > > Because, this debug "scheme" produces copious and otherwise > unnecessary volume of debug information in a working driver. And > because you can turn it on/off just for this driver. It is intended to > be used only for debugging this driver and the UAS device(s) to which > the driver communicates. 99.9% of the time, this setting will be a 'N' > out there. And this is what the Kconfig help entry says: "If unsure, > say 'N'." dev_dbg() is able to be turned on and off dynamically when the kernel is running without rebuilding anything, which is what you really want to have happen for a driver being used by normal users. It also produces the output in a standard manner that we have all agreed to follow. thanks, greg k-h -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
