Matthew Wilcox wrote:
scsi_print_msg got the transfer period translation to nanoseconds wrong. We have the right code to do this in scsi_transport_spi.c, but we can't guarantee that'll be linked in. scsi_print_msg is a bit of a misnomer anyway, since these messages are actually SPI messages, not SAM. So I've moved scsi_print_msg to scsi_transport_spi.c and renamed it to spi_print_msg. All drivers that use scsi_print_msg() got converted, and the Kconfig was adjusted to link in the SPI transport code for those drivers. Note that those drivers haven't been converted to use the SPI transport code ... but surely it's just a matter of time.
Strong ACK, though I think your change here should be split up: the patch to link against SPI should be separated from the scsi_print_msg() change.
This change will make it easier to start marginalizing SPI. Eventually all SPI drivers will link against the SPI transport class, and this is the necessary first step.
Jeff - : 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
