On Tue, Jun 28, 2011 at 12:29:36PM -0700, Andy Grover wrote: > Instead of converting rd_dr to be se_mem-free, maybe we should just > remove it. Do we really need *two* private ramdisk implementations, > especially when the kernel already has its own ramdisk code, that we > could access via the iblock backstore? > > This also lets us remove hooks like do_se_mem_map, that are in the common > path but only implemented by rd_dr. I always hated the hooks the direct ramdisk code had to the core code, so I'm absolutely in favour of it. The indirect ramdisk currently is used for things like the virtual lun0 handling, which requires it to be just there instead of setting up a block device. I think keeping the simple internal ramdisk around for that and benchmarking is fine. -- 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
