summaryrefslogtreecommitdiff
path: root/drivers/usb/core
diff options
context:
space:
mode:
authorAlan Stern <stern@rowland.harvard.edu>2010-02-26 11:49:39 -0500
committerGreg Kroah-Hartman <gregkh@suse.de>2010-03-02 14:55:07 -0800
commit49d6271b85a3e18062eaf4d6f8d899abe00a7725 (patch)
tree83b92caacf276bdd054a03320fd6f48cbe96c562 /drivers/usb/core
parentefe7daf2231a6beb59f0f641461d19fa62fb83ab (diff)
downloadlwn-49d6271b85a3e18062eaf4d6f8d899abe00a7725.tar.gz
lwn-49d6271b85a3e18062eaf4d6f8d899abe00a7725.zip
usb-storage: use max_hw_sectors instead of max_sectors
This patch (as1347) makes some adjustments to the way usb-storage handles the request-queue parameters. USB host controllers are able to handle arbitrarily long scatter-gather lists, since they are limited only by main memory and not by the controller hardware. Hence the sg_tablesize field in the host template can be increased to the maximum value. Drivers like usb-storage aren't supposed to touch the queue's max_sectors parameter; instead they are supposed to use the max_hw_sectors value. Accordingly, the patch replaces calls of queue_max_sectors() with calls of queue_max_hw_sectors(). Oddly enough, the blk_queue_max_sectors() routine is nevertheless still appropriate. The existing code imposes a limit of SCSI_DEFAULT_MAX_SECTORS (1024) on the values accepted by the max_sectors attribute file. There's no reason not to accept larger values, so the limit is removed. (It would be nice to change the file's name to max_hw_sectors, but the old name is already a well-established API.) Signed-off-by: Alan Stern <stern@rowland.harvard.edu> CC: Matthew Dharm <mdharm-usb@one-eyed-alien.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Diffstat (limited to 'drivers/usb/core')
0 files changed, 0 insertions, 0 deletions