
23 Aug
2012
23 Aug
'12
6:26 p.m.
Hi Mike,
On Wed, Aug 22, 2012 at 10:30 PM, Mike Frysinger vapier@gentoo.org wrote:
On Friday 17 August 2012 16:49:38 Joe Hershberger wrote:
* Break reads up into very small chunks so fw_printenv doesn't
* block the kernel long enough to starve other kernel tasks.
err, this sounds like a bug in your kernel driver. why should userspace be working around buggy drivers ?
The problem is something to do with the mtd driver blocking some resource on large reads that starves USB transfers. We never fully investigated it. It's true that this is a work-around. Is that so bad? Maybe so.
-Joe