[Bug 202022] slow (780KB/s) USB2.0 + VT82xxxx + yenta (cardbus pcmcia)

[bugzilla at redhat.com]

Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.

Summary: slow (780KB/s) USB2.0 + VT82xxxx + yenta (cardbus pcmcia)


https://bugzilla.redhat.com/show_bug.cgi?id=202022





------- Additional Comments From zaitcev at redhat.com  2008-04-06 21:16 EST -------
Let's look at this (in the first trace):

d2ffd680 2715166762 S Bi:1:002:2 -115 8192 <
d2ffd700 2715166775 S Bi:1:002:2 -115 16384 <
d2ffd780 2715166780 S Bi:1:002:2 -115 16384 <
d2ffd800 2715166785 S Bi:1:002:2 -115 16384 <
d2ffd880 2715166789 S Bi:1:002:2 -115 16384 <
d2ffd900 2715166795 S Bi:1:002:2 -115 16384 <
d2ffd980 2715166799 S Bi:1:002:2 -115 16384 <
d2ffda00 2715166804 S Bi:1:002:2 -115 16384 <
d2ffd680 2715175971 C Bi:1:002:2 0 8192 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d2ffd700 2715275971 C Bi:1:002:2 0 16384 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d2ffd780 2715275980 C Bi:1:002:2 0 16384 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d2ffd800 2715275985 C Bi:1:002:2 0 16384 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d2ffd880 2715373227 C Bi:1:002:2 0 16384 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d2ffd900 2715373236 C Bi:1:002:2 0 16384 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d2ffd980 2715373242 C Bi:1:002:2 0 16384 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d2ffda00 2715373246 C Bi:1:002:2 0 16384 = 00000000 00000000 00000000 00000000
00000000 00000000 00000000 00000000
d3183680 2715373287 S Bi:1:002:2 -115 13 <

It's clear as day that callbacks for big reads arrive in clumps
of about 3 at 100ms intervals, which is the value of TIMER_IO_WATCHDOG.

So, the system loses interrupts. Why? Beats me. I suspect that the
specific EHCI silicon doesn't like something that we do, but I don't
know what that might be. I haven't seen anything like this before.

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.