Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: "Doron Holan [MS]" <doronh@xxxxxxxxxxxxxxxxxxxx>
- Date: Tue, 19 Sep 2006 20:14:22 -0700
let's say i want to initiate i/o to the hw and while i am doing that, the Dx
irp comes in. now you have to stall the power irp and let the i/o finish.
basically, the power state change is done outside of the power policy state
machine and any PPO state machine which implements idle must be able to
control all state changes.
d
--
Please do not send e-mail directly to this alias. this alias is for
newsgroup purposes only.
This posting is provided "AS IS" with no warranties, and confers no rights.
<BubbaGump> wrote in message
news:kmuvg291e9kes9p7hvo64ueafqn7k2he53@xxxxxxxxxx
I'm not understanding in what instances idle detection does and
doesn't work or what state is mismatched and the consequences of the
mismatch.
On Mon, 18 Sep 2006 21:53:09 -0700, "Doron Holan [MS]"
<doronh@xxxxxxxxxxxxxxxxxxxx> wrote:
this is not a bug really, just a flawed design that works in very few
instances. the OS itself keeps the right state under the right lock. the
problem is that the driver maintains its own state on the
device and the OS has no knowledge of that state or the lock that
maintains
that state. apparently the storage stacks use the OS idle detection.
d
.
- Follow-Ups:
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: BubbaGump
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Maxim S. Shatskih
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- References:
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Maxim S. Shatskih
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Doron Holan [MS]
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Maxim S. Shatskih
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Doron Holan [MS]
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Doron Holan [MS]
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Doron Holan [MS]
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- From: Doron Holan [MS]
- Re: IoSkipCurrentIrpStackLocation() design flaw?
- Prev by Date: Re: KeFlushIoBuffers() before or after DMA?
- Next by Date: A working driver for Belkin Wifi adapters on x64?
- Previous by thread: Re: IoSkipCurrentIrpStackLocation() design flaw?
- Next by thread: Re: IoSkipCurrentIrpStackLocation() design flaw?
- Index(es):
Relevant Pages
|
Loading