NOW: GeekFest WAS: Thread status in a lot of i/o traffic

GeekFest is on my 61 acre wilderness ranch on top of Palomar Mountain in San
Diego California. This will be the fourth GeekFest and it gets better each
year.

Jamey Kirby, Windows DDK MVP
StorageCraft Inc.
xxxxx@storagecraft.com
http://www.storagecraft.com

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Sinha, Prokash
Sent: Thursday, January 29, 2004 9:54 AM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

Well the way I see it is pretty round. Those coming from the other side of
the
California river ( well if my metric is Mega-meter) might not like to make
it to east coast…

How about silicon valley ? Or may be we should pick about 10 places around
the world, and do a drawing so noone would feel about it …

-prokash

-----Original Message-----
From: Moreira, Alberto [mailto:xxxxx@compuware.com]
Sent: Thursday, January 29, 2004 9:24 AM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

Well, at some point in time maybe people should schedule something in the
East Coast ? After all, it’s right in between the West Coast and Europe. :slight_smile:

Alberto.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of
xxxxx@3Dlabs.com
Sent: Thursday, January 29, 2004 12:14 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

I met with SOME of the members of this list in Seattle/Redmond at
Microsoft’s Driver Developers conference.

Of course, that assumes that you go to the conference, and there will of
course be a lot of “other people” there too.

Proximity is certainly a large problem. I know where some of the list
members are, I also guess where some are (due to time of e-mails arriving,
not a good indicator, but still), and there is a big distance! This is why
the developers conference, and other such events, are very useful, because
for those of us that are working for a company (rather than consultants who
“are” the company), it means that the company pays for the trip, and you get
to meet some nice people too (oh, the MS people are nice TOO, and they give
you all sorts of things, CD’s, papers and food/drink). That’s just my
thoughts…


Mats

-----Original Message-----
From: Chuck Batson [mailto:xxxxx@cbatson.com]
Sent: Thursday, January 29, 2004 4:45 PM
To: Windows System Software Devs Interest List
Subject: Re: [ntdev] Thread status in a lot of i/o traffic

Does this group have anything like an annual get-together or
something?
Perhaps an “ice cream social”? I realize there’s probably some
geographic proximity issues, but it could be fun to meet at a
different
location every year. Especially if they’re warm places and the
get-togethers are in the winter. An open bar would work well
too. But
I digress…

Chuck

----- Original Message -----
From: “Roddy, Mark”
> To: “Windows System Software Devs Interest List”
> Sent: Thursday, January 29, 2004 11:30 PM
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
>
> > Somebody please make alberto stop.
> >
> >
> > =====================
> > Mark Roddy
> >
> >
> > > -----Original Message-----
> > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > Sent: Thursday, January 29, 2004 11:22 AM
> > > To: Windows System Software Devs Interest List
> > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > >
> > >
> > > I wouldn’t use KeSetTimer because I’m not guaranteed that my DPC
> > > gets control in time. I would install my own Timer Interrupt
> > > routine and do it right there and then. Time-pressing chores
> > > should be done in the upper half, not in the bottom half !
> > >
> > > Alberto.
> > >
> > >
> > > -----Original Message-----
> > > From: xxxxx@lists.osr.com
> > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S.
> > > Shatskih
> > > Sent: Thursday, January 29, 2004 11:16 AM
> > > To: Windows System Software Devs Interest List
> > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > >
> > >
> > > > Basically, all this thred does is it sends dummy hearbeat
> > > cmds to the
> > > > device.
> > >
> > > Why not use KeSetTimer for this?
> > >
> > > > & I/O errors take place. Apparently the thread doesnt
> get the CPU
> > > attention,
> > > > as it should.
> > >
> > > It should not. NT provides no guarantees on amount of CPU
> > > attention for a particular thread, i.e. it is not a realtime OS.
> > >
> > > > Is there any workaround to this?
> > >
> > > No, but switching to KeSetTimer will make things much better (it
> > > supercedes any thread scheduling and can be delayed only by the
> > > long and huge DPCs and ISRs).
> > >
> > > Note: the commodity installs of NT OSes disable the DMA on ATAPI
> > > CD drives - due to compatibility with ancient IDE controllers I
> > > think. So, if your product is targeted to commodity market - then
> > > it will fail miserably while running in parallel with heavy CD-ROM
> > > I/O, since the ATAPI driver will do all data transport as REP INSW
> > > opcode inside the ISR (higher execution level then timer
> > > callbacks, not to say threads).
> > >
> > > Redesign your hardware and make its timing requirements more
> > > relaxing. This is the only way.
> > >
> > > Maxim Shatskih, Windows DDK MVP
> > > StorageCraft Corporation
> > > xxxxx@storagecraft.com
> > > http://www.storagecraft.com
> > >
> > >
> > > —
> > > Questions? First check the Kernel Driver FAQ at
> > > http://www.osronline.com/article.cfm?id=256
> > >
> > > You are currently subscribed to ntdev as:
> > > xxxxx@compuware.com To unsubscribe send a blank email to
> > > xxxxx@lists.osr.com
> > >
> > >
> > >
> > > The contents of this e-mail are intended for the named addressee
> > > only. It contains information that may be confidential. Unless you
> > > are the named addressee or an authorized designee, you may not
> > > copy or use it, or disclose it to anyone else. If you received it
> > > in error please notify us immediately and then destroy it.
> > >
> > >
> > > —
> > > Questions? First check the Kernel Driver FAQ at
> > > http://www.osronline.com/article.cfm?id=256
> > >
> > > You are currently subscribed to ntdev as: xxxxx@stratus.com
> > > To unsubscribe send a blank email to
> > > xxxxx@lists.osr.com
> > >
> >
> > —
> > Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
> >
> > You are currently subscribed to ntdev as: xxxxx@cbatson.com To
> > unsubscribe send a blank email to
> xxxxx@lists.osr.com
> >
>
>
> —
> Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@3dlabs.com
To unsubscribe send a blank email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@maxtor.com
To unsubscribe send a blank email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@storagecraft.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

Well, is there another way ? This one’s dead simple.

Alberto.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Jamey Kirby
Sent: Thursday, January 29, 2004 2:12 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

DOS is dead dude!

Jamey Kirby, Windows DDK MVP
StorageCraft Inc.
xxxxx@storagecraft.com
http://www.storagecraft.com

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com] On Behalf Of Moreira, Alberto
Sent: Thursday, January 29, 2004 8:22 AM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

I wouldn’t use KeSetTimer because I’m not guaranteed that my DPC gets
control in time. I would install my own Timer Interrupt routine and do it
right there and then. Time-pressing chores should be done in the upper half,
not in the bottom half !

Alberto.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S. Shatskih
Sent: Thursday, January 29, 2004 11:16 AM
To: Windows System Software Devs Interest List
Subject: Re: [ntdev] Thread status in a lot of i/o traffic

Basically, all this thred does is it sends dummy hearbeat cmds to the
device.

Why not use KeSetTimer for this?

& I/O errors take place. Apparently the thread doesnt get the CPU
attention,
as it should.

It should not. NT provides no guarantees on amount of CPU attention for a
particular thread, i.e. it is not a realtime OS.

Is there any workaround to this?

No, but switching to KeSetTimer will make things much better (it supercedes
any
thread scheduling and can be delayed only by the long and huge DPCs and
ISRs).

Note: the commodity installs of NT OSes disable the DMA on ATAPI CD drives -
due to compatibility with ancient IDE controllers I think. So, if your
product
is targeted to commodity market - then it will fail miserably while running
in
parallel with heavy CD-ROM I/O, since the ATAPI driver will do all data
transport as REP INSW opcode inside the ISR (higher execution level then
timer
callbacks, not to say threads).

Redesign your hardware and make its timing requirements more relaxing. This
is
the only way.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@storagecraft.com
To unsubscribe send a blank email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.

The FAQ for this is easy: use Linux and write it the way you want it to
work.

“Don Burn” wrote in message news:xxxxx@ntdev…
> Mark,
>
> I had already decided to ignore him since we can’t make him stop.
> Perhaps it is time for a FAQ entry on bypassing the OS?
>
> Don Burn (MVP, Windows DDK)
> Windows 2k/XP/2k3 Filesystem and Driver Consulting
>
> ----- Original Message -----
> From: “Roddy, Mark”
> To: “Windows System Software Devs Interest List”
> Sent: Thursday, January 29, 2004 11:30 AM
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
>
> > Somebody please make alberto stop.
> >
> >
> > =====================
> > Mark Roddy
> >
> >
> > > -----Original Message-----
> > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > Sent: Thursday, January 29, 2004 11:22 AM
> > > To: Windows System Software Devs Interest List
> > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > >
> > >
> > > I wouldn’t use KeSetTimer because I’m not guaranteed that my
> > > DPC gets control in time. I would install my own Timer
> > > Interrupt routine and do it right there and then.
> > > Time-pressing chores should be done in the upper half, not in
> > > the bottom half !
> > >
> > > Alberto.
> > >
> > >
> > > -----Original Message-----
> > > From: xxxxx@lists.osr.com
> > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim
> > > S. Shatskih
> > > Sent: Thursday, January 29, 2004 11:16 AM
> > > To: Windows System Software Devs Interest List
> > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > >
> > >
> > > > Basically, all this thred does is it sends dummy hearbeat
> > > cmds to the
> > > > device.
> > >
> > > Why not use KeSetTimer for this?
> > >
> > > > & I/O errors take place. Apparently the thread doesnt get the
CPU
> > > attention,
> > > > as it should.
> > >
> > > It should not. NT provides no guarantees on amount of CPU
> > > attention for a particular thread, i.e. it is not a realtime OS.
> > >
> > > > Is there any workaround to this?
> > >
> > > No, but switching to KeSetTimer will make things much better
> > > (it supercedes any thread scheduling and can be delayed only
> > > by the long and huge DPCs and ISRs).
> > >
> > > Note: the commodity installs of NT OSes disable the DMA on
> > > ATAPI CD drives - due to compatibility with ancient IDE
> > > controllers I think. So, if your product is targeted to
> > > commodity market - then it will fail miserably while running
> > > in parallel with heavy CD-ROM I/O, since the ATAPI driver
> > > will do all data transport as REP INSW opcode inside the ISR
> > > (higher execution level then timer callbacks, not to say threads).
> > >
> > > Redesign your hardware and make its timing requirements more
> > > relaxing. This is the only way.
> > >
> > > Maxim Shatskih, Windows DDK MVP
> > > StorageCraft Corporation
> > > xxxxx@storagecraft.com
> > > http://www.storagecraft.com
> > >
> > >
> > > —
> > > Questions? First check the Kernel Driver FAQ at
> > > http://www.osronline.com/article.cfm?id=256
> > >
> > > You are currently subscribed to ntdev as:
> > > xxxxx@compuware.com To unsubscribe send a blank
> > > email to xxxxx@lists.osr.com
> > >
> > >
> > >
> > > The contents of this e-mail are intended for the named
> > > addressee only. It contains information that may be
> > > confidential. Unless you are the named addressee or an
> > > authorized designee, you may not copy or use it, or disclose
> > > it to anyone else. If you received it in error please notify
> > > us immediately and then destroy it.
> > >
> > >
> > > —
> > > Questions? First check the Kernel Driver FAQ at
> > > http://www.osronline.com/article.cfm?id=256
> > >
> > > You are currently subscribed to ntdev as:
> > > xxxxx@stratus.com To unsubscribe send a blank email to
> > > xxxxx@lists.osr.com
> > >
> >
> > —
> > Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
> >
> > You are currently subscribed to ntdev as: xxxxx@acm.org
> > To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>

Hawaii.

“Moreira, Alberto” wrote in message
news:xxxxx@ntdev…
> In that case, I vote for St. Moritz.
>
> Alberto.
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com]On Behalf Of Sinha, Prokash
> Sent: Thursday, January 29, 2004 12:54 PM
> To: Windows System Software Devs Interest List
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
>
> Well the way I see it is pretty round. Those coming from the other
side of
> the
> California river ( well if my metric is Mega-meter) might not like to
make
> it to east coast…
>
> How about silicon valley ? Or may be we should pick about 10 places
around
> the world, and do a drawing so noone would feel about it …
>
> -prokash
>
> -----Original Message-----
> From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> Sent: Thursday, January 29, 2004 9:24 AM
> To: Windows System Software Devs Interest List
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
>
>
> Well, at some point in time maybe people should schedule something in
the
> East Coast ? After all, it’s right in between the West Coast and
Europe. :slight_smile:
>
> Alberto.
>
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com]On Behalf Of
> xxxxx@3Dlabs.com
> Sent: Thursday, January 29, 2004 12:14 PM
> To: Windows System Software Devs Interest List
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
>
> I met with SOME of the members of this list in Seattle/Redmond at
> Microsoft’s Driver Developers conference.
>
> Of course, that assumes that you go to the conference, and there will
of
> course be a lot of “other people” there too.
>
> Proximity is certainly a large problem. I know where some of the list
> members are, I also guess where some are (due to time of e-mails
arriving,
> not a good indicator, but still), and there is a big distance! This is
why
> the developers conference, and other such events, are very useful,
because
> for those of us that are working for a company (rather than
consultants who
> “are” the company), it means that the company pays for the trip, and
you get
> to meet some nice people too (oh, the MS people are nice TOO, and they
give
> you all sorts of things, CD’s, papers and food/drink). That’s just my
> thoughts…
>
> –
> Mats
>
> > -----Original Message-----
> > From: Chuck Batson [mailto:xxxxx@cbatson.com]
> > Sent: Thursday, January 29, 2004 4:45 PM
> > To: Windows System Software Devs Interest List
> > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> >
> >
> > Does this group have anything like an annual get-together or
> > something?
> > Perhaps an “ice cream social”? I realize there’s probably some
> > geographic proximity issues, but it could be fun to meet at a
> > different
> > location every year. Especially if they’re warm places and the
> > get-togethers are in the winter. An open bar would work well
> > too. But
> > I digress…
> >
> > Chuck
> >
> > ----- Original Message -----
> > From: “Roddy, Mark”
> > To: “Windows System Software Devs Interest List”

> > Sent: Thursday, January 29, 2004 11:30 PM
> > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> >
> >
> > > Somebody please make alberto stop.
> > >
> > >
> > > =====================
> > > Mark Roddy
> > >
> > >
> > > > -----Original Message-----
> > > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > > Sent: Thursday, January 29, 2004 11:22 AM
> > > > To: Windows System Software Devs Interest List
> > > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > > >
> > > >
> > > > I wouldn’t use KeSetTimer because I’m not guaranteed that my DPC
> > > > gets control in time. I would install my own Timer Interrupt
> > > > routine and do it right there and then. Time-pressing chores
> > > > should be done in the upper half, not in the bottom half !
> > > >
> > > > Alberto.
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: xxxxx@lists.osr.com
> > > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S.
> > > > Shatskih
> > > > Sent: Thursday, January 29, 2004 11:16 AM
> > > > To: Windows System Software Devs Interest List
> > > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > > >
> > > >
> > > > > Basically, all this thred does is it sends dummy hearbeat
> > > > cmds to the
> > > > > device.
> > > >
> > > > Why not use KeSetTimer for this?
> > > >
> > > > > & I/O errors take place. Apparently the thread doesnt
> > get the CPU
> > > > attention,
> > > > > as it should.
> > > >
> > > > It should not. NT provides no guarantees on amount of CPU
> > > > attention for a particular thread, i.e. it is not a realtime OS.
> > > >
> > > > > Is there any workaround to this?
> > > >
> > > > No, but switching to KeSetTimer will make things much better (it
> > > > supercedes any thread scheduling and can be delayed only by the
> > > > long and huge DPCs and ISRs).
> > > >
> > > > Note: the commodity installs of NT OSes disable the DMA on ATAPI
> > > > CD drives - due to compatibility with ancient IDE controllers I
> > > > think. So, if your product is targeted to commodity market -
then
> > > > it will fail miserably while running in parallel with heavy
CD-ROM
> > > > I/O, since the ATAPI driver will do all data transport as REP
INSW
> > > > opcode inside the ISR (higher execution level then timer
> > > > callbacks, not to say threads).
> > > >
> > > > Redesign your hardware and make its timing requirements more
> > > > relaxing. This is the only way.
> > > >
> > > > Maxim Shatskih, Windows DDK MVP
> > > > StorageCraft Corporation
> > > > xxxxx@storagecraft.com
> > > > http://www.storagecraft.com
> > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as:
> > > > xxxxx@compuware.com To unsubscribe send a blank email
to
> > > > xxxxx@lists.osr.com
> > > >
> > > >
> > > >
> > > > The contents of this e-mail are intended for the named addressee
> > > > only. It contains information that may be confidential. Unless
you
> > > > are the named addressee or an authorized designee, you may not
> > > > copy or use it, or disclose it to anyone else. If you received
it
> > > > in error please notify us immediately and then destroy it.
> > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as: xxxxx@stratus.com
> > > > To unsubscribe send a blank email to
> > > > xxxxx@lists.osr.com
> > > >
> > >
> > > —
> > > Questions? First check the Kernel Driver FAQ at
> > http://www.osronline.com/article.cfm?id=256
> > >
> > > You are currently subscribed to ntdev as: xxxxx@cbatson.com
To
> > > unsubscribe send a blank email to
> > xxxxx@lists.osr.com
> > >
> >
> >
> > —
> > Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@3dlabs.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
xxxxx@compuware.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
>
> The contents of this e-mail are intended for the named addressee only.
It
> contains information that may be confidential. Unless you are the
named
> addressee or an authorized designee, you may not copy or use it, or
disclose
> it to anyone else. If you received it in error please notify us
immediately
> and then destroy it.
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@maxtor.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
xxxxx@compuware.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
>
> The contents of this e-mail are intended for the named addressee only.
It
> contains information that may be confidential. Unless you are the
named
> addressee or an authorized designee, you may not copy or use it, or
disclose
> it to anyone else. If you received it in error please notify us
immediately
> and then destroy it.
>
>

> ----------

From: xxxxx@stratus.com[SMTP:xxxxx@stratus.com]
Reply To: xxxxx@lists.osr.com
Sent: Thursday, January 29, 2004 5:30 PM
To: xxxxx@lists.osr.com
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

Somebody please make alberto stop.

d:\>net stop alberto
System error 1060 has occurred.

The specified service does not exist as an installed service.

Best regards,

Michal Vodicka
STMicroelectronics Design and Application s.r.o.
[michal.vodicka@st.com, http:://www.st.com]

I hope Linus is sane enough to never include such dirt to the official
kernel.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

----- Original Message -----
From: “David J. Craig”
Newsgroups: ntdev
To: “Windows System Software Devs Interest List”
Sent: Thursday, January 29, 2004 10:53 PM
Subject: Re:[ntdev] Thread status in a lot of i/o traffic

> The FAQ for this is easy: use Linux and write it the way you want it to
> work.
>
> “Don Burn” wrote in message news:xxxxx@ntdev…
> > Mark,
> >
> > I had already decided to ignore him since we can’t make him stop.
> > Perhaps it is time for a FAQ entry on bypassing the OS?
> >
> > Don Burn (MVP, Windows DDK)
> > Windows 2k/XP/2k3 Filesystem and Driver Consulting
> >
> > ----- Original Message -----
> > From: “Roddy, Mark”
> > To: “Windows System Software Devs Interest List”
> > Sent: Thursday, January 29, 2004 11:30 AM
> > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> >
> >
> > > Somebody please make alberto stop.
> > >
> > >
> > > =====================
> > > Mark Roddy
> > >
> > >
> > > > -----Original Message-----
> > > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > > Sent: Thursday, January 29, 2004 11:22 AM
> > > > To: Windows System Software Devs Interest List
> > > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > > >
> > > >
> > > > I wouldn’t use KeSetTimer because I’m not guaranteed that my
> > > > DPC gets control in time. I would install my own Timer
> > > > Interrupt routine and do it right there and then.
> > > > Time-pressing chores should be done in the upper half, not in
> > > > the bottom half !
> > > >
> > > > Alberto.
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: xxxxx@lists.osr.com
> > > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim
> > > > S. Shatskih
> > > > Sent: Thursday, January 29, 2004 11:16 AM
> > > > To: Windows System Software Devs Interest List
> > > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > > >
> > > >
> > > > > Basically, all this thred does is it sends dummy hearbeat
> > > > cmds to the
> > > > > device.
> > > >
> > > > Why not use KeSetTimer for this?
> > > >
> > > > > & I/O errors take place. Apparently the thread doesnt get the
> CPU
> > > > attention,
> > > > > as it should.
> > > >
> > > > It should not. NT provides no guarantees on amount of CPU
> > > > attention for a particular thread, i.e. it is not a realtime OS.
> > > >
> > > > > Is there any workaround to this?
> > > >
> > > > No, but switching to KeSetTimer will make things much better
> > > > (it supercedes any thread scheduling and can be delayed only
> > > > by the long and huge DPCs and ISRs).
> > > >
> > > > Note: the commodity installs of NT OSes disable the DMA on
> > > > ATAPI CD drives - due to compatibility with ancient IDE
> > > > controllers I think. So, if your product is targeted to
> > > > commodity market - then it will fail miserably while running
> > > > in parallel with heavy CD-ROM I/O, since the ATAPI driver
> > > > will do all data transport as REP INSW opcode inside the ISR
> > > > (higher execution level then timer callbacks, not to say threads).
> > > >
> > > > Redesign your hardware and make its timing requirements more
> > > > relaxing. This is the only way.
> > > >
> > > > Maxim Shatskih, Windows DDK MVP
> > > > StorageCraft Corporation
> > > > xxxxx@storagecraft.com
> > > > http://www.storagecraft.com
> > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as:
> > > > xxxxx@compuware.com To unsubscribe send a blank
> > > > email to xxxxx@lists.osr.com
> > > >
> > > >
> > > >
> > > > The contents of this e-mail are intended for the named
> > > > addressee only. It contains information that may be
> > > > confidential. Unless you are the named addressee or an
> > > > authorized designee, you may not copy or use it, or disclose
> > > > it to anyone else. If you received it in error please notify
> > > > us immediately and then destroy it.
> > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as:
> > > > xxxxx@stratus.com To unsubscribe send a blank email to
> > > > xxxxx@lists.osr.com
> > > >
> > >
> > > —
> > > Questions? First check the Kernel Driver FAQ at
> > http://www.osronline.com/article.cfm?id=256
> > >
> > > You are currently subscribed to ntdev as: xxxxx@acm.org
> > > To unsubscribe send a blank email to xxxxx@lists.osr.com
> >
> >
>
>
>
> —
> Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@storagecraft.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com

Too late - Linux processes can be enqueued into a wait queue as
interruptible or uninterruptible. Interruptible processes react to signals,
or to interrupts such as timers expiring. Because, hey, one might want to be
able to terminate one’s threads. :wink:

Alberto.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S. Shatskih
Sent: Thursday, January 29, 2004 3:54 PM
To: Windows System Software Devs Interest List
Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic

I hope Linus is sane enough to never include such dirt to the official
kernel.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

----- Original Message -----
From: “David J. Craig”
Newsgroups: ntdev
To: “Windows System Software Devs Interest List”
Sent: Thursday, January 29, 2004 10:53 PM
Subject: Re:[ntdev] Thread status in a lot of i/o traffic

> The FAQ for this is easy: use Linux and write it the way you want it to
> work.
>
> “Don Burn” wrote in message news:xxxxx@ntdev…
> > Mark,
> >
> > I had already decided to ignore him since we can’t make him stop.
> > Perhaps it is time for a FAQ entry on bypassing the OS?
> >
> > Don Burn (MVP, Windows DDK)
> > Windows 2k/XP/2k3 Filesystem and Driver Consulting
> >
> > ----- Original Message -----
> > From: “Roddy, Mark”
> > To: “Windows System Software Devs Interest List”
> > Sent: Thursday, January 29, 2004 11:30 AM
> > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> >
> >
> > > Somebody please make alberto stop.
> > >
> > >
> > > =====================
> > > Mark Roddy
> > >
> > >
> > > > -----Original Message-----
> > > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > > Sent: Thursday, January 29, 2004 11:22 AM
> > > > To: Windows System Software Devs Interest List
> > > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > > >
> > > >
> > > > I wouldn’t use KeSetTimer because I’m not guaranteed that my
> > > > DPC gets control in time. I would install my own Timer
> > > > Interrupt routine and do it right there and then.
> > > > Time-pressing chores should be done in the upper half, not in
> > > > the bottom half !
> > > >
> > > > Alberto.
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: xxxxx@lists.osr.com
> > > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim
> > > > S. Shatskih
> > > > Sent: Thursday, January 29, 2004 11:16 AM
> > > > To: Windows System Software Devs Interest List
> > > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > > >
> > > >
> > > > > Basically, all this thred does is it sends dummy hearbeat
> > > > cmds to the
> > > > > device.
> > > >
> > > > Why not use KeSetTimer for this?
> > > >
> > > > > & I/O errors take place. Apparently the thread doesnt get the
> CPU
> > > > attention,
> > > > > as it should.
> > > >
> > > > It should not. NT provides no guarantees on amount of CPU
> > > > attention for a particular thread, i.e. it is not a realtime OS.
> > > >
> > > > > Is there any workaround to this?
> > > >
> > > > No, but switching to KeSetTimer will make things much better
> > > > (it supercedes any thread scheduling and can be delayed only
> > > > by the long and huge DPCs and ISRs).
> > > >
> > > > Note: the commodity installs of NT OSes disable the DMA on
> > > > ATAPI CD drives - due to compatibility with ancient IDE
> > > > controllers I think. So, if your product is targeted to
> > > > commodity market - then it will fail miserably while running
> > > > in parallel with heavy CD-ROM I/O, since the ATAPI driver
> > > > will do all data transport as REP INSW opcode inside the ISR
> > > > (higher execution level then timer callbacks, not to say threads).
> > > >
> > > > Redesign your hardware and make its timing requirements more
> > > > relaxing. This is the only way.
> > > >
> > > > Maxim Shatskih, Windows DDK MVP
> > > > StorageCraft Corporation
> > > > xxxxx@storagecraft.com
> > > > http://www.storagecraft.com
> > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as:
> > > > xxxxx@compuware.com To unsubscribe send a blank
> > > > email to xxxxx@lists.osr.com
> > > >
> > > >
> > > >
> > > > The contents of this e-mail are intended for the named
> > > > addressee only. It contains information that may be
> > > > confidential. Unless you are the named addressee or an
> > > > authorized designee, you may not copy or use it, or disclose
> > > > it to anyone else. If you received it in error please notify
> > > > us immediately and then destroy it.
> > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as:
> > > > xxxxx@stratus.com To unsubscribe send a blank email to
> > > > xxxxx@lists.osr.com
> > > >
> > >
> > > —
> > > Questions? First check the Kernel Driver FAQ at
> > http://www.osronline.com/article.cfm?id=256
> > >
> > > You are currently subscribed to ntdev as: xxxxx@acm.org
> > > To unsubscribe send a blank email to xxxxx@lists.osr.com
> >
> >
>
>
>
> —
> Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@storagecraft.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.

Well thanks for trying.

=====================
Mark Roddy

-----Original Message-----
From: Michal Vodicka [mailto:xxxxx@veridicom.cz.nospam]
Sent: Thursday, January 29, 2004 3:47 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

> ----------
> From: xxxxx@stratus.com[SMTP:xxxxx@stratus.com]
> Reply To: xxxxx@lists.osr.com
> Sent: Thursday, January 29, 2004 5:30 PM
> To: xxxxx@lists.osr.com
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
> Somebody please make alberto stop.
>
d:\>net stop alberto
System error 1060 has occurred.

The specified service does not exist as an installed service.

Best regards,

Michal Vodicka
STMicroelectronics Design and Application s.r.o.
[michal.vodicka@st.com, http:://www.st.com]


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as:
xxxxx@stratus.com To unsubscribe send a blank email to
xxxxx@lists.osr.com

This is already in NT as Alertable/NonAlertable waits.

BTW - Linux’s wait queue (and also select()/poll()) implementation is
absolutely brain damaged, it only shows that the Linux co-authors never ever
heard on encapsulation.

The “dirt” is to hook the timer IRQ, not to terminate threads. No commodity
(non-lab) product must ever do this in any OS.

Lab products are other song. People will tolerate SoftICE being
incompatible with some video card (OK, lab machine can have a requirement of
this particular card, like CE/PC imposes) - but they will not tolerate the
same in some mass-market dongle’s driver.

If the mass-market hardware cannot run in parallel with heavy loaded CD
drive running in its default mode, since REP INSW in ATAPI driver’s ISR will
time the hardware out - then the hardware is broken. Same on serial ports, who
are nearly as chatty as the CD running in PIO mode.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

----- Original Message -----
From: “Moreira, Alberto”
To: “Windows System Software Devs Interest List”
Sent: Friday, January 30, 2004 12:02 AM
Subject: RE: Re:[ntdev] Thread status in a lot of i/o traffic

> Too late - Linux processes can be enqueued into a wait queue as
> interruptible or uninterruptible. Interruptible processes react to signals,
> or to interrupts such as timers expiring. Because, hey, one might want to be
> able to terminate one’s threads. :wink:
>
> Alberto.
>
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S. Shatskih
> Sent: Thursday, January 29, 2004 3:54 PM
> To: Windows System Software Devs Interest List
> Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic
>
>
> I hope Linus is sane enough to never include such dirt to the official
> kernel.
>
> Maxim Shatskih, Windows DDK MVP
> StorageCraft Corporation
> xxxxx@storagecraft.com
> http://www.storagecraft.com
>
> ----- Original Message -----
> From: “David J. Craig”
> Newsgroups: ntdev
> To: “Windows System Software Devs Interest List”
> Sent: Thursday, January 29, 2004 10:53 PM
> Subject: Re:[ntdev] Thread status in a lot of i/o traffic
>
>
> > The FAQ for this is easy: use Linux and write it the way you want it to
> > work.
> >
> > “Don Burn” wrote in message news:xxxxx@ntdev…
> > > Mark,
> > >
> > > I had already decided to ignore him since we can’t make him stop.
> > > Perhaps it is time for a FAQ entry on bypassing the OS?
> > >
> > > Don Burn (MVP, Windows DDK)
> > > Windows 2k/XP/2k3 Filesystem and Driver Consulting
> > >
> > > ----- Original Message -----
> > > From: “Roddy, Mark”
> > > To: “Windows System Software Devs Interest List”
> > > Sent: Thursday, January 29, 2004 11:30 AM
> > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > >
> > >
> > > > Somebody please make alberto stop.
> > > >
> > > >
> > > > =====================
> > > > Mark Roddy
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > > > Sent: Thursday, January 29, 2004 11:22 AM
> > > > > To: Windows System Software Devs Interest List
> > > > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > > > >
> > > > >
> > > > > I wouldn’t use KeSetTimer because I’m not guaranteed that my
> > > > > DPC gets control in time. I would install my own Timer
> > > > > Interrupt routine and do it right there and then.
> > > > > Time-pressing chores should be done in the upper half, not in
> > > > > the bottom half !
> > > > >
> > > > > Alberto.
> > > > >
> > > > >
> > > > > -----Original Message-----
> > > > > From: xxxxx@lists.osr.com
> > > > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim
> > > > > S. Shatskih
> > > > > Sent: Thursday, January 29, 2004 11:16 AM
> > > > > To: Windows System Software Devs Interest List
> > > > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > > > >
> > > > >
> > > > > > Basically, all this thred does is it sends dummy hearbeat
> > > > > cmds to the
> > > > > > device.
> > > > >
> > > > > Why not use KeSetTimer for this?
> > > > >
> > > > > > & I/O errors take place. Apparently the thread doesnt get the
> > CPU
> > > > > attention,
> > > > > > as it should.
> > > > >
> > > > > It should not. NT provides no guarantees on amount of CPU
> > > > > attention for a particular thread, i.e. it is not a realtime OS.
> > > > >
> > > > > > Is there any workaround to this?
> > > > >
> > > > > No, but switching to KeSetTimer will make things much better
> > > > > (it supercedes any thread scheduling and can be delayed only
> > > > > by the long and huge DPCs and ISRs).
> > > > >
> > > > > Note: the commodity installs of NT OSes disable the DMA on
> > > > > ATAPI CD drives - due to compatibility with ancient IDE
> > > > > controllers I think. So, if your product is targeted to
> > > > > commodity market - then it will fail miserably while running
> > > > > in parallel with heavy CD-ROM I/O, since the ATAPI driver
> > > > > will do all data transport as REP INSW opcode inside the ISR
> > > > > (higher execution level then timer callbacks, not to say threads).
> > > > >
> > > > > Redesign your hardware and make its timing requirements more
> > > > > relaxing. This is the only way.
> > > > >
> > > > > Maxim Shatskih, Windows DDK MVP
> > > > > StorageCraft Corporation
> > > > > xxxxx@storagecraft.com
> > > > > http://www.storagecraft.com
> > > > >
> > > > >
> > > > > —
> > > > > Questions? First check the Kernel Driver FAQ at
> > > > > http://www.osronline.com/article.cfm?id=256
> > > > >
> > > > > You are currently subscribed to ntdev as:
> > > > > xxxxx@compuware.com To unsubscribe send a blank
> > > > > email to xxxxx@lists.osr.com
> > > > >
> > > > >
> > > > >
> > > > > The contents of this e-mail are intended for the named
> > > > > addressee only. It contains information that may be
> > > > > confidential. Unless you are the named addressee or an
> > > > > authorized designee, you may not copy or use it, or disclose
> > > > > it to anyone else. If you received it in error please notify
> > > > > us immediately and then destroy it.
> > > > >
> > > > >
> > > > > —
> > > > > Questions? First check the Kernel Driver FAQ at
> > > > > http://www.osronline.com/article.cfm?id=256
> > > > >
> > > > > You are currently subscribed to ntdev as:
> > > > > xxxxx@stratus.com To unsubscribe send a blank email to
> > > > > xxxxx@lists.osr.com
> > > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as: xxxxx@acm.org
> > > > To unsubscribe send a blank email to xxxxx@lists.osr.com
> > >
> > >
> >
> >
> >
> > —
> > Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
> >
> > You are currently subscribed to ntdev as: xxxxx@storagecraft.com
> > To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@compuware.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
>
> The contents of this e-mail are intended for the named addressee only. It
> contains information that may be confidential. Unless you are the named
> addressee or an authorized designee, you may not copy or use it, or disclose
> it to anyone else. If you received it in error please notify us immediately
> and then destroy it.
>
>
> —
> Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@storagecraft.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com

Well, pls don’t throw that many stone at his head :-). He probably have only
one :-).
Think about self-resilence!!!, if a process creates bunch of threads and
waits for shit, is not there someway, kernel or someone come and clean it,
just like garbage collection … So once that path makes sense, then the
next step would be to have some kernel interface so that kernel-component
designer can have a choice to kill those ever waiting threads.

I am thinking about somewhat self-correcting behavior. Oh, well, pls don’t
throw stone at me, I’m loosing hair quite fast, so there is no cushion to
absorb the thrust :-).

-prokash

-----Original Message-----
From: Maxim S. Shatskih [mailto:xxxxx@storagecraft.com]
Sent: Thursday, January 29, 2004 1:58 PM
To: Windows System Software Devs Interest List
Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic

This is already in NT as Alertable/NonAlertable waits.

BTW - Linux’s wait queue (and also select()/poll()) implementation is
absolutely brain damaged, it only shows that the Linux co-authors never ever
heard on encapsulation.

The “dirt” is to hook the timer IRQ, not to terminate threads. No
commodity
(non-lab) product must ever do this in any OS.

Lab products are other song. People will tolerate SoftICE being
incompatible with some video card (OK, lab machine can have a requirement of
this particular card, like CE/PC imposes) - but they will not tolerate the
same in some mass-market dongle’s driver.

If the mass-market hardware cannot run in parallel with heavy loaded CD
drive running in its default mode, since REP INSW in ATAPI driver’s ISR will
time the hardware out - then the hardware is broken. Same on serial ports,
who are nearly as chatty as the CD running in PIO mode.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

----- Original Message -----
From: “Moreira, Alberto”
To: “Windows System Software Devs Interest List”
Sent: Friday, January 30, 2004 12:02 AM
Subject: RE: Re:[ntdev] Thread status in a lot of i/o traffic

> Too late - Linux processes can be enqueued into a wait queue as
> interruptible or uninterruptible. Interruptible processes react to
> signals, or to interrupts such as timers expiring. Because, hey, one
> might want to be able to terminate one’s threads. :wink:
>
> Alberto.
>
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S.
> Shatskih
> Sent: Thursday, January 29, 2004 3:54 PM
> To: Windows System Software Devs Interest List
> Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic
>
>
> I hope Linus is sane enough to never include such dirt to the
> official kernel.
>
> Maxim Shatskih, Windows DDK MVP
> StorageCraft Corporation
> xxxxx@storagecraft.com
> http://www.storagecraft.com
>
> ----- Original Message -----
> From: “David J. Craig”
> Newsgroups: ntdev
> To: “Windows System Software Devs Interest List”
> Sent: Thursday, January 29, 2004 10:53 PM
> Subject: Re:[ntdev] Thread status in a lot of i/o traffic
>
>
> > The FAQ for this is easy: use Linux and write it the way you want
> > it to work.
> >
> > “Don Burn” wrote in message news:xxxxx@ntdev…
> > > Mark,
> > >
> > > I had already decided to ignore him since we can’t make him
> > > stop. Perhaps it is time for a FAQ entry on bypassing the OS?
> > >
> > > Don Burn (MVP, Windows DDK)
> > > Windows 2k/XP/2k3 Filesystem and Driver Consulting
> > >
> > > ----- Original Message -----
> > > From: “Roddy, Mark”
> > > To: “Windows System Software Devs Interest List”
> > > Sent: Thursday, January 29, 2004 11:30 AM
> > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > >
> > >
> > > > Somebody please make alberto stop.
> > > >
> > > >
> > > > =====================
> > > > Mark Roddy
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > > > Sent: Thursday, January 29, 2004 11:22 AM
> > > > > To: Windows System Software Devs Interest List
> > > > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > > > >
> > > > >
> > > > > I wouldn’t use KeSetTimer because I’m not guaranteed that my
> > > > > DPC gets control in time. I would install my own Timer
> > > > > Interrupt routine and do it right there and then.
> > > > > Time-pressing chores should be done in the upper half, not in
> > > > > the bottom half !
> > > > >
> > > > > Alberto.
> > > > >
> > > > >
> > > > > -----Original Message-----
> > > > > From: xxxxx@lists.osr.com
> > > > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S.
> > > > > Shatskih
> > > > > Sent: Thursday, January 29, 2004 11:16 AM
> > > > > To: Windows System Software Devs Interest List
> > > > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > > > >
> > > > >
> > > > > > Basically, all this thred does is it sends dummy hearbeat
> > > > > cmds to the
> > > > > > device.
> > > > >
> > > > > Why not use KeSetTimer for this?
> > > > >
> > > > > > & I/O errors take place. Apparently the thread doesnt get
> > > > > > the
> > CPU
> > > > > attention,
> > > > > > as it should.
> > > > >
> > > > > It should not. NT provides no guarantees on amount of CPU
> > > > > attention for a particular thread, i.e. it is not a realtime
> > > > > OS.
> > > > >
> > > > > > Is there any workaround to this?
> > > > >
> > > > > No, but switching to KeSetTimer will make things much better
> > > > > (it supercedes any thread scheduling and can be delayed only
> > > > > by the long and huge DPCs and ISRs).
> > > > >
> > > > > Note: the commodity installs of NT OSes disable the DMA on
> > > > > ATAPI CD drives - due to compatibility with ancient IDE
> > > > > controllers I think. So, if your product is targeted to
> > > > > commodity market - then it will fail miserably while running
> > > > > in parallel with heavy CD-ROM I/O, since the ATAPI driver will
> > > > > do all data transport as REP INSW opcode inside the ISR
> > > > > (higher execution level then timer callbacks, not to say
> > > > > threads).
> > > > >
> > > > > Redesign your hardware and make its timing requirements more
> > > > > relaxing. This is the only way.
> > > > >
> > > > > Maxim Shatskih, Windows DDK MVP
> > > > > StorageCraft Corporation
> > > > > xxxxx@storagecraft.com
> > > > > http://www.storagecraft.com
> > > > >
> > > > >
> > > > > —
> > > > > Questions? First check the Kernel Driver FAQ at
> > > > > http://www.osronline.com/article.cfm?id=256
> > > > >
> > > > > You are currently subscribed to ntdev as:
> > > > > xxxxx@compuware.com To unsubscribe send a blank
> > > > > email to xxxxx@lists.osr.com
> > > > >
> > > > >
> > > > >
> > > > > The contents of this e-mail are intended for the named
> > > > > addressee only. It contains information that may be
> > > > > confidential. Unless you are the named addressee or an
> > > > > authorized designee, you may not copy or use it, or disclose
> > > > > it to anyone else. If you received it in error please notify
> > > > > us immediately and then destroy it.
> > > > >
> > > > >
> > > > > —
> > > > > Questions? First check the Kernel Driver FAQ at
> > > > > http://www.osronline.com/article.cfm?id=256
> > > > >
> > > > > You are currently subscribed to ntdev as:
> > > > > xxxxx@stratus.com To unsubscribe send a blank email to
> > > > > xxxxx@lists.osr.com
> > > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as: xxxxx@acm.org To
> > > > unsubscribe send a blank email to
> > > > xxxxx@lists.osr.com
> > >
> > >
> >
> >
> >
> > —
> > Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
> >
> > You are currently subscribed to ntdev as: xxxxx@storagecraft.com To
> > unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as:
> xxxxx@compuware.com To unsubscribe send a blank email to
> xxxxx@lists.osr.com
>
>
>
> The contents of this e-mail are intended for the named addressee only.
> It contains information that may be confidential. Unless you are the
> named addressee or an authorized designee, you may not copy or use it,
> or disclose it to anyone else. If you received it in error please
> notify us immediately and then destroy it.
>
>
> —
> Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@storagecraft.com To
> unsubscribe send a blank email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@maxtor.com To
unsubscribe send a blank email to xxxxx@lists.osr.com

> next step would be to have some kernel interface so that kernel-component

designer can have a choice to kill those ever waiting threads.

Admin can do this by KILL.EXE command.

Design the multithreaded apps properly, and there will be no need in such.
Code the drivers properly (IRP cancellation), and there will be no need in
such.

I am thinking about somewhat self-correcting behavior.

Well, I can dream about Win32’s (fantastic) AlertThread function, which will
turn the thread to a state where all waits will immediately return
ERROR_ALERTED and do not block. This ERROR_ALERTED would be a signal for a
thread to quit ASAP, doing necessary cleanup in the process.

Maybe this would be a good feature.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

> Well, I can dream about Win32’s (fantastic) AlertThread function,
which will

turn the thread to a state where all waits will immediately return
ERROR_ALERTED and do not block. This ERROR_ALERTED would be a signal
for a
thread to quit ASAP, doing necessary cleanup in the process.

Great idea! That would even be useful in user mode.

What I do, in the mean time, is achieve basically the same thing by
using a wrapper function for waiting that always includes an abort event
in the list of wait objects. Using a wrapper function is a good idea in
this case, because I’m less likely to absent-mindedly forget to include
the abort event when waiting if my wrapper does it automatically for me
every time.

Chuck

> What I do, in the mean time, is achieve basically the same thing by

using a wrapper function for waiting that always includes an abort event
in the list of wait objects.

Zero timeout is also suitable.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

> > What I do, in the mean time, is achieve basically the same thing by

> using a wrapper function for waiting that always includes an abort
event
> in the list of wait objects.

Zero timeout is also suitable.

That will work, but is less efficient than waiting on a sync. object.

Chuck

Yes, you got a point there, use Kill.exe by an admin :-).

I wish I could say the same thing to millions of drivers, that automatic
transmission is not required, use the stick shift.
I bet the next morning you might see a head line like this in major news
paper –
“AN EXCURCIATINGLY PATHETIC PERSON WAS BEATEN TO DEATH BY THOUSANDS OF
PEOPLE FOR HIS BAD
RECOMMENDATION ABOUT TRANSMISSION OF CARS IN THE 100 YEARS OF HISTORY” :-).

Most schools have 30 or 40 or 50 computers, and this could very well be used
as a DOS.

Personally, and wishfully, there should not be the admin concept ( not as
much the way we think) for comodity PCs. May be
someday the admin part would be semi-automatic, just a thought !!!. So
ignore it !

-prokash

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S. Shatskih
Sent: Thursday, January 29, 2004 3:09 PM
To: Windows System Software Devs Interest List
Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic

next step would be to have some kernel interface so that kernel-component
designer can have a choice to kill those ever waiting threads.

Admin can do this by KILL.EXE command.

Design the multithreaded apps properly, and there will be no need in such.
Code the drivers properly (IRP cancellation), and there will be no need in
such.

I am thinking about somewhat self-correcting behavior.

Well, I can dream about Win32’s (fantastic) AlertThread function, which will
turn the thread to a state where all waits will immediately return
ERROR_ALERTED and do not block. This ERROR_ALERTED would be a signal for a
thread to quit ASAP, doing necessary cleanup in the process.

Maybe this would be a good feature.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@garlic.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

I’m not sure what you’re trying to say here, but any user with
sufficient permission (say, perhaps, the user that created the process)
can terminate a process.

Chuck

----- Original Message -----
From: “Prokash Sinha”
To: “Windows System Software Devs Interest List”
Sent: Friday, January 30, 2004 10:22 PM
Subject: RE: Re:[ntdev] Thread status in a lot of i/o traffic

> Yes, you got a point there, use Kill.exe by an admin :-).
>
> I wish I could say the same thing to millions of drivers, that
automatic
> transmission is not required, use the stick shift.
> I bet the next morning you might see a head line like this in major
news
> paper –
> “AN EXCURCIATINGLY PATHETIC PERSON WAS BEATEN TO DEATH BY THOUSANDS OF
> PEOPLE FOR HIS BAD
> RECOMMENDATION ABOUT TRANSMISSION OF CARS IN THE 100 YEARS OF HISTORY”
:-).
>
> Most schools have 30 or 40 or 50 computers, and this could very well
be used
> as a DOS.
>
> Personally, and wishfully, there should not be the admin concept ( not
as
> much the way we think) for comodity PCs. May be
> someday the admin part would be semi-automatic, just a thought !!!. So
> ignore it !
>
> -prokash
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S.
Shatskih
> Sent: Thursday, January 29, 2004 3:09 PM
> To: Windows System Software Devs Interest List
> Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic
>
>
> > next step would be to have some kernel interface so that
kernel-component
> > designer can have a choice to kill those ever waiting threads.
>
> Admin can do this by KILL.EXE command.
>
> Design the multithreaded apps properly, and there will be no need in
such.
> Code the drivers properly (IRP cancellation), and there will be no
need in
> such.
>
> > I am thinking about somewhat self-correcting behavior.
>
> Well, I can dream about Win32’s (fantastic) AlertThread function,
which will
> turn the thread to a state where all waits will immediately return
> ERROR_ALERTED and do not block. This ERROR_ALERTED would be a signal
for a
> thread to quit ASAP, doing necessary cleanup in the process.
>
> Maybe this would be a good feature.
>
> Maxim Shatskih, Windows DDK MVP
> StorageCraft Corporation
> xxxxx@storagecraft.com
> http://www.storagecraft.com
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@garlic.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
>
>
> —
> Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@cbatson.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>

I may be talking out of context here, but I would implement this by adding a
“nextQ” to the process block. Processes are enqueued with nextQ=readyQueue.
A kill signal to that process has the effect of setting
nextQ=tobeTerminatedQueue. When time comes to move processes out of the wait
queue - and that’s a question of policy and it’s up to debate - I’d just
throw the process into its nextQ. Processes to be deleted would go to the
tobeTerminatedQueue, from which a terminator - again, to get control on a
policy basis - would delete it, return the stack to the available pool, and
so on. But then, I haven’t properly thought about the ins and outs at detail
level, so, there may be a hole lurking somewhere.

Which brings me the question, what’s so hard with TerminateThread that they
can’t properly dispose of the thread resources ?

Alberto.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S. Shatskih
Sent: Thursday, January 29, 2004 4:58 PM
To: Windows System Software Devs Interest List
Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic

This is already in NT as Alertable/NonAlertable waits.

BTW - Linux’s wait queue (and also select()/poll()) implementation is
absolutely brain damaged, it only shows that the Linux co-authors never ever
heard on encapsulation.

The “dirt” is to hook the timer IRQ, not to terminate threads. No
commodity
(non-lab) product must ever do this in any OS.

Lab products are other song. People will tolerate SoftICE being
incompatible with some video card (OK, lab machine can have a requirement of
this particular card, like CE/PC imposes) - but they will not tolerate the
same in some mass-market dongle’s driver.

If the mass-market hardware cannot run in parallel with heavy loaded CD
drive running in its default mode, since REP INSW in ATAPI driver’s ISR will
time the hardware out - then the hardware is broken. Same on serial ports,
who
are nearly as chatty as the CD running in PIO mode.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

----- Original Message -----
From: “Moreira, Alberto”
To: “Windows System Software Devs Interest List”
Sent: Friday, January 30, 2004 12:02 AM
Subject: RE: Re:[ntdev] Thread status in a lot of i/o traffic

> Too late - Linux processes can be enqueued into a wait queue as
> interruptible or uninterruptible. Interruptible processes react to
signals,
> or to interrupts such as timers expiring. Because, hey, one might want to
be
> able to terminate one’s threads. :wink:
>
> Alberto.
>
>
> -----Original Message-----
> From: xxxxx@lists.osr.com
> [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim S. Shatskih
> Sent: Thursday, January 29, 2004 3:54 PM
> To: Windows System Software Devs Interest List
> Subject: Re: Re:[ntdev] Thread status in a lot of i/o traffic
>
>
> I hope Linus is sane enough to never include such dirt to the official
> kernel.
>
> Maxim Shatskih, Windows DDK MVP
> StorageCraft Corporation
> xxxxx@storagecraft.com
> http://www.storagecraft.com
>
> ----- Original Message -----
> From: “David J. Craig”
> Newsgroups: ntdev
> To: “Windows System Software Devs Interest List”
> Sent: Thursday, January 29, 2004 10:53 PM
> Subject: Re:[ntdev] Thread status in a lot of i/o traffic
>
>
> > The FAQ for this is easy: use Linux and write it the way you want it to
> > work.
> >
> > “Don Burn” wrote in message news:xxxxx@ntdev…
> > > Mark,
> > >
> > > I had already decided to ignore him since we can’t make him stop.
> > > Perhaps it is time for a FAQ entry on bypassing the OS?
> > >
> > > Don Burn (MVP, Windows DDK)
> > > Windows 2k/XP/2k3 Filesystem and Driver Consulting
> > >
> > > ----- Original Message -----
> > > From: “Roddy, Mark”
> > > To: “Windows System Software Devs Interest List”
> > > Sent: Thursday, January 29, 2004 11:30 AM
> > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > >
> > >
> > > > Somebody please make alberto stop.
> > > >
> > > >
> > > > =====================
> > > > Mark Roddy
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: Moreira, Alberto [mailto:xxxxx@compuware.com]
> > > > > Sent: Thursday, January 29, 2004 11:22 AM
> > > > > To: Windows System Software Devs Interest List
> > > > > Subject: RE: [ntdev] Thread status in a lot of i/o traffic
> > > > >
> > > > >
> > > > > I wouldn’t use KeSetTimer because I’m not guaranteed that my
> > > > > DPC gets control in time. I would install my own Timer
> > > > > Interrupt routine and do it right there and then.
> > > > > Time-pressing chores should be done in the upper half, not in
> > > > > the bottom half !
> > > > >
> > > > > Alberto.
> > > > >
> > > > >
> > > > > -----Original Message-----
> > > > > From: xxxxx@lists.osr.com
> > > > > [mailto:xxxxx@lists.osr.com]On Behalf Of Maxim
> > > > > S. Shatskih
> > > > > Sent: Thursday, January 29, 2004 11:16 AM
> > > > > To: Windows System Software Devs Interest List
> > > > > Subject: Re: [ntdev] Thread status in a lot of i/o traffic
> > > > >
> > > > >
> > > > > > Basically, all this thred does is it sends dummy hearbeat
> > > > > cmds to the
> > > > > > device.
> > > > >
> > > > > Why not use KeSetTimer for this?
> > > > >
> > > > > > & I/O errors take place. Apparently the thread doesnt get the
> > CPU
> > > > > attention,
> > > > > > as it should.
> > > > >
> > > > > It should not. NT provides no guarantees on amount of CPU
> > > > > attention for a particular thread, i.e. it is not a realtime OS.
> > > > >
> > > > > > Is there any workaround to this?
> > > > >
> > > > > No, but switching to KeSetTimer will make things much better
> > > > > (it supercedes any thread scheduling and can be delayed only
> > > > > by the long and huge DPCs and ISRs).
> > > > >
> > > > > Note: the commodity installs of NT OSes disable the DMA on
> > > > > ATAPI CD drives - due to compatibility with ancient IDE
> > > > > controllers I think. So, if your product is targeted to
> > > > > commodity market - then it will fail miserably while running
> > > > > in parallel with heavy CD-ROM I/O, since the ATAPI driver
> > > > > will do all data transport as REP INSW opcode inside the ISR
> > > > > (higher execution level then timer callbacks, not to say threads).
> > > > >
> > > > > Redesign your hardware and make its timing requirements more
> > > > > relaxing. This is the only way.
> > > > >
> > > > > Maxim Shatskih, Windows DDK MVP
> > > > > StorageCraft Corporation
> > > > > xxxxx@storagecraft.com
> > > > > http://www.storagecraft.com
> > > > >
> > > > >
> > > > > —
> > > > > Questions? First check the Kernel Driver FAQ at
> > > > > http://www.osronline.com/article.cfm?id=256
> > > > >
> > > > > You are currently subscribed to ntdev as:
> > > > > xxxxx@compuware.com To unsubscribe send a blank
> > > > > email to xxxxx@lists.osr.com
> > > > >
> > > > >
> > > > >
> > > > > The contents of this e-mail are intended for the named
> > > > > addressee only. It contains information that may be
> > > > > confidential. Unless you are the named addressee or an
> > > > > authorized designee, you may not copy or use it, or disclose
> > > > > it to anyone else. If you received it in error please notify
> > > > > us immediately and then destroy it.
> > > > >
> > > > >
> > > > > —
> > > > > Questions? First check the Kernel Driver FAQ at
> > > > > http://www.osronline.com/article.cfm?id=256
> > > > >
> > > > > You are currently subscribed to ntdev as:
> > > > > xxxxx@stratus.com To unsubscribe send a blank email to
> > > > > xxxxx@lists.osr.com
> > > > >
> > > >
> > > > —
> > > > Questions? First check the Kernel Driver FAQ at
> > > http://www.osronline.com/article.cfm?id=256
> > > >
> > > > You are currently subscribed to ntdev as: xxxxx@acm.org
> > > > To unsubscribe send a blank email to xxxxx@lists.osr.com
> > >
> > >
> >
> >
> >
> > —
> > Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
> >
> > You are currently subscribed to ntdev as: xxxxx@storagecraft.com
> > To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
> —
> Questions? First check the Kernel Driver FAQ at
> http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@compuware.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com
>
>
>
> The contents of this e-mail are intended for the named addressee only. It
> contains information that may be confidential. Unless you are the named
> addressee or an authorized designee, you may not copy or use it, or
disclose
> it to anyone else. If you received it in error please notify us
immediately
> and then destroy it.
>
>
> —
> Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
>
> You are currently subscribed to ntdev as: xxxxx@storagecraft.com
> To unsubscribe send a blank email to xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.

Well, here’s another thread waiting for a signal that’s never going to
arrive. Time to TerminateThread it ? Oh, sucks, I forgot, it doesn’t work in
this OS. :slight_smile:

Alberto.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Roddy, Mark
Sent: Thursday, January 29, 2004 4:08 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

Well thanks for trying.

=====================
Mark Roddy

-----Original Message-----
From: Michal Vodicka [mailto:xxxxx@veridicom.cz.nospam]
Sent: Thursday, January 29, 2004 3:47 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

> ----------
> From: xxxxx@stratus.com[SMTP:xxxxx@stratus.com]
> Reply To: xxxxx@lists.osr.com
> Sent: Thursday, January 29, 2004 5:30 PM
> To: xxxxx@lists.osr.com
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
> Somebody please make alberto stop.
>
d:\>net stop alberto
System error 1060 has occurred.

The specified service does not exist as an installed service.

Best regards,

Michal Vodicka
STMicroelectronics Design and Application s.r.o.
[michal.vodicka@st.com, http:://www.st.com]


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as:
xxxxx@stratus.com To unsubscribe send a blank email to
xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.

Chuck:

Nothing could be better explained than these couple lines from the
lone-ranger, camafluaged as
Alberto :slight_smile:

-prokash

-----Original Message-----
From: Moreira, Alberto [mailto:xxxxx@compuware.com]
Sent: Friday, January 30, 2004 7:46 AM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

Well, here’s another thread waiting for a signal that’s never going to
arrive. Time to TerminateThread it ? Oh, sucks, I forgot, it doesn’t work in
this OS. :slight_smile:

Alberto.

-----Original Message-----
From: xxxxx@lists.osr.com
[mailto:xxxxx@lists.osr.com]On Behalf Of Roddy, Mark
Sent: Thursday, January 29, 2004 4:08 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

Well thanks for trying.

=====================
Mark Roddy

-----Original Message-----
From: Michal Vodicka [mailto:xxxxx@veridicom.cz.nospam]
Sent: Thursday, January 29, 2004 3:47 PM
To: Windows System Software Devs Interest List
Subject: RE: [ntdev] Thread status in a lot of i/o traffic

> ----------
> From: xxxxx@stratus.com[SMTP:xxxxx@stratus.com]
> Reply To: xxxxx@lists.osr.com
> Sent: Thursday, January 29, 2004 5:30 PM
> To: xxxxx@lists.osr.com
> Subject: RE: [ntdev] Thread status in a lot of i/o traffic
>
> Somebody please make alberto stop.
>
d:\>net stop alberto
System error 1060 has occurred.

The specified service does not exist as an installed service.

Best regards,

Michal Vodicka
STMicroelectronics Design and Application s.r.o.
[michal.vodicka@st.com, http:://www.st.com]


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as:
xxxxx@stratus.com To unsubscribe send a blank email to
xxxxx@lists.osr.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@maxtor.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

> I’m not sure what you’re trying to say here, but any user with

sufficient permission (say, perhaps, the user that created the process)
can terminate a process.

More exactly:

  • TerminateProcess must be called by the code running under the same Logon ID
    as the victim process
    OR
  • TerminateProcess must be called by the code running with SeDebugPrivilege.
    The latter privilege is assigned to Administrators only by default.

So, everybody can kill his/her own processes.

Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com