Re: [ros-diffs] [ion] 14047: Dispatching & Queue Rewrite II:

[Gunnar Dalsnes]

[e-mail] wrote:
> Dispatching & Queue Rewrite II:
> - Rewrote wait code. It is now cleaner, more optimized and faster. All waiting
>   functions are now clearly differentiated instead of sharing code. 

Imo, sharing the wait code is cleaner... (I remember the time i merged 
the old wait code into one routine, heh)

These functions
>   are called up to a dozen times a second, so having dedicated code for each of
>   them is a real boost in speed.
> - Fixed several queue issues, made a dedicated queue wait/wake function (you are not
>   supposed to use KeWaitFor on a queue, 

I used KeWaitXxx in the queue impl. because i thought it was good to 
have the wait code in one place only. Thou others find speed more 
important:-D

[Alex Ionescu]

Gunnar Dalsnes wrote:
> [e-mail] wrote:
> > Dispatching & Queue Rewrite II:
> > - Rewrote wait code. It is now cleaner, more optimized and 
> >
> > faster. All waiting
> >       functions are now clearly differentiated instead of sharing code.
>
> Imo, sharing the wait code is cleaner... (I remember the time i merged 
> the old wait code into one routine, heh)

I admit it's cleaner, but speed is more important, imo. Debate that can 
go on for ages however :)
> These functions
> >   are called up to a dozen times a second, so having dedicated 
> >
> > code for each of
> >       them is a real boost in speed.
> >     - Fixed several queue issues, made a dedicated queue wait/wake 
> > function (you are not
> >       supposed to use KeWaitFor on a queue,
>
> I used KeWaitXxx in the queue impl. because i thought it was good to 
> have the wait code in one place only. Thou others find speed more 
> important:-D

This introduces hacks into KeWaitXxx (remember != WrQueue?? :P) and 
also, if you look inside a checked build of NT, you will find 
ASSERT(Header->Type != QueueObject) inside KeWaitXxx so that really made 
me unshare the code :)

Now that these changes are in, I have rewritten thread termination 
(based on Windows Internals Chapter 11) and thread reaping to use 
hypercrtical work queues. The code for the latter is much 
simpler/smaller and also quite faster, since the OS manages everything. 
Termination also works MUCH better, gets rid of the "terminate type" 
hacks that were added, and also flushes/delivers all APCs, timers and 
mutants (we only flushed mutants!). Windows Internals described the 
process in line-by-line pseudo-code so it was a great help.

I should commit that tonight.

Best regards,
Alex Ionescu

[Hartmut Birr]

Alex Ionescu schrieb:
> This introduces hacks into KeWaitXxx (remember != WrQueue?? :P) and
> also, if you look inside a checked build of NT, you will find
> ASSERT(Header->Type != QueueObject) inside KeWaitXxx so that really
> made me unshare the code :)

Can you explain this a little bit more?

- Hartmut

[Alex Ionescu]

Hartmut Birr wrote:
> Alex Ionescu schrieb:
> > This introduces hacks into KeWaitXxx (remember != WrQueue?? :P) and
> > also, if you look inside a checked build of NT, you will find
> > ASSERT(Header->Type != QueueObject) inside KeWaitXxx so that really
> > made me unshare the code :)
>
> Can you explain this a little bit more?

In order to share the code, Gunnar had to insert a hack into the wait 
code not to do some things if the wait mode is WrQueue. By using a 
dedicated function, the hack can be removed.

Best regards,
Alex Ionescu

[Hartmut Birr]

Alex Ionescu schrieb:
> Hartmut Birr wrote:
> > Alex Ionescu schrieb:
> > > This introduces hacks into KeWaitXxx (remember != WrQueue?? :P) and
> > > also, if you look inside a checked build of NT, you will find
> > > ASSERT(Header->Type != QueueObject) inside KeWaitXxx so that really
> > > made me unshare the code :)
> >
> > Can you explain this a little bit more?
>
> In order to share the code, Gunnar had to insert a hack into the wait
> code not to do some things if the wait mode is WrQueue. By using a
> dedicated function, the hack can be removed.

Currently I'm not content with this     answer , I expect more an answer
to 'if you look inside a checked build of NT'.

- Hartmut

[Richard Campbell]

I'm not sure, but i believe he is referring to the 'checked build' 
versions of windows nt...that is, versions of NT that included debugging 
symbols.

Hartmut Birr wrote:
> Alex Ionescu schrieb:
> > Hartmut Birr wrote:
> > > Alex Ionescu schrieb:
> > > > This introduces hacks into KeWaitXxx (remember != WrQueue?? :P) and
> > > > also, if you look inside a checked build of NT, you will find
> > > > ASSERT(Header->Type != QueueObject) inside KeWaitXxx so that really
> > > > made me unshare the code :)
> > >
> > > Can you explain this a little bit more?
> >
> > In order to share the code, Gunnar had to insert a hack into the wait
> > code not to do some things if the wait mode is WrQueue. By using a
> > dedicated function, the hack can be removed.
>
> Currently I'm not content with this     answer , I expect more an answer
> to 'if you look inside a checked build of NT'.
> Hartmut
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Wesley Parish]

On Tue, 15 Mar 2005 11:53, Richard Campbell wrote:
> I'm not sure, but i believe he is referring to the 'checked build'
> versions of windows nt...that is, versions of NT that included debugging
> symbols.

That's the way "Inside Windows NT" uses "checked build".
> Hartmut Birr wrote:
> > Alex Ionescu schrieb:
> > > Hartmut Birr wrote:
> > > > Alex Ionescu schrieb:
> > > > > This introduces hacks into KeWaitXxx (remember != WrQueue?? :P) and
> > > > > also, if you look inside a checked build of NT, you will find
> > > > > ASSERT(Header->Type != QueueObject) inside KeWaitXxx so that really
> > > > > made me unshare the code :)
> > > >
> > > > Can you explain this a little bit more?
> > >
> > > In order to share the code, Gunnar had to insert a hack into the wait
> > > code not to do some things if the wait mode is WrQueue. By using a
> > > dedicated function, the hack can be removed.
> >
> > Currently I'm not content with this     answer , I expect more an answer
> > to 'if you look inside a checked build of NT'.
> > Hartmut
> >
> > Ros-dev mailing list
> > [e-mail]
> > http://reactos.com:8080/mailman/listinfo/ros-dev
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev
-- 
Clinersterton beademung, with all of love - RIP James Blish
-----
Mau e ki, he aha te mea nui?
You ask, what is the most important thing?
Maku e ki, he tangata, he tangata, he tangata.
I reply, it is people, it is people, it is people.

[Alex Ionescu]

Hartmut Birr wrote:
> Currently I'm not content with this     answer , I expect more an answer
> to 'if you look inside a checked build of NT'.
> Hartmut

What the fsck are you implying.

[Hartmut Birr]

Hartmut Birr schrieb:
> Alex Ionescu schrieb:
> > This introduces hacks into KeWaitXxx (remember != WrQueue?? :P) and
> > also, if you look inside a checked build of NT, you will find
> > ASSERT(Header->Type != QueueObject) inside KeWaitXxx so that really
> > made me unshare the code :)
>
> Can you explain this a little bit more?

Hi,

sometimes I find some pieces of code, with which isn't clearly for me,
from where does the knowledge come for the implementation. Such code
isn't used and it isn't necessary for the current status of ros. This is
only a statement and has nothing to do with the changes from revision
14047. If I see a discussion with an answer which does sound like 'look
inside windows and you will find some line of code...', sorry call me
paranoid, I must get this clarify. I am a little bit disappointed from
the following dialogue. Actually I've expected the following answer or
something like this:  'I wrote a test driver and it has hit an assertion
on Windows for queues'.

- Hartmut

[Jason Filby]

Hi guys

Due to the nature of the project, ReactOS is a legal minefield. It is
easy (and almost required) to point fingers and its also natural to
get offended at too much finger pointing. But we need to handle this a
lot better.

1) Until someone acts to prove differently, assume everyone is acting
in good faith and wants to help
2) Respect that others in the project have the right to query in
origins of your code (comes with the 'legal minefield' territory)
3) Separate the 'work' from the 'person' as much as possible during
this type of debate; its shouldn't have to get personal!

I realize that a lot of this is my fault; the IP paper is not yet
published - I'll rectify this shortly. When there are grey areas we
need to decide how we're going to act and publish this in the IP
paper.

We can't expect more proof from some committers than others; we must
decide on the guidelines for all developers and judge everyone
according to the same standard.

Regards
Jason

[Mike Nordell]

Jason Filby wrote:
[snip]
> I realize that a lot of this is my fault; the IP paper is not yet
> published

"IP"? Please don't muddy the waters even more by spreading FUD - even if
unintentionally, and even if due to ignorance (else it's lies). If it's
about copyright, wite copyright. If it's about patents, write patents. If
it's about Internet Protocol...

/Mike

[Richard Campbell]

Ermm....IP = intellectual property...

Mike Nordell wrote:
> Jason Filby wrote:
> [snip]
> > I realize that a lot of this is my fault; the IP paper is not yet
> > published
>
> "IP"? Please don't muddy the waters even more by spreading FUD - even if
> unintentionally, and even if due to ignorance (else it's lies). If it's
> about copyright, wite copyright. If it's about patents, write patents. If
> it's about Internet Protocol...
>
> /Mike
>
> Ros-dev mailing list
> [e-mail]
> http://reactos.com:8080/mailman/listinfo/ros-dev

[Alex Ionescu]

Richard Campbell wrote:
> Ermm....IP = intellectual property...

I think that was another one of Mike's sarcastic jokes that nobody ever 
gets ;)

Best regards,
Alex Ionescu

[Jakob Eriksson]

Alex Ionescu wrote:
> Richard Campbell wrote:
> > Ermm....IP = intellectual property...
>
> I think that was another one of Mike's sarcastic jokes that nobody 
> ever gets ;)

But I got it...

the thing is, IP is a term best reserved for use by lawyers, since it is
newspeak. There is no such thing as Intellectual Property.

It's applying property rights to something that is not property, but
information.

Information is regulated by copyrights and patents, not property laws.

I can steal your car but I can't steal you patent.
I can violate your patent though.

regards,
Jakob

(Plus it's a whole different discussion about quality of patents...)

[Mike Nordell]

Alex Ionescu wrote:
> Richard Campbell wrote:
> > Ermm....IP = intellectual property...
>
> I think that was another one of Mike's sarcastic jokes that nobody
> ever gets ;)

The only sarcasm was displaying the established use of IP for "Internet
Protocol". The rest was dead serious. If someone uses that
(american-lawyer-"invented"?) acronym to bunch together completely unrelated
(and NOT universally accepted, nether bunched together nor by themselves)
concepts, ideas, conventions and laws, instead of referring to the exact
concept they are intending, they are doing this out of two possible reasons:

1. Ignorance. Perhaps the person using the acronym isn't aware these are a
number of completely unrelated concepts bunched together? Perhaps it is, but
isn't aware that by using the acronym it's hurting others (by using the
acronym it gives the acronym credibility)? Perhaps it isn't a person, but a
sheep doing what "all others" (within visual range) do? The possibilities
are almost endless - but the result is still hurting others.

2. Malice, to (try to) spread the venom to make people believe there is some
such universally accepted umbrella term in the real world, where all the
concepts they try to cram under this umbrella most certainly are separate,
and completely unrelated issues.

Giving the ones using that acronym the benefit of a dubt, I believe most
uses are due to ignorance.

But whatever the reasons people use that acronym, it still is spreading of
FUD - and it is hurting others (too).

/Mike

[Alex Ionescu]

Hartmut Birr wrote:
> Hi,
>
> sometimes I find some pieces of code, with which isn't clearly for me,
> from where does the knowledge come for the implementation. Such code
> isn't used and it isn't necessary for the current status of ros. This is
> only a statement and has nothing to do with the changes from revision
> 14047. If I see a discussion with an answer which does sound like 'look
> inside windows and you will find some line of code...', sorry call me
> paranoid, I must get this clarify. I am a little bit disappointed from
> the following dialogue. Actually I've expected the following answer or
> something like this:  'I wrote a test driver and it has hit an assertion
> on Windows for queues'.
> Hartmut

Hi Hartmut,

Sorry for lashing out against you -- I've had an extremly bad week, if 
that's any excuse --. I've been sick, tired, unadmitted to my first 
choice university, and spent sleepness nights with the branch patching.

Both Gunnar and I knew about the beaviour of Queues && KeWaitXxx, my 
reply with him was an "inside joke" on the reason why I un-shared the 
code. Since he argued that speed is not imporant for him, but it is for 
me (our usual chat conversion ;), I brought up as an "inside joke", 
"Hey, btw, windows also asserts!". This was not the reason for the 
changes, and is just a joke I was playing on Gunnar. I understand that 
due to its nature it might've raisen some flags with you, and I 
probabaly overeacted. I don't mean this in an offending way, but your 
English sounded as if you were attacking me (or it might've been that 
bad week I've been having).

In any case, that assert had nothing to do with the changes and it comes 
out of a publically viewable information inside XP Checked Build.

Best regards,
Alex Ionescu