Linux-Development-Sys Digest #605, Volume #6 Fri, 9 Apr 99 14:14:16 EDT
Contents:
Re: persistent heap design advice please (Bernd Strieder)
Re: PCI DMA to user space possible? (Marc Lefranc)
Re: Arrgghh! How MUCH does it cost to set up Apache? ("Walter B. Burke")
Re: SMP Linux, Any Catches? ("Clint Byrum")
Re: linux 2.2.5, SMP, device driver (Andi Kleen)
Re: GDB broken in Redhat 5.2? (David B Anderson)
Re: Arrgghh! How MUCH does it cost to set up Apache? ("Clay Reiche")
dump and kernels 2.2.5 (SMP) ( master/slave protocol botched ) (Bruce)
Re: GDB broken in Redhat 5.2? (Thorsten Kukuk)
Re: 4 Gb memory? ([EMAIL PROTECTED])
Re: Took one guy 3 days, another 1 day, me 1 hour... (Phil Howard)
Re: PCI DMA to user space possible? (Peter Samuelson)
Network slowing with 2.2.x ? ("Troy Bennett")
Re: Is use of WINE acceptable for a short-term implementation of ThinkPad Modem?
(Andi Kleen)
scanning SCSI-cdrom ([EMAIL PROTECTED])
Re: Idea: Make a seperate "i686" tree for Redhat Linux 6.0 (Matthias Warkus)
----------------------------------------------------------------------------
From: Bernd Strieder <[EMAIL PROTECTED]>
Subject: Re: persistent heap design advice please
Date: Fri, 09 Apr 1999 16:31:58 +0200
Hi
Keith Morgan wrote:
> =
> I am interested in creating a persistent heap library and would appreci=
ate
> any hints or
> suggestions on how to proceed. The 'persistent heap' would be a region =
of
> virtual memory
> backed by a file and could be expanded or contracted.
One big problem within a persistant heap are the pointers among the data
in it and their need of relocation if the heap is mapped in later again
at another address. There might be no guarantee that something is mapped
in twice at the same place. So one would have to provide facilities for
relocation which highly depends on the users of the persistant heap to
keep the pointers right.
I think there are many other problems. Your solution might be error
prone for the users, because there are many things which make no sense
to store within a persistant heap, e.g. all kinds of handles to
resources like sockets, file-descriptors, window-handles. So this thing
does not feel like a heap, it will never be a `heap'. The domain where
this persistant heap could be used might be very small.
Nowadays an often discussed technique is serialization, this is what I
want with my data when storing it externally. One has to write code to
deal with the data and code to serialize the data to a file and read it
again. Not too difficult and easily portable. It is in the hands of the
designer to decide what data produced by the code can be serialized and
read in again later. It=B4s the only person who can decide that and how t=
o
replace those parts of the data during reading which could not be
serialized. If it is clear that the data is fully serializable, why not
look for a true database system to handle the data?
For me these problems would be enough to be discouraged to follow your
idea further.
Bye,
Bernd Strieder
------------------------------
From: Marc Lefranc <[EMAIL PROTECTED]>
Subject: Re: PCI DMA to user space possible?
Date: 09 Apr 1999 15:46:02 +0200
[EMAIL PROTECTED] (Peter Samuelson) writes:
>
> If Bryan is right and it's as simple as going through mmap() (and, as
> he said, "you have to implement your own nopage()" in the driver), I
> would consider that plenty good enough. Not being much of a kernel
> hacker myself, I wouldn't know whom to believe here.
But is this practical when you transfer by DMA say 30 Mbytes in one
shot (in my case this is to record chaotic time series with a A/D board) ?
I have followed the same road as Robin Kaiser, i.e. scatter/gather DMA
directly into user space (if memory serves me right, I pointed him at
that time to the routines for doing the phys/virt address
translation), and IMHO this is the most flexible way do to it.
Marc
--
_____________________________________________________________
Marc Lefranc, Charge de Recherche au CNRS
Laboratoire de Physique des Lasers, Atomes, Molecules
Bat P5, UFR de Physique
Universite des Sciences et Technologies de Lille
F-59655 Villeneuve d'Ascq CEDEX (FRANCE)
e-mail: [EMAIL PROTECTED] ; FAX : +33 (0)3 20 33 70 20
_____________________________________________________________
------------------------------
From: "Walter B. Burke" <[EMAIL PROTECTED]>
Crossposted-To: linux.admin.isp,alt.comp.linux.isp
Subject: Re: Arrgghh! How MUCH does it cost to set up Apache?
Date: Fri, 9 Apr 1999 10:12:51 -0500
That was my point. If you used aliasing or NAT you would only need one
registered IP and you could use all the private IPs you wanted internally.
The only problem is I have yet to find an ISP that would knock the price
down reasonably enough for using one IP. My router at home might as well be
using a dedicated connection because it's dialed out most of the time
anyway. Just my half pence....
Clay Reiche wrote in message ...
>So, what did they do? Just subnet for me? Could I have done that myself?
>Wouldn't I need to register the IPs with ARIN?
>Andrew Bates wrote in message <[EMAIL PROTECTED]>...
>>maybe so you wouldn't need a router on the network?
>>
>>"Walter B. Burke" wrote:
>>>
>>> Why would you need any more than one IP address?
>>> Clay Reiche wrote in message ...
>>> I've set up an Apache web server in Florida using a 384kbs ADSL. The
>>> phone company charged $99 for the installation AND the modem! Not bad!
>For
>>> ADSL they charge based on the amount of band width you require. 384kbps
>>> costs me $55 a month from the phone company. Then there's the ISP
>charge...
>>> I shopped for a week to find the best rate... $199 a month and they gave
>me
>>> a block of 64 static IP addresses!(Don't know what I'm gonna do with
them
>>> all, but I'm sure I'll find something...) They charged $49 for
>installation.
>>> I found one company that was cheaper, $149 a month, but they only gave
me
>>> one IP address...(they would charge $5 per additional IP) and the
>>> installation was $149! So, my monthly expense is $250 and the
>installation
>>> expense was $104. I know that you don't live in Florida, but maybe this
>can
>>> give you an idea...?
>>> Clay
>>> PS: I didn't even consider T1, prices are outrageous.(and I'm
running
>>> this out of my apartment.)
>>> Wayne Chunn wrote in message <[EMAIL PROTECTED]>...
>>> [EMAIL PROTECTED] wrote:
>>> I've been poking around for a bit on the relevant
>newsgroups,
>>> and
>>> trying search engines, but can't seem to get any information
>at
>>> all
>>> about how roughly much it would run to start a small web
>hosting
>>> service,
>>> exclusive of the hardware. I mean some idea, *any* idea, of
>>> current
>>> rates in the Rochester, NY area or thereabouts, for T1, ASDL
>>> (yeah, I..........snip
------------------------------
From: "Clint Byrum" <[EMAIL PROTECTED]>
Subject: Re: SMP Linux, Any Catches?
Date: Fri, 9 Apr 1999 09:03:43 -0700
Our system is pretty memory efficient, so We will really only require 3 megs
per user plus a base of 64 for daemons and OS. I guess that means we're
limited to about 600 users until they fix it. What do you mean 1GB (or 2
with path)?
Sascha Bohnenkamp wrote in message
<7ekjij$220$[EMAIL PROTECTED]>...
>>We have a system which we are porting to Linux from NT/HP-UX/AIX. All of
>the
>>aforementioned handle Multiprocessing extremely well(ok, two of the
>>aforementioned). Will Linux handle 200-300 user telnet/serial getty
>>connections well with 2 or 4 processors(Xeon 450's) and a Hardware RAID? I
>>would like to make sure so we can throw NT out the door. Currently we have
>>to go with HP or IBM for anything higher than 100 users.
>>
>>Also, we are tied to RedHat for our database.
>
>
>linux should do its should nice IF 1(or 2 with path) GB RAM is enough for
>your applications, because linux'86 does not proper handle mor, yet.
>
>
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Subject: Re: linux 2.2.5, SMP, device driver
Date: 09 Apr 1999 17:42:57 +0200
Lorenz Hahn <[EMAIL PROTECTED]> writes:
> Symptom: After setting a interrupt enable bit in the hardware a interrupt
> shoud occur. When I'm tracing through the application, everything is fine.
> When a delay (for(i=0;i<3000;i++)) is inserted after the hardware access,
> everything is fine in most cases. But when the same register is accessed
> short after the first access the interrupt enable bit is not set.
Looks like a race between you and the interrupt handler.
Do the hardware access inside a save_flags()/cli()/restore_flags(), or
cheaper on SMP use a spinlock (see Documentation/spinlocks.txt)
Adding a call to synchronize_irq() may also help.
> At the first look a cache problem, but why does it occur only when two CPU's
> are involved? Any known caveats?
Because then the interrupt handler will probably execute on another CPU.
Actually Intel SMP hardware should be cache coherent (because all CPUs snoop
on the bus for accesses), so it is more likely an interaction with the irq
handler.
-Andi
--
This is like TV. I don't like TV.
------------------------------
From: [EMAIL PROTECTED] (David B Anderson)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: GDB broken in Redhat 5.2?
Date: 9 Apr 1999 16:41:37 GMT
In article <7ekshq$5sr$[EMAIL PROTECTED]>,
Thorsten Kukuk <[EMAIL PROTECTED]> wrote:
>Hello,
>
>In comp.os.linux.development.system Louis S Leclerc -- Personal Account
><[EMAIL PROTECTED]> wrote:
>
>> Hi,
>
>> I have a problem with gdb, which I recently installed on Linux.
>
>> This is for a high visability project in a bank in which we
>> wish to prove Linux as a better alternative over NT4.0 (which
>> we currently use) for future installations, so your
>> assistance is appreciated if we are to succeed.
>
>> When stepping a program (any program, even "hello,world"),
>> when I 'step' to a c function call, the gdb gets stuck and
>> says in the bottom white window (ie. when I hit printf)
>
>Why do you wish to step into a C Library function ? Use
>n for (n)ext, and you could debug your program. Sometimes
>it helps to use the right command.
Of course, the suggestion to use 'next' is a good one.
But sometimes library calls are hidden in macros, and 'step'
is tempting and should work right...
Anyway, the problem is that if a function has not been
lazy-text-resolved yet (first call) and step gets you into
the run-time-linker the debugger gets confused:
stack back traces don't work and the xxgdb 'finish' button does not
work either (because stack back traces don't work).
gdb goes downhill from there.
The workaround is simple:
setenv LD_BIND_NOW 1
(sh/bash/ksh LD_BIND_NOW=1; export LD_BIND_NOW
or export LD_BIND_NOW=1)
and then start your app in that environment (that shell
where you set the LD_BIND_NOW).
Then ld.so/ld-linux.so resolves the calls at startup time
and you don't get into the run-time-linker.
You still step into libc, but the 'finish' button in
xxgdb will work.
[EMAIL PROTECTED]
------------------------------
From: "Clay Reiche" <[EMAIL PROTECTED]>
Crossposted-To: linux.admin.isp,alt.comp.linux.isp
Subject: Re: Arrgghh! How MUCH does it cost to set up Apache?
Date: Thu, 08 Apr 1999 23:02:38 GMT
I can't really justify 64 IP addresses right now and I told the ISP
that..... I do hope to offer a dial-up service at some point and I can dole
some out for that.... I can also use them to set up businesses with a
routable IP.... It's kinda nice to have all my PCs(3) with there own
globally routable IP address.
========Clay
Walter B. Burke wrote in message ...
>Why would you need any more than one IP address?
> Clay Reiche wrote in message ...
> I've set up an Apache web server in Florida using a 384kbs ADSL. The
>phone company charged $99 for the installation AND the modem! Not bad! For
>ADSL they charge based on the amount of band width you require. 384kbps
>costs me $55 a month from the phone company. Then there's the ISP charge...
>I shopped for a week to find the best rate... $199 a month and they gave me
>a block of 64 static IP addresses!(Don't know what I'm gonna do with them
>all, but I'm sure I'll find something...) They charged $49 for
installation.
>I found one company that was cheaper, $149 a month, but they only gave me
>one IP address...(they would charge $5 per additional IP) and the
>installation was $149! So, my monthly expense is $250 and the installation
>expense was $104. I know that you don't live in Florida, but maybe this can
>give you an idea...?
> Clay
> PS: I didn't even consider T1, prices are outrageous.(and I'm running
>this out of my apartment.)
> Wayne Chunn wrote in message <[EMAIL PROTECTED]>...
> [EMAIL PROTECTED] wrote:
> I've been poking around for a bit on the relevant
newsgroups,
>and
> trying search engines, but can't seem to get any information at
>all
> about how roughly much it would run to start a small web
hosting
>service,
> exclusive of the hardware. I mean some idea, *any* idea, of
>current
> rates in the Rochester, NY area or thereabouts, for T1, ASDL
>(yeah, I
> know -- but it might be okay if there were multiple lines,
>perhaps
> cheaper than the same capacity from a single T1 if I understand
>the
> racket the telephone companies are running on T1/T3 etc.), or
>whatever.
> The telephone company sites simply refuse to provide this
>information,
> and I've been through this sort of wild-goose chase enough
times
>to know
> that if the first few conceptually related sites will not
>provide
> some piece of information, then they're *all* going to be
hiding
> that information, or will be wanting to play head games on you
>before
> finally grudgingly admitting to even some vague price or
>another.
>
> Fuck that crap. I'll read the FAQ, if I can ever find one,
or
>a book.
> ANYTHING is better than a predatory salesman trying to probe my
>mental
> armour for a chink or flaw he can exploit to screw me for all
he
>can get.
>
> Speaking of which, a FAQ pointer or book recommendation for
>setting
> up Apache under Linux, to run a small server network with
>perhaps
> four or five 200MHz Pentium class junkers with maybe only 64M
>each,
> would be very useful. :)
>
> -----------== Posted via Deja News, The Discussion Network
>==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start
>Your Own
>
> I can't help you much with your primary request (pricing) but for
>your request in
> the last paragraph I would suggest the Special Edition of Using
>Linux Part 7, Chapters
> 35-37. Also, if you have access to several 200 MHz Pentiums you
may
>be interested
> in experimenting with Parallel Processing. Search the web for
info
>on current PP Linux
> projects if that interests you, there are quite a few.
> Later,
>
> Wayne
>
>
>
------------------------------
From: Bruce <[EMAIL PROTECTED]>
Subject: dump and kernels 2.2.5 (SMP) ( master/slave protocol botched )
Date: 09 Apr 1999 09:03:29 -0600
Hi,
I am having trouble getting dump to run on a new system
(redhat5.2 with linux-2.2.5). It ran previously under 2.0.36. I have
installed the required updated packages for 2.2.5. I really don't have
any idea if this is a problem with dump or a newer kernel.
When I run dump (locally) using a DLT tape drive, I get the following almost
immediately
[root@wolf bin]# dump 7ufbB /dev/nst0 20 20000000 /
DUMP: Date of this level 7 dump: Fri Apr 9 10:15:22 1999
DUMP: Date of last level 0 dump: Thu Apr 8 16:21:40 1999
DUMP: Dumping /dev/sda8 (/) to /dev/nst0
DUMP: mapping (Pass I) [regular files]
DUMP: mapping (Pass II) [directories]
DUMP: estimated 6079 tape blocks on 0.00 tape(s).
DUMP: dumping (Pass III) [directories]
DUMP: master/slave protocol botched.
DUMP: The ENTIRE dump is aborted.
I also built the source rpm and it ran once just fine. I tried it later
and it came up with the same "botched" errors. These errors occur regardless
of the drive being mounted read-only or read-write.
Any help would be appreciated.
System information ...
Dump rpm version ... dump-0.3-17
scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.10/3.2.4
<Adaptec AHA-294X Ultra2 SCSI host adapter>
scsi1 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.10/3.2.4
<Adaptec AIC-7890/1 Ultra2 SCSI host adapter>
scsi2 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.10/3.2.4
<Adaptec AIC-7860 Ultra SCSI host adapter>
scsi : 3 hosts.
Vendor: Quantum Model: DLT4000 Rev: CD50
Type: Sequential-Access ANSI SCSI revision: 02
Detected scsi tape st0 at scsi0, channel 0, id 6, lun 0
(scsi0:0:6:0) Synchronous at 10.0 Mbyte/sec, offset 15.
Vendor: IBM Model: DGHS18D Rev: 03C0
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sda at scsi1, channel 0, id 0, lun 0 (This is where the
filesystem resides)
(scsi1:0:0:0) Synchronous at 80.0 Mbyte/sec, offset 15.
Vendor: IBM Model: DGHS18D Rev: 03C0
Type: Direct-Access ANSI SCSI revision: 02
Does anyone have any ideas about this.
Thanks,
--bruce milner
------------------------------
From: Thorsten Kukuk <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: GDB broken in Redhat 5.2?
Date: 9 Apr 1999 12:41:30 GMT
Hello,
In comp.os.linux.development.system Louis S Leclerc -- Personal Account
<[EMAIL PROTECTED]> wrote:
> Hi,
> I have a problem with gdb, which I recently installed on Linux.
> This is for a high visability project in a bank in which we
> wish to prove Linux as a better alternative over NT4.0 (which
> we currently use) for future installations, so your
> assistance is appreciated if we are to succeed.
> When stepping a program (any program, even "hello,world"),
> when I 'step' to a c function call, the gdb gets stuck and
> says in the bottom white window (ie. when I hit printf)
Why do you wish to step into a C Library function ? Use
n for (n)ext, and you could debug your program. Sometimes
it helps to use the right command.
Thorsten
--
Thorsten Kukuk http://www.suse.de/~kukuk/ [EMAIL PROTECTED]
SuSE GmbH Schanzaeckerstr. 10 90443 Nuernberg
Linux is like a Vorlon. It is incredibly powerful, gives terse,
cryptic answers and has a lot of things going on in the background.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: 4 Gb memory?
Date: Fri, 09 Apr 1999 14:39:16 GMT
In article <[EMAIL PROTECTED]>,
Robert Krawitz <[EMAIL PROTECTED]> wrote:
> "G. Sumner Hayes" <[EMAIL PROTECTED]> writes:
>
> > Think about the memory model for a minute -- processes need a part of
> > the address space for themselves. The current default is 1GB kernel
> > memory mappings, 3GB maximum process size (having maxprocsize >
> > maxramsize makes sense when you consider virtual memory. Ingo's
> > patches make it an even 2/2 split. A 3/1 split might be possible but
> > probably not very useful; it's not as straightforward as going 2/2 as
> > you need to ensure that there aren't places in the kernel which assume
> > that you can map all the RAM into one process.
>
> Actually, the 4 GB virtual address space is split between the kernel
> (1G), physical memory (1G), and process address space (2G). The
> kernel maps all of physical memory into virtual memory so that it can
> always directly access all of physical memory. Ingo's patch allows 2G
> of physical memory and 1G of process address space.
Where is this patch? I have 2G system running 2.2.5 and need this patch.
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Phil Howard)
Subject: Re: Took one guy 3 days, another 1 day, me 1 hour...
Crossposted-To: comp.os.linux.advocacy
Date: Sun, 04 Apr 1999 07:32:55 GMT
On Fri, 02 Apr 1999 19:58:03 -0700 Mario Miyojim ([EMAIL PROTECTED]) wrote:
| I have used Linux for 2 years now. Started with RedHat 2.01, the 4.1, then 5.1,
| then 5.2. And I used my experience to install for other people, and orient friends
| and family remotely. Each time my installation difficulties changed with the
| environment and version (available hardware devices, my preferences, the
| user's specifications, whether dual-boot or not, etc.).
| My last installation of Linux at work was done in less than one hour, and it
| was from the Mandrake 5.3 CDROM that I bought from www.cheapbytes.com.
| This time is much more impressive, because there was no need to stop to think,
| I knew exactly all the particulars of my hardware, and I did not have to run
| rpm to install KDE 1.1. Everything was done automatically. It must be ideal
| for Linux beginners migrating from the Windows domain. I think Mandrake
| and Corel, and other initiatives are responses to long-time criticisms from
| people who know nothing but Microsoft.
Someone I was talking to was mentioning that the next coming user interface
is voice, referring to how the computer interacted with people in Star Trek
and other Sci Fi. I asked why that would be better. The response was that
you could just tell the computer what you wanted done, and didn't have to
go find the right menu or dialog.
--
Phil Howard KA9WGN
[EMAIL PROTECTED] [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: PCI DMA to user space possible?
Date: 9 Apr 1999 11:35:46 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Klaus Elend <[EMAIL PROTECTED]>]
> I do not like those special solutions where the user has to do an
> ioctl() to get some shared memory from the driver and then do more
> ioctl()-calls to transfer data. This approach results in software
> that is not portable between different types of hardware.
>From what I understand, it sounds like the user process is just using
mmap(). Not what I would call nonstandard.
> Therefore, I prefer to stick to the old read()/write() model as long
> as I can, which allows the user to malloc() a buffer and do a read()
> (and a write(), of course) with a pointer to that buffer.
Shared memory is already done through mmap() -- why shouldn't DMA be?
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: "Troy Bennett" <[EMAIL PROTECTED]>
Subject: Network slowing with 2.2.x ?
Date: Fri, 9 Apr 1999 09:54:29 -0700
I run a quake 2 server on RedHat 5.2 on a dual PPro 200Mhz.
It has been providing speedy connections until I installed the 2.2.x kernel.
I have been through three revs now and they all seem to considerably
increase the lag.
The system will run fine after a reboot for a about 20 min then its back to
lag city.
-Thanks for any help
------------------------------
From: Andi Kleen <[EMAIL PROTECTED]>
Crossposted-To: comp.emulators.ms-windows.wine
Subject: Re: Is use of WINE acceptable for a short-term implementation of ThinkPad
Modem?
Date: 09 Apr 1999 18:01:14 +0200
[crossposted to comp.emulators.ms-windows.wine, because it is really
more appropiate there]
"Keith Frechette" <[EMAIL PROTECTED]> writes:
> First, please allow me a quick introduction. I am IBM's development team
> lead responsible for the Windows 2000 modem support for the ThinkPad 770/600
> series laptops. I've been with IBM's DSP Integrated Solutions group
> (formerly the Mwave group) for the last seven years. Life is good.
>
> I've only recently begun my journey as a Linux advocate and developer. As a
> matter of fact, it was IBM and RedHat's joint annoucement in February that
> inspired me to go buy a copy of RedHat Linux 5.2 to see what the hubbub was
> all about. I installed it on a ThinkPad 600E, and I was impressed at how
> easy it was for me to install. (Granted, I am a long-time Windows developer;
> so that did give me an edge over, say, Grandma Jean.) However, I was very
> disappointed (and reasonably so) to see the lack of support for the 600E's
> internal modem. I said to myself, "What a bunch of slackers those RedHat
> folks are, not providing me with modem support." But then I figured, what
> the hey -- it's not their fault that I haven't given them any code!
>
> Although IBM has not committed to providing Linux modem drivers for its
> ThinkPad 770/600 series, I have begun my own after-hours research into the
> best approach for porting our existing Windows 95/98/NT/2000 drivers to
> Linux. Although a small kernel driver is required to do I/O to the modem,
> the bulk of the Windows drivers is user-mode code. As a quick and purely
> experimental path for getting the modem running on Linux and to help
> identify any architectural hurdles, I am thinking of creating a Linux kernel
> driver and then using WINE to run the existing user-mode Windows NT 4.0
> drivers, with possibly small modifications. If the modem works in that
> environment, porting the user-mode code to pure Linux should be relatively
> straight forward.
>
> Any thoughts on this approach and whether an alpha release implemented this
> way would be useful to the Linux community?
The basic approach sounds fine (assuming you have the source and use
Winelib to compile it with some changes, binary emulation is probably too
hairy). Using an user level process for a server is easy, using virtual
terminals (telnetd does it all the time). Linux user space only offers
soft realtime though, so if the driver is very timing dependent it may
be needed to move more of it into kernel space.
[BTW Linux 2.2 has a "soundmodem" driver that can do 9600baud over a
soundblaster]
[Disclaimer: I have neither seen the driver source of course, nor
I am a Wine guru, so this is just an (un?)informed guess. ]
One possible caveat though:
I would assume the DSP code is rather low-level windows code that is
very dependent on windows internal special functions. So far Wine has
been mostly tested with normal user level applications, less with very
system dependent drivers. Because Windows has a _lot_ of functions,
and Wine developers have finite time resources only a commonly used
subset used by the tested applications is implemented.
If the driver depends on lots of internal Windows pecularities (e.g. specific
timing requirements would come to mind) it may need some Wine work to get
it working.
Of course this really depends on the design of the driver, perhaps
it is really easy @) I think the project would be most likely doable.
-Andi
--
This is like TV. I don't like TV.
------------------------------
From: [EMAIL PROTECTED]
Subject: scanning SCSI-cdrom
Date: Fri, 09 Apr 1999 16:31:36 GMT
Hi,
I've got the problem to scan a whole SCSI-CDROM ( not an audio CD ) without
interruption. (this is like playing an audio one for 65 minutes..)
My guess is to either apply one of the READ (6,10) commands with a large
Transfer Length or to use the Search Data Commands (30h - 32h).
The first option is limited by "sg.h:#define SG_BIG_BUFF 32768" in the
driver source. This limits me to read only 13 Frames. Of course I can try to
increase this value in the kernel (can't I ?).
Anyway, the second way seems to be more promising. But the member of
sg_header structure "sg_hd->reply_len = OFF + out_size" is involved in
allocating scsi-buffer-memory in the kernel. So if it's bigger than
SG_BIG_BUFFER, the command won't work. I wonder if this is ok, because the
Search Data command you shouldn't need to store the scanned data in the
buffer.
On the other side, if I set out_size (or reply_length in the sg_header
structure) to a small value and try to pass the wanted number
of sectors for a search through the SCSI-command block, it doesn't
report a memory allocation error but it still scans for about 1 minute.
I don't know how the SCSI-command block are getting involved by the drivers.
But don't tell me to read the SCSI-Programming-Howto and SCSI-Manuals please,
that's what I've done the last few days...
Sorry, this is a little bit difficult to understand, it needs someone who is
familiar with this problem.
Cheers, Kay
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Matthias Warkus)
Crossposted-To:
comp.os.linux.misc,linux.redhat.misc,alt.linux,alt.os.linux,comp.os.linux.hardware
Subject: Re: Idea: Make a seperate "i686" tree for Redhat Linux 6.0
Date: Thu, 8 Apr 1999 23:34:10 +0200
Reply-To: [EMAIL PROTECTED]
It was the 8 Apr 1999 17:07:45 -0400...
..and Alexander Viro <[EMAIL PROTECTED]> wrote:
> In article <[EMAIL PROTECTED]>,
> Johan Kullstam <[EMAIL PROTECTED]> wrote:
> >not if you're from a commonwealth country - which includes new zealand
> >(where is the old zealand btw?). in *english* (as opposed to american
> ^^^^^^^ - Zeeland.
>
> Gaak... Across the North Sea (looking from England, that is).
> Netherlands. Heck, they *really* don't teach history and geography
> in schools, or what?
Not in the U.S. as far as I know.
mawa
--
Sometimes Usenet makes me feel like a crossbreed of Don Giovanni and
the Sorcerer's Apprentice; while around me something I created is
making all kinds of mess, I want to yell "Enough! Enough!" and wait
for the devil to get me. -- mawa
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
