Cygwin and Windows Vista Beta 2

Cygwin and Windows Vista Beta 2

[Chuck McDevitt]

Just an FYI:  Cygwin's fork() implementation has an intermittent bug
when running on Windows Vista Beta 2 
(Beta 2 will be released to the general public soon).

Here's an example error (from running  bash -l ):

      6 [main] ? (3572) c:\cygwin\bin\bash.exe: *** fatal error -
couldn't alloc
ate heap, Win32 error 0, base 0x800000, top 0x810000, reserve_size
61440, allocs
ize 65536, page_const 4096
     62 [main] bash 4888 child_copy: stack write copy failed,
0x22E300..0x230000
, done 0, windows pid 2286132, Win32 error 5
bash: fork: No error

Here's another example (doing an ls inside bash):

bash-3.1$ ls
     20 [main] ? (5248) c:\cygwin\bin\bash.exe: *** fatal error -
couldn't alloc
ate heap, Win32 error 0, base 0x800000, top 0x820000, reserve_size
126976, alloc
size 131072, page_const 4096
137604415 [main] bash 4436 child_copy: stack write copy failed,
0x22E940..0x2300
00, done 0, windows pid 2287732, Win32 error 5
bash: fork: No error

The odd thing is that it sometimes works, and sometimes gives this
error.

Is it possible there were some "reserved for future use" parameters on
some Win32 calls that aren't getting initialized properly, and are now
being used by Vista?




--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: Cygwin and Windows Vista Beta 2

[Larry Hall (Cygwin)]

Chuck McDevitt wrote:
> Just an FYI:  Cygwin's fork() implementation has an intermittent bug
> when running on Windows Vista Beta 2 
> (Beta 2 will be released to the general public soon).
> 
> Here's an example error (from running  bash -l ):
> 
>       6 [main] ? (3572) c:\cygwin\bin\bash.exe: *** fatal error -
> couldn't alloc
> ate heap, Win32 error 0, base 0x800000, top 0x810000, reserve_size
> 61440, allocs
> ize 65536, page_const 4096
>      62 [main] bash 4888 child_copy: stack write copy failed,
> 0x22E300..0x230000
> , done 0, windows pid 2286132, Win32 error 5
> bash: fork: No error
> 
> Here's another example (doing an ls inside bash):
> 
> bash-3.1$ ls
>      20 [main] ? (5248) c:\cygwin\bin\bash.exe: *** fatal error -
> couldn't alloc
> ate heap, Win32 error 0, base 0x800000, top 0x820000, reserve_size
> 126976, alloc
> size 131072, page_const 4096
> 137604415 [main] bash 4436 child_copy: stack write copy failed,
> 0x22E940..0x2300
> 00, done 0, windows pid 2287732, Win32 error 5
> bash: fork: No error
> 
> The odd thing is that it sometimes works, and sometimes gives this
> error.
> 
> Is it possible there were some "reserved for future use" parameters on
> some Win32 calls that aren't getting initialized properly, and are now
> being used by Vista?


As the main Cygwin page <www.cygwin.com> says:

"The Cygwin DLL works with all non-beta, non "release candidate", ix86 32 bit
versions of Windows since Windows 95, with the exception of Windows CE."

I offer that just in case there's any confusion about the state of Cygwin
and Vista.

With that in mind, have you tried a snapshot?
<http://cygwin.com/snapshots/>

-- 
Larry Hall                              http://www.rfk.com
RFK Partners, Inc.                      (508) 893-9779 - RFK Office
838 Washington Street                   (508) 893-9889 - FAX
Holliston, MA 01746

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: Cygwin and Windows Vista Beta 2

[Robin Walker]

--On 06 June 2006 20:11 -0400 Chuck McDevitt wrote:

> Just an FYI:  Cygwin's fork() implementation has an intermittent bug
> when running on Windows Vista Beta 2
>
> The odd thing is that it sometimes works, and sometimes gives this
> error.

It isn't just under Vista - this is an issue I am seeing under XP: see my 
thread on sshd, with the same intermittent errors in fork().

-- 
Robin Walker (Junior Bursar), Queens' College, Cambridge CB3 9ET, UK
rdhw@cam.ac.uk  http://www.queens.cam.ac.uk/  Tel:+44 1223 335528

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

RE: Cygwin and Windows Vista Beta 2

[Dave Korn]

On 07 June 2006 07:50, Robin Walker wrote:

> --On 06 June 2006 20:11 -0400 Chuck McDevitt wrote:
> 
>> Just an FYI:  Cygwin's fork() implementation has an intermittent bug
>> when running on Windows Vista Beta 2
>> 
>> The odd thing is that it sometimes works, and sometimes gives this
>> error.
> 
> It isn't just under Vista - this is an issue I am seeing under XP: see my
> thread on sshd, with the same intermittent errors in fork().


  Just because it's the same symptom does not make it possible to infer that
there is the same underlying cause.  The *symptom* in both cases is that a
fork operation fails because cygwin cannot reserve the corresponding memory
areas in the child process to make its address space a duplicate of the
parent's.  The particular memory areas that fail to be allocatable in each
case, and the reasons why those memory areas cannot be allocated in each case,
are the actual underlying problem, and are likely to derive from different
causes.

  In the case of Vista B2, I'd want to try switching off the ASLR feature
first.  I can see how that could easily cause problems if you didn't rebaseall
every single time you reboot the machine...


    cheers,
      DaveK
-- 
Can't think of a witty .sigline today....


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: Cygwin and Windows Vista Beta 2

[Robert Pendell]

[-- Attachment #1: Type: text/plain, Size: 1571 bytes --]

Chuck McDevitt wrote:
> Just an FYI:  Cygwin's fork() implementation has an intermittent bug
> when running on Windows Vista Beta 2 
> (Beta 2 will be released to the general public soon).
> 
> Here's an example error (from running  bash -l ):
> 
>       6 [main] ? (3572) c:\cygwin\bin\bash.exe: *** fatal error -
> couldn't alloc
> ate heap, Win32 error 0, base 0x800000, top 0x810000, reserve_size
> 61440, allocs
> ize 65536, page_const 4096
>      62 [main] bash 4888 child_copy: stack write copy failed,
> 0x22E300..0x230000
> , done 0, windows pid 2286132, Win32 error 5
> bash: fork: No error
> 
> Here's another example (doing an ls inside bash):
> 
> bash-3.1$ ls
>      20 [main] ? (5248) c:\cygwin\bin\bash.exe: *** fatal error -
> couldn't alloc
> ate heap, Win32 error 0, base 0x800000, top 0x820000, reserve_size
> 126976, alloc
> size 131072, page_const 4096
> 137604415 [main] bash 4436 child_copy: stack write copy failed,
> 0x22E940..0x2300
> 00, done 0, windows pid 2287732, Win32 error 5
> bash: fork: No error
> 
> The odd thing is that it sometimes works, and sometimes gives this
> error.
> 
> Is it possible there were some "reserved for future use" parameters on
> some Win32 calls that aren't getting initialized properly, and are now
> being used by Vista?
> 
> 
> 
> 

Maybe someone can offer to fix this now that Vista has gone into the
public beta phase.  I can post links here if someone asks for them.

-- 
Robert Pendell
shinji257@uplink.net

Thawte Web of Trust Notary
CAcert Assurer


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Cygwin and Windows Vista Beta 2 (FAQ alert)

[Christopher Faylor]

On Thu, Jun 08, 2006 at 12:01:01AM -0400, Robert Pendell wrote:
>Maybe someone can offer to fix this now that Vista has gone into the
>public beta phase.  I can post links here if someone asks for them.

It's extremely unlikely.  We don't usually try to get Cygwin working on
moving targets.  We have enough trouble with the stationary ones.

The Cygwin web page says this:

    "The Cygwin DLL works with all non-beta, non "release candidate", ix86
    32 bit versions of Windows since Windows 95, with the exception of
    Windows CE."

I guess this belongs as a FAQ entry as it comes up every time Microsoft
has a new beta.

Btw, what this means is that I normally ignore complaints about Cygwin
not working on beta versions of Windows.  I can't testify for Corinna,
however.  In know that in some rare occasions a Red Hat customer might
have an interest and will pay to get something working but this is not
a common occurrence.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/