Re: strace red herring ?

[Corinna Vinschen <corinna-cygwin at cygwin dot com>]

On Jul  6 17:40, Marco Atzeri wrote:
> strace extract from a simple session
>   strace -o mc.strace /usr/bin/mc.exe
> 
> some of the "child_copy: dll bss" report huge time steps,
> but I am almost sure the program was not waiting 15 seconds
> 
> 
> 18 526 [main] bash 5824 pinfo::thisproc: myself dwProcessId 5824
> 47 573 [main] bash 5824 child_copy: dll data - hp 0x238 low 0x180202000,
> high 0x18020E320, res 1
> 
> 15707110 15707683 [main] bash 5824 child_copy: dll bss - hp 0x238 low
> 0x1802D6000, high 0x18031FB30, res 1
> 
> 224 15707907 [main] bash 5824 child_copy: user heap - hp 0x238 low
> 0x600000000, high 0x600080000, res 1
> 24 15707931 [main] bash 5824 child_copy: done
> 
> Could be the counter wrong ?

Yes.  It seems this is the value copied from the parent :)

A fix requires to move `static hires_ns now' out of the
strace::microseconds method and to save/restore it in child_copy.

I'll look into it the next couple of days if nobody beats me to it.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Maintainer                 cygwin AT cygwin DOT com
Red Hat

Attachment: signature.asc
Description: PGP signature