[ofw] RE: OpenSM patches for saquery porting - 2 of 6 patches.

[Tzachi Dar]

Applied on 1926.

By the way, the file winosm_common.h already seems to be in windows
format.

Thanks
Tzachi 

> -----Original Message-----
> From: Stan C. Smith [mailto:stan.smith at intel.com] 
> Sent: Monday, February 02, 2009 9:17 PM
> To: Tzachi Dar; Hefty, Sean; Yevgeny Kliteynik
> Cc: ofw at lists.openfabrics.org
> Subject: RE: [ofw] RE: OpenSM patches for saquery porting - 2 
> of 6 patches.
> 
> 
> After spending more time looking into opensm, while the real 
> goal was saquery, the best approach for now is to not address 
> opensm getopt cleanup. OpenSM, ibtrapgen, opensmtest all use 
> the function getopt_long_only() which is not part of 
> etc\user\getopt.c. Removing getopt_long_only() can be 
> addressed in another patch for opensm.
> 
> Please commit previous OpenSM patches with the following 
> (.patch file included) superceeding patch #2.
> 
> BTW, the EOL char in the SVN file winosm_common.h is Linux, 
> would you kindly convert the EOL to Windows format?
> 
> Thank you,
> 
> Stan.
> 
> 
> diff U3 C:/Documents and Settings/scsmith/Local 
> Settings/Temp/winosm_common.h-revBASE.svn001.tmp.h 
> C:/Documents and Settings/scsmith/My 
> Documents/openIB-windows/SVN/gen1/trunk/ulp/opensm/user/includ
> e/vendor/winosm_common.h
> --- C:/Documents and Settings/scsmith/Local 
> Settings/Temp/winosm_common.h-revBASE.svn001.tmp.h	Mon Feb 
> 02 10:49:39 2009
> +++ C:/Documents and Settings/scsmith/My
> Documents/openIB-windows/SVN/gen1/trunk/ulp/opensm/user/includ
> e/vendor/winosm_common.h	Mon Feb 02 10:49:21 2009
> @@ -18,17 +18,16 @@
>  };
>  
>  
> +
> +#ifndef _GETOPT_H_
> +
>  # define no_argument		0
>  # define required_argument	1
>  # define optional_argument	2
> +
>  /* Global variables for getopt_long */
>  char *optarg;
> -/*
> -extern int optind;
> -extern int opterr;
> -extern int optopt;
> -extern int iArg;
> -  */
> +
>  struct option
>  {
>    const char *name;
> @@ -36,6 +35,7 @@
>    int *flag;
>    int val;
>  };
> +#endif
>  
>  
>  
> /*************************************************************
> ***********/
> @@ -162,9 +162,10 @@
>  
> /*************************************************************
> ****************/
>  
>  
> -
> -
> +#ifndef getpid
>  #define getpid() GetCurrentProcessId()
> +#endif
> +
>  #define sleep(sec) SleepEx((sec)*1000,TRUE)  #define 
> usleep(usec) SleepEx(usec/1000,TRUE)
>  //#define MT_ALIGN8	__declspec(align(8))
> @@ -172,7 +173,11 @@
>  #ifndef ETIMEDOUT
>  #define ETIMEDOUT		(10060)
>  #endif
> +
> +#ifndef strtoull
>  #define strtoull _strtoui64
> +#endif
> +
>  #define OSM_MAX_LOG_NAME_SIZE 2048
>  #define unlink(str) _unlink(str)
>  #define strnicmp _strnicmp
> 
> 
> 
> 
> 
> Tzachi Dar wrote:
> > Sean and Stan,
> > 
> > Have you reached a conclusion of how to use getopt?
> > 
> > Can we commit the original patch?
> > 
> > Thanks
> > Tzachi
> > 
> >> -----Original Message-----
> >> From: ofw-bounces at lists.openfabrics.org 
> >> [mailto:ofw-bounces at lists.openfabrics.org] On Behalf Of Sean Hefty
> >> Sent: Saturday, January 31, 2009 1:08 AM
> >> To: Smith, Stan; Yevgeny Kliteynik
> >> Cc: ofw at lists.openfabrics.org
> >> Subject: [ofw] RE: OpenSM patches for saquery porting - 2 of 6 
> >> patches.
> >> 
> >>> +
> >>> +#ifndef _GETOPT_H_
> >>> /* Global variables for getopt_long */  char *optarg; /* @@ -36,6 
> >>> +38,7 @@
> >>>   int *flag;
> >>>   int val;
> >>> };
> >>> +#endif
> >>> 
> >>> 
> >>> 
> >>> /************************************************************
> >>> ********** **/ @@ -43,11 +46,14 @@
> >>> get_char_option(const char* optstring,
> >>>                              char*const* argv,int argc,
> >>>                              int iArg, int* opt_ind,char* 
> opt_p); + 
> >>> int getopt_long_only(int argc, char *const*argv,
> >>>                          const char *optstring,
> >>>                          const struct option *longopts, int 
> >>> *longindex);
> >>> 
> >> 
> /********************************************************************
> >> *
> >>> *****/
> >> 
> >> The duplicate implementation of getopt should just be 
> removed, with 
> >> the code using the common one provided by etc\user\getopt.c and 
> >> inc\user\getopt.h.
> >> 
> >> 
> >> 
> >> _______________________________________________
> >> ofw mailing list
> >> ofw at lists.openfabrics.org
> >> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw
> 
>