svn commit: samba r15685 - in branches/SAMBA_4_0/source/torture/raw: .

Jeremy Allison jra at samba.org
Sat May 20 02:11:00 GMT 2006


On Thu, May 18, 2006 at 09:29:09AM +0000, metze at samba.org wrote:
> Author: metze
> Date: 2006-05-18 09:29:08 +0000 (Thu, 18 May 2006)
> New Revision: 15685
> 
> WebSVN: http://websvn.samba.org/cgi-bin/viewcvs.cgi?view=rev&root=samba&rev=15685
> 
> Log:
> let samba3 pass RAW-READ and ignore one bug
> 
> this demonstrates how the target:samba3=yes option
> passed in samba3's make test can be used in samba4's smbtorture
> 
> metze
> Modified:
>    branches/SAMBA_4_0/source/torture/raw/read.c
> 
> 
> Changeset:
> Modified: branches/SAMBA_4_0/source/torture/raw/read.c
> ===================================================================
> --- branches/SAMBA_4_0/source/torture/raw/read.c	2006-05-18 09:22:57 UTC (rev 15684)
> +++ branches/SAMBA_4_0/source/torture/raw/read.c	2006-05-18 09:29:08 UTC (rev 15685)
> @@ -475,7 +475,12 @@
>  	CHECK_STATUS(status, NT_STATUS_OK);
>  	CHECK_VALUE(io.readx.out.remaining, 0xFFFF);
>  	CHECK_VALUE(io.readx.out.compaction_mode, 0);
> -	CHECK_VALUE(io.readx.out.nread, 0);
> +	if (lp_parm_bool(-1, "target", "samba3", False)) {
> +		printf("SAMBA3: ignore wrong nread[%d] should be [%d]\n",
> +			io.readx.out.nread, 0);
> +	} else {
> +		CHECK_VALUE(io.readx.out.nread, 0);
> +	}

Actually, bug here is relative - I think Windows has the
bug here as it's returning the wrong value for nread :-).

I know Windows doesn't return a meaningful value here but
I'd rather keep Samba3 doing so unless we know it causes
an application problem.

Jeremy.


More information about the samba-technical mailing list