httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ryan Bloom <>
Subject Re: cvs commit: httpd-2.0/server core.c
Date Mon, 26 Nov 2001 16:57:39 GMT
On Monday 26 November 2001 07:43 am, Brian Pane wrote:
> Ryan Bloom wrote:
> >On Monday 26 November 2001 12:49 am, wrote:
> >>brianp      01/11/26 00:49:29
> >>
> >>  Modified:    server   core.c
> >>  Log:
> >>  Another fix for the core_output_filter() code that concatenates
> >>  small buckets: It's possible for the temporary brigade to
> >>  contain more than one bucket.  If this happens, we need to
> >>  recover the buckets after the first from the temporary brigade
> >>  before destroying it.
> >
> >That shouldn't be possible. The whole point of this loop is that the
> >data in the brigade is < the max size of one bucket. If we have created
> >two buckets in the temp brigade either our values are off, or there is
> >another bug.  Can you post the example that caused this bug? IMHO,
> >this is the wrong fix.
> You can trigger the bug with an shtml file consisting of alternating
> blocks of content and <!--#echo var="foo" --> directives.
> The example that I used is attached.
> Your description of the point of the loop doesn't match the code.
> The temporary brigade is created if we run out of slots in the
> iovec while scanning through a brigade.  The size of the data doesn't
> actually affect whether the temporary brigade will be created.  (I
> think that's a bug, but I'll leave it unchanged for now in case it
> works that way on purpose.)

You're right, my mistake.  Sorry about that.


Ryan Bloom
Covalent Technologies

View raw message