[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Curious "buffer_full" line within host_execute

From: David Malcolm
Subject: Re: Curious "buffer_full" line within host_execute
Date: Mon, 13 Oct 2014 19:49:02 -0400

On Tue, 2014-10-14 at 09:07 +1100, Ben Elliston wrote:
> On Mon, Oct 13, 2014 at 11:42:07AM -0400, David Malcolm wrote:
> > FWIW, for now, I'm attempting to fix my testsuite by including a patched
> > copy of host_execute, and simply removing the bogus line:
> >;a=commitdiff;h=5531d1df2ab9f50c8be72271c6c74ed3757c79db
> That seems a bit nasty (having a duplicated copy of host_execute) in
> the long term. 

Agreed; I don't want to have to carry a copy this code in my jit.exp;
it's merely short-term expediency for the sake of getting my testsuite
working within gcc's larger testsuite (feature freeze for gcc 5 is just
a few weeks away, so I didn't want to gate my changes on a DG release).

> It seems to me that we should patch the DejaGnu version
> as you've done?

That seems to me like a good approach - but I'm relatively unfamiliar
with DG, expect and tcl, so take my opinion with a suitably large pinch
of salt :)

BTW, I'm curious about to what extent others are using (A) host_execute,
and (B) the dejagnu.h header (my testsuite uses both).


reply via email to

[Prev in Thread] Current Thread [Next in Thread]