Re: [Discuss-gnuradio] Fix/Workaround teardown problem in cppdb feature branch

[Eric Blossom]

On Wed, Oct 08, 2008 at 04:40:59PM +0200, Stefan Brüns wrote:
> Hi,
> 
> I have noticed some problem in the cppdb branch (at least) in combination 
> with 
> the inband code.
> 
> when the usrp instances are destructed, the vector holding the daughterboard 
> instances is destructed as well, but there seems to be some problem doing the 
> destruction in the right order. Symptoms are: a) deadlock in the inband 
> examples, b) gdb stopping in siglongjmp, c) valgrind reporting accesses to 
> already freed memory.
> 
> The attached small patch fixes this problem.
> Stefan

Thanks.  FYI, I reworking a lot of this in a dev branch right now.

Eric


> 
> -- 
> Stefan Brüns  /  Bergstraße 21  /  52062 Aachen
> mailto:lurch at gmx.li  http://www.kawo1.rwth-aachen.de/~lurchi/
>    phone: +49 241 53809034     mobile: +49 151 50412019

> === modified file 'usrp/host/lib/legacy/usrp_basic.cc'
> --- usrp/host/lib/legacy/usrp_basic.cc        2008-09-24 17:59:43 +0000
> +++ usrp/host/lib/legacy/usrp_basic.cc        2008-10-08 14:15:58 +0000
> @@ -815,6 +815,8 @@
>    if (!usrp_9862_write_many_all (d_udh, rx_fini_regs, sizeof 
> (rx_fini_regs))){
>      fprintf (stderr, "usrp_basic_rx: failed to fini AD9862 RX regs\n");
>    }
> +  d_db[SLOT_RX_A].resize(0);
> +  d_db[SLOT_RX_B].resize(0);
>  }
>  
>  int
> @@ -1208,6 +1210,8 @@
>    if (!usrp_9862_write_many_all (d_udh, tx_fini_regs, sizeof 
> (tx_fini_regs))){
>      fprintf (stderr, "usrp_basic_tx: failed to fini AD9862 TX regs\n");
>    }
> +  d_db[SLOT_TX_A].resize(0);
> +  d_db[SLOT_TX_B].resize(0);
>  }
>  
>  int