Re: [ping] Re: Fix PR debug/66728

2015-11-02 Thread Richard Sandiford
Richard Sandiford writes: > gcc/ > PR debug/66728 > * dwarf2out.c (loc_descriptor): Remove redundant GET_MODE of > CONST_WIDE_INTs. Handle BLKmode for CONST_WIDE_INT too. > (add_const_value_attribute): Add a mode parameter. > Check it for CONST_INT and CONST_WIDE_INT

Re: [ping] Re: Fix PR debug/66728

2015-11-02 Thread Richard Sandiford
Bernd Schmidt writes: > On 10/28/2015 02:06 PM, Richard Sandiford wrote: >> Ulrich Weigand writes: >>> seems this still hasn't gone upstream ... Any news? >> >> Ah, sorry, I should have been pinging it. I think it's still waiting >> for review. > > Hmm, unfortunately I have a hard time making s

Re: [ping] Re: Fix PR debug/66728

2015-10-28 Thread Ulrich Weigand
Bernd Schmidt wrote: > This is all a bit html mangled, but this and the other change in > loc_descriptor seem to have the effect of doing nothing when called with > BLKmode. What is the desired effect of this on the debug information, > and how is it achieved? Sorry for the mangling caused by

Re: [ping] Re: Fix PR debug/66728

2015-10-28 Thread Bernd Schmidt
On 10/28/2015 02:06 PM, Richard Sandiford wrote: Ulrich Weigand writes: seems this still hasn't gone upstream ... Any news? Ah, sorry, I should have been pinging it. I think it's still waiting for review. Hmm, unfortunately I have a hard time making sense of this patch. Some extra explan

Re: [ping] Re: Fix PR debug/66728

2015-10-28 Thread Richard Sandiford
Ulrich Weigand writes: > Hi Richard, > > seems this still hasn't gone upstream ... Any news? Ah, sorry, I should have been pinging it. I think it's still waiting for review. Thanks, Richard >> This is yet another bug caused by rtx having modeless scalar integer >> constants. We need to use c

[ping] Re: Fix PR debug/66728

2015-10-28 Thread Ulrich Weigand
Hi Richard, seems this still hasn't gone upstream ... Any news? Thanks, Ulrich > This is yet another bug caused by rtx having modeless scalar integer > constants. We need to use context to find the actual mode of a > CONST_INT or CONST_WIDE_INT. > > Getting a mode is especially awkward here.