Re: PR debug/60655, debug loc expressions

2014-10-24 Thread Maciej W. Rozycki
On Mon, 20 Oct 2014, Alan Modra wrote: > You were correct to be suspicious that we weren't simplifying as we > should. After more time in the debugger than I care to admit, I found > the underlying cause. > > One of the var loc expressions is > (plus:SI (plus:SI (not:SI (debug_expr:SI D#9)) >

Re: PR debug/60655, debug loc expressions

2014-10-20 Thread Jakub Jelinek
On Mon, Oct 20, 2014 at 09:16:57PM +1030, Alan Modra wrote: > PR debug/60655 > * simplify-rtx.c (simplify_plus_minus): Delete unused "input_ops". > Increase "ops" array size. Correct array size tests. Init > n_constants in loop. Break out of innermost loop when finding >

Re: PR debug/60655, debug loc expressions

2014-10-20 Thread Alan Modra
On Thu, Oct 16, 2014 at 09:07:58AM +0200, Jakub Jelinek wrote: > So, please find the spot where we forget to simplify stuff, and put the > simplification there. You were correct to be suspicious that we weren't simplifying as we should. After more time in the debugger than I care to admit, I foun

Re: PR debug/60655, debug loc expressions

2014-10-16 Thread Jakub Jelinek
On Thu, Oct 16, 2014 at 05:25:57PM +1030, Alan Modra wrote: > Ping? > https://gcc.gnu.org/ml/gcc-patches/2014-09/msg00704.html I think the simplification should be done when constructing the expressions, i.e. if possible in the simplification callback or so if it isn't performed at some level. Bec

Re: PR debug/60655, debug loc expressions

2014-10-15 Thread Alan Modra
Ping? https://gcc.gnu.org/ml/gcc-patches/2014-09/msg00704.html -- Alan Modra Australia Development Lab, IBM

Re: PR debug/60655, debug loc expressions

2014-09-10 Thread Maciej W. Rozycki
On Wed, 10 Sep 2014, Ramana Radhakrishnan wrote: > > 2014-04-10 Ramana Radhakrishnan > > > > PR debug/60655 > > * config/arm/arm.c (TARGET_CONST_NOT_OK_FOR_DEBUG_P): Define > > (arm_const_not_ok_for_debug_p): Reject MINUS with SYM_REF's > > ameliorating the cases where it can b

Re: PR debug/60655, debug loc expressions

2014-09-10 Thread Ramana Radhakrishnan
2014-04-10 Ramana Radhakrishnan PR debug/60655 * config/arm/arm.c (TARGET_CONST_NOT_OK_FOR_DEBUG_P): Define (arm_const_not_ok_for_debug_p): Reject MINUS with SYM_REF's ameliorating the cases where it can be. can I suppose be reverted too. That was always s

Re: PR debug/60655, debug loc expressions

2014-09-10 Thread Maciej W. Rozycki
On Wed, 10 Sep 2014, Alan Modra wrote: > I always build powerpc64-linux with --enable-targets=powerpc-linux and > regression test with RUNTESTFLAGS=--target_board=unix/'{-m32,-m64}', > so my "bootstrapped and regression tested powerpc64-linux" claim > includes a -m32 regression test too. Not quit

Re: PR debug/60655, debug loc expressions

2014-09-10 Thread Alan Modra
On Wed, Sep 10, 2014 at 01:43:22PM +0100, Maciej W. Rozycki wrote: > Thanks for your work on this issue, I have tested your change with my > usual powerpc-gnu-linux multilibs with the old and new result for > gcc.c-torture/compile/pr60655-2.c noted on the right: > > -mcpu=603e

Re: PR debug/60655, debug loc expressions

2014-09-10 Thread Maciej W. Rozycki
On Tue, 9 Sep 2014, Alan Modra wrote: > This cures PR60655 on PowerPC by passing the horrible debug_loc > expressions we have through simplify_rtx. Not only do we get > reg10 + &.LANCHOR0 + const(0x14f - &.LANCHOR0) and > reg10 + &modulus + const(~&d_data), > the two expressions that

Re: PR debug/60655, debug loc expressions

2014-09-09 Thread Alan Modra
On Tue, Sep 09, 2014 at 04:42:04PM +0200, Richard Biener wrote: > On Tue, Sep 9, 2014 at 4:30 PM, Jakub Jelinek wrote: > > On Tue, Sep 09, 2014 at 04:25:23PM +0200, Richard Biener wrote: > >> why wasn't 'result' built using simplify_gen_* in the first place? I also > > > > It is built using cseli

Re: PR debug/60655, debug loc expressions

2014-09-09 Thread Richard Biener
On Tue, Sep 9, 2014 at 4:30 PM, Jakub Jelinek wrote: > On Tue, Sep 09, 2014 at 04:25:23PM +0200, Richard Biener wrote: >> why wasn't 'result' built using simplify_gen_* in the first place? I also > > It is built using cselib_expand_value_rtx_cb, which calls the various > simplify_*_operation and

Re: PR debug/60655, debug loc expressions

2014-09-09 Thread Jakub Jelinek
On Tue, Sep 09, 2014 at 04:25:23PM +0200, Richard Biener wrote: > why wasn't 'result' built using simplify_gen_* in the first place? I also It is built using cselib_expand_value_rtx_cb, which calls the various simplify_*_operation and simplify_rtx too. > note that debug_insns can have all sorts

Re: PR debug/60655, debug loc expressions

2014-09-09 Thread Richard Biener
On Tue, Sep 9, 2014 at 1:50 PM, Alan Modra wrote: > On Fri, Sep 05, 2014 at 11:00:04AM +0930, Alan Modra wrote: >> Of course it would be better to repair the damage done to debug info >> rather than rejecting it outright.. > > This cures PR60655 on PowerPC by passing the horrible debug_loc > expre

PR debug/60655, debug loc expressions

2014-09-09 Thread Alan Modra
On Fri, Sep 05, 2014 at 11:00:04AM +0930, Alan Modra wrote: > Of course it would be better to repair the damage done to debug info > rather than rejecting it outright.. This cures PR60655 on PowerPC by passing the horrible debug_loc expressions we have through simplify_rtx. Not only do we get