The C++ sequence point rules for compound assignment (+= and such) are
slightly different from those for C: the value of the LHS is not loaded
until after the RHS value (and LHS location) are evaluated. Since we
don't have tree codes for compound assignment, I need to handle this in
C++ by pre
This appears to do the right thing. I didn't bother with
markers for unwind info, since pdp11 is limited to a.out
and thus will never use dwarf2.
There are improvements that could be made. I added some
comments to that effect but did not otherwise change the
code generation.
Note that the exist
On Fri, Jul 08, 2011 at 05:49:36PM -0400, Jason Merrill wrote:
> The C++ sequence point rules for compound assignment (+= and such)
> are slightly different from those for C: the value of the LHS is not
> loaded until after the RHS value (and LHS location) are evaluated.
> Since we don't have tree
> As suggested in c++/49353, this patch stops copying linkage flags to
> thunks in function_and_variable_visibility, replacing that with just
> asserts to make sure that the front end is setting things up properly.
>
> To make the front end's job easier, I've added
> cgraph_add_to_same_comdat
On Wed, 6 Jul 2011, Tom G. Christensen wrote:
> Latest results for 4.5.x
Thanks, Tom, this is on now.
Gerald
Both resulting from the discussion in PR49680, neither
of which (probably) fix the original reported problem.
Tested via cross-compile to mips-sgi-irix6.5. Committed.
r~
Richard Henderson (2):
dwarf2cfi: Insert notes correctly wrt tablejumps.
dwarf2cfi: Flush queued saves at the end of th
There's little point at holding onto these saves indefinitely.
PR bootstrap/49680
* dwarf2cfi.c (create_cfi_notes): Flush queued saves at the
end of the prologue.
---
gcc/ChangeLog |4
gcc/dwarf2cfi.c |7 +++
2 files changed, 11 insertions(+), 0 deletions(-)
di
PR bootstrap/49680
* dwarf2cfi.c (dwarf2out_frame_debug): Insert cfi notes after
any tablejump vector.
---
gcc/ChangeLog |6 ++
gcc/dwarf2cfi.c | 12 ++--
2 files changed, 16 insertions(+), 2 deletions(-)
diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 8ad8058
On Fri, 8 Jul 2011, Paolo Bonzini wrote:
On 07/08/2011 05:51 AM, Dimitrios Apostolou wrote:
+ /* first write DF_REF_BASE */
This is not necessary. These uses are written to use_vec, while the uses
from REG_EQUIV and REG_EQUAL are written to eq_use_vec (see
df_ref_create_structure).
The fix of streaming DECL_CHAIN in pph for VAR_DECL and FUNCTION_DECL was
introduced to fix "member not found" errors for structs (and we don't have any
tests with unions (we probably should...), but I believe they work the same).
The fix was too general and was actually interfering with somethi
Diego, I know you're working in that area, but the bug fix I'm working on is
kinda coming up to it too.
I realize you readded those two lines (in patch below) because of an ICE
(although I still think they don't make sense, i.e. adding the bindings of a
namespace to itself when they're already
101 - 111 of 111 matches
Mail list logo
