On 17/10/14 09:21, Eric Botcazou wrote:
> Hi,
>
> some OSes, for example VxWorks 6, still use DWARF unwinding on the ARM, which
> means that they use __builtin_eh_return (EABI unwinding doesn't). The
> builtin
> is implemented by means of {arm|thumb}_set_return_address, which can generate
> a
> [ thinking out loud ] So, I can’t help but wonder if c6x, mips, nios2 and sh
> now have the exact same problem (or could if they switched code-gen
> some)...
ARM is very peculiar here since it goes through a stack slot to copy a value
into a register; other architectures, e.g. SPARC, do a bare
On Oct 17, 2014, at 1:21 AM, Eric Botcazou wrote:
> This nevertheless used to work because the blockage insn emitted by the RTL
> epilogue was acting as a "wild load" but this got broken by Richard's patch
> which removed the "wild load" trick.
> The attached patch fixes the breakage by marking
Hi,
some OSes, for example VxWorks 6, still use DWARF unwinding on the ARM, which
means that they use __builtin_eh_return (EABI unwinding doesn't). The builtin
is implemented by means of {arm|thumb}_set_return_address, which can generate
a store if LR has been stored on function entry. The pr
