Displaying 3 results from an estimated 3 matches for "fetchadd".
2007 Jul 09
1
[LLVMdev] Proposal for atomic and synchronization instructions
...hese semantics, while the
L/S membars are offered to varying levels of detail on several
architectures. As Itanium is not yet a fully functional target, it was
not prioritized.
Moreover, as the only instructions (to my knowledge) on Itanium to
have memory synchronization components are cmpxchg and fetchadd, these
could be implemented correctly when implementing the lowering for the
instructions in this proposal, while still providing full memory
barriers when needed outside of the atomic instructions. If there is
serious demand for building memory semantics into the atomic
instructions, "aquire&...
2007 Jul 09
0
[LLVMdev] Proposal for atomic and synchronization instructions
On Monday 09 July 2007 19:33, Scott Michel wrote:
> Torvald Riegel wrote:
> > Hi,
> >
> > I'd like to see support for something like this. I have some comments,
> > and I think there is existing work that you can reuse.
>
> "reuse within the compiler."
within the LLVM compiler framework, to be precise.
>
> > "While the processor may
2007 Jul 09
5
[LLVMdev] Proposal for atomic and synchronization instructions
Torvald Riegel wrote:
> Hi,
>
> I'd like to see support for something like this. I have some comments, and I
> think there is existing work that you can reuse.
"reuse within the compiler."
> "While the processor may spin and attempt the atomic operation more than once
> before it is successful, research indicates this is extremely uncommon."
> I