On Thu, 1 Aug 2019, Alexander Monakov wrote:
> On Thu, 1 Aug 2019, Richard Biener wrote:
>
> > Ugh. Yeah, it's probably some inlining parameter. OTOH I assumed
> > the inliner would always inline into a simple wrapper if that makes
> > the wrapped function unused - of course if we can't make th
On Thu, 1 Aug 2019, Richard Biener wrote:
> Ugh. Yeah, it's probably some inlining parameter. OTOH I assumed
> the inliner would always inline into a simple wrapper if that makes
> the wrapped function unused - of course if we can't make that
> function static then it cannot know... it looks li
On Thu, 1 Aug 2019, Alexander Monakov wrote:
> On Thu, 1 Aug 2019, Richard Biener wrote:
>
> > So - were you able to allocate some cycles for this? I think
> > we've settled with doing the implementation qsort_r-style while
> > implementing the wrapping with fancy C++ templates.
>
> Yes. It wo
On Thu, 1 Aug 2019, Richard Biener wrote:
> So - were you able to allocate some cycles for this? I think
> we've settled with doing the implementation qsort_r-style while
> implementing the wrapping with fancy C++ templates.
Yes. It works, but not quite as perfectly as I'd hoped. I found two i
On Tue, 23 Jul 2019, Richard Biener wrote:
> On Tue, 23 Jul 2019, Alexander Monakov wrote:
>
> > On Tue, 23 Jul 2019, Richard Biener wrote:
> >
> > > > Can you let me handle sort.cc changes once there's agreement
> > > > which way to go?
So - were you able to allocate some cycles for this? I t
On Tue, 23 Jul 2019, Alexander Monakov wrote:
> On Tue, 23 Jul 2019, Richard Biener wrote:
>
> > > Can you let me handle sort.cc changes once there's agreement
> > > which way to go?
> >
> > Sure - feel free to give option 3 a shot. I'm just worried
> > about the system.h redirect which would b
On Tue, 23 Jul 2019, Richard Biener wrote:
> > Can you let me handle sort.cc changes once there's agreement
> > which way to go?
>
> Sure - feel free to give option 3 a shot. I'm just worried
> about the system.h redirect which would be then disabled for C?
*nod* My preference would be to poi
On Tue, 23 Jul 2019, Alexander Monakov wrote:
> On Tue, 23 Jul 2019, Richard Biener wrote:
>
> > OK, so like below. Slight complication is that C++ doesn't allow
> > the void * to int (*)() casting which means we probably have to go
> > with the template wrapper. I have to check if we use gcc_q
On Tue, 23 Jul 2019, Richard Biener wrote:
> OK, so like below. Slight complication is that C++ doesn't allow
> the void * to int (*)() casting which means we probably have to go
> with the template wrapper. I have to check if we use gcc_qsort
> from C code, the prototypes are currently in syste
On Mon, 22 Jul 2019, Alexander Monakov wrote:
> Hi!
>
> On Mon, 22 Jul 2019, Richard Biener wrote:
>
> > I've noticed that we have quite some instances using qsort
> > and passing down state to the comparator via a global var.
> > Now that we have our own qsort implementation I wonder what
> > t
Hi!
On Mon, 22 Jul 2019, Richard Biener wrote:
> I've noticed that we have quite some instances using qsort
> and passing down state to the comparator via a global var.
> Now that we have our own qsort implementation I wonder what
> the best course of action is to add a variant passing down
> suc
This removes label_for_bb, it's allocated and deallocated in a single
function so we can just pass it down as needed.
Bootstrap / regtest running on x86_64-unknown-linux-gnu.
I've noticed that we have quite some instances using qsort
and passing down state to the comparator via a global var.
No
12 matches
Mail list logo
