On Thu, Sep 12, 2013 at 01:32:07PM +0200, Marc-André Lureau wrote:
> On Wed, Sep 11, 2013 at 11:09 AM, Christophe Fergeau
> wrote:
> > On Tue, Sep 10, 2013 at 04:43:39PM +0200, Marc-André Lureau wrote:
> >> From: Marc-André Lureau
> >>
> >> Thos function shows up in some profiling results, it see
On Thu, Sep 12, 2013 at 1:32 PM, Marc-André Lureau
wrote:
> On Wed, Sep 11, 2013 at 11:09 AM, Christophe Fergeau
> wrote:
> (fyi, 5% is about the same amount of cpu used for glz or quic decoding
> in this recording)
scrap that, they take much more if you cumulate all their functions.
--
Ma
On Wed, Sep 11, 2013 at 11:09 AM, Christophe Fergeau
wrote:
> On Tue, Sep 10, 2013 at 04:43:39PM +0200, Marc-André Lureau wrote:
>> From: Marc-André Lureau
>>
>> Thos function shows up in some profiling results, it seems we can
>> trivially replace it with a precomputed array of 256bytes.
>
> I f
On Tue, Sep 10, 2013 at 04:43:39PM +0200, Marc-André Lureau wrote:
> From: Marc-André Lureau
>
> Thos function shows up in some profiling results, it seems we can
> trivially replace it with a precomputed array of 256bytes.
I forgot to ask, is the function less obvious in the profiling results
a
On Tue, Sep 10, 2013 at 04:43:39PM +0200, Marc-André Lureau wrote:
> From: Marc-André Lureau
>
> Thos function shows up in some profiling results, it seems we can
> trivially replace it with a precomputed array of 256bytes.
> ---
> common/canvas_base.c | 43 ++
From: Marc-André Lureau
Thos function shows up in some profiling results, it seems we can
trivially replace it with a precomputed array of 256bytes.
---
common/canvas_base.c | 43 +++
1 file changed, 43 insertions(+)
diff --git a/common/canvas_base.c b/co
