On 5/20/20 1:49 PM, Jan Hubicka wrote:
On related note, I remember us discussing that std compression has
problem with bigger headers then zlib. Since we stream our header that
says if section is compressed, I wonder if we could teach stream-out
phase to skip compression if it is not benefical,
On 5/20/20 1:49 PM, Jan Hubicka wrote:
On related note, I remember us discussing that std compression has
problem with bigger headers then zlib. Since we stream our header that
says if section is compressed, I wonder if we could teach stream-out
phase to skip compression if it is not benefical,
On Wed, 20 May 2020, Jan Hubicka wrote:
> Hi,
> >
> > The patch looks reasonable - the wins are not entirely clear to me
> > since you mix in LTO streaming format optimizations. You promised
> > to improve collisions on the SCC merging side but those are not
> > spectacular (if present at all) i
Hi,
>
> The patch looks reasonable - the wins are not entirely clear to me
> since you mix in LTO streaming format optimizations. You promised
> to improve collisions on the SCC merging side but those are not
> spectacular (if present at all) in the above data?
Thanks for looking into that. Coll
> On 5/19/20 10:46 PM, Jan Hubicka wrote:
> > Martin, the zstd compression breaks the compression statistics (it works
> > when
> > GCC is configured for zlib)
>
> Hello.
>
> Can you please be please more concrete? I can help with, but I don't know
> what's broken.
Sure, the compression statisti
On Tue, 19 May 2020, Jan Hubicka wrote:
> Hi,
> this is new incarantion of patch to identify unmergeable tree at streaming out
> time rather than streaming in and to avoid pickling them to sccs with with
> hash
> codes.
>
> Building cc1 plus this patch reduces:
>
> [WPA] read 4452927 SCCs of av
On 5/19/20 10:46 PM, Jan Hubicka wrote:
Martin, the zstd compression breaks the compression statistics (it works when
GCC is configured for zlib)
Hello.
Can you please be please more concrete? I can help with, but I don't know
what's broken.
Thanks,
Martin
Hi,
this is new incarantion of patch to identify unmergeable tree at streaming out
time rather than streaming in and to avoid pickling them to sccs with with hash
codes.
Building cc1 plus this patch reduces:
[WPA] read 4452927 SCCs of average size 1.986030
[WPA] 8843646 tree bodies read in total
