Quoting Tvrtko Ursulin (2017-12-19 11:15:39)
> From: Tvrtko Ursulin <tvrtko.ursu...@intel.com>
>
> A couple of small optimizations which altogether bring around 30%
> improvement in my testing.
>
> 1. Do less string processing on tracepoints names and push more of the
> check into the if-ladder.
>
> 2. Pull out common db key and ctx processing and cache common values in
> local vars.
>
> 3. Key value pair parsing is faster with a regexp.
>
> 4. Avoid sorting the db hash multiple times if possible.
>
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursu...@intel.com>
> Cc: John Harrison <john.c.harri...@intel.com>
> ---
> scripts/trace.pl | 114
> ++++++++++++++++++++++++-------------------------------
> 1 file changed, 49 insertions(+), 65 deletions(-)
> ---
> John please check if this helps with your monster traces before we decide
> if it is worth putting this in. Well, there isn't much of a downside
> except touching something which works and risking breaking it.
Could you run the same trace and do something like
time trace1 < input > output1.html
time trace2 < input > output2.html
cmp output1.html output2.html
?
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
