-
Testing latest master on an m1 mac gives an error while using ProfileView, while normal profiling works fine.
Profile:
```julia
@profile plot(sin)
Profile.print()
Overhead ╎ [+additional indent…
-
Here's an example: http://bl.ocks.org/tjdecke/5558084
Or: http://bl.ocks.org/mbostock/3202354
And this one would also be nice: https://github.com/lbrucher/d3-tree-heatmap -- it looks like an upside-d…
otisg updated
4 years ago
-
I would like to use flamegraphs to analyze where a process I am running is spending all of its time. In this instance, I do not care about whether that time is on or off cpu, I just want to know where…
-
It seems that managing the glyph cache we keep around for sal-layout is far slower to manage than even the shaping logic there, which is quite amazing:
.
Ideally, this would be integrated into the main python launcher, which i…
TheJJ updated
2 years ago
-
In the past there were some attempts at profiling the code and analysing the flamegraphs to look for inefficiencies. One of the visible issues were copies of data coming from lmdb: lmdb read gives you…
-
Split from https://github.com/cockroachdb/cockroach/issues/101299
A performance regression of ~3-5% was observed in `execStmtInOpenState`. The flamegraphs below suggest the regression was caused by…
-
Hi,
I just wrote an article about the Inferno options I use in my memory profiler, and why I think that gives better output: https://pythonspeed.com/articles/a-better-flamegraph/
Some if you're …
-
_Not guaranteed improvement_
Look into general optimization of queries performance of the builders methods which assemble queries from the client size. Not the query itself that will be sent to th…
-
`abstutil::Timer` could probably be partly replaced.
- Hierarchial
- Strongly prefer explicitly plumbing a ctx, not a global singleton
- Glue it to file reading for bytes read progress bar
- Col…