Overview of my bug reports on add-on performance measurements · 2011-04-09 19:52 by Wladimir Palant
In a comment to my previous blog post Jorge Villalobos asked me to file bugs on everything I think is wrong with the add-on performance measurements. So that’s what I did. The resulting list is long and I didn’t feel like creating a tracking bug so I better list the bugs in a blog post so that everybody (especially me) can find them.
By the way, Nils Maier’s comment in my previous blog post is worth reading. I think Nils got a point here.
Bug 647954 – When extracting add-ons files should be written in binary mode
Bug 648222 – Talos needs to enforce browser shutdown when testing add-on performance
Bug 648225 – Performance of platform-dependent add-ons is not tested
Bug 648229 – Compatibility info of the add-ons isn’t properly considered during performance testing
Bug 648732 – Testing add-ons unpacked creates unrealistic results
Bug 648734 – Effect of measurement fluctuations on Talos should be minimized
Bug 648737 – Add-on developers need access to the raw results of performance measurements
Bug 648742 – Displaying percentage values for ‘slowness’ is misleading
Bug 648745 – The boundary of 5% for slow add-ons is arbitrary
Bug 648749 – Talos should capture error console output in the logs
Bug 648751 – Performance results with negative impact on some platforms should not be suppressed
Bug 648752 – Add-ons causing higher real-world slowdown than measured on Talos need to be recognized
Bug 648793 – Test results for Download Statusbar on OS X are unrealistically low
Also, there is a bug report filed by Dave Garrett:
Update: Nils Maier created that tracking bug for me:
Commenting is closed for this article.