Feed on
Posts
Comments

Archive for the 'Uncategorized' Category

Mochitests are now Valgrind-clean

One of the things I’ve been doing these past few months is running Mochitests on Valgrind on a semi-regular basis — roughly weekly — and trying to drive the resultant error count down to zero. I did this some years back, in the approach to Firefox 4, but the situation has drifted since then. Anyway, […]

Valgrind support for MacOS X 10.10 (Yosemite)

Valgrind support on Yosemite has improved significantly in the past few months. It is now feasible to run Firefox on Valgrind on Yosemite. Support for 10.9 (Mavericks) has also improved, but 64-bit Yosemite remains the primary focus of support work. For various reasons, MacOS is a difficult target for Valgrind. So you’ll find it little […]

Last August I asked the question “How fast can CFI/EXIDX-based stack unwinding be?” At the time I was experimenting with native unwinding using our in-tree Breakpad copy, but getting dismal performance results. The posting observed that Breakpad’s CFI unwinder is around 30 times slower than Valgrind’s CFI unwinder, and looked in detail at the reasons […]

In last week’s episode of The Wonderful World of CFI Unwinding we looked at how fast it could be done.  This week we’re going to look at how compactly the unwind information can be stored. But I know you want a summary up front, so here it is: I reckon it’s possible to store enough […]

This is a long post.  Here’s a summary. Valgrind unwinds CFI typically 30 times faster than Breakpad.  Bad though that sounds, it’s really encouraging.  If we can get even half that speedup — that is, 15 times faster than at present — CFI/EXIDX unwinding in SPS will be a lot more useful than it currently […]

SPS profiler backend news, 12 August 2013

In short: more adaptation to real-use cases, and more unwind speedups. Benoit redid time-unit measurements so as to support profiling with sampling intervals of less than one millisecond.  This will help profiling graphics and animation code (820048). Gijs Kruitbosch made it possible to control the size of the profiler’s circular buffer using the MOZ_PROFILER_ENTRIES environment […]

SPS Profiler backend news, 8 July 2013

There’s been quite some progress since the last report, focussed on Firefox for Android. Anton Kovalyov landed changes to the GUI that show the event timelines for multiple threads in a synchronised way.  The multithreaded aspects of the profiler are stabilising nicely. Multi-thread profiling support is ready.  Select ‘Multi-Thread’ when profiling to watch any registered […]

Profiler backend news, 3 June 2013

Since the last report, work has focused on making multithreaded profiling more usable, using ARM EXIDX unwind data for Android nightlies, and more analysis of unwind failures.  The latter two lines of investigation happened concurrently because there was some doubt as to how good EXIDX is, so I wanted to find out to what extent […]

Profiler backend news, 30 April 2013

There has been a lot of activity since the previous report, mostly along three lines of development: multithreaded profiling, getting profiling working for Android nightlies, and improving the quality of the stack traces. We now have native unwinds working out-of-the-box for 32- and 64-bit Linux nightlies, and very nearly working for Android nightlies. Multithreaded profiling: […]

Profiler backend news, 10 April 2013

Here is some news about work on Firefox’s built-in time profiler (SPS), and in particular the work being done to use native stack unwinding by using the in-tree copy of the Breakpad unwinding library. There has been a lot of activity in the two and a half weeks since we got back from Paris.  Some […]

Next »