The subsample alignment of resampled buffers provides seamless playback even
when buffer durations are not an integer number of track ticks.
--HG--
extra : rebase_source : 0fcd52e8a9560de881aa73931cf22a02f984d748
The resampling filter means that the buffer influences a greater number of
samples than indicated by just its length. Including the full influence of
the linear filter means that adjacent buffers aligned appropriately will
behave as if they were one extended buffer.
The buffers are not yet aligned more carefully than track ticks, so buffers
play back seamlessly only if their sample rates and lengths are such that
their duration is an integer number of track ticks.
Knowing how far the filter extends before the start time requires
initializing the resampler before buffer processing.
The patch also includes the input latency in the first resampler input
buffer sample count estimate to reduce the number to calls required
to start the resampler.
--HG--
extra : rebase_source : 16d5af79bc5621be830f5956b51f7ff59d490575
This allows a client to align output samples consistently for independent
resampling of contiguous input buffers.
--HG--
extra : rebase_source : 58caba11ac7f78418b173a98556276d3b4644603
While it seems a little silly since 'perspective' doesn't require
atomicity (rather, it adds an additional transformation to any 3-D
descendants, which already require atomicity), the spec requires it, and
it matches WebKit.
There are three categories of improvements:
(1) using size_t* rather than unsigned long* (and "%zX" rather than
"%lX"), to better support platforms where sizeof(long) !=
sizeof(void*), such as Win64 (untested, though). This is a
non-issue for 64-bit Linux (where I tested) and Mac.
(2) Using the correct amount of 0-padding when printing addresses to
show how much memory space is being printed. In other words, using
"%016zX" on 64-bit platforms instead of "%08zX". This change is
cosmetic-only, though it makes the logs much more understandable.
(3) [in leaksoup.cpp only] Fixing an occurrence of assuming that
sizeof(int) == sizeof(void*). This occurrence led to printing only
the lower half of each word in the output, after doing a correct
analysis of the memory graph.
This patch is patching three files:
(A) nsTraceMalloc.cpp, which is the in-process Gecko trace-malloc code
that generates the memory dumps.
(B) adreader.cpp, which is shared utility code for reading such a
memory dump (currently used only by leaksoup.cpp)
(C) leaksoup.cpp, which reads in such a memory dump, performs a
strongly connected components analysis of the memory graph, and
writes it back out, HTML-ized, with the roots listed at the top.
A fourth file appears to need no modification since it only looks at the
stack part of the dump and not the contents of the memory:
(D) diffbloatdump.pl, which diffs two bloat dumps and produces a stack
tree showing the change in allocations between them
Now that bug 975295 removed the obsolete wrapper class, we can rename
nsTraceRefcntImpl back to its correct pre-XPCOM-glue name,
nsTraceRefcnt.
The best part is that the one place where indentation should have needed
fixing, nsTraceRefcnt::DemangleSymbol, never had its indentation fixed
for the previous renaming.
--HG--
rename : xpcom/base/nsTraceRefcntImpl.cpp => xpcom/base/nsTraceRefcnt.cpp
rename : xpcom/base/nsTraceRefcntImpl.h => xpcom/base/nsTraceRefcnt.h
This makes sense since the file no longer contains anything with the
nsTraceRefcnt name in it, and it will allow renaming nsTraceRefcntImpl
back to nsTraceRefcnt.
Now that the first patch in this bug changed things so that 'none' is
reliably stored as eStyleUnit_None (rather than being stored as a 0
length when it comes from the initial value), we know mChildPerspective
is always > 0 when it is eStyleUnit_Coord, and there's no point making
the additional check.
qimportbz is now pulled from its new canonical home. The old location can
be cleaned if the user agrees to it. The new location is updated in the
hgrc.
DONTBUILD (NPOTB)
