This is a reland of 91bab5588c
This CL contains two major changes w.r.t to the original CL:
The random state is removed from the Smi root list and we pre-seed the RNG
on each sort with the length of the array.
To cut down on the length of the arguments list and to keep track of the
random state across recursive calls, we move most of the sort arguments into
a FixedArray and reload from the array for each recursion.
Original change's description:
> [array] Use random middle element to determine pivot during sorting
>
> This CL adds a "random state" to the Smi Root list and implements a
> basic Linear congruential pseudo random number generator in Torque.
>
> The RNG is used to determine the pivot element for sorting. This will
> prevent the worst cases for certain data layouts.
>
> Drive-by-fix: Make sorting of ranges and execution pauses for profviz
> deterministic by adding a secondary sorting criteria.
>
> Bug: v8:7382
> Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
> Change-Id: Ieb871e98e74bdb803f821b0cd35d2f67ee0f2868
> Reviewed-on: https://chromium-review.googlesource.com/1082193
> Reviewed-by: Hannes Payer <hpayer@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Camillo Bruni <cbruni@chromium.org>
> Commit-Queue: Simon Zünd <szuend@google.com>
> Cr-Commit-Position: refs/heads/master@{#53524}
Bug: v8:7382
Change-Id: Ia7bef7ed1c0e904ffe43bc428e702f64f9c6a60b
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1087888
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Simon Zünd <szuend@google.com>
Cr-Commit-Position: refs/heads/master@{#53583}
This reverts commit 91bab5588c.
Reason for revert: Seems to break a layout test:
https://ci.chromium.org/buildbot/client.v8.fyi/V8-Blink%20Linux%2064/23895
See also:
https://github.com/v8/v8/wiki/Blink-layout-tests
Original change's description:
> [array] Use random middle element to determine pivot during sorting
>
> This CL adds a "random state" to the Smi Root list and implements a
> basic Linear congruential pseudo random number generator in Torque.
>
> The RNG is used to determine the pivot element for sorting. This will
> prevent the worst cases for certain data layouts.
>
> Drive-by-fix: Make sorting of ranges and execution pauses for profviz
> deterministic by adding a secondary sorting criteria.
>
> Bug: v8:7382
> Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
> Change-Id: Ieb871e98e74bdb803f821b0cd35d2f67ee0f2868
> Reviewed-on: https://chromium-review.googlesource.com/1082193
> Reviewed-by: Hannes Payer <hpayer@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Camillo Bruni <cbruni@chromium.org>
> Commit-Queue: Simon Zünd <szuend@google.com>
> Cr-Commit-Position: refs/heads/master@{#53524}
TBR=hpayer@chromium.org,cbruni@chromium.org,jgruber@chromium.org,szuend@google.com
Change-Id: I54f5d3f719428fd089ff12ff217d1c819f9ad1f7
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:7382
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1088506
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#53542}
This CL adds a "random state" to the Smi Root list and implements a
basic Linear congruential pseudo random number generator in Torque.
The RNG is used to determine the pivot element for sorting. This will
prevent the worst cases for certain data layouts.
Drive-by-fix: Make sorting of ranges and execution pauses for profviz
deterministic by adding a secondary sorting criteria.
Bug: v8:7382
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Change-Id: Ieb871e98e74bdb803f821b0cd35d2f67ee0f2868
Reviewed-on: https://chromium-review.googlesource.com/1082193
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Commit-Queue: Simon Zünd <szuend@google.com>
Cr-Commit-Position: refs/heads/master@{#53524}
The --trace-turbo flag would cause a crash when compiling a WASM_FUNCTION. It
was caused by assuming the OptimizedCompilationInfo had a SharedFunctionInfo
attached if the code isn't a stub and wasm functions are not considered as such.
In order to test this, we've added a new flag to specify were to dump JSON
files: --trace-turbo-path. This is used to make sure we do not leave lots of
files behind in the top-level directory. It should be useful as standalone
feature too.
Change-Id: Ia9442638d28100bea45a8683fb233803cc5393f2
Reviewed-on: https://chromium-review.googlesource.com/1030555
Commit-Queue: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#53153}
Implicitly escape all output that is passed to Log::MessageBuilder.
We escape non-printable characters and the log field separator ','
using the \x00 and \u0000 escape sequences.
Example:
Before: event-foo,"space: ","comma: ,","double quotes: """
After: event-foo,space: ,comma: \x2C,double quotes: "
This might slightly impact human readability of the log files in
extreme cases. However, most strings do not contain any escaped
characters.
Bug:
Change-Id: Ic78f6d9932367d02f9f3c3f70b41b5c283bdf880
Reviewed-on: https://chromium-review.googlesource.com/728332
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#48805}
This adds optimization and deoptimization counts to the Web UI. Also, the function timeline
now shows optimization and deoptimization marks.
Review-Url: https://codereview.chromium.org/2753543006
Cr-Commit-Position: refs/heads/master@{#44033}
- lower the cutoff-threshold for bottom up graphs to 1%
- show all builtins/stubs/ics by default
R=jkummerow@chromium.org
Change-Id: I522173155b817ae20a37c40ecf411f2b55d82105
Reviewed-on: https://chromium-review.googlesource.com/452482
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#43717}
This is mostly a performance experiment. If it provides no speedup,
it can be reverted to keep IC miss events in timeline plots.
Otherwise, the RuntimeCallStats system is the replacement tool for
investigating performance issues related to IC misses.
This effectively reverts 1f8adc15 / r21736.
Review-Url: https://codereview.chromium.org/2480343002
Cr-Commit-Position: refs/heads/master@{#40893}
An additional slide offset is exported into `shared-library`, which is used to
symbolize C++ stack on systems with ASLR (OS X).
This patch adds slide offset support in dumpcpp script.
BUG=v8:5048
Review-Url: https://codereview.chromium.org/2006813002
Cr-Commit-Position: refs/heads/master@{#36574}
When exporting `shared-library` in profile log, additionally export a
slide offset. This is required to parse profile logs generated on
systems with ASLR (OS X), otherwise it is impossible to assign C++
symbol names to their addresses in the log.
See: https://github.com/nodejs/node/issues/6466
BUG=
Review-Url: https://codereview.chromium.org/1934453003
Cr-Commit-Position: refs/heads/master@{#35921}
After histrogram timer added time recaling functionality,
some events, e.g. parse, histogram timer generates event log ending with 'MicroSeconds'.
Since ProfViz can't recorgnize it, this patch cuts off 'MicroSeconds' postfix.
R=vogelheim@chromium.org, yangguo@chromium.org
BUG=chromium:
LOG=N
Review URL: https://codereview.chromium.org/1771293002
Cr-Commit-Position: refs/heads/master@{#34710}
(1) --prof-cpp: Collects ticks like --prof, but ignores code creation events to reduce distortion (so all JS ticks will be "unaccounted"). Useful for profiling C++ code.
(2) --timed-range flag for tick processor: Ignores ticks before the first and after the last call to Date.now(). Useful for focusing on the timed section of a test.
Review URL: https://codereview.chromium.org/802333002
Cr-Commit-Position: refs/heads/master@{#26168}
Their values are not used neither by the tick processor nor by CpuProfiler so it is just a waste of space.
TickSample used to be a transport for grabbed register values to TickSample::Trace, now they are passed in a special structure RegisterState which is allocated on the stack for the sampling period.
Some common pieces were moved from platform-dependent code into Sampler::SampleStack and TickSample::Init.
BUG=None
R=jkummerow@chromium.org, loislo@chromium.org
Review URL: https://codereview.chromium.org/18620002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15484 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
Encapsulate the helper functions in mjsunit.js.
Now only exposes the exception class and the assertXXX functions.
Make assertEquals use === instead of ==.
This prevents a lot of possiblefalse positives in tests, and avoids
having to do assertTrue(expected === actual) when you need it.
Fixed some tests that were either buggy or assuming == test.
Review URL: http://codereview.chromium.org/6869007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@7628 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The main issue was due to multiple recompilations of functions. Now
code objects are grouped by function using SFI object address.
JSFunction objects are no longer tracked, instead we track SFI object
moves. To pick a correct code version, we now sample return addresses
instead of JSFunction addresses.
tools/{linux|mac|windows}-tickprocessor scripts differentiate
between code optimization states for the same function
(using * and ~ prefixes introduced earlier).
DevTools CPU profiler treats all variants of function code as
a single function.
ll_prof treats each optimized variant as a separate entry, because
it can disassemble each one of them.
tickprocessor.py not updated -- it is deprecated and will be removed.
BUG=v8/1087,b/3178160
TEST=all existing tests pass, including Chromium layout tests
Review URL: http://codereview.chromium.org/6551011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6902 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
The problem appeared due to a fact that stubs doesn't create a stack
frame, reusing the stack frame of the caller function. When building
stack traces, the current function is retrieved from PC, and its
callees are retrieved by traversing the stack backwards. Thus, for
stubs, the stub itself was discovered via PC, and then stub's caller's
caller was retrieved from stack.
To fix this problem, a pointer to JSFunction object is now captured
from the topmost stack frame, and is saved into stack trace log
record. Then a simple heuristics is applied whether a referred
function should be added to decoded stack, or not, to avoid reporting
the same function twice (from PC and from the pointer.)
BUG=553
TEST=added to mjsunit/tools/tickprocessor
Review URL: http://codereview.chromium.org/546089
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3673 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
MSVS names '.map' file using only module's name, so both 'a.exe' and 'a.dll' will have 'a.map' file. To distinguish an originating module, we're now checking for image base which is always 00400000 for .exe files, and not 00400000 for .dlls.
Verified that windows-tick-processor can now process logs from Chromium using .map file generated for 'chrome.dll', an that it still works for V8's 'shell.exe'.
Review URL: http://codereview.chromium.org/172044
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2699 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
'nm' is now called with an option to report function code sizes. Static code entries are restricted to the sizes reported, and the remaining unnamed code is attributed to a library as a whole. This makes reports more accurate, as some functions are tiny, but has chunks of unnamed code behind them.
This change doesn't affect reporting on Windows, as in .map files function code sizes aren't specified.
Review URL: http://codereview.chromium.org/149513
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@2455 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
