With this change, the tool will find and mark reordered packets in the
plot. Furthermore, the instantaneous send bitrate will be correct even
for reordered packets.
BUG=webrtc:2692
R=tina.legrand@webrtc.org
Review URL: https://codereview.webrtc.org/1172533004.
Cr-Commit-Position: refs/heads/master@{#9443}
One of the unit tests added in
https://webrtc-codereview.appspot.com/50079004/ is failing
on Windows since os.sep is a backslash on Windows.
The code is based on the contents of the DEPS file rather than
the filesystem, so the right thing is to use '/' instead of os.sep.
The PyLint blacklist also didn't work on Windows, causing it
to process a massive list of files during presubmit.
I also added a bunch of new entries to speed up lint execution on
all platforms.
TESTED=Ran the presubmit with this CL on Windows and Linux.
R=phoglund@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/52029004
Cr-Commit-Position: refs/heads/master@{#9353}
Registers transport on construction removing the need for ViESender as a
hop and removing a potential deadlock by removing RegisterSendTransport.
BUG=1695, 2999
R=stefan@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/57449004
Cr-Commit-Position: refs/heads/master@{#9309}
The purpose of the tool is to analyze the output from the command line
tool rtp_analyze. That is, starting with an rtpdump or pcap file, it
is processed through rtp_analyze to produce a text output, which is
then used as input to this new Matlab function.
BUG=2692
TBR=tina.legrand@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/47339004
Cr-Commit-Position: refs/heads/master@{#9306}
The previous script only looked up the LKGR and generated a
commit message. This CL renames the script and makes it update
the DEPS file, commit locally, upload CL and send tryjobs.
BUG=4688
R=phoglund@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/50079004
Cr-Commit-Position: refs/heads/master@{#9284}
Add pylintrc file based on
https://code.google.com/p/chromium/codesearch#chromium/src/tools/perf/pylintrc
bit tightened up quite a bit (the one in depot_tools is far
more relaxed).
Remove a few excluded directories from pylint check and fixed/
suppressed all warnings generated.
Add GN format check + formatted all GN files using 'gn format'.
Cleanup redundant rules in tools/PRESUBMIT.py
TESTED=Ran 'git cl presubmit -vv', fixed the PyLint violations.
Ran it again with a modification in webrtc/build/webrtc.gni, formatted
all the GN files and ran it again.
R=henrika@webrtc.org, phoglund@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/50069004
Cr-Commit-Position: refs/heads/master@{#9274}
Having nice CL descriptions that explains what a roll of
chromium_revision in WebRTC's DEPS file actually changes is
useful to save time tracing down breakages.
This script can be used to generate such CL descriptions.
In the future it might be used for a complete auto-rolling
solution for this.
BUG=4688
R=phoglund@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/27889004
Cr-Commit-Position: refs/heads/master@{#9270}
This will make it easier to track which revision is
in a certain Chrome release, since you don't have to
look up the Git hash every time.
Also rename svn_revision to commit_position to make
it more clear what the number is in the post-SVN era.
TESTED=tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /ssd/chrome/src --verbose --ignore-checks --dry-run --close-previous-roll
and verified in the modified DEPS file that the comment was set.
R=tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/49439004
Cr-Commit-Position: refs/heads/master@{#8756}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8756 4adac7df-926f-26a2-2b94-8c16560cd09d
The --close-previous-roll makes it possible to always
close a previously created roll when creating a new one.
This way it will be possible to avoid getting a pile of
open CLs created and never closed for all failed
roll attempts, which is useful for automation.
I also moved some variables out of the AutoRoller
class that doesn't neeed to be there.
BUG=chromium:433305
TESTED=Ran:
tools/autoroller/roll_webrtc_in_chromium.py --verbose --close-previous-roll
and verified it actually closed an existing roll.
R=tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/40349004
Cr-Commit-Position: refs/heads/master@{#8726}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8726 4adac7df-926f-26a2-2b94-8c16560cd09d
Add a --wait-for-trybots flag that will have the script look for
a previously created roll and poll the trybot status until at least
one trybot has failed or all have compelted successfully.
BUG=chromium:433305
TESTED=Ran:
tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /ssd/chrome/src --verbose
tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /ssd/chrome/src --verbose --wait-for-trybots
R=tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/45659004
Cr-Commit-Position: refs/heads/master@{#8718}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8718 4adac7df-926f-26a2-2b94-8c16560cd09d
Send the created CL to the CQ right away upon creation.
Add a --no-commit flag that can be specified to avoid that.
Remove the trybot status feature (--status) and the commit CL
feature (--commit).
BUG=chromium:433305
TESTED=Ran:
tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /ssd/chrome/src --dry-run --ignore-checks
and verified it didn't trigger tryjobs or sent to CQ. Then I ran:
tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /ssd/chrome/src --ignore-checks
and verified it was sent to CQ.
R=tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/42739004
Cr-Commit-Position: refs/heads/master@{#8715}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8715 4adac7df-926f-26a2-2b94-8c16560cd09d
* Add optional --chromium-checkout flag to point out the Chromium checkout to
work in (automatic fallback to current working dir if that's a Chromium checkout).
* Introduce logging and --verbose flag.
* Make functions private and adopt Python style guide.
* Made --dry-run perform the presubmit checks and other stuff for --commit, except the actual commit.
* Add EDITOR=true environment to 'git cl upload' to avoid launching an interactive editor.
* Removed interactive invocation of 'git cl upload' and 'git cl land'
BUG=chromium:433305
TESTED=Muliple local runs using:
tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /path/to/chrome/src --dry-run --verbose
tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /path/to/chrome/src --dry-run --abort
tools/autoroller/roll_webrtc_in_chromium.py --chromium-checkout /path/to/chrome/src --dry-run --status
cd /path/to/chrome/src
/path/to/webrtc/src/tools/autoroller/roll_webrtc_in_chromium.py --dry-run --verbose
R=phoglund@webrtc.org, tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/46559004
Cr-Commit-Position: refs/heads/master@{#8694}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8694 4adac7df-926f-26a2-2b94-8c16560cd09d
Adding a pristine copy (only changed license and ownerless
TODOs) of the script we've used to roll the WebRTC revision
in Chromium DEPS.
This script assumes being located inside a Chromium checkout,
so a follow-up CL will be created for review of the required
changes for that before it can be used.
R=tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/47499004
Cr-Commit-Position: refs/heads/master@{#8675}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8675 4adac7df-926f-26a2-2b94-8c16560cd09d
The roll seems to cause leaks on our Linux Memcheck bot.
Added a suppression needed for Trusty in order to run
memcheck similar to the bot (that runs Precise).
Leave all the other source code edits from r8596 in place.
See also http://chromegw/i/client.webrtc/builders/Linux%20Memcheck/builds/3343TBR=pbos@webrtc.org
TESTED=Can no longer repro memcheck failure with this patch applied:
GYP_DEFINES="build_for_tool=memcheck" webrtc/build/gyp_webrtc
ninja -C out/Release libjingle_peerconnection_unittest
tools/valgrind-webrtc/webrtc_tests.sh --test libjingle_peerconnection_unittest --tool memcheck --target Release --build-dir out --gtest_filter=WebRtcSessionTest.TestIncorrectMLinesInLocalAnswer
Review URL: https://webrtc-codereview.appspot.com/47419004
Cr-Commit-Position: refs/heads/master@{#8612}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8612 4adac7df-926f-26a2-2b94-8c16560cd09d
Without this patch, Valgrind's Memcheck was complaining that the test
for whether we should return -1 following the call to
WebRtcIsac_DecodeLb made a conditional branch or move based on the
value of numSamplesLB, which was uninitialized if WebRtcIsac_DecodeLb
failed.
However, as can be seen in the source, the control flow only depends
on the value of numSamplesLB if numDecodedBytesLB >= 0; i.e., if
WebRtcIsac_DecodeLb returned successfully, in which case numSamplesLB
is always initialized. The discrepancy is due to the fact that
Valgrind works on the generated machine code, which contains spurious
such dependencies. The generated code for this test:
if ((numDecodedBytesLB < 0) || (numDecodedBytesLB > lenEncodedLBBytes) ||
(numSamplesLB > MAX_FRAMESAMPLES)) {
looks like this:
95: 0f bf 45 d6 movswl -0x2a(%rbp),%eax
99: 3d c0 03 00 00 cmp $0x3c0,%eax
9e: 0f 8f 45 01 00 00 jg 1e9 <Decode+0x1e9>
a4: 44 89 f0 mov %r14d,%eax
a7: c1 e0 10 shl $0x10,%eax
aa: 0f 88 39 01 00 00 js 1e9 <Decode+0x1e9>
b0: 41 0f bf ce movswl %r14w,%ecx
b4: 89 8d 98 e1 ff ff mov %ecx,-0x1e68(%rbp)
ba: 41 0f bf c7 movswl %r15w,%eax
be: 39 c1 cmp %eax,%ecx
c0: 0f 8f 23 01 00 00 jg 1e9 <Decode+0x1e9>
Note how the compiler has seemingly ignored the C language's guarantee
that the arguments to || must be evaluated in left-to-right order, and
compares numSamplesLB (%eax) with MAX_FRAMESAMPLES (0x3c0, a.k.a. 960)
before the other two conditions! If the uninitialized value in
numSamplesLB happens to be greater than 960, we'll jump to
Decode+0x1e9 (where we'll return -1) without even looking at the other
two conditions. Has the compiler generated broken code?
Well, no. If numDecodedBytesLB is < 0 so that numSamplesLB is
uninitialized, we'll end up jumping to 1e9 whether that value is
greater than 960 or not; we'll just do it with different jump
instructions. This is entirely invisible as far as the C language is
concerned, but the dependency on the uninitialized value is visible at
the machine code level, which is why Memcheck complains.
This patch solves the problem by pragmatically initializing
numSamplesLB before the call even though it isn't necessary other than
for placating Memcheck.
BUG=4143
R=henrik.lundin@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/36309004
Cr-Commit-Position: refs/heads/master@{#8492}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8492 4adac7df-926f-26a2-2b94-8c16560cd09d
This test is too slow to execute:
[ RUN ] EndToEndTest.SendsAndReceivesH264
e:\b\build\slave\win-drmem\build\src\webrtc\video\end_to_end_tests.cc(287): error: Value of: Wait()
Actual: 3
Expected: kEventSignaled
Which is: 1
Timed out while waiting for enough frames to be decoded.
[ FAILED ] EndToEndTest.SendsAndReceivesH264 (72812 ms)
BUG=3159
R=pbos@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/38599004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8063 4adac7df-926f-26a2-2b94-8c16560cd09d
