Roll src/third_party/libsrtp 6446144:f0957e8 (svn 292694:295151)

Roll src/third_party/libsrtp 6446144:9c53f85 (svn 292694:295151)

This updates libsrtp to 1.5.2 and uses the OpenSSL/BoringSSL crypto code.
See https://codereview.chromium.org/889083003/ and https://codereview.chromium.org/1098043003/

Summary of changes available at:
https://chromium.googlesource.com/chromium/deps/libsrtp/+log/6446144..9c53f85

BUG=328475
Committed: https://crrev.com/94c5e80d3a385e223279b8ddb3ed8aa19d69a0e9
Cr-Commit-Position: refs/heads/master@{#328972}

[fancycode, 2015-03-04 17:25:47]

mail@joachim-bauch.de changed reviewers:
+ jiayl@chromium.org, juberti@chromium.org

[fancycode, 2015-03-04 17:25:48]

This is the followup of https://codereview.chromium.org/889083003/ to get the
libsrtp update into Chromium.

[fancycode, 2015-03-04 17:27:06]

The CQ bit was checked by mail@joachim-bauch.de

[fancycode, 2015-03-04 17:27:06]

The CQ bit was unchecked by mail@joachim-bauch.de

[juberti2, 2015-03-04 21:30:13]

On 2015/03/04 17:25:48, fancycode wrote:
> This is the followup of https://codereview.chromium.org/889083003/ to get the
> libsrtp update into Chromium.

lgtm, you'll need jiayl lgtm though

[jiayl, 2015-03-04 21:41:33]

Can you make a svn checkout instead of git? That will make committing the change
easier since the git repo does not allow direct submit.

To checkout svn, create .gclient in the root dir with:

solutions = [
  {
    "managed": False,
    "name": "src",
    "url": "svn://svn.chromium.org/chrome/trunk/deps/third_party/libsrtp",
    "custom_deps": {},
    "deps_file": "DEPS",
    "safesync_url": "",
  },
]

Run 'gclient sync', make your changes, run 'gcl change XXX' and 'gcl upload XXX'
to upload the patch.

[fancycode, 2015-03-04 21:51:49]

On 2015/03/04 21:41:33, jiayl wrote:
> Can you make a svn checkout instead of git? That will make committing the
change
> easier since the git repo does not allow direct submit.

Unfortunately I'm not allowed to access this SVN url (what should I use as
username/password?):
svn: E170001: Unable to connect to a repository at URL
'svn://svn.chromium.org/chrome/trunk/deps/third_party/libsrtp'
svn: E170001: Can't get username or password

Also from what I understand, this would checkout the libsrtp dependency, right?
The changes already landed there in https://codereview.chromium.org/936663005/
so I'm now trying to roll the new version in Chromium with this CL.

[jiayl, 2015-03-04 21:53:11]

On 2015/03/04 21:51:49, fancycode wrote:
> On 2015/03/04 21:41:33, jiayl wrote:
> > Can you make a svn checkout instead of git? That will make committing the
> change
> > easier since the git repo does not allow direct submit.
> 
> Unfortunately I'm not allowed to access this SVN url (what should I use as
> username/password?):
> svn: E170001: Unable to connect to a repository at URL
> 'svn://svn.chromium.org/chrome/trunk/deps/third_party/libsrtp'
> svn: E170001: Can't get username or password
> 
> Also from what I understand, this would checkout the libsrtp dependency,
right?
> The changes already landed there in https://codereview.chromium.org/936663005/
> so I'm now trying to roll the new version in Chromium with this CL.

Sorry, I misunderstood.
lgtm

[jiayl, 2015-03-04 21:53:15]

The CQ bit was checked by jiayl@chromium.org

[commit-bot: I haz the power, 2015-03-04 21:54:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/981593002/1

[commit-bot: I haz the power, 2015-03-04 22:20:37]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-03-04 22:20:38]

Try jobs failed on following builders:
  android_aosp on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/android_aosp/bu...)

[fancycode, 2015-03-04 22:47:33]

On 2015/03/04 22:20:38, I haz the power (commit-bot) wrote:
> Try jobs failed on following builders:
>   android_aosp on tryserver.chromium.linux (JOB_FAILED,
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/android_aosp/bu...)

Ok, looks like the roll needs more changes in WebRTC when compiled for Chromium.
I'll create CLs for that shortly and will comment here again once they have
landed.

[fancycode, 2015-03-26 00:18:12]

On 2015/03/04 22:47:33, fancycode wrote:
> Ok, looks like the roll needs more changes in WebRTC when compiled for
Chromium.
> I'll create CLs for that shortly and will comment here again once they have
> landed.

The changes to WebRTC landed in
https://chromium.googlesource.com/external/webrtc/+/a197a5eed68ad29e42cde8305...
and rolled in Chromium in
https://chromium.googlesource.com/chromium/src.git/+/851489733b9a58323a034cb9...

I rebased the patch and successfully compiled Chromium on Linux, could you
please trigger the try bots again to check the other platforms?

[jiayl, 2015-03-26 18:19:14]

The CQ bit was checked by jiayl@chromium.org

[jiayl, 2015-03-26 18:19:14]

The patchset sent to the CQ was uploaded after l-g-t-m from jiayl@chromium.org,
juberti@chromium.org
Link to the patchset: https://codereview.chromium.org/981593002/#ps20001 (title:
"rebased")

[commit-bot: I haz the power, 2015-03-26 18:19:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/981593002/20001

[commit-bot: I haz the power, 2015-03-26 18:32:32]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-03-26 18:32:33]

Try jobs failed on following builders:
  linux_chromium_gn_dbg on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[jiayl, 2015-03-26 18:37:32]

On 2015/03/26 18:32:33, I haz the power (commit-bot) wrote:
> Try jobs failed on following builders:
>   linux_chromium_gn_dbg on tryserver.chromium.linux (JOB_FAILED,
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

Compile failed:

FAILED: /b/build/goma/gomacc
../../third_party/llvm-build/Release+Asserts/bin/clang++ -Wl,--fatal-warnings
-m64 -fPIC -Wl,-z,noexecstack -Wl,-z,now -Wl,-z,relro -Wl,-z,defs -pthread
-B../../third_party/binutils/Linux_x64/Release/bin -fuse-ld=gold -Wl,--icf=safe
-Wl,-rpath=\$ORIGIN/ -Wl,-rpath-link= -Wl,--disable-new-dtags -o ./rtpw
-Wl,--start-group @./rtpw.rsp  -Wl,--end-group  -ldl 
../../third_party/libsrtp/srtp/test/rtpw.c:456:error: undefined reference to
'base64_string_to_octet_string'
clang:error: linker command failed with exit code 1 (use -v to see invocation)
ninja: build stopped: subcommand failed.

[fancycode, 2015-03-26 23:25:13]

Thanks for triggering the bots, I noticed I missed updating the GN build script
:-/

This is done now to match the changes in the GYP script from
https://codereview.chromium.org/936663005/ and works with GN for me on Linux
Precise.

Could you please trigger again?

[jiayl, 2015-03-26 23:49:23]

On 2015/03/26 23:25:13, fancycode wrote:
> Thanks for triggering the bots, I noticed I missed updating the GN build
script
> :-/
> 
> This is done now to match the changes in the GYP script from
> https://codereview.chromium.org/936663005/ and works with GN for me on Linux
> Precise.
> 
> Could you please trigger again?

It now needs lgtm from an owner of the new file.

BTW, you can also trigger the commit queue from the web UI.

[fancycode, 2015-03-26 23:52:08]

mail@joachim-bauch.de changed reviewers:
+ thakis@chromium.org

[fancycode, 2015-03-26 23:54:36]

On 2015/03/26 23:49:23, jiayl wrote:
> It now needs lgtm from an owner of the new file.

Thanks for the heads-up. 

thakis@, I added you to the reviewers list as suggested by git-cl, could you
please take a look?

> BTW, you can also trigger the commit queue from the web UI.

Oh, I wasn't aware of that - I thought you need special commit permissions...

[Nico, 2015-03-26 23:59:16]

lgtm, but if that switch isn't necessary I'd omit it.

https://codereview.chromium.org/981593002/diff/80001/build/secondary/third_pa...
File build/secondary/third_party/libsrtp/BUILD.gn (right):

https://codereview.chromium.org/981593002/diff/80001/build/secondary/third_pa...
build/secondary/third_party/libsrtp/BUILD.gn:7: use_srtp_openssl = true
Does anything ever set this to false? Should it be called "use_srtp_boringssl"?

[fancycode, 2015-03-27 00:09:23]

On 2015/03/26 23:59:16, Nico wrote:
> lgtm, but if that switch isn't necessary I'd omit it.

Thanks, I wanted to keep this compatible with the GYP version where
OpenSSL/BoringSSL is optional. Also third-party users of the libsrtp/WebRTC code
might want to use the builtin libsrtp crypto code for whatever reason, so they
can override the flag.

>
https://codereview.chromium.org/981593002/diff/80001/build/secondary/third_pa...
> File build/secondary/third_party/libsrtp/BUILD.gn (right):
> 
>
https://codereview.chromium.org/981593002/diff/80001/build/secondary/third_pa...
> build/secondary/third_party/libsrtp/BUILD.gn:7: use_srtp_openssl = true
> Does anything ever set this to false? Should it be called
"use_srtp_boringssl"?

Renamed flag to "use_srtp_boringssl".

[fancycode, 2015-03-27 00:09:29]

The CQ bit was checked by mail@joachim-bauch.de

[fancycode, 2015-03-27 00:09:29]

The patchset sent to the CQ was uploaded after l-g-t-m from
juberti@chromium.org, jiayl@chromium.org, thakis@chromium.org
Link to the patchset: https://codereview.chromium.org/981593002/#ps100001
(title: "Renamed flag.")

[commit-bot: I haz the power, 2015-03-27 00:10:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/981593002/100001

[commit-bot: I haz the power, 2015-03-27 00:46:41]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-03-27 00:46:42]

Try jobs failed on following builders:
  win8_chromium_rel on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_rel...)

[fancycode, 2015-05-07 17:10:43]

I updated the CL to use the latest changes from libsrtp which should hopefully
fix the remaining Windows issues. Also the GN script has been updated to match
the GYP version.

Could you ptal?

[fancycode, 2015-05-08 07:37:42]

The CQ bit was checked by mail@joachim-bauch.de to run a CQ dry run

[fancycode, 2015-05-08 07:37:43]

The patchset sent to the CQ was uploaded after l-g-t-m from jiayl@chromium.org,
thakis@chromium.org, juberti@chromium.org
Link to the patchset: https://codereview.chromium.org/981593002/#ps120001
(title: "Roll to libsrtp 1.5.2")

[commit-bot: I haz the power, 2015-05-08 07:37:53]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/981593002/120001

[commit-bot: I haz the power, 2015-05-08 09:03:01]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-05-08 09:03:02]

Dry run: This issue passed the CQ dry run.

[fancycode, 2015-05-08 09:07:13]

On 2015/05/08 09:03:02, I haz the power (commit-bot) wrote:
> Dry run: This issue passed the CQ dry run.

Just noticed I can trigger a CQ dry run.

All looks good, however I'm not sure if I need to wait on another lgtm to
actually commit the change...

[Nico, 2015-05-08 17:03:28]

jiayl: The "ptal" was for you, I think.

(fancycode: When saying "ping" or "ptal" or something, always prefix that with
who you expect to do that)

[jiayl, 2015-05-08 17:06:48]

lgtm

[jiayl, 2015-05-08 17:06:56]

The CQ bit was checked by jiayl@chromium.org

[commit-bot: I haz the power, 2015-05-08 17:08:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/981593002/120001

[commit-bot: I haz the power, 2015-05-08 17:26:01]

Committed patchset #7 (id:120001)

[commit-bot: I haz the power, 2015-05-08 17:26:59]

Patchset 7 (id:??) landed as
https://crrev.com/94c5e80d3a385e223279b8ddb3ed8aa19d69a0e9
Cr-Commit-Position: refs/heads/master@{#328972}

[Garrett Casto, 2015-05-08 17:48:54]

A revert of this CL (patchset #7 id:120001) has been created in
https://codereview.chromium.org/1131323002/ by gcasto@chromium.org.

The reason for reverting is: Broke the Windows GN bots
(https://build.chromium.org/p/chromium.win/builders/Win8%20GN).

[Nico, 2015-05-08 17:53:26]

Thanks for the revert. That's the 2nd time in 2 days that bot breaks after
the CQ landed something. Filed
http://code.google.com/p/chromium/issues/detail?id=486089 for that.

On Fri, May 8, 2015 at 10:48 AM, <gcasto@chromium.org> wrote:

> A revert of this CL (patchset #7 id:120001) has been created in
> https://codereview.chromium.org/1131323002/ by gcasto@chromium.org.
>
> The reason for reverting is: Broke the Windows GN bots
> (https://build.chromium.org/p/chromium.win/builders/Win8%20GN).
>
> https://codereview.chromium.org/981593002/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Nico, 2015-05-14 20:17:58]

On 2015/05/08 17:48:54, Garrett Casto wrote:
> A revert of this CL (patchset #7 id:120001) has been created in
> https://codereview.chromium.org/1131323002/ by mailto:gcasto@chromium.org.
> 
> The reason for reverting is: Broke the Windows GN bots
> (https://build.chromium.org/p/chromium.win/builders/Win8%20GN).

(gcasto: In the future, please link to the failing build itself, not just to the
bot. I'm trying to reproduce this failure atm, and now I have to hunt around to
find it.)

[fancycode, 2015-05-14 20:32:52]

If I remember correctly, problem was about not finding some include in
"openssl/" while compiling srtpfilter from webrtc on that GN bot.

> Am 14.05.2015 um 22:17 schrieb thakis@chromium.org:
> 
>> On 2015/05/08 17:48:54, Garrett Casto wrote:
>> A revert of this CL (patchset #7 id:120001) has been created in
>> https://codereview.chromium.org/1131323002/ by mailto:gcasto@chromium.org.
> 
>> The reason for reverting is: Broke the Windows GN bots
>> (https://build.chromium.org/p/chromium.win/builders/Win8%20GN).
> 
> (gcasto: In the future, please link to the failing build itself, not just to
the
> bot. I'm trying to reproduce this failure atm, and now I have to hunt around
to
> find it.)
> 
> https://codereview.chromium.org/981593002/
> 

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Nico, 2015-05-14 20:40:51]

Thanks! Building target "libjingle_webrtc_common" does repro the problem
for me:

C:\src\chrome\src>ninja -C out\gn libjingle_webrtc_common
ninja: Entering directory `out\gn'
[89/373] CXX
obj/third_party/libjingle/source/talk/session/media/libjingle_webrtc_common.externalhmac.obj
FAILED: ninja -t msvc -e environment.x86 --
"c:\src\depot_tools\win_toolchain\vs2013_files\VC\bin\amd64_x86/cl.exe"
/nologo /showIncludes /F
C
@obj/third_party/libjingle/source/talk/session/media/libjingle_webrtc_common.externalhmac.obj.rsp
/c ../../third_party/libjingle/source/ta
lk/session/media/externalhmac.cc
/Foobj/third_party/libjingle/source/talk/session/media/libjingle_webrtc_common.externalhmac.obj
/Fdobj/thir
d_party/libjingle/libjingle_webrtc_common_cc.pdb
c:\src\chrome\src\third_party\libsrtp\srtp\crypto\include\aes_icm_ossl.h(50)
: fatal error C1083: Cannot open include file: 'openssl/evp.h':
 No such file or directory
[89/373] CXX
obj/third_party/libjingle/source/talk/session/media/libjingle_webrtc_common.channel.obj

I'll try your new patch now.



On Thu, May 14, 2015 at 1:32 PM, Joachim Bauch <mail@joachim-bauch.de>
wrote:

> If I remember correctly, problem was about not finding some include in
> "openssl/" while compiling srtpfilter from webrtc on that GN bot.
>
> > Am 14.05.2015 um 22:17 schrieb thakis@chromium.org:
> >
> >> On 2015/05/08 17:48:54, Garrett Casto wrote:
> >> A revert of this CL (patchset #7 id:120001) has been created in
> >> https://codereview.chromium.org/1131323002/ by mailto:
> gcasto@chromium.org.
> >
> >> The reason for reverting is: Broke the Windows GN bots
> >> (https://build.chromium.org/p/chromium.win/builders/Win8%20GN).
> >
> > (gcasto: In the future, please link to the failing build itself, not
> just to the
> > bot. I'm trying to reproduce this failure atm, and now I have to hunt
> around to
> > find it.)
> >
> > https://codereview.chromium.org/981593002/
> >
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.