993569 - Update Mozilla 31 to use NSS 3.16.1 (once it's ready)

Status: RESOLVED FIXED

(Core :: Security: PSM, defect, P1)

Description

[Camilo Viecco (:cviecco)]

Update Mozilla 31 to use NSS 3.16.1 (once it's ready)

Starting with beta 1 today.

Comment 1

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/66b7259580f4

Comment 2

[Wan-Teh Chang]

Camilo: I forgot to create the NSS_3_16_1_BETA2 tag yesterday,
so I pushed NSS_3_16_1_BETA2 to mozilla-inbound myself:
https://hg.mozilla.org/integration/mozilla-inbound/rev/949b53bb379c

Comment 3

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/949b53bb379c

Comment 4

[Octoploid]

I get:
...Executing: c++ -Wall -Wpointer-arith -Woverloaded-virtual -Werror=return-type -Werror=int-to-pointer-cast -Wtype-limits -Wempty-body -Wsign-compare -Wno-invalid-offsetof -Wca
st-align -flto=4 -ffunction-sections -fdata-sections -march=native -fno-exceptions -fno-strict-aliasing -fno-rtti -fno-exceptions -fno-math-errno -std=gnu++0x -pthread -pipe 
-DNDEBUG -DTRIMMED -O3 -fomit-frame-pointer -o BadCertServer /var/tmp/moz-build-dir/security/manager/ssl/tests/unit/tlsserver/cmd/tmpaq1Xb1.list -lpthread -Wl,--hash-style=gn
u,--as-needed,--gc-sections,--icf=all,--icf-iterations=3 -Wl,-z,noexecstack -Wl,--build-id -Wl,-rpath-link,/var/tmp/moz-build-dir/dist/bin -Wl,-rpath-link,/usr/lib -L../../..
/../../../../dist/bin -L../../../../../../../dist/lib -L/var/tmp/moz-build-dir/dist/lib -lnspr4 -lplc4 -lplds4 -L../../../../../../../dist/bin -L../../../../../../../dist/lib
 -lcrmf -lsmime3 -lssl3 -lnss3 -lnssutil3 ../../../../../../../dist/lib/libmozalloc.so ../../../../../../pkix/test/lib/libpkixtestutil.a -Wl,--whole-archive ../../../../../..
/../dist/lib/libmozglue.a -Wl,--no-whole-archive -rdynamic -ldlmake[5]: *** [GenerateOCSPResponse] Error 1

/var/tmp/moz-build-dir/security/manager/ssl/tests/unit/tlsserver/cmd/tmpaq1Xb1.list:
    INPUT("BadCertServer.o")
    INPUT("../../../../../../pkix/Unified_cpp_security_pkix0.o")
    INPUT("../lib/Unified_cpp_unit_tlsserver_lib0.o")

c++: error: ../../../../../../pkix/test/lib/libpkixtestutil.a: No such file or directory
/var/tmp/mozilla-central/config/rules.mk:802: recipe for target 'BadCertServer' failed
make[5]: *** [BadCertServer] Error 1

Comment 5

[Wan-Teh Chang]

Octoploid: it seems that libpkixtestutil.a wasn't built successfully.
Can you search in your build output log for "pkixtestutil.cpp" and
"libpkixtestutil.a"? That should tell us if pkixtestutil.cpp has
compilation errors. Thanks.

Comment 6

[Wan-Teh Chang]

Octoploid: the build problem you ran into seems like bug 1001585.
Please try the patch in that bug.

Comment 7

[Octoploid]

Wan-Teh: Many thanks. The patch from bug 1001585 fixes the issue.

Comment 8

[Wan-Teh Chang]

NSS_3_16_1_BETA3 pushed to mozilla-inbound:
https://hg.mozilla.org/integration/mozilla-inbound/rev/d9a6a378349a

Comment 9

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/d9a6a378349a

Comment 10

[Wan-Teh Chang]

Attachment: Update to NSS_3_16_1_BETA4: disable the new Intel AES assembly code for Windows

Patch pushed to mozilla-inbound:
https://hg.mozilla.org/integration/mozilla-inbound/rev/ca5aed9f1c37

Comment 11

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/ca5aed9f1c37

Comment 12

[Wan-Teh Chang]

Comment on attachment 8414870 [details] [diff] [review]
Update to NSS_3_16_1_BETA4: disable the new Intel AES assembly code for Windows

[Approval Request Comment]
Bug caused by (feature/regressing bug #): New x86 AES assembly code
enabled in this patchset on Apr. 28 morning:
https://hg.mozilla.org/mozilla-central/rev/d9a6a378349a

User impact if declined: Potential memory corruption on Windows
computers with Intel CPUs that support the AES-NI and AVX instructions.

Testing completed (on m-c, etc.): On mozilla-central since Apr. 30.

Risk to taking this patch (and alternatives if risky): None. This reverts
the problematic change.

String or IDL/UUID changes made by this patch: None

Comment 13

[u279076]

Wan-Teh, can you have a look at bug 1005988. I'm wondering if *this* bug has caused that crash.

Comment 14

[Wan-Teh Chang]

Comment on attachment 8414870 [details] [diff] [review]
Update to NSS_3_16_1_BETA4: disable the new Intel AES assembly code for Windows

Anthony: yes, bug 1005988 will be fixed after I push this patch
to mozilla-aurora. I will do this today.

Comment 15

[Wan-Teh Chang]

Comment on attachment 8414870 [details] [diff] [review]
Update to NSS_3_16_1_BETA4: disable the new Intel AES assembly code for Windows

NSS_3_16_1_RTM and NSPR_4_10_5_RTM pushed to mozilla-aurora:
https://hg.mozilla.org/releases/mozilla-aurora/rev/0e571cde7fec

Comment 16

[Wan-Teh Chang]

NSS_3_16_1_RTM and NSPR_4_10_5_RTM pushed to mozilla-inbound:
https://hg.mozilla.org/integration/mozilla-inbound/rev/e5695a4f7ab7

Comment 17

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/e5695a4f7ab7