Update Bitcoin to use OpenSSL 1.0.1c #1837

fanquake commented at 3:32 AM on September 19, 2012: member

I noticed that 0.7.0 for OSX is compiled with 1.0.1c, so I thought it might be an idea to update the makefiles, readme etc. I'm assuming that theres no reason that unix and windows builds cannot use 1.0.1b.

Update to use OpenSSL 1.0.1c f4b92c8ede

Diapolo commented at 6:43 AM on September 19, 2012: none

I've been using 1.0.1c for my own Windows builds for months and have not found any problems with it.

Update msw build notes with boost 1.5.0 b7a8802c2a

in doc/build-msw.txt:None in f4b92c8ede outdated

  23 | @@ -24,7 +24,7 @@ Dependencies
  24 |  Libraries you need to download separately and build:
  25 |  
  26 |                  default path               download
  27 | -OpenSSL         \openssl-1.0.1b-mgw        http://www.openssl.org/source/
  28 | +OpenSSL         \openssl-1.0.1c-mgw        http://www.openssl.org/source/
  29 |  Berkeley DB     \db-4.8.30.NC-mgw          http://www.oracle.com/technology/software/products/berkeley-db/index.html
  30 |  Boost           \boost-1.47.0-mgw          http://www.boost.org/users/download/

Diapolo commented at 6:44 AM on September 19, 2012:

We are at 1.50 for Windows AFAIK? Perhaps you can create another small patch for this.

Diapolo commented at 8:52 AM on September 19, 2012: none

Thanks :), looks good.

BitcoinPullTester commented at 8:29 PM on September 19, 2012: none

Automatic sanity-testing: FAILED BUILD/TEST, see http://jenkins.bluematt.me/pull-tester/b7a8802c2acd61394b63856d7ff7ca942df81a00 for binaries and test log.

This could happen for one of several reasons:

It chanages paths in makefile.linux-mingw or otherwise changes build scripts in a way that made them incompatible with the automated testing scripts
It does not build on either Linux i386 or Win32 (via MinGW cross compile)
The test suite fails on either Linux i386 or Win32
The block test-cases failed (lookup the first bNN identifier which failed in https://github.com/TheBlueMatt/test-scripts/blob/master/FullBlockTestGenerator.java)

laanwj commented at 10:23 AM on September 20, 2012: member

Please increase the deps version number, ie to bitcoin-deps-0.0.5.zip (both in deps-win32.yml and gitian-win32.yml)

Bump deps version number to 0.0.5 05c1c08e29

BitcoinPullTester commented at 12:00 AM on September 21, 2012: none

Automatic sanity-testing: FAILED BUILD/TEST, see http://jenkins.bluematt.me/pull-tester/05c1c08e295f067a5fbd32fb7ad80fe01d2aa49d for binaries and test log.

This could happen for one of several reasons:

It chanages paths in makefile.linux-mingw or otherwise changes build scripts in a way that made them incompatible with the automated testing scripts
It does not build on either Linux i386 or Win32 (via MinGW cross compile)
The test suite fails on either Linux i386 or Win32
The block test-cases failed (lookup the first bNN identifier which failed in https://github.com/TheBlueMatt/test-scripts/blob/master/FullBlockTestGenerator.java)

laanwj commented at 5:33 AM on September 21, 2012: member

Ok great,

Looks like @TheBlueMatt's has to install openssl1.0.1c in /mnt/mingw/openssl-1.0.1c, before the pull tester can test this.

Correct 1.5.0 -> 1.50.0 9cde889992

in doc/build-msw.txt:None in 05c1c08e29 outdated

  35 | @@ -36,9 +36,9 @@ Boost          MIT-like license
  36 |  miniupnpc      New (3-clause) BSD license
  37 |  
  38 |  Versions used in this release:
  39 | -OpenSSL      1.0.1b
  40 | +OpenSSL      1.0.1c
  41 |  Berkeley DB  4.8.30.NC
  42 | -Boost        1.47.0
  43 | +Boost        1.5.0

sipa commented at 12:30 PM on September 21, 2012:

I think 1.5.0 is not the same thing as 1.50.0.

Diapolo commented at 12:37 PM on September 21, 2012:

You are right, should be 1.50.0.

fanquake commented at 12:39 PM on September 21, 2012:

Sorry, I'll correct that

BitcoinPullTester commented at 7:36 PM on September 21, 2012: none

Automatic sanity-testing: FAILED BUILD/TEST, see http://jenkins.bluematt.me/pull-tester/9cde889992e16527c99c262831292b944a2faabe for binaries and test log.

This could happen for one of several reasons:

It chanages paths in makefile.linux-mingw or otherwise changes build scripts in a way that made them incompatible with the automated testing scripts
It does not build on either Linux i386 or Win32 (via MinGW cross compile)
The test suite fails on either Linux i386 or Win32
The block test-cases failed (lookup the first bNN identifier which failed in https://github.com/TheBlueMatt/test-scripts/blob/master/FullBlockTestGenerator.java)

fanquake commented at 6:51 AM on September 22, 2012: member

Closing this. Will merge all into a single commit in a new pull.

fanquake closed this on Sep 22, 2012

laanwj commented at 6:55 AM on September 22, 2012: member

You don't need to close your pull for that. You can just "rebase -i", squash the commits and push -f to the current branch.

fanquake commented at 7:04 AM on September 22, 2012: member

I would have. But those commits were in the master branch, instead of a separate one. I've been meaning to start using branches for pulls for a while now, so thought I might just sort myself out and do that instead.

TheBlueMatt commented at 9:30 PM on September 22, 2012: member

1.0.0c build for pull-tester, should test properly now.

KolbyML referenced this in commit d1682ef4f6 on Dec 5, 2020

DrahtBot locked this on Sep 8, 2021