[Libreoffice-commits] core.git: Branch 'libreoffice-5-3' - jvmfwk/plugins

Michael Stahl mstahl at redhat.com
Wed Mar 22 11:50:59 UTC 2017


 jvmfwk/plugins/sunmajor/pluginlib/vendorbase.cxx |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

New commits:
commit c78d39066ace32e0324365bcf10882aa6e976a03
Author: Michael Stahl <mstahl at redhat.com>
Date:   Wed Mar 22 00:35:19 2017 +0100

    jvmfwk: fix JVM detection on 64-bit Windows
    
    _WIN32 is also defined on 64-bit Windows so reorder these ifdefs.
    
    Should i be surprised that this breaks several dbaccess tests
    for me but all tinderboxes are green?
    
    (regression from 9143dd4ebe37b608e43d04434cf831624bf55b65)
    
    Change-Id: Id917952d3135768355af711688ff70bf6c019a6e
    (cherry picked from commit 2aa20cfb7a11dd8d86372af4065a5887a0b752ca)
    Reviewed-on: https://gerrit.libreoffice.org/35530
    Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt at cib.de>
    Tested-by: Jenkins <ci at libreoffice.org>

diff --git a/jvmfwk/plugins/sunmajor/pluginlib/vendorbase.cxx b/jvmfwk/plugins/sunmajor/pluginlib/vendorbase.cxx
index ef9da176ebc4..27b3ddcd7422 100644
--- a/jvmfwk/plugins/sunmajor/pluginlib/vendorbase.cxx
+++ b/jvmfwk/plugins/sunmajor/pluginlib/vendorbase.cxx
@@ -215,10 +215,10 @@ bool VendorBase::isValidArch() const
     // It is not defined what the exact values are.
     // Oracle JRE 8 has "x86" and "amd64", the others were found at http://lopica.sourceforge.net/os.html .
     // There might still be missing some options; we need to extend the check once we find out.
-#if defined _WIN32
-    return m_sArch == "x86" || m_sArch == "i386" || m_sArch == "i686";
-#elif defined _WIN64
+#if defined _WIN64
     return m_sArch == "amd64" || m_sArch == "x86_64";
+#elif defined _WIN32
+    return m_sArch == "x86" || m_sArch == "i386" || m_sArch == "i686";
 #else
     (void)this;
     return true;


More information about the Libreoffice-commits mailing list