Horton

Re: OK, so culicidae is *still* broken

Craig Ringer <craig@2ndquadrant.com>

View thread View on postgr.es
From: Craig Ringer <craig@2ndquadrant.com>
To: Andres Freund <andres@anarazel.de>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2017-04-24T12:40:18Z
Lists: pgsql-hackers 
On 16 April 2017 at 05:18, Andres Freund <andres@anarazel.de> wrote:

> Because of ASLR of the main executable (i.e. something like PIE).  It'll
> supposedly become harder (as in only running in compatibility modes) if
> binaries don't enable that.  It's currently disabled somewhere in the VC
> project generated.

I thought we passed /DYNAMICBASE:NO directly , but I don't see that in
the code. A look at the git logs shows that we disabled it in
7f3e17b48 by emitting
<RandomizedBaseAddress>false</RandomizedBaseAddress> in the MSBuild
project. That'll pass /DYNAMICBASE:NO to the linker.

See https://msdn.microsoft.com/en-us/library/bb384887.aspx

It's rather better than the old registry hack, but it's a compat
option we're likely to lose at some point.

-- 
 Craig Ringer                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

Commits

  1. Provide a way to control SysV shmem attach address in EXEC_BACKEND builds.

  2. Disable RandomizedBaseAddress on MSVC builds