Re: [HACKERS] OK, so culicidae is *still* broken - Mailing list pgsql-hackers

From Tom Lane
Subject Re: [HACKERS] OK, so culicidae is *still* broken
Date
Msg-id 18989.1492721360@sss.pgh.pa.us
Whole thread Raw
In response to Re: [HACKERS] OK, so culicidae is *still* broken  (Andres Freund <andres@anarazel.de>)
List pgsql-hackers
Andres Freund <andres@anarazel.de> writes:
> On 2017-04-20 16:57:03 +0530, Amit Kapila wrote:
>> Agreed.  I have done some further study by using VMMap tool in Windows
>> and it seems to me that all 64-bit processes use address range
>> (0000000000010000 ~ 000007FFFFFE0000).  I have attached two screen
>> shots to show the address range (lower range and upper range).  You
>> need to refer the lower half of the window in attached screenshots.
>> At this stage, I am not completely sure whether we can assume some
>> address out of this range to use in MapViewOfFileEx.  Let me know your
>> thoughts.

> That seems like a fairly restricted range (good!), and squares with
> memories of reading about this (can't find the reference anymore).  Just
> using something like 0x00000F0000000000 as the address might work
> sufficiently well.  What happens if you just hardcode that in the first
> MapViewOfFileEx call, and change RandomizedBaseAddress="FALSE" in
> src/tools/msvc/VCBuildProject.pm to true?

The material I found about Linux x86_64 addressing explains that the
underlying hardware address space is only 48 bits.  Linux chooses to
normalize this by sign-extending into the upper 16 bits, so that
valid 64-bit addresses are0000000000000000 .. 00007FFFFFFFFFFF
andFFFF800000000000 .. FFFFFFFFFFFFFFFF

If I'm counting the bits correctly, Windows is choosing to use only
1/16th of the lower half of the available address space, which seems
a bit odd.  However, the main point here is that there should be
quite a bit of available daylight, if they will let us use addresses
above 000007FFFFFE0000 at all.  I'd be inclined to do our own low-tech
ASLR by using addresses with a random component, say 00000Fxxxxxx0000.
        regards, tom lane



pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: [HACKERS] [POC] hash partitioning
Next
From: Claudio Freire
Date:
Subject: Re: [HACKERS] Vacuum: allow usage of more than 1GB of work mem