We have over 150 casinos that are using our system running aix 32 bit mode with huge 32 bit DB2 databases. A lot are still using AIX 5.2 which isn't even supported by IBM any more, but we can't seem to get them to move up to newer versions of anything. So we still have to support aix 5.2 32 bit kernels

anyway, with your patch I recompiled everything and ran the make test, which did complete all tests successfully. Ithen recompiled the latest openssh using the new openssl compile and reran my original tests and the ssh is now working as expected.

dean

On 05/16/2012 05:45 AM, Andy Polyakov via RT wrote:
I rebooted the box, a IBM 44 Model 170 with a "PowerPC_POWER3" CPU with
"realmem  2097152",   just to make sure there was nothing else affecting
the compile.

I then recompiled using the "./config" with no flags.  I then ran the
"make test" and it got into an endless loop (attachment maketest3.script)

I then set the environmental flag ( export OPENSSL_ppccap=0 ) as
suggested and ran the "make test" again. (attachment maketest.script).
This time the test ran to completion with "ALL TESTS SUCCESSFUL"
Power3 is 64-bit capable CPU, but my understanding is that AIX 5.x
kernel is available in two flavors, 32- and 64-bit ones. With this in
mind it's possible that your system is configured to boot "pure" 32-bit
kernel. If this is the case, then it can "feel" no obligation to
preserve upper halves of general-purpose registers upon context switch.
Latter means that otherwise kernels were observed to preserve upper
halves *even* for 32-bit applications, and module in question,
ppc64-mont.pl, actually relies on it. Can you confirm that
http://cvs.openssl.org/chngview?cn=22574 fixes the problem?



<<attachment: DCarter.vcf>>

Reply via email to