Vms x86 #108
Vms x86 #108
Conversation
|
Hi Bill, Thanks for PR. I'm actually working on VMS x86 port in the last days :) Looking into the long size issue right now, the rest was quite trivial. I did not do much changes in build_vms.com, though, maybe a few lines. I suggest that I merge my changes into your PR, as most of them are common. I have only access to Alpha and x86 in emulators, cannot test on IA64. Did IEEE floating denormalisation cause any issues on x86? |
|
Alexander, The only test issues I had were the 3 that were already failing, plus the one caused by the size mismatch. |
C++ compiler has different size of long types compared to C compiler. Apparently, there is no easy way to fix this incompatible behaviour.
|
|
Do the 3 tests still fail on vms?
bill
…On Wed, May 15, 2024 at 7:22 PM Alexander Saprykin ***@***.***> wrote:
Merged #108 <#108> into master.
—
Reply to this email directly, view it on GitHub
<#108 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNZYPCWMRV3Y46NKBZZDZDZCPUZRAVCNFSM6AAAAABHYUMMDWVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJSHAZDIOBRGE3TCMA>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Yes. The only difference is that semaphore test passes one time when compiled with 32-bit pointers. Each next time it fails. |
I did this port of the library to vms_x86 about 6 months ago. Not many changes. I did rewrite the vms_build procedure. I held off requesting a merge to await a fix for the 3 failing tests. I commented in a group somewhere and got a reply from a developer that sounded like it would be fixed in the next release, but I was not able to confirm the fix. I also noted that the one new test failure was due to C and C++ disagreeing on sizes.
My VMS_x86 license expired a month ago and I am still awaiting a new license.
Feel free to reject this if you have already done the port.
bill