| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 7825.1 | io_zero() returns wrong value | SMURF::THOMPSON | | Thu Nov 14 1996 14:03 | 9 |
| 7825.2 | Any new status... | ADISSW::FERRARA | | Fri Feb 21 1997 17:10 | 20 |
|
What's up with this issue?
I am on a AXPvme 230 running UNIX 4.0b.
I too get a -1 back from io_zero, when, based on a bus analyzer
it looks like it cleared the correct REGION of I/O space.
In addition, I expected to see 32-bit transfers out on the VMEbus
when I see all BYTE transfers...this takes a long time to zero
out a megabyte or 2 of memory out in I/O space.
Has anyone encountered this behavior?
-BobF
CSS Real-time Engineering
|
| 7825.3 | io_zero return value wrong on axpvme 230 as well | SMURF::THOMPSON | | Tue Feb 25 1997 10:12 | 11 |
| The issue submitted (io_zero incorrect status on alphaserver 1000) has
been fixed in the development pools. The associated QAR has been
reassigned to the support folks for development of a patch.
I just looked through the code and it appears that io_zero is returning
no status on the axpvme 230 as well. Do you need to check the status?
We will address this so it matches the documentation, but in the
meantime can you remove the return value check from your code?
-Pete Thompson
|
| 7825.4 | What's the plan? | ADISSW::FERRARA | | Wed Feb 26 1997 13:37 | 9 |
|
Yes, I can remove the check of the return status.
What is your intended fix? To fix the code to return the
correct status or change the documentation to reflect the
current code?
-Bob
|
| 7825.5 | Fix the return value | WASTED::map | Mark Parenti, Unix Engineering Group | Mon Mar 03 1997 10:23 | 4 |
| The fix will be to return the correct status as documented.
Mark Parenti
UEG
|