mbox series

[kvm-unit-tests,v2,0/6] s390x SIGP fixes

Message ID 20220311173822.1234617-1-farman@linux.ibm.com (mailing list archive)
Headers show
Series s390x SIGP fixes | expand

Message

Eric Farman March 11, 2022, 5:38 p.m. UTC 
Hi all,

Here is a second version of the SIGNAL PROCESSOR tests, addressing
the comments that were made to v1.

The biggest change is Patch 6, which has effectively been replaced.
Considering comments from Nico and Janosch, and making sure
consumers of smp_sigp get their SIGP retried, I opted to convert
smp_sigp() to perform the CC2 retry under the covers, and remove
smp_sigp_retry() itself. This effectively reverts patch 1, but I
thought the end result was cleaner and less prone to confusion
going forward. If a test needs a no-retry interface going forward,
that would be one thing, but the only places that do that are
the testing of STOP and RESTART itself, which handle this as seen
in patches 4 and 5.

So, thoughts?

v2:
 - Patch 1-2: Applied r-b from Claudio, Janosch, and Nico (thank you!)
 - [JF] Patch 3: Clarified commit message that it's dealing with semantics
   rather than a bugfix.
 - [NB] Patch 4-5: Added return code checks of the _nowait() routines
   (Claudio: I appreciate the r-b on 4, but didn't apply it because of the
   new checks that were made here)
 - [EF] Patch 4-5: Use the non-retry sigp() call directly, rather than
   smp_sigp()
v1: https://lore.kernel.org/r/20220303210425.1693486-1-farman@linux.ibm.com/

Eric Farman (6):
  lib: s390x: smp: Retry SIGP SENSE on CC2
  s390x: smp: Test SIGP RESTART against stopped CPU
  s390x: smp: Fix checks for SIGP STOP STORE STATUS
  s390x: smp: Create and use a non-waiting CPU stop
  s390x: smp: Create and use a non-waiting CPU restart
  lib: s390x: smp: Remove smp_sigp_retry

 lib/s390x/smp.c | 63 ++++++++++++++++++++++++++++++++++++++++++-------
 lib/s390x/smp.h |  3 ++-
 s390x/smp.c     | 53 +++++++++++++++++++++++++----------------
 3 files changed, 89 insertions(+), 30 deletions(-)