Re: [libvirt] [PATCH 19/21] tests: Add virresctrltest
On 11/13/2017 03:50 AM, Martin Kletzander wrote:
This test has 2 different uses:
1) DO_TEST_FREE initializes capabilities from vircaps2xmldata (since it exists
there already) and then requests list of free bitmaps (all unallocated
space) from virresctrl.c
2) DO_TEST_ALLOC takes capabilities from vircaps2xmldata, and uses resctrl info
to request an allocation from virresctrl.c for a VM from
genericxml2xmlindata.
Desirable outputs are saved in virresctrldata.
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
tests/Makefile.am | 8 +-
.../resctrl--cachetune/vcpus-0-1.alloc | 1 +
.../resctrl--cachetune/vcpus-3.alloc | 1 +
.../resctrl-cdp--cachetune-cdp/vcpus-0-1.alloc | 2 +
.../resctrl-cdp--cachetune-cdp/vcpus-2.alloc | 1 +
.../resctrl-cdp--cachetune-cdp/vcpus-3.alloc | 1 +
tests/virresctrldata/resctrl-cdp.schemata | 2 +
.../virresctrldata/resctrl-skx-twocaches.schemata | 1 +
tests/virresctrldata/resctrl-skx.schemata | 1 +
tests/virresctrldata/resctrl.schemata | 1 +
tests/virresctrltest.c | 277 +++++++++++++++++++++
11 files changed, 295 insertions(+), 1 deletion(-)
create mode 100644 tests/virresctrldata/resctrl--cachetune/vcpus-0-1.alloc
create mode 100644 tests/virresctrldata/resctrl--cachetune/vcpus-3.alloc
create mode 100644 tests/virresctrldata/resctrl-cdp--cachetune-cdp/vcpus-0-1.alloc
create mode 100644 tests/virresctrldata/resctrl-cdp--cachetune-cdp/vcpus-2.alloc
create mode 100644 tests/virresctrldata/resctrl-cdp--cachetune-cdp/vcpus-3.alloc
create mode 100644 tests/virresctrldata/resctrl-cdp.schemata
create mode 100644 tests/virresctrldata/resctrl-skx-twocaches.schemata
create mode 100644 tests/virresctrldata/resctrl-skx.schemata
create mode 100644 tests/virresctrldata/resctrl.schemata
create mode 100644 tests/virresctrltest.c
I'm not sure any mor amount of looking at this will help me figure out
how this all works. I mean I understand how the test works, but how the
CAT works - I'm lost. I honestly didn't follow the previous conversation
on all this and trying to figure out how those schemata impact the XML
or vice versa - thoroughly and completely lost. I can understand the
need/desire for the technology but wonder if someone over engineered the
simple rope swing the customer was looking to create (not the libvirt
patches per se). There's so much detail in each cache line that one
really needs to "know" a lot more than I (currently) have the desire to
read up on. I can only imagine (as you noted a few patches earlier) how
much thinking about CAT(s ;-)) has dominated your life lately.
From the aspect of it looks like a test, it works, then yeap - good to
go. I'll be intrigued to see how qe tests/breaks this ;-)!
Reviewed-by: John Ferlan <jferlan(a)redhat.com>, but not really fully
understaood,
John