Signed-off-by: Daniel Black
---
v4 Message inline rather than macro. greppability nitpick accepted with
grace and the magic of %s/CAPABILITY_HINT("\([^"]*\)")/" try appending -machine
\1"/g.
v3 Hint, and not all messages, (can't use error_append_hint on
error_fat
ss Bandwidth Initiator
Domains:1 Target Domains:2 Base:1
[1.987229] Initiator-Target[0-0]:200 MB/s
[1.987356] Initiator-Target[0-1]:0 MB/s
[ 1.987549] HMAT: Cache: Domain:0 Size:131072 Attrs:0008 SMBIOS Handles:0
[1.988393] HMAT: Cache: Domain:1 Size:131072 Attrs:0008 SMBIOS Handles:0
Leaving default latency/bw as 0 if ok as spec says '0: the corresponding latency
or bandwidth information is not provided.'. Potentially the kernel could
display this better.
Also note https://marc.info/?l=linux-acpi&m=156506549410279&w=2 submitted as
hmat_build_table_structs only calls build_hmat_mpda with flags=0 or
HMAT_PROX_INIT_VALID (0x1) which is right looking at ACPI-6.3. An Ack/(Nack if
I'm wrong) there would be good to have both kernel and this patch series
working together.
for entire series:
Reviewed-by: Daniel Black
Liu Jingqi, Tao Xu,
Apologies to the late response on a patch on what is already a v8 patch.
The specification of latency and bandwidth is very much following the
ACPI specification.
For a qemu interface I think this should be in more human measurements
(time for latency and a bandwidth rate f
ferences in the kernel early boot.
Small enhancements where made to error messages to be more explicit
about errors in node specification.
CC: Tao Xu
CC: Liu Jingqi
Signed-off-by: Daniel Black
---
Based-on: 20190614155626.27932-1-tao3...@intel.com
([PATCH RESEND v8 00/11] Build ACPI Heter
been left as is for now.
Reviewed-by: Greg Kurz
Signed-off-by: Daniel Black
---
tcg (broken only for cap-{cfpc|sbbc|ibs}:
$ ppc64-softmmu/qemu-system-ppc64 -nodefaults -nographic -machine
pseries,accel=tcg
qemu-system-ppc64: warning: TCG doesn't support requested feature,
cap-cfpc=worka
On Thu, 1 Aug 2019 12:41:59 +0200
Greg Kurz wrote:
> On Thu, 1 Aug 2019 13:38:19 +1000
> Daniel Black wrote:
>
> > Its not immediately obvious how cap-X=Y setting need to be applied
> > to the command line so, for spapr capability error messages, this
>
: Daniel Black
---
v2 Improved error message thanks David Gibson
Left the "appending" as its not obvious more that one -machine
arguement is allowed.
---
hw/ppc/spapr_caps.c | 47 ++---
1 file changed, 32 insertions(+), 15 deletions(-)
diff --gi
Its not immediately obvious how cap-X=y setting need to be applied
to the command line so this has been clarified to "appending to the
machine name" in spapr capability error messages.
The wrong value messages have been left as is, as the user has found
the right location.
Signed-off-
On Wed, 31 Jul 2019 18:40:33 +1000
Daniel Black wrote:
> Replace all nodeid assumptions with lookups in the numa_info array
Tao Xu,
Sorry, only just noticed your patches of the move of numa_info and other
elements into MachineState. I approve of it. Lets see if I can follow
all of Igo
in node specification.
Signed-off-by: Daniel Black
---
Below is a TCG test script that covers ppc64le/x86_64/aarch64 architecutes:
Uncomment run* lines as needed. Manually complete steps at end to
verify:
Caution, brutal killalls contained within.
#!/bin/bash
set -x -v
QEMUHOME=${HOME}/re
10 matches
Mail list logo
