* [PATCH] x86/gen-cpuid: Distinguish default vs max in feature annotations
@ 2020-05-08 15:27 Andrew Cooper
2020-05-11 7:43 ` Jan Beulich
0 siblings, 1 reply; 3+ messages in thread
From: Andrew Cooper @ 2020-05-08 15:27 UTC (permalink / raw)
To: Xen-devel; +Cc: Andrew Cooper, Wei Liu, Jan Beulich, Roger Pau Monné
Allow lowercase a/s/h to be used to annotate a non-default feature.
However, until the toolstack migration logic is fixed, it is not safe to
activate yet. Tolerate the annotations, but ignore them for now.
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
---
CC: Jan Beulich <JBeulich@suse.com>
CC: Wei Liu <wl@xen.org>
CC: Roger Pau Monné <roger.pau@citrix.com>
This patch has been pending the toolstack work for several months now, and we
want to start using the max tags for x86 emul work.
---
xen/include/public/arch-x86/cpufeatureset.h | 2 ++
xen/tools/gen-cpuid.py | 5 +++++
2 files changed, 7 insertions(+)
diff --git a/xen/include/public/arch-x86/cpufeatureset.h b/xen/include/public/arch-x86/cpufeatureset.h
index e2749245f3..0ffab6c57b 100644
--- a/xen/include/public/arch-x86/cpufeatureset.h
+++ b/xen/include/public/arch-x86/cpufeatureset.h
@@ -87,6 +87,8 @@ enum {
* 'A' = All guests.
* 'S' = All HVM guests (not PV guests).
* 'H' = HVM HAP guests (not PV or HVM Shadow guests).
+ * Upper case => Available by default
+ * Lower case => Can be opted-in to, but not available by default.
*/
/* Intel-defined CPU features, CPUID level 0x00000001.edx, word 0 */
diff --git a/xen/tools/gen-cpuid.py b/xen/tools/gen-cpuid.py
index af5610a5e6..d90a2d85c7 100755
--- a/xen/tools/gen-cpuid.py
+++ b/xen/tools/gen-cpuid.py
@@ -23,6 +23,7 @@ def __init__(self, input, output):
self.raw = {
'!': set(),
'A': set(), 'S': set(), 'H': set(),
+ 'a': set(), 's': set(), 'h': set(),
}
# State calculated
@@ -133,9 +134,13 @@ def crunch_numbers(state):
state.hvm_shadow_def = state.pv_def | state.raw['S']
state.hvm_hap_def = state.hvm_shadow_def | state.raw['H']
+ # TODO: Ignore def/max split until the toolstack migration logic is fixed
state.pv_max = state.pv_def
state.hvm_shadow_max = state.hvm_shadow_def
state.hvm_hap_max = state.hvm_hap_def
+ # state.pv_max = state.raw['A'] | state.raw['a']
+ # state.hvm_shadow_max = state.pv_max | state.raw['S'] | state.raw['s']
+ # state.hvm_hap_max = state.hvm_shadow_max | state.raw['H'] | state.raw['h']
#
# Feature dependency information.
--
2.11.0
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] x86/gen-cpuid: Distinguish default vs max in feature annotations
2020-05-08 15:27 [PATCH] x86/gen-cpuid: Distinguish default vs max in feature annotations Andrew Cooper
@ 2020-05-11 7:43 ` Jan Beulich
2020-05-15 13:19 ` Andrew Cooper
0 siblings, 1 reply; 3+ messages in thread
From: Jan Beulich @ 2020-05-11 7:43 UTC (permalink / raw)
To: Andrew Cooper; +Cc: Xen-devel, Wei Liu, Roger Pau Monné
On 08.05.2020 17:27, Andrew Cooper wrote:
> @@ -133,9 +134,13 @@ def crunch_numbers(state):
> state.hvm_shadow_def = state.pv_def | state.raw['S']
> state.hvm_hap_def = state.hvm_shadow_def | state.raw['H']
>
> + # TODO: Ignore def/max split until the toolstack migration logic is fixed
> state.pv_max = state.pv_def
> state.hvm_shadow_max = state.hvm_shadow_def
> state.hvm_hap_max = state.hvm_hap_def
> + # state.pv_max = state.raw['A'] | state.raw['a']
> + # state.hvm_shadow_max = state.pv_max | state.raw['S'] | state.raw['s']
> + # state.hvm_hap_max = state.hvm_shadow_max | state.raw['H'] | state.raw['h']
While in comment form it doesn't matter yet, for actually enabling
this it would seem to me to be more expressive as
state.pv_max = state.pv_def | state.raw['a']
state.hvm_shadow_max = state.hvm_shadow_def | state.pv_max | state.raw['s']
state.hvm_hap_max = state.hvm_hap_def | state.hvm_shadow_max | state.raw['h']
Thoughts?
Jan
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] x86/gen-cpuid: Distinguish default vs max in feature annotations
2020-05-11 7:43 ` Jan Beulich
@ 2020-05-15 13:19 ` Andrew Cooper
0 siblings, 0 replies; 3+ messages in thread
From: Andrew Cooper @ 2020-05-15 13:19 UTC (permalink / raw)
To: Jan Beulich; +Cc: Xen-devel, Wei Liu, Roger Pau Monné
On 11/05/2020 08:43, Jan Beulich wrote:
> [CAUTION - EXTERNAL EMAIL] DO NOT reply, click links, or open attachments unless you have verified the sender and know the content is safe.
>
> On 08.05.2020 17:27, Andrew Cooper wrote:
>> @@ -133,9 +134,13 @@ def crunch_numbers(state):
>> state.hvm_shadow_def = state.pv_def | state.raw['S']
>> state.hvm_hap_def = state.hvm_shadow_def | state.raw['H']
>>
>> + # TODO: Ignore def/max split until the toolstack migration logic is fixed
>> state.pv_max = state.pv_def
>> state.hvm_shadow_max = state.hvm_shadow_def
>> state.hvm_hap_max = state.hvm_hap_def
>> + # state.pv_max = state.raw['A'] | state.raw['a']
>> + # state.hvm_shadow_max = state.pv_max | state.raw['S'] | state.raw['s']
>> + # state.hvm_hap_max = state.hvm_shadow_max | state.raw['H'] | state.raw['h']
> While in comment form it doesn't matter yet, for actually enabling
> this it would seem to me to be more expressive as
>
> state.pv_max = state.pv_def | state.raw['a']
> state.hvm_shadow_max = state.hvm_shadow_def | state.pv_max | state.raw['s']
> state.hvm_hap_max = state.hvm_hap_def | state.hvm_shadow_max | state.raw['h']
>
> Thoughts?
The format/layout of metadata prohibits that from having any kind of
difference.
Furthermore, expressing it this way doesn't require anyone to look at
the *_def derivation to figure out what is going on.
I'm going to commit it in this form, but will tidy it up to be properly
tabulated in a future patch, to make it easier to read.
~Andrew
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-05-15 13:19 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-08 15:27 [PATCH] x86/gen-cpuid: Distinguish default vs max in feature annotations Andrew Cooper
2020-05-11 7:43 ` Jan Beulich
2020-05-15 13:19 ` Andrew Cooper
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.