Re: [PATCH 15/17] x86/shadow: drop SH_type_l2h_pae_shadow

2021-01-25 Thread Jan Beulich
On 22.01.2021 21:02, Tim Deegan wrote: > At 17:31 +0100 on 22 Jan (1611336662), Jan Beulich wrote: >> Because of this having been benign (due to none of the callback >> tables specifying non-NULL entries there), wouldn't it make >> sense to dimension the tables by SH_type_max_shadow + 1 only? >>

Re: [PATCH 15/17] x86/shadow: drop SH_type_l2h_pae_shadow

2021-01-25 Thread Jan Beulich
On 22.01.2021 21:02, Tim Deegan wrote: > At 17:31 +0100 on 22 Jan (1611336662), Jan Beulich wrote: >> On 22.01.2021 14:11, Tim Deegan wrote: >>> At 16:10 +0100 on 14 Jan (1610640627), Jan Beulich wrote: hash_{domain,vcpu}_foreach() have a use each of literal 15. It's not clear to me what

Re: [PATCH 15/17] x86/shadow: drop SH_type_l2h_pae_shadow

2021-01-22 Thread Tim Deegan
Hi, At 17:31 +0100 on 22 Jan (1611336662), Jan Beulich wrote: > On 22.01.2021 14:11, Tim Deegan wrote: > > At 16:10 +0100 on 14 Jan (1610640627), Jan Beulich wrote: > >> hash_{domain,vcpu}_foreach() have a use each of literal 15. It's not > >> clear to me what the proper replacement constant

Re: [PATCH 15/17] x86/shadow: drop SH_type_l2h_pae_shadow

2021-01-22 Thread Jan Beulich
On 22.01.2021 14:11, Tim Deegan wrote: > At 16:10 +0100 on 14 Jan (1610640627), Jan Beulich wrote: >> This is a remnant from 32-bit days, having no place anymore where a >> shadow of this type would be created. >> >> Signed-off-by: Jan Beulich >> --- >> hash_{domain,vcpu}_foreach() have a use

Re: [PATCH 15/17] x86/shadow: drop SH_type_l2h_pae_shadow

2021-01-22 Thread Tim Deegan
Hi, At 16:10 +0100 on 14 Jan (1610640627), Jan Beulich wrote: > This is a remnant from 32-bit days, having no place anymore where a > shadow of this type would be created. > > Signed-off-by: Jan Beulich > --- > hash_{domain,vcpu}_foreach() have a use each of literal 15. It's not > clear to me

[PATCH 15/17] x86/shadow: drop SH_type_l2h_pae_shadow

2021-01-14 Thread Jan Beulich
This is a remnant from 32-bit days, having no place anymore where a shadow of this type would be created. Signed-off-by: Jan Beulich --- hash_{domain,vcpu}_foreach() have a use each of literal 15. It's not clear to me what the proper replacement constant would be, as it doesn't look as if