| Message ID | 20230213163351.30704-4-minipli@grsecurity.net (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/5] KVM: x86: Shrink struct kvm_pmu | expand |
On Mon, Feb 13, 2023, Mathias Krause wrote: > Move the 'capacity' member around to make use of the padding hole on 64 > bit systems instead of introducing yet another one. > > This allows us to save 8 bytes per instance for 64 bit builds of which, > e.g., x86's struct kvm_vcpu_arch has a few. > > Signed-off-by: Mathias Krause <minipli@grsecurity.net> > --- > include/linux/kvm_types.h | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/include/linux/kvm_types.h b/include/linux/kvm_types.h > index 76de36e56cdf..8e4f8fa31457 100644 > --- a/include/linux/kvm_types.h > +++ b/include/linux/kvm_types.h > @@ -92,10 +92,10 @@ struct gfn_to_pfn_cache { > */ > struct kvm_mmu_memory_cache { > int nobjs; > + int capacity; > gfp_t gfp_zero; > gfp_t gfp_custom; > struct kmem_cache *kmem_cache; > - int capacity; > void **objects; If we touch this, I vote to do a more aggressive cleanup and place nobjs next to objects, e.g. struct kvm_mmu_memory_cache { gfp_t gfp_zero; gfp_t gfp_custom; struct kmem_cache *kmem_cache; int capacity; int nobjs; void **objects; };
On 13.02.23 18:08, Sean Christopherson wrote: > On Mon, Feb 13, 2023, Mathias Krause wrote: >> Move the 'capacity' member around to make use of the padding hole on 64 >> bit systems instead of introducing yet another one. >> >> This allows us to save 8 bytes per instance for 64 bit builds of which, >> e.g., x86's struct kvm_vcpu_arch has a few. >> >> Signed-off-by: Mathias Krause <minipli@grsecurity.net> >> --- >> include/linux/kvm_types.h | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/include/linux/kvm_types.h b/include/linux/kvm_types.h >> index 76de36e56cdf..8e4f8fa31457 100644 >> --- a/include/linux/kvm_types.h >> +++ b/include/linux/kvm_types.h >> @@ -92,10 +92,10 @@ struct gfn_to_pfn_cache { >> */ >> struct kvm_mmu_memory_cache { >> int nobjs; >> + int capacity; >> gfp_t gfp_zero; >> gfp_t gfp_custom; >> struct kmem_cache *kmem_cache; >> - int capacity; >> void **objects; > > If we touch this, I vote to do a more aggressive cleanup and place nobjs next > to objects, e.g. > > struct kvm_mmu_memory_cache { > gfp_t gfp_zero; > gfp_t gfp_custom; > struct kmem_cache *kmem_cache; > int capacity; > int nobjs; > void **objects; > }; Ok
diff --git a/include/linux/kvm_types.h b/include/linux/kvm_types.h index 76de36e56cdf..8e4f8fa31457 100644 --- a/include/linux/kvm_types.h +++ b/include/linux/kvm_types.h @@ -92,10 +92,10 @@ struct gfn_to_pfn_cache { */ struct kvm_mmu_memory_cache { int nobjs; + int capacity; gfp_t gfp_zero; gfp_t gfp_custom; struct kmem_cache *kmem_cache; - int capacity; void **objects; }; #endif
Move the 'capacity' member around to make use of the padding hole on 64 bit systems instead of introducing yet another one. This allows us to save 8 bytes per instance for 64 bit builds of which, e.g., x86's struct kvm_vcpu_arch has a few. Signed-off-by: Mathias Krause <minipli@grsecurity.net> --- include/linux/kvm_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)