Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Mark Brown
On Tue, Dec 17, 2024 at 10:00:44AM -0800, Oliver Upton wrote: > On Tue, Dec 17, 2024 at 03:10:28PM +, Mark Brown wrote: > > No, this isn't a new use - a Fixes: tag indicates that the referenced > > commit introduced the problem being fixed and that is exactly what's > > going on here. Like I

Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Oliver Upton
On Tue, Dec 17, 2024 at 03:10:28PM +, Mark Brown wrote: > On Tue, Dec 17, 2024 at 01:54:39PM +, Marc Zyngier wrote: > > Mark Brown wrote: > > > > The selftests are shipped as part of the kernel source and frequently > > > used for testing the kernel, it's all one source base and we want t

Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Marc Zyngier
On Tue, 17 Dec 2024 15:10:28 +, Mark Brown wrote: > > [1 ] > On Tue, Dec 17, 2024 at 01:54:39PM +, Marc Zyngier wrote: > > Mark Brown wrote: > > > > The selftests are shipped as part of the kernel source and frequently > > > used for testing the kernel, it's all one source base and we

Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Mark Brown
On Tue, Dec 17, 2024 at 01:54:39PM +, Marc Zyngier wrote: > Mark Brown wrote: > > The selftests are shipped as part of the kernel source and frequently > > used for testing the kernel, it's all one source base and we want to > > ensure that for example the test fix gets backported if the rele

Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Marc Zyngier
On Tue, 17 Dec 2024 13:12:12 +, Mark Brown wrote: > > [1 ] > On Tue, Dec 17, 2024 at 08:30:37AM +, Marc Zyngier wrote: > > Mark Brown wrote: > > > > Fixes: 03c7527e97f7 ("KVM: arm64: Do not allow ID_AA64MMFR0_EL1.ASIDbits > > > to be overridden") > > > A patch for a test doesn't fix

Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Mark Brown
On Tue, Dec 17, 2024 at 08:30:37AM +, Marc Zyngier wrote: > Mark Brown wrote: > > Fixes: 03c7527e97f7 ("KVM: arm64: Do not allow ID_AA64MMFR0_EL1.ASIDbits to > > be overridden") > A patch for a test doesn't fix anything in the kernel. The selftests are shipped as part of the kernel source

Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Oliver Upton
On Mon, 16 Dec 2024 19:28:24 +, Mark Brown wrote: > In commit 03c7527e97f7 ("KVM: arm64: Do not allow ID_AA64MMFR0_EL1.ASIDbits > to be overridden") we made that bitfield in the ID registers unwritable > however the change neglected to make the corresponding update to set_id_regs > resulting in

Re: [PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-17 Thread Marc Zyngier
On Mon, 16 Dec 2024 19:28:24 +, Mark Brown wrote: > > In commit 03c7527e97f7 ("KVM: arm64: Do not allow ID_AA64MMFR0_EL1.ASIDbits > to be overridden") we made that bitfield in the ID registers unwritable > however the change neglected to make the corresponding update to set_id_regs > resultin

[PATCH] KVM: arm64: Fix set_id_regs selftest for ASIDBITS becoming unwritable

2024-12-16 Thread Mark Brown
In commit 03c7527e97f7 ("KVM: arm64: Do not allow ID_AA64MMFR0_EL1.ASIDbits to be overridden") we made that bitfield in the ID registers unwritable however the change neglected to make the corresponding update to set_id_regs resulting in it failing: # ok 56 ID_AA64MMFR0_EL1_BIGEND # Test Asse