| Message ID | 1416931805-23223-4-git-send-email-alex.bennee@linaro.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Tue, Nov 25, 2014 at 04:10:01PM +0000, Alex Bennée wrote: > This commit adds a stub function to support the KVM_SET_GUEST_DEBUG > ioctl. Currently any operation flag will return EINVAL. Actual > functionality will be added with further patches. Technically the stub is already there, and you're extending it to start looking at control flags, but still not doing anything yet. > > Signed-off-by: Alex Bennée <alex.bennee@linaro.org>. > > diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt > index 7610eaa..2c6386e 100644 > --- a/Documentation/virtual/kvm/api.txt > +++ b/Documentation/virtual/kvm/api.txt > @@ -2570,7 +2570,7 @@ handled. > 4.87 KVM_SET_GUEST_DEBUG > > Capability: KVM_CAP_SET_GUEST_DEBUG > -Architectures: x86, s390, ppc > +Architectures: x86, s390, ppc, arm64 > Type: vcpu ioctl > Parameters: struct kvm_guest_debug (in) > Returns: 0 on success; -1 on error > diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c > index 9e193c8..a0ff410 100644 > --- a/arch/arm/kvm/arm.c > +++ b/arch/arm/kvm/arm.c > @@ -180,6 +180,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) > case KVM_CAP_ARM_PSCI: > case KVM_CAP_ARM_PSCI_0_2: > case KVM_CAP_READONLY_MEM: > + case KVM_CAP_SET_GUEST_DEBUG: > r = 1; > break; > case KVM_CAP_COALESCED_MMIO: > @@ -302,7 +303,34 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) > int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, > struct kvm_guest_debug *dbg) > { > - return -EINVAL; > + /* If it's not enabled clear all flags */ > + if (!(dbg->control & KVM_GUESTDBG_ENABLE)) { > + vcpu->guest_debug = 0; > + return 0; > + } > + > + vcpu->guest_debug = dbg->control; > + kvm_info("%s: guest_debug is 0x%lx\n", __func__, vcpu->guest_debug); > + > + /* Single Step */ > + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) { > + kvm_info("SS requested, not yet implemented\n"); > + return -EINVAL; > + } > + > + /* Software Break Points */ > + if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) { > + kvm_info("SW BP support requested, not yet implemented\n"); > + return -EINVAL; > + } > + > + /* Hardware assisted Break and Watch points */ > + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) { > + kvm_info("HW BP support requested, not yet implemented\n"); > + return -EINVAL; > + } > + > + return 0; > } I guess all the kvm_info's were useful for developing this patch series, but do we still need them? > > > -- > 2.1.3 > > _______________________________________________ > kvmarm mailing list > kvmarm@lists.cs.columbia.edu > https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
Andrew Jones <drjones@redhat.com> writes: > On Tue, Nov 25, 2014 at 04:10:01PM +0000, Alex Bennée wrote: >> This commit adds a stub function to support the KVM_SET_GUEST_DEBUG >> ioctl. Currently any operation flag will return EINVAL. Actual >> functionality will be added with further patches. > > Technically the stub is already there, and you're extending it to > start looking at control flags, but still not doing anything yet. Sure we do: >> int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, >> struct kvm_guest_debug *dbg) >> { >> - return -EINVAL; >> + /* If it's not enabled clear all flags */ >> + if (!(dbg->control & KVM_GUESTDBG_ENABLE)) { >> + vcpu->guest_debug = 0; >> + return 0; >> + } That's some class non-functionality right there ;-) >> + vcpu->guest_debug = dbg->control; >> + kvm_info("%s: guest_debug is 0x%lx\n", __func__, vcpu->guest_debug); >> + >> + /* Single Step */ >> + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) { >> + kvm_info("SS requested, not yet implemented\n"); >> + return -EINVAL; >> + } >> + >> + /* Software Break Points */ >> + if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) { >> + kvm_info("SW BP support requested, not yet implemented\n"); >> + return -EINVAL; >> + } >> + >> + /* Hardware assisted Break and Watch points */ >> + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) { >> + kvm_info("HW BP support requested, not yet implemented\n"); >> + return -EINVAL; >> + } >> + >> + return 0; >> } > > I guess all the kvm_info's were useful for developing this patch series, > but do we still need them? They also served the very useful roll of stopping checkpatch.pl bitching about my reluctance to remove braces from the if () { } clauses. However I take your point. I can certainly remove the kvm_info() statements as each bit of functionality is added while leaving this one to help when someone is bisecting and confused.
On 26/11/2014 16:03, Alex Bennée wrote: > They also served the very useful roll of stopping checkpatch.pl bitching > about my reluctance to remove braces from the if () { } clauses. However > I take your point. I can certainly remove the kvm_info() statements as > each bit of functionality is added while leaving this one to help when > someone is bisecting and confused. Good idea. Paolo
On Tue, Nov 25, 2014 at 04:10:01PM +0000, Alex Bennée wrote: > This commit adds a stub function to support the KVM_SET_GUEST_DEBUG > ioctl. Currently any operation flag will return EINVAL. Actual > functionality will be added with further patches. > > Signed-off-by: Alex Bennée <alex.bennee@linaro.org>. > > diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt > index 7610eaa..2c6386e 100644 > --- a/Documentation/virtual/kvm/api.txt > +++ b/Documentation/virtual/kvm/api.txt > @@ -2570,7 +2570,7 @@ handled. > 4.87 KVM_SET_GUEST_DEBUG > > Capability: KVM_CAP_SET_GUEST_DEBUG > -Architectures: x86, s390, ppc > +Architectures: x86, s390, ppc, arm64 > Type: vcpu ioctl > Parameters: struct kvm_guest_debug (in) > Returns: 0 on success; -1 on error > diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c > index 9e193c8..a0ff410 100644 > --- a/arch/arm/kvm/arm.c > +++ b/arch/arm/kvm/arm.c > @@ -180,6 +180,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) > case KVM_CAP_ARM_PSCI: > case KVM_CAP_ARM_PSCI_0_2: > case KVM_CAP_READONLY_MEM: > + case KVM_CAP_SET_GUEST_DEBUG: Don't we usually wait with introducing visibility to user space until you actually do support the ioctl? > r = 1; > break; > case KVM_CAP_COALESCED_MMIO: > @@ -302,7 +303,34 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) > int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, > struct kvm_guest_debug *dbg) > { > - return -EINVAL; > + /* If it's not enabled clear all flags */ > + if (!(dbg->control & KVM_GUESTDBG_ENABLE)) { > + vcpu->guest_debug = 0; > + return 0; > + } > + > + vcpu->guest_debug = dbg->control; > + kvm_info("%s: guest_debug is 0x%lx\n", __func__, vcpu->guest_debug); > + > + /* Single Step */ > + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) { > + kvm_info("SS requested, not yet implemented\n"); > + return -EINVAL; > + } > + > + /* Software Break Points */ > + if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) { > + kvm_info("SW BP support requested, not yet implemented\n"); > + return -EINVAL; > + } > + > + /* Hardware assisted Break and Watch points */ > + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) { > + kvm_info("HW BP support requested, not yet implemented\n"); > + return -EINVAL; > + } > + > + return 0; > } > > > -- > 2.1.3 >
diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index 7610eaa..2c6386e 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -2570,7 +2570,7 @@ handled. 4.87 KVM_SET_GUEST_DEBUG Capability: KVM_CAP_SET_GUEST_DEBUG -Architectures: x86, s390, ppc +Architectures: x86, s390, ppc, arm64 Type: vcpu ioctl Parameters: struct kvm_guest_debug (in) Returns: 0 on success; -1 on error diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 9e193c8..a0ff410 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -180,6 +180,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_ARM_PSCI: case KVM_CAP_ARM_PSCI_0_2: case KVM_CAP_READONLY_MEM: + case KVM_CAP_SET_GUEST_DEBUG: r = 1; break; case KVM_CAP_COALESCED_MMIO: @@ -302,7 +303,34 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) { - return -EINVAL; + /* If it's not enabled clear all flags */ + if (!(dbg->control & KVM_GUESTDBG_ENABLE)) { + vcpu->guest_debug = 0; + return 0; + } + + vcpu->guest_debug = dbg->control; + kvm_info("%s: guest_debug is 0x%lx\n", __func__, vcpu->guest_debug); + + /* Single Step */ + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) { + kvm_info("SS requested, not yet implemented\n"); + return -EINVAL; + } + + /* Software Break Points */ + if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) { + kvm_info("SW BP support requested, not yet implemented\n"); + return -EINVAL; + } + + /* Hardware assisted Break and Watch points */ + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) { + kvm_info("HW BP support requested, not yet implemented\n"); + return -EINVAL; + } + + return 0; }
This commit adds a stub function to support the KVM_SET_GUEST_DEBUG ioctl. Currently any operation flag will return EINVAL. Actual functionality will be added with further patches. Signed-off-by: Alex Bennée <alex.bennee@linaro.org>.