On Mon, Jan 25, 2016 at 11:12:36AM +1100, David Gibson wrote:

On Thu, Jan 21, 2016 at 06:39:36PM +1100, Alexey Kardashevskiy wrote:

quoted

Upcoming multi-tce support (H_PUT_TCE_INDIRECT/H_STUFF_TCE hypercalls)
will validate TCE (not to have unexpected bits) and IO address
(to be within the DMA window boundaries).

This introduces helpers to validate TCE and IO address. The helpers are
exported as they compile into vmlinux (to work in realmode) and will be
used later by KVM kernel module in virtual mode.

Signed-off-by: Alexey Kardashevskiy <redacted>
---
Changes:
v2:
* added note to the commit log about why new helpers are exported
* did not add a note that xxx_validate() validate TCEs for KVM (not for
host kernel DMA) as the helper names and file location tell what are
they for
---
 arch/powerpc/include/asm/kvm_ppc.h  |  4 ++
 arch/powerpc/kvm/book3s_64_vio_hv.c | 92 ++++++++++++++++++++++++++++++++-----
 2 files changed, 84 insertions(+), 12 deletions(-)

diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h
index 2241d53..9513911 100644
--- a/arch/powerpc/include/asm/kvm_ppc.h
+++ b/arch/powerpc/include/asm/kvm_ppc.h

@@ -166,6 +166,10 @@ extern int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu);
 
 extern long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm,
 				struct kvm_create_spapr_tce *args);
+extern long kvmppc_ioba_validate(struct kvmppc_spapr_tce_table *stt,
+		unsigned long ioba, unsigned long npages);
+extern long kvmppc_tce_validate(struct kvmppc_spapr_tce_table *tt,
+		unsigned long tce);
 extern long kvmppc_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
 			     unsigned long ioba, unsigned long tce);
 extern long kvmppc_h_get_tce(struct kvm_vcpu *vcpu, unsigned long liobn,

diff --git a/arch/powerpc/kvm/book3s_64_vio_hv.c b/arch/powerpc/kvm/book3s_64_vio_hv.c
index e142171..8cd3a95 100644
--- a/arch/powerpc/kvm/book3s_64_vio_hv.c
+++ b/arch/powerpc/kvm/book3s_64_vio_hv.c

@@ -36,6 +36,7 @@
 #include <asm/kvm_host.h>
 #include <asm/udbg.h>
 #include <asm/iommu.h>
+#include <asm/tce.h>
 
 #define TCES_PER_PAGE	(PAGE_SIZE / sizeof(u64))
 

@@ -64,18 +65,90 @@ static struct kvmppc_spapr_tce_table *kvmppc_find_table(struct kvm_vcpu *vcpu,
  * WARNING: This will be called in real-mode on HV KVM and virtual
  *          mode on PR KVM
  */
-static long kvmppc_ioba_validate(struct kvmppc_spapr_tce_table *stt,
+long kvmppc_ioba_validate(struct kvmppc_spapr_tce_table *stt,
 		unsigned long ioba, unsigned long npages)
 {
-	unsigned long mask = (1ULL << IOMMU_PAGE_SHIFT_4K) - 1;
+	unsigned long mask = IOMMU_PAGE_MASK_4K;
 	unsigned long idx = ioba >> IOMMU_PAGE_SHIFT_4K;
 	unsigned long size = stt->window_size >> IOMMU_PAGE_SHIFT_4K;
 
-	if ((ioba & mask) || (idx + npages > size))
+	if ((ioba & ~mask) || (idx + npages > size))
 		return H_PARAMETER;
 
 	return H_SUCCESS;
 }
+EXPORT_SYMBOL_GPL(kvmppc_ioba_validate);
+
+/*
+ * Validates TCE address.
+ * At the moment flags and page mask are validated.
+ * As the host kernel does not access those addresses (just puts them
+ * to the table and user space is supposed to process them), we can skip
+ * checking other things (such as TCE is a guest RAM address or the page
+ * was actually allocated).
+ *
+ * WARNING: This will be called in real-mode on HV KVM and virtual
+ *          mode on PR KVM
+ */
+long kvmppc_tce_validate(struct kvmppc_spapr_tce_table *stt,
 unsigned long tce)

It would be nice to write this in terms of kvmppc_ioba_validate() above.

Duh, sorry.  Realised shortly afterwards that's nonsense.  One is
looking at the IOBA, the other at the real address.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson