Fix SEV user-space mapping of unencrypted coherent memory
Subsystems (2)
v3 cover Fix SEV user-space mapping of unencrypted coherent memory
This patchset fixes dma_mmap_coherent() mapping of unencrypted memory in otherwise encrypted environments, where it would incorrectly map that memory as encrypted.
With SEV and sometimes with SME encryption, The dma api coherent memory is typically unencrypted, meaning the linear kernel map has the encryption bit cleared. However, default page protection returned from vm_get_page_prot() has the encryption bit set. So to compute the correct page protection we need to clear the encryption bit.
Also, in order for the encryption bit setting to survive across do_mmap() and mprotect_fixup(), We need to make pgprot_modify() aware of it and not touch it. Therefore make sme_me_mask part of _PAGE_CHG_MASK and make sure pgprot_modify() preserves also cleared bits that are part of _PAGE_CHG_MASK, not just set bits. The use of pgprot_modify() is currently quite limited and easy to audit.
(Note that the encryption status is not logically encoded in the pfn but in the page protection even if an address line in the physical address is used).
The patchset has seen some sanity testing by exporting dma_pgprot() and using it in the vmwgfx mmap handler with SEV enabled.
As far as I can tell there are no current users of dma_mmap_coherent() with SEV or SME encryption which means that there is no need to CC stable.
Changes since:
RFC:
- Make sme_me_mask port of _PAGE_CHG_MASK rather than using it by its own in pgprot_modify().
v2:
- Clarify which use-cases this patchset actually fixes.
v3:
- Use _PAGE_ENC instead of sme_me_mask in the definition of _PAGE_CHG_MASK
PATCH v3 1/2 x86: Don't let pgprot_modify() change the page encryption bit
From: Thomas Hellstrom <[email protected]>
When SEV or SME is enabled and active, vm_get_page_prot() typically returns with the encryption bit set. This means that users of pgprot_modify(, vm_get_page_prot()) (mprotect_fixup, do_mmap) end up with a value of vma->vm_pg_prot that is not consistent with the intended protection of the PTEs. This is also important for fault handlers that rely on the VMA vm_page_prot to set the page protection. Fix this by not allowing pgprot_modify() to change the encryption bit, similar to how it's done for PAT bits.
--- arch/x86/include/asm/pgtable.h | 7 +++++-- arch/x86/include/asm/pgtable_types.h | 2 +- 2 files changed, 6 insertions(+), 3 deletions(-)
arch/x86/include/asm/pgtable.h +5 -2
@@ -624,12 +624,15 @@ static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)return __pmd (val); } -/* mprotect needs to preserve PAT bits when updating vm_page_prot */ +/* +* mprotect needs to preserve PAT and encryption bits when updating +* vm_page_prot +*/ #define pgprot_modify pgprot_modifystatic inline pgprot_t pgprot_modify (pgprot_t oldprot,pgprot_t newprot) {pgprotval_t preservebits =pgprot_val (oldprot) & _PAGE_CHG_MASK; -pgprotval_t addbits =pgprot_val (newprot); +pgprotval_t addbits =pgprot_val (newprot) & ~_PAGE_CHG_MASK;return __pgprot (preservebits | addbits); }
arch/x86/include/asm/pgtable_types.h +1 -1
@@ -123,7 +123,7 @@*/ #define _PAGE_CHG_MASK (PTE_PFN_MASK | _PAGE_PCD | _PAGE_PWT | \ _PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY | \ - _PAGE_SOFT_DIRTY | _PAGE_DEVMAP) + _PAGE_SOFT_DIRTY | _PAGE_DEVMAP | _PAGE_ENC)#define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE)/*
Signature · 1 line
2.20.1
Re: [PATCH v3 1/2] x86: Don't let pgprot_modify() change the page encryption bit
arch/x86/include/asm/pgtable_types.h line 127
Quoted 11 lines · Thomas Hellström (VMware)
arch/x86/include/asm/pgtable_types.h +1
-1@@ -123,7 +123,7 @@*/ #define _PAGE_CHG_MASK (PTE_PFN_MASK | _PAGE_PCD | _PAGE_PWT | \ _PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY | \ - _PAGE_SOFT_DIRTY | _PAGE_DEVMAP) + _PAGE_SOFT_DIRTY | _PAGE_DEVMAP | _PAGE_ENC)#define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE)
My only nit with what remains is that it expands the infestation of things that look like a simple macro but are not.
I'm debating whether we want to go fix that now, though.
Re: [PATCH v3 1/2] x86: Don't let pgprot_modify() change the page encryption bit
Hi,
Quoted 15 lines · Dave Hansen
arch/x86/include/asm/pgtable_types.h line 127
Quoted 11 lines · Thomas Hellström (VMware)
arch/x86/include/asm/pgtable_types.h +1
-1@@ -123,7 +123,7 @@*/ #define _PAGE_CHG_MASK (PTE_PFN_MASK | _PAGE_PCD | _PAGE_PWT | \ _PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY | \ - _PAGE_SOFT_DIRTY | _PAGE_DEVMAP) + _PAGE_SOFT_DIRTY | _PAGE_DEVMAP | _PAGE_ENC)#define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE)My only nit with what remains is that it expands the infestation of things that look like a simple macro but are not.
Quoted 2 lines
I'm debating whether we want to go fix that now, though.
Any chance for an ack on this? It's really a small change that, as we've found out, fixes an existing problem.
Thanks,
Thomas
Dave Hansen
Reviewed-byRe: [PATCH v3 1/2] x86: Don't let pgprot_modify() change the page encryption bit
Quoted 18 lines · Thomas Hellstrom
Hi,
Quoted 15 lines · Dave Hansen
arch/x86/include/asm/pgtable_types.h line 127
Quoted 11 lines · Thomas Hellström (VMware)
arch/x86/include/asm/pgtable_types.h +1
-1@@ -123,7 +123,7 @@*/ #define _PAGE_CHG_MASK (PTE_PFN_MASK | _PAGE_PCD | _PAGE_PWT | \ _PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY | \ - _PAGE_SOFT_DIRTY | _PAGE_DEVMAP) + _PAGE_SOFT_DIRTY | _PAGE_DEVMAP | _PAGE_ENC)#define _HPAGE_CHG_MASK (_PAGE_CHG_MASK | _PAGE_PSE)My only nit with what remains is that it expands the infestation of things that look like a simple macro but are not.
Quoted 4 lines
Quoted 2 lines
I'm debating whether we want to go fix that now, though.
Any chance for an ack on this? It's really a small change that, as we've found out, fixes an existing problem.
Yes, feel free to add:
PATCH v3 2/2 dma-mapping: Fix dma_pgprot() for unencrypted coherent pages
In reply to Thomas Hellström (VMware)From: Thomas Hellstrom <[email protected]>
When dma_mmap_coherent() sets up a mapping to unencrypted coherent memory under SEV encryption and sometimes under SME encryption, it will actually set up an encrypted mapping rather than an unencrypted, causing devices that DMAs from that memory to read encrypted contents. Fix this.
When force_dma_unencrypted() returns true, the linear kernel map of the coherent pages have had the encryption bit explicitly cleared and the page content is unencrypted. Make sure that any additional PTEs we set up to these pages also have the encryption bit cleared by having dma_pgprot() return a protection with the encryption bit cleared in this case.
--- kernel/dma/mapping.c | 2 ++ 1 file changed, 2 insertions(+)
kernel/dma/mapping.c +2 -0
@@ -157,6 +157,8 @@ EXPORT_SYMBOL(dma_get_sgtable_attrs);*/ pgprot_t dma_pgprot (struct device *dev,pgprot_t prot,unsigned long attrs) { +if (force_dma_unencrypted (dev)) + prot =pgprot_decrypted (prot);if (dev_is_dma_coherent (dev) || (IS_ENABLED (CONFIG_DMA_NONCOHERENT_CACHE_SYNC ) && (attrs &DMA_ATTR_NON_CONSISTENT )))
Signature · 1 line
2.20.1
This looks sensibe to me. x86 folks let me know if you guys want to pick this up or if I should take it.
Quoted 2 lines · Thomas Hellström (VMware)
As far as I can tell there are no current users of
dma_mmap_coherent()with SEV or SME encryption which means that there is no need to CC stable.
Note that the usbfs access really should use it, as the current version is completely broken on non-coherent architectures. That already triggered on x86 dma_mmap_coherent fix. So I'd actually like to see a stable backport to avoid hitting surprises.