Skip to contentSkip to conversation

dma-buf: move system_cc_shared heap under separate Kconfig

PATCH In review Sumit Semwal 5 participants 6 messages Jun 10 – Jun 11

Subsystems (4)
lore
Share & tools
lore

Sumit Semwal

author maintainer
#0 lore raw

PATCH 1/1 dma-buf: move system_cc_shared heap under separate Kconfig

From: Arnd Bergmann <[email protected]>

While system heap and system_cc_shared heap share a lot of code and hence the same source file, their users have different needs.

system heap users need it to be a loadable module, while system_cc_shared heap users don't.

Building as a loadable module breaks system_cc_shared heap on powerpc and s390 due to un-exported set_memory_encrypted / set_memory_decrypted functions.

Fix these by reorganising code to put the system_cc_shared heap under a new Kconfig symbol, which allows either building both into the kernel, or leave encryption up to the consumers of the system heap.

Fixes fd55edff8a0a ("dma-buf: heaps: system: Turn the heap into a module")
---
 drivers/dma-buf/heaps/Kconfig       |  8 ++++++++
 drivers/dma-buf/heaps/system_heap.c | 16 ++++++++++------
 2 files changed, 18 insertions(+), 6 deletions(-)
drivers/dma-buf/heaps/Kconfig +8 -0
index e273fb18feca..a39decdcf067 100644
@@ -5,6 +5,14 @@ config DMABUF_HEAPS_SYSTEM
 	  Choose this option to enable the system dmabuf heap. The system heap
 	  is backed by pages from the buddy allocator. If in doubt, say Y.
 
+config DMABUF_HEAPS_CC_SYSTEM
+	bool "DMA-BUF System Heap for decrypted CoCo VMs"
+	depends on DMABUF_HEAPS && ARCH_HAS_MEM_ENCRYPT && DMABUF_HEAPS_SYSTEM=y
+	help
+	  Choose this option to enable the system_cc_shared dmabuf heap. This
+	  allows allocating shared (decrypted) memory for confidential computing
+	  (CoCo) VMs.
+
 config DMABUF_HEAPS_CMA
 	tristate "DMA-BUF CMA Heap"
 	depends on DMABUF_HEAPS && DMA_CMA
drivers/dma-buf/heaps/system_heap.c +10 -6
index c92bdec356fc..71d9028cc3df 100644
@@ -48,6 +48,9 @@ struct dma_heap_attachment {
 	bool cc_shared;
 };
 
+#define cc_shared_buffer(b) (IS_ENABLED(CONFIG_DMABUF_HEAPS_CC_SYSTEM) && \
+				(b)->cc_shared)
+
 #define LOW_ORDER_GFP (GFP_HIGHUSER | __GFP_ZERO)
 #define HIGH_ORDER_GFP  (((GFP_HIGHUSER | __GFP_ZERO | __GFP_NOWARN \
 				| __GFP_NORETRY) & ~__GFP_RECLAIM) \
@@ -161,7 +164,7 @@ static struct sg_table *system_heap_map_dma_buf(struct dma_buf_attachment *attac
 	unsigned long attrs;
 	int ret;
 
-	attrs = a->cc_shared ? DMA_ATTR_CC_SHARED : 0;
+	attrs = cc_shared_buffer(a) ? DMA_ATTR_CC_SHARED : 0;
 	ret = dma_map_sgtable(attachment->dev, table, direction, attrs);
 	if (ret)
 		return ERR_PTR(ret);
@@ -233,7 +236,7 @@ static int system_heap_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma)
 	int i, ret;
 
 	prot = vma->vm_page_prot;
-	if (buffer->cc_shared)
+	if (cc_shared_buffer(buffer))
 		prot = pgprot_decrypted(prot);
 
 	for_each_sgtable_sg(table, sg, i) {
@@ -282,7 +285,7 @@ static void *system_heap_do_vmap(struct system_heap_buffer *buffer)
 	}
 
 	prot = PAGE_KERNEL;
-	if (buffer->cc_shared)
+	if (cc_shared_buffer(buffer))
 		prot = pgprot_decrypted(prot);
 	vaddr = vmap(pages, npages, VM_MAP, prot);
 	vfree(pages);
@@ -349,7 +352,7 @@ static void system_heap_dma_buf_release(struct dma_buf *dmabuf)
 		 * Intentionally leak pages that cannot be re-encrypted
 		 * to prevent shared memory from being reused.
 		 */
-		if (buffer->cc_shared &&
+		if (cc_shared_buffer(buffer) &&
 		    system_heap_set_page_encrypted(page))
 			continue;
 
@@ -456,7 +459,7 @@ static struct dma_buf *system_heap_allocate(struct dma_heap *heap,
 		list_del(&page->lru);
 	}
 
-	if (cc_shared) {
+	if (cc_shared_buffer(buffer)) {
 		for_each_sgtable_sg(table, sg, i) {
 			ret = system_heap_set_page_decrypted(sg_page(sg));
 			if (ret)
@@ -485,7 +488,7 @@ static struct dma_buf *system_heap_allocate(struct dma_heap *heap,
 		 * Intentionally leak pages that cannot be re-encrypted
 		 * to prevent shared memory from being reused.
 		 */
-		if (buffer->cc_shared &&
+		if (cc_shared_buffer(buffer) &&
 		    system_heap_set_page_encrypted(p))
 			continue;
 		__free_pages(p, compound_order(p));
@@ -525,6 +528,7 @@ static int __init system_heap_create(void)
 		return PTR_ERR(sys_heap);
 
 	if (IS_ENABLED(CONFIG_HIGHMEM) ||
+	    !IS_ENABLED(CONFIG_DMABUF_HEAPS_CC_SYSTEM) ||
 	    !cc_platform_has(CC_ATTR_MEM_ENCRYPT))
 		return 0;
 
Signature · 1 line
2.43.0
#1 lore raw
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Unconditional page re-encryption in error path causes potential memory leak and invalid hypercalls --

commit 611c6660a2bb2f9f324ce349cfce6b2530846d28
Author: Arnd Bergmann <[email protected]>

dma-buf: move system_cc_shared heap under separate Kconfig

This commit moves the system_cc_shared heap into its own Kconfig symbol, DMABUF_HEAPS_CC_SYSTEM. This resolves a build failure on architectures like powerpc and s390 when the system heap is built as a loadable module, since the set_memory_encrypted and set_memory_decrypted functions are not exported to modules.

Quoted 4 lines
drivers/dma-buf/heaps/system_heap.c +0 -0
index c92bdec356fcc..71d9028cc3dfc 100644

[ ... ]

drivers/dma-buf/heaps/system_heap.c static struct dma_buf *system_heap_allocate(struct dma_heap *heap, · line 494

Quoted 9 lines
drivers/dma-buf/heaps/system_heap.c +1 -1
@@ -485,7 +488,7 @@ static struct dma_buf *system_heap_allocate(struct dma_heap *heap,
 		 * Intentionally leak pages that cannot be re-encrypted
 		 * to prevent shared memory from being reused.
 		 */
-		if (buffer->cc_shared &&
+		if (cc_shared_buffer(buffer) &&
 		    system_heap_set_page_encrypted(p))
 			continue;
 		__free_pages(p, compound_order(p));

[Severity: High]
This isn't a bug introduced by this patch, but does this error path unintentionally re-encrypt pages that were never decrypted?

If system_heap_set_page_decrypted() fails during the earlier loop in system_heap_allocate(), it jumps to the free_pages label. However, the cleanup loop here uses for_each_sgtable_sg(table, sg, i), which iterates over the entire scatterlist starting from index 0.

This means pages after the failure point, which were never decrypted, will still be passed to system_heap_set_page_encrypted(). If the underlying set_memory_encrypted() call fails on already-encrypted pages (e.g., due to hypervisor state enforcement), the code skips __free_pages() and leaks the memory.

Could this loop be modified to only re-encrypt the pages that were successfully decrypted before the failure?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
reviewer Reviewed-by on v1 1/1
#2 lore raw
In reply to Sumit Semwal
Quoted 13 lines · Sumit Semwal

From: Arnd Bergmann <[email protected]>

While system heap and system_cc_shared heap share a lot of code and hence the same source file, their users have different needs.

system heap users need it to be a loadable module, while system_cc_shared heap users don't.

Building as a loadable module breaks system_cc_shared heap on powerpc and s390 due to un-exported set_memory_encrypted / set_memory_decrypted functions.

Quoted 8 lines

Fix these by reorganising code to put the system_cc_shared heap under a new Kconfig symbol, which allows either building both into the kernel, or leave encryption up to the consumers of the system heap.

Fixes fd55edff8a0a ("dma-buf: heaps: system: Turn the heap into a module")
Reviewed-by on v1 1/1
#4 lore raw
In reply to Sumit Semwal
Quoted 30 lines · Wed, Jun 10, 2026 at 04:23:29PM +0200, [email protected] wrote:

From: Arnd Bergmann <[email protected]>

While system heap and system_cc_shared heap share a lot of code and hence the same source file, their users have different needs.

system heap users need it to be a loadable module, while system_cc_shared heap users don't.

Building as a loadable module breaks system_cc_shared heap on powerpc and s390 due to un-exported set_memory_encrypted / set_memory_decrypted functions.

Fix these by reorganising code to put the system_cc_shared heap under a new Kconfig symbol, which allows either building both into the kernel, or leave encryption up to the consumers of the system heap.

Fixes fd55edff8a0a ("dma-buf: heaps: system: Turn the heap into a module")

---
drivers/dma-buf/heaps/Kconfig | 8 ++++++++
drivers/dma-buf/heaps/system_heap.c | 16 ++++++++++------

2 files changed, 18 insertions(+), 6 deletions(-)
drivers/dma-buf/heaps/Kconfig +0 -0
index e273fb18feca..a39decdcf067 100644

drivers/dma-buf/heaps/Kconfig config DMABUF_HEAPS_SYSTEM · line 8

Quoted 4 lines
drivers/dma-buf/heaps/Kconfig +1 -0
@@ -5,6 +5,14 @@ config DMABUF_HEAPS_SYSTEM
 	  Choose this option to enable the system dmabuf heap. The system heap
 	  is backed by pages from the buddy allocator. If in doubt, say Y.
 
+config DMABUF_HEAPS_CC_SYSTEM

Nit: "DMABUF_HEAPS_SYSTEM_CC_SHARED" to be consistent with the heap name?

With or without it:

author maintainer
#5 lore raw

Hi Jiri,

Quoted 35 lines · Jiri Pirko
Quoted 33 lines · Wed, Jun 10, 2026 at 04:23:29PM +0200, [email protected] wrote:

From: Arnd Bergmann <[email protected]>

While system heap and system_cc_shared heap share a lot of code and hence the same source file, their users have different needs.

system heap users need it to be a loadable module, while system_cc_shared heap users don't.

Building as a loadable module breaks system_cc_shared heap on powerpc and s390 due to un-exported set_memory_encrypted / set_memory_decrypted functions.

Fix these by reorganising code to put the system_cc_shared heap under a new Kconfig symbol, which allows either building both into the kernel, or leave encryption up to the consumers of the system heap.

Fixes fd55edff8a0a ("dma-buf: heaps: system: Turn the heap into a module")

---
drivers/dma-buf/heaps/Kconfig | 8 ++++++++
drivers/dma-buf/heaps/system_heap.c | 16 ++++++++++------

2 files changed, 18 insertions(+), 6 deletions(-)
drivers/dma-buf/heaps/Kconfig +0 -0
index e273fb18feca..a39decdcf067 100644
@@ -5,6 +5,14 @@ config DMABUF_HEAPS_SYSTEM
        Choose this option to enable the system dmabuf heap. The system heap
        is backed by pages from the buddy allocator. If in doubt, say Y.
 
Quoted 8 lines

drivers/dma-buf/heaps/Kconfig config DMABUF_HEAPS_SYSTEM · line 8

Quoted 1 line
drivers/dma-buf/heaps/Kconfig +1 -0
+config DMABUF_HEAPS_CC_SYSTEM

Nit: "DMABUF_HEAPS_SYSTEM_CC_SHARED" to be consistent with the heap name?

With or without it:

Thanks for catching this; I'll fix this while pushing to drm-misc-next-fixes in a few minutes.

Best,
Sumit.