xen/arm: page: Avoid pointer overflow on cache clean & invalidate

On Arm32, after cleaning and invalidating the last dcache line of the top
domheap page i.e. VA = 0xfffff000 (as a result of flushing the page to
RAM), we end up adding the value of a dcache line size to the pointer
once again, which results in a pointer arithmetic overflow (with 64B line
size, operation 0xffffffc0 + 0x40 overflows to 0x0). Such behavior is
undefined and given the wide range of compiler versions we support, it is
difficult to determine what could happen in such scenario.

Modify clean_and_invalidate_dcache_va_range() as well as
clean_dcache_va_range() and invalidate_dcache_va_range() due to similarity
of handling to prevent pointer arithmetic overflow. Modify the loops to
use an additional variable to store the index of the next cacheline.
Add an assert to prevent passing a region that wraps around which is
illegal and would end up in a page fault anyway (region 0-2MB is
unmapped). Lastly, return early if size passed is 0.

Note that on Arm64, we don't have this problem given that the max VA
space we support is 48-bits.

This is XSA-447 / CVE-2023-46837.

Signed-off-by: Michal Orzel <michal.orzel@amd.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
master commit: 190b7f49af6487a9665da63d43adc9d9a5fbd01e
master date: 2023-12-12 14:01:00 +0100

diff --git a/xen/include/asm-arm/page.h b/xen/include/asm-arm/page.h
index eff5883ef87b8966835c1dd23b1ccd94a3cf4406..b6784417ed347bb7a302877e75395da5c84e8565 100644 (file)
--- a/xen/include/asm-arm/page.h
+++ b/xen/include/asm-arm/page.h
@@ -153,6 +153,13 @@ static inline size_t read_dcache_line_bytes(void)
 static inline int invalidate_dcache_va_range(const void *p, unsigned long size)
 {
     size_t cacheline_mask = dcache_line_bytes - 1;
+    unsigned long idx = 0;
+
+    if ( !size )
+        return 0;
+
+    /* Passing a region that wraps around is illegal */
+    ASSERT(((uintptr_t)p + size - 1) >= (uintptr_t)p);
 
     dsb(sy);           /* So the CPU issues all writes to the range */
 
@@ -165,11 +172,11 @@ static inline int invalidate_dcache_va_range(const void *p, unsigned long size)
     }
 
     for ( ; size >= dcache_line_bytes;
-            p += dcache_line_bytes, size -= dcache_line_bytes )
-        asm volatile (__invalidate_dcache_one(0) : : "r" (p));
+            idx += dcache_line_bytes, size -= dcache_line_bytes )
+        asm volatile (__invalidate_dcache_one(0) : : "r" (p + idx));
 
     if ( size > 0 )
-        asm volatile (__clean_and_invalidate_dcache_one(0) : : "r" (p));
+        asm volatile (__clean_and_invalidate_dcache_one(0) : : "r" (p + idx));
 
     dsb(sy);           /* So we know the flushes happen before continuing */
 
@@ -179,14 +186,21 @@ static inline int invalidate_dcache_va_range(const void *p, unsigned long size)
 static inline int clean_dcache_va_range(const void *p, unsigned long size)
 {
     size_t cacheline_mask = dcache_line_bytes - 1;
+    unsigned long idx = 0;
+
+    if ( !size )
+        return 0;
+
+    /* Passing a region that wraps around is illegal */
+    ASSERT(((uintptr_t)p + size - 1) >= (uintptr_t)p);
 
     dsb(sy);           /* So the CPU issues all writes to the range */
     size += (uintptr_t)p & cacheline_mask;
     size = (size + cacheline_mask) & ~cacheline_mask;
     p = (void *)((uintptr_t)p & ~cacheline_mask);
     for ( ; size >= dcache_line_bytes;
-            p += dcache_line_bytes, size -= dcache_line_bytes )
-        asm volatile (__clean_dcache_one(0) : : "r" (p));
+            idx += dcache_line_bytes, size -= dcache_line_bytes )
+        asm volatile (__clean_dcache_one(0) : : "r" (p + idx));
     dsb(sy);           /* So we know the flushes happen before continuing */
     /* ARM callers assume that dcache_* functions cannot fail. */
     return 0;
@@ -196,14 +210,21 @@ static inline int clean_and_invalidate_dcache_va_range
     (const void *p, unsigned long size)
 {
     size_t cacheline_mask = dcache_line_bytes - 1;
+    unsigned long idx = 0;
+
+    if ( !size )
+        return 0;
+
+    /* Passing a region that wraps around is illegal */
+    ASSERT(((uintptr_t)p + size - 1) >= (uintptr_t)p);
 
     dsb(sy);         /* So the CPU issues all writes to the range */
     size += (uintptr_t)p & cacheline_mask;
     size = (size + cacheline_mask) & ~cacheline_mask;
     p = (void *)((uintptr_t)p & ~cacheline_mask);
     for ( ; size >= dcache_line_bytes;
-            p += dcache_line_bytes, size -= dcache_line_bytes )
-        asm volatile (__clean_and_invalidate_dcache_one(0) : : "r" (p));
+            idx += dcache_line_bytes, size -= dcache_line_bytes )
+        asm volatile (__clean_and_invalidate_dcache_one(0) : : "r" (p + idx));
     dsb(sy);         /* So we know the flushes happen before continuing */
     /* ARM callers assume that dcache_* functions cannot fail. */
     return 0;