libfdt: Internally perform potentially unaligned loads

Commits 6dcb8ba4 "libfdt: Add helpers for accessing unaligned words"
introduced changes to support unaligned reads for ARM platforms and
11738cf01f15 "libfdt: Don't use memcpy to handle unaligned reads on ARM"
improved the performance of these helpers.

On further discussion, while there are potential cases where we could be
used on platforms that do not fixup unaligned reads for us, making this
choice the default is very expensive in terms of binary size and access
time.  To address this, introduce and use new fdt{32,64}_ld_ functions
that call fdt{32,64}_to_cpu() as was done prior to the above mentioned
commits.  Leave the existing load functions as unaligned-safe and
include comments in both cases.

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Message-Id: <20201211022736.31657-1-trini@konsulko.com>
Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>

2 files changed, 29 insertions, 10 deletions

diff --git a/cpukit/dtc/libfdt/fdt_ro.c b/cpukit/dtc/libfdt/fdt_ro.c
index 91cc6fefe3..17584da257 100644
--- a/cpukit/dtc/libfdt/fdt_ro.c
+++ b/cpukit/dtc/libfdt/fdt_ro.c
@@ -181,8 +181,8 @@ int fdt_get_mem_rsv(const void *fdt, int n, uint64_t *address, uint64_t *size)
 	if (!can_assume(VALID_INPUT) && !re)
 		return -FDT_ERR_BADOFFSET;
 
-	*address = fdt64_ld(&re->address);
-	*size = fdt64_ld(&re->size);
+	*address = fdt64_ld_(&re->address);
+	*size = fdt64_ld_(&re->size);
 	return 0;
 }
 
@@ -192,7 +192,7 @@ int fdt_num_mem_rsv(const void *fdt)
 	const struct fdt_reserve_entry *re;
 
 	for (i = 0; (re = fdt_mem_rsv(fdt, i)) != NULL; i++) {
-		if (fdt64_ld(&re->size) == 0)
+		if (fdt64_ld_(&re->size) == 0)
 			return i;
 	}
 	return -FDT_ERR_TRUNCATED;
@@ -370,7 +370,7 @@ static const struct fdt_property *fdt_get_property_by_offset_(const void *fdt,
 	prop = fdt_offset_ptr_(fdt, offset);
 
 	if (lenp)
-		*lenp = fdt32_ld(&prop->len);
+		*lenp = fdt32_ld_(&prop->len);
 
 	return prop;
 }
@@ -408,7 +408,7 @@ static const struct fdt_property *fdt_get_property_namelen_(const void *fdt,
 			offset = -FDT_ERR_INTERNAL;
 			break;
 		}
-		if (fdt_string_eq_(fdt, fdt32_ld(&prop->nameoff),
+		if (fdt_string_eq_(fdt, fdt32_ld_(&prop->nameoff),
 				   name, namelen)) {
 			if (poffset)
 				*poffset = offset;
@@ -461,7 +461,7 @@ const void *fdt_getprop_namelen(const void *fdt, int nodeoffset,
 
 	/* Handle realignment */
 	if (!can_assume(LATEST) && fdt_version(fdt) < 0x10 &&
-	    (poffset + sizeof(*prop)) % 8 && fdt32_ld(&prop->len) >= 8)
+	    (poffset + sizeof(*prop)) % 8 && fdt32_ld_(&prop->len) >= 8)
 		return prop->data + 4;
 	return prop->data;
 }
@@ -479,7 +479,7 @@ const void *fdt_getprop_by_offset(const void *fdt, int offset,
 		int namelen;
 
 		if (!can_assume(VALID_INPUT)) {
-			name = fdt_get_string(fdt, fdt32_ld(&prop->nameoff),
+			name = fdt_get_string(fdt, fdt32_ld_(&prop->nameoff),
 					      &namelen);
 			if (!name) {
 				if (lenp)
@@ -488,13 +488,13 @@ const void *fdt_getprop_by_offset(const void *fdt, int offset,
 			}
 			*namep = name;
 		} else {
-			*namep = fdt_string(fdt, fdt32_ld(&prop->nameoff));
+			*namep = fdt_string(fdt, fdt32_ld_(&prop->nameoff));
 		}
 	}
 
 	/* Handle realignment */
 	if (!can_assume(LATEST) && fdt_version(fdt) < 0x10 &&
-	    (offset + sizeof(*prop)) % 8 && fdt32_ld(&prop->len) >= 8)
+	    (offset + sizeof(*prop)) % 8 && fdt32_ld_(&prop->len) >= 8)
 		return prop->data + 4;
 	return prop->data;
 }
@@ -519,7 +519,7 @@ uint32_t fdt_get_phandle(const void *fdt, int nodeoffset)
 			return 0;
 	}
 
-	return fdt32_ld(php);
+	return fdt32_ld_(php);
 }
 
 const char *fdt_get_alias_namelen(const void *fdt,
diff --git a/cpukit/dtc/libfdt/libfdt_internal.h b/cpukit/dtc/libfdt/libfdt_internal.h
index d4e0bd49c0..8b580aa507 100644
--- a/cpukit/dtc/libfdt/libfdt_internal.h
+++ b/cpukit/dtc/libfdt/libfdt_internal.h
@@ -46,6 +46,25 @@ static inline struct fdt_reserve_entry *fdt_mem_rsv_w_(void *fdt, int n)
 	return (void *)(uintptr_t)fdt_mem_rsv_(fdt, n);
 }
 
+/*
+ * Internal helpers to access tructural elements of the device tree blob
+ * (rather than for exaple reading integers from within property values).  We
+ * assume that we are either given a naturally aligned address for the platform
+ * or if we are not, we are on a platform where unaligned memory reads will be
+ * handled in a graceful manner.  If this is not the case there are _unaligned
+ * versions of these functions that follow and can be used.
+ *
+ */
+static inline uint32_t fdt32_ld_(const fdt32_t *p)
+{
+	return fdt32_to_cpu(*p);
+}
+
+static inline uint64_t fdt64_ld_(const fdt64_t *p)
+{
+	return fdt64_to_cpu(*p);
+}
+
 #define FDT_SW_MAGIC		(~FDT_MAGIC)
 
 /**********************************************************************/