[U-Boot] [PATCH 1/5] sandbox: Add instructions about building on 32-bit machines

[U-Boot] [PATCH 1/5] sandbox: Add instructions about building on 32-bit machines

[Simon Glass]

Sandbox is built with 64-bit ints by default. This doesn't work properly on
32-bit machines.

Signed-off-by: Simon Glass <sjg@chromium.org>
---

 board/sandbox/README.sandbox | 3 +++
 include/configs/sandbox.h    | 5 ++++-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/board/sandbox/README.sandbox b/board/sandbox/README.sandbox
index 9fe3bf1..ed820d3 100644
--- a/board/sandbox/README.sandbox
+++ b/board/sandbox/README.sandbox
@@ -44,6 +44,9 @@ Note:
       make sandbox_defconfig all NO_SDL=1
       ./u-boot
 
+   If you are building on a 32-bit machine you may get errors from __ffs.h
+   about shifting more than the machine word size. Edit the config file
+   include/configs/sandbox.h and change CONFIG_SANDBOX_BITS_PER_LONG to 32.
 
 U-Boot will start on your computer, showing a sandbox emulation of the serial
 console:
diff --git a/include/configs/sandbox.h b/include/configs/sandbox.h
index 4de89f8..197d8bb 100644
--- a/include/configs/sandbox.h
+++ b/include/configs/sandbox.h
@@ -27,7 +27,10 @@
 
 #define CONFIG_SYS_STDIO_DEREGISTER
 
-/* Number of bits in a C 'long' on this architecture */
+/*
+ * Number of bits in a C 'long' on this architecture. Set this to 32 when
+ * building on a 32-bit machine.
+ */
 #define CONFIG_SANDBOX_BITS_PER_LONG	64
 
 #define CONFIG_LMB
-- 
2.8.0.rc3.226.g39d4020

[U-Boot] [PATCH 2/5] lzmadec: Use the same type as the lzma call

[Simon Glass]

With sandbox on 32-bit the size_t type can be a little inconsistent. Use
the same type as the caller expects to avoid a compiler warning.

Signed-off-by: Simon Glass <sjg@chromium.org>
---

 cmd/lzmadec.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/cmd/lzmadec.c b/cmd/lzmadec.c
index 1ad9ed6..c78df82 100644
--- a/cmd/lzmadec.c
+++ b/cmd/lzmadec.c
@@ -20,7 +20,7 @@
 static int do_lzmadec(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 {
 	unsigned long src, dst;
-	unsigned long src_len = ~0UL, dst_len = ~0UL;
+	SizeT src_len = ~0UL, dst_len = ~0UL;
 	int ret;
 
 	switch (argc) {
@@ -40,7 +40,8 @@ static int do_lzmadec(cmd_tbl_t *cmdtp, int flag, int argc, char *const argv[])
 
 	if (ret != SZ_OK)
 		return 1;
-	printf("Uncompressed size: %ld = 0x%lX\n", src_len, src_len);
+	printf("Uncompressed size: %ld = %#lX\n", (ulong)src_len,
+	       (ulong)src_len);
 	setenv_hex("filesize", src_len);
 
 	return 0;
-- 
2.8.0.rc3.226.g39d4020

[U-Boot] [PATCH 3/5] part_efi: Fix compiler warning on 32-bit sandbox

[Simon Glass]

This fixes a mismatch between the %zu format and the type used on sandbox.

Signed-off-by: Simon Glass <sjg@chromium.org>
---

 disk/part_efi.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/disk/part_efi.c b/disk/part_efi.c
index 0af1e92..01f71be 100644
--- a/disk/part_efi.c
+++ b/disk/part_efi.c
@@ -886,9 +886,10 @@ static gpt_entry *alloc_read_gpt_entries(struct blk_desc *dev_desc,
 	count = le32_to_cpu(pgpt_head->num_partition_entries) *
 		le32_to_cpu(pgpt_head->sizeof_partition_entry);
 
-	debug("%s: count = %u * %u = %zu\n", __func__,
+	debug("%s: count = %u * %u = %lu\n", __func__,
 	      (u32) le32_to_cpu(pgpt_head->num_partition_entries),
-	      (u32) le32_to_cpu(pgpt_head->sizeof_partition_entry), count);
+	      (u32) le32_to_cpu(pgpt_head->sizeof_partition_entry),
+	      (ulong)count);
 
 	/* Allocate memory for PTE, remember to FREE */
 	if (count != 0) {
@@ -897,9 +898,8 @@ static gpt_entry *alloc_read_gpt_entries(struct blk_desc *dev_desc,
 	}
 
 	if (count == 0 || pte == NULL) {
-		printf("%s: ERROR: Can't allocate 0x%zX "
-		       "bytes for GPT Entries\n",
-			__func__, count);
+		printf("%s: ERROR: Can't allocate %#lX bytes for GPT Entries\n",
+		       __func__, (ulong)count);
 		return NULL;
 	}
 
-- 
2.8.0.rc3.226.g39d4020

[U-Boot] [PATCH 4/5] hashtable: Fix compiler warning on 32-bit sandbox

[Simon Glass]

This fixes a mismatch between the %zu format and the type used on sandbox.

Signed-off-by: Simon Glass <sjg@chromium.org>
---

 lib/hashtable.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/lib/hashtable.c b/lib/hashtable.c
index 435e2a6..4e52b36 100644
--- a/lib/hashtable.c
+++ b/lib/hashtable.c
@@ -602,8 +602,8 @@ ssize_t hexport_r(struct hsearch_data *htab, const char sep, int flag,
 		return (-1);
 	}
 
-	debug("EXPORT  table = %p, htab.size = %d, htab.filled = %d, "
-		"size = %zu\n", htab, htab->size, htab->filled, size);
+	debug("EXPORT  table = %p, htab.size = %d, htab.filled = %d, size = %lu\n",
+	      htab, htab->size, htab->filled, (ulong)size);
 	/*
 	 * Pass 1:
 	 * search used entries,
@@ -657,8 +657,8 @@ ssize_t hexport_r(struct hsearch_data *htab, const char sep, int flag,
 	/* Check if the user supplied buffer size is sufficient */
 	if (size) {
 		if (size < totlen + 1) {	/* provided buffer too small */
-			printf("Env export buffer too small: %zu, "
-				"but need %zu\n", size, totlen + 1);
+			printf("Env export buffer too small: %lu, but need %lu\n",
+			       (ulong)size, (ulong)totlen + 1);
 			__set_errno(ENOMEM);
 			return (-1);
 		}
@@ -790,7 +790,7 @@ int himport_r(struct hsearch_data *htab,
 
 	/* we allocate new space to make sure we can write to the array */
 	if ((data = malloc(size + 1)) == NULL) {
-		debug("himport_r: can't malloc %zu bytes\n", size + 1);
+		debug("himport_r: can't malloc %lu bytes\n", (ulong)size + 1);
 		__set_errno(ENOMEM);
 		return 0;
 	}
-- 
2.8.0.rc3.226.g39d4020

[U-Boot] [PATCH 5/5] dtoc: Correct the type widening code in fdt_fallback

[Simon Glass]

This code does not match the fdt version in fdt.py. When dtoc is unable to
use the Python libfdt library, it uses the fallback version, which does not
widen arrays correctly.

Fix this to avoid a warning 'excess elements in array initialize' in
dt-platdata.c which happens on some platforms.

Reported-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
---

 tools/dtoc/fdt_fallback.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/dtoc/fdt_fallback.py b/tools/dtoc/fdt_fallback.py
index 14decf3..9ed11e4 100644
--- a/tools/dtoc/fdt_fallback.py
+++ b/tools/dtoc/fdt_fallback.py
@@ -71,6 +71,12 @@ class Prop:
         if type(newprop.value) == list and type(self.value) != list:
             self.value = newprop.value
 
+        if type(self.value) == list and len(newprop.value) > len(self.value):
+            val = fdt_util.GetEmpty(self.type)
+            while len(self.value) < len(newprop.value):
+                self.value.append(val)
+
+
 class Node:
     """A device tree node
 
-- 
2.8.0.rc3.226.g39d4020

[U-Boot] [PATCH 1/5] sandbox: Add instructions about building on 32-bit machines

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:45AM -0600, Simon Glass wrote:

> Sandbox is built with 64-bit ints by default. This doesn't work properly on
> 32-bit machines.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>

Reviewed-by: Tom Rini <trini@konsulko.com>

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160722/308812c3/attachment.sig>

[U-Boot] [PATCH 2/5] lzmadec: Use the same type as the lzma call

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:46AM -0600, Simon Glass wrote:

> With sandbox on 32-bit the size_t type can be a little inconsistent. Use
> the same type as the caller expects to avoid a compiler warning.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>

Reviewed-by: Tom Rini <trini@konsulko.com>

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160722/8bad12e0/attachment.sig>

[U-Boot] [PATCH 3/5] part_efi: Fix compiler warning on 32-bit sandbox

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:47AM -0600, Simon Glass wrote:

> This fixes a mismatch between the %zu format and the type used on sandbox.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>

Reviewed-by: Tom Rini <trini@konsulko.com>

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160722/937f7415/attachment.sig>

[U-Boot] [PATCH 4/5] hashtable: Fix compiler warning on 32-bit sandbox

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:48AM -0600, Simon Glass wrote:

> This fixes a mismatch between the %zu format and the type used on sandbox.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>

Reviewed-by: Tom Rini <trini@konsulko.com>

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160722/6718e941/attachment.sig>

[U-Boot] [PATCH 5/5] dtoc: Correct the type widening code in fdt_fallback

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:49AM -0600, Simon Glass wrote:

> This code does not match the fdt version in fdt.py. When dtoc is unable to
> use the Python libfdt library, it uses the fallback version, which does not
> widen arrays correctly.
> 
> Fix this to avoid a warning 'excess elements in array initialize' in
> dt-platdata.c which happens on some platforms.
> 
> Reported-by: Tom Rini <trini@konsulko.com>
> Signed-off-by: Simon Glass <sjg@chromium.org>

Tested-by: Tom Rini <trini@konsulko.com>

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160722/3719ea98/attachment.sig>

[U-Boot] [U-Boot, 1/5] sandbox: Add instructions about building on 32-bit machines

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:45AM -0600, Simon Glass wrote:

> Sandbox is built with 64-bit ints by default. This doesn't work properly on
> 32-bit machines.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>
> Reviewed-by: Tom Rini <trini@konsulko.com>

Applied to u-boot/master, thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160725/6b8c7d40/attachment.sig>

[U-Boot] [U-Boot, 2/5] lzmadec: Use the same type as the lzma call

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:46AM -0600, Simon Glass wrote:

> With sandbox on 32-bit the size_t type can be a little inconsistent. Use
> the same type as the caller expects to avoid a compiler warning.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>
> Reviewed-by: Tom Rini <trini@konsulko.com>

Applied to u-boot/master, thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160725/4ba51b37/attachment.sig>

[U-Boot] [U-Boot, 3/5] part_efi: Fix compiler warning on 32-bit sandbox

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:47AM -0600, Simon Glass wrote:

> This fixes a mismatch between the %zu format and the type used on sandbox.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>
> Reviewed-by: Tom Rini <trini@konsulko.com>

Applied to u-boot/master, thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160725/78f9e95b/attachment.sig>

[U-Boot] [U-Boot, 4/5] hashtable: Fix compiler warning on 32-bit sandbox

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:48AM -0600, Simon Glass wrote:

> This fixes a mismatch between the %zu format and the type used on sandbox.
> 
> Signed-off-by: Simon Glass <sjg@chromium.org>
> Reviewed-by: Tom Rini <trini@konsulko.com>

Applied to u-boot/master, thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160725/454f3ca2/attachment.sig>

[U-Boot] [U-Boot, 5/5] dtoc: Correct the type widening code in fdt_fallback

[Tom Rini]

On Fri, Jul 22, 2016 at 09:22:49AM -0600, Simon Glass wrote:

> This code does not match the fdt version in fdt.py. When dtoc is unable to
> use the Python libfdt library, it uses the fallback version, which does not
> widen arrays correctly.
> 
> Fix this to avoid a warning 'excess elements in array initialize' in
> dt-platdata.c which happens on some platforms.
> 
> Reported-by: Tom Rini <trini@konsulko.com>
> Signed-off-by: Simon Glass <sjg@chromium.org>
> Tested-by: Tom Rini <trini@konsulko.com>

Applied to u-boot/master, thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20160725/9e6bf2d4/attachment.sig>