[PATCH] firmware: Prevent _start from not being at the beginning

[PATCH] firmware: Prevent _start from not being at the beginning

[Xiang W]

There are multiple .entry sections in opensbi, and _start may be put
after other .entry sections, which may cause problems. The patch fix
this issue.

Signed-off-by: Xiang W <wxjstz@126.com>
---
 firmware/fw_base.S   | 2 +-
 firmware/fw_base.ldS | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/firmware/fw_base.S b/firmware/fw_base.S
index 09894e6..7ade6a5 100644
--- a/firmware/fw_base.S
+++ b/firmware/fw_base.S
@@ -42,7 +42,7 @@
 999:
 .endm
 
-	.section .entry, "ax", %progbits
+	.section .entry0, "ax", %progbits
 	.align 3
 	.globl _start
 	.globl _start_warm
diff --git a/firmware/fw_base.ldS b/firmware/fw_base.ldS
index fb47984..c1dc743 100644
--- a/firmware/fw_base.ldS
+++ b/firmware/fw_base.ldS
@@ -18,6 +18,7 @@
 	.text :
  	{
 		PROVIDE(_text_start = .);
+		*(.entry0)
 		*(.entry)
 		*(.text)
 		. = ALIGN(8);
-- 
2.43.0

[PATCH] firmware: Prevent _start from not being at the beginning

[Andreas Schwab]

On M?r 05 2024, Xiang W wrote:

> There are multiple .entry sections in opensbi, and _start may be put
> after other .entry sections, which may cause problems. The patch fix
> this issue.

That should not be a problem.  Since it is passed as the first object
file to the linker it is guaranteed to be placed first.

-- 
Andreas Schwab, SUSE Labs, schwab at suse.de
GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE  1748 E4D4 88E3 0EEA B9D7
"And now for something completely different."

[PATCH] firmware: Prevent _start from not being at the beginning

[Xiang W]

? 2024-03-05???? 11:43 +0100?Andreas Schwab???
> On M?r 05 2024, Xiang W wrote:
> 
> > There are multiple .entry sections in opensbi, and _start may be put
> > after other .entry sections, which may cause problems. The patch fix
> > this issue.
> 
> That should not be a problem.? Since it is passed as the first object
> file to the linker it is guaranteed to be placed first.
> 
Thank you for your reply

Is this the default behavior of the linker or is there any standard
provision? If there are no standards, modifications will still be?
needed.

Regards,
Xiang W