On Mon, May 28, 2012 at 10:30:05PM +0200, Sam Ravnborg wrote:
On Mon, May 28, 2012 at 07:33:37PM +0100, Vincent Sanders wrote:
quoted
From: Vincent Sanders <redacted>
The constructor symbol name is different between platforms. Allow this
to be selected by configuration and set suitable default values.
Signed-off-by: Vincent Sanders <redacted>
---
include/asm-generic/vmlinux.lds.h | 6 +++---
init/Kconfig | 6 ++++++
kernel/module.c | 2 +-
3 files changed, 10 insertions(+), 4 deletions(-)
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 8aeadf6..fd34808 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -471,9 +471,9 @@
}
#ifdef CONFIG_CONSTRUCTORS
-#define KERNEL_CTORS() . = ALIGN(8); \
- VMLINUX_SYMBOL(__ctors_start) = .; \
- *(.ctors) \
+#define KERNEL_CTORS() . = ALIGN(8); \
+ VMLINUX_SYMBOL(__ctors_start) = .; \
+ *(CONFIG_CONSTRUCTORS_NAME) \
VMLINUX_SYMBOL(__ctors_end) = .;
What is wrong with adding both "standard" names for ctors uncnditionally?
Like this:
quoted
*(.ctors) \
+ *(.init_array) \
Persoinally I have no strong opinion and would alter approach.
As I understand it the previous objections to the "put them all in"
approach (this change has been round the loop before) were:
- It contaminates the x86 symbol namespace and if init_array was used on
not ARM it might lead to surprises
- If another architecture needs something different again then we are
going to end up with another symbol.
- The module loading code has to be changed to allow either symbol.
The only technical issue really is the module symbol handling and I am
unsure how to robustly handle that.
--
Regards Vincent
http://www.kyllikki.org/