Hi,
The enclosed patch allows to remove the DMI scanning code when
CONFIG_EMBEDDED is defined. It's basically the dma_blacklist patch of
Linux-Tiny ported to 2.6.25-rc1, with the required modifications. It
allows to remove ~10k from the kernel code/data size.
On top of this patch, I've tested if removing the big dmi tables in the
code (for example in arch/x86/kernel/reboot.c) would allow to make more
space optimizations. However, it seems that simply defining
dmi_check_system() to an empty static inlined function already allows
gcc to optimize out the dmi tables, because there are not present in
the code. Is that possible, or is my understanding incorrect ?
Of course, your comments on this patch are more than welcome.
Thanks!
Thomas
---
Turn CONFIG_DMI into a selectable option if EMBEDDED is defined, in
order to be able to remove the DMI table scanning code if it's not
needed, and then reduce the kernel code size.
With CONFIG_DMI (i.e before) :
text data bss dec hex filename
1076076 128656 98304 1303036 13e1fc vmlinux
Without CONFIG_DMI (i.e after) :
text data bss dec hex filename
1068092 126308 98304 1292704 13b9a0 vmlinux
Result:
text data bss dec hex filename
-7984 -2348 0 -10332 -285c vmlinux
The new option appears in "Processor type and features", only when
CONFIG_EMBEDDED is defined.
Signed-Off-By: Thomas Petazzoni <[email protected]>
---
arch/x86/Kconfig | 12 +++++++++---
include/linux/dmi.h | 1 +
2 files changed, 10 insertions(+), 3 deletions(-)
Index: linux/arch/x86/Kconfig
===================================================================
--- linux.orig/arch/x86/Kconfig 2008-02-11 16:51:50.000000000 +0100
+++ linux/arch/x86/Kconfig 2008-02-11 17:48:15.000000000 +0100
@@ -88,9 +88,6 @@
config ARCH_MAY_HAVE_PC_FDC
def_bool y
-config DMI
- def_bool y
-
config RWSEM_GENERIC_SPINLOCK
def_bool !X86_XADD
@@ -433,6 +430,15 @@
# Mark as embedded because too many people got it wrong.
# The code disables itself when not needed.
+config DMI
+ default y
+ bool "Enable DMI scanning" if EMBEDDED
+ help
+ Enabled scanning of DMI to identify machine quirks. Say Y
+ here unless you have verified that your setup is not
+ affected by entries in the DMI blacklist. Required by PNP
+ BIOS code.
+
config GART_IOMMU
bool "GART IOMMU support" if EMBEDDED
default y
Index: linux/include/linux/dmi.h
===================================================================
--- linux.orig/include/linux/dmi.h 2008-02-11 16:51:30.000000000 +0100
+++ linux/include/linux/dmi.h 2008-02-11 17:40:31.000000000 +0100
@@ -90,6 +90,7 @@
static inline const char * dmi_get_system_info(int field) { return NULL; }
static inline const struct dmi_device * dmi_find_device(int type, const char *name,
const struct dmi_device *from) { return NULL; }
+static inline void dmi_scan_machine(void) { return; }
static inline int dmi_get_year(int year) { return 0; }
static inline int dmi_name_in_vendors(const char *s) { return 0; }
#define dmi_available 0
--
Thomas Petazzoni, Free Electrons
Free Embedded Linux Training Materials
on http://free-electrons.com/training
(More than 1500 pages!)
On Mon, 2008-02-11 at 17:58 +0100, Thomas Petazzoni wrote:
> Hi,
>
> The enclosed patch allows to remove the DMI scanning code when
> CONFIG_EMBEDDED is defined. It's basically the dma_blacklist patch of
> Linux-Tiny ported to 2.6.25-rc1, with the required modifications. It
> allows to remove ~10k from the kernel code/data size.
Looks ok. Please preserve original authorship (ie me) in some fashion in
your description.
> On top of this patch, I've tested if removing the big dmi tables in the
> code (for example in arch/x86/kernel/reboot.c) would allow to make more
> space optimizations. However, it seems that simply defining
> dmi_check_system() to an empty static inlined function already allows
> gcc to optimize out the dmi tables, because there are not present in
> the code. Is that possible, or is my understanding incorrect ?
That's possible with modern gccs, yes.
--
Mathematics is the supreme nostalgia of our time.
Hi Matt,
Thanks for your review.
Le Mon, 11 Feb 2008 13:01:18 -0600,
Matt Mackall <[email protected]> a écrit :
> Looks ok. Please preserve original authorship (ie me) in some fashion
> in your description.
Andrew seems to already have picked up the patch in -mm (I received an
e-mail + x86-configurable-dmi-scanning-code.patch added to -mm tree).
If I resend a new one with proper credits, will Andrew pick it up and
replace the old version by the new version ?
> > On top of this patch, I've tested if removing the big dmi tables in
> > the code (for example in arch/x86/kernel/reboot.c) would allow to
> > make more space optimizations. However, it seems that simply
> > defining dmi_check_system() to an empty static inlined function
> > already allows gcc to optimize out the dmi tables, because there
> > are not present in the code. Is that possible, or is my
> > understanding incorrect ?
>
> That's possible with modern gccs, yes.
I was compiling with 4.2.3, which I suppose is a modern gcc.
Thanks,
Thomas
--
Thomas Petazzoni, Free Electrons
Free Embedded Linux Training Materials
on http://free-electrons.com/training
(More than 1500 pages!)
On Tue, 12 Feb 2008 08:33:23 +0100 Thomas Petazzoni <[email protected]> wrote:
> If I resend a new one with proper credits, will Andrew pick it up and
> replace the old version by the new version ?
yup.