mtd: only use __xipram annotation when XIP_KERNEL is set

Message ID 1453736525-1959191-2-git-send-email-arnd@arndb.de
State New
Headers show

Commit Message

Arnd Bergmann Jan. 25, 2016, 3:41 p.m.
When XIP_KERNEL is enabled, some functions are defined in the .data
ELF section because we require them to be in RAM whenever we communicate
with the flash chip. However this causes problems when FTRACE is
enabled and gcc emits calls to __gnu_mcount_nc in the function
prolog:

drivers/built-in.o: In function `cfi_chip_setup':
:(.data+0x272fc): relocation truncated to fit: R_ARM_CALL against symbol `__gnu_mcount_nc' defined in .text section in arch/arm/kernel/built-in.o
drivers/built-in.o: In function `cfi_probe_chip':
:(.data+0x27de8): relocation truncated to fit: R_ARM_CALL against symbol `__gnu_mcount_nc' defined in .text section in arch/arm/kernel/built-in.o
/tmp/ccY172rP.s: Assembler messages:
/tmp/ccY172rP.s:70: Warning: ignoring changed section attributes for .data
/tmp/ccY172rP.s: Error: 1 warning, treating warnings as errors
make[5]: *** [drivers/mtd/chips/cfi_probe.o] Error 1
/tmp/ccK4rjeO.s: Assembler messages:
/tmp/ccK4rjeO.s:421: Warning: ignoring changed section attributes for .data
/tmp/ccK4rjeO.s: Error: 1 warning, treating warnings as errors
make[5]: *** [drivers/mtd/chips/cfi_util.o] Error 1
/tmp/ccUvhCYR.s: Assembler messages:
/tmp/ccUvhCYR.s:1895: Warning: ignoring changed section attributes for .data
/tmp/ccUvhCYR.s: Error: 1 warning, treating warnings as errors

Specifically, this does not work because the .data section is not
marked executable, which leads LD to not generate trampolines for
long calls.

This moves the __xipram functions into their own .xiptext section instead.
The section is still placed next to .data and located in RAM but is marked
executable, which avoids the build errors.

Also, we only need to place the XIP functions into a separate section
if both CONFIG_XIP_KERNEL and CONFIG_MTD_XIP are set: When only MTD_XIP
is used, the whole kernel is still in RAM and we do not need to worry
about pulling out the rug under it. When only XIP_KERNEL but not MTD_XIP
is set, the kernel is in some form of ROM, but we never write to it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>

---
 include/asm-generic/vmlinux.lds.h |  1 +
 include/linux/mtd/xip.h           | 10 ++++++----
 2 files changed, 7 insertions(+), 4 deletions(-)

-- 
2.7.0


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Comments

Arnd Bergmann March 5, 2016, 12:33 a.m. | #1
On Friday 04 March 2016 16:22:03 Brian Norris wrote:
> Hi Arnd,

> 

> On Sat, Mar 05, 2016 at 01:19:21AM +0100, Arnd Bergmann wrote:

> > On Friday 04 March 2016 16:02:25 Brian Norris wrote:

> > > On Mon, Jan 25, 2016 at 04:41:50PM +0100, Arnd Bergmann wrote:

> > > > When XIP_KERNEL is enabled, some functions are defined in the .data

> > > > ELF section because we require them to be in RAM whenever we communicate

> > > > with the flash chip. However this causes problems when FTRACE is

> > > > enabled and gcc emits calls to __gnu_mcount_nc in the function

> > > > prolog:

> > > > 

> > > > drivers/built-in.o: In function `cfi_chip_setup':

> > > > :(.data+0x272fc): relocation truncated to fit: R_ARM_CALL against symbol `__gnu_mcount_nc' defined in .text section in arch/arm/kernel/built-in.o

> > > > drivers/built-in.o: In function `cfi_probe_chip':

> > > > :(.data+0x27de8): relocation truncated to fit: R_ARM_CALL against symbol `__gnu_mcount_nc' defined in .text section in arch/arm/kernel/built-in.o

> > > > /tmp/ccY172rP.s: Assembler messages:

> > > > /tmp/ccY172rP.s:70: Warning: ignoring changed section attributes for .data

> > > > /tmp/ccY172rP.s: Error: 1 warning, treating warnings as errors

> > > > make[5]: *** [drivers/mtd/chips/cfi_probe.o] Error 1

> > > > /tmp/ccK4rjeO.s: Assembler messages:

> > > > /tmp/ccK4rjeO.s:421: Warning: ignoring changed section attributes for .data

> > > > /tmp/ccK4rjeO.s: Error: 1 warning, treating warnings as errors

> > > > make[5]: *** [drivers/mtd/chips/cfi_util.o] Error 1

> > > > /tmp/ccUvhCYR.s: Assembler messages:

> > > > /tmp/ccUvhCYR.s:1895: Warning: ignoring changed section attributes for .data

> > > > /tmp/ccUvhCYR.s: Error: 1 warning, treating warnings as errors

> > > 

> > > Can you provide a sample .config that DOES build correctly with

> > > XIP_KERNEL enabled + this patch? My first attempt yields some other

> > > failures I don't care to fixup right now...

> > > 

> > > Anyway, I don't doubt you have a good fix here, so I can probably take

> > > it. Any review from others would be welcome though.

> > 

> > I found the config in the attachment in my logs.

> 

> Thanks...

> 

> > To get this config working, I also needed this hunk from my set of

> > old unsubmitted patches:

> 

> ...but, does anyone care about XIP / MTD_XIP then, if the first two

> examples we have both have long-standing build issues?

> 


Probably not. I just checked the third user (omap1), and it seems that one
has been broken since 2009 with 941132606c76 ("OMAP: Remove OMAP_IO_ADDRESS,
use OMAP1_IO_ADDRESS and OMAP2_IO_ADDRESS instead"), when Tony replaced
the inline omap_readl() with an extern function, thus breaking the
implementation of xip_irqpending() that must be inlined.

The other two have apparently been broken since 2011, by other patches
that also broke compilation.

	Arnd

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
Arnd Bergmann March 6, 2016, 7:56 p.m. | #2
On Saturday 05 March 2016, Robert Jarzmik wrote:
> Arnd Bergmann <arnd@arndb.de> writes:

> 

> > I guess we can bring back the macros for the case that MTD_XIP and XIP_KERNEL

> > are both enabled.

> >

> > 	Arnd

> I wouldn't have ICMR and ICIP exposed to drivers, Eric's original move looks

> corect to me.


Yes, it's a reasonable cleanup that we have done in many places, except
this one introduced a regression for MTD_XIP.

> On the other hand, I'm wondering if xip_irqpending(), xip_currtime() and

> xip_cpu_idle() should be declared as functions in mtd-xip.h, and be part of

> something like arch/arm/mach-pxa/xip.c ...


They have to at least be marked __xipram and must not reference anything
in .text or .rodata, but there might be additional requirements.

	Arnd

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Patch hide | download patch | download mbox

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index c4bd0e2c173c..73a2c72e4252 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -199,6 +199,7 @@ 
 
 /* .data section */
 #define DATA_DATA							\
+	*(.xiptext)							\
 	*(.data)							\
 	*(.ref.data)							\
 	*(.data..shared_aligned) /* percpu related */			\
diff --git a/include/linux/mtd/xip.h b/include/linux/mtd/xip.h
index abed4dec5c2f..e373690cce0a 100644
--- a/include/linux/mtd/xip.h
+++ b/include/linux/mtd/xip.h
@@ -30,7 +30,9 @@ 
  * obviously not be running from flash.  The __xipram is therefore marking
  * those functions so they get relocated to ram.
  */
-#define __xipram noinline __attribute__ ((__section__ (".data")))
+#ifdef CONFIG_XIP_KERNEL
+#define __xipram noinline __attribute__ ((__section__ (".xiptext")))
+#endif
 
 /*
  * Each architecture has to provide the following macros.  They must access
@@ -90,10 +92,10 @@ 
 #define xip_cpu_idle()  do { } while (0)
 #endif
 
-#else
+#endif /* CONFIG_MTD_XIP */
 
+#ifndef __xipram
 #define __xipram
-
-#endif /* CONFIG_MTD_XIP */
+#endif
 
 #endif /* __LINUX_MTD_XIP_H__ */