diff mbox

v3.11-rc4: OMAP1/Amstrad Delta (E3) crash

Message ID 20130807230856.GJ23053@n2100.arm.linux.org.uk (mailing list archive)
State New, archived
Headers show

Commit Message

Russell King - ARM Linux Aug. 7, 2013, 11:08 p.m. UTC
On Wed, Aug 07, 2013 at 10:05:21PM +0300, Aaro Koskinen wrote:
> Hi,
> 
> On Wed, Aug 07, 2013 at 02:26:09AM +0300, Aaro Koskinen wrote:
> > I get the following on Amstrad Delta (E3) when booting unpatched
> > v3.11-rc4. The kernel config is after the crash.
> 
> This is regression introduced between -rc3 and -rc4. According to the
> git bisect, the first bad commit is:
> 
> 	commit e39e3f3ebfef03450cf7bfa7a974a8c61f7980c8
> 	Author: Russell King <rmk+kernel@arm.linux.org.uk>
> 	Date:   Tue Jul 9 01:03:17 2013 +0100
> 
> 	    ARM: update FIQ support for relocation of vectors
> 
> If I revert the commit from v3.11-rc4, the board boots fine.
> 
> A.
> 
> > Uncompressing Linux... done, booting the kernel.
> > [    0.000000] Booting Linux on physical CPU 0x0
> > [    0.000000] Initializing cgroup subsys cpu
> > [    0.000000] Linux version 3.11.0-rc4-e3 (aaro@blackmetal) (gcc version 4.7.3 (GCC) ) #4 PREEMPT Wed Aug 7 02:03:27 EEST 2013
> > [    0.000000] bootconsole [earlycon0] enabled
> > [    0.000000] OMAP1510
> > [    0.000000]  revision 2 handled as 15xx id: bc058c9b93111a16

Hmm, so ARM925T which is VIVT...

> > [    0.238868] Installing fiq handler from c001b110, length 0x164
> > [    0.245750] Unable to handle kernel paging request at virtual address ffff1224

In one way, that's good news - that's public independent proof that the
code associated with the vectors is well protected from kernel writes. :)

Can you try this (untested) patch please?  We need to write into this
page via the page's alias, rather than directly to it.  Thanks.

 arch/arm/kernel/fiq.c |    8 ++------
 1 files changed, 2 insertions(+), 6 deletions(-)


--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Aaro Koskinen Aug. 8, 2013, 12:05 a.m. UTC | #1
Hi,

On Thu, Aug 08, 2013 at 12:08:56AM +0100, Russell King - ARM Linux wrote:
> On Wed, Aug 07, 2013 at 10:05:21PM +0300, Aaro Koskinen wrote:
> > On Wed, Aug 07, 2013 at 02:26:09AM +0300, Aaro Koskinen wrote:
> > > I get the following on Amstrad Delta (E3) when booting unpatched
> > > v3.11-rc4. The kernel config is after the crash.
> > 
> > This is regression introduced between -rc3 and -rc4. According to the
> > git bisect, the first bad commit is:
> > 
> > 	commit e39e3f3ebfef03450cf7bfa7a974a8c61f7980c8
> > 	Author: Russell King <rmk+kernel@arm.linux.org.uk>
> > 	Date:   Tue Jul 9 01:03:17 2013 +0100
> > 
> > 	    ARM: update FIQ support for relocation of vectors
> > 
> > If I revert the commit from v3.11-rc4, the board boots fine.
> > 
> > A.
> > 
> > > Uncompressing Linux... done, booting the kernel.
> > > [    0.000000] Booting Linux on physical CPU 0x0
> > > [    0.000000] Initializing cgroup subsys cpu
> > > [    0.000000] Linux version 3.11.0-rc4-e3 (aaro@blackmetal) (gcc version 4.7.3 (GCC) ) #4 PREEMPT Wed Aug 7 02:03:27 EEST 2013
> > > [    0.000000] bootconsole [earlycon0] enabled
> > > [    0.000000] OMAP1510
> > > [    0.000000]  revision 2 handled as 15xx id: bc058c9b93111a16
> 
> Hmm, so ARM925T which is VIVT...
> 
> > > [    0.238868] Installing fiq handler from c001b110, length 0x164
> > > [    0.245750] Unable to handle kernel paging request at virtual address ffff1224
> 
> In one way, that's good news - that's public independent proof that the
> code associated with the vectors is well protected from kernel writes. :)
> 
> Can you try this (untested) patch please?  We need to write into this
> page via the page's alias, rather than directly to it.  Thanks.

That seems to work - the board boots fine to shell as before. Feel free
to add Reported-by: or Tested-by: from me to the patch.

Thanks,

A.

>  arch/arm/kernel/fiq.c |    8 ++------
>  1 files changed, 2 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/arm/kernel/fiq.c b/arch/arm/kernel/fiq.c
> index 25442f4..fc79202 100644
> --- a/arch/arm/kernel/fiq.c
> +++ b/arch/arm/kernel/fiq.c
> @@ -84,17 +84,13 @@ int show_fiq_list(struct seq_file *p, int prec)
>  
>  void set_fiq_handler(void *start, unsigned int length)
>  {
> -#if defined(CONFIG_CPU_USE_DOMAINS)
> -	void *base = (void *)0xffff0000;
> -#else
>  	void *base = vectors_page;
> -#endif
>  	unsigned offset = FIQ_OFFSET;
>  
>  	memcpy(base + offset, start, length);
> +	if (!cache_is_vipt_nonaliasing())
> +		flush_icache_range(base + offset, offset + length);
>  	flush_icache_range(0xffff0000 + offset, 0xffff0000 + offset + length);
> -	if (!vectors_high())
> -		flush_icache_range(offset, offset + length);
>  }
>  
>  int claim_fiq(struct fiq_handler *f)
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Russell King - ARM Linux Aug. 8, 2013, 11:02 a.m. UTC | #2
On Thu, Aug 08, 2013 at 03:05:03AM +0300, Aaro Koskinen wrote:
> That seems to work - the board boots fine to shell as before. Feel free
> to add Reported-by: or Tested-by: from me to the patch.

Thanks, done.  Should be in linux-next by tomorrow.
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/arch/arm/kernel/fiq.c b/arch/arm/kernel/fiq.c
index 25442f4..fc79202 100644
--- a/arch/arm/kernel/fiq.c
+++ b/arch/arm/kernel/fiq.c
@@ -84,17 +84,13 @@  int show_fiq_list(struct seq_file *p, int prec)
 
 void set_fiq_handler(void *start, unsigned int length)
 {
-#if defined(CONFIG_CPU_USE_DOMAINS)
-	void *base = (void *)0xffff0000;
-#else
 	void *base = vectors_page;
-#endif
 	unsigned offset = FIQ_OFFSET;
 
 	memcpy(base + offset, start, length);
+	if (!cache_is_vipt_nonaliasing())
+		flush_icache_range(base + offset, offset + length);
 	flush_icache_range(0xffff0000 + offset, 0xffff0000 + offset + length);
-	if (!vectors_high())
-		flush_icache_range(offset, offset + length);
 }
 
 int claim_fiq(struct fiq_handler *f)