Albert,
That's not a correct characterization of the bug.
The incoherent cache lines are from before the relocation stage. If U-Boot is relocating from RAM, and later copies the OS there without invalidating those lines, then that's a bug in U-Boot.
Michael
On Mon, Jul 29, 2013 at 3:19 AM, Albert ARIBAUD albert.u.boot@aribaud.net wrote:
(although this patch is more than two *years* old, it never got properly answered to. I am doing so here to make sure future readers know why it was not applied and won't be.)
Hi Michael,
On Thu, 17 Mar 2011 15:46:55 -0400, Michael Spang mspang@csclub.uwaterloo.ca wrote:
If U-Boot is loaded from RAM and the OS is loaded into an overlapping region, the instruction cache is not coherent when that OS is started. We must therefore invalidate the instruction cache in addition to cleaning the data cache.
Signed-off-by: Michael Spang mspang@csclub.uwaterloo.ca
arch/arm/lib/cache.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/arch/arm/lib/cache.c b/arch/arm/lib/cache.c index 30686fe..047786a 100644 --- a/arch/arm/lib/cache.c +++ b/arch/arm/lib/cache.c @@ -37,6 +37,8 @@ void flush_cache (unsigned long dummy1, unsigned long dummy2) asm("0: mrc p15, 0, r15, c7, c10, 3\n\t" "bne 0b\n" : : : "memory"); /* disable write buffer as well (page 2-22) */ asm("mcr p15, 0, %0, c7, c10, 4" : : "r" (0));
/* invalidate icache for coherence with cleaned dcache */asm("mcr p15, 0, %0, c7, c5, 0" : : "r" (0));#endif #ifdef CONFIG_OMAP34XX void v7_flush_cache_all(void);
This patch has obviously not been applied, and won't be, because of two reasons: i) overwriting part of U-Boot when loading the kernel is a bug which should not be papered over, and ii) if we had to do this anyway, the right place to do it would have been where the issue might occur, that is, in the OS boot sequence, not in the cache handling functions.
Apologies for not properly answering in due time.
Amicalement,
Albert.