提交 33573c0e 编写于 作者: P Paul Mundt

sh: Fix occasional flush_cache_4096() stack corruption.

IRQs disabling in flush_cache_4096 for cache purge. Under certain
workloads we would get an IRQ in the middle of a purge operation,
and the cachelines would remain in an inconsistent state, leading
to occasional stack corruption.
Signed-off-by: NTakeo Takahashi <takahashi.takeo@renesas.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>
上级 f3c25758
......@@ -221,22 +221,20 @@ void flush_cache_sigtramp(unsigned long addr)
static inline void flush_cache_4096(unsigned long start,
unsigned long phys)
{
unsigned long flags, exec_offset = 0;
/*
* All types of SH-4 require PC to be in P2 to operate on the I-cache.
* Some types of SH-4 require PC to be in P2 to operate on the D-cache.
*/
if ((cpu_data->flags & CPU_HAS_P2_FLUSH_BUG) ||
(start < CACHE_OC_ADDRESS_ARRAY)) {
unsigned long flags;
local_irq_save(flags);
__flush_cache_4096(start | SH_CACHE_ASSOC,
P1SEGADDR(phys), 0x20000000);
local_irq_restore(flags);
} else {
__flush_cache_4096(start | SH_CACHE_ASSOC,
P1SEGADDR(phys), 0);
}
(start < CACHE_OC_ADDRESS_ARRAY))
exec_offset = 0x20000000;
local_irq_save(flags);
__flush_cache_4096(start | SH_CACHE_ASSOC,
P1SEGADDR(phys), exec_offset);
local_irq_restore(flags);
}
/*
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册