michaelh/zephyr
1
0
Fork 0
Code Releases Activity

intel_adsp: meteorlake: Initialize stack flush pointer SR

Browse source 
The simulator seems to drop garbage addresses (somewhere in the ROM it
looks like) into this SR at arbitrary times.  I don't know if this is
a hardware exception handler that we can't turn off, or a simulator
bug, or what.  But our code that assumes it will be cleared to zero or
valid is breaking.  Set it every time in every context switch for now
pending someone figuring out what's going wrong.

Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This commit is contained in:
Andy Ross 2022-01-07 13:44:52 -08:00 • committed by Anas Nashif
commit 910c96b7d8
1 changed files with 9 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

9
arch/xtensa/include/kernel_arch_func.h
View file

@ -34,6 +34,11 @@ static ALWAYS_INLINE void arch_kernel_init(void)
* regions due to boot firmware * regions due to boot firmware
*/ */
z_xtensa_cache_flush_inv_all(); z_xtensa_cache_flush_inv_all();
/* Our cache top stash location might have junk in it from a
* pre-boot environment. Must be zero or valid!
*/
WSR(ZSR_FLUSH_STR, 0);
#endif #endif
cpu0->nested = 0; cpu0->nested = 0;
@ -75,6 +80,10 @@ static ALWAYS_INLINE void arch_cohere_stacks(struct k_thread *old_thread,
size_t nsz = new_thread->stack_info.size; size_t nsz = new_thread->stack_info.size;
size_t nsp = (size_t) new_thread->switch_handle; size_t nsp = (size_t) new_thread->switch_handle;
int zero = 0;
__asm__ volatile("wsr %0, " ZSR_FLUSH_STR :: "r"(zero));
if (old_switch_handle != NULL) { if (old_switch_handle != NULL) {
int32_t a0save; int32_t a0save;