This patch adds workaround for ARM errata 798870 which says "If back-to-back speculative cache line fills (fill A and fill B) are issued from the L1 data cache of a CPU to the L2 cache, the second request (fill B) is then cancelled, and the second request would have detected a hazard against a recent write or eviction (write B) to the same cache line as fill B then the L2 logic might deadlock."
Signed-off-by: Kimoon Kim kimoon.kim@samsung.com Signed-off-by: Akshay Saraswat akshay.s@samsung.com --- arch/arm/cpu/armv7/exynos/lowlevel_init.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)
diff --git a/arch/arm/cpu/armv7/exynos/lowlevel_init.c b/arch/arm/cpu/armv7/exynos/lowlevel_init.c index 43c957b..7073c5c 100644 --- a/arch/arm/cpu/armv7/exynos/lowlevel_init.c +++ b/arch/arm/cpu/armv7/exynos/lowlevel_init.c @@ -45,6 +45,28 @@ enum {
#ifdef CONFIG_EXYNOS5420 /* + * Set L2ACTLR[7] to reissue any memory transaction in the L2 that has been + * stalled for 1024 cycles to verify that its hazard condition still exists. + */ +void set_l2cache(void) +{ + uint32_t val; + + /* Read MIDR for Primary Part Number*/ + mrc_midr(val); + val = (val >> 4); + val &= 0xf; + + /* L2ACTLR[7]: Enable hazard detect timeout for A15 */ + if (val == 0xf) { + mrc_l2_aux_ctlr(val); + val |= (1 << 7); + mcr_l2_aux_ctlr(val); + mrc_l2_ctlr(val); + } +} + +/* * Pointer to this function is stored in iRam which is used * for jump and power down of a specific core. */