From: Zhen Lei Date: Mon, 31 May 2021 20:35:53 +0800 Subject: iommu/arm-smmu-v3: Decrease the queue size of evtq and priq Git-commit: f115f3c0d5d846f69b2bc2d86653117f305b6066 Patch-mainline: v5.14-rc1 References: bsc#1189210 Commit d25f6ead162e ("iommu/arm-smmu-v3: Increase maximum size of queues") expands the cmdq queue size to improve the success rate of concurrent command queue space allocation by multiple cores. However, this extension does not apply to evtq and priq, because for both of them, the SMMU driver is the consumer. Instead, memory resources are wasted. Therefore, the queue size of evtq and priq is restored to the original setting, one page. Fixes: d25f6ead162e ("iommu/arm-smmu-v3: Increase maximum size of queues") Signed-off-by: Zhen Lei Link: https://lore.kernel.org/r/20210531123553.9602-1-thunder.leizhen@huawei.com Signed-off-by: Will Deacon Acked-by: Joerg Roedel --- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) --- a/drivers/iommu/arm-smmu-v3.c +++ b/drivers/iommu/arm-smmu-v3.c @@ -200,6 +200,7 @@ #else #define Q_MAX_SZ_SHIFT (PAGE_SHIFT + MAX_ORDER - 1) #endif +#define Q_MIN_SZ_SHIFT (PAGE_SHIFT) /* * Stream table. @@ -348,14 +349,14 @@ /* Event queue */ #define EVTQ_ENT_SZ_SHIFT 5 #define EVTQ_ENT_DWORDS ((1 << EVTQ_ENT_SZ_SHIFT) >> 3) -#define EVTQ_MAX_SZ_SHIFT (Q_MAX_SZ_SHIFT - EVTQ_ENT_SZ_SHIFT) +#define EVTQ_MAX_SZ_SHIFT (Q_MIN_SZ_SHIFT - EVTQ_ENT_SZ_SHIFT) #define EVTQ_0_ID GENMASK_ULL(7, 0) /* PRI queue */ #define PRIQ_ENT_SZ_SHIFT 4 #define PRIQ_ENT_DWORDS ((1 << PRIQ_ENT_SZ_SHIFT) >> 3) -#define PRIQ_MAX_SZ_SHIFT (Q_MAX_SZ_SHIFT - PRIQ_ENT_SZ_SHIFT) +#define PRIQ_MAX_SZ_SHIFT (Q_MIN_SZ_SHIFT - PRIQ_ENT_SZ_SHIFT) #define PRIQ_0_SID GENMASK_ULL(31, 0) #define PRIQ_0_SSID GENMASK_ULL(51, 32)