From: Zhen Lei thunder.leizhen@huawei.com

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8Z6DC CVE: NA

-------------------------------------------------------------------------

Ensure that each core exclusively occupies an ECMDQ and all of them are enabled during initialization. During this initialization process, any errors will result in a fallback to using normal CMDQ.

When GERROR is triggered by ECMDQ, all ECMDQs need to be traversed: the ECMDQs with errors will be processed and the ECMDQs without errors will be skipped directly.

Compared with register SMMU_CMDQ_PROD, register SMMU_ECMDQ_PROD has one more 'EN' bit and one more 'ERRACK' bit. Therefore, an extra member 'ecmdq_prod' is added to record the values of these two bits. Each time register SMMU_ECMDQ_PROD is updated, the value of 'ecmdq_prod' is ORed. After the error indicated by SMMU_GERROR.CMDQP_ERR is fixed, the 'ERRACK' bit needs to be toggled to resume the corresponding ECMDQ. Therefore, a rwlock is used to protect the write operation to bit 'ERRACK' during error handling and the read operation to bit 'ERRACK' during command insertion.

Signed-off-by: Zhen Lei thunder.leizhen@huawei.com Signed-off-by: Zhang Zekun zhangzekun11@huawei.com --- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 210 +++++++++++++++++++- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 36 ++++ 2 files changed, 245 insertions(+), 1 deletion(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c index 25b5439ca6cf..f756184bf53d 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c @@ -400,6 +400,14 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)

static struct arm_smmu_cmdq *arm_smmu_get_cmdq(struct arm_smmu_device *smmu) { + if (smmu->ecmdq_enabled) { + struct arm_smmu_ecmdq *ecmdq; + + ecmdq = *this_cpu_ptr(smmu->ecmdq); + + return &ecmdq->cmdq; + } + return &smmu->cmdq; }

@@ -482,6 +490,38 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu) __arm_smmu_cmdq_skip_err(smmu, &smmu->cmdq.q); }

+static void arm_smmu_ecmdq_skip_err(struct arm_smmu_device *smmu) +{ + int i; + u32 prod, cons; + struct arm_smmu_queue *q; + struct arm_smmu_ecmdq *ecmdq; + + for (i = 0; i < smmu->nr_ecmdq; i++) { + unsigned long flags; + + ecmdq = *per_cpu_ptr(smmu->ecmdq, i); + q = &ecmdq->cmdq.q; + + prod = readl_relaxed(q->prod_reg); + cons = readl_relaxed(q->cons_reg); + if (((prod ^ cons) & ECMDQ_CONS_ERR) == 0) + continue; + + __arm_smmu_cmdq_skip_err(smmu, q); + + write_lock_irqsave(&q->ecmdq_lock, flags); + q->ecmdq_prod &= ~ECMDQ_PROD_ERRACK; + q->ecmdq_prod |= cons & ECMDQ_CONS_ERR; + + prod = readl_relaxed(q->prod_reg); + prod &= ~ECMDQ_PROD_ERRACK; + prod |= cons & ECMDQ_CONS_ERR; + writel(prod, q->prod_reg); + write_unlock_irqrestore(&q->ecmdq_lock, flags); + } +} + /* * Command queue locking. * This is a form of bastardised rwlock with the following major changes: @@ -878,7 +918,13 @@ static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu, * d. Advance the hardware prod pointer * Control dependency ordering from the entries becoming valid. */ - writel_relaxed(prod, cmdq->q.prod_reg); + if (smmu->ecmdq_enabled) { + read_lock(&cmdq->q.ecmdq_lock); + writel_relaxed(prod | cmdq->q.ecmdq_prod, cmdq->q.prod_reg); + read_unlock(&cmdq->q.ecmdq_lock); + } else { + writel_relaxed(prod, cmdq->q.prod_reg); + }

/* * e. Tell the next owner we're done @@ -1757,6 +1803,9 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev) if (active & GERROR_CMDQ_ERR) arm_smmu_cmdq_skip_err(smmu);

+ if (active & GERROR_CMDQP_ERR) + arm_smmu_ecmdq_skip_err(smmu); + writel(gerror, smmu->base + ARM_SMMU_GERRORN); return IRQ_HANDLED; } @@ -3027,6 +3076,20 @@ static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu) return 0; }

+static int arm_smmu_ecmdq_init(struct arm_smmu_cmdq *cmdq) +{ + unsigned int nents = 1 << cmdq->q.llq.max_n_shift; + + atomic_set(&cmdq->owner_prod, 0); + atomic_set(&cmdq->lock, 0); + + cmdq->valid_map = (atomic_long_t *)bitmap_zalloc(nents, GFP_KERNEL); + if (!cmdq->valid_map) + return -ENOMEM; + + return 0; +} + static int arm_smmu_init_queues(struct arm_smmu_device *smmu) { int ret; @@ -3474,6 +3537,7 @@ static int arm_smmu_device_disable(struct arm_smmu_device *smmu)

static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool resume) { + int i; int ret; u32 reg, enables; struct arm_smmu_cmdq_ent cmd; @@ -3518,6 +3582,28 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool resume) writel_relaxed(smmu->cmdq.q.llq.prod, smmu->base + ARM_SMMU_CMDQ_PROD); writel_relaxed(smmu->cmdq.q.llq.cons, smmu->base + ARM_SMMU_CMDQ_CONS);

+ for (i = 0; i < smmu->nr_ecmdq; i++) { + struct arm_smmu_ecmdq *ecmdq; + struct arm_smmu_queue *q; + + ecmdq = *per_cpu_ptr(smmu->ecmdq, i); + q = &ecmdq->cmdq.q; + + writeq_relaxed(q->q_base, ecmdq->base + ARM_SMMU_ECMDQ_BASE); + writel_relaxed(q->llq.prod, ecmdq->base + ARM_SMMU_ECMDQ_PROD); + writel_relaxed(q->llq.cons, ecmdq->base + ARM_SMMU_ECMDQ_CONS); + + /* enable ecmdq */ + writel(ECMDQ_PROD_EN, q->prod_reg); + ret = readl_relaxed_poll_timeout(q->cons_reg, reg, reg & ECMDQ_CONS_ENACK, + 1, ARM_SMMU_POLL_TIMEOUT_US); + if (ret) { + dev_err(smmu->dev, "ecmdq[%d] enable failed\n", i); + smmu->ecmdq_enabled = 0; + break; + } + } + enables = CR0_CMDQEN; ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0, ARM_SMMU_CR0ACK); @@ -3607,6 +3693,115 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool resume) return 0; }

+static int arm_smmu_ecmdq_layout(struct arm_smmu_device *smmu) +{ + int cpu; + struct arm_smmu_ecmdq *ecmdq; + + if (num_possible_cpus() <= smmu->nr_ecmdq) { + ecmdq = devm_alloc_percpu(smmu->dev, *ecmdq); + if (!ecmdq) + return -ENOMEM; + + for_each_possible_cpu(cpu) + *per_cpu_ptr(smmu->ecmdq, cpu) = per_cpu_ptr(ecmdq, cpu); + + /* A core requires at most one ECMDQ */ + smmu->nr_ecmdq = num_possible_cpus(); + + return 0; + } + + return -ENOSPC; +} + +static int arm_smmu_ecmdq_probe(struct arm_smmu_device *smmu) +{ + int ret, cpu; + u32 i, nump, numq, gap; + u32 reg, shift_increment; + u64 addr, smmu_dma_base; + void __iomem *cp_regs, *cp_base; + + /* IDR6 */ + reg = readl_relaxed(smmu->base + ARM_SMMU_IDR6); + nump = 1 << FIELD_GET(IDR6_LOG2NUMP, reg); + numq = 1 << FIELD_GET(IDR6_LOG2NUMQ, reg); + smmu->nr_ecmdq = nump * numq; + gap = ECMDQ_CP_RRESET_SIZE >> FIELD_GET(IDR6_LOG2NUMQ, reg); + + smmu_dma_base = (vmalloc_to_pfn(smmu->base) << PAGE_SHIFT); + cp_regs = ioremap(smmu_dma_base + ARM_SMMU_ECMDQ_CP_BASE, PAGE_SIZE); + if (!cp_regs) + return -ENOMEM; + + for (i = 0; i < nump; i++) { + u64 val, pre_addr; + + val = readq_relaxed(cp_regs + 32 * i); + if (!(val & ECMDQ_CP_PRESET)) { + iounmap(cp_regs); + dev_err(smmu->dev, "ecmdq control page %u is memory mode\n", i); + return -EFAULT; + } + + if (i && ((val & ECMDQ_CP_ADDR) != (pre_addr + ECMDQ_CP_RRESET_SIZE))) { + iounmap(cp_regs); + dev_err(smmu->dev, "ecmdq_cp memory region is not contiguous\n"); + return -EFAULT; + } + + pre_addr = val & ECMDQ_CP_ADDR; + } + + addr = readl_relaxed(cp_regs) & ECMDQ_CP_ADDR; + iounmap(cp_regs); + + cp_base = devm_ioremap(smmu->dev, smmu_dma_base + addr, ECMDQ_CP_RRESET_SIZE * nump); + if (!cp_base) + return -ENOMEM; + + smmu->ecmdq = devm_alloc_percpu(smmu->dev, struct arm_smmu_ecmdq *); + if (!smmu->ecmdq) + return -ENOMEM; + + ret = arm_smmu_ecmdq_layout(smmu); + if (ret) + return ret; + + shift_increment = order_base_2(num_possible_cpus() / smmu->nr_ecmdq); + + addr = 0; + for_each_possible_cpu(cpu) { + struct arm_smmu_ecmdq *ecmdq; + struct arm_smmu_queue *q; + + ecmdq = *per_cpu_ptr(smmu->ecmdq, cpu); + ecmdq->base = cp_base + addr; + + q = &ecmdq->cmdq.q; + + q->llq.max_n_shift = ECMDQ_MAX_SZ_SHIFT + shift_increment; + ret = arm_smmu_init_one_queue(smmu, q, ecmdq->base, ARM_SMMU_ECMDQ_PROD, + ARM_SMMU_ECMDQ_CONS, CMDQ_ENT_DWORDS, "ecmdq"); + if (ret) + return ret; + + q->ecmdq_prod = ECMDQ_PROD_EN; + rwlock_init(&q->ecmdq_lock); + + ret = arm_smmu_ecmdq_init(&ecmdq->cmdq); + if (ret) { + dev_err(smmu->dev, "ecmdq[%d] init failed\n", i); + return ret; + } + + addr += gap; + } + + return 0; +} + #define IIDR_IMPLEMENTER_ARM 0x43b #define IIDR_PRODUCTID_ARM_MMU_600 0x483 #define IIDR_PRODUCTID_ARM_MMU_700 0x487 @@ -3755,6 +3950,9 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu) return -ENXIO; }

+ if (reg & IDR1_ECMDQ) + smmu->features |= ARM_SMMU_FEAT_ECMDQ; + /* Queue sizes, capped to ensure natural alignment */ smmu->cmdq.q.llq.max_n_shift = min_t(u32, CMDQ_MAX_SZ_SHIFT, FIELD_GET(IDR1_CMDQS, reg)); @@ -3862,6 +4060,16 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)

dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n", smmu->ias, smmu->oas, smmu->features); + + if (smmu->features & ARM_SMMU_FEAT_ECMDQ) { + int err; + + err = arm_smmu_ecmdq_probe(smmu); + if (err) { + dev_err(smmu->dev, "suppress ecmdq feature, errno=%d\n", err); + smmu->ecmdq_enabled = 0; + } + } return 0; }

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h index 269b6fa705d4..ed45909bc372 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h @@ -41,6 +41,7 @@ #define IDR0_S2P (1 << 0)

#define ARM_SMMU_IDR1 0x4 +#define IDR1_ECMDQ (1 << 31) #define IDR1_TABLES_PRESET (1 << 30) #define IDR1_QUEUES_PRESET (1 << 29) #define IDR1_REL (1 << 28) @@ -113,6 +114,7 @@ #define ARM_SMMU_IRQ_CTRLACK 0x54

#define ARM_SMMU_GERROR 0x60 +#define GERROR_CMDQP_ERR (1 << 9) #define GERROR_SFM_ERR (1 << 8) #define GERROR_MSI_GERROR_ABT_ERR (1 << 7) #define GERROR_MSI_PRIQ_ABT_ERR (1 << 6) @@ -129,6 +131,26 @@ #define ARM_SMMU_GERROR_IRQ_CFG1 0x70 #define ARM_SMMU_GERROR_IRQ_CFG2 0x74

+#define ARM_SMMU_IDR6 0x190 +#define IDR6_LOG2NUMP GENMASK(27, 24) +#define IDR6_LOG2NUMQ GENMASK(19, 16) +#define IDR6_BA_DOORBELLS GENMASK(9, 0) + +#define ARM_SMMU_ECMDQ_BASE 0x00 +#define ARM_SMMU_ECMDQ_PROD 0x08 +#define ARM_SMMU_ECMDQ_CONS 0x0c +#define ECMDQ_MAX_SZ_SHIFT 8 +#define ECMDQ_PROD_EN (1 << 31) +#define ECMDQ_CONS_ENACK (1 << 31) +#define ECMDQ_CONS_ERR (1 << 23) +#define ECMDQ_PROD_ERRACK (1 << 23) + +#define ARM_SMMU_ECMDQ_CP_BASE 0x4000 +#define ECMDQ_CP_ADDR GENMASK_ULL(51, 12) +#define ECMDQ_CP_CMDQGS GENMASK_ULL(2, 1) +#define ECMDQ_CP_PRESET (1UL << 0) +#define ECMDQ_CP_RRESET_SIZE 0x10000 + #define ARM_SMMU_STRTAB_BASE 0x80 #define STRTAB_BASE_RA (1UL << 62) #define STRTAB_BASE_ADDR_MASK GENMASK_ULL(51, 6) @@ -527,6 +549,8 @@ struct arm_smmu_ll_queue { struct arm_smmu_queue { struct arm_smmu_ll_queue llq; int irq; /* Wired interrupt */ + u32 ecmdq_prod; + rwlock_t ecmdq_lock;

__le64 *base; dma_addr_t base_dma; @@ -552,6 +576,11 @@ struct arm_smmu_cmdq { atomic_t lock; };

+struct arm_smmu_ecmdq { + struct arm_smmu_cmdq cmdq; + void __iomem *base; +}; + struct arm_smmu_cmdq_batch { u64 cmds[CMDQ_BATCH_ENTRIES * CMDQ_ENT_DWORDS]; int num; @@ -646,6 +675,7 @@ struct arm_smmu_device { #define ARM_SMMU_FEAT_SVA (1 << 17) #define ARM_SMMU_FEAT_E2H (1 << 18) #define ARM_SMMU_FEAT_NESTING (1 << 19) +#define ARM_SMMU_FEAT_ECMDQ (1 << 20) u32 features;

#define ARM_SMMU_OPT_SKIP_PREFETCH (1 << 0) @@ -654,6 +684,12 @@ struct arm_smmu_device { #define ARM_SMMU_OPT_CMDQ_FORCE_SYNC (1 << 3) u32 options;

+ union { + u32 nr_ecmdq; + u32 ecmdq_enabled; + }; + struct arm_smmu_ecmdq *__percpu *ecmdq; + struct arm_smmu_cmdq cmdq; struct arm_smmu_evtq evtq; struct arm_smmu_priq priq;

From: Zhen Lei thunder.leizhen@huawei.com

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8Z6DC CVE: NA

-------------------------------------------------------------------------

When a core can exclusively own an ECMDQ, competition with other cores does not need to be considered during command insertion. Therefore, we can delete the part of arm_smmu_cmdq_issue_cmdlist() that deals with multi-core contention and generate a more efficient ECMDQ-specific function arm_smmu_ecmdq_issue_cmdlist().

Signed-off-by: Zhen Lei thunder.leizhen@huawei.com Signed-off-by: Zhang Zekun zhangzekun11@huawei.com --- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 85 +++++++++++++++++++++ drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 1 + 2 files changed, 86 insertions(+)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c index 40c9b8cf6a03..6055165f23f9 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c @@ -830,6 +830,87 @@ static void arm_smmu_cmdq_write_entries(struct arm_smmu_cmdq *cmdq, u64 *cmds, } }

+/* + * The function is used when the current core exclusively occupies an ECMDQ. + * This is a reduced version of arm_smmu_cmdq_issue_cmdlist(), which eliminates + * a lot of unnecessary inter-core competition considerations. + */ +static int arm_smmu_ecmdq_issue_cmdlist(struct arm_smmu_device *smmu, + struct arm_smmu_cmdq *cmdq, + u64 *cmds, int n, bool sync) +{ + u32 prod; + unsigned long flags; + struct arm_smmu_ll_queue llq = { + .max_n_shift = cmdq->q.llq.max_n_shift, + }, head; + int ret = 0; + + /* 1. Allocate some space in the queue */ + local_irq_save(flags); + llq.val = READ_ONCE(cmdq->q.llq.val); + do { + u64 old; + + while (!queue_has_space(&llq, n + sync)) { + local_irq_restore(flags); + if (arm_smmu_cmdq_poll_until_not_full(smmu, &llq)) + dev_err_ratelimited(smmu->dev, "ECMDQ timeout\n"); + local_irq_save(flags); + } + + head.cons = llq.cons; + head.prod = queue_inc_prod_n(&llq, n + sync); + + old = cmpxchg_relaxed(&cmdq->q.llq.val, llq.val, head.val); + if (old == llq.val) + break; + + llq.val = old; + } while (1); + + /* 2. Write our commands into the queue */ + arm_smmu_cmdq_write_entries(cmdq, cmds, llq.prod, n); + if (sync) { + u64 cmd_sync[CMDQ_ENT_DWORDS]; + + prod = queue_inc_prod_n(&llq, n); + arm_smmu_cmdq_build_sync_cmd(cmd_sync, smmu, &cmdq->q, prod); + queue_write(Q_ENT(&cmdq->q, prod), cmd_sync, CMDQ_ENT_DWORDS); + } + + /* 3. Ensuring commands are visible first */ + dma_wmb(); + + /* 4. Advance the hardware prod pointer */ + read_lock(&cmdq->q.ecmdq_lock); + writel_relaxed(head.prod | cmdq->q.ecmdq_prod, cmdq->q.prod_reg); + read_unlock(&cmdq->q.ecmdq_lock); + + /* 5. If we are inserting a CMD_SYNC, we must wait for it to complete */ + if (sync) { + llq.prod = queue_inc_prod_n(&llq, n); + ret = arm_smmu_cmdq_poll_until_sync(smmu, &llq); + if (ret) { + dev_err_ratelimited(smmu->dev, + "CMD_SYNC timeout at 0x%08x [hwprod 0x%08x, hwcons 0x%08x]\n", + llq.prod, + readl_relaxed(cmdq->q.prod_reg), + readl_relaxed(cmdq->q.cons_reg)); + } + + /* + * Update cmdq->q.llq.cons, to improve the success rate of + * queue_has_space() when some new commands are inserted next + * time. + */ + WRITE_ONCE(cmdq->q.llq.cons, llq.cons); + } + + local_irq_restore(flags); + return ret; +} + /* * This is the actual insertion function, and provides the following * ordering guarantees to callers: @@ -857,6 +938,9 @@ static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu, struct arm_smmu_ll_queue llq, head; int ret = 0;

+ if (!cmdq->shared) + return arm_smmu_ecmdq_issue_cmdlist(smmu, cmdq, cmds, n, sync); + llq.max_n_shift = cmdq->q.llq.max_n_shift;

/* 1. Allocate some space in the queue */ @@ -3090,6 +3174,7 @@ static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu) struct arm_smmu_cmdq *cmdq = &smmu->cmdq; unsigned int nents = 1 << cmdq->q.llq.max_n_shift;

+ cmdq->shared = 1; atomic_set(&cmdq->owner_prod, 0); atomic_set(&cmdq->lock, 0);

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h index ed45909bc372..209795fce12e 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h @@ -574,6 +574,7 @@ struct arm_smmu_cmdq { atomic_long_t *valid_map; atomic_t owner_prod; atomic_t lock; + int shared; };

struct arm_smmu_ecmdq {

From: Zhen Lei thunder.leizhen@huawei.com

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8Z6DC

--------------------------------

Add boot option arm_smmu_v3.disable_ecmdq to support disabling ECMDQ during startup. For example, add arm_smmu_v3.disable_ecmdq=1 in command line to disable ECMDQ and use normal CMDQ.

Signed-off-by: Zhen Lei thunder.leizhen@huawei.com Signed-off-by: Zhang Zekun zhangzekun11@huawei.com --- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c index 6fc1204ea6f1..11ab5fae24eb 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c @@ -41,6 +41,10 @@ module_param(disable_msipolling, bool, 0444); MODULE_PARM_DESC(disable_msipolling, "Disable MSI-based polling for CMD_SYNC completion.");

+static bool disable_ecmdq; +module_param(disable_ecmdq, bool, 0444); +MODULE_PARM_DESC(disable_ecmdq, "Disable the use of ECMDQs"); + #ifdef CONFIG_SMMU_BYPASS_DEV struct smmu_bypass_device { unsigned short vendor; @@ -4211,7 +4215,7 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu) dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n", smmu->ias, smmu->oas, smmu->features);

- if (smmu->features & ARM_SMMU_FEAT_ECMDQ) { + if (smmu->features & ARM_SMMU_FEAT_ECMDQ && !disable_ecmdq) { int err;

err = arm_smmu_ecmdq_probe(smmu);