[PATCH] habanalabs: move asic property to correct structure

[PATCH] habanalabs: move asic property to correct structure

[Oded Gabbay]

Whether an ASIC has MMU towards its DRAM is an ASIC property, so
move it to the asic fixed properties structure.

Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/debugfs.c    | 2 +-
 drivers/misc/habanalabs/common/habanalabs.h | 4 ++--
 drivers/misc/habanalabs/common/memory.c     | 6 +++---
 drivers/misc/habanalabs/common/mmu_v1.c     | 4 ++--
 drivers/misc/habanalabs/gaudi/gaudi.c       | 3 +--
 drivers/misc/habanalabs/goya/goya.c         | 2 +-
 6 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/misc/habanalabs/common/debugfs.c b/drivers/misc/habanalabs/common/debugfs.c
index 104b9686e57b..3d9bd86312e2 100644
--- a/drivers/misc/habanalabs/common/debugfs.c
+++ b/drivers/misc/habanalabs/common/debugfs.c
@@ -598,7 +598,7 @@ static bool hl_is_device_va(struct hl_device *hdev, u64 addr)
 	if (!hdev->mmu_enable)
 		goto out;
 
-	if (hdev->dram_supports_virtual_memory &&
+	if (prop->dram_supports_virtual_memory &&
 		(addr >= prop->dmmu.start_addr && addr < prop->dmmu.end_addr))
 		return true;
 
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 822f90087e04..a1d82de60ef6 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -411,6 +411,7 @@ struct hl_mmu_properties {
  *                        false otherwise
  * @fw_security_status_valid: security status bits are valid and can be fetched
  *                            from BOOT_DEV_STS0
+ * @dram_supports_virtual_memory: is there an MMU towards the DRAM
  */
 struct asic_fixed_properties {
 	struct hw_queue_properties	*hw_queues_props;
@@ -467,6 +468,7 @@ struct asic_fixed_properties {
 	u8				completion_queues_count;
 	u8				fw_security_disabled;
 	u8				fw_security_status_valid;
+	u8				dram_supports_virtual_memory;
 };
 
 /**
@@ -1772,7 +1774,6 @@ struct hl_mmu_funcs {
  * @heartbeat: is heartbeat sanity check towards CPU-CP enabled.
  * @reset_on_lockup: true if a reset should be done in case of stuck CS, false
  *                   otherwise.
- * @dram_supports_virtual_memory: is MMU enabled towards DRAM.
  * @dram_default_page_mapping: is DRAM default page mapping enabled.
  * @memory_scrub: true to perform device memory scrub in various locations,
  *                such as context-switch, context close, page free, etc.
@@ -1886,7 +1887,6 @@ struct hl_device {
 	u8				hard_reset_pending;
 	u8				heartbeat;
 	u8				reset_on_lockup;
-	u8				dram_supports_virtual_memory;
 	u8				dram_default_page_mapping;
 	u8				memory_scrub;
 	u8				pmmu_huge_range;
diff --git a/drivers/misc/habanalabs/common/memory.c b/drivers/misc/habanalabs/common/memory.c
index 01e93e83d3af..f27ca80d3c3c 100644
--- a/drivers/misc/habanalabs/common/memory.c
+++ b/drivers/misc/habanalabs/common/memory.c
@@ -1351,7 +1351,7 @@ int hl_mem_ioctl(struct hl_fpriv *hpriv, void *data)
 		 * the user's input, the driver can't ensure the validity
 		 * of this accounting.
 		 */
-		if (!hdev->dram_supports_virtual_memory) {
+		if (!hdev->asic_prop.dram_supports_virtual_memory) {
 			atomic64_add(args->in.alloc.mem_size,
 					&ctx->dram_phys_mem);
 			atomic64_add(args->in.alloc.mem_size,
@@ -1380,7 +1380,7 @@ int hl_mem_ioctl(struct hl_fpriv *hpriv, void *data)
 		 * the user's input, the driver can't ensure the validity
 		 * of this accounting.
 		 */
-		if (!hdev->dram_supports_virtual_memory) {
+		if (!hdev->asic_prop.dram_supports_virtual_memory) {
 			atomic64_sub(args->in.alloc.mem_size,
 					&ctx->dram_phys_mem);
 			atomic64_sub(args->in.alloc.mem_size,
@@ -1915,7 +1915,7 @@ void hl_vm_ctx_fini(struct hl_ctx *ctx)
 	 * because the user notifies us on allocations. If the user is no more,
 	 * all DRAM is available
 	 */
-	if (!ctx->hdev->dram_supports_virtual_memory)
+	if (!ctx->hdev->asic_prop.dram_supports_virtual_memory)
 		atomic64_set(&ctx->hdev->dram_used_mem, 0);
 }
 
diff --git a/drivers/misc/habanalabs/common/mmu_v1.c b/drivers/misc/habanalabs/common/mmu_v1.c
index ac3784523baa..5f62cb158eef 100644
--- a/drivers/misc/habanalabs/common/mmu_v1.c
+++ b/drivers/misc/habanalabs/common/mmu_v1.c
@@ -262,7 +262,7 @@ static int dram_default_mapping_init(struct hl_ctx *ctx)
 		hop2_pte_addr, hop3_pte_addr, pte_val;
 	int rc, i, j, hop3_allocated = 0;
 
-	if ((!hdev->dram_supports_virtual_memory) ||
+	if ((!prop->dram_supports_virtual_memory) ||
 			(!hdev->dram_default_page_mapping) ||
 			(ctx->asid == HL_KERNEL_ASID_ID))
 		return 0;
@@ -362,7 +362,7 @@ static void dram_default_mapping_fini(struct hl_ctx *ctx)
 		hop2_pte_addr, hop3_pte_addr;
 	int i, j;
 
-	if ((!hdev->dram_supports_virtual_memory) ||
+	if ((!prop->dram_supports_virtual_memory) ||
 			(!hdev->dram_default_page_mapping) ||
 			(ctx->asid == HL_KERNEL_ASID_ID))
 		return;
diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index 1135b2cc5860..8d6cffd28381 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -462,6 +462,7 @@ static int gaudi_get_fixed_properties(struct hl_device *hdev)
 	prop->mmu_hop_table_size = HOP_TABLE_SIZE;
 	prop->mmu_hop0_tables_total_size = HOP0_TABLES_TOTAL_SIZE;
 	prop->dram_page_size = PAGE_SIZE_2MB;
+	prop->dram_supports_virtual_memory = false;
 
 	prop->pmmu.hop0_shift = HOP0_SHIFT;
 	prop->pmmu.hop1_shift = HOP1_SHIFT;
@@ -3573,8 +3574,6 @@ static int gaudi_mmu_init(struct hl_device *hdev)
 	if (gaudi->hw_cap_initialized & HW_CAP_MMU)
 		return 0;
 
-	hdev->dram_supports_virtual_memory = false;
-
 	for (i = 0 ; i < prop->max_asid ; i++) {
 		hop0_addr = prop->mmu_pgt_addr +
 				(i * prop->mmu_hop_table_size);
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
index 002fc53fb39d..bf21f05f7849 100644
--- a/drivers/misc/habanalabs/goya/goya.c
+++ b/drivers/misc/habanalabs/goya/goya.c
@@ -410,6 +410,7 @@ int goya_get_fixed_properties(struct hl_device *hdev)
 	prop->mmu_hop_table_size = HOP_TABLE_SIZE;
 	prop->mmu_hop0_tables_total_size = HOP0_TABLES_TOTAL_SIZE;
 	prop->dram_page_size = PAGE_SIZE_2MB;
+	prop->dram_supports_virtual_memory = true;
 
 	prop->dmmu.hop0_shift = HOP0_SHIFT;
 	prop->dmmu.hop1_shift = HOP1_SHIFT;
@@ -2481,7 +2482,6 @@ int goya_mmu_init(struct hl_device *hdev)
 	if (goya->hw_cap_initialized & HW_CAP_MMU)
 		return 0;
 
-	hdev->dram_supports_virtual_memory = true;
 	hdev->dram_default_page_mapping = true;
 
 	for (i = 0 ; i < prop->max_asid ; i++) {
-- 
2.17.1

[PATCH] habanalabs: use host va range for internal pools

[Oded Gabbay]

From: Ofir Bitton <obitton@habana.ai>

Instead of using a dedicated va range for each internal pool,
we introduce a new way for reserving a va block from an existing
va range. This is a more generic way of reserving va blocks for
future use.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/habanalabs.h |  4 ++
 drivers/misc/habanalabs/common/memory.c     | 79 +++++++++++++++++++++
 drivers/misc/habanalabs/gaudi/gaudi.c       | 18 +++--
 drivers/misc/habanalabs/gaudi/gaudiP.h      |  5 +-
 4 files changed, 98 insertions(+), 8 deletions(-)

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 87060cd2c525..822f90087e04 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -2109,6 +2109,10 @@ void hl_vm_ctx_fini(struct hl_ctx *ctx);
 int hl_vm_init(struct hl_device *hdev);
 void hl_vm_fini(struct hl_device *hdev);
 
+u64 hl_reserve_va_block(struct hl_device *hdev, struct hl_ctx *ctx,
+		enum hl_va_range_type type, u32 size);
+int hl_unreserve_va_block(struct hl_device *hdev, struct hl_ctx *ctx,
+		u64 start_addr, u64 size);
 int hl_pin_host_memory(struct hl_device *hdev, u64 addr, u64 size,
 			struct hl_userptr *userptr);
 void hl_unpin_host_memory(struct hl_device *hdev, struct hl_userptr *userptr);
diff --git a/drivers/misc/habanalabs/common/memory.c b/drivers/misc/habanalabs/common/memory.c
index 02233899336f..01e93e83d3af 100644
--- a/drivers/misc/habanalabs/common/memory.c
+++ b/drivers/misc/habanalabs/common/memory.c
@@ -620,6 +620,85 @@ static u64 get_va_block(struct hl_device *hdev, struct hl_va_range *va_range,
 	return res_valid_start;
 }
 
+/*
+ * hl_reserve_va_block() - reserve a virtual block of a given size.
+ * @hdev: pointer to the habanalabs device structure.
+ * @ctx: current context
+ * @type: virtual addresses range type.
+ * @size: requested block size.
+ *
+ * This function does the following:
+ * - Iterate on the virtual block list to find a suitable virtual block for the
+ *   given size.
+ * - Reserve the requested block and update the list.
+ * - Return the start address of the virtual block.
+ */
+u64 hl_reserve_va_block(struct hl_device *hdev, struct hl_ctx *ctx,
+		enum hl_va_range_type type, u32 size)
+{
+	return get_va_block(hdev, ctx->va_range[type], size, 0,
+			ctx->va_range[type]->page_size);
+}
+
+/**
+ * hl_get_va_range_type() - get va_range type for the given address and size.
+ * @address: The start address of the area we want to validate.
+ * @size: The size in bytes of the area we want to validate.
+ * @type: returned va_range type
+ *
+ * Return: true if the area is inside a valid range, false otherwise.
+ */
+static int hl_get_va_range_type(struct hl_ctx *ctx, u64 address, u64 size,
+			enum hl_va_range_type *type)
+{
+	int i;
+
+	for (i = 0 ; i < HL_VA_RANGE_TYPE_MAX; i++) {
+		if (hl_mem_area_inside_range(address, size,
+				ctx->va_range[i]->start_addr,
+				ctx->va_range[i]->end_addr)) {
+			*type = i;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
+/*
+ * hl_unreserve_va_block - wrapper for add_va_block for unreserving a va block
+ *
+ * @hdev: pointer to the habanalabs device structure
+ * @ctx: current context
+ * @start: start virtual address
+ * @end: end virtual address
+ *
+ * This function does the following:
+ * - Takes the list lock and calls add_va_block_locked
+ */
+int hl_unreserve_va_block(struct hl_device *hdev, struct hl_ctx *ctx,
+		u64 start_addr, u64 size)
+{
+	enum hl_va_range_type type;
+	int rc;
+
+	rc = hl_get_va_range_type(ctx, start_addr, size, &type);
+	if (rc) {
+		dev_err(hdev->dev,
+			"cannot find va_range for va %#llx size %llu",
+			start_addr, size);
+		return rc;
+	}
+
+	rc = add_va_block(hdev, ctx->va_range[type], start_addr,
+						start_addr + size - 1);
+	if (rc)
+		dev_warn(hdev->dev,
+			"add va block failed for vaddr: 0x%llx\n", start_addr);
+
+	return rc;
+}
+
 /*
  * get_sg_info - get number of pages and the DMA address from SG list
  *
diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index 086d8363f3ae..1135b2cc5860 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -7768,7 +7768,11 @@ static int gaudi_internal_cb_pool_init(struct hl_device *hdev,
 		goto destroy_internal_cb_pool;
 	}
 
-	hdev->internal_cb_va_base = VA_HOST_SPACE_INTERNAL_CB_START;
+	hdev->internal_cb_va_base = hl_reserve_va_block(hdev, ctx,
+			HL_VA_RANGE_TYPE_HOST, HOST_SPACE_INTERNAL_CB_SZ);
+
+	if (!hdev->internal_cb_va_base)
+		goto destroy_internal_cb_pool;
 
 	mutex_lock(&ctx->mmu_lock);
 
@@ -7776,7 +7780,7 @@ static int gaudi_internal_cb_pool_init(struct hl_device *hdev,
 	 * is aligned to HOST_SPACE_INTERNAL_CB_SZ
 	 */
 	for (off = 0 ; off < HOST_SPACE_INTERNAL_CB_SZ ; off += PAGE_SIZE_4KB) {
-		va = VA_HOST_SPACE_INTERNAL_CB_START + off;
+		va = hdev->internal_cb_va_base + off;
 		pa = hdev->internal_cb_pool_dma_addr + off;
 		flush_pte = (off + PAGE_SIZE_4KB) >= HOST_SPACE_INTERNAL_CB_SZ;
 		rc = hl_mmu_map(ctx, va, pa, PAGE_SIZE_4KB, flush_pte);
@@ -7796,13 +7800,16 @@ static int gaudi_internal_cb_pool_init(struct hl_device *hdev,
 
 unmap:
 	for (; off >= 0 ; off -= PAGE_SIZE_4KB) {
-		va = VA_HOST_SPACE_INTERNAL_CB_START + off;
+		va = hdev->internal_cb_va_base + off;
 		flush_pte = (off - (s32) PAGE_SIZE_4KB) < 0;
 		if (hl_mmu_unmap(ctx, va, PAGE_SIZE_4KB, flush_pte))
 			dev_warn_ratelimited(hdev->dev,
 					"failed to unmap va 0x%llx\n", va);
 	}
 
+	hl_unreserve_va_block(hdev, ctx, hdev->internal_cb_va_base,
+			HOST_SPACE_INTERNAL_CB_SZ);
+
 	hdev->asic_funcs->mmu_invalidate_cache(hdev, true, VM_TYPE_USERPTR);
 
 	mutex_unlock(&ctx->mmu_lock);
@@ -7832,7 +7839,7 @@ static void gaudi_internal_cb_pool_fini(struct hl_device *hdev,
 	mutex_lock(&ctx->mmu_lock);
 
 	for (off = 0 ; off < HOST_SPACE_INTERNAL_CB_SZ ; off += PAGE_SIZE_4KB) {
-		va = VA_HOST_SPACE_INTERNAL_CB_START + off;
+		va = hdev->internal_cb_va_base + off;
 
 		if (off + PAGE_SIZE_4KB >= HOST_SPACE_INTERNAL_CB_SZ)
 			flush_pte = true;
@@ -7842,6 +7849,9 @@ static void gaudi_internal_cb_pool_fini(struct hl_device *hdev,
 					"failed to unmap va 0x%llx\n", va);
 	}
 
+	hl_unreserve_va_block(hdev, ctx, hdev->internal_cb_va_base,
+			HOST_SPACE_INTERNAL_CB_SZ);
+
 	hdev->asic_funcs->mmu_invalidate_cache(hdev, true, VM_TYPE_USERPTR);
 
 	mutex_unlock(&ctx->mmu_lock);
diff --git a/drivers/misc/habanalabs/gaudi/gaudiP.h b/drivers/misc/habanalabs/gaudi/gaudiP.h
index e882cf7f8cf7..1a5e681c720d 100644
--- a/drivers/misc/habanalabs/gaudi/gaudiP.h
+++ b/drivers/misc/habanalabs/gaudi/gaudiP.h
@@ -153,12 +153,9 @@
 
 /* Virtual address space */
 #define VA_HOST_SPACE_START	0x1000000000000ull	/* 256TB */
-#define VA_HOST_SPACE_END	0x3FF7FFFE00000ull	/* 1PB - 1TB */
+#define VA_HOST_SPACE_END	0x3FF8000000000ull	/* 1PB - 512GB */
 #define VA_HOST_SPACE_SIZE	(VA_HOST_SPACE_END - \
 					VA_HOST_SPACE_START) /* 767TB */
-
-#define VA_HOST_SPACE_INTERNAL_CB_START	0x3FF7FFFE00000ull /* 1PB - 1TB - 2MB */
-#define VA_HOST_SPACE_INTERNAL_CB_END	0x3FF8000000000ull /* 1PB - 1TB */
 #define HOST_SPACE_INTERNAL_CB_SZ	SZ_2M
 
 #define HW_CAP_PLL		BIT(0)
-- 
2.17.1