iommu/vt-d: add wrapper functions for page allocations

In order to improve observability and accountability of IOMMU layer, we
must account the number of pages that are allocated by functions that
are calling directly into buddy allocator.

This is achieved by first wrapping the allocation related functions into a
separate inline functions in new file:

drivers/iommu/iommu-pages.h

Convert all page allocation calls under iommu/intel to use these new
functions.

Signed-off-by: Pasha Tatashin <[email protected]>
Acked-by: David Rientjes <[email protected]>
Tested-by: Bagas Sanjaya <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Joerg Roedel <[email protected]>

Author: soleen

drivers/iommu/intel/dmar.c

@@ -32,6 +32,7 @@
 
 #include "iommu.h"
 #include "../irq_remapping.h"
+#include "../iommu-pages.h"
 #include "perf.h"
 #include "trace.h"
 #include "perfmon.h"
@@ -1187,7 +1188,7 @@ static void free_iommu(struct intel_iommu *iommu)
 	}
 
 	if (iommu->qi) {
-		free_page((unsigned long)iommu->qi->desc);
+		iommu_free_page(iommu->qi->desc);
 		kfree(iommu->qi->desc_status);
 		kfree(iommu->qi);
 	}
@@ -1755,7 +1756,8 @@ static void __dmar_enable_qi(struct intel_iommu *iommu)
 int dmar_enable_qi(struct intel_iommu *iommu)
 {
 	struct q_inval *qi;
-	struct page *desc_page;
+	void *desc;
+	int order;
 
 	if (!ecap_qis(iommu->ecap))
 		return -ENOENT;
@@ -1776,19 +1778,19 @@ int dmar_enable_qi(struct intel_iommu *iommu)
 	 * Need two pages to accommodate 256 descriptors of 256 bits each
 	 * if the remapping hardware supports scalable mode translation.
 	 */
-	desc_page = alloc_pages_node(iommu->node, GFP_ATOMIC | __GFP_ZERO,
-				     !!ecap_smts(iommu->ecap));
-	if (!desc_page) {
+	order = ecap_smts(iommu->ecap) ? 1 : 0;
+	desc = iommu_alloc_pages_node(iommu->node, GFP_ATOMIC, order);
+	if (!desc) {
 		kfree(qi);
 		iommu->qi = NULL;
 		return -ENOMEM;
 	}
 
-	qi->desc = page_address(desc_page);
+	qi->desc = desc;
 
 	qi->desc_status = kcalloc(QI_LENGTH, sizeof(int), GFP_ATOMIC);
 	if (!qi->desc_status) {
-		free_page((unsigned long) qi->desc);
+		iommu_free_page(qi->desc);
 		kfree(qi);
 		iommu->qi = NULL;
 		return -ENOMEM;

drivers/iommu/intel/iommu.c

@@ -27,6 +27,7 @@
 #include "iommu.h"
 #include "../dma-iommu.h"
 #include "../irq_remapping.h"
+#include "../iommu-pages.h"
 #include "pasid.h"
 #include "cap_audit.h"
 #include "perfmon.h"
@@ -298,22 +299,6 @@ static int __init intel_iommu_setup(char *str)
 }
 __setup("intel_iommu=", intel_iommu_setup);
 
-void *alloc_pgtable_page(int node, gfp_t gfp)
-{
-	struct page *page;
-	void *vaddr = NULL;
-
-	page = alloc_pages_node(node, gfp | __GFP_ZERO, 0);
-	if (page)
-		vaddr = page_address(page);
-	return vaddr;
-}
-
-void free_pgtable_page(void *vaddr)
-{
-	free_page((unsigned long)vaddr);
-}
-
 static int domain_type_is_si(struct dmar_domain *domain)
 {
 	return domain->domain.type == IOMMU_DOMAIN_IDENTITY;
@@ -545,7 +530,7 @@ struct context_entry *iommu_context_addr(struct intel_iommu *iommu, u8 bus,
 		if (!alloc)
 			return NULL;
 
-		context = alloc_pgtable_page(iommu->node, GFP_ATOMIC);
+		context = iommu_alloc_page_node(iommu->node, GFP_ATOMIC);
 		if (!context)
 			return NULL;
 
@@ -719,17 +704,17 @@ static void free_context_table(struct intel_iommu *iommu)
 	for (i = 0; i < ROOT_ENTRY_NR; i++) {
 		context = iommu_context_addr(iommu, i, 0, 0);
 		if (context)
-			free_pgtable_page(context);
+			iommu_free_page(context);
 
 		if (!sm_supported(iommu))
 			continue;
 
 		context = iommu_context_addr(iommu, i, 0x80, 0);
 		if (context)
-			free_pgtable_page(context);
+			iommu_free_page(context);
 	}
 
-	free_pgtable_page(iommu->root_entry);
+	iommu_free_page(iommu->root_entry);
 	iommu->root_entry = NULL;
 }
 
@@ -867,7 +852,7 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 		if (!dma_pte_present(pte)) {
 			uint64_t pteval;
 
-			tmp_page = alloc_pgtable_page(domain->nid, gfp);
+			tmp_page = iommu_alloc_page_node(domain->nid, gfp);
 
 			if (!tmp_page)
 				return NULL;
@@ -879,7 +864,7 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 
 			if (cmpxchg64(&pte->val, 0ULL, pteval))
 				/* Someone else set it while we were thinking; use theirs. */
-				free_pgtable_page(tmp_page);
+				iommu_free_page(tmp_page);
 			else
 				domain_flush_cache(domain, pte, sizeof(*pte));
 		}
@@ -992,7 +977,7 @@ static void dma_pte_free_level(struct dmar_domain *domain, int level,
 		      last_pfn < level_pfn + level_size(level) - 1)) {
 			dma_clear_pte(pte);
 			domain_flush_cache(domain, pte, sizeof(*pte));
-			free_pgtable_page(level_pte);
+			iommu_free_page(level_pte);
 		}
 next:
 		pfn += level_size(level);
@@ -1016,7 +1001,7 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 
 	/* free pgd */
 	if (start_pfn == 0 && last_pfn == DOMAIN_MAX_PFN(domain->gaw)) {
-		free_pgtable_page(domain->pgd);
+		iommu_free_page(domain->pgd);
 		domain->pgd = NULL;
 	}
 }
@@ -1118,7 +1103,7 @@ static int iommu_alloc_root_entry(struct intel_iommu *iommu)
 {
 	struct root_entry *root;
 
-	root = alloc_pgtable_page(iommu->node, GFP_ATOMIC);
+	root = iommu_alloc_page_node(iommu->node, GFP_ATOMIC);
 	if (!root) {
 		pr_err("Allocating root entry for %s failed\n",
 			iommu->name);
@@ -1841,7 +1826,7 @@ static void domain_exit(struct dmar_domain *domain)
 		LIST_HEAD(freelist);
 
 		domain_unmap(domain, 0, DOMAIN_MAX_PFN(domain->gaw), &freelist);
-		put_pages_list(&freelist);
+		iommu_put_pages_list(&freelist);
 	}
 
 	if (WARN_ON(!list_empty(&domain->devices)))
@@ -2497,7 +2482,7 @@ static int copy_context_table(struct intel_iommu *iommu,
 			if (!old_ce)
 				goto out;
 
-			new_ce = alloc_pgtable_page(iommu->node, GFP_KERNEL);
+			new_ce = iommu_alloc_page_node(iommu->node, GFP_KERNEL);
 			if (!new_ce)
 				goto out_unmap;
 
@@ -3426,7 +3411,7 @@ static int intel_iommu_memory_notifier(struct notifier_block *nb,
 					start_vpfn, mhp->nr_pages,
 					list_empty(&freelist), 0);
 			rcu_read_unlock();
-			put_pages_list(&freelist);
+			iommu_put_pages_list(&freelist);
 		}
 		break;
 	}
@@ -3833,7 +3818,7 @@ static int md_domain_init(struct dmar_domain *domain, int guest_width)
 	domain->max_addr = 0;
 
 	/* always allocate the top pgd */
-	domain->pgd = alloc_pgtable_page(domain->nid, GFP_ATOMIC);
+	domain->pgd = iommu_alloc_page_node(domain->nid, GFP_ATOMIC);
 	if (!domain->pgd)
 		return -ENOMEM;
 	domain_flush_cache(domain, domain->pgd, PAGE_SIZE);
@@ -3987,7 +3972,7 @@ int prepare_domain_attach_device(struct iommu_domain *domain,
 		pte = dmar_domain->pgd;
 		if (dma_pte_present(pte)) {
 			dmar_domain->pgd = phys_to_virt(dma_pte_addr(pte));
-			free_pgtable_page(pte);
+			iommu_free_page(pte);
 		}
 		dmar_domain->agaw--;
 	}
@@ -4141,7 +4126,7 @@ static void intel_iommu_tlb_sync(struct iommu_domain *domain,
 	if (dmar_domain->nested_parent)
 		parent_domain_flush(dmar_domain, start_pfn, nrpages,
 				    list_empty(&gather->freelist));
-	put_pages_list(&gather->freelist);
+	iommu_put_pages_list(&gather->freelist);
 }
 
 static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,

drivers/iommu/intel/iommu.h

@@ -1085,8 +1085,6 @@ void domain_update_iommu_cap(struct dmar_domain *domain);
 
 int dmar_ir_support(void);
 
-void *alloc_pgtable_page(int node, gfp_t gfp);
-void free_pgtable_page(void *vaddr);
 void iommu_flush_write_buffer(struct intel_iommu *iommu);
 struct iommu_domain *intel_nested_domain_alloc(struct iommu_domain *parent,
 					       const struct iommu_user_data *user_data);

drivers/iommu/intel/irq_remapping.c

@@ -22,6 +22,7 @@
 
 #include "iommu.h"
 #include "../irq_remapping.h"
+#include "../iommu-pages.h"
 #include "cap_audit.h"
 
 enum irq_mode {
@@ -527,7 +528,7 @@ static int intel_setup_irq_remapping(struct intel_iommu *iommu)
 	struct ir_table *ir_table;
 	struct fwnode_handle *fn;
 	unsigned long *bitmap;
-	struct page *pages;
+	void *ir_table_base;
 
 	if (iommu->ir_table)
 		return 0;
@@ -536,9 +537,9 @@ static int intel_setup_irq_remapping(struct intel_iommu *iommu)
 	if (!ir_table)
 		return -ENOMEM;
 
-	pages = alloc_pages_node(iommu->node, GFP_KERNEL | __GFP_ZERO,
-				 INTR_REMAP_PAGE_ORDER);
-	if (!pages) {
+	ir_table_base = iommu_alloc_pages_node(iommu->node, GFP_KERNEL,
+					       INTR_REMAP_PAGE_ORDER);
+	if (!ir_table_base) {
 		pr_err("IR%d: failed to allocate pages of order %d\n",
 		       iommu->seq_id, INTR_REMAP_PAGE_ORDER);
 		goto out_free_table;
@@ -573,7 +574,7 @@ static int intel_setup_irq_remapping(struct intel_iommu *iommu)
 	else
 		iommu->ir_domain->msi_parent_ops = &dmar_msi_parent_ops;
 
-	ir_table->base = page_address(pages);
+	ir_table->base = ir_table_base;
 	ir_table->bitmap = bitmap;
 	iommu->ir_table = ir_table;
 
@@ -622,7 +623,7 @@ static int intel_setup_irq_remapping(struct intel_iommu *iommu)
 out_free_bitmap:
 	bitmap_free(bitmap);
 out_free_pages:
-	__free_pages(pages, INTR_REMAP_PAGE_ORDER);
+	iommu_free_pages(ir_table_base, INTR_REMAP_PAGE_ORDER);
 out_free_table:
 	kfree(ir_table);
 
@@ -643,8 +644,7 @@ static void intel_teardown_irq_remapping(struct intel_iommu *iommu)
 			irq_domain_free_fwnode(fn);
 			iommu->ir_domain = NULL;
 		}
-		free_pages((unsigned long)iommu->ir_table->base,
-			   INTR_REMAP_PAGE_ORDER);
+		iommu_free_pages(iommu->ir_table->base, INTR_REMAP_PAGE_ORDER);
 		bitmap_free(iommu->ir_table->bitmap);
 		kfree(iommu->ir_table);
 		iommu->ir_table = NULL;

drivers/iommu/intel/pasid.c

@@ -20,6 +20,7 @@
 
 #include "iommu.h"
 #include "pasid.h"
+#include "../iommu-pages.h"
 
 /*
  * Intel IOMMU system wide PASID name space:
@@ -38,7 +39,7 @@ int intel_pasid_alloc_table(struct device *dev)
 {
 	struct device_domain_info *info;
 	struct pasid_table *pasid_table;
-	struct page *pages;
+	struct pasid_dir_entry *dir;
 	u32 max_pasid = 0;
 	int order, size;
 
@@ -59,14 +60,13 @@ int intel_pasid_alloc_table(struct device *dev)
 
 	size = max_pasid >> (PASID_PDE_SHIFT - 3);
 	order = size ? get_order(size) : 0;
-	pages = alloc_pages_node(info->iommu->node,
-				 GFP_KERNEL | __GFP_ZERO, order);
-	if (!pages) {
+	dir = iommu_alloc_pages_node(info->iommu->node, GFP_KERNEL, order);
+	if (!dir) {
 		kfree(pasid_table);
 		return -ENOMEM;
 	}
 
-	pasid_table->table = page_address(pages);
+	pasid_table->table = dir;
 	pasid_table->order = order;
 	pasid_table->max_pasid = 1 << (order + PAGE_SHIFT + 3);
 	info->pasid_table = pasid_table;
@@ -97,10 +97,10 @@ void intel_pasid_free_table(struct device *dev)
 	max_pde = pasid_table->max_pasid >> PASID_PDE_SHIFT;
 	for (i = 0; i < max_pde; i++) {
 		table = get_pasid_table_from_pde(&dir[i]);
-		free_pgtable_page(table);
+		iommu_free_page(table);
 	}
 
-	free_pages((unsigned long)pasid_table->table, pasid_table->order);
+	iommu_free_pages(pasid_table->table, pasid_table->order);
 	kfree(pasid_table);
 }
 
@@ -146,7 +146,7 @@ static struct pasid_entry *intel_pasid_get_entry(struct device *dev, u32 pasid)
 retry:
 	entries = get_pasid_table_from_pde(&dir[dir_index]);
 	if (!entries) {
-		entries = alloc_pgtable_page(info->iommu->node, GFP_ATOMIC);
+		entries = iommu_alloc_page_node(info->iommu->node, GFP_ATOMIC);
 		if (!entries)
 			return NULL;
 
@@ -158,7 +158,7 @@ static struct pasid_entry *intel_pasid_get_entry(struct device *dev, u32 pasid)
 		 */
 		if (cmpxchg64(&dir[dir_index].val, 0ULL,
 			      (u64)virt_to_phys(entries) | PASID_PTE_PRESENT)) {
-			free_pgtable_page(entries);
+			iommu_free_page(entries);
 			goto retry;
 		}
 		if (!ecap_coherent(info->iommu->ecap)) {

drivers/iommu/intel/svm.c

@@ -22,6 +22,7 @@
 #include "iommu.h"
 #include "pasid.h"
 #include "perf.h"
+#include "../iommu-pages.h"
 #include "trace.h"
 
 static irqreturn_t prq_event_thread(int irq, void *d);
@@ -63,16 +64,14 @@ svm_lookup_device_by_dev(struct intel_svm *svm, struct device *dev)
 int intel_svm_enable_prq(struct intel_iommu *iommu)
 {
 	struct iopf_queue *iopfq;
-	struct page *pages;
 	int irq, ret;
 
-	pages = alloc_pages_node(iommu->node, GFP_KERNEL | __GFP_ZERO, PRQ_ORDER);
-	if (!pages) {
+	iommu->prq = iommu_alloc_pages_node(iommu->node, GFP_KERNEL, PRQ_ORDER);
+	if (!iommu->prq) {
 		pr_warn("IOMMU: %s: Failed to allocate page request queue\n",
 			iommu->name);
 		return -ENOMEM;
 	}
-	iommu->prq = page_address(pages);
 
 	irq = dmar_alloc_hwirq(IOMMU_IRQ_ID_OFFSET_PRQ + iommu->seq_id, iommu->node, iommu);
 	if (irq <= 0) {
@@ -117,7 +116,7 @@ int intel_svm_enable_prq(struct intel_iommu *iommu)
 	dmar_free_hwirq(irq);
 	iommu->pr_irq = 0;
 free_prq:
-	free_pages((unsigned long)iommu->prq, PRQ_ORDER);
+	iommu_free_pages(iommu->prq, PRQ_ORDER);
 	iommu->prq = NULL;
 
 	return ret;
@@ -140,7 +139,7 @@ int intel_svm_finish_prq(struct intel_iommu *iommu)
 		iommu->iopf_queue = NULL;
 	}
 
-	free_pages((unsigned long)iommu->prq, PRQ_ORDER);
+	iommu_free_pages(iommu->prq, PRQ_ORDER);
 	iommu->prq = NULL;
 
 	return 0;