Merge branch 'memory-observability' into x86/amd

Author: joergroedel

Documentation/admin-guide/cgroup-v2.rst

@@ -1432,7 +1432,7 @@ PAGE_SIZE multiple when read back.
 	  sec_pagetables
 		Amount of memory allocated for secondary page tables,
 		this currently includes KVM mmu allocations on x86
-		and arm64.
+		and arm64 and IOMMU page tables.
 
 	  percpu (npn)
 		Amount of memory used for storing per-cpu kernel

Documentation/filesystems/proc.rst

@@ -1110,8 +1110,8 @@ KernelStack
 PageTables
               Memory consumed by userspace page tables
 SecPageTables
-              Memory consumed by secondary page tables, this currently
-              currently includes KVM mmu allocations on x86 and arm64.
+              Memory consumed by secondary page tables, this currently includes
+              KVM mmu and IOMMU allocations on x86 and arm64.
 NFS_Unstable
               Always zero. Previous counted pages which had been written to
               the server, but has not been committed to stable storage.

drivers/iommu/amd/amd_iommu.h

@@ -162,14 +162,6 @@ static inline int get_pci_sbdf_id(struct pci_dev *pdev)
 	return PCI_SEG_DEVID_TO_SBDF(seg, devid);
 }
 
-static inline void *alloc_pgtable_page(int nid, gfp_t gfp)
-{
-	struct page *page;
-
-	page = alloc_pages_node(nid, gfp | __GFP_ZERO, 0);
-	return page ? page_address(page) : NULL;
-}
-
 /*
  * This must be called after device probe completes. During probe
  * use rlookup_amd_iommu() get the iommu.

drivers/iommu/amd/init.c

@@ -36,6 +36,7 @@
 
 #include "amd_iommu.h"
 #include "../irq_remapping.h"
+#include "../iommu-pages.h"
 
 /*
  * definitions for the ACPI scanning code
@@ -649,8 +650,8 @@ static int __init find_last_devid_acpi(struct acpi_table_header *table, u16 pci_
 /* Allocate per PCI segment device table */
 static inline int __init alloc_dev_table(struct amd_iommu_pci_seg *pci_seg)
 {
-	pci_seg->dev_table = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO | GFP_DMA32,
-						      get_order(pci_seg->dev_table_size));
+	pci_seg->dev_table = iommu_alloc_pages(GFP_KERNEL | GFP_DMA32,
+					       get_order(pci_seg->dev_table_size));
 	if (!pci_seg->dev_table)
 		return -ENOMEM;
 
@@ -659,17 +660,16 @@ static inline int __init alloc_dev_table(struct amd_iommu_pci_seg *pci_seg)
 
 static inline void free_dev_table(struct amd_iommu_pci_seg *pci_seg)
 {
-	free_pages((unsigned long)pci_seg->dev_table,
-		    get_order(pci_seg->dev_table_size));
+	iommu_free_pages(pci_seg->dev_table,
+			 get_order(pci_seg->dev_table_size));
 	pci_seg->dev_table = NULL;
 }
 
 /* Allocate per PCI segment IOMMU rlookup table. */
 static inline int __init alloc_rlookup_table(struct amd_iommu_pci_seg *pci_seg)
 {
-	pci_seg->rlookup_table = (void *)__get_free_pages(
-						GFP_KERNEL | __GFP_ZERO,
-						get_order(pci_seg->rlookup_table_size));
+	pci_seg->rlookup_table = iommu_alloc_pages(GFP_KERNEL,
+						   get_order(pci_seg->rlookup_table_size));
 	if (pci_seg->rlookup_table == NULL)
 		return -ENOMEM;
 
@@ -678,16 +678,15 @@ static inline int __init alloc_rlookup_table(struct amd_iommu_pci_seg *pci_seg)
 
 static inline void free_rlookup_table(struct amd_iommu_pci_seg *pci_seg)
 {
-	free_pages((unsigned long)pci_seg->rlookup_table,
-		   get_order(pci_seg->rlookup_table_size));
+	iommu_free_pages(pci_seg->rlookup_table,
+			 get_order(pci_seg->rlookup_table_size));
 	pci_seg->rlookup_table = NULL;
 }
 
 static inline int __init alloc_irq_lookup_table(struct amd_iommu_pci_seg *pci_seg)
 {
-	pci_seg->irq_lookup_table = (void *)__get_free_pages(
-					     GFP_KERNEL | __GFP_ZERO,
-					     get_order(pci_seg->rlookup_table_size));
+	pci_seg->irq_lookup_table = iommu_alloc_pages(GFP_KERNEL,
+						      get_order(pci_seg->rlookup_table_size));
 	kmemleak_alloc(pci_seg->irq_lookup_table,
 		       pci_seg->rlookup_table_size, 1, GFP_KERNEL);
 	if (pci_seg->irq_lookup_table == NULL)
@@ -699,17 +698,17 @@ static inline int __init alloc_irq_lookup_table(struct amd_iommu_pci_seg *pci_se
 static inline void free_irq_lookup_table(struct amd_iommu_pci_seg *pci_seg)
 {
 	kmemleak_free(pci_seg->irq_lookup_table);
-	free_pages((unsigned long)pci_seg->irq_lookup_table,
-		   get_order(pci_seg->rlookup_table_size));
+	iommu_free_pages(pci_seg->irq_lookup_table,
+			 get_order(pci_seg->rlookup_table_size));
 	pci_seg->irq_lookup_table = NULL;
 }
 
 static int __init alloc_alias_table(struct amd_iommu_pci_seg *pci_seg)
 {
 	int i;
 
-	pci_seg->alias_table = (void *)__get_free_pages(GFP_KERNEL,
-					get_order(pci_seg->alias_table_size));
+	pci_seg->alias_table = iommu_alloc_pages(GFP_KERNEL,
+						 get_order(pci_seg->alias_table_size));
 	if (!pci_seg->alias_table)
 		return -ENOMEM;
 
@@ -724,8 +723,8 @@ static int __init alloc_alias_table(struct amd_iommu_pci_seg *pci_seg)
 
 static void __init free_alias_table(struct amd_iommu_pci_seg *pci_seg)
 {
-	free_pages((unsigned long)pci_seg->alias_table,
-		   get_order(pci_seg->alias_table_size));
+	iommu_free_pages(pci_seg->alias_table,
+			 get_order(pci_seg->alias_table_size));
 	pci_seg->alias_table = NULL;
 }
 
@@ -736,8 +735,8 @@ static void __init free_alias_table(struct amd_iommu_pci_seg *pci_seg)
  */
 static int __init alloc_command_buffer(struct amd_iommu *iommu)
 {
-	iommu->cmd_buf = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
-						  get_order(CMD_BUFFER_SIZE));
+	iommu->cmd_buf = iommu_alloc_pages(GFP_KERNEL,
+					   get_order(CMD_BUFFER_SIZE));
 
 	return iommu->cmd_buf ? 0 : -ENOMEM;
 }
@@ -834,19 +833,19 @@ static void iommu_disable_command_buffer(struct amd_iommu *iommu)
 
 static void __init free_command_buffer(struct amd_iommu *iommu)
 {
-	free_pages((unsigned long)iommu->cmd_buf, get_order(CMD_BUFFER_SIZE));
+	iommu_free_pages(iommu->cmd_buf, get_order(CMD_BUFFER_SIZE));
 }
 
 void *__init iommu_alloc_4k_pages(struct amd_iommu *iommu, gfp_t gfp,
 				  size_t size)
 {
 	int order = get_order(size);
-	void *buf = (void *)__get_free_pages(gfp, order);
+	void *buf = iommu_alloc_pages(gfp, order);
 
 	if (buf &&
 	    check_feature(FEATURE_SNP) &&
 	    set_memory_4k((unsigned long)buf, (1 << order))) {
-		free_pages((unsigned long)buf, order);
+		iommu_free_pages(buf, order);
 		buf = NULL;
 	}
 
@@ -856,7 +855,7 @@ void *__init iommu_alloc_4k_pages(struct amd_iommu *iommu, gfp_t gfp,
 /* allocates the memory where the IOMMU will log its events to */
 static int __init alloc_event_buffer(struct amd_iommu *iommu)
 {
-	iommu->evt_buf = iommu_alloc_4k_pages(iommu, GFP_KERNEL | __GFP_ZERO,
+	iommu->evt_buf = iommu_alloc_4k_pages(iommu, GFP_KERNEL,
 					      EVT_BUFFER_SIZE);
 
 	return iommu->evt_buf ? 0 : -ENOMEM;
@@ -890,14 +889,14 @@ static void iommu_disable_event_buffer(struct amd_iommu *iommu)
 
 static void __init free_event_buffer(struct amd_iommu *iommu)
 {
-	free_pages((unsigned long)iommu->evt_buf, get_order(EVT_BUFFER_SIZE));
+	iommu_free_pages(iommu->evt_buf, get_order(EVT_BUFFER_SIZE));
 }
 
 static void free_ga_log(struct amd_iommu *iommu)
 {
 #ifdef CONFIG_IRQ_REMAP
-	free_pages((unsigned long)iommu->ga_log, get_order(GA_LOG_SIZE));
-	free_pages((unsigned long)iommu->ga_log_tail, get_order(8));
+	iommu_free_pages(iommu->ga_log, get_order(GA_LOG_SIZE));
+	iommu_free_pages(iommu->ga_log_tail, get_order(8));
 #endif
 }
 
@@ -942,13 +941,11 @@ static int iommu_init_ga_log(struct amd_iommu *iommu)
 	if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir))
 		return 0;
 
-	iommu->ga_log = (u8 *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
-					get_order(GA_LOG_SIZE));
+	iommu->ga_log = iommu_alloc_pages(GFP_KERNEL, get_order(GA_LOG_SIZE));
 	if (!iommu->ga_log)
 		goto err_out;
 
-	iommu->ga_log_tail = (u8 *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
-					get_order(8));
+	iommu->ga_log_tail = iommu_alloc_pages(GFP_KERNEL, get_order(8));
 	if (!iommu->ga_log_tail)
 		goto err_out;
 
@@ -961,15 +958,15 @@ static int iommu_init_ga_log(struct amd_iommu *iommu)
 
 static int __init alloc_cwwb_sem(struct amd_iommu *iommu)
 {
-	iommu->cmd_sem = iommu_alloc_4k_pages(iommu, GFP_KERNEL | __GFP_ZERO, 1);
+	iommu->cmd_sem = iommu_alloc_4k_pages(iommu, GFP_KERNEL, 1);
 
 	return iommu->cmd_sem ? 0 : -ENOMEM;
 }
 
 static void __init free_cwwb_sem(struct amd_iommu *iommu)
 {
 	if (iommu->cmd_sem)
-		free_page((unsigned long)iommu->cmd_sem);
+		iommu_free_page((void *)iommu->cmd_sem);
 }
 
 static void iommu_enable_xt(struct amd_iommu *iommu)
@@ -1034,7 +1031,6 @@ static bool __copy_device_table(struct amd_iommu *iommu)
 	u32 lo, hi, devid, old_devtb_size;
 	phys_addr_t old_devtb_phys;
 	u16 dom_id, dte_v, irq_v;
-	gfp_t gfp_flag;
 	u64 tmp;
 
 	/* Each IOMMU use separate device table with the same size */
@@ -1068,9 +1064,8 @@ static bool __copy_device_table(struct amd_iommu *iommu)
 	if (!old_devtb)
 		return false;
 
-	gfp_flag = GFP_KERNEL | __GFP_ZERO | GFP_DMA32;
-	pci_seg->old_dev_tbl_cpy = (void *)__get_free_pages(gfp_flag,
-						    get_order(pci_seg->dev_table_size));
+	pci_seg->old_dev_tbl_cpy = iommu_alloc_pages(GFP_KERNEL | GFP_DMA32,
+						     get_order(pci_seg->dev_table_size));
 	if (pci_seg->old_dev_tbl_cpy == NULL) {
 		pr_err("Failed to allocate memory for copying old device table!\n");
 		memunmap(old_devtb);
@@ -2769,8 +2764,8 @@ static void early_enable_iommus(void)
 
 		for_each_pci_segment(pci_seg) {
 			if (pci_seg->old_dev_tbl_cpy != NULL) {
-				free_pages((unsigned long)pci_seg->old_dev_tbl_cpy,
-						get_order(pci_seg->dev_table_size));
+				iommu_free_pages(pci_seg->old_dev_tbl_cpy,
+						 get_order(pci_seg->dev_table_size));
 				pci_seg->old_dev_tbl_cpy = NULL;
 			}
 		}
@@ -2783,8 +2778,8 @@ static void early_enable_iommus(void)
 		pr_info("Copied DEV table from previous kernel.\n");
 
 		for_each_pci_segment(pci_seg) {
-			free_pages((unsigned long)pci_seg->dev_table,
-				   get_order(pci_seg->dev_table_size));
+			iommu_free_pages(pci_seg->dev_table,
+					 get_order(pci_seg->dev_table_size));
 			pci_seg->dev_table = pci_seg->old_dev_tbl_cpy;
 		}
 
@@ -2989,8 +2984,8 @@ static bool __init check_ioapic_information(void)
 
 static void __init free_dma_resources(void)
 {
-	free_pages((unsigned long)amd_iommu_pd_alloc_bitmap,
-		   get_order(MAX_DOMAIN_ID/8));
+	iommu_free_pages(amd_iommu_pd_alloc_bitmap,
+			 get_order(MAX_DOMAIN_ID / 8));
 	amd_iommu_pd_alloc_bitmap = NULL;
 
 	free_unity_maps();
@@ -3062,9 +3057,8 @@ static int __init early_amd_iommu_init(void)
 	/* Device table - directly used by all IOMMUs */
 	ret = -ENOMEM;
 
-	amd_iommu_pd_alloc_bitmap = (void *)__get_free_pages(
-					    GFP_KERNEL | __GFP_ZERO,
-					    get_order(MAX_DOMAIN_ID/8));
+	amd_iommu_pd_alloc_bitmap = iommu_alloc_pages(GFP_KERNEL,
+						      get_order(MAX_DOMAIN_ID / 8));
 	if (amd_iommu_pd_alloc_bitmap == NULL)
 		goto out;
 

drivers/iommu/amd/io_pgtable.c

@@ -22,6 +22,7 @@
 
 #include "amd_iommu_types.h"
 #include "amd_iommu.h"
+#include "../iommu-pages.h"
 
 static void v1_tlb_flush_all(void *cookie)
 {
@@ -156,7 +157,7 @@ static bool increase_address_space(struct protection_domain *domain,
 	bool ret = true;
 	u64 *pte;
 
-	pte = alloc_pgtable_page(domain->nid, gfp);
+	pte = iommu_alloc_page_node(domain->nid, gfp);
 	if (!pte)
 		return false;
 
@@ -187,7 +188,7 @@ static bool increase_address_space(struct protection_domain *domain,
 
 out:
 	spin_unlock_irqrestore(&domain->lock, flags);
-	free_page((unsigned long)pte);
+	iommu_free_page(pte);
 
 	return ret;
 }
@@ -250,7 +251,7 @@ static u64 *alloc_pte(struct protection_domain *domain,
 
 		if (!IOMMU_PTE_PRESENT(__pte) ||
 		    pte_level == PAGE_MODE_NONE) {
-			page = alloc_pgtable_page(domain->nid, gfp);
+			page = iommu_alloc_page_node(domain->nid, gfp);
 
 			if (!page)
 				return NULL;
@@ -259,7 +260,7 @@ static u64 *alloc_pte(struct protection_domain *domain,
 
 			/* pte could have been changed somewhere. */
 			if (!try_cmpxchg64(pte, &__pte, __npte))
-				free_page((unsigned long)page);
+				iommu_free_page(page);
 			else if (IOMMU_PTE_PRESENT(__pte))
 				*updated = true;
 
@@ -431,7 +432,7 @@ static int iommu_v1_map_pages(struct io_pgtable_ops *ops, unsigned long iova,
 	}
 
 	/* Everything flushed out, free pages now */
-	put_pages_list(&freelist);
+	iommu_put_pages_list(&freelist);
 
 	return ret;
 }
@@ -580,7 +581,7 @@ static void v1_free_pgtable(struct io_pgtable *iop)
 	/* Make changes visible to IOMMUs */
 	amd_iommu_domain_update(dom);
 
-	put_pages_list(&freelist);
+	iommu_put_pages_list(&freelist);
 }
 
 static struct io_pgtable *v1_alloc_pgtable(struct io_pgtable_cfg *cfg, void *cookie)