iommu/amd: Consolidate feature detection and reporting logic

Currently, IOMMU driver assumes capabilities on all IOMMU instances to be
homogeneous. During early_amd_iommu_init(), the driver probes all IVHD
blocks and do sanity check to make sure that only features common among all
IOMMU instances are supported. This is tracked in the global amd_iommu_efr
and amd_iommu_efr2, which should be used whenever the driver need to check
hardware capabilities.

Therefore, introduce check_feature() and check_feature2(), and modify
the driver to adopt the new helper functions.

In addition, clean up the print_iommu_info() to avoid reporting redundant
EFR/EFR2 for each IOMMU instance.

Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Vasant Hegde <vasant.hegde@amd.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Link: https://lore.kernel.org/r/20230921092147.5930-9-vasant.hegde@amd.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>

Author: ssuthiku-amd

drivers/iommu/amd/amd_iommu.h

@@ -87,9 +87,19 @@ static inline bool is_rd890_iommu(struct pci_dev *pdev)
 	       (pdev->device == PCI_DEVICE_ID_RD890_IOMMU);
 }
 
-static inline bool iommu_feature(struct amd_iommu *iommu, u64 mask)
+static inline bool check_feature(u64 mask)
 {
-	return !!(iommu->features & mask);
+	return (amd_iommu_efr & mask);
+}
+
+static inline bool check_feature2(u64 mask)
+{
+	return (amd_iommu_efr2 & mask);
+}
+
+static inline int check_feature_gpt_level(void)
+{
+	return ((amd_iommu_efr >> FEATURE_GATS_SHIFT) & FEATURE_GATS_MASK);
 }
 
 static inline u64 iommu_virt_to_phys(void *vaddr)
@@ -145,8 +155,5 @@ void amd_iommu_domain_set_pgtable(struct protection_domain *domain,
 				  u64 *root, int mode);
 struct dev_table_entry *get_dev_table(struct amd_iommu *iommu);
 
-extern u64 amd_iommu_efr;
-extern u64 amd_iommu_efr2;
-
 extern bool amd_iommu_snp_en;
 #endif

drivers/iommu/amd/amd_iommu_types.h

@@ -897,6 +897,10 @@ extern bool amd_iommu_force_isolation;
 /* Max levels of glxval supported */
 extern int amd_iommu_max_glx_val;
 
+/* Global EFR and EFR2 registers */
+extern u64 amd_iommu_efr;
+extern u64 amd_iommu_efr2;
+
 /*
  * This function flushes all internal caches of
  * the IOMMU used by this driver.

drivers/iommu/amd/init.c

@@ -270,7 +270,7 @@ int amd_iommu_get_num_iommus(void)
  * Iterate through all the IOMMUs to get common EFR
  * masks among all IOMMUs and warn if found inconsistency.
  */
-static void get_global_efr(void)
+static __init void get_global_efr(void)
 {
 	struct amd_iommu *iommu;
 
@@ -302,16 +302,6 @@ static void get_global_efr(void)
 	pr_info("Using global IVHD EFR:%#llx, EFR2:%#llx\n", amd_iommu_efr, amd_iommu_efr2);
 }
 
-static bool check_feature_on_all_iommus(u64 mask)
-{
-	return !!(amd_iommu_efr & mask);
-}
-
-static inline int check_feature_gpt_level(void)
-{
-	return ((amd_iommu_efr >> FEATURE_GATS_SHIFT) & FEATURE_GATS_MASK);
-}
-
 /*
  * For IVHD type 0x11/0x40, EFR is also available via IVHD.
  * Default to IVHD EFR since it is available sooner
@@ -397,7 +387,7 @@ static void iommu_set_cwwb_range(struct amd_iommu *iommu)
 	u64 start = iommu_virt_to_phys((void *)iommu->cmd_sem);
 	u64 entry = start & PM_ADDR_MASK;
 
-	if (!check_feature_on_all_iommus(FEATURE_SNP))
+	if (!check_feature(FEATURE_SNP))
 		return;
 
 	/* Note:
@@ -867,7 +857,7 @@ static void *__init iommu_alloc_4k_pages(struct amd_iommu *iommu,
 	void *buf = (void *)__get_free_pages(gfp, order);
 
 	if (buf &&
-	    check_feature_on_all_iommus(FEATURE_SNP) &&
+	    check_feature(FEATURE_SNP) &&
 	    set_memory_4k((unsigned long)buf, (1 << order))) {
 		free_pages((unsigned long)buf, order);
 		buf = NULL;
@@ -1046,7 +1036,7 @@ static void iommu_enable_xt(struct amd_iommu *iommu)
 
 static void iommu_enable_gt(struct amd_iommu *iommu)
 {
-	if (!iommu_feature(iommu, FEATURE_GT))
+	if (!check_feature(FEATURE_GT))
 		return;
 
 	iommu_feature_enable(iommu, CONTROL_GT_EN);
@@ -1985,7 +1975,7 @@ static void init_iommu_perf_ctr(struct amd_iommu *iommu)
 	u64 val;
 	struct pci_dev *pdev = iommu->dev;
 
-	if (!iommu_feature(iommu, FEATURE_PC))
+	if (!check_feature(FEATURE_PC))
 		return;
 
 	amd_iommu_pc_present = true;
@@ -2012,8 +2002,7 @@ static ssize_t amd_iommu_show_features(struct device *dev,
 				       struct device_attribute *attr,
 				       char *buf)
 {
-	struct amd_iommu *iommu = dev_to_amd_iommu(dev);
-	return sysfs_emit(buf, "%llx:%llx\n", iommu->features2, iommu->features);
+	return sysfs_emit(buf, "%llx:%llx\n", amd_iommu_efr, amd_iommu_efr2);
 }
 static DEVICE_ATTR(features, S_IRUGO, amd_iommu_show_features, NULL);
 
@@ -2049,22 +2038,22 @@ static void __init late_iommu_features_init(struct amd_iommu *iommu)
 	features = readq(iommu->mmio_base + MMIO_EXT_FEATURES);
 	features2 = readq(iommu->mmio_base + MMIO_EXT_FEATURES2);
 
-	if (!iommu->features) {
-		iommu->features = features;
-		iommu->features2 = features2;
+	if (!amd_iommu_efr) {
+		amd_iommu_efr = features;
+		amd_iommu_efr2 = features2;
 		return;
 	}
 
 	/*
 	 * Sanity check and warn if EFR values from
 	 * IVHD and MMIO conflict.
 	 */
-	if (features != iommu->features ||
-	    features2 != iommu->features2) {
+	if (features != amd_iommu_efr ||
+	    features2 != amd_iommu_efr2) {
 		pr_warn(FW_WARN
 			"EFR mismatch. Use IVHD EFR (%#llx : %#llx), EFR2 (%#llx : %#llx).\n",
-			features, iommu->features,
-			features2, iommu->features2);
+			features, amd_iommu_efr,
+			features2, amd_iommu_efr2);
 	}
 }
 
@@ -2090,20 +2079,20 @@ static int __init iommu_init_pci(struct amd_iommu *iommu)
 
 	late_iommu_features_init(iommu);
 
-	if (iommu_feature(iommu, FEATURE_GT)) {
+	if (check_feature(FEATURE_GT)) {
 		int glxval;
 		u32 max_pasid;
 		u64 pasmax;
 
-		pasmax = iommu->features & FEATURE_PASID_MASK;
+		pasmax = amd_iommu_efr & FEATURE_PASID_MASK;
 		pasmax >>= FEATURE_PASID_SHIFT;
 		max_pasid  = (1 << (pasmax + 1)) - 1;
 
 		amd_iommu_max_pasid = min(amd_iommu_max_pasid, max_pasid);
 
 		BUG_ON(amd_iommu_max_pasid & ~PASID_MASK);
 
-		glxval   = iommu->features & FEATURE_GLXVAL_MASK;
+		glxval   = amd_iommu_efr & FEATURE_GLXVAL_MASK;
 		glxval >>= FEATURE_GLXVAL_SHIFT;
 
 		if (amd_iommu_max_glx_val == -1)
@@ -2112,13 +2101,13 @@ static int __init iommu_init_pci(struct amd_iommu *iommu)
 			amd_iommu_max_glx_val = min(amd_iommu_max_glx_val, glxval);
 	}
 
-	if (iommu_feature(iommu, FEATURE_GT) &&
-	    iommu_feature(iommu, FEATURE_PPR)) {
+	if (check_feature(FEATURE_GT) &&
+	    check_feature(FEATURE_PPR)) {
 		iommu->is_iommu_v2   = true;
 		amd_iommu_v2_present = true;
 	}
 
-	if (iommu_feature(iommu, FEATURE_PPR) && alloc_ppr_log(iommu))
+	if (check_feature(FEATURE_PPR) && alloc_ppr_log(iommu))
 		return -ENOMEM;
 
 	if (iommu->cap & (1UL << IOMMU_CAP_NPCACHE)) {
@@ -2130,8 +2119,8 @@ static int __init iommu_init_pci(struct amd_iommu *iommu)
 	init_iommu_perf_ctr(iommu);
 
 	if (amd_iommu_pgtable == AMD_IOMMU_V2) {
-		if (!iommu_feature(iommu, FEATURE_GIOSUP) ||
-		    !iommu_feature(iommu, FEATURE_GT)) {
+		if (!check_feature(FEATURE_GIOSUP) ||
+		    !check_feature(FEATURE_GT)) {
 			pr_warn("Cannot enable v2 page table for DMA-API. Fallback to v1.\n");
 			amd_iommu_pgtable = AMD_IOMMU_V1;
 		}
@@ -2181,35 +2170,29 @@ static int __init iommu_init_pci(struct amd_iommu *iommu)
 
 static void print_iommu_info(void)
 {
+	int i;
 	static const char * const feat_str[] = {
 		"PreF", "PPR", "X2APIC", "NX", "GT", "[5]",
 		"IA", "GA", "HE", "PC"
 	};
-	struct amd_iommu *iommu;
-
-	for_each_iommu(iommu) {
-		struct pci_dev *pdev = iommu->dev;
-		int i;
-
-		pci_info(pdev, "Found IOMMU cap 0x%x\n", iommu->cap_ptr);
 
-		if (iommu->cap & (1 << IOMMU_CAP_EFR)) {
-			pr_info("Extended features (%#llx, %#llx):", iommu->features, iommu->features2);
+	if (amd_iommu_efr) {
+		pr_info("Extended features (%#llx, %#llx):", amd_iommu_efr, amd_iommu_efr2);
 
-			for (i = 0; i < ARRAY_SIZE(feat_str); ++i) {
-				if (iommu_feature(iommu, (1ULL << i)))
-					pr_cont(" %s", feat_str[i]);
-			}
+		for (i = 0; i < ARRAY_SIZE(feat_str); ++i) {
+			if (check_feature(1ULL << i))
+				pr_cont(" %s", feat_str[i]);
+		}
 
-			if (iommu->features & FEATURE_GAM_VAPIC)
-				pr_cont(" GA_vAPIC");
+		if (check_feature(FEATURE_GAM_VAPIC))
+			pr_cont(" GA_vAPIC");
 
-			if (iommu->features & FEATURE_SNP)
-				pr_cont(" SNP");
+		if (check_feature(FEATURE_SNP))
+			pr_cont(" SNP");
 
-			pr_cont("\n");
-		}
+		pr_cont("\n");
 	}
+
 	if (irq_remapping_enabled) {
 		pr_info("Interrupt remapping enabled\n");
 		if (amd_iommu_xt_mode == IRQ_REMAP_X2APIC_MODE)
@@ -2907,7 +2890,7 @@ static void enable_iommus_vapic(void)
 	}
 
 	if (AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir) &&
-	    !check_feature_on_all_iommus(FEATURE_GAM_VAPIC)) {
+	    !check_feature(FEATURE_GAM_VAPIC)) {
 		amd_iommu_guest_ir = AMD_IOMMU_GUEST_IR_LEGACY_GA;
 		return;
 	}
@@ -3819,7 +3802,7 @@ int amd_iommu_snp_enable(void)
 		return -EINVAL;
 	}
 
-	amd_iommu_snp_en = check_feature_on_all_iommus(FEATURE_SNP);
+	amd_iommu_snp_en = check_feature(FEATURE_SNP);
 	if (!amd_iommu_snp_en)
 		return -EINVAL;
 

drivers/iommu/amd/iommu.c

@@ -1295,7 +1295,7 @@ static void amd_iommu_flush_irt_all(struct amd_iommu *iommu)
 
 void iommu_flush_all_caches(struct amd_iommu *iommu)
 {
-	if (iommu_feature(iommu, FEATURE_IA)) {
+	if (check_feature(FEATURE_IA)) {
 		amd_iommu_flush_all(iommu);
 	} else {
 		amd_iommu_flush_dte_all(iommu);
@@ -1639,7 +1639,7 @@ static void set_dte_entry(struct amd_iommu *iommu, u16 devid,
 		flags |= DTE_FLAG_IOTLB;
 
 	if (ppr) {
-		if (iommu_feature(iommu, FEATURE_EPHSUP))
+		if (check_feature(FEATURE_EPHSUP))
 			pte_root |= 1ULL << DEV_ENTRY_PPR;
 	}