Automatic merge of 'next' into merge (2024-06-17 15:38)

arch/powerpc/include/asm/iommu.h

@@ -31,6 +31,8 @@
 #define DIRECT64_PROPNAME "linux,direct64-ddr-window-info"
 #define DMA64_PROPNAME "linux,dma64-ddr-window-info"
 
+#define	MIN_DDW_VPMEM_DMA_WINDOW	SZ_2G
+
 /* Boot time flags */
 extern int iommu_is_off;
 extern int iommu_force_on;

arch/powerpc/include/asm/plpar_wrappers.h

@@ -18,16 +18,6 @@ static inline long poll_pending(void)
 	return plpar_hcall_norets(H_POLL_PENDING);
 }
 
-static inline u8 get_cede_latency_hint(void)
-{
-	return get_lppaca()->cede_latency_hint;
-}
-
-static inline void set_cede_latency_hint(u8 latency_hint)
-{
-	get_lppaca()->cede_latency_hint = latency_hint;
-}
-
 static inline long cede_processor(void)
 {
 	/*
@@ -37,24 +27,6 @@ static inline long cede_processor(void)
 	return plpar_hcall_norets_notrace(H_CEDE);
 }
 
-static inline long extended_cede_processor(unsigned long latency_hint)
-{
-	long rc;
-	u8 old_latency_hint = get_cede_latency_hint();
-
-	set_cede_latency_hint(latency_hint);
-
-	rc = cede_processor();
-
-	/* Ensure that H_CEDE returns with IRQs on */
-	if (WARN_ON(IS_ENABLED(CONFIG_PPC_IRQ_SOFT_MASK_DEBUG) && !(mfmsr() & MSR_EE)))
-		__hard_irq_enable();
-
-	set_cede_latency_hint(old_latency_hint);
-
-	return rc;
-}
-
 static inline long vpa_call(unsigned long flags, unsigned long cpu,
 		unsigned long vpa)
 {

arch/powerpc/mm/drmem.c

@@ -491,10 +491,8 @@ static int __init drmem_init(void)
 	const __be32 *prop;
 
 	dn = of_find_node_by_path("/ibm,dynamic-reconfiguration-memory");
-	if (!dn) {
-		pr_info("No dynamic reconfiguration memory found\n");
+	if (!dn)
 		return 0;
-	}
 
 	if (init_drmem_lmb_size(dn)) {
 		of_node_put(dn);

arch/powerpc/mm/numa.c

@@ -896,7 +896,7 @@ static int __init numa_setup_drmem_lmb(struct drmem_lmb *lmb,
 
 static int __init parse_numa_properties(void)
 {
-	struct device_node *memory;
+	struct device_node *memory, *pci;
 	int default_nid = 0;
 	unsigned long i;
 	const __be32 *associativity;
@@ -1010,6 +1010,18 @@ static int __init parse_numa_properties(void)
 			goto new_range;
 	}
 
+	for_each_node_by_name(pci, "pci") {
+		int nid = NUMA_NO_NODE;
+
+		associativity = of_get_associativity(pci);
+		if (associativity) {
+			nid = associativity_to_nid(associativity);
+			initialize_form1_numa_distance(associativity);
+		}
+		if (likely(nid >= 0) && !node_online(nid))
+			node_set_online(nid);
+	}
+
 	/*
 	 * Now do the same thing for each MEMBLOCK listed in the
 	 * ibm,dynamic-memory property in the

arch/powerpc/platforms/pseries/iommu.c

@@ -1304,14 +1304,15 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
 	struct ddw_query_response query;
 	struct ddw_create_response create;
 	int page_shift;
-	u64 win_addr;
+	u64 win_addr, dynamic_offset = 0;
 	const char *win_name;
 	struct device_node *dn;
 	u32 ddw_avail[DDW_APPLICABLE_SIZE];
 	struct dma_win *window;
 	struct property *win64;
 	struct failed_ddw_pdn *fpdn;
 	bool default_win_removed = false, direct_mapping = false;
+	bool dynamic_mapping = false;
 	bool pmem_present;
 	struct pci_dn *pci = PCI_DN(pdn);
 	struct property *default_win = NULL;
@@ -1407,7 +1408,6 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
 		goto out_failed;
 	}
 
-
 	/*
 	 * The "ibm,pmemory" can appear anywhere in the address space.
 	 * Assuming it is still backed by page structs, try MAX_PHYSMEM_BITS
@@ -1432,14 +1432,42 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
 			1ULL << page_shift);
 
 		len = order_base_2(query.largest_available_block << page_shift);
-		win_name = DMA64_PROPNAME;
+
+		dynamic_mapping = true;
 	} else {
 		direct_mapping = !default_win_removed ||
 			(len == MAX_PHYSMEM_BITS) ||
 			(!pmem_present && (len == max_ram_len));
-		win_name = direct_mapping ? DIRECT64_PROPNAME : DMA64_PROPNAME;
+
+		/* DDW is big enough to direct map RAM. If there is vPMEM, check
+		 * if enough space is left in DDW where we can dynamically
+		 * allocate TCEs for vPMEM. For now, this Hybrid sharing of DDW
+		 * is only for SR-IOV devices.
+		 */
+		if (default_win_removed && pmem_present && !direct_mapping) {
+			/* DDW is big enough to be split */
+			if ((query.largest_available_block << page_shift) >=
+			     MIN_DDW_VPMEM_DMA_WINDOW + (1ULL << max_ram_len)) {
+				direct_mapping = true;
+
+				/* offset of the Dynamic part of DDW */
+				dynamic_offset = 1ULL << max_ram_len;
+			}
+
+			/* DDW will at least have dynamic allocation */
+			dynamic_mapping = true;
+
+			/* create max size DDW possible */
+			len = order_base_2(query.largest_available_block
+							<< page_shift);
+		}
 	}
 
+	/* Even if the DDW is split into both direct mapped RAM and dynamically
+	 * mapped vPMEM, the DDW property in OF will be marked as Direct.
+	 */
+	win_name = direct_mapping ? DIRECT64_PROPNAME : DMA64_PROPNAME;
+
 	ret = create_ddw(dev, ddw_avail, &create, page_shift, len);
 	if (ret != 0)
 		goto out_failed;
@@ -1467,9 +1495,9 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
 	if (!window)
 		goto out_del_prop;
 
-	if (direct_mapping) {
-		window->direct = true;
+	window->direct = direct_mapping;
 
+	if (direct_mapping) {
 		/* DDW maps the whole partition, so enable direct DMA mapping */
 		ret = walk_system_ram_range(0, memblock_end_of_DRAM() >> PAGE_SHIFT,
 					    win64->value, tce_setrange_multi_pSeriesLP_walk);
@@ -1481,12 +1509,13 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
 			clean_dma_window(pdn, win64->value);
 			goto out_del_list;
 		}
-	} else {
+	}
+
+	if (dynamic_mapping) {
 		struct iommu_table *newtbl;
 		int i;
 		unsigned long start = 0, end = 0;
-
-		window->direct = false;
+		u64 dynamic_addr, dynamic_len;
 
 		for (i = 0; i < ARRAY_SIZE(pci->phb->mem_resources); i++) {
 			const unsigned long mask = IORESOURCE_MEM_64 | IORESOURCE_MEM;
@@ -1506,8 +1535,15 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
 			goto out_del_list;
 		}
 
-		iommu_table_setparms_common(newtbl, pci->phb->bus->number, create.liobn, win_addr,
-					    1UL << len, page_shift, NULL, &iommu_table_lpar_multi_ops);
+		/* If the DDW is split between directly mapped RAM and Dynamic
+		 * mapped for TCES, offset into the DDW where the dynamic part
+		 * begins.
+		 */
+		dynamic_addr = win_addr + dynamic_offset;
+		dynamic_len = (1UL << len) - dynamic_offset;
+		iommu_table_setparms_common(newtbl, pci->phb->bus->number, create.liobn,
+					    dynamic_addr, dynamic_len, page_shift, NULL,
+					    &iommu_table_lpar_multi_ops);
 		iommu_init_table(newtbl, pci->phb->node, start, end);
 
 		pci->table_group->tables[1] = newtbl;
@@ -1559,13 +1595,12 @@ static bool enable_ddw(struct pci_dev *dev, struct device_node *pdn)
 out_unlock:
 	mutex_unlock(&dma_win_init_mutex);
 
-	/*
-	 * If we have persistent memory and the window size is only as big
-	 * as RAM, then we failed to create a window to cover persistent
-	 * memory and need to set the DMA limit.
+	/* If we have persistent memory and the window size is not big enough
+	 * to directly map both RAM and vPMEM, then we need to set DMA limit.
 	 */
-	if (pmem_present && direct_mapping && len == max_ram_len)
-		dev->dev.bus_dma_limit = dev->dev.archdata.dma_offset + (1ULL << len);
+	if (pmem_present && direct_mapping && len != MAX_PHYSMEM_BITS)
+		dev->dev.bus_dma_limit = dev->dev.archdata.dma_offset +
+						(1ULL << max_ram_len);
 
 	return direct_mapping;
 }

arch/powerpc/platforms/pseries/pci_dlpar.c

@@ -11,6 +11,7 @@
 
 #include <linux/pci.h>
 #include <linux/export.h>
+#include <linux/node.h>
 #include <asm/pci-bridge.h>
 #include <asm/ppc-pci.h>
 #include <asm/firmware.h>
@@ -21,9 +22,22 @@
 struct pci_controller *init_phb_dynamic(struct device_node *dn)
 {
 	struct pci_controller *phb;
+	int nid;
 
 	pr_debug("PCI: Initializing new hotplug PHB %pOF\n", dn);
 
+	nid = of_node_to_nid(dn);
+	if (likely((nid) >= 0)) {
+		if (!node_online(nid)) {
+			if (__register_one_node(nid)) {
+				pr_err("PCI: Failed to register node %d\n", nid);
+			} else {
+				update_numa_distance(dn);
+				node_set_online(nid);
+			}
+		}
+	}
+
 	phb = pcibios_alloc_controller(dn);
 	if (!phb)
 		return NULL;

arch/powerpc/platforms/pseries/vas.c

@@ -38,7 +38,27 @@ static long hcall_return_busy_check(long rc)
 {
 	/* Check if we are stalled for some time */
 	if (H_IS_LONG_BUSY(rc)) {
-		msleep(get_longbusy_msecs(rc));
+		unsigned int ms;
+		/*
+		 * Allocate, Modify and Deallocate HCALLs returns
+		 * H_LONG_BUSY_ORDER_1_MSEC or H_LONG_BUSY_ORDER_10_MSEC
+		 * for the long delay. So the sleep time should always
+		 * be either 1 or 10msecs, but in case if the HCALL
+		 * returns the long delay > 10 msecs, clamp the sleep
+		 * time to 10msecs.
+		 */
+		ms = clamp(get_longbusy_msecs(rc), 1, 10);
+
+		/*
+		 * msleep() will often sleep at least 20 msecs even
+		 * though the hypervisor suggests that the OS reissue
+		 * HCALLs after 1 or 10msecs. Also the delay hint from
+		 * the HCALL is just a suggestion. So OK to pause for
+		 * less time than the hinted delay. Use usleep_range()
+		 * to ensure we don't sleep much longer than actually
+		 * needed.
+		 */
+		usleep_range(ms * (USEC_PER_MSEC / 10), ms * USEC_PER_MSEC);
 		rc = H_BUSY;
 	} else if (rc == H_BUSY) {
 		cond_resched();