efHeterogeneousSynchronization temperatureMemoryDefinition.c 姚强 commit at 2020-11-27

efHeterogeneousSynchronization/dataSharedMemory/temperatureMemoryDefinition.c

@@ -255,3 +255,171 @@ struct pci_ops cia_pci_ops =
 void
 cia_pci_tbi(struct pci_controller *hose, dma_addr_t start, dma_addr_t end)
 {
+	wmb();
+	*(vip)CIA_IOC_PCI_TBIA = 3;	/* Flush all locked and unlocked.  */
+	mb();
+	*(vip)CIA_IOC_PCI_TBIA;
+}
+
+/*
+ * On PYXIS, even if the tbia works, we cannot use it. It effectively locks
+ * the chip (as well as direct write to the tag registers) if there is a
+ * SG DMA operation in progress. This is true at least for PYXIS rev. 1,
+ * so always use the method below.
+ */
+/*
+ * This is the method NT and NetBSD use.
+ *
+ * Allocate mappings, and put the chip into DMA loopback mode to read a
+ * garbage page.  This works by causing TLB misses, causing old entries to
+ * be purged to make room for the new entries coming in for the garbage page.
+ */
+
+#define CIA_BROKEN_TBIA_BASE	0x30000000
+#define CIA_BROKEN_TBIA_SIZE	1024
+
+/* Always called with interrupts disabled */
+void
+cia_pci_tbi_try2(struct pci_controller *hose,
+		 dma_addr_t start, dma_addr_t end)
+{
+	void __iomem *bus_addr;
+	int ctrl;
+
+	/* Put the chip into PCI loopback mode.  */
+	mb();
+	ctrl = *(vip)CIA_IOC_CIA_CTRL;
+	*(vip)CIA_IOC_CIA_CTRL = ctrl | CIA_CTRL_PCI_LOOP_EN;
+	mb();
+	*(vip)CIA_IOC_CIA_CTRL;
+	mb();
+
+	/* Read from PCI dense memory space at TBI_ADDR, skipping 32k on
+	   each read.  This forces SG TLB misses.  NetBSD claims that the
+	   TLB entries are not quite LRU, meaning that we need to read more
+	   times than there are actual tags.  The 2117x docs claim strict
+	   round-robin.  Oh well, we've come this far...  */
+	/* Even better - as seen on the PYXIS rev 1 the TLB tags 0-3 can
+	   be filled by the TLB misses *only once* after being invalidated
+	   (by tbia or direct write). Next misses won't update them even
+	   though the lock bits are cleared. Tags 4-7 are "quite LRU" though,
+	   so use them and read at window 3 base exactly 4 times. Reading
+	   more sometimes makes the chip crazy.  -ink */
+
+	bus_addr = cia_ioremap(CIA_BROKEN_TBIA_BASE, 32768 * 4);
+
+	cia_readl(bus_addr + 0x00000);
+	cia_readl(bus_addr + 0x08000);
+	cia_readl(bus_addr + 0x10000);
+	cia_readl(bus_addr + 0x18000);
+
+	cia_iounmap(bus_addr);
+
+	/* Restore normal PCI operation.  */
+	mb();
+	*(vip)CIA_IOC_CIA_CTRL = ctrl;
+	mb();
+	*(vip)CIA_IOC_CIA_CTRL;
+	mb();
+}
+
+static inline void
+cia_prepare_tbia_workaround(int window)
+{
+	unsigned long *ppte, pte;
+	long i;
+
+	/* Use minimal 1K map. */
+	ppte = __alloc_bootmem(CIA_BROKEN_TBIA_SIZE, 32768, 0);
+	pte = (virt_to_phys(ppte) >> (PAGE_SHIFT - 1)) | 1;
+
+	for (i = 0; i < CIA_BROKEN_TBIA_SIZE / sizeof(unsigned long); ++i)
+		ppte[i] = pte;
+
+	*(vip)CIA_IOC_PCI_Wn_BASE(window) = CIA_BROKEN_TBIA_BASE | 3;
+	*(vip)CIA_IOC_PCI_Wn_MASK(window)
+	  = (CIA_BROKEN_TBIA_SIZE*1024 - 1) & 0xfff00000;
+	*(vip)CIA_IOC_PCI_Tn_BASE(window) = virt_to_phys(ppte) >> 2;
+}
+
+static void __init
+verify_tb_operation(void)
+{
+	static int page[PAGE_SIZE/4]
+		__attribute__((aligned(PAGE_SIZE)))
+		__initdata = { 0 };
+
+	struct pci_iommu_arena *arena = pci_isa_hose->sg_isa;
+	int ctrl, addr0, tag0, pte0, data0;
+	int temp, use_tbia_try2 = 0;
+	void __iomem *bus_addr;
+
+	/* pyxis -- tbia is broken */
+	if (pci_isa_hose->dense_io_base)
+		use_tbia_try2 = 1;
+
+	/* Put the chip into PCI loopback mode.  */
+	mb();
+	ctrl = *(vip)CIA_IOC_CIA_CTRL;
+	*(vip)CIA_IOC_CIA_CTRL = ctrl | CIA_CTRL_PCI_LOOP_EN;
+	mb();
+	*(vip)CIA_IOC_CIA_CTRL;
+	mb();
+
+	/* Write a valid entry directly into the TLB registers.  */
+
+	addr0 = arena->dma_base;
+	tag0 = addr0 | 1;
+	pte0 = (virt_to_phys(page) >> (PAGE_SHIFT - 1)) | 1;
+
+	*(vip)CIA_IOC_TB_TAGn(0) = tag0;
+	*(vip)CIA_IOC_TB_TAGn(1) = 0;
+	*(vip)CIA_IOC_TB_TAGn(2) = 0;
+	*(vip)CIA_IOC_TB_TAGn(3) = 0;
+	*(vip)CIA_IOC_TB_TAGn(4) = 0;
+	*(vip)CIA_IOC_TB_TAGn(5) = 0;
+	*(vip)CIA_IOC_TB_TAGn(6) = 0;
+	*(vip)CIA_IOC_TB_TAGn(7) = 0;
+	*(vip)CIA_IOC_TBn_PAGEm(0,0) = pte0;
+	*(vip)CIA_IOC_TBn_PAGEm(0,1) = 0;
+	*(vip)CIA_IOC_TBn_PAGEm(0,2) = 0;
+	*(vip)CIA_IOC_TBn_PAGEm(0,3) = 0;
+	mb();
+
+	/* Get a usable bus address */
+	bus_addr = cia_ioremap(addr0, 8*PAGE_SIZE);
+
+	/* First, verify we can read back what we've written.  If
+	   this fails, we can't be sure of any of the other testing
+	   we're going to do, so bail.  */
+	/* ??? Actually, we could do the work with machine checks.
+	   By passing this register update test, we pretty much
+	   guarantee that cia_pci_tbi_try1 works.  If this test
+	   fails, cia_pci_tbi_try2 might still work.  */
+
+	temp = *(vip)CIA_IOC_TB_TAGn(0);
+	if (temp != tag0) {
+		printk("pci: failed tb register update test "
+		       "(tag0 %#x != %#x)\n", temp, tag0);
+		goto failed;
+	}
+	temp = *(vip)CIA_IOC_TB_TAGn(1);
+	if (temp != 0) {
+		printk("pci: failed tb register update test "
+		       "(tag1 %#x != 0)\n", temp);
+		goto failed;
+	}
+	temp = *(vip)CIA_IOC_TBn_PAGEm(0,0);
+	if (temp != pte0) {
+		printk("pci: failed tb register update test "
+		       "(pte0 %#x != %#x)\n", temp, pte0);
+		goto failed;
+	}
+	printk("pci: passed tb register update test\n");
+
+	/* Second, verify we can actually do I/O through this entry.  */
+
+	data0 = 0xdeadbeef;
+	page[0] = data0;
+	mcheck_expected(0) = 1;
+	mcheck_taken(0) = 0;