s390/pci: Implement ioremap_wc/prot() with MIO

With our current support for the new MIO PCI instructions, write
combining/write back MMIO memory can be obtained via the pci_iomap_wc()
and pci_iomap_wc_range() functions.
This is achieved by using the write back address for a specific bar
as provided in clp_store_query_pci_fn()

These functions are however not widely used and instead drivers often
rely on ioremap_wc() and ioremap_prot(), which on other platforms enable
write combining using a PTE flag set through the pgrprot value.

While we do not have a write combining flag in the low order flag bits
of the PTE like x86_64 does, with MIO support, there is a write back bit
in the physical address (bit 1 on z15) and thus also the PTE.
Which bit is used to toggle write back and whether it is available at
all, is however not fixed in the architecture. Instead we get this
information from the CLP Store Logical Processor Characteristics for PCI
command. When the write back bit is not provided we fall back to the
existing behavior.
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Reviewed-by: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>

Documentation/features/vm/ioremap_prot/arch-support.txt

@@ -24,7 +24,7 @@
     |      parisc: | TODO |
     |     powerpc: |  ok  |
     |       riscv: | TODO |
-    |        s390: | TODO |
+    |        s390: |  ok  |
     |          sh: |  ok  |
     |       sparc: | TODO |
     |          um: | TODO |

arch/s390/include/asm/clp.h

@@ -5,6 +5,9 @@
 /* CLP common request & response block size */
 #define CLP_BLK_SIZE			PAGE_SIZE
 
+/* Call Logical Processor - Command Code */
+#define CLP_SLPC		0x0001
+
 #define CLP_LPS_BASE	0
 #define CLP_LPS_PCI	2
 

arch/s390/include/asm/io.h

@@ -12,6 +12,7 @@
 
 #include <linux/kernel.h>
 #include <asm/page.h>
+#include <asm/pgtable.h>
 #include <asm/pci_io.h>
 
 #define xlate_dev_mem_ptr xlate_dev_mem_ptr
@@ -26,7 +27,10 @@ void unxlate_dev_mem_ptr(phys_addr_t phys, void *addr);
 
 #define IO_SPACE_LIMIT 0
 
+void __iomem *ioremap_prot(phys_addr_t addr, size_t size, unsigned long prot);
 void __iomem *ioremap(phys_addr_t addr, size_t size);
+void __iomem *ioremap_wc(phys_addr_t addr, size_t size);
+void __iomem *ioremap_wt(phys_addr_t addr, size_t size);
 void iounmap(volatile void __iomem *addr);
 
 static inline void __iomem *ioport_map(unsigned long port, unsigned int nr)
@@ -52,6 +56,10 @@ static inline void ioport_unmap(void __iomem *p)
 #define pci_iomap_wc pci_iomap_wc
 #define pci_iomap_wc_range pci_iomap_wc_range
 
+#define ioremap ioremap
+#define ioremap_wt ioremap_wt
+#define ioremap_wc ioremap_wc
+
 #define memcpy_fromio(dst, src, count)	zpci_memcpy_fromio(dst, src, count)
 #define memcpy_toio(dst, src, count)	zpci_memcpy_toio(dst, src, count)
 #define memset_io(dst, val, count)	zpci_memset_io(dst, val, count)

arch/s390/include/asm/pci.h

@@ -208,6 +208,7 @@ int zpci_unregister_ioat(struct zpci_dev *, u8);
 void zpci_remove_reserved_devices(void);
 
 /* CLP */
+int clp_setup_writeback_mio(void);
 int clp_scan_pci_devices(void);
 int clp_rescan_pci_devices(void);
 int clp_rescan_pci_devices_simple(u32 *fid);

arch/s390/include/asm/pci_clp.h

@@ -7,6 +7,7 @@
 /*
  * Call Logical Processor - Command Codes
  */
+#define CLP_SLPC		0x0001
 #define CLP_LIST_PCI		0x0002
 #define CLP_QUERY_PCI_FN	0x0003
 #define CLP_QUERY_PCI_FNGRP	0x0004
@@ -51,6 +52,19 @@ struct clp_fh_list_entry {
 
 extern bool zpci_unique_uid;
 
+struct clp_rsp_slpc_pci {
+	struct clp_rsp_hdr hdr;
+	u32 reserved2[4];
+	u32 lpif[8];
+	u32 reserved3[4];
+	u32 vwb		:  1;
+	u32		:  1;
+	u32 mio_wb	:  6;
+	u32		: 24;
+	u32 reserved5[3];
+	u32 lpic[8];
+} __packed;
+
 /* List PCI functions request */
 struct clp_req_list_pci {
 	struct clp_req_hdr hdr;
@@ -172,6 +186,11 @@ struct clp_rsp_set_pci {
 } __packed;
 
 /* Combined request/response block structures used by clp insn */
+struct clp_req_rsp_slpc_pci {
+	struct clp_req_slpc request;
+	struct clp_rsp_slpc_pci response;
+} __packed;
+
 struct clp_req_rsp_list_pci {
 	struct clp_req_list_pci request;
 	struct clp_rsp_list_pci response;

arch/s390/include/asm/pgtable.h

@@ -1186,6 +1186,12 @@ void gmap_pmdp_invalidate(struct mm_struct *mm, unsigned long vmaddr);
 void gmap_pmdp_idte_local(struct mm_struct *mm, unsigned long vmaddr);
 void gmap_pmdp_idte_global(struct mm_struct *mm, unsigned long vmaddr);
 
+#define pgprot_writecombine	pgprot_writecombine
+pgprot_t pgprot_writecombine(pgprot_t prot);
+
+#define pgprot_writethrough	pgprot_writethrough
+pgprot_t pgprot_writethrough(pgprot_t prot);
+
 /*
  * Certain architectures need to do special things when PTEs
  * within a page table are directly modified.  Thus, the following
@@ -1209,7 +1215,8 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
 static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot)
 {
 	pte_t __pte;
-	pte_val(__pte) = physpage + pgprot_val(pgprot);
+
+	pte_val(__pte) = physpage | pgprot_val(pgprot);
 	if (!MACHINE_HAS_NX)
 		pte_val(__pte) &= ~_PAGE_NOEXEC;
 	return pte_mkyoung(__pte);

arch/s390/include/asm/setup.h

@@ -94,6 +94,9 @@ extern unsigned long vmalloc_size;
 extern unsigned long max_physmem_end;
 extern unsigned long __swsusp_reset_dma;
 
+/* The Write Back bit position in the physaddr is given by the SLPC PCI */
+extern unsigned long mio_wb_bit_mask;
+
 #define MACHINE_IS_VM		(S390_lowcore.machine_flags & MACHINE_FLAG_VM)
 #define MACHINE_IS_KVM		(S390_lowcore.machine_flags & MACHINE_FLAG_KVM)
 #define MACHINE_IS_LPAR		(S390_lowcore.machine_flags & MACHINE_FLAG_LPAR)

arch/s390/kernel/setup.c

@@ -127,6 +127,12 @@ unsigned long MODULES_END;
 struct lowcore *lowcore_ptr[NR_CPUS];
 EXPORT_SYMBOL(lowcore_ptr);
 
+/*
+ * The Write Back bit position in the physaddr is given by the SLPC PCI.
+ * Leaving the mask zero always uses write through which is safe
+ */
+unsigned long mio_wb_bit_mask __ro_after_init;
+
 /*
  * This is set up by the setup-routine at boot-time
  * for S390 need to find out, what we have to setup

arch/s390/mm/pgtable.c

@@ -24,6 +24,26 @@
 #include <asm/mmu_context.h>
 #include <asm/page-states.h>
 
+pgprot_t pgprot_writecombine(pgprot_t prot)
+{
+	/*
+	 * mio_wb_bit_mask may be set on a different CPU, but it is only set
+	 * once at init and only read afterwards.
+	 */
+	return __pgprot(pgprot_val(prot) | mio_wb_bit_mask);
+}
+EXPORT_SYMBOL_GPL(pgprot_writecombine);
+
+pgprot_t pgprot_writethrough(pgprot_t prot)
+{
+	/*
+	 * mio_wb_bit_mask may be set on a different CPU, but it is only set
+	 * once at init and only read afterwards.
+	 */
+	return __pgprot(pgprot_val(prot) & ~mio_wb_bit_mask);
+}
+EXPORT_SYMBOL_GPL(pgprot_writethrough);
+
 static inline void ptep_ipte_local(struct mm_struct *mm, unsigned long addr,
 				   pte_t *ptep, int nodat)
 {

arch/s390/pci/pci.c

@@ -226,7 +226,7 @@ void __iowrite64_copy(void __iomem *to, const void *from, size_t count)
        zpci_memcpy_toio(to, from, count);
 }
 
-void __iomem *ioremap(phys_addr_t addr, size_t size)
+static void __iomem *__ioremap(phys_addr_t addr, size_t size, pgprot_t prot)
 {
 	unsigned long offset, vaddr;
 	struct vm_struct *area;
@@ -247,14 +247,37 @@ void __iomem *ioremap(phys_addr_t addr, size_t size)
 		return NULL;
 
 	vaddr = (unsigned long) area->addr;
-	if (ioremap_page_range(vaddr, vaddr + size, addr, PAGE_KERNEL)) {
+	if (ioremap_page_range(vaddr, vaddr + size, addr, prot)) {
 		free_vm_area(area);
 		return NULL;
 	}
 	return (void __iomem *) ((unsigned long) area->addr + offset);
 }
+
+void __iomem *ioremap_prot(phys_addr_t addr, size_t size, unsigned long prot)
+{
+	return __ioremap(addr, size, __pgprot(prot));
+}
+EXPORT_SYMBOL(ioremap_prot);
+
+void __iomem *ioremap(phys_addr_t addr, size_t size)
+{
+	return __ioremap(addr, size, PAGE_KERNEL);
+}
 EXPORT_SYMBOL(ioremap);
 
+void __iomem *ioremap_wc(phys_addr_t addr, size_t size)
+{
+	return __ioremap(addr, size, pgprot_writecombine(PAGE_KERNEL));
+}
+EXPORT_SYMBOL(ioremap_wc);
+
+void __iomem *ioremap_wt(phys_addr_t addr, size_t size)
+{
+	return __ioremap(addr, size, pgprot_writethrough(PAGE_KERNEL));
+}
+EXPORT_SYMBOL(ioremap_wt);
+
 void iounmap(volatile void __iomem *addr)
 {
 	if (static_branch_likely(&have_mio))
@@ -784,6 +807,9 @@ static int zpci_mem_init(void)
 	if (!zpci_iomap_bitmap)
 		goto error_iomap_bitmap;
 
+	if (static_branch_likely(&have_mio))
+		clp_setup_writeback_mio();
+
 	return 0;
 error_iomap_bitmap:
 	kfree(zpci_iomap_start);

arch/s390/pci/pci_clp.c

@@ -292,6 +292,40 @@ static int clp_set_pci_fn(struct zpci_dev *zdev, u8 nr_dma_as, u8 command)
 	return rc;
 }
 
+int clp_setup_writeback_mio(void)
+{
+	struct clp_req_rsp_slpc_pci *rrb;
+	u8  wb_bit_pos;
+	int rc;
+
+	rrb = clp_alloc_block(GFP_KERNEL);
+	if (!rrb)
+		return -ENOMEM;
+
+	memset(rrb, 0, sizeof(*rrb));
+	rrb->request.hdr.len = sizeof(rrb->request);
+	rrb->request.hdr.cmd = CLP_SLPC;
+	rrb->response.hdr.len = sizeof(rrb->response);
+
+	rc = clp_req(rrb, CLP_LPS_PCI);
+	if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) {
+		if (rrb->response.vwb) {
+			wb_bit_pos = rrb->response.mio_wb;
+			set_bit_inv(wb_bit_pos, &mio_wb_bit_mask);
+			zpci_dbg(3, "wb bit: %d\n", wb_bit_pos);
+		} else {
+			zpci_dbg(3, "wb bit: n.a.\n");
+		}
+
+	} else {
+		zpci_err("SLPC PCI:\n");
+		zpci_err_clp(rrb->response.hdr.rsp, rc);
+		rc = -EIO;
+	}
+	clp_free_block(rrb);
+	return rc;
+}
+
 int clp_enable_fh(struct zpci_dev *zdev, u8 nr_dma_as)
 {
 	int rc;
@@ -495,7 +529,7 @@ static int clp_base_command(struct clp_req *req, struct clp_req_hdr *lpcb)
 	}
 }
 
-static int clp_pci_slpc(struct clp_req *req, struct clp_req_rsp_slpc *lpcb)
+static int clp_pci_slpc(struct clp_req *req, struct clp_req_rsp_slpc_pci *lpcb)
 {
 	unsigned long limit = PAGE_SIZE - sizeof(lpcb->request);