[RFC 14/16] NOVA: Read-only pmem devices

From: Steven Swanson
Date: Thu Aug 03 2017 - 03:50:43 EST

Add (and implement) a module command line option to nd_pmem to support read-only pmem devices.

Signed-off-by: Steven Swanson <swanson@xxxxxxxxxxx>
---
arch/x86/include/asm/io.h | 1 +
arch/x86/mm/ioremap.c | 25 ++++++++++++++++++-------
drivers/nvdimm/pmem.c | 14 ++++++++++++--
include/linux/io.h | 2 ++
kernel/memremap.c | 24 ++++++++++++++++++++++++
mm/memory.c | 2 +-
mm/mmap.c | 1 +
7 files changed, 59 insertions(+), 10 deletions(-)

diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h
index 7afb0e2f07f4..7aae48f2e4f1 100644
--- a/arch/x86/include/asm/io.h
+++ b/arch/x86/include/asm/io.h
@@ -173,6 +173,7 @@ extern void __iomem *ioremap_uc(resource_size_t offset, unsigned long size);
#define ioremap_uc ioremap_uc

extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size);
+extern void __iomem *ioremap_cache_ro(resource_size_t phys_addr, unsigned long size);
extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size, unsigned long prot_val);

/**
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index bbc558b88a88..bcd473801817 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -81,7 +81,8 @@ static int __ioremap_check_ram(unsigned long start_pfn, unsigned long nr_pages,
* caller shouldn't need to know that small detail.
*/
static void __iomem *__ioremap_caller(resource_size_t phys_addr,
- unsigned long size, enum page_cache_mode pcm, void *caller)
+ unsigned long size, enum page_cache_mode pcm, void *caller,
+ int readonly)
{
unsigned long offset, vaddr;
resource_size_t pfn, last_pfn, last_addr;
@@ -172,6 +173,9 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr,
break;
}

+ if (readonly)
+ prot = __pgprot((unsigned long)prot.pgprot & ~_PAGE_RW);
+
/*
* Ok, go for it..
*/
@@ -239,7 +243,7 @@ void __iomem *ioremap_nocache(resource_size_t phys_addr, unsigned long size)
enum page_cache_mode pcm = _PAGE_CACHE_MODE_UC_MINUS;

return __ioremap_caller(phys_addr, size, pcm,
- __builtin_return_address(0));
+ __builtin_return_address(0), 0);
}
EXPORT_SYMBOL(ioremap_nocache);

@@ -272,7 +276,7 @@ void __iomem *ioremap_uc(resource_size_t phys_addr, unsigned long size)
enum page_cache_mode pcm = _PAGE_CACHE_MODE_UC;

return __ioremap_caller(phys_addr, size, pcm,
- __builtin_return_address(0));
+ __builtin_return_address(0), 0);
}
EXPORT_SYMBOL_GPL(ioremap_uc);

@@ -289,7 +293,7 @@ EXPORT_SYMBOL_GPL(ioremap_uc);
void __iomem *ioremap_wc(resource_size_t phys_addr, unsigned long size)
{
return __ioremap_caller(phys_addr, size, _PAGE_CACHE_MODE_WC,
- __builtin_return_address(0));
+ __builtin_return_address(0), 0);
}
EXPORT_SYMBOL(ioremap_wc);

@@ -306,23 +310,30 @@ EXPORT_SYMBOL(ioremap_wc);
void __iomem *ioremap_wt(resource_size_t phys_addr, unsigned long size)
{
return __ioremap_caller(phys_addr, size, _PAGE_CACHE_MODE_WT,
- __builtin_return_address(0));
+ __builtin_return_address(0), 0);
}
EXPORT_SYMBOL(ioremap_wt);

void __iomem *ioremap_cache(resource_size_t phys_addr, unsigned long size)
{
return __ioremap_caller(phys_addr, size, _PAGE_CACHE_MODE_WB,
- __builtin_return_address(0));
+ __builtin_return_address(0), 0);
}
EXPORT_SYMBOL(ioremap_cache);

+void __iomem *ioremap_cache_ro(resource_size_t phys_addr, unsigned long size)
+{
+ return __ioremap_caller(phys_addr, size, _PAGE_CACHE_MODE_WB,
+ __builtin_return_address(0), 1);
+}
+EXPORT_SYMBOL(ioremap_cache_ro);
+
void __iomem *ioremap_prot(resource_size_t phys_addr, unsigned long size,
unsigned long prot_val)
{
return __ioremap_caller(phys_addr, size,
pgprot2cachemode(__pgprot(prot_val)),
- __builtin_return_address(0));
+ __builtin_return_address(0), 0);
}
EXPORT_SYMBOL(ioremap_prot);

diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
index c544d466ea51..a6b29c731c53 100644
--- a/drivers/nvdimm/pmem.c
+++ b/drivers/nvdimm/pmem.c
@@ -35,6 +35,11 @@
#include "pfn.h"
#include "nd.h"

+int readonly;
+
+module_param(readonly, int, S_IRUGO);
+MODULE_PARM_DESC(readonly, "Mount readonly");
+
static struct device *to_dev(struct pmem_device *pmem)
{
/*
@@ -324,9 +329,14 @@ static int pmem_attach_disk(struct device *dev,
addr = devm_memremap_pages(dev, &nsio->res,
&q->q_usage_counter, NULL);
pmem->pfn_flags |= PFN_MAP;
- } else
- addr = devm_memremap(dev, pmem->phys_addr,
+ } else {
+ if (readonly == 0)
+ addr = devm_memremap(dev, pmem->phys_addr,
pmem->size, ARCH_MEMREMAP_PMEM);
+ else
+ addr = devm_memremap_ro(dev, pmem->phys_addr,
+ pmem->size, ARCH_MEMREMAP_PMEM);
+ }

/*
* At release time the queue must be frozen before
diff --git a/include/linux/io.h b/include/linux/io.h
index 2195d9ea4aaa..00641aef9ab3 100644
--- a/include/linux/io.h
+++ b/include/linux/io.h
@@ -86,6 +86,8 @@ void devm_ioremap_release(struct device *dev, void *res);

void *devm_memremap(struct device *dev, resource_size_t offset,
size_t size, unsigned long flags);
+void *devm_memremap_ro(struct device *dev, resource_size_t offset,
+ size_t size, unsigned long flags);
void devm_memunmap(struct device *dev, void *addr);

void *__devm_memremap_pages(struct device *dev, struct resource *res);
diff --git a/kernel/memremap.c b/kernel/memremap.c
index 23a6483c3666..68371a9a40e5 100644
--- a/kernel/memremap.c
+++ b/kernel/memremap.c
@@ -162,6 +162,30 @@ void *devm_memremap(struct device *dev, resource_size_t offset,
}
EXPORT_SYMBOL(devm_memremap);

+void *devm_memremap_ro(struct device *dev, resource_size_t offset,
+ size_t size, unsigned long flags)
+{
+ void **ptr, *addr;
+
+ printk("%s\n", __func__);
+ ptr = devres_alloc_node(devm_memremap_release, sizeof(*ptr), GFP_KERNEL,
+ dev_to_node(dev));
+ if (!ptr)
+ return ERR_PTR(-ENOMEM);
+
+ addr = ioremap_cache_ro(offset, size);
+ if (addr) {
+ *ptr = addr;
+ devres_add(dev, ptr);
+ } else {
+ devres_free(ptr);
+ return ERR_PTR(-ENXIO);
+ }
+
+ return addr;
+}
+EXPORT_SYMBOL(devm_memremap_ro);
+
void devm_memunmap(struct device *dev, void *addr)
{
WARN_ON(devres_release(dev, devm_memremap_release,
diff --git a/mm/memory.c b/mm/memory.c
index bb11c474857e..625623a90f08 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1793,7 +1793,7 @@ static int remap_pte_range(struct mm_struct *mm, pmd_t *pmd,
return -ENOMEM;
arch_enter_lazy_mmu_mode();
do {
- BUG_ON(!pte_none(*pte));
+// BUG_ON(!pte_none(*pte));
set_pte_at(mm, addr, pte, pte_mkspecial(pfn_pte(pfn, prot)));
pfn++;
} while (pte++, addr += PAGE_SIZE, addr != end);
diff --git a/mm/mmap.c b/mm/mmap.c
index a5e3dcd75e79..5423e3340e59 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -126,6 +126,7 @@ void vma_set_page_prot(struct vm_area_struct *vma)
/* remove_protection_ptes reads vma->vm_page_prot without mmap_sem */
WRITE_ONCE(vma->vm_page_prot, vm_page_prot);
}
+EXPORT_SYMBOL(vma_set_page_prot);

/*
* Requires inode->i_mapping->i_mmap_rwsem