On 2020-01-08 5:43 a.m., David Hildenbrand wrote:

On 07.01.20 21:59, Logan Gunthorpe wrote:

quoted

In prepartion to support a pgprot_t argument for arch_add_memory().

Cc: Heiko Carstens <redacted>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Christian Borntraeger <redacted>
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
---
 arch/s390/include/asm/pgtable.h |  3 ++-
 arch/s390/mm/extmem.c           |  3 ++-
 arch/s390/mm/init.c             |  2 +-
 arch/s390/mm/vmem.c             | 10 +++++-----
 4 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 7b03037a8475..e667a1a96879 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h

@@ -1640,7 +1640,8 @@ static inline swp_entry_t __swp_entry(unsigned long type, unsigned long offset)
 
 #define kern_addr_valid(addr)   (1)
 
-extern int vmem_add_mapping(unsigned long start, unsigned long size);
+extern int vmem_add_mapping(unsigned long start, unsigned long size,
+			    pgprot_t prot);
 extern int vmem_remove_mapping(unsigned long start, unsigned long size);
 extern int s390_enable_sie(void);
 extern int s390_enable_skey(void);

diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c
index fd0dae9d10f4..6cf7029a7b35 100644
--- a/arch/s390/mm/extmem.c
+++ b/arch/s390/mm/extmem.c

@@ -313,7 +313,8 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long
 		goto out_free;
 	}
 
-	rc = vmem_add_mapping(seg->start_addr, seg->end - seg->start_addr + 1);
+	rc = vmem_add_mapping(seg->start_addr, seg->end - seg->start_addr + 1,
+			      PAGE_KERNEL);
 
 	if (rc)
 		goto out_free;

diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index a0c88c1c9ad0..ef19522ddad2 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c

@@ -277,7 +277,7 @@ int arch_add_memory(int nid, u64 start, u64 size,
 	if (WARN_ON_ONCE(modifiers->altmap))
 		return -EINVAL;
 
-	rc = vmem_add_mapping(start, size);
+	rc = vmem_add_mapping(start, size, PAGE_KERNEL);
 	if (rc)
 		return rc;
 

diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index b403fa14847d..8a5e95f184a2 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c

@@ -66,7 +66,7 @@ pte_t __ref *vmem_pte_alloc(void)
 /*
  * Add a physical memory range to the 1:1 mapping.
  */
-static int vmem_add_mem(unsigned long start, unsigned long size)
+static int vmem_add_mem(unsigned long start, unsigned long size, pgprot_t prot)
 {
 	unsigned long pgt_prot, sgt_prot, r3_prot;
 	unsigned long pages4k, pages1m, pages2g;

@@ -79,7 +79,7 @@ static int vmem_add_mem(unsigned long start, unsigned long size)
 	pte_t *pt_dir;
 	int ret = -ENOMEM;
 
-	pgt_prot = pgprot_val(PAGE_KERNEL);
+	pgt_prot = pgprot_val(prot);
 	sgt_prot = pgprot_val(SEGMENT_KERNEL);
 	r3_prot = pgprot_val(REGION3_KERNEL);

So, if we map as huge/gigantic pages, the protection would be discarded?
That looks wrong.

s390x does not support ZONE_DEVICE yet. Maybe simply bail out for s390x
as you do for sh to make your life easier?

Yeah, ok, makes sense to me; I'll change it for v3.

Logan