s390/smp: rework absolute lowcore access

Temporary unsetting of the prefix page in memcpy_absolute() routine
poses a risk of executing code path with unexpectedly disabled prefix
page. This rework avoids the prefix page uninstalling and disabling
of normal and machine check interrupts when accessing the absolute
zero memory.

Although memcpy_absolute() routine can access the whole memory, it is
only used to update the absolute zero lowcore. This rework therefore
introduces a new mechanism for the absolute zero lowcore access and
scraps memcpy_absolute() routine for good.

Instead, an area is reserved in the virtual memory that is used for
the absolute lowcore access only. That area holds an array of 8KB
virtual mappings - one per CPU. Whenever a CPU is brought online, the
corresponding item is mapped to the real address of the previously
installed prefix page.

The absolute zero lowcore access works like this: a CPU calls the
new primitive get_abs_lowcore() to obtain its 8KB mapping as a
pointer to the struct lowcore. Virtual address references to that
pointer get translated to the real addresses of the prefix page,
which in turn gets swapped with the absolute zero memory addresses
due to prefixing. Once the pointer is not needed it must be released
with put_abs_lowcore() primitive:

	struct lowcore *abs_lc;
	unsigned long flags;

	abs_lc = get_abs_lowcore(&flags);
	abs_lc->... = ...;
	put_abs_lowcore(abs_lc, flags);

To ensure the described mechanism works large segment- and region-
table entries must be avoided for the 8KB mappings. Failure to do
so results in usage of Region-Frame Absolute Address (RFAA) or
Segment-Frame Absolute Address (SFAA) large page fields. In that
case absolute addresses would be used to address the prefix page
instead of the real ones and the prefixing would get bypassed.

Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>

1 files changed, 97 insertions, 0 deletions

diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c
index 1c86de9dd87f..a50a809e024f 100644
--- a/arch/s390/mm/vmem.c
+++ b/arch/s390/mm/vmem.c
@@ -561,6 +561,103 @@ int vmem_add_mapping(unsigned long start, unsigned long size)
 }
 
 /*
+ * Allocate new or return existing page-table entry, but do not map it
+ * to any physical address. If missing, allocate segment- and region-
+ * table entries along. Meeting a large segment- or region-table entry
+ * while traversing is an error, since the function is expected to be
+ * called against virtual regions reserverd for 4KB mappings only.
+ */
+static pte_t *vmem_get_alloc_pte(unsigned long addr, bool alloc)
+{
+	pte_t *ptep = NULL;
+	pgd_t *pgd;
+	p4d_t *p4d;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+
+	pgd = pgd_offset_k(addr);
+	if (pgd_none(*pgd)) {
+		if (!alloc)
+			goto out;
+		p4d = vmem_crst_alloc(_REGION2_ENTRY_EMPTY);
+		if (!p4d)
+			goto out;
+		pgd_populate(&init_mm, pgd, p4d);
+	}
+	p4d = p4d_offset(pgd, addr);
+	if (p4d_none(*p4d)) {
+		if (!alloc)
+			goto out;
+		pud = vmem_crst_alloc(_REGION3_ENTRY_EMPTY);
+		if (!pud)
+			goto out;
+		p4d_populate(&init_mm, p4d, pud);
+	}
+	pud = pud_offset(p4d, addr);
+	if (pud_none(*pud)) {
+		if (!alloc)
+			goto out;
+		pmd = vmem_crst_alloc(_SEGMENT_ENTRY_EMPTY);
+		if (!pmd)
+			goto out;
+		pud_populate(&init_mm, pud, pmd);
+	} else if (WARN_ON_ONCE(pud_large(*pud))) {
+		goto out;
+	}
+	pmd = pmd_offset(pud, addr);
+	if (pmd_none(*pmd)) {
+		if (!alloc)
+			goto out;
+		pte = vmem_pte_alloc();
+		if (!pte)
+			goto out;
+		pmd_populate(&init_mm, pmd, pte);
+	} else if (WARN_ON_ONCE(pmd_large(*pmd))) {
+		goto out;
+	}
+	ptep = pte_offset_kernel(pmd, addr);
+out:
+	return ptep;
+}
+
+int __vmem_map_4k_page(unsigned long addr, unsigned long phys, pgprot_t prot, bool alloc)
+{
+	pte_t *ptep, pte;
+
+	if (!IS_ALIGNED(addr, PAGE_SIZE))
+		return -EINVAL;
+	ptep = vmem_get_alloc_pte(addr, alloc);
+	if (!ptep)
+		return -ENOMEM;
+	__ptep_ipte(addr, ptep, 0, 0, IPTE_GLOBAL);
+	pte = mk_pte_phys(phys, prot);
+	set_pte(ptep, pte);
+	return 0;
+}
+
+int vmem_map_4k_page(unsigned long addr, unsigned long phys, pgprot_t prot)
+{
+	int rc;
+
+	mutex_lock(&vmem_mutex);
+	rc = __vmem_map_4k_page(addr, phys, prot, true);
+	mutex_unlock(&vmem_mutex);
+	return rc;
+}
+
+void vmem_unmap_4k_page(unsigned long addr)
+{
+	pte_t *ptep;
+
+	mutex_lock(&vmem_mutex);
+	ptep = virt_to_kpte(addr);
+	__ptep_ipte(addr, ptep, 0, 0, IPTE_GLOBAL);
+	pte_clear(&init_mm, addr, ptep);
+	mutex_unlock(&vmem_mutex);
+}
+
+/*
  * map whole physical memory to virtual memory (identity mapping)
  * we reserve enough space in the vmalloc area for vmemmap to hotplug
  * additional memory segments.