[PATCH stable 4.9 2/2] arch: pgtable: define MAX_POSSIBLE_PHYSMEM_BITS where needed

From: Florian Fainelli
Date: Wed Nov 03 2021 - 16:57:52 EST

From: Arnd Bergmann <arnd@xxxxxxxx>

[ Upstream commit cef397038167ac15d085914493d6c86385773709 ]

Stefan Agner reported a bug when using zsram on 32-bit Arm machines
with RAM above the 4GB address boundary:

Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = a27bd01c
[00000000] *pgd=236a0003, *pmd=1ffa64003
Internal error: Oops: 207 [#1] SMP ARM
Modules linked in: mdio_bcm_unimac(+) brcmfmac cfg80211 brcmutil raspberrypi_hwmon hci_uart crc32_arm_ce bcm2711_thermal phy_generic genet
CPU: 0 PID: 123 Comm: mkfs.ext4 Not tainted 5.9.6 #1
Hardware name: BCM2711
PC is at zs_map_object+0x94/0x338
LR is at zram_bvec_rw.constprop.0+0x330/0xa64
pc : [<c0602b38>] lr : [<c0bda6a0>] psr: 60000013
sp : e376bbe0 ip : 00000000 fp : c1e2921c
r10: 00000002 r9 : c1dda730 r8 : 00000000
r7 : e8ff7a00 r6 : 00000000 r5 : 02f9ffa0 r4 : e3710000
r3 : 000fdffe r2 : c1e0ce80 r1 : ebf979a0 r0 : 00000000
Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
Control: 30c5383d Table: 235c2a80 DAC: fffffffd
Process mkfs.ext4 (pid: 123, stack limit = 0x495a22e6)
Stack: (0xe376bbe0 to 0xe376c000)

As it turns out, zsram needs to know the maximum memory size, which
is defined in MAX_PHYSMEM_BITS when CONFIG_SPARSEMEM is set, or in
MAX_POSSIBLE_PHYSMEM_BITS on the x86 architecture.

The same problem will be hit on all 32-bit architectures that have a
physical address space larger than 4GB and happen to not enable sparsemem
and include asm/sparsemem.h from asm/pgtable.h.

After the initial discussion, I suggested just always defining
MAX_POSSIBLE_PHYSMEM_BITS whenever CONFIG_PHYS_ADDR_T_64BIT is
set, or provoking a build error otherwise. This addresses all
configurations that can currently have this runtime bug, but
leaves all other configurations unchanged.

I looked up the possible number of bits in source code and
datasheets, here is what I found:

- on ARC, CONFIG_ARC_HAS_PAE40 controls whether 32 or 40 bits are used
- on ARM, CONFIG_LPAE enables 40 bit addressing, without it we never
support more than 32 bits, even though supersections in theory allow
up to 40 bits as well.
- on MIPS, some MIPS32r1 or later chips support 36 bits, and MIPS32r5
XPA supports up to 60 bits in theory, but 40 bits are more than
anyone will ever ship
- On PowerPC, there are three different implementations of 36 bit
addressing, but 32-bit is used without CONFIG_PTE_64BIT
- On RISC-V, the normal page table format can support 34 bit
addressing. There is no highmem support on RISC-V, so anything
above 2GB is unused, but it might be useful to eventually support
CONFIG_ZRAM for high pages.

Fixes: 61989a80fb3a ("staging: zsmalloc: zsmalloc memory allocation library")
Fixes: 02390b87a945 ("mm/zsmalloc: Prepare to variable MAX_PHYSMEM_BITS")
Acked-by: Thomas Bogendoerfer <tsbogend@xxxxxxxxxxxxxxxx>
Reviewed-by: Stefan Agner <stefan@xxxxxxxx>
Tested-by: Stefan Agner <stefan@xxxxxxxx>
Acked-by: Mike Rapoport <rppt@xxxxxxxxxxxxx>
Link: https://lore.kernel.org/linux-mm/bdfa44bf1c570b05d6c70898e2bbb0acf234ecdf.1604762181.git.stefan@xxxxxxxx/
Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
[florian: patch arch/powerpc/include/asm/pte-common.h for 4.9.y
removed arch/riscv/include/asm/pgtable.h which does not exist]
Signed-off-by: Florian Fainelli <f.fainelli@xxxxxxxxx>
---
arch/arc/include/asm/pgtable.h | 2 ++
arch/arm/include/asm/pgtable-2level.h | 2 ++
arch/arm/include/asm/pgtable-3level.h | 2 ++
arch/mips/include/asm/pgtable-32.h | 3 +++
arch/powerpc/include/asm/pte-common.h | 2 ++
include/asm-generic/pgtable.h | 13 +++++++++++++
6 files changed, 24 insertions(+)

diff --git a/arch/arc/include/asm/pgtable.h b/arch/arc/include/asm/pgtable.h
index c10f5cb203e6..81198a6773c6 100644
--- a/arch/arc/include/asm/pgtable.h
+++ b/arch/arc/include/asm/pgtable.h
@@ -137,8 +137,10 @@

#ifdef CONFIG_ARC_HAS_PAE40
#define PTE_BITS_NON_RWX_IN_PD1 (0xff00000000 | PAGE_MASK | _PAGE_CACHEABLE)
+#define MAX_POSSIBLE_PHYSMEM_BITS 40
#else
#define PTE_BITS_NON_RWX_IN_PD1 (PAGE_MASK | _PAGE_CACHEABLE)
+#define MAX_POSSIBLE_PHYSMEM_BITS 32
#endif

/**************************************************************************
diff --git a/arch/arm/include/asm/pgtable-2level.h b/arch/arm/include/asm/pgtable-2level.h
index 92fd2c8a9af0..6154902bed83 100644
--- a/arch/arm/include/asm/pgtable-2level.h
+++ b/arch/arm/include/asm/pgtable-2level.h
@@ -78,6 +78,8 @@
#define PTE_HWTABLE_OFF (PTE_HWTABLE_PTRS * sizeof(pte_t))
#define PTE_HWTABLE_SIZE (PTRS_PER_PTE * sizeof(u32))

+#define MAX_POSSIBLE_PHYSMEM_BITS 32
+
/*
* PMD_SHIFT determines the size of the area a second-level page table can map
* PGDIR_SHIFT determines what a third-level page table entry can map
diff --git a/arch/arm/include/asm/pgtable-3level.h b/arch/arm/include/asm/pgtable-3level.h
index 2a029bceaf2f..35807e611b6e 100644
--- a/arch/arm/include/asm/pgtable-3level.h
+++ b/arch/arm/include/asm/pgtable-3level.h
@@ -37,6 +37,8 @@
#define PTE_HWTABLE_OFF (0)
#define PTE_HWTABLE_SIZE (PTRS_PER_PTE * sizeof(u64))

+#define MAX_POSSIBLE_PHYSMEM_BITS 40
+
/*
* PGDIR_SHIFT determines the size a top-level page table entry can map.
*/
diff --git a/arch/mips/include/asm/pgtable-32.h b/arch/mips/include/asm/pgtable-32.h
index c0be540e83cb..2c6df5a92e1e 100644
--- a/arch/mips/include/asm/pgtable-32.h
+++ b/arch/mips/include/asm/pgtable-32.h
@@ -110,6 +110,7 @@ static inline void pmd_clear(pmd_t *pmdp)

#if defined(CONFIG_XPA)

+#define MAX_POSSIBLE_PHYSMEM_BITS 40
#define pte_pfn(x) (((unsigned long)((x).pte_high >> _PFN_SHIFT)) | (unsigned long)((x).pte_low << _PAGE_PRESENT_SHIFT))
static inline pte_t
pfn_pte(unsigned long pfn, pgprot_t prot)
@@ -125,6 +126,7 @@ pfn_pte(unsigned long pfn, pgprot_t prot)

#elif defined(CONFIG_PHYS_ADDR_T_64BIT) && defined(CONFIG_CPU_MIPS32)

+#define MAX_POSSIBLE_PHYSMEM_BITS 36
#define pte_pfn(x) ((unsigned long)((x).pte_high >> 6))

static inline pte_t pfn_pte(unsigned long pfn, pgprot_t prot)
@@ -139,6 +141,7 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t prot)

#else

+#define MAX_POSSIBLE_PHYSMEM_BITS 32
#ifdef CONFIG_CPU_VR41XX
#define pte_pfn(x) ((unsigned long)((x).pte >> (PAGE_SHIFT + 2)))
#define pfn_pte(pfn, prot) __pte(((pfn) << (PAGE_SHIFT + 2)) | pgprot_val(prot))
diff --git a/arch/powerpc/include/asm/pte-common.h b/arch/powerpc/include/asm/pte-common.h
index 4ba26dd259fd..0d81cd9dd60e 100644
--- a/arch/powerpc/include/asm/pte-common.h
+++ b/arch/powerpc/include/asm/pte-common.h
@@ -101,8 +101,10 @@ static inline bool pte_user(pte_t pte)
*/
#if defined(CONFIG_PPC32) && defined(CONFIG_PTE_64BIT)
#define PTE_RPN_MASK (~((1ULL<<PTE_RPN_SHIFT)-1))
+#define MAX_POSSIBLE_PHYSMEM_BITS 36
#else
#define PTE_RPN_MASK (~((1UL<<PTE_RPN_SHIFT)-1))
+#define MAX_POSSIBLE_PHYSMEM_BITS 32
#endif

/* _PAGE_CHG_MASK masks of bits that are to be preserved across
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 0a4c2d4d9f8d..b43fa9d95a7a 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -847,6 +847,19 @@ static inline bool arch_has_pfn_modify_check(void)
#define io_remap_pfn_range remap_pfn_range
#endif

+#if !defined(MAX_POSSIBLE_PHYSMEM_BITS) && !defined(CONFIG_64BIT)
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+/*
+ * ZSMALLOC needs to know the highest PFN on 32-bit architectures
+ * with physical address space extension, but falls back to
+ * BITS_PER_LONG otherwise.
+ */
+#error Missing MAX_POSSIBLE_PHYSMEM_BITS definition
+#else
+#define MAX_POSSIBLE_PHYSMEM_BITS 32
+#endif
+#endif
+
#ifndef has_transparent_hugepage
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
#define has_transparent_hugepage() 1
--
2.25.1