Skip to contentSkip to conversation

Add support for 32-bit physical address

PATCH In review Ayan Kumar Halder 4 participants 17 messages Jun 2, 2023 – Jun 16, 2023

lore
Share & tools
lore
#0 lore raw

v8 cover Add support for 32-bit physical address

Hi All,

Please have a look at https://lists.xenproject.org/archives/html/xen-devel/2022-11/msg01465.html
for the context.

The benefits of using 32 bit physical addresses are as follows :-

1. It helps to use Xen on platforms (for eg R52) which supports 32-bit physical addresses and has no support for large physical address extension. On 32-bit MPU systems which supports flat-mapping (for eg R52), it helps to translate 32 bit VA into 32 bit PA.

2. It also helps in code optimization when the underlying platform does not use large physical address extension.

The following points are to be noted :-
1. Device tree always use uint64_t for address and size. The caller needs to translate between uint64_t and unsigned long (when 32 bit physical addressing is used).
2. Currently, we have enabled this option for Arm_32 as the MMU for Arm_64 uses 48-bit physical addressing.

Changes from :

v1 - 1. Reordered the patches such that the first three patches fixes issues in the existing codebase. These can be applied independent of the remaining patches in this serie.

2. Dropped translate_dt_address_size() for the address/size translation between paddr_t and u64 (as parsed from the device tree). Also, dropped the check for truncation (while converting u64 to paddr_t).
Instead now we have modified device_tree_get_reg() and typecasted the return for dt_read_number(), to obtain paddr_t. Also, introduced wrappers for fdt_get_mem_rsv() and dt_device_get_address() for the same purpose. These can be found in patch 4/11 and patch 6/11.

3. Split "Other adaptations required to support 32bit paddr" into the following individual patches for each adaptation :
xen/arm: smmu: Use writeq_relaxed_non_atomic() for writing to
SMMU_CBn_TTBR0
xen/arm: guest_walk: LPAE specific bits should be enclosed within

    "ifndef CONFIG_ARM_PA_32"

4. Introduced "xen/arm: p2m: Enable support for 32bit IPA".

v2 - 1. Dropped patches 1/11, 2/11 and 3/11 from the v2 as it has already been committed (except 2/11 - "[XEN v5] xen/arm: Use the correct format specifier" which is waiting to be committed).

2. Introduced a new patch "xen/drivers: ns16550: Use paddr_t for io_base/io_size".

v3 - 1. Combined the patches from https://lists.xenproject.org/archives/html/xen-devel/2023-02/msg00656.html in this series.

v4 - 1. Dropped "xen/drivers: ns16550: Use paddr_t for io_base/io_size" from the patch series.

2. Introduced "xen/arm: domain_build: Check if the address fits the range of physical address".

3. "xen/arm: Use the correct format specifier" has been committed in v4.

v5 - 1. Based on the comments on "[XEN v5 08/10] xen/arm: domain_build: Check if the address fits the range of physical address", the patch has been modified and split into the following :-

a. xen: dt: Replace u64 with uint64_t as the callback function parameters
for dt_for_each_range()
b. xen/arm: pci: Use 'uint64_t' as the datatype for the function
parameters.
c. xen/arm: domain_build: Check if the address fits the range of physical
address

v6 - 1. Reordered the patches such that only the patches which are dependent on "CONFIG_PHYS_ADDR_T_32" appear after the Kconfig option is introduced.

v7 - 1. Changes from "[XEN v7 01/11] xen/arm: domain_build: Track unallocated pages using the frame number " till "[XEN v7 06/11] xen: dt: Replace u64 with uint64_t as the callback function parameters for dt_for_each_range()" have been committed. So the remaining 5 patches are sent out.

Ayan Kumar Halder (5):
xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64 xen/arm: Introduce choice to enable 64/32 bit physical addressing xen/arm: guest_walk: LPAE specific bits should be enclosed within

    "ifndef CONFIG_PHYS_ADDR_T_32"

xen/arm: Restrict zeroeth_table_offset for ARM_64 xen/arm: p2m: Enable support for 32bit IPA for ARM_32

 xen/arch/Kconfig                     |  3 ++
 xen/arch/arm/Kconfig                 | 33 ++++++++++++++++++
 xen/arch/arm/guest_walk.c            |  2 ++
 xen/arch/arm/include/asm/lpae.h      |  4 +++
 xen/arch/arm/include/asm/p2m.h       |  6 ----
 xen/arch/arm/include/asm/page-bits.h |  6 +---
 xen/arch/arm/include/asm/types.h     | 14 ++++++++
 xen/arch/arm/mm.c                    | 12 +++----
 xen/arch/arm/p2m.c                   | 51 +++++++++++++++++++++-------
 9 files changed, 101 insertions(+), 30 deletions(-)
Signature · 1 line
2.17.1
#1 lore raw

PATCH v8 1/5 xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

Restructure the code so that one can use pa_range_info[] table for both ARM_32 as well as ARM_64.

Also, removed the hardcoding for P2M_ROOT_ORDER and P2M_ROOT_LEVEL as p2m_root_order can be obtained from the pa_range_info[].root_order and p2m_root_level can be obtained from pa_range_info[].sl0.

Refer ARM DDI 0406C.d ID040418, B3-1345, "Use of concatenated first-level translation tables

...However, a 40-bit input address range with a translation granularity of 4KB requires a total of 28 bits of address resolution. Therefore, a stage 2 translation that supports a 40-bit input address range requires two concatenated first-level translation tables,..."

Thus, root-order is 1 for 40-bit IPA on ARM_32.

Refer ARM DDI 0406C.d ID040418, B3-1348,

"Determining the required first lookup level for stage 2 translations

For a stage 2 translation, the output address range from the stage 1 translations determines the required input address range for the stage 2 translation. The permitted values of VTCR.SL0 are:

0b00 Stage 2 translation lookup must start at the second level. 0b01 Stage 2 translation lookup must start at the first level.

VTCR.T0SZ must indicate the required input address range. The size of the input address region is 2^(32-T0SZ) bytes."

Thus VTCR.SL0 = 1 (maximum value) and VTCR.T0SZ = -8 when the size of input address region is 2^40 bytes.

Thus, pa_range_info[].t0sz = 1 (VTCR.S) | 8 (VTCR.T0SZ) ie 11000b which is 24.

VTCR.T0SZ, is bits [5:0] for ARM_64.
VTCR.T0SZ is bits [3:0] and S(sign extension), bit[4] for ARM_32.

For this, we have used struct bitfields to convert pa_range_info[].t0sz to its ARM_32 variant.

pa_range_info[] is indexed by ID_AA64MMFR0_EL1.PARange which is present in Arm64 only. This is the reason we do not specify the indices for ARM_32. Also, we
duplicated the entry "{ 40, 24/*24*/, 1, 1 }" between ARM_64 and
ARM_32. This is done to avoid introducing extra #if-defs.

---
Changes from -

v3 - 1. New patch introduced in v4.
2. Restructure the code such that pa_range_info[] is used both by ARM_32 as well as ARM_64.

v4 - 1. Removed the hardcoded definitions of P2M_ROOT_ORDER and P2M_ROOT_LEVEL. The reason being root_order will not be always 1 (See the next patch).
2. Updated the commit message to explain t0sz, sl0 and root_order values for 32-bit IPA on Arm32.
3. Some sanity fixes.

v5 - pa_range_info is indexed by system_cpuinfo.mm64.pa_range. ie when PARange is 0, the PA size is 32, 1 -> 36 and so on. So pa_range_info[] has been updated accordingly.
For ARM_32 pa_range_info[0] = 0 and pa_range_info[1] = 0 as we do not support 32-bit, 36-bit physical address range yet.

v6 - 1. Added pa_range_info[] entries for ARM_32 without indices. Some entry may be duplicated between ARM_64 and ARM_32.
2. Recalculate p2m_ipa_bits for ARM_32 from T0SZ (similar to ARM_64).
3. Introduced an union to reinterpret T0SZ bits between ARM_32 and ARM_64.

v7 - 1. Used struct bifield instead of union to reinterpret T0SZ bits between ARM_32 and ARM_64.
2. Removed the invalid entry for ARM_32.

 xen/arch/arm/include/asm/p2m.h |  6 ----
 xen/arch/arm/p2m.c             | 50 +++++++++++++++++++++++++---------
 2 files changed, 37 insertions(+), 19 deletions(-)
xen/arch/arm/include/asm/p2m.h +0 -6
index f67e9ddc72..940495d42b 100644
@@ -14,16 +14,10 @@
 /* Holds the bit size of IPAs in p2m tables.  */
 extern unsigned int p2m_ipa_bits;
 
-#ifdef CONFIG_ARM_64
 extern unsigned int p2m_root_order;
 extern unsigned int p2m_root_level;
 #define P2M_ROOT_ORDER    p2m_root_order
 #define P2M_ROOT_LEVEL p2m_root_level
-#else
-/* First level P2M is always 2 consecutive pages */
-#define P2M_ROOT_ORDER    1
-#define P2M_ROOT_LEVEL 1
-#endif
 
 struct domain;
 
xen/arch/arm/p2m.c +37 -13

111 lines not on this page · raw message

Signature · 1 line
2.17.1
#10 lore raw

PATCH v8 2/5 xen/arm: Introduce choice to enable 64/32 bit physical addressing

In reply to Ayan Kumar Halder

Some Arm based hardware platforms which does not support LPAE (eg Cortex-R52), uses 32 bit physical addresses.
Also, users may choose to use 32 bits to represent physical addresses for optimization.

To support the above use cases, we have introduced arch independent config to choose if the physical address can be represented using 32 bits (PHYS_ADDR_T_32) or 64 bits (!PHYS_ADDR_T_32). For now only ARM_32 provides support to enable 32 bit physical addressing.

When PHYS_ADDR_T_32 is defined, PADDR_BITS is set to 32. Note that we use "unsigned long" (not "uint32_t") to denote the datatype of physical address. This is done to avoid using a cast each time PAGE_* macros are used on paddr_t. For eg PAGE_SIZE is defined as unsigned long. Thus, each time PAGE_SIZE is used with paddr_t, the result will be "unsigned long".
On 32-bit architecture, "unsigned long" is 32-bit wide. Thus, it can be used to denote physical address.

When PHYS_ADDR_T_32 is not defined for ARM_32, PADDR_BITS is set to 40. For ARM_64, PADDR_BITS is set to 48.
The last two are same as the current configuration used today on Xen.

---
Changes from -
v1 - 1. Extracted from "[XEN v1 8/9] xen/arm: Other adaptations required to support 32bit paddr".

v2 - 1. Introduced Kconfig choice. ARM_64 can select PHYS_ADDR_64 only whereas ARM_32 can select PHYS_ADDR_32 or PHYS_ADDR_64.
2. For CONFIG_ARM_PA_32, paddr_t is defined as 'unsigned long'.

v3 - 1. Allow user to define PADDR_BITS by selecting different config options ARM_PA_BITS_32, ARM_PA_BITS_40 and ARM_PA_BITS_48.
2. Add the choice under "Architecture Features".

v4 - 1. Removed PHYS_ADDR_T_64 as !PHYS_ADDR_T_32 means PHYS_ADDR_T_32.

v5 - 1. Removed ARM_PA_BITS_48 as there is no choice for ARM_64.
2. In ARM_PA_BITS_32, "help" is moved to last, and "depends on" before "select".

v6 - 1. Explained why we use "unsigned long" to represent physical address for ARM_32.

v7 - 1. Updated the reasoning for using "unsigned long" for paddr_t.
2. Added R-b by Michal.

 xen/arch/Kconfig                     |  3 +++
 xen/arch/arm/Kconfig                 | 33 ++++++++++++++++++++++++++++
 xen/arch/arm/include/asm/page-bits.h |  6 +----
 xen/arch/arm/include/asm/types.h     | 14 ++++++++++++
 xen/arch/arm/mm.c                    |  5 +++++
 5 files changed, 56 insertions(+), 5 deletions(-)
xen/arch/Kconfig +3 -0
index 7028f7b74f..67ba38f32f 100644
@@ -1,6 +1,9 @@
 config 64BIT
 	bool
 
+config PHYS_ADDR_T_32
+	bool
+
 config NR_CPUS
 	int "Maximum number of CPUs"
 	range 1 4095
xen/arch/arm/Kconfig +33 -0
index 239d3aed3c..1e87fe0247 100644
@@ -26,6 +26,39 @@ config ARCH_DEFCONFIG
 
 menu "Architecture Features"
 
+choice
+	prompt "Physical address space size" if ARM_32
+	default ARM_PA_BITS_40 if ARM_32
+	help
+	  User can choose to represent the width of physical address. This can
+	  sometimes help in optimizing the size of image when user chooses a
+	  smaller size to represent physical address.
+
+config ARM_PA_BITS_32
+	bool "32-bit"
+	depends on ARM_32
+	select PHYS_ADDR_T_32
+	help
+	  On platforms where any physical address can be represented within 32 bits,
+	  user should choose this option. This will help in reduced size of the
+	  binary.
+	  Xen uses "unsigned long" and not "uint32_t" to denote the datatype of
+	  physical address. This is done to avoid using a cast each time PAGE_*
+	  macros are used on paddr_t. For eg PAGE_SIZE is defined as unsigned long.
+	  On 32-bit architecture, "unsigned long" is 32-bit wide. Thus, it can be
+	  used to denote physical address.
+
+config ARM_PA_BITS_40
+	bool "40-bit"
+	depends on ARM_32
+endchoice
+
+config PADDR_BITS
+	int
+	default 32 if ARM_PA_BITS_32
+	default 40 if ARM_PA_BITS_40
+	default 48 if ARM_64
+
 source "arch/Kconfig"
 
 config ACPI
xen/arch/arm/include/asm/page-bits.h +1 -5
index 5d6477e599..deb381ceeb 100644
@@ -3,10 +3,6 @@
 
 #define PAGE_SHIFT              12
 
-#ifdef CONFIG_ARM_64
-#define PADDR_BITS              48
-#else
-#define PADDR_BITS              40
-#endif
+#define PADDR_BITS              CONFIG_PADDR_BITS
 
 #endif /* __ARM_PAGE_SHIFT_H__ */
xen/arch/arm/include/asm/types.h +14 -0
index e218ed77bd..fb6618ef24 100644
@@ -34,9 +34,23 @@ typedef signed long long s64;
 typedef unsigned long long u64;
 typedef u32 vaddr_t;
 #define PRIvaddr PRIx32
+#if defined(CONFIG_PHYS_ADDR_T_32)
+
+/*
+ * We use "unsigned long" and not "uint32_t" to denote the type. This is done
+ * to avoid having a cast each time PAGE_* macros are used on paddr_t. For eg
+ * PAGE_SIZE is defined as unsigned long.
+ * On 32-bit architecture, "unsigned long" is 32-bit wide. Thus, we can use it
+ * to denote physical address.
+ */
+typedef unsigned long paddr_t;
+#define INVALID_PADDR (~0UL)
+#define PRIpaddr "08lx"
+#else
 typedef u64 paddr_t;
 #define INVALID_PADDR (~0ULL)
 #define PRIpaddr "016llx"
+#endif
 typedef u32 register_t;
 #define PRIregister "08x"
 #elif defined (CONFIG_ARM_64)
xen/arch/arm/mm.c +5 -0
index 74f6ff2c6f..5ef5fd8c49 100644
@@ -703,6 +703,11 @@ void __init setup_frametable_mappings(paddr_t ps, paddr_t pe)
     const unsigned long mapping_size = frametable_size < MB(32) ? MB(2) : MB(32);
     int rc;
 
+    /*
+     * The size of paddr_t should be sufficient for the complete range of
+     * physical address.
+     */
+    BUILD_BUG_ON((sizeof(paddr_t) * BITS_PER_BYTE) < PADDR_BITS);
     BUILD_BUG_ON(sizeof(struct page_info) != PAGE_INFO_SIZE);
 
     if ( frametable_size > FRAMETABLE_SIZE )
Signature · 1 line
2.17.1
#12 lore raw

PATCH v8 3/5 xen/arm: guest_walk: LPAE specific bits should be enclosed within "ifndef CONFIG_PHYS_ADDR_T_32"

In reply to Ayan Kumar Halder

As the previous patch introduces CONFIG_PHYS_ADDR_T_32 to support 32 bit physical addresses, the code specific to "Large Physical Address Extension" (ie LPAE) should be enclosed within "ifndef CONFIG_PHYS_ADDR_T_32".

Refer xen/arch/arm/include/asm/short-desc.h, "short_desc_l1_supersec_t"
unsigned int extbase1:4; /* Extended base address, PA[35:32] */
unsigned int extbase2:4; /* Extended base address, PA[39:36] */

Thus, extbase1 and extbase2 are not valid when 32 bit physical addresses are supported.

---
Changes from -
v1 - 1. Extracted from "[XEN v1 8/9] xen/arm: Other adaptations required to support 32bit paddr".

v2 - 1. Reordered this patch so that it appears after CONFIG_ARM_PA_32 is introduced (in 6/9).

v3 - 1. Updated the commit message.
2. Added Ack.

v4 - 1. No changes.

v5 - 1. No changes.

v6 - 1. No changes.

v7 - 1. No changes.

 xen/arch/arm/guest_walk.c | 2 ++
 1 file changed, 2 insertions(+)
xen/arch/arm/guest_walk.c +2 -0
index 43d3215304..c80a0ce55b 100644
@@ -154,8 +154,10 @@ static bool guest_walk_sd(const struct vcpu *v,
             mask = (1ULL << L1DESC_SUPERSECTION_SHIFT) - 1;
             *ipa = gva & mask;
             *ipa |= (paddr_t)(pte.supersec.base) << L1DESC_SUPERSECTION_SHIFT;
+#ifndef CONFIG_PHYS_ADDR_T_32
             *ipa |= (paddr_t)(pte.supersec.extbase1) << L1DESC_SUPERSECTION_EXT_BASE1_SHIFT;
             *ipa |= (paddr_t)(pte.supersec.extbase2) << L1DESC_SUPERSECTION_EXT_BASE2_SHIFT;
+#endif /* CONFIG_PHYS_ADDR_T_32 */
         }
 
         /* Set permissions so that the caller can check the flags by herself. */
Signature · 1 line
2.17.1
#13 lore raw

PATCH v8 4/5 xen/arm: Restrict zeroeth_table_offset for ARM_64

In reply to Ayan Kumar Halder

When 32 bit physical addresses are used (ie PHYS_ADDR_T_32=y), "va >> ZEROETH_SHIFT" causes an overflow.
Also, there is no zeroeth level page table on Arm32.

Also took the opportunity to clean up dump_pt_walk(). One could use DECLARE_OFFSETS() macro instead of declaring an array of page table offsets.

---
Changes from -

v1 - Removed the duplicate declaration for DECLARE_OFFSETS.

v2 - 1. Reworded the commit message.
2. Use CONFIG_ARM_PA_32 to restrict zeroeth_table_offset.

v3 - 1. Added R-b and Ack.

v4 - 1. Removed R-b and Ack as we use CONFIG_PHYS_ADDR_T_32 instead of CONFIG_ARM_PA_BITS_32. This is to be in parity with our earlier patches where we use CONFIG_PHYS_ADDR_T_32 to denote 32-bit physical addr support.

v5 - 1. Added R-b and Ack.

v6 - 1. No changes.

v7 - 1. No changes.

 xen/arch/arm/include/asm/lpae.h | 4 ++++
 xen/arch/arm/mm.c               | 7 +------
 2 files changed, 5 insertions(+), 6 deletions(-)
xen/arch/arm/include/asm/lpae.h +4 -0
index 3fdd5d0de2..7d2f6fd1bd 100644
@@ -259,7 +259,11 @@ lpae_t mfn_to_xen_entry(mfn_t mfn, unsigned int attr);
 #define first_table_offset(va)  TABLE_OFFSET(first_linear_offset(va))
 #define second_table_offset(va) TABLE_OFFSET(second_linear_offset(va))
 #define third_table_offset(va)  TABLE_OFFSET(third_linear_offset(va))
+#ifdef CONFIG_PHYS_ADDR_T_32
+#define zeroeth_table_offset(va)  0
+#else
 #define zeroeth_table_offset(va)  TABLE_OFFSET(zeroeth_linear_offset(va))
+#endif
 
 /*
  * Macros to define page-tables:
xen/arch/arm/mm.c +1 -6
index 5ef5fd8c49..e460249736 100644
@@ -233,12 +233,7 @@ void dump_pt_walk(paddr_t ttbr, paddr_t addr,
 {
     static const char *level_strs[4] = { "0TH", "1ST", "2ND", "3RD" };
     const mfn_t root_mfn = maddr_to_mfn(ttbr);
-    const unsigned int offsets[4] = {
-        zeroeth_table_offset(addr),
-        first_table_offset(addr),
-        second_table_offset(addr),
-        third_table_offset(addr)
-    };
+    DECLARE_OFFSETS(offsets, addr);
     lpae_t pte, *mapping;
     unsigned int level, root_table;
 
Signature · 1 line
2.17.1
#14 lore raw

PATCH v8 5/5 xen/arm: p2m: Enable support for 32bit IPA for ARM_32

In reply to Ayan Kumar Halder

Refer ARM DDI 0406C.d ID040418, B3-1345,

"A stage 2 translation with an input address range of 31-34 bits can start the translation either:

- With a first-level lookup, accessing a first-level translation table with 2-16 entries.

- With a second-level lookup, accessing a set of concatenated second-level translation tables"

Thus, for 32 bit IPA, there will be no concatenated root level tables. So, the root-order is 0.

Also, Refer ARM DDI 0406C.d ID040418, B3-1348
"Determining the required first lookup level for stage 2 translations

For a stage 2 translation, the output address range from the stage 1 translations determines the required input address range for the stage 2 translation. The permitted values of VTCR.SL0 are:
0b00 Stage 2 translation lookup must start at the second level. 0b01 Stage 2 translation lookup must start at the first level.

VTCR.T0SZ must indicate the required input address range. The size of the input address region is 2^(32-T0SZ) bytes."

Thus VTCR.SL0 = 1 (maximum value) and VTCR.T0SZ = 0 when the size of input address region is 2^32 bytes.

---
Changes from -

v1 - New patch.

v2 - 1. Added Ack.

v3 - 1. Dropped Ack.
2. Rebased the patch based on the previous change.

v4 - 1. t0sz is 0 for 32-bit IPA on Arm32.
2. Updated the commit message to explain t0sz, sl0 and root_order.

v5 - 1. Rebased on top of the changes in the previous patch.

v6 - 1. Removed the index for ARM_32.

v7 - 1. No changes.

 xen/arch/arm/p2m.c | 1 +
 1 file changed, 1 insertion(+)
xen/arch/arm/p2m.c +1 -0
index 76388ba54b..a969068a68 100644
@@ -2265,6 +2265,7 @@ void __init setup_virt_paging(void)
         [6] = { 52,      12/*12*/,  4,          2 },
         [7] = { 0 }  /* Invalid */
 #else
+        { 32,      0/*0*/,    0,          1 },
         { 40,      24/*24*/,  1,          1 }
 #endif
     };
Signature · 1 line
2.17.1
#2 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

In reply to Ayan Kumar Halder

Hi Ayan,

Quoted 4 lines · Ayan Kumar Halder

Restructure the code so that one can use pa_range_info[] table for both ARM_32 as well as ARM_64.

I grepped for ARM_{32,64} in our code base and could not find any use in source files except for things introduced by this commit. While I'm ok with it in a commit message I think for consistency we should be using arm32/arm64 in the code.

Quoted 155 lines

Also, removed the hardcoding for P2M_ROOT_ORDER and P2M_ROOT_LEVEL as p2m_root_order can be obtained from the pa_range_info[].root_order and p2m_root_level can be obtained from pa_range_info[].sl0.

Refer ARM DDI 0406C.d ID040418, B3-1345, "Use of concatenated first-level translation tables

...However, a 40-bit input address range with a translation granularity of 4KB requires a total of 28 bits of address resolution. Therefore, a stage 2 translation that supports a 40-bit input address range requires two concatenated first-level translation tables,..."

Thus, root-order is 1 for 40-bit IPA on ARM_32.

Refer ARM DDI 0406C.d ID040418, B3-1348,

"Determining the required first lookup level for stage 2 translations

For a stage 2 translation, the output address range from the stage 1 translations determines the required input address range for the stage 2 translation. The permitted values of VTCR.SL0 are:

0b00 Stage 2 translation lookup must start at the second level. 0b01 Stage 2 translation lookup must start at the first level.

VTCR.T0SZ must indicate the required input address range. The size of the input address region is 2^(32-T0SZ) bytes."

Thus VTCR.SL0 = 1 (maximum value) and VTCR.T0SZ = -8 when the size of input address region is 2^40 bytes.

Thus, pa_range_info[].t0sz = 1 (VTCR.S) | 8 (VTCR.T0SZ) ie 11000b which is 24.

VTCR.T0SZ, is bits [5:0] for ARM_64.
VTCR.T0SZ is bits [3:0] and S(sign extension), bit[4] for ARM_32.

For this, we have used struct bitfields to convert pa_range_info[].t0sz to its ARM_32 variant.

pa_range_info[] is indexed by ID_AA64MMFR0_EL1.PARange which is present in Arm64 only. This is the reason we do not specify the indices for ARM_32. Also, we
duplicated the entry "{ 40, 24/*24*/, 1, 1 }" between ARM_64 and
ARM_32. This is done to avoid introducing extra #if-defs.

---
Changes from -

v3 - 1. New patch introduced in v4.
2. Restructure the code such that pa_range_info[] is used both by ARM_32 as well as ARM_64.

v4 - 1. Removed the hardcoded definitions of P2M_ROOT_ORDER and P2M_ROOT_LEVEL. The reason being root_order will not be always 1 (See the next patch).
2. Updated the commit message to explain t0sz, sl0 and root_order values for 32-bit IPA on Arm32.
3. Some sanity fixes.

v5 - pa_range_info is indexed by system_cpuinfo.mm64.pa_range. ie when PARange is 0, the PA size is 32, 1 -> 36 and so on. So pa_range_info[] has been updated accordingly.
For ARM_32 pa_range_info[0] = 0 and pa_range_info[1] = 0 as we do not support 32-bit, 36-bit physical address range yet.

v6 - 1. Added pa_range_info[] entries for ARM_32 without indices. Some entry may be duplicated between ARM_64 and ARM_32.
2. Recalculate p2m_ipa_bits for ARM_32 from T0SZ (similar to ARM_64).
3. Introduced an union to reinterpret T0SZ bits between ARM_32 and ARM_64.

v7 - 1. Used struct bifield instead of union to reinterpret T0SZ bits between ARM_32 and ARM_64.
2. Removed the invalid entry for ARM_32.

 xen/arch/arm/include/asm/p2m.h |  6 ----
 xen/arch/arm/p2m.c             | 50 +++++++++++++++++++++++++---------
 2 files changed, 37 insertions(+), 19 deletions(-)
xen/arch/arm/include/asm/p2m.h +0 -6
index f67e9ddc72..940495d42b 100644
@@ -14,16 +14,10 @@
 /* Holds the bit size of IPAs in p2m tables.  */
 extern unsigned int p2m_ipa_bits;
 
-#ifdef CONFIG_ARM_64
 extern unsigned int p2m_root_order;
 extern unsigned int p2m_root_level;
 #define P2M_ROOT_ORDER    p2m_root_order
 #define P2M_ROOT_LEVEL p2m_root_level
-#else
-/* First level P2M is always 2 consecutive pages */
-#define P2M_ROOT_ORDER    1
-#define P2M_ROOT_LEVEL 1
-#endif
 
 struct domain;
 
xen/arch/arm/p2m.c +9 -11
index 418997843d..76388ba54b 100644
@@ -19,9 +19,9 @@
 
 #define INVALID_VMID 0 /* VMID 0 is reserved */
 
-#ifdef CONFIG_ARM_64
 unsigned int __read_mostly p2m_root_order;
 unsigned int __read_mostly p2m_root_level;
+#ifdef CONFIG_ARM_64
 static unsigned int __read_mostly max_vmid = MAX_VMID_8_BIT;
 /* VMID is by default 8 bit width on AArch64 */
 #define MAX_VMID       max_vmid
@@ -2247,16 +2247,6 @@ void __init setup_virt_paging(void)
     /* Setup Stage 2 address translation */
     register_t val = VTCR_RES1|VTCR_SH0_IS|VTCR_ORGN0_WBWA|VTCR_IRGN0_WBWA;
 
-#ifdef CONFIG_ARM_32
-    if ( p2m_ipa_bits < 40 )
-        panic("P2M: Not able to support %u-bit IPA at the moment\n",
-              p2m_ipa_bits);
-
-    printk("P2M: 40-bit IPA\n");
-    p2m_ipa_bits = 40;
-    val |= VTCR_T0SZ(0x18); /* 40 bit IPA */
-    val |= VTCR_SL0(0x1); /* P2M starts at first level */
-#else /* CONFIG_ARM_64 */
     static const struct {
         unsigned int pabits; /* Physical Address Size */
         unsigned int t0sz;   /* Desired T0SZ, minimum in comment */
@@ -2265,6 +2255,7 @@ void __init setup_virt_paging(void)
     } pa_range_info[] __initconst = {
         /* T0SZ minimum and SL0 maximum from ARM DDI 0487H.a Table D5-6 */
         /*      PA size, t0sz(min), root-order, sl0(max) */
+#ifdef CONFIG_ARM_64
         [0] = { 32,      32/*32*/,  0,          1 },
         [1] = { 36,      28/*28*/,  0,          1 },
         [2] = { 40,      24/*24*/,  1,          1 },
@@ -2273,11 +2264,28 @@ void __init setup_virt_paging(void)
         [5] = { 48,      16/*16*/,  0,          2 },
         [6] = { 52,      12/*12*/,  4,          2 },
         [7] = { 0 }  /* Invalid */
+#else
+        { 40,      24/*24*/,  1,          1 }
+#endif
     };
 
     unsigned int i;
     unsigned int pa_range = 0x10; /* Larger than any possible value */
 
+#ifdef CONFIG_ARM_32
+    /*
+     * Typecast pa_range_info[].t0sz into ARM_32 bit variant.
+     *

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2288

Quoted 10 lines
xen/arch/arm/p2m.c +10 -0
+     * VTCR.T0SZ is bits [3:0] and S(sign extension), bit[4] for ARM_32.
+     * Thus, pa_range_info[].t0sz is translated to its ARM_32 variant using
+     * struct bitfields.
+     */
+    struct
+    {
+        signed int val:5;
+    } t0sz_32;
+#else
+

no need for this empty line

Quoted 12 lines
xen/arch/arm/p2m.c +1 -0
     /*
      * Restrict "p2m_ipa_bits" if needed. As P2M table is always configured
      * with IPA bits == PA bits, compare against "pabits".
@@ -2291,6 +2299,7 @@ void __init setup_virt_paging(void)
      */
     if ( system_cpuinfo.mm64.vmid_bits == MM64_VMID_16_BITS_SUPPORT )
         max_vmid = MAX_VMID_16_BIT;
+#endif
 
     /* Choose suitable "pa_range" according to the resulted "p2m_ipa_bits". */
     for ( i = 0; i < ARRAY_SIZE(pa_range_info); i++ )

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2314

Quoted 2 lines
xen/arch/arm/p2m.c +0 -0
@@ -2304,26 +2313,41 @@ void __init setup_virt_paging(void)
 
     /* pa_range is 4 bits but we don't support all modes */

this comment makes sense really only on arm64 as it refers to PARange field of ID_AA64MMFR0_EL1.

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2321

Quoted 8 lines
xen/arch/arm/p2m.c +6 -1
     if ( pa_range >= ARRAY_SIZE(pa_range_info) || !pa_range_info[pa_range].pabits )
-        panic("Unknown encoding of ID_AA64MMFR0_EL1.PARange %x\n", pa_range);
+    {
+        /*
+         * In case of ARM_64, we do not support this encoding of
+         * ID_AA64MMFR0_EL1.PARange
+         */
+        panic("Unsupported value for p2m_ipa_bits = 0x%x\n", p2m_ipa_bits);

NIT: Putting variable names in messages visible by users is not a great idea IMO. "Unsupported IPA size" would read better. Furthermore, I do not think printing IPA size in hex is beneficial. I would use "%u bits" (i.e. 32 bits reads better than 0x20 bits).

Quoted 34 lines
xen/arch/arm/p2m.c +11 -2
+    }
 
+#ifdef CONFIG_ARM_64
     val |= VTCR_PS(pa_range);
     val |= VTCR_TG0_4K;
 
     /* Set the VS bit only if 16 bit VMID is supported. */
     if ( MAX_VMID == MAX_VMID_16_BIT )
         val |= VTCR_VS;
+#endif
+
     val |= VTCR_SL0(pa_range_info[pa_range].sl0);
     val |= VTCR_T0SZ(pa_range_info[pa_range].t0sz);
 
     p2m_root_order = pa_range_info[pa_range].root_order;
     p2m_root_level = 2 - pa_range_info[pa_range].sl0;
+
+#ifdef CONFIG_ARM_64
     p2m_ipa_bits = 64 - pa_range_info[pa_range].t0sz;
+#else
+    t0sz_32.val = pa_range_info[pa_range].t0sz;
+    p2m_ipa_bits = 32 - t0sz_32.val;
+#endif
 
     printk("P2M: %d-bit IPA with %d-bit PA and %d-bit VMID\n",
            p2m_ipa_bits,
            pa_range_info[pa_range].pabits,
            ( MAX_VMID == MAX_VMID_16_BIT ) ? 16 : 8);
-#endif
+
     printk("P2M: %d levels with order-%d root, VTCR 0x%"PRIregister"\n",
            4 - P2M_ROOT_LEVEL, P2M_ROOT_ORDER, val);
 
--
Quoted 3 lines

2.17.1

~Michal

Reviewed-by on v8 5/5
#15 lore raw

Re: [XEN v8 5/5] xen/arm: p2m: Enable support for 32bit IPA for ARM_32

In reply to Ayan Kumar Halder
Quoted 25 lines · Ayan Kumar Halder

Refer ARM DDI 0406C.d ID040418, B3-1345,

"A stage 2 translation with an input address range of 31-34 bits can start the translation either:

- With a first-level lookup, accessing a first-level translation table with 2-16 entries.

- With a second-level lookup, accessing a set of concatenated second-level translation tables"

Thus, for 32 bit IPA, there will be no concatenated root level tables. So, the root-order is 0.

Also, Refer ARM DDI 0406C.d ID040418, B3-1348
"Determining the required first lookup level for stage 2 translations

For a stage 2 translation, the output address range from the stage 1 translations determines the required input address range for the stage 2 translation. The permitted values of VTCR.SL0 are:
0b00 Stage 2 translation lookup must start at the second level. 0b01 Stage 2 translation lookup must start at the first level.

Quoted 7 lines

VTCR.T0SZ must indicate the required input address range. The size of the input address region is 2^(32-T0SZ) bytes."

Thus VTCR.SL0 = 1 (maximum value) and VTCR.T0SZ = 0 when the size of input address region is 2^32 bytes.

~Michal

#3 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

In reply to Michal Orzel
Quoted 1 line · Michal Orzel

Hi Ayan,

Hi Michal,

Quoted 8 lines · Ayan Kumar Halder
Quoted 3 lines

Restructure the code so that one can use pa_range_info[] table for both ARM_32 as well as ARM_64.

I grepped for ARM_{32,64} in our code base and could not find any use in source files except for things introduced by this commit. While I'm ok with it in a commit message I think for consistency we should be using arm32/arm64 in the code.

AFAIU, arm32/arm64 refers to the Architecture. ARM_32/ARM_64 refers to the configuration.

If you see the original code (xen/arch/arm/include/asm/p2m.h, xen/arch/arm/p2m.c)

ARM_32/ARM_64 has been used.

Thus, I used ARM_32/ARM_64 in this commit. Let me know if it makes sense. The rest of your comments look sane to me.

- Ayan

Quoted 233 lines
Quoted 155 lines

Also, removed the hardcoding for P2M_ROOT_ORDER and P2M_ROOT_LEVEL as p2m_root_order can be obtained from the pa_range_info[].root_order and p2m_root_level can be obtained from pa_range_info[].sl0.

Refer ARM DDI 0406C.d ID040418, B3-1345, "Use of concatenated first-level translation tables

...However, a 40-bit input address range with a translation granularity of 4KB requires a total of 28 bits of address resolution. Therefore, a stage 2 translation that supports a 40-bit input address range requires two concatenated first-level translation tables,..."

Thus, root-order is 1 for 40-bit IPA on ARM_32.

Refer ARM DDI 0406C.d ID040418, B3-1348,

"Determining the required first lookup level for stage 2 translations

For a stage 2 translation, the output address range from the stage 1 translations determines the required input address range for the stage 2 translation. The permitted values of VTCR.SL0 are:

0b00 Stage 2 translation lookup must start at the second level. 0b01 Stage 2 translation lookup must start at the first level.

VTCR.T0SZ must indicate the required input address range. The size of the input address region is 2^(32-T0SZ) bytes."

Thus VTCR.SL0 = 1 (maximum value) and VTCR.T0SZ = -8 when the size of input address region is 2^40 bytes.

Thus, pa_range_info[].t0sz = 1 (VTCR.S) | 8 (VTCR.T0SZ) ie 11000b which is 24.

VTCR.T0SZ, is bits [5:0] for ARM_64.
VTCR.T0SZ is bits [3:0] and S(sign extension), bit[4] for ARM_32.

For this, we have used struct bitfields to convert pa_range_info[].t0sz to its ARM_32 variant.

pa_range_info[] is indexed by ID_AA64MMFR0_EL1.PARange which is present in Arm64 only. This is the reason we do not specify the indices for ARM_32. Also, we
duplicated the entry "{ 40, 24/*24*/, 1, 1 }" between ARM_64 and
ARM_32. This is done to avoid introducing extra #if-defs.

---
Changes from -

v3 - 1. New patch introduced in v4.
2. Restructure the code such that pa_range_info[] is used both by ARM_32 as well as ARM_64.

v4 - 1. Removed the hardcoded definitions of P2M_ROOT_ORDER and P2M_ROOT_LEVEL. The reason being root_order will not be always 1 (See the next patch).
2. Updated the commit message to explain t0sz, sl0 and root_order values for 32-bit IPA on Arm32.
3. Some sanity fixes.

v5 - pa_range_info is indexed by system_cpuinfo.mm64.pa_range. ie when PARange is 0, the PA size is 32, 1 -> 36 and so on. So pa_range_info[] has been updated accordingly.
For ARM_32 pa_range_info[0] = 0 and pa_range_info[1] = 0 as we do not support 32-bit, 36-bit physical address range yet.

v6 - 1. Added pa_range_info[] entries for ARM_32 without indices. Some entry may be duplicated between ARM_64 and ARM_32.
2. Recalculate p2m_ipa_bits for ARM_32 from T0SZ (similar to ARM_64).
3. Introduced an union to reinterpret T0SZ bits between ARM_32 and ARM_64.

v7 - 1. Used struct bifield instead of union to reinterpret T0SZ bits between ARM_32 and ARM_64.
2. Removed the invalid entry for ARM_32.

  xen/arch/arm/include/asm/p2m.h |  6 ----
  xen/arch/arm/p2m.c             | 50 +++++++++++++++++++++++++---------
  2 files changed, 37 insertions(+), 19 deletions(-)
xen/arch/arm/include/asm/p2m.h +0 -6
index f67e9ddc72..940495d42b 100644
@@ -14,16 +14,10 @@
  /* Holds the bit size of IPAs in p2m tables.  */
  extern unsigned int p2m_ipa_bits;
 
-#ifdef CONFIG_ARM_64
  extern unsigned int p2m_root_order;
  extern unsigned int p2m_root_level;
  #define P2M_ROOT_ORDER    p2m_root_order
  #define P2M_ROOT_LEVEL p2m_root_level
-#else
-/* First level P2M is always 2 consecutive pages */
-#define P2M_ROOT_ORDER    1
-#define P2M_ROOT_LEVEL 1
-#endif
 
  struct domain;
 
xen/arch/arm/p2m.c +9 -11
index 418997843d..76388ba54b 100644
@@ -19,9 +19,9 @@
 
  #define INVALID_VMID 0 /* VMID 0 is reserved */
 
-#ifdef CONFIG_ARM_64
  unsigned int __read_mostly p2m_root_order;
  unsigned int __read_mostly p2m_root_level;
+#ifdef CONFIG_ARM_64
  static unsigned int __read_mostly max_vmid = MAX_VMID_8_BIT;
  /* VMID is by default 8 bit width on AArch64 */
  #define MAX_VMID       max_vmid
@@ -2247,16 +2247,6 @@ void __init setup_virt_paging(void)
      /* Setup Stage 2 address translation */
      register_t val = VTCR_RES1|VTCR_SH0_IS|VTCR_ORGN0_WBWA|VTCR_IRGN0_WBWA;
 
-#ifdef CONFIG_ARM_32
-    if ( p2m_ipa_bits < 40 )
-        panic("P2M: Not able to support %u-bit IPA at the moment\n",
-              p2m_ipa_bits);
-
-    printk("P2M: 40-bit IPA\n");
-    p2m_ipa_bits = 40;
-    val |= VTCR_T0SZ(0x18); /* 40 bit IPA */
-    val |= VTCR_SL0(0x1); /* P2M starts at first level */
-#else /* CONFIG_ARM_64 */
      static const struct {
          unsigned int pabits; /* Physical Address Size */
          unsigned int t0sz;   /* Desired T0SZ, minimum in comment */
@@ -2265,6 +2255,7 @@ void __init setup_virt_paging(void)
      } pa_range_info[] __initconst = {
          /* T0SZ minimum and SL0 maximum from ARM DDI 0487H.a Table D5-6 */
          /*      PA size, t0sz(min), root-order, sl0(max) */
+#ifdef CONFIG_ARM_64
          [0] = { 32,      32/*32*/,  0,          1 },
          [1] = { 36,      28/*28*/,  0,          1 },
          [2] = { 40,      24/*24*/,  1,          1 },
@@ -2273,11 +2264,28 @@ void __init setup_virt_paging(void)
          [5] = { 48,      16/*16*/,  0,          2 },
          [6] = { 52,      12/*12*/,  4,          2 },
          [7] = { 0 }  /* Invalid */
+#else
+        { 40,      24/*24*/,  1,          1 }
+#endif
      };
 
      unsigned int i;
      unsigned int pa_range = 0x10; /* Larger than any possible value */
 
+#ifdef CONFIG_ARM_32
+    /*
+     * Typecast pa_range_info[].t0sz into ARM_32 bit variant.
+     *

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2288

Quoted 10 lines
xen/arch/arm/p2m.c +10 -0
+     * VTCR.T0SZ is bits [3:0] and S(sign extension), bit[4] for ARM_32.
+     * Thus, pa_range_info[].t0sz is translated to its ARM_32 variant using
+     * struct bitfields.
+     */
+    struct
+    {
+        signed int val:5;
+    } t0sz_32;
+#else
+

no need for this empty line

Quoted 12 lines
xen/arch/arm/p2m.c +1 -0
      /*
       * Restrict "p2m_ipa_bits" if needed. As P2M table is always configured
       * with IPA bits == PA bits, compare against "pabits".
@@ -2291,6 +2299,7 @@ void __init setup_virt_paging(void)
       */
      if ( system_cpuinfo.mm64.vmid_bits == MM64_VMID_16_BITS_SUPPORT )
          max_vmid = MAX_VMID_16_BIT;
+#endif
 
      /* Choose suitable "pa_range" according to the resulted "p2m_ipa_bits". */
      for ( i = 0; i < ARRAY_SIZE(pa_range_info); i++ )

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2314

Quoted 2 lines
xen/arch/arm/p2m.c +0 -0
@@ -2304,26 +2313,41 @@ void __init setup_virt_paging(void)
 
      /* pa_range is 4 bits but we don't support all modes */

this comment makes sense really only on arm64 as it refers to PARange field of ID_AA64MMFR0_EL1.

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2321

Quoted 8 lines
xen/arch/arm/p2m.c +6 -1
      if ( pa_range >= ARRAY_SIZE(pa_range_info) || !pa_range_info[pa_range].pabits )
-        panic("Unknown encoding of ID_AA64MMFR0_EL1.PARange %x\n", pa_range);
+    {
+        /*
+         * In case of ARM_64, we do not support this encoding of
+         * ID_AA64MMFR0_EL1.PARange
+         */
+        panic("Unsupported value for p2m_ipa_bits = 0x%x\n", p2m_ipa_bits);

NIT: Putting variable names in messages visible by users is not a great idea IMO. "Unsupported IPA size" would read better. Furthermore, I do not think printing IPA size in hex is beneficial. I would use "%u bits" (i.e. 32 bits reads better than 0x20 bits).

Quoted 34 lines
xen/arch/arm/p2m.c +11 -2
+    }
 
+#ifdef CONFIG_ARM_64
      val |= VTCR_PS(pa_range);
      val |= VTCR_TG0_4K;
 
      /* Set the VS bit only if 16 bit VMID is supported. */
      if ( MAX_VMID == MAX_VMID_16_BIT )
          val |= VTCR_VS;
+#endif
+
      val |= VTCR_SL0(pa_range_info[pa_range].sl0);
      val |= VTCR_T0SZ(pa_range_info[pa_range].t0sz);
 
      p2m_root_order = pa_range_info[pa_range].root_order;
      p2m_root_level = 2 - pa_range_info[pa_range].sl0;
+
+#ifdef CONFIG_ARM_64
      p2m_ipa_bits = 64 - pa_range_info[pa_range].t0sz;
+#else
+    t0sz_32.val = pa_range_info[pa_range].t0sz;
+    p2m_ipa_bits = 32 - t0sz_32.val;
+#endif
 
      printk("P2M: %d-bit IPA with %d-bit PA and %d-bit VMID\n",
             p2m_ipa_bits,
             pa_range_info[pa_range].pabits,
             ( MAX_VMID == MAX_VMID_16_BIT ) ? 16 : 8);
-#endif
+
      printk("P2M: %d levels with order-%d root, VTCR 0x%"PRIregister"\n",
             4 - P2M_ROOT_LEVEL, P2M_ROOT_ORDER, val);
 
--
Quoted 3 lines

2.17.1

~Michal

#4 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

Hi Ayan,

Quoted 13 lines · Ayan Kumar Halder
Quoted 1 line · Michal Orzel

Hi Ayan,

Hi Michal,

Quoted 8 lines · Ayan Kumar Halder
Quoted 3 lines

Restructure the code so that one can use pa_range_info[] table for both ARM_32 as well as ARM_64.

I grepped for ARM_{32,64} in our code base and could not find any use in source files except for things introduced by this commit. While I'm ok with it in a commit message I think for consistency we should be using arm32/arm64 in the code.

Quoted 8 lines

AFAIU, arm32/arm64 refers to the Architecture. ARM_32/ARM_64 refers to the configuration.

If you see the original code (xen/arch/arm/include/asm/p2m.h, xen/arch/arm/p2m.c)

ARM_32/ARM_64 has been used.

Thus, I used ARM_32/ARM_64 in this commit. Let me know if it makes sense. The rest of your comments look sane to me.

In text, we commonly don't use the name of the config. Instead we use the name of the architecture (i.e. arm32/arm64) because this is a strict correspondence.

I agree with Michal, about using arm32/arm64 rather than ARM_32/ARM_64 in the comments at least to stay consistent with the rest of the code.

Assuming there is no other changes required in this patch (I haven't looked at it yet), then I am happy to handle this request on commit if you are OK with it.

Cheers,

Signature · 1 line
Julien Grall
#5 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

Quoted 1 line · Julien Grall

Hi Ayan,

Hi Julien,

Quoted 19 lines · Ayan Kumar Halder
Quoted 17 lines · Michal Orzel
Quoted 1 line

Hi Ayan,

Hi Michal,

Quoted 8 lines · Ayan Kumar Halder
Quoted 3 lines

Restructure the code so that one can use pa_range_info[] table for both ARM_32 as well as ARM_64.

I grepped for ARM_{32,64} in our code base and could not find any use in source files except for things introduced by this commit. While I'm ok with it in a commit message I think for consistency we should be using arm32/arm64 in the code.

AFAIU, arm32/arm64 refers to the Architecture. ARM_32/ARM_64 refers to the configuration.

If you see the original code (xen/arch/arm/include/asm/p2m.h, xen/arch/arm/p2m.c)

Quoted 7 lines
Quoted 4 lines

ARM_32/ARM_64 has been used.

Thus, I used ARM_32/ARM_64 in this commit. Let me know if it makes sense. The rest of your comments look sane to me.

In text, we commonly don't use the name of the config. Instead we use the name of the architecture (i.e. arm32/arm64) because this is a strict correspondence.

I agree with Michal, about using arm32/arm64 rather than ARM_32/ARM_64 in the comments at least to stay consistent with the rest of the code.

I see what you mean.

Quoted 2 lines

Assuming there is no other changes required in this patch (I haven't looked at it yet), then I am happy to handle this request on commit if you are OK with it.

I see that the rest of the comments from Michal are related to style (ie no functional changes). Thus, I am happy for you to fix them on commit.

- Ayan

Quoted 3 lines

Cheers,

Acked-by on v8 2/5
#11 lore raw

Re: [XEN v8 2/5] xen/arm: Introduce choice to enable 64/32 bit physical addressing

In reply to Ayan Kumar Halder

Hi Ayan,

Quoted 20 lines · Ayan Kumar Halder

Some Arm based hardware platforms which does not support LPAE (eg Cortex-R52), uses 32 bit physical addresses.
Also, users may choose to use 32 bits to represent physical addresses for optimization.

To support the above use cases, we have introduced arch independent config to choose if the physical address can be represented using 32 bits (PHYS_ADDR_T_32) or 64 bits (!PHYS_ADDR_T_32). For now only ARM_32 provides support to enable 32 bit physical addressing.

When PHYS_ADDR_T_32 is defined, PADDR_BITS is set to 32. Note that we use "unsigned long" (not "uint32_t") to denote the datatype of physical address. This is done to avoid using a cast each time PAGE_* macros are used on paddr_t. For eg PAGE_SIZE is defined as unsigned long. Thus, each time PAGE_SIZE is used with paddr_t, the result will be "unsigned long".
On 32-bit architecture, "unsigned long" is 32-bit wide. Thus, it can be used to denote physical address.

Quoted 6 lines

When PHYS_ADDR_T_32 is not defined for ARM_32, PADDR_BITS is set to 40. For ARM_64, PADDR_BITS is set to 48.
The last two are same as the current configuration used today on Xen.

Cheers,

Signature · 1 line
Julien Grall
Acked-by on v8 5/5
#16 lore raw

Re: [XEN v8 5/5] xen/arm: p2m: Enable support for 32bit IPA for ARM_32

In reply to Ayan Kumar Halder

Hi Ayan,

Quoted 23 lines · Ayan Kumar Halder

Refer ARM DDI 0406C.d ID040418, B3-1345,

"A stage 2 translation with an input address range of 31-34 bits can start the translation either:

- With a first-level lookup, accessing a first-level translation table with 2-16 entries.

- With a second-level lookup, accessing a set of concatenated second-level translation tables"

Thus, for 32 bit IPA, there will be no concatenated root level tables. So, the root-order is 0.

Also, Refer ARM DDI 0406C.d ID040418, B3-1348
"Determining the required first lookup level for stage 2 translations

For a stage 2 translation, the output address range from the stage 1 translations determines the required input address range for the stage 2 translation. The permitted values of VTCR.SL0 are:
0b00 Stage 2 translation lookup must start at the second level. 0b01 Stage 2 translation lookup must start at the first level.

Quoted 7 lines

VTCR.T0SZ must indicate the required input address range. The size of the input address region is 2^(32-T0SZ) bytes."

Thus VTCR.SL0 = 1 (maximum value) and VTCR.T0SZ = 0 when the size of input address region is 2^32 bytes.

Cheers,

Signature · 1 line
Julien Grall
#6 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

In reply to Michal Orzel

Hi Michal,

I notice you posted some comments but didn't add a Acked-by/Reviewed-by. Can you indicate if you are happy with the patch so long your comments are addressed?

If so, are you OK if I deal with them on commit?

Cheers,

Signature · 1 line
Julien Grall
Reviewed-by
#7 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

Hi Julien,

Quoted 9 lines · Julien Grall

Hi Michal,

I notice you posted some comments but didn't add a Acked-by/Reviewed-by. Can you indicate if you are happy with the patch so long your comments are addressed?

If so, are you OK if I deal with them on commit?

I thought I added my tag but clearly not. With the remarks addressed on commit:

~Michal

#9 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

In reply to Michal Orzel

Hi,

Quoted 24 lines · Michal Orzel
Quoted 12 lines

/*
* Restrict "p2m_ipa_bits" if needed. As P2M table is always configured
* with IPA bits == PA bits, compare against "pabits".

xen/arch/arm/p2m.c +1 -0
@@ -2291,6 +2299,7 @@ void __init setup_virt_paging(void)
       */
      if ( system_cpuinfo.mm64.vmid_bits == MM64_VMID_16_BITS_SUPPORT )
          max_vmid = MAX_VMID_16_BIT;
+#endif
 
      /* Choose suitable "pa_range" according to the resulted "p2m_ipa_bits". */
      for ( i = 0; i < ARRAY_SIZE(pa_range_info); i++ )

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2314

Quoted 2 lines
xen/arch/arm/p2m.c +0 -0
@@ -2304,26 +2313,41 @@ void __init setup_virt_paging(void)
 
      /* pa_range is 4 bits but we don't support all modes */

this comment makes sense really only on arm64 as it refers to PARange field of ID_AA64MMFR0_EL1.

xen/arch/arm/p2m.c void __init setup_virt_paging(void) · line 2321

Quoted 8 lines
xen/arch/arm/p2m.c +6 -1
      if ( pa_range >= ARRAY_SIZE(pa_range_info) || !pa_range_info[pa_range].pabits )
-        panic("Unknown encoding of ID_AA64MMFR0_EL1.PARange %x\n", pa_range);
+    {
+        /*
+         * In case of ARM_64, we do not support this encoding of
+         * ID_AA64MMFR0_EL1.PARange
+         */
+        panic("Unsupported value for p2m_ipa_bits = 0x%x\n", p2m_ipa_bits);
Quoted 3 lines

NIT: Putting variable names in messages visible by users is not a great idea IMO. "Unsupported IPA size" would read better. Furthermore, I do not think printing IPA size in hex is beneficial. I would use "%u bits" (i.e. 32 bits reads better than 0x20 bits).

I went with the following:

- /* pa_range is 4 bits but we don't support all modes */
+ /* Check if we found the associated entry in the array */

     if ( pa_range >= ARRAY_SIZE(pa_range_info) || 
!pa_range_info[pa_range].pabits )
-    {
-        /*
-         * In case of ARM_64, we do not support this encoding of
-         * ID_AA64MMFR0_EL1.PARange
-         */

- panic("Unsupported value for p2m_ipa_bits = 0x%x\n", p2m_ipa_bits);
- }
+ panic("%-bit P2M is not supported\n", p2m_ipa_bits);

This should be generic enough.

Signature · 1 line
Julien Grall
#8 lore raw

Re: [XEN v8 1/5] xen/arm: p2m: Use the pa_range_info table to support ARM_32 and ARM_64

In reply to Michal Orzel
Quoted 7 lines · Michal Orzel

Hi Julien,

Quoted 4 lines · Julien Grall

Hi Michal,

Quoted 7 lines
Quoted 5 lines

I notice you posted some comments but didn't add a Acked-by/Reviewed-by. Can you indicate if you are happy with the patch so long your comments are addressed?

If so, are you OK if I deal with them on commit?

I thought I added my tag but clearly not. With the remarks addressed on commit:

Thanks. The series is now committed.

Cheers,

Signature · 1 line
Julien Grall