* [PATCH v3 0/2] xen/arm: Split MMU code as the prepration of MPU work form Arm32
@ 2023-11-07 11:02 Ayan Kumar Halder
2023-11-07 11:02 ` [PATCH v3 1/2] xen/arm32: head: Introduce enable_{boot,secondary}_cpu_mm() Ayan Kumar Halder
2023-11-07 11:02 ` [PATCH v3 2/2] xen/arm32: head Split and move MMU-specific head.S to mmu/head.S Ayan Kumar Halder
0 siblings, 2 replies; 5+ messages in thread
From: Ayan Kumar Halder @ 2023-11-07 11:02 UTC (permalink / raw)
To: xen-devel
Cc: sstabellini, julien, bertrand.marquis, michal.orzel, henry.wang,
Volodymyr_Babchuk, Ayan Kumar Halder
Hi,
These are the set of patches based on top of
"[PATCH v8 0/8] xen/arm: Split MMU code as the prepration of MPU work".
This is the preparation work to add MPU support on Arm32.
There are two more dependencies for this series :-
1. "[XEN v1 1/4] xen/arm: arm32: Move pt_enforce_wxn() so that it can be bundled with other MMU functionality"
is merged into "[PATCH v8 3/8] xen/arm: Fold mmu_init_secondary_cpu() to head.S"
as per the discussion on [1].
2. "[XEN v4] xen/arm32: head: Replace load_paddr with adr_l when they are equivalent"
[1] - https://lore.kernel.org/all/f098a07d-fa19-4b40-bfac-7b121524382a@xen.org/#t
Changes from :-
v1 - Dropped "[XEN v1 1/4] xen/arm: arm32: Move pt_enforce_wxn() so that it can be bundled with other MMU functionality"
and "[XEN v1 4/4] xen/arm: traps.c: Enclose VMSA specific registers within CONFIG_MMU".
v2 - Changes mentioned in individual patches.
Ayan Kumar Halder (2):
xen/arm32: head: Introduce enable_{boot,secondary}_cpu_mm()
xen/arm32: head Split and move MMU-specific head.S to mmu/head.S
xen/arch/arm/arm32/head.S | 578 +-----------------------
xen/arch/arm/arm32/mmu/Makefile | 1 +
xen/arch/arm/arm32/mmu/head.S | 573 +++++++++++++++++++++++
xen/arch/arm/include/asm/arm32/macros.h | 58 ++-
4 files changed, 641 insertions(+), 569 deletions(-)
create mode 100644 xen/arch/arm/arm32/mmu/head.S
--
2.25.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v3 1/2] xen/arm32: head: Introduce enable_{boot,secondary}_cpu_mm()
2023-11-07 11:02 [PATCH v3 0/2] xen/arm: Split MMU code as the prepration of MPU work form Arm32 Ayan Kumar Halder
@ 2023-11-07 11:02 ` Ayan Kumar Halder
2023-11-07 11:23 ` Michal Orzel
2023-11-07 11:02 ` [PATCH v3 2/2] xen/arm32: head Split and move MMU-specific head.S to mmu/head.S Ayan Kumar Halder
1 sibling, 1 reply; 5+ messages in thread
From: Ayan Kumar Halder @ 2023-11-07 11:02 UTC (permalink / raw)
To: xen-devel
Cc: sstabellini, julien, bertrand.marquis, michal.orzel, henry.wang,
Volodymyr_Babchuk, Ayan Kumar Halder, Michal Orzel, Julien Grall
All the MMU related functionality have been clubbed together in
enable_boot_cpu_mm() for booting primary cpu and enable_secondary_cpu_mm() for
booting secondary cpus.
This is done in preparation for moving the code related to MMU in MMU specific
file and in order to support non MMU cpus in future.
This is based on d2f8df5b3ede ("xen/arm64: head.S: Introduce enable_{boot,secondary}_cpu_mm()").
Signed-off-by: Ayan Kumar Halder <ayan.kumar.halder@amd.com>
Reviewed-by: Michal Orzel <michal.orzel@amd.com>
Acked-by: Julien Grall <jgrall@amazon.com>
---
Changes from :-
v1 - 1. Added a proper commit message.
2. Some style and other fixes suggested in v1.
v2 - 1. Updated the comment on top of enable_boot_cpu_mm() and
enable_secondary_cpu_mm() ie mentioned the input and output registers.
2. Updated the comment inside enable_boot_cpu_mm().
xen/arch/arm/arm32/head.S | 102 ++++++++++++++++++++++++++++++--------
1 file changed, 80 insertions(+), 22 deletions(-)
diff --git a/xen/arch/arm/arm32/head.S b/xen/arch/arm/arm32/head.S
index 2d7e690bf5..2204ea6dce 100644
--- a/xen/arch/arm/arm32/head.S
+++ b/xen/arch/arm/arm32/head.S
@@ -201,13 +201,11 @@ past_zImage:
bl check_cpu_mode
bl cpu_init
- bl create_page_tables
- /* Address in the runtime mapping to jump to after the MMU is enabled */
mov_w lr, primary_switched
- b enable_mmu
+ b enable_boot_cpu_mm
+
primary_switched:
- bl setup_fixmap
#ifdef CONFIG_EARLY_PRINTK
/* Use a virtual address to access the UART. */
mov_w r11, EARLY_UART_VIRTUAL_ADDRESS
@@ -249,27 +247,11 @@ GLOBAL(init_secondary)
#endif
bl check_cpu_mode
bl cpu_init
- bl create_page_tables
- /* Address in the runtime mapping to jump to after the MMU is enabled */
mov_w lr, secondary_switched
- b enable_mmu
-secondary_switched:
- /*
- * Non-boot CPUs need to move on to the proper pagetables, which were
- * setup in prepare_secondary_mm.
- *
- * XXX: This is not compliant with the Arm Arm.
- */
- mov_w r4, init_ttbr /* VA of HTTBR value stashed by CPU 0 */
- ldrd r4, r5, [r4] /* Actual value */
- dsb
- mcrr CP64(r4, r5, HTTBR)
- dsb
- isb
- flush_xen_tlb_local r0
- pt_enforce_wxn r0
+ b enable_secondary_cpu_mm
+secondary_switched:
#ifdef CONFIG_EARLY_PRINTK
/* Use a virtual address to access the UART. */
mov_w r11, EARLY_UART_VIRTUAL_ADDRESS
@@ -692,6 +674,82 @@ ready_to_switch:
mov pc, lr
ENDPROC(switch_to_runtime_mapping)
+/*
+ * Enable mm (turn on the data cache and the MMU) for secondary CPUs.
+ * The function will return to the virtual address provided in LR (e.g. the
+ * runtime mapping).
+ *
+ * Inputs:
+ * r9 : paddr(start)
+ * r10: phys offset
+ * lr : Virtual address to return to.
+ *
+ * Output:
+ * r12: Was a temporary mapping created?
+ *
+ * Clobbers r0 - r6
+ */
+enable_secondary_cpu_mm:
+ mov r6, lr
+
+ bl create_page_tables
+
+ /* Address in the runtime mapping to jump to after the MMU is enabled */
+ mov_w lr, 1f
+ b enable_mmu
+1:
+ /*
+ * Non-boot CPUs need to move on to the proper pagetables, which were
+ * setup in prepare_secondary_mm.
+ *
+ * XXX: This is not compliant with the Arm Arm.
+ */
+ mov_w r4, init_ttbr /* VA of HTTBR value stashed by CPU 0 */
+ ldrd r4, r5, [r4] /* Actual value */
+ dsb
+ mcrr CP64(r4, r5, HTTBR)
+ dsb
+ isb
+ flush_xen_tlb_local r0
+ pt_enforce_wxn r0
+
+ /* Return to the virtual address requested by the caller. */
+ mov pc, r6
+ENDPROC(enable_secondary_cpu_mm)
+
+/*
+ * Enable mm (turn on the data cache and the MMU) for the boot CPU.
+ * The function will return to the virtual address provided in LR (e.g. the
+ * runtime mapping).
+ *
+ * Inputs:
+ * r9 : paddr(start)
+ * r10: phys offset
+ * lr : Virtual address to return to.
+ *
+ * Output:
+ * r12: Was a temporary mapping created?
+ *
+ * Clobbers r0 - r6
+ */
+enable_boot_cpu_mm:
+ mov r6, lr
+
+ bl create_page_tables
+
+ /* Address in the runtime mapping to jump to after the MMU is enabled */
+ mov_w lr, 1f
+ b enable_mmu
+1:
+ /*
+ * Prepare the fixmap. The function will return to the virtual address
+ * requested by the caller.
+ */
+ mov lr, r6
+
+ b setup_fixmap
+ENDPROC(enable_boot_cpu_mm)
+
/*
* Remove the 1:1 map from the page-tables. It is not easy to keep track
* where the 1:1 map was mapped, so we will look for the top-level entry
--
2.25.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH v3 2/2] xen/arm32: head Split and move MMU-specific head.S to mmu/head.S
2023-11-07 11:02 [PATCH v3 0/2] xen/arm: Split MMU code as the prepration of MPU work form Arm32 Ayan Kumar Halder
2023-11-07 11:02 ` [PATCH v3 1/2] xen/arm32: head: Introduce enable_{boot,secondary}_cpu_mm() Ayan Kumar Halder
@ 2023-11-07 11:02 ` Ayan Kumar Halder
2023-11-07 12:26 ` Michal Orzel
1 sibling, 1 reply; 5+ messages in thread
From: Ayan Kumar Halder @ 2023-11-07 11:02 UTC (permalink / raw)
To: xen-devel
Cc: sstabellini, julien, bertrand.marquis, michal.orzel, henry.wang,
Volodymyr_Babchuk, Ayan Kumar Halder
The MMU specific code in head.S will not be used on MPU systems.
Instead of introducing more #ifdefs which will bring complexity
to the code, move MMU related code to mmu/head.S and keep common
code in head.S. Two notes while moving:
- As "fail" in original head.S is very simple and this name is too
easy to be conflicted, duplicate it in mmu/head.S instead of
exporting it.
- Realigned ".macro ret" so that the alignment matches to the other
macros.
- Rename puts to asm_puts, putn to asm_putn (this denotes that the
macros are used within the context of assembly only).
- Use ENTRY() for enable_secondary_cpu_mm, enable_boot_cpu_mm,
setup_fixmap, asm_puts, asm_putn as they will be used externally.
Also move the assembly macros shared by head.S and mmu/head.S to
macros.h.
This is based on 6734327d76be ("xen/arm64: Split and move MMU-specific head.S to mmu/head.S").
Signed-off-by: Ayan Kumar Halder <ayan.kumar.halder@amd.com>
---
Changes from v1 :-
1. Added a commit message
2. Moved load_paddr to mmu/head.S
v2 :-
1. Renamed puts to asm_puts and putn to asm_putn. Exported asm_putn().
2. Moved XEN_TEMPORARY_OFFSET to head.S.
3. Some style issues.
xen/arch/arm/arm32/head.S | 628 +-----------------------
xen/arch/arm/arm32/mmu/Makefile | 1 +
xen/arch/arm/arm32/mmu/head.S | 573 +++++++++++++++++++++
xen/arch/arm/include/asm/arm32/macros.h | 58 ++-
4 files changed, 637 insertions(+), 623 deletions(-)
create mode 100644 xen/arch/arm/arm32/mmu/head.S
diff --git a/xen/arch/arm/arm32/head.S b/xen/arch/arm/arm32/head.S
index 2204ea6dce..1448d092d1 100644
--- a/xen/arch/arm/arm32/head.S
+++ b/xen/arch/arm/arm32/head.S
@@ -22,86 +22,10 @@
#define ZIMAGE_MAGIC_NUMBER 0x016f2818
-#define PT_PT 0xf7f /* nG=1 AF=1 SH=11 AP=01 NS=1 ATTR=111 T=1 P=1 */
-#define PT_MEM 0xf7d /* nG=1 AF=1 SH=11 AP=01 NS=1 ATTR=111 T=0 P=1 */
-#define PT_MEM_L3 0xf7f /* nG=1 AF=1 SH=11 AP=01 NS=1 ATTR=111 T=1 P=1 */
-#define PT_DEV 0xe71 /* nG=1 AF=1 SH=10 AP=01 NS=1 ATTR=100 T=0 P=1 */
-#define PT_DEV_L3 0xe73 /* nG=1 AF=1 SH=10 AP=01 NS=1 ATTR=100 T=1 P=1 */
-
-#define PT_UPPER(x) (PT_##x & 0xf00)
-#define PT_LOWER(x) (PT_##x & 0x0ff)
-
-/* Convenience defines to get slot used by Xen mapping. */
-#define XEN_FIRST_SLOT first_table_offset(XEN_VIRT_START)
-#define XEN_SECOND_SLOT second_table_offset(XEN_VIRT_START)
-
-/* Offset between the early boot xen mapping and the runtime xen mapping */
-#define XEN_TEMPORARY_OFFSET (TEMPORARY_XEN_VIRT_START - XEN_VIRT_START)
-
#if defined(CONFIG_EARLY_PRINTK) && defined(CONFIG_EARLY_PRINTK_INC)
#include CONFIG_EARLY_PRINTK_INC
#endif
-/*
- * Move an immediate constant into a 32-bit register using movw/movt
- * instructions.
- */
-.macro mov_w reg, word
- movw \reg, #:lower16:\word
- movt \reg, #:upper16:\word
-.endm
-
-/*
- * Pseudo-op for PC relative adr <reg>, <symbol> where <symbol> is
- * within the range +/- 4GB of the PC.
- *
- * @dst: destination register
- * @sym: name of the symbol
- */
-.macro adr_l, dst, sym
- mov_w \dst, \sym - .Lpc\@
- .set .Lpc\@, .+ 8 /* PC bias */
- add \dst, \dst, pc
-.endm
-
-.macro load_paddr rb, sym
- mov_w \rb, \sym
- add \rb, \rb, r10
-.endm
-
-/*
- * Flush local TLBs
- *
- * @tmp: Scratch register
- *
- * See asm/arm32/flushtlb.h for the explanation of the sequence.
- */
-.macro flush_xen_tlb_local tmp
- dsb nshst
- mcr CP32(\tmp, TLBIALLH)
- dsb nsh
- isb
-.endm
-
-/*
- * Enforce Xen page-tables do not contain mapping that are both
- * Writable and eXecutables.
- *
- * This should be called on each secondary CPU.
- */
-.macro pt_enforce_wxn tmp
- mrc CP32(\tmp, HSCTLR)
- orr \tmp, \tmp, #SCTLR_Axx_ELx_WXN
- dsb
- mcr CP32(\tmp, HSCTLR)
- /*
- * The TLBs may cache SCTLR_EL2.WXN. So ensure it is synchronized
- * before flushing the TLBs.
- */
- isb
- flush_xen_tlb_local \tmp
-.endm
-
/*
* Common register usage in this file:
* r0 -
@@ -121,38 +45,6 @@
* r14 - LR
* r15 - PC
*/
-#ifdef CONFIG_EARLY_PRINTK
-/*
- * Macro to print a string to the UART, if there is one.
- *
- * Clobbers r0 - r3
- */
-#define PRINT(_s) \
- mov r3, lr ;\
- adr_l r0, 98f ;\
- bl puts ;\
- mov lr, r3 ;\
- RODATA_STR(98, _s)
-
-/*
- * Macro to print the value of register \rb
- *
- * Clobbers r0 - r4
- */
-.macro print_reg rb
- mov r0, \rb
- mov r4, lr
- bl putn
- mov lr, r4
-.endm
-
-#else /* CONFIG_EARLY_PRINTK */
-#define PRINT(s)
-
-.macro print_reg rb
-.endm
-
-#endif /* !CONFIG_EARLY_PRINTK */
.section .text.header, "ax", %progbits
.arm
@@ -355,480 +247,6 @@ cpu_init_done:
mov pc, r5 /* Return address is in r5 */
ENDPROC(cpu_init)
-/*
- * Macro to find the slot number at a given page-table level
- *
- * slot: slot computed
- * virt: virtual address
- * lvl: page-table level
- *
- * Note that ubxf is unpredictable when the end bit is above 32-bit. So we
- * can't use it for first level offset.
- */
-.macro get_table_slot, slot, virt, lvl
- .if \lvl == 1
- lsr \slot, \virt, #XEN_PT_LEVEL_SHIFT(\lvl)
- .else
- ubfx \slot, \virt, #XEN_PT_LEVEL_SHIFT(\lvl), #XEN_PT_LPAE_SHIFT
- .endif
-.endm
-
-/*
- * Macro to create a page table entry in \ptbl to \tbl (physical
- * address)
- *
- * ptbl: table symbol where the entry will be created
- * tbl: physical address of the table to point to
- * virt: virtual address
- * lvl: page-table level
- *
- * Preserves \virt
- * Clobbers \tbl, r1 - r3
- *
- * Note that \tbl and \virt should be in a register other than r1 - r3
- */
-.macro create_table_entry_from_paddr, ptbl, tbl, virt, lvl
- get_table_slot r1, \virt, \lvl /* r1 := slot in \tbl */
- lsl r1, r1, #3 /* r1 := slot offset in \tbl */
-
- movw r2, #PT_PT /* r2:r3 := right for linear PT */
- orr r2, r2, \tbl /* + \tbl paddr */
- mov r3, #0
-
- adr_l \tbl, \ptbl /* \tbl := (v,p)addr of \ptbl */
-
- strd r2, r3, [\tbl, r1]
-.endm
-
-
-/*
- * Macro to create a page table entry in \ptbl to \tbl (symbol)
- *
- * ptbl: table symbol where the entry will be created
- * tbl: table symbol to point to
- * virt: virtual address
- * lvl: page-table level
- *
- * Preserves \virt
- * Clobbers r1 - r4
- *
- * Also use r10 for the phys offset.
- *
- * Note that \virt should be in a register other than r1 - r4
- */
-.macro create_table_entry, ptbl, tbl, virt, lvl
- load_paddr r4, \tbl
- create_table_entry_from_paddr \ptbl, r4, \virt, \lvl
- .endm
-
-/*
- * Macro to create a mapping entry in \tbl to \paddr. Only mapping in 3rd
- * level table (i.e page granularity) is supported.
- *
- * ptbl: table symbol where the entry will be created
- * virt: virtual address
- * phys: physical address
- * type: mapping type. If not specified it will be normal memory (PT_MEM_L3)
- *
- * Preserves \virt, \phys
- * Clobbers r1 - r4
- *
- * Note that \virt and \paddr should be in other registers than r1 - r4
- * and be distinct.
- */
-.macro create_mapping_entry, ptbl, virt, phys, type=PT_MEM_L3
- mov_w r2, XEN_PT_LPAE_ENTRY_MASK
- lsr r1, \virt, #THIRD_SHIFT
- and r1, r1, r2 /* r1 := slot in \tlb */
- lsl r1, r1, #3 /* r1 := slot offset in \tlb */
-
- lsr r4, \phys, #THIRD_SHIFT
- lsl r4, r4, #THIRD_SHIFT /* r4 := PAGE_ALIGNED(phys) */
-
- movw r2, #\type /* r2:r3 := right for section PT */
- orr r2, r2, r4 /* + PAGE_ALIGNED(phys) */
- mov r3, #0
-
- adr_l r4, \ptbl
-
- strd r2, r3, [r4, r1]
-.endm
-
-/*
- * Rebuild the boot pagetable's first-level entries. The structure
- * is described in mm.c.
- *
- * After the CPU enables paging it will add the fixmap mapping
- * to these page tables, however this may clash with the 1:1
- * mapping. So each CPU must rebuild the page tables here with
- * the 1:1 in place.
- *
- * Inputs:
- * r9 : paddr(start)
- * r10: phys offset
- *
- * Output:
- * r12: Was a temporary mapping created?
- *
- * Clobbers r0 - r5
- */
-create_page_tables:
- /* Prepare the page-tables for mapping Xen */
- mov_w r0, XEN_VIRT_START
-
- /*
- * We need to use a stash register because
- * create_table_entry_paddr() will clobber the register storing
- * the physical address of the table to point to.
- */
- adr_l r5, boot_third
- mov_w r4, XEN_VIRT_START
-.rept XEN_NR_ENTRIES(2)
- mov r0, r5 /* r0 := paddr(l3 table) */
- create_table_entry_from_paddr boot_second, r0, r4, 2
- add r4, r4, #XEN_PT_LEVEL_SIZE(2) /* r4 := Next vaddr */
- add r5, r5, #PAGE_SIZE /* r5 := Next table */
-.endr
-
- /*
- * Find the size of Xen in pages and multiply by the size of a
- * PTE. This will then be compared in the mapping loop below.
- *
- * Note the multiplication is just to avoid using an extra
- * register/instruction per iteration.
- */
- mov_w r0, _start /* r0 := vaddr(_start) */
- mov_w r1, _end /* r1 := vaddr(_end) */
- sub r0, r1, r0 /* r0 := effective size of Xen */
- lsr r0, r0, #PAGE_SHIFT /* r0 := Number of pages for Xen */
- lsl r0, r0, #3 /* r0 := Number of pages * PTE size */
-
- /* Setup boot_third: */
- adr_l r4, boot_third
-
- lsr r2, r9, #THIRD_SHIFT /* Base address for 4K mapping */
- lsl r2, r2, #THIRD_SHIFT
- orr r2, r2, #PT_UPPER(MEM_L3) /* r2:r3 := map */
- orr r2, r2, #PT_LOWER(MEM_L3)
- mov r3, #0x0
-
- /* ... map of vaddr(start) in boot_third */
- mov r1, #0
-1: strd r2, r3, [r4, r1] /* Map vaddr(start) */
- add r2, r2, #PAGE_SIZE /* Next page */
- add r1, r1, #8 /* Next slot */
- cmp r1, r0 /* Loop until we map all of Xen */
- blo 1b
-
- /*
- * Setup the 1:1 mapping so we can turn the MMU on. Note that
- * only the first page of Xen will be part of the 1:1 mapping.
- */
- create_table_entry boot_pgtable, boot_second_id, r9, 1
- create_table_entry boot_second_id, boot_third_id, r9, 2
- create_mapping_entry boot_third_id, r9, r9
-
- /*
- * Find the first slot used. If the slot is not the same
- * as TEMPORARY_AREA_FIRST_SLOT, then we will want to switch
- * to the temporary mapping before jumping to the runtime
- * virtual mapping.
- */
- get_table_slot r1, r9, 1 /* r1 := first slot */
- cmp r1, #TEMPORARY_AREA_FIRST_SLOT
- bne use_temporary_mapping
-
- mov_w r0, XEN_VIRT_START
- create_table_entry boot_pgtable, boot_second, r0, 1
- mov r12, #0 /* r12 := temporary mapping not created */
- mov pc, lr
-
-use_temporary_mapping:
- /*
- * The identity mapping is not using the first slot
- * TEMPORARY_AREA_FIRST_SLOT. Create a temporary mapping.
- * See switch_to_runtime_mapping for more details.
- */
- PRINT("- Create temporary mapping -\r\n")
-
- /* Map boot_second (cover Xen mappings) to the temporary 1st slot */
- mov_w r0, TEMPORARY_XEN_VIRT_START
- create_table_entry boot_pgtable, boot_second, r0, 1
-
- mov r12, #1 /* r12 := temporary mapping created */
- mov pc, lr
-ENDPROC(create_page_tables)
-
-/*
- * Turn on the Data Cache and the MMU. The function will return
- * to the virtual address provided in LR (e.g. the runtime mapping).
- *
- * Inputs:
- * r9 : paddr(start)
- * r12 : Was the temporary mapping created?
- * lr : Virtual address to return to
- *
- * Clobbers r0 - r5
- */
-enable_mmu:
- PRINT("- Turning on paging -\r\n")
-
- /*
- * The state of the TLBs is unknown before turning on the MMU.
- * Flush them to avoid stale one.
- */
- flush_xen_tlb_local r0
-
- /* Write Xen's PT's paddr into the HTTBR */
- adr_l r0, boot_pgtable
- mov r1, #0 /* r0:r1 is paddr (boot_pagetable) */
- mcrr CP64(r0, r1, HTTBR)
- isb
-
- mrc CP32(r0, HSCTLR)
- /* Enable MMU and D-cache */
- orr r0, r0, #(SCTLR_Axx_ELx_M|SCTLR_Axx_ELx_C)
- dsb /* Flush PTE writes and finish reads */
- mcr CP32(r0, HSCTLR) /* now paging is enabled */
- isb /* Now, flush the icache */
-
- /*
- * The MMU is turned on and we are in the 1:1 mapping. Switch
- * to the runtime mapping.
- */
- mov r5, lr /* Save LR before overwritting it */
- mov_w lr, 1f /* Virtual address in the runtime mapping */
- b switch_to_runtime_mapping
-1:
- mov lr, r5 /* Restore LR */
- /*
- * At this point, either the 1:1 map or the temporary mapping
- * will be present. The former may clash with other parts of the
- * Xen virtual memory layout. As both of them are not used
- * anymore, remove them completely to avoid having to worry
- * about replacing existing mapping afterwards.
- *
- * On return this will jump to the virtual address requested by
- * the caller.
- */
- teq r12, #0
- beq remove_identity_mapping
- b remove_temporary_mapping
-ENDPROC(enable_mmu)
-
-/*
- * Switch to the runtime mapping. The logic depends on whether the
- * runtime virtual region is clashing with the physical address
- *
- * - If it is not clashing, we can directly jump to the address in
- * the runtime mapping.
- * - If it is clashing, create_page_tables() would have mapped Xen to
- * a temporary virtual address. We need to switch to the temporary
- * mapping so we can remove the identity mapping and map Xen at the
- * correct position.
- *
- * Inputs
- * r9: paddr(start)
- * r12: Was a temporary mapping created?
- * lr: Address in the runtime mapping to jump to
- *
- * Clobbers r0 - r4
- */
-switch_to_runtime_mapping:
- /*
- * Jump to the runtime mapping if the virt and phys are not
- * clashing
- */
- teq r12, #0
- beq ready_to_switch
-
- /* We are still in the 1:1 mapping. Jump to the temporary Virtual address. */
- mov_w r0, 1f
- add r0, r0, #XEN_TEMPORARY_OFFSET /* r0 := address in temporary mapping */
- mov pc, r0
-
-1:
- /* Remove boot_second_id */
- mov r2, #0
- mov r3, #0
- adr_l r0, boot_pgtable
- get_table_slot r1, r9, 1 /* r1 := first slot */
- lsl r1, r1, #3 /* r1 := first slot offset */
- strd r2, r3, [r0, r1]
-
- flush_xen_tlb_local r0
-
- /* Map boot_second into boot_pgtable */
- mov_w r0, XEN_VIRT_START
- create_table_entry boot_pgtable, boot_second, r0, 1
-
- /* Ensure any page table updates are visible before continuing */
- dsb nsh
- /*
- * The function will return on the runtime mapping. So we want
- * to prevent instruction fetch before the dsb completes.
- */
- isb
-
-ready_to_switch:
- mov pc, lr
-ENDPROC(switch_to_runtime_mapping)
-
-/*
- * Enable mm (turn on the data cache and the MMU) for secondary CPUs.
- * The function will return to the virtual address provided in LR (e.g. the
- * runtime mapping).
- *
- * Inputs:
- * r9 : paddr(start)
- * r10: phys offset
- * lr : Virtual address to return to.
- *
- * Output:
- * r12: Was a temporary mapping created?
- *
- * Clobbers r0 - r6
- */
-enable_secondary_cpu_mm:
- mov r6, lr
-
- bl create_page_tables
-
- /* Address in the runtime mapping to jump to after the MMU is enabled */
- mov_w lr, 1f
- b enable_mmu
-1:
- /*
- * Non-boot CPUs need to move on to the proper pagetables, which were
- * setup in prepare_secondary_mm.
- *
- * XXX: This is not compliant with the Arm Arm.
- */
- mov_w r4, init_ttbr /* VA of HTTBR value stashed by CPU 0 */
- ldrd r4, r5, [r4] /* Actual value */
- dsb
- mcrr CP64(r4, r5, HTTBR)
- dsb
- isb
- flush_xen_tlb_local r0
- pt_enforce_wxn r0
-
- /* Return to the virtual address requested by the caller. */
- mov pc, r6
-ENDPROC(enable_secondary_cpu_mm)
-
-/*
- * Enable mm (turn on the data cache and the MMU) for the boot CPU.
- * The function will return to the virtual address provided in LR (e.g. the
- * runtime mapping).
- *
- * Inputs:
- * r9 : paddr(start)
- * r10: phys offset
- * lr : Virtual address to return to.
- *
- * Output:
- * r12: Was a temporary mapping created?
- *
- * Clobbers r0 - r6
- */
-enable_boot_cpu_mm:
- mov r6, lr
-
- bl create_page_tables
-
- /* Address in the runtime mapping to jump to after the MMU is enabled */
- mov_w lr, 1f
- b enable_mmu
-1:
- /*
- * Prepare the fixmap. The function will return to the virtual address
- * requested by the caller.
- */
- mov lr, r6
-
- b setup_fixmap
-ENDPROC(enable_boot_cpu_mm)
-
-/*
- * Remove the 1:1 map from the page-tables. It is not easy to keep track
- * where the 1:1 map was mapped, so we will look for the top-level entry
- * exclusive to the 1:1 map and remove it.
- *
- * Inputs:
- * r9 : paddr(start)
- *
- * Clobbers r0 - r3
- */
-remove_identity_mapping:
- /* r2:r3 := invalid page-table entry */
- mov r2, #0x0
- mov r3, #0x0
-
- /* Find the first slot used and remove it */
- get_table_slot r1, r9, 1 /* r1 := first slot */
- mov_w r0, boot_pgtable /* r0 := root table */
- lsl r1, r1, #3 /* r1 := Slot offset */
- strd r2, r3, [r0, r1]
-
- flush_xen_tlb_local r0
- mov pc, lr
-ENDPROC(remove_identity_mapping)
-
-/*
- * Remove the temporary mapping of Xen starting at TEMPORARY_XEN_VIRT_START.
- *
- * Clobbers r0 - r3
- */
-remove_temporary_mapping:
- /* r2:r3 := invalid page-table entry */
- mov r2, #0
- mov r3, #0
-
- adr_l r0, boot_pgtable
- mov_w r1, TEMPORARY_XEN_VIRT_START
- get_table_slot r1, r1, 1 /* r1 := first slot */
- lsl r1, r1, #3 /* r1 := first slot offset */
- strd r2, r3, [r0, r1]
-
- flush_xen_tlb_local r0
-
- mov pc, lr
-ENDPROC(remove_temporary_mapping)
-
-/*
- * Map the UART in the fixmap (when earlyprintk is used) and hook the
- * fixmap table in the page tables.
- *
- * The fixmap cannot be mapped in create_page_tables because it may
- * clash with the 1:1 mapping.
- *
- * Inputs:
- * r10: Physical offset
- * r11: Early UART base physical address
- *
- * Clobbers r0 - r4
- */
-setup_fixmap:
-#if defined(CONFIG_EARLY_PRINTK)
- /* Add UART to the fixmap table */
- mov_w r0, EARLY_UART_VIRTUAL_ADDRESS
- create_mapping_entry xen_fixmap, r0, r11, type=PT_DEV_L3
-#endif
- /* Map fixmap into boot_second */
- mov_w r0, FIXMAP_ADDR(0)
- create_table_entry boot_second, xen_fixmap, r0, 2
- /* Ensure any page table updates made above have occurred. */
- dsb nshst
- /*
- * The fixmap area will be used soon after. So ensure no hardware
- * translation happens before the dsb completes.
- */
- isb
-
- mov pc, lr
-ENDPROC(setup_fixmap)
-
/*
* Setup the initial stack and jump to the C world
*
@@ -856,38 +274,6 @@ fail: PRINT("- Boot failed -\r\n")
b 1b
ENDPROC(fail)
-/*
- * Switch TTBR
- * r1:r0 ttbr
- *
- * TODO: This code does not comply with break-before-make.
- */
-ENTRY(switch_ttbr)
- dsb /* Ensure the flushes happen before
- * continuing */
- isb /* Ensure synchronization with previous
- * changes to text */
- mcr CP32(r0, TLBIALLH) /* Flush hypervisor TLB */
- mcr CP32(r0, ICIALLU) /* Flush I-cache */
- mcr CP32(r0, BPIALL) /* Flush branch predictor */
- dsb /* Ensure completion of TLB+BP flush */
- isb
-
- mcrr CP64(r0, r1, HTTBR)
-
- dsb /* ensure memory accesses do not cross
- * over the TTBR0 write */
- isb /* Ensure synchronization with previous
- * changes to text */
- mcr CP32(r0, TLBIALLH) /* Flush hypervisor TLB */
- mcr CP32(r0, ICIALLU) /* Flush I-cache */
- mcr CP32(r0, BPIALL) /* Flush branch predictor */
- dsb /* Ensure completion of TLB+BP flush */
- isb
-
- mov pc, lr
-ENDPROC(switch_ttbr)
-
#ifdef CONFIG_EARLY_PRINTK
/*
* Initialize the UART. Should only be called on the boot CPU.
@@ -912,14 +298,14 @@ ENDPROC(init_uart)
* r11: Early UART base address
* Clobbers r0-r1
*/
-puts:
+ENTRY(asm_puts)
early_uart_ready r11, r1
ldrb r1, [r0], #1 /* Load next char */
teq r1, #0 /* Exit on nul */
moveq pc, lr
early_uart_transmit r11, r1
- b puts
-ENDPROC(puts)
+ b asm_puts
+ENDPROC(asm_puts)
/*
* Print a 32-bit number in hex. Specific to the PL011 UART.
@@ -927,7 +313,7 @@ ENDPROC(puts)
* r11: Early UART base address
* Clobbers r0-r3
*/
-putn:
+ENTRY(asm_putn)
adr_l r1, hex
mov r3, #8
1:
@@ -939,7 +325,7 @@ putn:
subs r3, r3, #1
bne 1b
mov pc, lr
-ENDPROC(putn)
+ENDPROC(asm_putn)
RODATA_STR(hex, "0123456789abcdef")
@@ -947,8 +333,8 @@ RODATA_STR(hex, "0123456789abcdef")
ENTRY(early_puts)
init_uart:
-puts:
-putn: mov pc, lr
+asm_puts:
+asm_putn: mov pc, lr
#endif /* !CONFIG_EARLY_PRINTK */
diff --git a/xen/arch/arm/arm32/mmu/Makefile b/xen/arch/arm/arm32/mmu/Makefile
index b18cec4836..a8a750a3d0 100644
--- a/xen/arch/arm/arm32/mmu/Makefile
+++ b/xen/arch/arm/arm32/mmu/Makefile
@@ -1 +1,2 @@
+obj-y += head.o
obj-y += mm.o
diff --git a/xen/arch/arm/arm32/mmu/head.S b/xen/arch/arm/arm32/mmu/head.S
new file mode 100644
index 0000000000..6d427328f3
--- /dev/null
+++ b/xen/arch/arm/arm32/mmu/head.S
@@ -0,0 +1,573 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * xen/arch/arm/arm32/mmu/head.S
+ *
+ * Arm32 MMU specific start-of-day code.
+ */
+
+#include <asm/page.h>
+#include <asm/early_printk.h>
+
+#define PT_PT 0xf7f /* nG=1 AF=1 SH=11 AP=01 NS=1 ATTR=111 T=1 P=1 */
+#define PT_MEM 0xf7d /* nG=1 AF=1 SH=11 AP=01 NS=1 ATTR=111 T=0 P=1 */
+#define PT_MEM_L3 0xf7f /* nG=1 AF=1 SH=11 AP=01 NS=1 ATTR=111 T=1 P=1 */
+#define PT_DEV 0xe71 /* nG=1 AF=1 SH=10 AP=01 NS=1 ATTR=100 T=0 P=1 */
+#define PT_DEV_L3 0xe73 /* nG=1 AF=1 SH=10 AP=01 NS=1 ATTR=100 T=1 P=1 */
+
+#define PT_UPPER(x) (PT_##x & 0xf00)
+#define PT_LOWER(x) (PT_##x & 0x0ff)
+
+/* Convenience defines to get slot used by Xen mapping. */
+#define XEN_FIRST_SLOT first_table_offset(XEN_VIRT_START)
+#define XEN_SECOND_SLOT second_table_offset(XEN_VIRT_START)
+
+/* Offset between the early boot xen mapping and the runtime xen mapping */
+#define XEN_TEMPORARY_OFFSET (TEMPORARY_XEN_VIRT_START - XEN_VIRT_START)
+
+.macro load_paddr rb, sym
+ mov_w \rb, \sym
+ add \rb, \rb, r10
+.endm
+
+/*
+ * Flush local TLBs
+ *
+ * @tmp: Scratch register
+ *
+ * See asm/arm32/flushtlb.h for the explanation of the sequence.
+ */
+.macro flush_xen_tlb_local tmp
+ dsb nshst
+ mcr CP32(\tmp, TLBIALLH)
+ dsb nsh
+ isb
+.endm
+
+/*
+ * Enforce Xen page-tables do not contain mapping that are both
+ * Writable and eXecutables.
+ *
+ * This should be called on each secondary CPU.
+ */
+.macro pt_enforce_wxn tmp
+ mrc CP32(\tmp, HSCTLR)
+ orr \tmp, \tmp, #SCTLR_Axx_ELx_WXN
+ dsb
+ mcr CP32(\tmp, HSCTLR)
+ /*
+ * The TLBs may cache SCTLR_EL2.WXN. So ensure it is synchronized
+ * before flushing the TLBs.
+ */
+ isb
+ flush_xen_tlb_local \tmp
+.endm
+
+/* Macro to find the slot number at a given page-table level
+ *
+ * slot: slot computed
+ * virt: virtual address
+ * lvl: page-table level
+ *
+ * Note that ubxf is unpredictable when the end bit is above 32-bit. So we
+ * can't use it for first level offset.
+ */
+.macro get_table_slot, slot, virt, lvl
+ .if \lvl == 1
+ lsr \slot, \virt, #XEN_PT_LEVEL_SHIFT(\lvl)
+ .else
+ ubfx \slot, \virt, #XEN_PT_LEVEL_SHIFT(\lvl), #XEN_PT_LPAE_SHIFT
+ .endif
+.endm
+
+/*
+ * Macro to create a page table entry in \ptbl to \tbl (physical
+ * address)
+ *
+ * ptbl: table symbol where the entry will be created
+ * tbl: physical address of the table to point to
+ * virt: virtual address
+ * lvl: page-table level
+ *
+ * Preserves \virt
+ * Clobbers \tbl, r1 - r3
+ *
+ * Note that \tbl and \virt should be in a register other than r1 - r3
+ */
+.macro create_table_entry_from_paddr, ptbl, tbl, virt, lvl
+ get_table_slot r1, \virt, \lvl /* r1 := slot in \tbl */
+ lsl r1, r1, #3 /* r1 := slot offset in \tbl */
+
+ movw r2, #PT_PT /* r2:r3 := right for linear PT */
+ orr r2, r2, \tbl /* + \tbl paddr */
+ mov r3, #0
+
+ adr_l \tbl, \ptbl /* \tbl := (v,p)addr of \ptbl */
+
+ strd r2, r3, [\tbl, r1]
+.endm
+
+
+/*
+ * Macro to create a page table entry in \ptbl to \tbl (symbol)
+ *
+ * ptbl: table symbol where the entry will be created
+ * tbl: table symbol to point to
+ * virt: virtual address
+ * lvl: page-table level
+ *
+ * Preserves \virt
+ * Clobbers r1 - r4
+ *
+ * Also use r10 for the phys offset.
+ *
+ * Note that \virt should be in a register other than r1 - r4
+ */
+.macro create_table_entry, ptbl, tbl, virt, lvl
+ load_paddr r4, \tbl
+ create_table_entry_from_paddr \ptbl, r4, \virt, \lvl
+ .endm
+
+/*
+ * Macro to create a mapping entry in \tbl to \paddr. Only mapping in 3rd
+ * level table (i.e page granularity) is supported.
+ *
+ * ptbl: table symbol where the entry will be created
+ * virt: virtual address
+ * phys: physical address
+ * type: mapping type. If not specified it will be normal memory (PT_MEM_L3)
+ *
+ * Preserves \virt, \phys
+ * Clobbers r1 - r4
+ *
+ * Note that \virt and \paddr should be in other registers than r1 - r4
+ * and be distinct.
+ */
+.macro create_mapping_entry, ptbl, virt, phys, type=PT_MEM_L3
+ mov_w r2, XEN_PT_LPAE_ENTRY_MASK
+ lsr r1, \virt, #THIRD_SHIFT
+ and r1, r1, r2 /* r1 := slot in \tlb */
+ lsl r1, r1, #3 /* r1 := slot offset in \tlb */
+
+ lsr r4, \phys, #THIRD_SHIFT
+ lsl r4, r4, #THIRD_SHIFT /* r4 := PAGE_ALIGNED(phys) */
+
+ movw r2, #\type /* r2:r3 := right for section PT */
+ orr r2, r2, r4 /* + PAGE_ALIGNED(phys) */
+ mov r3, #0
+
+ adr_l r4, \ptbl
+
+ strd r2, r3, [r4, r1]
+.endm
+
+.section .text.idmap, "ax", %progbits
+
+/*
+ * Rebuild the boot pagetable's first-level entries. The structure
+ * is described in mm.c.
+ *
+ * After the CPU enables paging it will add the fixmap mapping
+ * to these page tables, however this may clash with the 1:1
+ * mapping. So each CPU must rebuild the page tables here with
+ * the 1:1 in place.
+ *
+ * Inputs:
+ * r9 : paddr(start)
+ * r10: phys offset
+ *
+ * Output:
+ * r12: Was a temporary mapping created?
+ *
+ * Clobbers r0 - r5
+ */
+create_page_tables:
+ /* Prepare the page-tables for mapping Xen */
+ mov_w r0, XEN_VIRT_START
+
+ /*
+ * We need to use a stash register because
+ * create_table_entry_paddr() will clobber the register storing
+ * the physical address of the table to point to.
+ */
+ adr_l r5, boot_third
+ mov_w r4, XEN_VIRT_START
+.rept XEN_NR_ENTRIES(2)
+ mov r0, r5 /* r0 := paddr(l3 table) */
+ create_table_entry_from_paddr boot_second, r0, r4, 2
+ add r4, r4, #XEN_PT_LEVEL_SIZE(2) /* r4 := Next vaddr */
+ add r5, r5, #PAGE_SIZE /* r5 := Next table */
+.endr
+
+ /*
+ * Find the size of Xen in pages and multiply by the size of a
+ * PTE. This will then be compared in the mapping loop below.
+ *
+ * Note the multiplication is just to avoid using an extra
+ * register/instruction per iteration.
+ */
+ mov_w r0, _start /* r0 := vaddr(_start) */
+ mov_w r1, _end /* r1 := vaddr(_end) */
+ sub r0, r1, r0 /* r0 := effective size of Xen */
+ lsr r0, r0, #PAGE_SHIFT /* r0 := Number of pages for Xen */
+ lsl r0, r0, #3 /* r0 := Number of pages * PTE size */
+
+ /* Setup boot_third: */
+ adr_l r4, boot_third
+
+ lsr r2, r9, #THIRD_SHIFT /* Base address for 4K mapping */
+ lsl r2, r2, #THIRD_SHIFT
+ orr r2, r2, #PT_UPPER(MEM_L3) /* r2:r3 := map */
+ orr r2, r2, #PT_LOWER(MEM_L3)
+ mov r3, #0x0
+
+ /* ... map of vaddr(start) in boot_third */
+ mov r1, #0
+1: strd r2, r3, [r4, r1] /* Map vaddr(start) */
+ add r2, r2, #PAGE_SIZE /* Next page */
+ add r1, r1, #8 /* Next slot */
+ cmp r1, r0 /* Loop until we map all of Xen */
+ blo 1b
+
+ /*
+ * Setup the 1:1 mapping so we can turn the MMU on. Note that
+ * only the first page of Xen will be part of the 1:1 mapping.
+ */
+ create_table_entry boot_pgtable, boot_second_id, r9, 1
+ create_table_entry boot_second_id, boot_third_id, r9, 2
+ create_mapping_entry boot_third_id, r9, r9
+
+ /*
+ * Find the first slot used. If the slot is not the same
+ * as TEMPORARY_AREA_FIRST_SLOT, then we will want to switch
+ * to the temporary mapping before jumping to the runtime
+ * virtual mapping.
+ */
+ get_table_slot r1, r9, 1 /* r1 := first slot */
+ cmp r1, #TEMPORARY_AREA_FIRST_SLOT
+ bne use_temporary_mapping
+
+ mov_w r0, XEN_VIRT_START
+ create_table_entry boot_pgtable, boot_second, r0, 1
+ mov r12, #0 /* r12 := temporary mapping not created */
+ mov pc, lr
+
+use_temporary_mapping:
+ /*
+ * The identity mapping is not using the first slot
+ * TEMPORARY_AREA_FIRST_SLOT. Create a temporary mapping.
+ * See switch_to_runtime_mapping for more details.
+ */
+ PRINT("- Create temporary mapping -\r\n")
+
+ /* Map boot_second (cover Xen mappings) to the temporary 1st slot */
+ mov_w r0, TEMPORARY_XEN_VIRT_START
+ create_table_entry boot_pgtable, boot_second, r0, 1
+
+ mov r12, #1 /* r12 := temporary mapping created */
+ mov pc, lr
+ENDPROC(create_page_tables)
+
+/*
+ * Turn on the Data Cache and the MMU. The function will return
+ * to the virtual address provided in LR (e.g. the runtime mapping).
+ *
+ * Inputs:
+ * r9 : paddr(start)
+ * r12 : Was the temporary mapping created?
+ * lr : Virtual address to return to
+ *
+ * Clobbers r0 - r5
+ */
+enable_mmu:
+ PRINT("- Turning on paging -\r\n")
+
+ /*
+ * The state of the TLBs is unknown before turning on the MMU.
+ * Flush them to avoid stale one.
+ */
+ flush_xen_tlb_local r0
+
+ /* Write Xen's PT's paddr into the HTTBR */
+ adr_l r0, boot_pgtable
+ mov r1, #0 /* r0:r1 is paddr (boot_pagetable) */
+ mcrr CP64(r0, r1, HTTBR)
+ isb
+
+ mrc CP32(r0, HSCTLR)
+ /* Enable MMU and D-cache */
+ orr r0, r0, #(SCTLR_Axx_ELx_M|SCTLR_Axx_ELx_C)
+ dsb /* Flush PTE writes and finish reads */
+ mcr CP32(r0, HSCTLR) /* now paging is enabled */
+ isb /* Now, flush the icache */
+
+ /*
+ * The MMU is turned on and we are in the 1:1 mapping. Switch
+ * to the runtime mapping.
+ */
+ mov r5, lr /* Save LR before overwritting it */
+ mov_w lr, 1f /* Virtual address in the runtime mapping */
+ b switch_to_runtime_mapping
+1:
+ mov lr, r5 /* Restore LR */
+ /*
+ * At this point, either the 1:1 map or the temporary mapping
+ * will be present. The former may clash with other parts of the
+ * Xen virtual memory layout. As both of them are not used
+ * anymore, remove them completely to avoid having to worry
+ * about replacing existing mapping afterwards.
+ *
+ * On return this will jump to the virtual address requested by
+ * the caller.
+ */
+ teq r12, #0
+ beq remove_identity_mapping
+ b remove_temporary_mapping
+ENDPROC(enable_mmu)
+
+/*
+ * Switch to the runtime mapping. The logic depends on whether the
+ * runtime virtual region is clashing with the physical address
+ *
+ * - If it is not clashing, we can directly jump to the address in
+ * the runtime mapping.
+ * - If it is clashing, create_page_tables() would have mapped Xen to
+ * a temporary virtual address. We need to switch to the temporary
+ * mapping so we can remove the identity mapping and map Xen at the
+ * correct position.
+ *
+ * Inputs
+ * r9: paddr(start)
+ * r12: Was a temporary mapping created?
+ * lr: Address in the runtime mapping to jump to
+ *
+ * Clobbers r0 - r4
+ */
+switch_to_runtime_mapping:
+ /*
+ * Jump to the runtime mapping if the virt and phys are not
+ * clashing
+ */
+ teq r12, #0
+ beq ready_to_switch
+
+ /* We are still in the 1:1 mapping. Jump to the temporary Virtual address. */
+ mov_w r0, 1f
+ add r0, r0, #XEN_TEMPORARY_OFFSET /* r0 := address in temporary mapping */
+ mov pc, r0
+
+1:
+ /* Remove boot_second_id */
+ mov r2, #0
+ mov r3, #0
+ adr_l r0, boot_pgtable
+ get_table_slot r1, r9, 1 /* r1 := first slot */
+ lsl r1, r1, #3 /* r1 := first slot offset */
+ strd r2, r3, [r0, r1]
+
+ flush_xen_tlb_local r0
+
+ /* Map boot_second into boot_pgtable */
+ mov_w r0, XEN_VIRT_START
+ create_table_entry boot_pgtable, boot_second, r0, 1
+
+ /* Ensure any page table updates are visible before continuing */
+ dsb nsh
+ /*
+ * The function will return on the runtime mapping. So we want
+ * to prevent instruction fetch before the dsb completes.
+ */
+ isb
+
+ready_to_switch:
+ mov pc, lr
+ENDPROC(switch_to_runtime_mapping)
+
+/*
+ * Enable mm (turn on the data cache and the MMU) for secondary CPUs.
+ * The function will return to the virtual address provided in LR (e.g. the
+ * runtime mapping).
+ *
+ * Inputs:
+ * r9 : paddr(start)
+ * r10: phys offset
+ * lr : Virtual address to return to.
+ *
+ * Output:
+ * r12: Was a temporary mapping created?
+ *
+ * Clobbers r0 - r6
+ */
+ENTRY(enable_secondary_cpu_mm)
+ mov r6, lr
+
+ bl create_page_tables
+
+ /* Address in the runtime mapping to jump to after the MMU is enabled */
+ mov_w lr, 1f
+ b enable_mmu
+1:
+ /*
+ * Non-boot CPUs need to move on to the proper pagetables, which were
+ * setup in prepare_secondary_mm.
+ *
+ * XXX: This is not compliant with the Arm Arm.
+ */
+ mov_w r4, init_ttbr /* VA of HTTBR value stashed by CPU 0 */
+ ldrd r4, r5, [r4] /* Actual value */
+ dsb
+ mcrr CP64(r4, r5, HTTBR)
+ dsb
+ isb
+ flush_xen_tlb_local r0
+ pt_enforce_wxn r0
+
+ /* Return to the virtual address requested by the caller. */
+ mov pc, r6
+ENDPROC(enable_secondary_cpu_mm)
+
+/*
+ * Enable mm (turn on the data cache and the MMU) for the boot CPU.
+ * The function will return to the virtual address provided in LR (e.g. the
+ * runtime mapping).
+ *
+ * Inputs:
+ * r9 : paddr(start)
+ * r10: phys offset
+ * lr : Virtual address to return to.
+ *
+ * Output:
+ * r12: Was a temporary mapping created?
+ *
+ * Clobbers r0 - r6
+ */
+ENTRY(enable_boot_cpu_mm)
+ mov r6, lr
+
+ bl create_page_tables
+
+ /* Address in the runtime mapping to jump to after the MMU is enabled */
+ mov_w lr, 1f
+ b enable_mmu
+1:
+ /* Return to the virtual address requested by the caller. */
+ mov lr, r6
+
+ b setup_fixmap
+ENDPROC(enable_boot_cpu_mm)
+
+/*
+ * Remove the 1:1 map from the page-tables. It is not easy to keep track
+ * where the 1:1 map was mapped, so we will look for the top-level entry
+ * exclusive to the 1:1 map and remove it.
+ *
+ * Inputs:
+ * r9 : paddr(start)
+ *
+ * Clobbers r0 - r3
+ */
+remove_identity_mapping:
+ /* r2:r3 := invalid page-table entry */
+ mov r2, #0x0
+ mov r3, #0x0
+
+ /* Find the first slot used and remove it */
+ get_table_slot r1, r9, 1 /* r1 := first slot */
+ mov_w r0, boot_pgtable /* r0 := root table */
+ lsl r1, r1, #3 /* r1 := Slot offset */
+ strd r2, r3, [r0, r1]
+
+ flush_xen_tlb_local r0
+ mov pc, lr
+ENDPROC(remove_identity_mapping)
+
+/*
+ * Remove the temporary mapping of Xen starting at TEMPORARY_XEN_VIRT_START.
+ *
+ * Clobbers r0 - r3
+ */
+remove_temporary_mapping:
+ /* r2:r3 := invalid page-table entry */
+ mov r2, #0
+ mov r3, #0
+
+ adr_l r0, boot_pgtable
+ mov_w r1, TEMPORARY_XEN_VIRT_START
+ get_table_slot r1, r1, 1 /* r1 := first slot */
+ lsl r1, r1, #3 /* r1 := first slot offset */
+ strd r2, r3, [r0, r1]
+
+ flush_xen_tlb_local r0
+
+ mov pc, lr
+ENDPROC(remove_temporary_mapping)
+
+/*
+ * Map the UART in the fixmap (when earlyprintk is used) and hook the
+ * fixmap table in the page tables.
+ *
+ * The fixmap cannot be mapped in create_page_tables because it may
+ * clash with the 1:1 mapping.
+ *
+ * Inputs:
+ * r10: Physical offset
+ * r11: Early UART base physical address
+ *
+ * Clobbers r0 - r4
+ */
+setup_fixmap:
+#if defined(CONFIG_EARLY_PRINTK)
+ /* Add UART to the fixmap table */
+ mov_w r0, EARLY_UART_VIRTUAL_ADDRESS
+ create_mapping_entry xen_fixmap, r0, r11, type=PT_DEV_L3
+#endif
+ /* Map fixmap into boot_second */
+ mov_w r0, FIXMAP_ADDR(0)
+ create_table_entry boot_second, xen_fixmap, r0, 2
+ /* Ensure any page table updates made above have occurred. */
+ dsb nshst
+ /*
+ * The fixmap area will be used soon after. So ensure no hardware
+ * translation happens before the dsb completes.
+ */
+ isb
+
+ mov pc, lr
+ENDPROC(setup_fixmap)
+
+/* Fail-stop */
+fail: PRINT("- Boot failed -\r\n")
+1: wfe
+ b 1b
+ENDPROC(fail)
+
+/*
+ * Switch TTBR
+ * r1:r0 ttbr
+ *
+ * TODO: This code does not comply with break-before-make.
+ */
+ENTRY(switch_ttbr)
+ dsb /* Ensure the flushes happen before
+ * continuing */
+ isb /* Ensure synchronization with previous
+ * changes to text */
+ mcr CP32(r0, TLBIALLH) /* Flush hypervisor TLB */
+ mcr CP32(r0, ICIALLU) /* Flush I-cache */
+ mcr CP32(r0, BPIALL) /* Flush branch predictor */
+ dsb /* Ensure completion of TLB+BP flush */
+ isb
+
+ mcrr CP64(r0, r1, HTTBR)
+
+ dsb /* ensure memory accesses do not cross
+ * over the TTBR0 write */
+ isb /* Ensure synchronization with previous
+ * changes to text */
+ mcr CP32(r0, TLBIALLH) /* Flush hypervisor TLB */
+ mcr CP32(r0, ICIALLU) /* Flush I-cache */
+ mcr CP32(r0, BPIALL) /* Flush branch predictor */
+ dsb /* Ensure completion of TLB+BP flush */
+ isb
+
+ mov pc, lr
+ENDPROC(switch_ttbr)
diff --git a/xen/arch/arm/include/asm/arm32/macros.h b/xen/arch/arm/include/asm/arm32/macros.h
index a4e20aa520..c6e390cc5f 100644
--- a/xen/arch/arm/include/asm/arm32/macros.h
+++ b/xen/arch/arm/include/asm/arm32/macros.h
@@ -1,8 +1,62 @@
#ifndef __ASM_ARM_ARM32_MACROS_H
#define __ASM_ARM_ARM32_MACROS_H
- .macro ret
+.macro ret
mov pc, lr
- .endm
+.endm
+/*
+ * Move an immediate constant into a 32-bit register using movw/movt
+ * instructions.
+ */
+.macro mov_w reg, word
+ movw \reg, #:lower16:\word
+ movt \reg, #:upper16:\word
+.endm
+
+/*
+ * Pseudo-op for PC relative adr <reg>, <symbol> where <symbol> is
+ * within the range +/- 4GB of the PC.
+ *
+ * @dst: destination register
+ * @sym: name of the symbol
+ */
+.macro adr_l, dst, sym
+ mov_w \dst, \sym - .Lpc\@
+ .set .Lpc\@, .+ 8 /* PC bias */
+ add \dst, \dst, pc
+.endm
+
+#ifdef CONFIG_EARLY_PRINTK
+/*
+ * Macro to print a string to the UART, if there is one.
+ *
+ * Clobbers r0 - r3
+ */
+#define PRINT(_s) \
+ mov r3, lr ;\
+ adr_l r0, 98f ;\
+ bl puts ;\
+ mov lr, r3 ;\
+ RODATA_STR(98, _s)
+
+/*
+ * Macro to print the value of register \rb
+ *
+ * Clobbers r0 - r4
+ */
+.macro print_reg rb
+ mov r0, \rb
+ mov r4, lr
+ bl asm_putn
+ mov lr, r4
+.endm
+
+#else /* CONFIG_EARLY_PRINTK */
+#define PRINT(s)
+
+.macro print_reg rb
+.endm
+
+#endif /* !CONFIG_EARLY_PRINTK */
#endif /* __ASM_ARM_ARM32_MACROS_H */
--
2.25.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH v3 1/2] xen/arm32: head: Introduce enable_{boot,secondary}_cpu_mm()
2023-11-07 11:02 ` [PATCH v3 1/2] xen/arm32: head: Introduce enable_{boot,secondary}_cpu_mm() Ayan Kumar Halder
@ 2023-11-07 11:23 ` Michal Orzel
0 siblings, 0 replies; 5+ messages in thread
From: Michal Orzel @ 2023-11-07 11:23 UTC (permalink / raw)
To: Ayan Kumar Halder, xen-devel
Cc: sstabellini, julien, bertrand.marquis, michal.orzel, henry.wang,
Volodymyr_Babchuk, Julien Grall
Hi Ayan,
On 07/11/2023 12:02, Ayan Kumar Halder wrote:
> All the MMU related functionality have been clubbed together in
> enable_boot_cpu_mm() for booting primary cpu and enable_secondary_cpu_mm() for
> booting secondary cpus.
> This is done in preparation for moving the code related to MMU in MMU specific
> file and in order to support non MMU cpus in future.
>
> This is based on d2f8df5b3ede ("xen/arm64: head.S: Introduce enable_{boot,secondary}_cpu_mm()").
>
> Signed-off-by: Ayan Kumar Halder <ayan.kumar.halder@amd.com>
> Reviewed-by: Michal Orzel <michal.orzel@amd.com>
> Acked-by: Julien Grall <jgrall@amazon.com>
> ---
>
> Changes from :-
>
> v1 - 1. Added a proper commit message.
> 2. Some style and other fixes suggested in v1.
>
> v2 - 1. Updated the comment on top of enable_boot_cpu_mm() and
> enable_secondary_cpu_mm() ie mentioned the input and output registers.
> 2. Updated the comment inside enable_boot_cpu_mm().
>
> xen/arch/arm/arm32/head.S | 102 ++++++++++++++++++++++++++++++--------
> 1 file changed, 80 insertions(+), 22 deletions(-)
>
> diff --git a/xen/arch/arm/arm32/head.S b/xen/arch/arm/arm32/head.S
> index 2d7e690bf5..2204ea6dce 100644
> --- a/xen/arch/arm/arm32/head.S
> +++ b/xen/arch/arm/arm32/head.S
> @@ -201,13 +201,11 @@ past_zImage:
>
> bl check_cpu_mode
> bl cpu_init
> - bl create_page_tables
>
> - /* Address in the runtime mapping to jump to after the MMU is enabled */
> mov_w lr, primary_switched
> - b enable_mmu
> + b enable_boot_cpu_mm
> +
> primary_switched:
> - bl setup_fixmap
> #ifdef CONFIG_EARLY_PRINTK
> /* Use a virtual address to access the UART. */
> mov_w r11, EARLY_UART_VIRTUAL_ADDRESS
> @@ -249,27 +247,11 @@ GLOBAL(init_secondary)
> #endif
> bl check_cpu_mode
> bl cpu_init
> - bl create_page_tables
>
> - /* Address in the runtime mapping to jump to after the MMU is enabled */
> mov_w lr, secondary_switched
> - b enable_mmu
> -secondary_switched:
> - /*
> - * Non-boot CPUs need to move on to the proper pagetables, which were
> - * setup in prepare_secondary_mm.
> - *
> - * XXX: This is not compliant with the Arm Arm.
> - */
> - mov_w r4, init_ttbr /* VA of HTTBR value stashed by CPU 0 */
> - ldrd r4, r5, [r4] /* Actual value */
> - dsb
> - mcrr CP64(r4, r5, HTTBR)
> - dsb
> - isb
> - flush_xen_tlb_local r0
> - pt_enforce_wxn r0
> + b enable_secondary_cpu_mm
>
> +secondary_switched:
> #ifdef CONFIG_EARLY_PRINTK
> /* Use a virtual address to access the UART. */
> mov_w r11, EARLY_UART_VIRTUAL_ADDRESS
> @@ -692,6 +674,82 @@ ready_to_switch:
> mov pc, lr
> ENDPROC(switch_to_runtime_mapping)
>
> +/*
> + * Enable mm (turn on the data cache and the MMU) for secondary CPUs.
> + * The function will return to the virtual address provided in LR (e.g. the
> + * runtime mapping).
> + *
> + * Inputs:
> + * r9 : paddr(start)
> + * r10: phys offset
> + * lr : Virtual address to return to.
> + *
> + * Output:
> + * r12: Was a temporary mapping created?
> + *
> + * Clobbers r0 - r6
> + */
> +enable_secondary_cpu_mm:
> + mov r6, lr
> +
> + bl create_page_tables
> +
> + /* Address in the runtime mapping to jump to after the MMU is enabled */
> + mov_w lr, 1f
> + b enable_mmu
> +1:
> + /*
> + * Non-boot CPUs need to move on to the proper pagetables, which were
> + * setup in prepare_secondary_mm.
> + *
> + * XXX: This is not compliant with the Arm Arm.
> + */
> + mov_w r4, init_ttbr /* VA of HTTBR value stashed by CPU 0 */
> + ldrd r4, r5, [r4] /* Actual value */
> + dsb
> + mcrr CP64(r4, r5, HTTBR)
> + dsb
> + isb
> + flush_xen_tlb_local r0
> + pt_enforce_wxn r0
> +
> + /* Return to the virtual address requested by the caller. */
> + mov pc, r6
> +ENDPROC(enable_secondary_cpu_mm)
> +
> +/*
> + * Enable mm (turn on the data cache and the MMU) for the boot CPU.
> + * The function will return to the virtual address provided in LR (e.g. the
> + * runtime mapping).
> + *
> + * Inputs:
> + * r9 : paddr(start)
> + * r10: phys offset
> + * lr : Virtual address to return to.
> + *
> + * Output:
> + * r12: Was a temporary mapping created?
> + *
> + * Clobbers r0 - r6
> + */
> +enable_boot_cpu_mm:
> + mov r6, lr
> +
> + bl create_page_tables
> +
> + /* Address in the runtime mapping to jump to after the MMU is enabled */
> + mov_w lr, 1f
> + b enable_mmu
> +1:
> + /*
> + * Prepare the fixmap. The function will return to the virtual address
> + * requested by the caller.
> + */
This comment should be above branch instruction and not here.
~Michal
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v3 2/2] xen/arm32: head Split and move MMU-specific head.S to mmu/head.S
2023-11-07 11:02 ` [PATCH v3 2/2] xen/arm32: head Split and move MMU-specific head.S to mmu/head.S Ayan Kumar Halder
@ 2023-11-07 12:26 ` Michal Orzel
0 siblings, 0 replies; 5+ messages in thread
From: Michal Orzel @ 2023-11-07 12:26 UTC (permalink / raw)
To: Ayan Kumar Halder, xen-devel
Cc: sstabellini, julien, bertrand.marquis, michal.orzel, henry.wang,
Volodymyr_Babchuk
Hi Ayan,
On 07/11/2023 12:02, Ayan Kumar Halder wrote:
>
>
> The MMU specific code in head.S will not be used on MPU systems.
> Instead of introducing more #ifdefs which will bring complexity
> to the code, move MMU related code to mmu/head.S and keep common
> code in head.S. Two notes while moving:
> - As "fail" in original head.S is very simple and this name is too
> easy to be conflicted, duplicate it in mmu/head.S instead of
> exporting it.
> - Realigned ".macro ret" so that the alignment matches to the other
> macros.
> - Rename puts to asm_puts, putn to asm_putn (this denotes that the
> macros are used within the context of assembly only).
> - Use ENTRY() for enable_secondary_cpu_mm, enable_boot_cpu_mm,
> setup_fixmap, asm_puts, asm_putn as they will be used externally.
>
> Also move the assembly macros shared by head.S and mmu/head.S to
> macros.h.
>
> This is based on 6734327d76be ("xen/arm64: Split and move MMU-specific head.S to mmu/head.S").
>
> Signed-off-by: Ayan Kumar Halder <ayan.kumar.halder@amd.com>
For now, I won't provide Rb given that the baseline is still under development and not very clear to me.
Just a few remarks:
[...]
> -
> #ifdef CONFIG_EARLY_PRINTK
> /*
> * Initialize the UART. Should only be called on the boot CPU.
> @@ -912,14 +298,14 @@ ENDPROC(init_uart)
> * r11: Early UART base address
You could follow the arm64 patch and add:
"Note: This function must be called from assembly."
to make the usage of this function clear.
Same for asm_putn.
> * Clobbers r0-r1
> */
> -puts:
> +ENTRY(asm_puts)
> early_uart_ready r11, r1
> ldrb r1, [r0], #1 /* Load next char */
> teq r1, #0 /* Exit on nul */
> moveq pc, lr
> early_uart_transmit r11, r1
> - b puts
> -ENDPROC(puts)
> + b asm_puts
> +ENDPROC(asm_puts)
>
> /*
> * Print a 32-bit number in hex. Specific to the PL011 UART.
The second sentence can be dropped. I don't see anything PL011 specific here.
> @@ -927,7 +313,7 @@ ENDPROC(puts)
> * r11: Early UART base address
> * Clobbers r0-r3
> */
> -putn:
> +ENTRY(asm_putn)
> adr_l r1, hex
> mov r3, #8
> 1:
> @@ -939,7 +325,7 @@ putn:
> subs r3, r3, #1
> bne 1b
> mov pc, lr
> -ENDPROC(putn)
> +ENDPROC(asm_putn)
>
> RODATA_STR(hex, "0123456789abcdef")
>
> @@ -947,8 +333,8 @@ RODATA_STR(hex, "0123456789abcdef")
>
> ENTRY(early_puts)
> init_uart:
> -puts:
> -putn: mov pc, lr
> +asm_puts:
> +asm_putn: mov pc, lr
Both asm_puts and asm_putn are used only within #ifdef so you can drop the stubs
[...]
> diff --git a/xen/arch/arm/include/asm/arm32/macros.h b/xen/arch/arm/include/asm/arm32/macros.h
> index a4e20aa520..c6e390cc5f 100644
> --- a/xen/arch/arm/include/asm/arm32/macros.h
> +++ b/xen/arch/arm/include/asm/arm32/macros.h
> @@ -1,8 +1,62 @@
> #ifndef __ASM_ARM_ARM32_MACROS_H
> #define __ASM_ARM_ARM32_MACROS_H
>
> - .macro ret
> +.macro ret
> mov pc, lr
> - .endm
> +.endm
>
> +/*
> + * Move an immediate constant into a 32-bit register using movw/movt
> + * instructions.
> + */
> +.macro mov_w reg, word
> + movw \reg, #:lower16:\word
> + movt \reg, #:upper16:\word
> +.endm
> +
> +/*
> + * Pseudo-op for PC relative adr <reg>, <symbol> where <symbol> is
> + * within the range +/- 4GB of the PC.
> + *
> + * @dst: destination register
> + * @sym: name of the symbol
> + */
> +.macro adr_l, dst, sym
> + mov_w \dst, \sym - .Lpc\@
> + .set .Lpc\@, .+ 8 /* PC bias */
> + add \dst, \dst, pc
> +.endm
> +
> +#ifdef CONFIG_EARLY_PRINTK
> +/*
> + * Macro to print a string to the UART, if there is one.
> + *
> + * Clobbers r0 - r3
> + */
> +#define PRINT(_s) \
> + mov r3, lr ;\
> + adr_l r0, 98f ;\
> + bl puts ;\
This should be a call to asm_puts
~Michal
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2023-11-07 12:27 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-11-07 11:02 [PATCH v3 0/2] xen/arm: Split MMU code as the prepration of MPU work form Arm32 Ayan Kumar Halder
2023-11-07 11:02 ` [PATCH v3 1/2] xen/arm32: head: Introduce enable_{boot,secondary}_cpu_mm() Ayan Kumar Halder
2023-11-07 11:23 ` Michal Orzel
2023-11-07 11:02 ` [PATCH v3 2/2] xen/arm32: head Split and move MMU-specific head.S to mmu/head.S Ayan Kumar Halder
2023-11-07 12:26 ` Michal Orzel
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.