arm64: efi: boot with MMU and caches on if possible
Ard Biesheuvel
maintainercover arm64: efi: boot with MMU and caches on if possible
This small series is what remains now that much of the prerequisite changes that are shared with other work have landed. This is a follow-up to [0], and implements the necessary changes that allow the EFI stub to enter the kernel proper with MMU and caches enabled.
This is possible because the EFI spec mandates that all of memory is mapped, and so we can rely on this mapping to set up the initial ID map, instead of taking down the MMU first, and populating those page tables using non-cacheable accesses that require special care in terms of cache coherency. This also means that there is no need to clean the executable image to the point of coherency (with the exception of the contents of .idmap.text, which contains the code that switches from the firmware's ID map to the kernel one)
Given that the image will typically be mapped by the firmware according to the section descriptors in the PE/COFF header (R-X for .text and .rodata, RW- for .data and .bss), this also has a slight robustness advantage.
Note that this does not update the documented boot protocol [yet], although any loader could take advantage of this, and load the kernel image at any 64k aligned physical offset, and enter the image with the MMU still enabled, either at EL2 or EL1.
[0] https://lore.kernel.org/linux-arm-kernel/[email protected]/
Ard Biesheuvel (6):
arm64: lds: reduce effective minimum image alignment to 64k
arm64: kernel: move ID map out of .text mapping
arm64: head: record the MMU state at primary entry
arm64: head: avoid cache invalidation when entering with the MMU on
arm64: head: clean the ID map page to the PoC
arm64: efi/libstub: enter with the MMU on if executing in place
arch/arm64/include/asm/efi.h | 7 --- arch/arm64/kernel/efi-entry.S | 4 ++ arch/arm64/kernel/head.S | 45 ++++++++++++++++++-- arch/arm64/kernel/vmlinux.lds.S | 13 +++++- arch/arm64/mm/proc.S | 2 - drivers/firmware/efi/libstub/arm64-stub.c | 2 +- include/linux/efi.h | 6 +-- 7 files changed, 58 insertions(+), 21 deletions(-)
Signature · 7 lines
2.35.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Ard Biesheuvel
maintainerPATCH 1/6 arm64: lds: reduce effective minimum image alignment to 64k
Our segment alignment is 64k for all configurations, and coincidentally, this is the largest alignment supported by the PE/COFF executable format used by EFI. This means that generally, there is no need to move the image around in memory after it has been loaded by the firmware, which can be advantageous as it also permits us to rely on the memory attributes set by the firmware (R-X for [_text, __inittext_end] and RW- for [__initdata_begin, _end]. This means we can jump right from the EFI stub into the image with the MMU and caches enabled.
However, the minimum alignment of the image is actually 128k on 64k pages configurations with CONFIG_VMAP_STACK=y, due to the existence of a single 128k aligned object in the image, which is the stack of the init task.
Let's work around this by adding some padding before the init stack allocation, so we can round down the stack pointer to a suitably aligned value if the image is not aligned to 128k in memory.
Note that this does not affect the boot protocol, which still requires 2 MiB alignment for bare metal boot.
--- arch/arm64/include/asm/efi.h | 7 ------- arch/arm64/kernel/head.S | 3 +++ arch/arm64/kernel/vmlinux.lds.S | 11 ++++++++++- drivers/firmware/efi/libstub/arm64-stub.c | 2 +- include/linux/efi.h | 6 +----- 5 files changed, 15 insertions(+), 14 deletions(-)
arch/arm64/include/asm/efi.h +0 -7
@@ -57,13 +57,6 @@ efi_status_t __efi_rt_asm_wrapper(void *, const char *, ...);/* arch specific definitions used by the stub code */ -/* -* In some configurations (e.g. VMAP_STACK && 64K pages), stacks built into the -* kernel need greater alignment than we require the segments to be padded to. -*/ -#define EFI_KIMG_ALIGN \ - (SEGMENT_ALIGN > THREAD_ALIGN ? SEGMENT_ALIGN : THREAD_ALIGN) -/* * On arm64, we have to ensure that the initrd ends up in the linear region, * which is a 1 GB aligned region of size '1UL << (VA_BITS_MIN - 1)' that is
arch/arm64/kernel/head.S +3 -0
@@ -394,6 +394,9 @@ SYM_FUNC_END(create_kernel_mapping) msr sp_el0, \tsk ldr \tmp1, [\tsk, #TSK_STACK] +#if THREAD_ALIGN > SEGMENT_ALIGN + bic \tmp1, \tmp1, #THREAD_ALIGN - 1 +#endif add sp, \tmp1, #THREAD_SIZE sub sp, sp, #PT_REGS_SIZE
arch/arm64/kernel/vmlinux.lds.S +10 -1
@@ -274,7 +274,16 @@ SECTIONS _data = .; _sdata = .; - RW_DATA(L1_CACHE_BYTES, PAGE_SIZE, THREAD_ALIGN) +#if THREAD_ALIGN > SEGMENT_ALIGN + /* + * Add some padding for the init stack so we can fix up any potential + * misalignment at runtime. In practice, this can only occur on 64k + * pages configurations with CONFIG_VMAP_STACK=y. + */ + . += THREAD_ALIGN - SEGMENT_ALIGN; + ASSERT(. == init_stack, "init_stack not at start of RW_DATA as expected") +#endif + RW_DATA(L1_CACHE_BYTES, PAGE_SIZE, SEGMENT_ALIGN) /* * Data written with the MMU off but read with the MMU on requires
drivers/firmware/efi/libstub/arm64-stub.c +1 -1
@@ -98,7 +98,7 @@ efi_status_t handle_kernel_image(unsigned long *image_addr,* 2M alignment if KASLR was explicitly disabled, even if it was not * going to be activated to begin with. */ -u64 min_kimg_align = efi_nokaslr ?MIN_KIMG_ALIGN :EFI_KIMG_ALIGN ; +u64 min_kimg_align = efi_nokaslr ?MIN_KIMG_ALIGN :SEGMENT_ALIGN ;if (IS_ENABLED (CONFIG_RANDOMIZE_BASE )) {efi_guid_t li_fixed_proto =LINUX_EFI_LOADED_IMAGE_FIXED_GUID ;
include/linux/efi.h +1 -5
@@ -416,11 +416,7 @@ void efi_native_runtime_setup(void);/* * This GUID may be installed onto the kernel image's handle as a NULL protocol * to signal to the stub that the placement of the image should be respected, -* and moving the image in physical memory is undesirable. To ensure -* compatibility with 64k pages kernels with virtually mapped stacks, and to -* avoid defeating physical randomization, this protocol should only be -* installed if the image was placed at a randomized 128k aligned address in -* memory. +* and moving the image in physical memory is undesirable. */ #define LINUX_EFI_LOADED_IMAGE_FIXED_GUID EFI_GUID(0xf5a37b6d, 0x3344, 0x42a5, 0xb6, 0xbb, 0x97, 0x86, 0x48, 0xc1, 0x89, 0x0a)
Signature · 7 lines
2.35.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Ard Biesheuvel
maintainerPATCH 2/6 arm64: kernel: move ID map out of .text mapping
In reply to Ard BiesheuvelReorganize the ID map slightly so that only code that is executed via the 1:1 mapping remains. This allows us to move the ID map out of the .text segment, as it will no longer need executable permissions via the kernel mapping.
--- arch/arm64/kernel/head.S | 5 ++++- arch/arm64/kernel/vmlinux.lds.S | 2 +- arch/arm64/mm/proc.S | 2 -- 3 files changed, 5 insertions(+), 4 deletions(-)
arch/arm64/kernel/head.S +4 -1
@@ -464,7 +464,7 @@ SYM_FUNC_END(__primary_switched) * end early head section, begin head code that is also used for * hotplug and needs to have the same protections as the text region */ - .section ".idmap.text","awx" + .text /* * Starting from EL2 or EL1, configure the CPU to execute at the highest
@@ -556,6 +556,7 @@ SYM_FUNC_START_LOCAL(set_cpu_boot_mode_flag) ret SYM_FUNC_END(set_cpu_boot_mode_flag) + .section ".idmap.text","awx" /* * This provides a "holding pen" for platforms to hold all secondary * cores are held until we're ready for them to initialise.
@@ -600,6 +601,7 @@ SYM_FUNC_START_LOCAL(secondary_startup) br x8 SYM_FUNC_END(secondary_startup) + .text SYM_FUNC_START_LOCAL(__secondary_switched) mov x0, x20 bl set_cpu_boot_mode_flag
@@ -659,6 +661,7 @@ SYM_FUNC_END(__secondary_too_slow) * Checks if the selected granule size is supported by the CPU. * If it isn't, park the CPU */ + .section ".idmap.text","awx" SYM_FUNC_START(__enable_mmu) mrs x3, ID_AA64MMFR0_EL1 ubfx x3, x3, #ID_AA64MMFR0_TGRAN_SHIFT, 4
arch/arm64/kernel/vmlinux.lds.S +1 -1
@@ -168,7 +168,6 @@ SECTIONS LOCK_TEXT KPROBES_TEXT HYPERVISOR_TEXT - IDMAP_TEXT *(.gnu.warning) . = ALIGN(16); *(.got) /* Global offset table */
@@ -195,6 +194,7 @@ SECTIONS TRAMP_TEXT HIBERNATE_TEXT KEXEC_TEXT + IDMAP_TEXT . = ALIGN(PAGE_SIZE); }
arch/arm64/mm/proc.S +0 -2
@@ -107,7 +107,6 @@ SYM_FUNC_END(cpu_do_suspend) * * x0: Address of context pointer */ - .pushsection ".idmap.text", "awx" SYM_FUNC_START(cpu_do_resume) ldp x2, x3, [x0] ldp x4, x5, [x0, #16]
@@ -163,7 +162,6 @@ alternative_else_nop_endif isb ret SYM_FUNC_END(cpu_do_resume) - .popsection #endif .pushsection ".idmap.text", "awx"
Signature · 7 lines
2.35.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Ard Biesheuvel
maintainerPATCH 3/6 arm64: head: record the MMU state at primary entry
In reply to Ard BiesheuvelPrepare for being able to deal with primary entry with the MMU and caches enabled, by recording whether or not we entered with the MMU on in register x19.
While at it, add disable_mmu_workaround macro invocations to init_kernel_el, as its manipulation of SCTLR_ELx may amount to disabling of the MMU after subsequent patches.
--- arch/arm64/kernel/head.S | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)
arch/arm64/kernel/head.S +16 -0
@@ -77,6 +77,7 @@ * primary lowlevel boot path: * * Register Scope Purpose + * x19 primary_entry() .. start_kernel() whether we entered with the MMU on * x20 primary_entry() .. __primary_switch() CPU boot mode * x21 primary_entry() .. start_kernel() FDT pointer passed at boot in x0 * x22 create_idmap() .. start_kernel() ID map VA of the DT blob
@@ -85,6 +86,7 @@ * x28 create_idmap() callee preserved temp register */ SYM_CODE_START(primary_entry) + bl record_mmu_state bl preserve_boot_args bl init_kernel_el // w0=cpu_boot_mode mov x20, x0
@@ -107,6 +109,17 @@ SYM_CODE_START(primary_entry) b __primary_switch SYM_CODE_END(primary_entry) +SYM_CODE_START_LOCAL(record_mmu_state) + mrs x19, CurrentEL + cmp x19, #CurrentEL_EL2 + mrs x19, sctlr_el1 + b.ne 0f + mrs x19, sctlr_el2 +0: tst x19, #SCTLR_ELx_M + cset w19, ne + ret +SYM_CODE_END(record_mmu_state) + /* * Preserve the arguments passed by the bootloader in x0 .. x3 */
@@ -484,6 +497,7 @@ SYM_FUNC_START(init_kernel_el) SYM_INNER_LABEL(init_el1, SYM_L_LOCAL) mov_q x0, INIT_SCTLR_EL1_MMU_OFF + pre_disable_mmu_workaround msr sctlr_el1, x0 isb mov_q x0, INIT_PSTATE_EL1
@@ -515,6 +529,7 @@ SYM_INNER_LABEL(init_el2, SYM_L_LOCAL) /* Switching to VHE requires a sane SCTLR_EL1 as a start */ mov_q x0, INIT_SCTLR_EL1_MMU_OFF + pre_disable_mmu_workaround msr_s SYS_SCTLR_EL12, x0 /*
@@ -530,6 +545,7 @@ SYM_INNER_LABEL(init_el2, SYM_L_LOCAL) 1: mov_q x0, INIT_SCTLR_EL1_MMU_OFF + pre_disable_mmu_workaround msr sctlr_el1, x0 msr elr_el2, lr
Signature · 7 lines
2.35.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Ard Biesheuvel
maintainerPATCH 4/6 arm64: head: avoid cache invalidation when entering with the MMU on
In reply to Ard BiesheuvelIf we enter with the MMU on, there is no need for explicit cache invalidation for stores to memory, as they will be coherent with the caches.
Let's take advantage of this, and create the ID map with the MMU still enabled if that is how we entered, and avoid any cache invalidation calls in that case.
--- arch/arm64/kernel/head.S | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
arch/arm64/kernel/head.S +5 -2
@@ -88,9 +88,9 @@ SYM_CODE_START(primary_entry) bl record_mmu_state bl preserve_boot_args + bl create_idmap bl init_kernel_el // w0=cpu_boot_mode mov x20, x0 - bl create_idmap /* * The following calls CPU setup code, see arch/arm64/mm/proc.S for
@@ -130,11 +130,13 @@ SYM_CODE_START_LOCAL(preserve_boot_args) stp x21, x1, [x0] // x0 .. x3 at kernel entry stp x2, x3, [x0, #16] + cbnz x19, 0f // skip cache invalidation if MMU is on dmb sy // needed before dc ivac with // MMU off add x1, x0, #0x20 // 4 x 8 bytes b dcache_inval_poc // tail call +0: ret SYM_CODE_END(preserve_boot_args) SYM_FUNC_START_LOCAL(clear_page_tables)
@@ -371,12 +373,13 @@ SYM_FUNC_START_LOCAL(create_idmap) * accesses (MMU disabled), invalidate those tables again to * remove any speculatively loaded cache lines. */ + cbnz x19, 0f // skip cache invalidation if MMU is on dmb sy adrp x0, init_idmap_pg_dir adrp x1, init_idmap_pg_end bl dcache_inval_poc - ret x28 +0: ret x28 SYM_FUNC_END(create_idmap) SYM_FUNC_START_LOCAL(create_kernel_mapping)
Signature · 7 lines
2.35.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Ard Biesheuvel
maintainerPATCH 5/6 arm64: head: clean the ID map page to the PoC
In reply to Ard BiesheuvelIf we enter with the MMU and caches enabled, the caller may not have performed any cache maintenance. So clean the ID mapped page to the PoC, to ensure that instruction and data accesses with the MMU off see the correct data.
Note that this means primary_entry() itself needs to be moved into the ID map as well, as we will return from init_kernel_el() with the MMU and caches off.
--- arch/arm64/kernel/head.S | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-)
arch/arm64/kernel/head.S +13 -1
@@ -70,7 +70,7 @@ __EFI_PE_HEADER - __INIT + .section ".idmap.text","awx" /* * The following callee saved general purpose registers are used on the
@@ -89,6 +89,17 @@ SYM_CODE_START(primary_entry) bl record_mmu_state bl preserve_boot_args bl create_idmap + + /* + * If we entered with the MMU and caches on, clean the ID mapped part + * of the primary boot code to the PoC so we can safely execute it with + * the MMU off. + */ + cbz x19, 0f + adrp x0, __idmap_text_start + adr_l x1, __idmap_text_end + bl dcache_clean_poc +0: bl init_kernel_el // w0=cpu_boot_mode mov x20, x0
@@ -109,6 +120,7 @@ SYM_CODE_START(primary_entry) b __primary_switch SYM_CODE_END(primary_entry) + __INIT SYM_CODE_START_LOCAL(record_mmu_state) mrs x19, CurrentEL cmp x19, #CurrentEL_EL2
Signature · 7 lines
2.35.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Ard Biesheuvel
maintainerPATCH 6/6 arm64: efi/libstub: enter with the MMU on if executing in place
In reply to Ard BiesheuvelIf the kernel image has not been moved from the place where it was loaded by the firmware, just call the kernel entrypoint directly, and keep the MMU and caches enabled. This removes the need for any cache invalidation in the entry path.
--- arch/arm64/kernel/efi-entry.S | 4 ++++ 1 file changed, 4 insertions(+)
arch/arm64/kernel/efi-entry.S +4 -0
@@ -23,6 +23,10 @@ SYM_CODE_START(efi_enter_kernel) add x19, x0, x2 // relocated Image entrypoint mov x20, x1 // DTB address + adrp x3, _text // just call the entrypoint + cmp x0, x3 // directly if the image was + b.eq 2f // not moved around in memory + /* * Clean the copied Image to the PoC, and ensure it is not shadowed by * stale icache entries from before relocation.
Signature · 7 lines
2.35.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel