[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMzpN2iU07o8rBzZMpDv1gB00HUc2CRR1o3m=EuiQ6E16QcPFg@mail.gmail.com>
Date: Sat, 14 Oct 2017 13:19:08 -0400
From: Brian Gerst <brgerst@...il.com>
To: "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>
Cc: Ingo Molnar <mingo@...hat.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
"the arch/x86 maintainers" <x86@...nel.org>,
Thomas Gleixner <tglx@...utronix.de>,
"H. Peter Anvin" <hpa@...or.com>,
Andy Lutomirski <luto@...capital.net>,
Cyrill Gorcunov <gorcunov@...nvz.org>,
Borislav Petkov <bp@...e.de>, Andi Kleen <ak@...ux.intel.com>,
Linux-MM <linux-mm@...ck.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCHv2, RFC] x86/boot/compressed/64: Handle 5-level paging boot
if kernel is above 4G
On Fri, Oct 13, 2017 at 8:23 AM, Kirill A. Shutemov
<kirill.shutemov@...ux.intel.com> wrote:
> This patch addresses shortcoming in current boot process on machines
> that supports 5-level paging.
>
> If bootloader enables 64-bit mode with 4-level paging, we need to
> switch over to 5-level paging. The switching requires disabling paging.
> It works fine if kernel itself is loaded below 4G.
>
> If bootloader put the kernel above 4G (not sure if anybody does this),
> we would loose control as soon as paging is disabled as code becomes
> unreachable.
>
> This patch implements trampoline in lower memory to handle this
> situation.
>
> Apart from trampoline itself we also need place to store top level page
> table in lower memory as we don't have a way to load 64-bit value into
> CR3 from 32-bit mode. We only really need 8-bytes there as we only use
> the very first entry of the page table.
>
> place_trampoline() would choose an address for the trampoline page.
> The implementation is based on reserve_bios_regions(). We take a page
> next to end of lowmem.
>
> We only need the page for very short time, until main kernel image
> setup its own page tables.
>
> Signed-off-by: Kirill A. Shutemov <kirill.shutemov@...ux.intel.com>
> ---
> arch/x86/boot/compressed/head_64.S | 87 ++++++++++++++++++++++++++------------
> arch/x86/boot/compressed/misc.c | 25 +++++++++++
> 2 files changed, 84 insertions(+), 28 deletions(-)
>
> diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S
> index cefe4958fda9..961c72755986 100644
> --- a/arch/x86/boot/compressed/head_64.S
> +++ b/arch/x86/boot/compressed/head_64.S
> @@ -288,8 +288,23 @@ ENTRY(startup_64)
> leaq boot_stack_end(%rbx), %rsp
>
> #ifdef CONFIG_X86_5LEVEL
> +/*
> + * We need trampoline in lower memory switch from 4- to 5-level paging for
> + * cases when bootloader put kernel above 4G, but didn't enable 5-level paging
> + * for us.
> + *
> + * We also have to have top page table in lower memory as we don't have a way
> + * to load 64-bit value into CR3 from 32-bit mode. We only need 8-bytes there
> + * as we only use the very first entry of the page table.
> + *
> + * The same page can be used to place both trampoline code and top level page
> + * table. place_trampoline() will find suitable place for the trampoline page.
> + * Code will be placed with offset 0x100 from beginning of the page.
> + */
> +#define LVL5_TRAMPOLINE_CODE 0x100
> +
> /* Preserve RBX across CPUID */
> - movq %rbx, %r8
> + movq %rbx, %r15
>
> /* Check if leaf 7 is supported */
> xorl %eax, %eax
> @@ -307,9 +322,6 @@ ENTRY(startup_64)
> andl $(1 << 16), %ecx
> jz lvl5
>
> - /* Restore RBX */
> - movq %r8, %rbx
> -
> /* Check if 5-level paging has already been enabled */
> movq %cr4, %rax
> testl $X86_CR4_LA57, %eax
> @@ -323,34 +335,53 @@ ENTRY(startup_64)
> * long mode would trigger #GP. So we need to switch off long mode
> * first.
> *
> - * NOTE: This is not going to work if bootloader put us above 4G
> - * limit.
> + * We use trampoline in lower memory to handle situation when
> + * bootloader put the kernel image above 4G.
> *
> * The first step is go into compatibility mode.
> */
>
> - /* Clear additional page table */
> - leaq lvl5_pgtable(%rbx), %rdi
> - xorq %rax, %rax
> - movq $(PAGE_SIZE/8), %rcx
> - rep stosq
> + /*
> + * Find sitable place for trampoline.
> + * The address will be stored in RBX.
> + */
> + call place_trampoline
> + movq %rax, %rbx
> +
> + /* Preserve RSI, to be used by movsb below */
> + movq %rsi, %r14
> +
> + /* Copy trampoline code in place */
> + leaq lvl5_trampoline_src(%rip), %rsi
> + leaq LVL5_TRAMPOLINE_CODE(%rbx), %rdi
> + movq $(lvl5_trampoline_end - lvl5_trampoline_src), %rcx
> + rep movsb
> +
> + /* Restore RSI */
> + movq %r14, %rsi
>
> /*
> - * Setup current CR3 as the first and only entry in a new top level
> + * Setup current CR3 as the first and the only entry in a new top level
> * page table.
> */
> movq %cr3, %rdi
> leaq 0x7 (%rdi), %rax
> - movq %rax, lvl5_pgtable(%rbx)
> + movq %rax, (%rbx)
> +
> + /*
> + * Load address of lvl5 into RDI.
> + * It will be used to return address from trampoline.
> + */
> + leaq lvl5(%rip), %rdi
>
> /* Switch to compatibility mode (CS.L = 0 CS.D = 1) via far return */
> pushq $__KERNEL32_CS
> - leaq compatible_mode(%rip), %rax
> + leaq LVL5_TRAMPOLINE_CODE(%rbx), %rax
> pushq %rax
> lretq
> lvl5:
> /* Restore RBX */
> - movq %r8, %rbx
> + movq %r15, %rbx
> #endif
>
> /* Zero EFLAGS */
> @@ -488,9 +519,9 @@ relocated:
> */
> jmp *%rax
>
> - .code32
> #ifdef CONFIG_X86_5LEVEL
> -compatible_mode:
> + .code32
> +lvl5_trampoline_src:
> /* Setup data and stack segments */
> movl $__KERNEL_DS, %eax
> movl %eax, %ds
> @@ -502,7 +533,7 @@ compatible_mode:
> movl %eax, %cr0
>
> /* Point CR3 to 5-level paging */
> - leal lvl5_pgtable(%ebx), %eax
> + leal (%ebx), %eax
> movl %eax, %cr3
>
> /* Enable PAE and LA57 mode */
> @@ -510,23 +541,27 @@ compatible_mode:
> orl $(X86_CR4_PAE | X86_CR4_LA57), %eax
> movl %eax, %cr4
>
> - /* Calculate address we are running at */
> - call 1f
> -1: popl %edi
> - subl $1b, %edi
> + /* Calculate address of lvl5_enabled once we are in trampoline */
> + leal lvl5_enabled - lvl5_trampoline_src + LVL5_TRAMPOLINE_CODE (%ebx), %eax
>
> /* Prepare stack for far return to Long Mode */
> pushl $__KERNEL_CS
> - leal lvl5(%edi), %eax
> - push %eax
> + pushl %eax
>
> /* Enable paging back */
> movl $(X86_CR0_PG | X86_CR0_PE), %eax
> movl %eax, %cr0
>
> lret
> +
> + .code64
> +lvl5_enabled:
> + /* Return from trampoline */
> + jmp *%rdi
> +lvl5_trampoline_end:
> #endif
>
> + .code32
> no_longmode:
> /* This isn't an x86-64 CPU so hang */
> 1:
> @@ -584,7 +619,3 @@ boot_stack_end:
> .balign 4096
> pgtable:
> .fill BOOT_PGT_SIZE, 1, 0
> -#ifdef CONFIG_X86_5LEVEL
> -lvl5_pgtable:
> - .fill PAGE_SIZE, 1, 0
> -#endif
> diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
> index c14217cd0155..809c91837521 100644
> --- a/arch/x86/boot/compressed/misc.c
> +++ b/arch/x86/boot/compressed/misc.c
> @@ -415,3 +415,28 @@ void fortify_panic(const char *name)
> {
> error("detected buffer overflow");
> }
> +
> +#ifdef CONFIG_X86_5LEVEL
> +
> +#define BIOS_START_MIN 0x20000U /* 128K, less than this is insane */
> +#define BIOS_START_MAX 0x9f000U /* 640K, absolute maximum */
> +
> +asmlinkage __visible unsigned int place_trampoline()
> +{
> + unsigned int bios_start, ebda_start;
> +
> + /* Based on reserve_bios_regions() */
> +
> + ebda_start = *(unsigned short *)0x40e << 4;
> + bios_start = *(unsigned short *)0x413 << 10;
> +
> + if (bios_start < BIOS_START_MIN || bios_start > BIOS_START_MAX)
> + bios_start = BIOS_START_MAX;
> +
> + if (ebda_start > BIOS_START_MIN && ebda_start < bios_start)
> + bios_start = ebda_start;
> +
> + /* Place trampoline one page below end of low memory, alinged to 4k */
> + return round_down(bios_start - PAGE_SIZE, PAGE_SIZE);
> +}
> +#endif
> --
> 2.14.2
>
>From what we've seen with the TLB flush rework, having potential
garbage in the page tables that speculative reads can see can cause
bad things like machine checks. It would be best to have a second
temporary page just for the page table (and properly cleared).
The trampoline also needs its own stack, in case the stack pointer was
above 4G. That could be at the end of the code page, since you only
need 8 bytes.
--
Brian Gerst
Powered by blists - more mailing lists