Re: [PATCH] [2/11] x86: Fix alternatives and kprobes to remap write-protected kernel text

[Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>]

Hi Andi,

I would recommend to prefix with __init every function that is assumed
"safe" because executed at boot time. It would minimize the risks of
buggy usage of these text-modification primitives. That includes
text_poke() and a big chunk of the apply alternative code. (it should
never be executed on a live system, except at boot time, right ?)

Since the constraints are completely different between boot time code
patching and live code patching, there might be some need for two
text_poke versions.

Mathieu

* Andi Kleen (ak@suse.de) wrote:
> 
> Reenable kprobes and alternative patching when the kernel text is write
> protected by DEBUG_RODATA
> Add a general utility function to change write protected text.
> The new function remaps the code using vmap to write it and takes
> care of CPU synchronization. It also does CLFLUSH to make icache recovery
> faster. 
> 
> There are some limitations on when the function can be used, see
> the comment.
> 
> This is a newer version that also changes the paravirt_ops code.
> text_poke also supports multi byte patching now.
> 
> Contains bug fixes from Zach Amsden and suggestions from
> Mathieu Desnoyers.
> 
> Cc: jbeulich@novell.com
> Cc: jeremy@goop.org
> Cc: Mathieu Desnoyers <compudj@krystal.dyndns.org>
> Cc: zach@vmware.com
> 
> Signed-off-by: Andi Kleen <ak@suse.de>
> 
> ---
>  arch/i386/kernel/alternative.c   |   33 +++++++++++++++++++++++++++++++--
>  arch/i386/kernel/kprobes.c       |    9 +++------
>  arch/i386/mm/init.c              |   14 +++-----------
>  arch/x86_64/kernel/kprobes.c     |   10 +++-------
>  arch/x86_64/mm/init.c            |   10 ----------
>  arch/x86_64/mm/pageattr.c        |    2 +-
>  include/asm-i386/alternative.h   |    2 ++
>  include/asm-x86_64/alternative.h |    2 ++
>  include/asm-x86_64/pgtable.h     |    2 ++
>  9 files changed, 47 insertions(+), 37 deletions(-)
> 
> Index: linux/arch/x86_64/kernel/kprobes.c
> ===================================================================
> --- linux.orig/arch/x86_64/kernel/kprobes.c
> +++ linux/arch/x86_64/kernel/kprobes.c
> @@ -39,9 +39,9 @@
>  #include <linux/module.h>
>  #include <linux/kdebug.h>
>  
> -#include <asm/cacheflush.h>
>  #include <asm/pgtable.h>
>  #include <asm/uaccess.h>
> +#include <asm/alternative.h>
>  
>  void jprobe_return_end(void);
>  static void __kprobes arch_copy_kprobe(struct kprobe *p);
> @@ -209,16 +209,12 @@ static void __kprobes arch_copy_kprobe(s
>  
>  void __kprobes arch_arm_kprobe(struct kprobe *p)
>  {
> -	*p->addr = BREAKPOINT_INSTRUCTION;
> -	flush_icache_range((unsigned long) p->addr,
> -			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
> +	text_poke(p->addr, ((unsigned char []){BREAKPOINT_INSTRUCTION}), 1);
>  }
>  
>  void __kprobes arch_disarm_kprobe(struct kprobe *p)
>  {
> -	*p->addr = p->opcode;
> -	flush_icache_range((unsigned long) p->addr,
> -			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
> +	text_poke(p->addr, &p->opcode, 1);
>  }
>  
>  void __kprobes arch_remove_kprobe(struct kprobe *p)
> Index: linux/arch/i386/kernel/alternative.c
> ===================================================================
> --- linux.orig/arch/i386/kernel/alternative.c
> +++ linux/arch/i386/kernel/alternative.c
> @@ -2,8 +2,12 @@
>  #include <linux/sched.h>
>  #include <linux/spinlock.h>
>  #include <linux/list.h>
> +#include <linux/kprobes.h>
> +#include <linux/mm.h>
> +#include <linux/vmalloc.h>
>  #include <asm/alternative.h>
>  #include <asm/sections.h>
> +#include <asm/pgtable.h>
>  
>  #ifdef CONFIG_HOTPLUG_CPU
>  static int smp_alt_once;
> @@ -150,7 +154,7 @@ static void nop_out(void *insns, unsigne
>  		unsigned int noplen = len;
>  		if (noplen > ASM_NOP_MAX)
>  			noplen = ASM_NOP_MAX;
> -		memcpy(insns, noptable[noplen], noplen);
> +		text_poke(insns, noptable[noplen], noplen);
>  		insns += noplen;
>  		len -= noplen;
>  	}
> @@ -202,7 +206,7 @@ static void alternatives_smp_lock(u8 **s
>  			continue;
>  		if (*ptr > text_end)
>  			continue;
> -		**ptr = 0xf0; /* lock prefix */
> +		text_poke(*ptr, ((unsigned char []){0xf0}), 1); /* add lock prefix */
>  	};
>  }
>  
> @@ -360,10 +364,6 @@ void apply_paravirt(struct paravirt_patc
>  		/* Pad the rest with nops */
>  		nop_out(p->instr + used, p->len - used);
>  	}
> -
> -	/* Sync to be conservative, in case we patched following
> -	 * instructions */
> -	sync_core();
>  }
>  extern struct paravirt_patch_site __start_parainstructions[],
>  	__stop_parainstructions[];
> @@ -406,3 +406,30 @@ void __init alternative_instructions(voi
>   	apply_paravirt(__parainstructions, __parainstructions_end);
>  	local_irq_restore(flags);
>  }
> +
> +/*
> + * Warning: 
> + * When you use this code to patch more than one byte of an instruction
> + * you need to make sure that other CPUs cannot execute this code in parallel.
> + * And on the local CPU you need to be protected again NMI or MCE handlers
> + * seeing an inconsistent instruction while you patch.
> + */
> +void __kprobes text_poke(void *oaddr, unsigned char *opcode, int len)
> +{
> +        u8 *addr = oaddr;
> +	if (!pte_write(*lookup_address((unsigned long)addr))) {
> +		struct page *p[2] = { virt_to_page(addr), virt_to_page(addr+PAGE_SIZE) };
> +		addr = vmap(p, 2, VM_MAP, PAGE_KERNEL);
> +		if (!addr)
> +			return;
> +		addr += ((unsigned long)oaddr) % PAGE_SIZE;
> +	}
> +	memcpy(addr, opcode, len);
> +	sync_core();
> +	/* Not strictly needed, but can speed CPU recovery up. Ignore cross cacheline
> +	   case. */
> +	if (cpu_has_clflush)
> +		asm("clflush (%0) " :: "r" (oaddr) : "memory");
> +	if (addr != oaddr)
> +		vunmap(addr);
> +}
> Index: linux/arch/x86_64/mm/pageattr.c
> ===================================================================
> --- linux.orig/arch/x86_64/mm/pageattr.c
> +++ linux/arch/x86_64/mm/pageattr.c
> @@ -13,7 +13,7 @@
>  #include <asm/tlbflush.h>
>  #include <asm/io.h>
>  
> -static inline pte_t *lookup_address(unsigned long address) 
> +pte_t *lookup_address(unsigned long address)
>  { 
>  	pgd_t *pgd = pgd_offset_k(address);
>  	pud_t *pud;
> Index: linux/include/asm-i386/alternative.h
> ===================================================================
> --- linux.orig/include/asm-i386/alternative.h
> +++ linux/include/asm-i386/alternative.h
> @@ -149,4 +149,6 @@ apply_paravirt(struct paravirt_patch_sit
>  #define __parainstructions_end	NULL
>  #endif
>  
> +extern void text_poke(void *addr, unsigned char *opcode, int len);
> +
>  #endif /* _I386_ALTERNATIVE_H */
> Index: linux/include/asm-x86_64/alternative.h
> ===================================================================
> --- linux.orig/include/asm-x86_64/alternative.h
> +++ linux/include/asm-x86_64/alternative.h
> @@ -154,4 +154,6 @@ apply_paravirt(struct paravirt_patch *st
>  #define __parainstructions_end NULL
>  #endif
>  
> +extern void text_poke(void *addr, unsigned char *opcode, int len);
> +
>  #endif /* _X86_64_ALTERNATIVE_H */
> Index: linux/include/asm-x86_64/pgtable.h
> ===================================================================
> --- linux.orig/include/asm-x86_64/pgtable.h
> +++ linux/include/asm-x86_64/pgtable.h
> @@ -403,6 +403,8 @@ extern struct list_head pgd_list;
>  
>  extern int kern_addr_valid(unsigned long addr); 
>  
> +pte_t *lookup_address(unsigned long addr);
> +
>  #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
>  		remap_pfn_range(vma, vaddr, pfn, size, prot)
>  
> Index: linux/arch/i386/kernel/kprobes.c
> ===================================================================
> --- linux.orig/arch/i386/kernel/kprobes.c
> +++ linux/arch/i386/kernel/kprobes.c
> @@ -35,6 +35,7 @@
>  #include <asm/cacheflush.h>
>  #include <asm/desc.h>
>  #include <asm/uaccess.h>
> +#include <asm/alternative.h>
>  
>  void jprobe_return_end(void);
>  
> @@ -169,16 +170,12 @@ int __kprobes arch_prepare_kprobe(struct
>  
>  void __kprobes arch_arm_kprobe(struct kprobe *p)
>  {
> -	*p->addr = BREAKPOINT_INSTRUCTION;
> -	flush_icache_range((unsigned long) p->addr,
> -			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
> +	text_poke(p->addr, ((unsigned char []){BREAKPOINT_INSTRUCTION}), 1);
>  }
>  
>  void __kprobes arch_disarm_kprobe(struct kprobe *p)
>  {
> -	*p->addr = p->opcode;
> -	flush_icache_range((unsigned long) p->addr,
> -			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
> +	text_poke(p->addr, &p->opcode, 1);
>  }
>  
>  void __kprobes arch_remove_kprobe(struct kprobe *p)
> Index: linux/arch/i386/mm/init.c
> ===================================================================
> --- linux.orig/arch/i386/mm/init.c
> +++ linux/arch/i386/mm/init.c
> @@ -800,17 +800,9 @@ void mark_rodata_ro(void)
>  	unsigned long start = PFN_ALIGN(_text);
>  	unsigned long size = PFN_ALIGN(_etext) - start;
>  
> -#ifndef CONFIG_KPROBES
> -#ifdef CONFIG_HOTPLUG_CPU
> -	/* It must still be possible to apply SMP alternatives. */
> -	if (num_possible_cpus() <= 1)
> -#endif
> -	{
> -		change_page_attr(virt_to_page(start),
> -		                 size >> PAGE_SHIFT, PAGE_KERNEL_RX);
> -		printk("Write protecting the kernel text: %luk\n", size >> 10);
> -	}
> -#endif
> +	change_page_attr(virt_to_page(start),
> +	                 size >> PAGE_SHIFT, PAGE_KERNEL_RX);
> +	printk("Write protecting the kernel text: %luk\n", size >> 10);
>  	start += size;
>  	size = (unsigned long)__end_rodata - start;
>  	change_page_attr(virt_to_page(start),
> Index: linux/arch/x86_64/mm/init.c
> ===================================================================
> --- linux.orig/arch/x86_64/mm/init.c
> +++ linux/arch/x86_64/mm/init.c
> @@ -600,16 +600,6 @@ void mark_rodata_ro(void)
>  {
>  	unsigned long start = (unsigned long)_stext, end;
>  
> -#ifdef CONFIG_HOTPLUG_CPU
> -	/* It must still be possible to apply SMP alternatives. */
> -	if (num_possible_cpus() > 1)
> -		start = (unsigned long)_etext;
> -#endif
> -
> -#ifdef CONFIG_KPROBES
> -	start = (unsigned long)__start_rodata;
> -#endif
> -	
>  	end = (unsigned long)__end_rodata;
>  	start = (start + PAGE_SIZE - 1) & PAGE_MASK;
>  	end &= PAGE_MASK;
> Index: linux/arch/i386/kernel/paravirt.c
> ===================================================================
> --- linux.orig/arch/i386/kernel/paravirt.c
> +++ linux/arch/i386/kernel/paravirt.c
> @@ -124,20 +124,28 @@ unsigned paravirt_patch_ignore(unsigned 
>  	return len;
>  }
>  
> +struct branch { 
> +	unsigned char opcode;
> +	u32 delta;
> +} __attribute__((packed)); 
> +
>  unsigned paravirt_patch_call(void *target, u16 tgt_clobbers,
>  			     void *site, u16 site_clobbers,
>  			     unsigned len)
>  {
>  	unsigned char *call = site;
>  	unsigned long delta = (unsigned long)target - (unsigned long)(call+5);
> +	struct branch b;
>  
>  	if (tgt_clobbers & ~site_clobbers)
>  		return len;	/* target would clobber too much for this site */
>  	if (len < 5)
>  		return len;	/* call too long for patch site */
>  
> -	*call++ = 0xe8;		/* call */
> -	*(unsigned long *)call = delta;
> +	b.opcode = 0xe8; /* call */
> +	b.delta = delta;
> +	BUILD_BUG_ON(sizeof(b) != 5);
> +	text_poke(call, (unsigned char *)&b, 5);
>  
>  	return 5;
>  }
> @@ -150,8 +158,9 @@ unsigned paravirt_patch_jmp(void *target
>  	if (len < 5)
>  		return len;	/* call too long for patch site */
>  
> -	*jmp++ = 0xe9;		/* jmp */
> -	*(unsigned long *)jmp = delta;
> +	b.opcode = 0xe9;	/* jmp */
> +	b.delta = delta;
> +	text_poke(call, (unsigned char *)&b, 5);
>  
>  	return 5;
>  }
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

-- 
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/