text_poke, vmap and vmalloc on x86_64

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

Hi,

I am trying to figure out why the following code seems to work fine on
x86_32, but the update fails to happen on x86_64. Any idea why ? From
what I see the memcpy seems to fall in a memory area which is not the
same as the one I test at the end of the function. It only fails for
module addresses, which are vmalloc'd. Once I fix this, I'll be able to
send the patchset.

(from my own alternative.c) :

void *__kprobes text_poke(void *addr, const void *opcode, size_t len)
{
        unsigned long flags;
        char *vaddr;
        int nr_pages = 2;
        struct page *pages[2];
        int i;

        if (*((uint8_t *)addr - 1) != BREAKPOINT_INSTRUCTION) {
                BUG_ON(len > sizeof(long));
                BUG_ON((((long)addr + len - 1) & ~(sizeof(long) - 1))
                        - ((long)addr & ~(sizeof(long) - 1)));
        }
        if (is_vmalloc_addr(addr)) {
                pages[0] = vmalloc_to_page(addr);
                pages[1] = vmalloc_to_page(addr + PAGE_SIZE);
        } else {
                pages[0] = virt_to_page(addr);
                pages[1] = virt_to_page(addr + PAGE_SIZE);
        }
        BUG_ON(!pages[0]);
        if (!pages[1])
                nr_pages = 1;
        vaddr = vmap(pages, nr_pages, VM_MAP, PAGE_KERNEL);
        BUG_ON(!vaddr);
        local_irq_save(flags);
        memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len);
        local_irq_restore(flags);
        vunmap(vaddr);
        sync_core();
        /* Could also do a CLFLUSH here to speed up CPU recovery; but
           that causes hangs on some VIA CPUs. */
        for (i = 0; i < len; i++)
                BUG_ON(((char *)addr)[i] != ((char *)opcode)[i]);
        return addr;
}

Mathieu

-- 
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68