int __init construct_dom0(
struct domain *d,
unsigned long _image_start, unsigned long image_len,
unsigned long _initrd_start, unsigned long initrd_len,
char *cmdline)
{
int i, rc, compatible, compat32, order, machine;
struct cpu_user_regs *regs;
unsigned long pfn, mfn;
unsigned long nr_pages;
unsigned long nr_pt_pages;
unsigned long alloc_spfn;
unsigned long alloc_epfn;
unsigned long count;
struct page_info *page = NULL;
start_info_t *si;
struct vcpu *v = d->vcpu[0];
unsigned long long value;
#if defined(__i386__)
char *image_start = (char *)_image_start; /* use lowmem mappings */
char *initrd_start = (char *)_initrd_start; /* use lowmem mappings */
#elif defined(__x86_64__)
char *image_start = __va(_image_start);
char *initrd_start = __va(_initrd_start);
#endif
#if CONFIG_PAGING_LEVELS >= 4
l4_pgentry_t *l4tab = NULL, *l4start = NULL;
#endif
l3_pgentry_t *l3tab = NULL, *l3start = NULL;
l2_pgentry_t *l2tab = NULL, *l2start = NULL;
l1_pgentry_t *l1tab = NULL, *l1start = NULL;
/*
* This fully describes the memory layout of the initial domain. All
* *_start address are page-aligned, except v_start (and v_end) which are
* superpage-aligned.
*/
struct elf_binary elf;
struct elf_dom_parms parms;
unsigned long vkern_start;
unsigned long vkern_end;
unsigned long vinitrd_start;
unsigned long vinitrd_end;
unsigned long vphysmap_start;
unsigned long vphysmap_end;
unsigned long vstartinfo_start;
unsigned long vstartinfo_end;
unsigned long vstack_start;
unsigned long vstack_end;
unsigned long vpt_start;
unsigned long vpt_end;
unsigned long v_start;
unsigned long v_end;
/* Machine address of next candidate page-table page. */
unsigned long mpt_alloc;
/* Sanity! */
BUG_ON(d->domain_id != 0);
BUG_ON(d->vcpu[0] == NULL);
BUG_ON(v->is_initialised);
printk("*** LOADING DOMAIN 0 ***\n");
d->max_pages = ~0U;
nr_pages = compute_dom0_nr_pages();
if ( (rc = elf_init(&elf, image_start, image_len)) != 0 )
return rc;
#ifdef VERBOSE
elf_set_verbose(&elf);
#endif
elf_parse_binary(&elf);
if ( (rc = elf_xen_parse(&elf, &parms)) != 0 )
return rc;
/* compatibility check */
compatible = 0;
compat32 = 0;
machine = elf_uval(&elf, elf.ehdr, e_machine);
switch (CONFIG_PAGING_LEVELS) {
case 3: /* x86_32p */
if (parms.pae == PAEKERN_bimodal)
parms.pae = PAEKERN_extended_cr3;
printk(" Xen kernel: 32-bit, PAE, lsb\n");
if (elf_32bit(&elf) && parms.pae && machine == EM_386)
compatible = 1;
break;
case 4: /* x86_64 */
printk(" Xen kernel: 64-bit, lsb, compat32\n");
if (elf_32bit(&elf) && parms.pae == PAEKERN_bimodal)
parms.pae = PAEKERN_extended_cr3;
if (elf_32bit(&elf) && parms.pae && machine == EM_386)
{
compat32 = 1;
compatible = 1;
}
if (elf_64bit(&elf) && machine == EM_X86_64)
compatible = 1;
break;
}
printk(" Dom0 kernel: %s%s, %s, paddr 0x%" PRIx64 " -> 0x%" PRIx64 "\n",
elf_64bit(&elf) ? "64-bit" : "32-bit",
parms.pae ? ", PAE" : "",
elf_msb(&elf) ? "msb" : "lsb",
elf.pstart, elf.pend);
if ( elf.bsd_symtab_pstart )
printk(" Dom0 symbol map 0x%" PRIx64 " -> 0x%" PRIx64 "\n",
elf.bsd_symtab_pstart, elf.bsd_symtab_pend);
if ( !compatible )
{
printk("Mismatch between Xen and DOM0 kernel\n");
return -EINVAL;
}
#if defined(__x86_64__)
if ( compat32 )
{
l1_pgentry_t gdt_l1e;
d->arch.is_32bit_pv = d->arch.has_32bit_shinfo = 1;
v->vcpu_info = (void *)&d->shared_info->compat.vcpu_info[0];
if ( nr_pages != (unsigned int)nr_pages )
nr_pages = UINT_MAX;
/*
* Map compatibility Xen segments into every VCPU's GDT. See
* arch_domain_create() for further comments.
*/
gdt_l1e = l1e_from_page(virt_to_page(compat_gdt_table),
PAGE_HYPERVISOR);
for ( i = 0; i < MAX_VIRT_CPUS; i++ )
d->arch.mm_perdomain_pt[((i << GDT_LDT_VCPU_SHIFT) +
FIRST_RESERVED_GDT_PAGE)] = gdt_l1e;
flush_tlb_one_local(GDT_LDT_VIRT_START + FIRST_RESERVED_GDT_BYTE);
}
#endif
if ( parms.pae == PAEKERN_extended_cr3 )
set_bit(VMASST_TYPE_pae_extended_cr3, &d->vm_assist);
if ( (parms.virt_hv_start_low != UNSET_ADDR) && elf_32bit(&elf) )
{
unsigned long mask = (1UL << L2_PAGETABLE_SHIFT) - 1;
value = (parms.virt_hv_start_low + mask) & ~mask;
BUG_ON(!is_pv_32bit_domain(d));
#if defined(__i386__)
if ( value > HYPERVISOR_VIRT_START )
panic("Domain 0 expects too high a hypervisor start address.\n");
#else
if ( value > __HYPERVISOR_COMPAT_VIRT_START )
panic("Domain 0 expects too high a hypervisor start address.\n");
HYPERVISOR_COMPAT_VIRT_START(d) =
max_t(unsigned int, m2p_compat_vstart, value);
#endif
}
uint64_t elf_round_up(struct elf_binary *elf, uint64_t addr)
{
uint64_t elf_round = (elf_64bit(elf) ? 8 : 4) - 1;
return (addr + elf_round) & ~elf_round;
}
static int xc_dom_load_elf_symtab(struct xc_dom_image *dom,
struct elf_binary *elf, int load)
{
struct elf_binary syms;
const elf_shdr *shdr, *shdr2;
xen_vaddr_t symtab, maxaddr;
char *hdr;
size_t size;
int h, count, type, i, tables = 0;
if ( elf_swap(elf) )
{
xc_dom_printf("%s: non-native byte order, bsd symtab not supported\n",
__FUNCTION__);
return 0;
}
if ( load )
{
if ( !dom->bsd_symtab_start )
return 0;
size = dom->kernel_seg.vend - dom->bsd_symtab_start;
hdr = xc_dom_vaddr_to_ptr(dom, dom->bsd_symtab_start);
*(int *)hdr = size - sizeof(int);
}
else
{
size = sizeof(int) + elf_size(elf, elf->ehdr) +
elf_shdr_count(elf) * elf_size(elf, shdr);
hdr = xc_dom_malloc(dom, size);
if ( hdr == NULL )
return 0;
dom->bsd_symtab_start = elf_round_up(&syms, dom->kernel_seg.vend);
}
memcpy(hdr + sizeof(int),
elf->image,
elf_size(elf, elf->ehdr));
memcpy(hdr + sizeof(int) + elf_size(elf, elf->ehdr),
elf->image + elf_uval(elf, elf->ehdr, e_shoff),
elf_shdr_count(elf) * elf_size(elf, shdr));
if ( elf_64bit(elf) )
{
Elf64_Ehdr *ehdr = (Elf64_Ehdr *)(hdr + sizeof(int));
ehdr->e_phoff = 0;
ehdr->e_phentsize = 0;
ehdr->e_phnum = 0;
ehdr->e_shoff = elf_size(elf, elf->ehdr);
ehdr->e_shstrndx = SHN_UNDEF;
}
else
{
Elf32_Ehdr *ehdr = (Elf32_Ehdr *)(hdr + sizeof(int));
ehdr->e_phoff = 0;
ehdr->e_phentsize = 0;
ehdr->e_phnum = 0;
ehdr->e_shoff = elf_size(elf, elf->ehdr);
ehdr->e_shstrndx = SHN_UNDEF;
}
if ( elf_init(&syms, hdr + sizeof(int), size - sizeof(int)) )
return -1;
if ( xc_dom_logfile )
elf_set_logfile(&syms, xc_dom_logfile, 1);
symtab = dom->bsd_symtab_start + sizeof(int);
maxaddr = elf_round_up(&syms, symtab + elf_size(&syms, syms.ehdr) +
elf_shdr_count(&syms) * elf_size(&syms, shdr));
xc_dom_printf("%s/%s: bsd_symtab_start=%" PRIx64 ", kernel.end=0x%" PRIx64
" -- symtab=0x%" PRIx64 ", maxaddr=0x%" PRIx64 "\n",
__FUNCTION__, load ? "load" : "parse",
dom->bsd_symtab_start, dom->kernel_seg.vend,
symtab, maxaddr);
count = elf_shdr_count(&syms);
for ( h = 0; h < count; h++ )
{
shdr = elf_shdr_by_index(&syms, h);
type = elf_uval(&syms, shdr, sh_type);
if ( type == SHT_STRTAB )
{
/* Look for a strtab @i linked to symtab @h. */
for ( i = 0; i < count; i++ )
{
shdr2 = elf_shdr_by_index(&syms, i);
if ( (elf_uval(&syms, shdr2, sh_type) == SHT_SYMTAB) &&
(elf_uval(&syms, shdr2, sh_link) == h) )
break;
}
/* Skip symtab @h if we found no corresponding strtab @i. */
if ( i == count )
{
if ( elf_64bit(&syms) )
*(Elf64_Off*)(&shdr->e64.sh_offset) = 0;
else
*(Elf32_Off*)(&shdr->e32.sh_offset) = 0;
continue;
}
}
if ( (type == SHT_STRTAB) || (type == SHT_SYMTAB) )
{
/* Mangled to be based on ELF header location. */
if ( elf_64bit(&syms) )
*(Elf64_Off*)(&shdr->e64.sh_offset) = maxaddr - symtab;
else
*(Elf32_Off*)(&shdr->e32.sh_offset) = maxaddr - symtab;
size = elf_uval(&syms, shdr, sh_size);
maxaddr = elf_round_up(&syms, maxaddr + size);
tables++;
xc_dom_printf("%s: h=%d %s, size=0x%zx, maxaddr=0x%" PRIx64 "\n",
__FUNCTION__, h,
type == SHT_SYMTAB ? "symtab" : "strtab",
size, maxaddr);
if ( load )
{
shdr2 = elf_shdr_by_index(elf, h);
memcpy((void*)elf_section_start(&syms, shdr),
elf_section_start(elf, shdr2),
size);
}
}
/* Name is NULL. */
if ( elf_64bit(&syms) )
*(Elf64_Half*)(&shdr->e64.sh_name) = 0;
else
*(Elf32_Word*)(&shdr->e32.sh_name) = 0;
}
if ( tables == 0 )
{
xc_dom_printf("%s: no symbol table present\n", __FUNCTION__);
dom->bsd_symtab_start = 0;
return 0;
}
if ( !load )
dom->kernel_seg.vend = maxaddr;
return 0;
}
int __init construct_dom0(
struct domain *d,
const module_t *image, unsigned long image_headroom,
module_t *initrd,
void *(*bootstrap_map)(const module_t *),
char *cmdline)
{
int i, cpu, rc, compatible, compat32, order, machine;
struct cpu_user_regs *regs;
unsigned long pfn, mfn;
unsigned long nr_pages;
unsigned long nr_pt_pages;
unsigned long alloc_spfn;
unsigned long alloc_epfn;
unsigned long initrd_pfn = -1, initrd_mfn = 0;
unsigned long count;
struct page_info *page = NULL;
start_info_t *si;
struct vcpu *v = d->vcpu[0];
unsigned long long value;
char *image_base = bootstrap_map(image);
unsigned long image_len = image->mod_end;
char *image_start = image_base + image_headroom;
unsigned long initrd_len = initrd ? initrd->mod_end : 0;
#if CONFIG_PAGING_LEVELS < 4
module_t mpt;
void *mpt_ptr;
#else
l4_pgentry_t *l4tab = NULL, *l4start = NULL;
#endif
l3_pgentry_t *l3tab = NULL, *l3start = NULL;
l2_pgentry_t *l2tab = NULL, *l2start = NULL;
l1_pgentry_t *l1tab = NULL, *l1start = NULL;
/*
* This fully describes the memory layout of the initial domain. All
* *_start address are page-aligned, except v_start (and v_end) which are
* superpage-aligned.
*/
struct elf_binary elf;
struct elf_dom_parms parms;
unsigned long vkern_start;
unsigned long vkern_end;
unsigned long vinitrd_start;
unsigned long vinitrd_end;
unsigned long vphysmap_start;
unsigned long vphysmap_end;
unsigned long vstartinfo_start;
unsigned long vstartinfo_end;
unsigned long vstack_start;
unsigned long vstack_end;
unsigned long vpt_start;
unsigned long vpt_end;
unsigned long v_start;
unsigned long v_end;
/* Machine address of next candidate page-table page. */
paddr_t mpt_alloc;
/* Sanity! */
BUG_ON(d->domain_id != 0);
BUG_ON(d->vcpu[0] == NULL);
BUG_ON(v->is_initialised);
printk("*** LOADING DOMAIN 0 ***\n");
d->max_pages = ~0U;
if ( (rc = bzimage_parse(image_base, &image_start, &image_len)) != 0 )
return rc;
if ( (rc = elf_init(&elf, image_start, image_len)) != 0 )
return rc;
#ifdef VERBOSE
elf_set_verbose(&elf);
#endif
elf_parse_binary(&elf);
if ( (rc = elf_xen_parse(&elf, &parms)) != 0 )
return rc;
/* compatibility check */
compatible = compat32 = 0;
machine = elf_uval(&elf, elf.ehdr, e_machine);
switch (CONFIG_PAGING_LEVELS) {
case 3: /* x86_32p */
if (parms.pae == PAEKERN_bimodal)
parms.pae = PAEKERN_extended_cr3;
printk(" Xen kernel: 32-bit, PAE, lsb\n");
if (elf_32bit(&elf) && parms.pae && machine == EM_386)
compatible = 1;
break;
case 4: /* x86_64 */
printk(" Xen kernel: 64-bit, lsb, compat32\n");
if (elf_32bit(&elf) && parms.pae == PAEKERN_bimodal)
parms.pae = PAEKERN_extended_cr3;
if (elf_32bit(&elf) && parms.pae && machine == EM_386)
{
compat32 = 1;
compatible = 1;
}
if (elf_64bit(&elf) && machine == EM_X86_64)
compatible = 1;
break;
}
printk(" Dom0 kernel: %s%s, %s, paddr 0x%" PRIx64 " -> 0x%" PRIx64 "\n",
elf_64bit(&elf) ? "64-bit" : "32-bit",
parms.pae ? ", PAE" : "",
elf_msb(&elf) ? "msb" : "lsb",
elf.pstart, elf.pend);
if ( elf.bsd_symtab_pstart )
printk(" Dom0 symbol map 0x%" PRIx64 " -> 0x%" PRIx64 "\n",
elf.bsd_symtab_pstart, elf.bsd_symtab_pend);
if ( !compatible )
{
printk("Mismatch between Xen and DOM0 kernel\n");
return -EINVAL;
}
if ( parms.elf_notes[XEN_ELFNOTE_SUPPORTED_FEATURES].type != XEN_ENT_NONE &&
!test_bit(XENFEAT_dom0, parms.f_supported) )
{
printk("Kernel does not support Dom0 operation\n");
return -EINVAL;
}
#if defined(__x86_64__)
if ( compat32 )
{
d->arch.is_32bit_pv = d->arch.has_32bit_shinfo = 1;
v->vcpu_info = (void *)&d->shared_info->compat.vcpu_info[0];
if ( setup_compat_arg_xlat(v) != 0 )
BUG();
}
#endif
nr_pages = compute_dom0_nr_pages(d, &parms, initrd_len);
if ( parms.pae == PAEKERN_extended_cr3 )
set_bit(VMASST_TYPE_pae_extended_cr3, &d->vm_assist);
if ( (parms.virt_hv_start_low != UNSET_ADDR) && elf_32bit(&elf) )
{
unsigned long mask = (1UL << L2_PAGETABLE_SHIFT) - 1;
value = (parms.virt_hv_start_low + mask) & ~mask;
BUG_ON(!is_pv_32bit_domain(d));
#if defined(__i386__)
if ( value > HYPERVISOR_VIRT_START )
panic("Domain 0 expects too high a hypervisor start address.\n");
#else
if ( value > __HYPERVISOR_COMPAT_VIRT_START )
panic("Domain 0 expects too high a hypervisor start address.\n");
HYPERVISOR_COMPAT_VIRT_START(d) =
max_t(unsigned int, m2p_compat_vstart, value);
#endif
}
static int __init pvh_load_kernel(struct domain *d, const module_t *image,
unsigned long image_headroom,
module_t *initrd, void *image_base,
char *cmdline, paddr_t *entry,
paddr_t *start_info_addr)
{
void *image_start = image_base + image_headroom;
unsigned long image_len = image->mod_end;
struct elf_binary elf;
struct elf_dom_parms parms;
paddr_t last_addr;
struct hvm_start_info start_info = { 0 };
struct hvm_modlist_entry mod = { 0 };
struct vcpu *v = d->vcpu[0];
int rc;
if ( (rc = bzimage_parse(image_base, &image_start, &image_len)) != 0 )
{
printk("Error trying to detect bz compressed kernel\n");
return rc;
}
if ( (rc = elf_init(&elf, image_start, image_len)) != 0 )
{
printk("Unable to init ELF\n");
return rc;
}
#ifdef VERBOSE
elf_set_verbose(&elf);
#endif
elf_parse_binary(&elf);
if ( (rc = elf_xen_parse(&elf, &parms)) != 0 )
{
printk("Unable to parse kernel for ELFNOTES\n");
return rc;
}
if ( parms.phys_entry == UNSET_ADDR32 )
{
printk("Unable to find XEN_ELFNOTE_PHYS32_ENTRY address\n");
return -EINVAL;
}
printk("OS: %s version: %s loader: %s bitness: %s\n", parms.guest_os,
parms.guest_ver, parms.loader,
elf_64bit(&elf) ? "64-bit" : "32-bit");
/* Copy the OS image and free temporary buffer. */
elf.dest_base = (void *)(parms.virt_kstart - parms.virt_base);
elf.dest_size = parms.virt_kend - parms.virt_kstart;
elf_set_vcpu(&elf, v);
rc = elf_load_binary(&elf);
if ( rc < 0 )
{
printk("Failed to load kernel: %d\n", rc);
printk("Xen dom0 kernel broken ELF: %s\n", elf_check_broken(&elf));
return rc;
}
last_addr = ROUNDUP(parms.virt_kend - parms.virt_base, PAGE_SIZE);
if ( initrd != NULL )
{
rc = hvm_copy_to_guest_phys(last_addr, mfn_to_virt(initrd->mod_start),
initrd->mod_end, v);
if ( rc )
{
printk("Unable to copy initrd to guest\n");
return rc;
}
mod.paddr = last_addr;
mod.size = initrd->mod_end;
last_addr += ROUNDUP(initrd->mod_end, PAGE_SIZE);
}
/* Free temporary buffers. */
discard_initial_images();
if ( cmdline != NULL )
{
rc = hvm_copy_to_guest_phys(last_addr, cmdline, strlen(cmdline) + 1, v);
if ( rc )
{
printk("Unable to copy guest command line\n");
return rc;
}
start_info.cmdline_paddr = last_addr;
/*
* Round up to 32/64 bits (depending on the guest kernel bitness) so
* the modlist/start_info is aligned.
*/
last_addr += ROUNDUP(strlen(cmdline) + 1, elf_64bit(&elf) ? 8 : 4);
}
if ( initrd != NULL )
{
rc = hvm_copy_to_guest_phys(last_addr, &mod, sizeof(mod), v);
if ( rc )
{
printk("Unable to copy guest modules\n");
return rc;
}
start_info.modlist_paddr = last_addr;
start_info.nr_modules = 1;
last_addr += sizeof(mod);
}
start_info.magic = XEN_HVM_START_MAGIC_VALUE;
start_info.flags = SIF_PRIVILEGED | SIF_INITDOMAIN;
rc = hvm_copy_to_guest_phys(last_addr, &start_info, sizeof(start_info), v);
if ( rc )
{
printk("Unable to copy start info to guest\n");
return rc;
}
*entry = parms.phys_entry;
*start_info_addr = last_addr;
return 0;
}
