[xen staging-4.15] x86/cet: Remove writeable mapping of the BSPs shadow stack

[patchbot@xxxxxxx]

commit 74aeb555745b27575c3f82051a83942ee05bd842
Author:     Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
AuthorDate: Tue Mar 15 12:07:18 2022 +0000
Commit:     Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
CommitDate: Fri Mar 25 17:10:38 2022 +0000

    x86/cet: Remove writeable mapping of the BSPs shadow stack
    
    An unintended consequence of the BSP using cpu0_stack[] is that writeable
    mappings to the BSPs shadow stacks are retained in the bss.  This renders
    CET-SS almost useless, as an attacker can update both return addresses and 
the
    ret will not fault.
    
    We specifically don't want to shatter the superpage mapping .data and .bss, 
so
    the only way to fix this is to not have the BSP stack in the main Xen image.
    
    Break cpu_alloc_stack() out of cpu_smpboot_alloc(), and dynamically allocate
    the BSP stack as early as reasonable in __start_xen().  As a consequence,
    there is no need to delay the BSP's memguard_guard_stack() call.
    
    Copy the top of cpu info block just before switching to use the new stack.
    Fix a latent bug by setting %rsp to info->guest_cpu_user_regs rather than
    ->es; this would be buggy if reinit_bsp_stack() called schedule() (which
    rewrites the GPR block) directly, but luckily it doesn't.
    
    Finally, move cpu0_stack[] into .init, so it can be reclaimed after boot.
    
    Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
    Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
    (cherry picked from commit 37786b23b027ab83051175cb8ce9ac86cacfc58e)
---
 xen/arch/x86/setup.c      | 20 +++++++++++++-------
 xen/arch/x86/smpboot.c    | 26 +++++++++++++++++++-------
 xen/arch/x86/xen.lds.S    |  2 +-
 xen/include/asm-x86/smp.h |  2 ++
 4 files changed, 35 insertions(+), 15 deletions(-)

diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c
index c8cf9ce7b1..30d028af6d 100644
--- a/xen/arch/x86/setup.c
+++ b/xen/arch/x86/setup.c
@@ -150,7 +150,7 @@ unsigned long __read_mostly xen_phys_start;
 
 unsigned long __read_mostly xen_virt_end;
 
-char __section(".bss.stack_aligned") __aligned(STACK_SIZE)
+char __section(".init.bss.stack_aligned") __aligned(STACK_SIZE)
     cpu0_stack[STACK_SIZE];
 
 struct cpuinfo_x86 __read_mostly boot_cpu_data = { 0, 0, 0, 0, -1 };
@@ -702,7 +702,6 @@ static void __init noreturn reinit_bsp_stack(void)
     percpu_traps_init();
 
     stack_base[0] = stack;
-    memguard_guard_stack(stack);
 
     rc = setup_cpu_root_pgt(0);
     if ( rc )
@@ -868,6 +867,8 @@ void __init noreturn __start_xen(unsigned long mbi_p)
 {
     char *memmap_type = NULL;
     char *cmdline, *kextra, *loader;
+    void *bsp_stack;
+    struct cpu_info *info = get_cpu_info(), *bsp_info;
     unsigned int initrdidx, num_parked = 0;
     multiboot_info_t *mbi;
     module_t *mod;
@@ -900,7 +901,7 @@ void __init noreturn __start_xen(unsigned long mbi_p)
     /* Full exception support from here on in. */
 
     rdmsrl(MSR_EFER, this_cpu(efer));
-    asm volatile ( "mov %%cr4,%0" : "=r" (get_cpu_info()->cr4) );
+    asm volatile ( "mov %%cr4,%0" : "=r" (info->cr4) );
 
     /* Enable NMIs.  Our loader (e.g. Tboot) may have left them disabled. */
     enable_nmis();
@@ -1728,6 +1729,10 @@ void __init noreturn __start_xen(unsigned long mbi_p)
      */
     vm_init();
 
+    bsp_stack = cpu_alloc_stack(0);
+    if ( !bsp_stack )
+        panic("No memory for BSP stack\n");
+
     console_init_ring();
     vesa_init();
 
@@ -1983,17 +1988,18 @@ void __init noreturn __start_xen(unsigned long mbi_p)
 
     if ( bsp_delay_spec_ctrl )
     {
-        struct cpu_info *info = get_cpu_info();
-
         info->spec_ctrl_flags &= ~SCF_use_shadow;
         barrier();
         wrmsrl(MSR_SPEC_CTRL, default_xen_spec_ctrl);
         info->last_spec_ctrl = default_xen_spec_ctrl;
     }
 
-    /* Jump to the 1:1 virtual mappings of cpu0_stack. */
+    /* Copy the cpu info block, and move onto the BSP stack. */
+    bsp_info = get_cpu_info_from_stack((unsigned long)bsp_stack);
+    *bsp_info = *info;
+
     asm volatile ("mov %[stk], %%rsp; jmp %c[fn]" ::
-                  [stk] "g" (__va(__pa(get_stack_bottom()))),
+                  [stk] "g" (&bsp_info->guest_cpu_user_regs),
                   [fn] "i" (reinit_bsp_stack) : "memory");
     unreachable();
 }
diff --git a/xen/arch/x86/smpboot.c b/xen/arch/x86/smpboot.c
index 2a2e093cff..7bb54c2207 100644
--- a/xen/arch/x86/smpboot.c
+++ b/xen/arch/x86/smpboot.c
@@ -1008,6 +1008,23 @@ static void cpu_smpboot_free(unsigned int cpu, bool 
remove)
     }
 }
 
+void *cpu_alloc_stack(unsigned int cpu)
+{
+    nodeid_t node = cpu_to_node(cpu);
+    unsigned int memflags = 0;
+    void *stack;
+
+    if ( node != NUMA_NO_NODE )
+        memflags = MEMF_node(node);
+
+    stack = alloc_xenheap_pages(STACK_ORDER, memflags);
+
+    if ( stack )
+        memguard_guard_stack(stack);
+
+    return stack;
+}
+
 static int cpu_smpboot_alloc(unsigned int cpu)
 {
     struct cpu_info *info;
@@ -1020,15 +1037,10 @@ static int cpu_smpboot_alloc(unsigned int cpu)
     if ( node != NUMA_NO_NODE )
         memflags = MEMF_node(node);
 
-    if ( stack_base[cpu] == NULL )
-    {
-        stack_base[cpu] = alloc_xenheap_pages(STACK_ORDER, memflags);
-        if ( !stack_base[cpu] )
+    if ( stack_base[cpu] == NULL &&
+         (stack_base[cpu] = cpu_alloc_stack(cpu)) == NULL )
             goto out;
 
-        memguard_guard_stack(stack_base[cpu]);
-    }
-
     info = get_cpu_info_from_stack((unsigned long)stack_base[cpu]);
     info->processor_id = cpu;
     info->per_cpu_offset = __per_cpu_offset[cpu];
diff --git a/xen/arch/x86/xen.lds.S b/xen/arch/x86/xen.lds.S
index fcc6646487..7c9758c8cc 100644
--- a/xen/arch/x86/xen.lds.S
+++ b/xen/arch/x86/xen.lds.S
@@ -205,6 +205,7 @@ SECTIONS
   } :text
   DECL_SECTION(.init.data) {
 #endif
+       *(.init.bss.stack_aligned)
 
        *(.init.rodata)
        *(.init.rodata.*)
@@ -293,7 +294,6 @@ SECTIONS
 
   DECL_SECTION(.bss) {
        __bss_start = .;
-       *(.bss.stack_aligned)
        *(.bss.page_aligned*)
        . = ALIGN(PAGE_SIZE);
        __per_cpu_start = .;
diff --git a/xen/include/asm-x86/smp.h b/xen/include/asm-x86/smp.h
index f7485f602e..9628d7842d 100644
--- a/xen/include/asm-x86/smp.h
+++ b/xen/include/asm-x86/smp.h
@@ -85,6 +85,8 @@ extern cpumask_t **socket_cpumask;
 extern unsigned int disabled_cpus;
 extern bool unaccounted_cpus;
 
+void *cpu_alloc_stack(unsigned int cpu);
+
 #endif /* !__ASSEMBLY__ */
 
 #endif
--
generated by git-patchbot for /home/xen/git/xen.git#staging-4.15