[PATCH] paravirt: Add startup infrastructure for paravirtualization

1) Each hypervisor writes a probe function to detect whether we are
   running under that hypervisor.  paravirt_probe() registers this
   function.

2) If vmlinux is booted with ring != 0, we call all the probe
   functions (with registers except %esp intact) in link order: the
   winner will not return.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>

5 files changed, 50 insertions, 0 deletions

diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 406612136049..1e8988e558c5 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -39,6 +39,8 @@ obj-$(CONFIG_VM86)		+= vm86.o
 obj-$(CONFIG_EARLY_PRINTK)      += early_printk.o
 obj-$(CONFIG_HPET_TIMER)        += hpet.o
 obj-$(CONFIG_K8_NB)             += k8.o
+
+# Make sure this is linked after any other paravirt_ops structs: see head.S
 obj-$(CONFIG_PARAVIRT)          += paravirt.o
 
 EXTRA_AFLAGS   := -traditional
diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index 5b14e95ac8b9..edef5084ce17 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -55,6 +55,12 @@
  */
 ENTRY(startup_32)
 
+#ifdef CONFIG_PARAVIRT
+        movl %cs, %eax
+        testl $0x3, %eax
+        jnz startup_paravirt
+#endif
+
 /*
  * Set segments to known values.
  */
@@ -486,6 +492,33 @@ ignore_int:
 #endif
         iret
 
+#ifdef CONFIG_PARAVIRT
+startup_paravirt:
+        cld
+        movl $(init_thread_union+THREAD_SIZE),%esp
+
+        /* We take pains to preserve all the regs. */
+        pushl   %edx
+        pushl   %ecx
+        pushl   %eax
+
+        /* paravirt.o is last in link, and that probe fn never returns */
+        pushl   $__start_paravirtprobe
+1:
+        movl    0(%esp), %eax
+        pushl   (%eax)
+        movl    8(%esp), %eax
+        call    *(%esp)
+        popl    %eax
+
+        movl    4(%esp), %eax
+        movl    8(%esp), %ecx
+        movl    12(%esp), %edx
+
+        addl    $4, (%esp)
+        jmp     1b
+#endif
+
 /*
  * Real beginning of normal "text" segment
  */
diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c
index d46460426446..5a9bd3250a1a 100644
--- a/arch/i386/kernel/paravirt.c
+++ b/arch/i386/kernel/paravirt.c
@@ -19,6 +19,7 @@
 #include <linux/module.h>
 #include <linux/efi.h>
 #include <linux/bcd.h>
+#include <linux/start_kernel.h>
 
 #include <asm/bug.h>
 #include <asm/paravirt.h>
@@ -387,6 +388,9 @@ static int __init print_banner(void)
 }
 core_initcall(print_banner);
 
+/* We simply declare start_kernel to be the paravirt probe of last resort. */
+paravirt_probe(start_kernel);
+
 struct paravirt_ops paravirt_ops = {
         .name = "bare hardware",
         .paravirt_enabled = 0,
diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index 5c69cf0e5944..877dc5cfe3a8 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -65,6 +65,12 @@ SECTIONS
         CONSTRUCTORS
         } :data
 
+  __start_paravirtprobe = .;
+  .paravirtprobe : AT(ADDR(.paravirtprobe) - LOAD_OFFSET) {
+        *(.paravirtprobe)
+  }
+  __stop_paravirtprobe = .;
+
   . = ALIGN(4096);
   .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) {
         __nosave_begin = .;
diff --git a/include/asm-i386/paravirt.h b/include/asm-i386/paravirt.h
index 081194751ade..dd707d8c8270 100644
--- a/include/asm-i386/paravirt.h
+++ b/include/asm-i386/paravirt.h
@@ -120,6 +120,11 @@ struct paravirt_ops
         void (fastcall *iret)(void);
 };
 
+/* Mark a paravirt probe function. */
+#define paravirt_probe(fn)                                              \
+ static asmlinkage void (*__paravirtprobe_##fn)(void) __attribute_used__ \
+                __attribute__((__section__(".paravirtprobe"))) = fn
+
 extern struct paravirt_ops paravirt_ops;
 
 #define paravirt_enabled() (paravirt_ops.paravirt_enabled)