x86/paravirt: add sysret/sysexit pvops for returning to 32-bit compatibility userspace
In a 64-bit system, we need separate sysret/sysexit operations to
return to a 32-bit userspace.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citirx.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
index a19aba8..06c451a 100644
--- a/arch/x86/kernel/asm-offsets_64.c
+++ b/arch/x86/kernel/asm-offsets_64.c
@@ -62,7 +62,9 @@
OFFSET(PV_IRQ_irq_disable, pv_irq_ops, irq_disable);
OFFSET(PV_IRQ_irq_enable, pv_irq_ops, irq_enable);
OFFSET(PV_CPU_iret, pv_cpu_ops, iret);
- OFFSET(PV_CPU_usergs_sysret, pv_cpu_ops, usergs_sysret);
+ OFFSET(PV_CPU_usergs_sysret32, pv_cpu_ops, usergs_sysret32);
+ OFFSET(PV_CPU_usergs_sysret64, pv_cpu_ops, usergs_sysret64);
+ OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit);
OFFSET(PV_CPU_swapgs, pv_cpu_ops, swapgs);
OFFSET(PV_MMU_read_cr2, pv_mmu_ops, read_cr2);
#endif
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 18447a3..880ffe5 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -59,7 +59,7 @@
#endif
#ifdef CONFIG_PARAVIRT
-ENTRY(native_usergs_sysret)
+ENTRY(native_usergs_sysret64)
swapgs
sysretq
#endif /* CONFIG_PARAVIRT */
@@ -275,7 +275,7 @@
RESTORE_ARGS 0,-ARG_SKIP,1
/*CFI_REGISTER rflags,r11*/
movq %gs:pda_oldrsp, %rsp
- USERGS_SYSRET
+ USERGS_SYSRET64
CFI_RESTORE_STATE
/* Handle reschedules */
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index b0b17f0..bf1067e 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -141,7 +141,8 @@
ret = paravirt_patch_nop();
else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) ||
type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) ||
- type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret))
+ type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret32) ||
+ type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret64))
/* If operation requires a jmp, then jmp */
ret = paravirt_patch_jmp(insnbuf, opfunc, addr, len);
else
@@ -193,7 +194,8 @@
/* These are in entry.S */
extern void native_iret(void);
extern void native_irq_enable_sysexit(void);
-extern void native_usergs_sysret(void);
+extern void native_usergs_sysret32(void);
+extern void native_usergs_sysret64(void);
static int __init print_banner(void)
{
@@ -329,10 +331,10 @@
.write_idt_entry = native_write_idt_entry,
.load_sp0 = native_load_sp0,
-#ifdef CONFIG_X86_32
.irq_enable_sysexit = native_irq_enable_sysexit,
-#else
- .usergs_sysret = native_usergs_sysret,
+#ifdef CONFIG_X86_64
+ .usergs_sysret32 = native_usergs_sysret32,
+ .usergs_sysret64 = native_usergs_sysret64,
#endif
.iret = native_iret,
.swapgs = native_swapgs,
diff --git a/arch/x86/kernel/paravirt_patch_64.c b/arch/x86/kernel/paravirt_patch_64.c
index d4c0712..061d01d 100644
--- a/arch/x86/kernel/paravirt_patch_64.c
+++ b/arch/x86/kernel/paravirt_patch_64.c
@@ -14,8 +14,9 @@
DEF_NATIVE(pv_cpu_ops, clts, "clts");
DEF_NATIVE(pv_cpu_ops, wbinvd, "wbinvd");
-/* the three commands give us more control to how to return from a syscall */
-DEF_NATIVE(pv_cpu_ops, usergs_sysret, "swapgs; sysretq;");
+DEF_NATIVE(pv_cpu_ops, irq_enable_sysexit, "swapgs; sti; sysexit");
+DEF_NATIVE(pv_cpu_ops, usergs_sysret64, "swapgs; sysretq");
+DEF_NATIVE(pv_cpu_ops, usergs_sysret32, "swapgs; sysretl");
DEF_NATIVE(pv_cpu_ops, swapgs, "swapgs");
unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
@@ -35,7 +36,9 @@
PATCH_SITE(pv_irq_ops, irq_enable);
PATCH_SITE(pv_irq_ops, irq_disable);
PATCH_SITE(pv_cpu_ops, iret);
- PATCH_SITE(pv_cpu_ops, usergs_sysret);
+ PATCH_SITE(pv_cpu_ops, irq_enable_sysexit);
+ PATCH_SITE(pv_cpu_ops, usergs_sysret32);
+ PATCH_SITE(pv_cpu_ops, usergs_sysret64);
PATCH_SITE(pv_cpu_ops, swapgs);
PATCH_SITE(pv_mmu_ops, read_cr2);
PATCH_SITE(pv_mmu_ops, read_cr3);
