x86, vdso: Reimplement vdso.so preparation in build-time C

Currently, vdso.so files are prepared and analyzed by a combination
of objcopy, nm, some linker script tricks, and some simple ELF
parsers in the kernel.  Replace all of that with plain C code that
runs at build time.

All five vdso images now generate .c files that are compiled and
linked in to the kernel image.

This should cause only one userspace-visible change: the loaded vDSO
images are stripped more heavily than they used to be.  Everything
outside the loadable segment is dropped.  In particular, this causes
the section table and section name strings to be missing.  This
should be fine: real dynamic loaders don't load or inspect these
tables anyway.  The result is roughly equivalent to eu-strip's
--strip-sections option.

The purpose of this change is to enable the vvar and hpet mappings
to be moved to the page following the vDSO load segment.  Currently,
it is possible for the section table to extend into the page after
the load segment, so, if we map it, it risks overlapping the vvar or
hpet page.  This happens whenever the load segment is just under a
multiple of PAGE_SIZE.

The only real subtlety here is that the old code had a C file with
inline assembler that did 'call VDSO32_vsyscall' and a linker script
that defined 'VDSO32_vsyscall = __kernel_vsyscall'.  This most
likely worked by accident: the linker script entry defines a symbol
associated with an address as opposed to an alias for the real
dynamic symbol __kernel_vsyscall.  That caused ld to relocate the
reference at link time instead of leaving an interposable dynamic
relocation.  Since the VDSO32_vsyscall hack is no longer needed, I
now use 'call __kernel_vsyscall', and I added -Bsymbolic to make it
work.  vdso2c will generate an error and abort the build if the
resulting image contains any dynamic relocations, so we won't
silently generate bad vdso images.

(Dynamic relocations are a problem because nothing will even attempt
to relocate the vdso.)

Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/2c4fcf45524162a34d87fdda1eb046b2a5cecee7.1399317206.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>

arch/x86/ia32/ia32_signal.c

@@ -383,8 +383,8 @@ int ia32_setup_frame(int sig, struct ksignal *ksig,
 	} else {
 		/* Return stub is in 32bit vsyscall page */
 		if (current->mm->context.vdso)
-			restorer = VDSO32_SYMBOL(current->mm->context.vdso,
-						 sigreturn);
+			restorer = current->mm->context.vdso +
+				selected_vdso32->sym___kernel_sigreturn;
 		else
 			restorer = &frame->retcode;
 	}
@@ -462,8 +462,8 @@ int ia32_setup_rt_frame(int sig, struct ksignal *ksig,
 		if (ksig->ka.sa.sa_flags & SA_RESTORER)
 			restorer = ksig->ka.sa.sa_restorer;
 		else
-			restorer = VDSO32_SYMBOL(current->mm->context.vdso,
-						 rt_sigreturn);
+			restorer = current->mm->context.vdso +
+				selected_vdso32->sym___kernel_rt_sigreturn;
 		put_user_ex(ptr_to_compat(restorer), &frame->pretcode);
 
 		/*

arch/x86/include/asm/elf.h

@@ -299,15 +299,15 @@ do {									\
 do {									\
 	if (vdso64_enabled)						\
 		NEW_AUX_ENT(AT_SYSINFO_EHDR,				\
-			    (unsigned long)current->mm->context.vdso);	\
+			    (unsigned long __force)current->mm->context.vdso); \
 } while (0)
 
 /* As a historical oddity, the x32 and x86_64 vDSOs are controlled together. */
 #define ARCH_DLINFO_X32							\
 do {									\
 	if (vdso64_enabled)						\
 		NEW_AUX_ENT(AT_SYSINFO_EHDR,				\
-			    (unsigned long)current->mm->context.vdso);	\
+			    (unsigned long __force)current->mm->context.vdso); \
 } while (0)
 
 #define AT_SYSINFO		32
@@ -325,7 +325,8 @@ else									\
 #define VDSO_CURRENT_BASE	((unsigned long)current->mm->context.vdso)
 
 #define VDSO_ENTRY							\
-	((unsigned long)VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall))
+	((unsigned long)current->mm->context.vdso +			\
+	 selected_vdso32->sym___kernel_vsyscall)
 
 struct linux_binprm;
 

arch/x86/include/asm/mmu.h

@@ -18,7 +18,7 @@ typedef struct {
 #endif
 
 	struct mutex lock;
-	void *vdso;
+	void __user *vdso;
 } mm_context_t;
 
 #ifdef CONFIG_SMP

arch/x86/include/asm/vdso.h

@@ -3,63 +3,43 @@
 
 #include <asm/page_types.h>
 #include <linux/linkage.h>
+#include <linux/init.h>
 
-#ifdef __ASSEMBLER__
+#ifndef __ASSEMBLER__
 
-#define DEFINE_VDSO_IMAGE(symname, filename)				\
-__PAGE_ALIGNED_DATA ;							\
-	.globl symname##_start, symname##_end ;				\
-	.align PAGE_SIZE ;						\
-	symname##_start: ;						\
-	.incbin filename ;						\
-	symname##_end: ;						\
-	.align PAGE_SIZE /* extra data here leaks to userspace. */ ;	\
-									\
-.previous ;								\
-									\
-	.globl symname##_pages ;					\
-	.bss ;								\
-	.align 8 ;							\
-	.type symname##_pages, @object ;				\
-	symname##_pages: ;						\
-	.zero (symname##_end - symname##_start + PAGE_SIZE - 1) / PAGE_SIZE * (BITS_PER_LONG / 8) ; \
-	.size symname##_pages, .-symname##_pages
+struct vdso_image {
+	void *data;
+	unsigned long size;   /* Always a multiple of PAGE_SIZE */
+	struct page **pages;  /* Big enough for data/size page pointers */
 
-#else
+	unsigned long alt, alt_len;
 
-#define DECLARE_VDSO_IMAGE(symname)				\
-	extern char symname##_start[], symname##_end[];		\
-	extern struct page *symname##_pages[]
+	unsigned long sym_VDSO32_NOTE_MASK;
+	unsigned long sym___kernel_sigreturn;
+	unsigned long sym___kernel_rt_sigreturn;
+	unsigned long sym___kernel_vsyscall;
+	unsigned long sym_VDSO32_SYSENTER_RETURN;
+};
 
-#if defined CONFIG_X86_32 || defined CONFIG_COMPAT
+#ifdef CONFIG_X86_64
+extern const struct vdso_image vdso_image_64;
+#endif
 
-#include <asm/vdso32.h>
+#ifdef CONFIG_X86_X32
+extern const struct vdso_image vdso_image_x32;
+#endif
 
-DECLARE_VDSO_IMAGE(vdso32_int80);
+#if defined CONFIG_X86_32 || defined CONFIG_COMPAT
+extern const struct vdso_image vdso_image_32_int80;
 #ifdef CONFIG_COMPAT
-DECLARE_VDSO_IMAGE(vdso32_syscall);
+extern const struct vdso_image vdso_image_32_syscall;
 #endif
-DECLARE_VDSO_IMAGE(vdso32_sysenter);
+extern const struct vdso_image vdso_image_32_sysenter;
 
-/*
- * Given a pointer to the vDSO image, find the pointer to VDSO32_name
- * as that symbol is defined in the vDSO sources or linker script.
- */
-#define VDSO32_SYMBOL(base, name)					\
-({									\
-	extern const char VDSO32_##name[];				\
-	(void __user *)(VDSO32_##name + (unsigned long)(base));		\
-})
+extern const struct vdso_image *selected_vdso32;
 #endif
 
-/*
- * These symbols are defined with the addresses in the vsyscall page.
- * See vsyscall-sigreturn.S.
- */
-extern void __user __kernel_sigreturn;
-extern void __user __kernel_rt_sigreturn;
-
-void __init patch_vdso32(void *vdso, size_t len);
+extern void __init init_vdso_image(const struct vdso_image *image);
 
 #endif /* __ASSEMBLER__ */
 

arch/x86/kernel/signal.c

@@ -298,7 +298,8 @@ __setup_frame(int sig, struct ksignal *ksig, sigset_t *set,
 	}
 
 	if (current->mm->context.vdso)
-		restorer = VDSO32_SYMBOL(current->mm->context.vdso, sigreturn);
+		restorer = current->mm->context.vdso +
+			selected_vdso32->sym___kernel_sigreturn;
 	else
 		restorer = &frame->retcode;
 	if (ksig->ka.sa.sa_flags & SA_RESTORER)
@@ -361,7 +362,8 @@ static int __setup_rt_frame(int sig, struct ksignal *ksig,
 		save_altstack_ex(&frame->uc.uc_stack, regs->sp);
 
 		/* Set up to return from userspace.  */
-		restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn);
+		restorer = current->mm->context.vdso +
+			selected_vdso32->sym___kernel_sigreturn;
 		if (ksig->ka.sa.sa_flags & SA_RESTORER)
 			restorer = ksig->ka.sa.sa_restorer;
 		put_user_ex(restorer, &frame->pretcode);

arch/x86/mm/init_64.c

@@ -1223,7 +1223,8 @@ int in_gate_area_no_mm(unsigned long addr)
 
 const char *arch_vma_name(struct vm_area_struct *vma)
 {
-	if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
+	if (vma->vm_mm && vma->vm_start ==
+	    (long __force)vma->vm_mm->context.vdso)
 		return "[vdso]";
 	if (vma == &gate_vma)
 		return "[vsyscall]";

arch/x86/vdso/.gitignore

@@ -1,8 +1,7 @@
 vdso.lds
-vdso-syms.lds
 vdsox32.lds
-vdsox32-syms.lds
-vdso32-syms.lds
 vdso32-syscall-syms.lds
 vdso32-sysenter-syms.lds
 vdso32-int80-syms.lds
+vdso-image-*.c
+vdso2c

arch/x86/vdso/Makefile

@@ -24,15 +24,30 @@ vobj64s := $(filter-out $(vobjx32s-compat),$(vobjs-y))
 
 # files to link into kernel
 obj-y				+= vma.o
-obj-$(VDSO64-y)			+= vdso.o
-obj-$(VDSOX32-y)		+= vdsox32.o
-obj-$(VDSO32-y)			+= vdso32.o vdso32-setup.o
+
+# vDSO images to build
+vdso_img-$(VDSO64-y)		+= 64
+vdso_img-$(VDSOX32-y)		+= x32
+vdso_img-$(VDSO32-y)		+= 32-int80
+vdso_img-$(CONFIG_COMPAT)	+= 32-syscall
+vdso_img-$(VDSO32-y)		+= 32-sysenter
+
+obj-$(VDSO32-y)			+= vdso32-setup.o
 
 vobjs := $(foreach F,$(vobj64s),$(obj)/$F)
 
 $(obj)/vdso.o: $(obj)/vdso.so
 
-targets += vdso.so vdso.so.dbg vdso.lds $(vobjs-y)
+targets += vdso.lds $(vobjs-y)
+
+# Build the vDSO image C files and link them in.
+vdso_img_objs := $(vdso_img-y:%=vdso-image-%.o)
+vdso_img_cfiles := $(vdso_img-y:%=vdso-image-%.c)
+vdso_img_sodbg := $(vdso_img-y:%=vdso%.so.dbg)
+obj-y += $(vdso_img_objs)
+targets += $(vdso_img_cfiles)
+targets += $(vdso_img_sodbg)
+.SECONDARY: $(vdso_img-y:%=$(obj)/vdso-image-%.c)
 
 export CPPFLAGS_vdso.lds += -P -C
 
@@ -41,14 +56,18 @@ VDSO_LDFLAGS_vdso.lds = -m64 -Wl,-soname=linux-vdso.so.1 \
 			-Wl,-z,max-page-size=4096 -Wl,-z,common-page-size=4096 \
 			$(DISABLE_LTO)
 
-$(obj)/vdso.o: $(src)/vdso.S $(obj)/vdso.so
-
-$(obj)/vdso.so.dbg: $(src)/vdso.lds $(vobjs) FORCE
+$(obj)/vdso64.so.dbg: $(src)/vdso.lds $(vobjs) FORCE
 	$(call if_changed,vdso)
 
-$(obj)/%.so: OBJCOPYFLAGS := -S
-$(obj)/%.so: $(obj)/%.so.dbg FORCE
-	$(call if_changed,objcopy)
+hostprogs-y			+= vdso2c
+
+quiet_cmd_vdso2c = VDSO2C  $@
+define cmd_vdso2c
+	$(obj)/vdso2c $< $@
+endef
+
+$(obj)/vdso-image-%.c: $(obj)/vdso%.so.dbg $(obj)/vdso2c FORCE
+	$(call if_changed,vdso2c)
 
 #
 # Don't omit frame pointers for ease of userspace debugging, but do
@@ -68,22 +87,6 @@ CFLAGS_REMOVE_vclock_gettime.o = -pg
 CFLAGS_REMOVE_vgetcpu.o = -pg
 CFLAGS_REMOVE_vvar.o = -pg
 
-targets += vdso-syms.lds
-obj-$(VDSO64-y)			+= vdso-syms.lds
-
-#
-# Match symbols in the DSO that look like VDSO*; produce a file of constants.
-#
-sed-vdsosym := -e 's/^00*/0/' \
-	-e 's/^\([0-9a-fA-F]*\) . \(VDSO[a-zA-Z0-9_]*\)$$/\2 = 0x\1;/p'
-quiet_cmd_vdsosym = VDSOSYM $@
-define cmd_vdsosym
-	$(NM) $< | LC_ALL=C sed -n $(sed-vdsosym) | LC_ALL=C sort > $@
-endef
-
-$(obj)/%-syms.lds: $(obj)/%.so.dbg FORCE
-	$(call if_changed,vdsosym)
-
 #
 # X32 processes use x32 vDSO to access 64bit kernel data.
 #
@@ -94,9 +97,6 @@ $(obj)/%-syms.lds: $(obj)/%.so.dbg FORCE
 # so that it can reach 64bit address space with 64bit pointers.
 #
 
-targets += vdsox32-syms.lds
-obj-$(VDSOX32-y)		+= vdsox32-syms.lds
-
 CPPFLAGS_vdsox32.lds = $(CPPFLAGS_vdso.lds)
 VDSO_LDFLAGS_vdsox32.lds = -Wl,-m,elf32_x86_64 \
 			   -Wl,-soname=linux-vdso.so.1 \
@@ -113,17 +113,14 @@ quiet_cmd_x32 = X32     $@
 $(obj)/%-x32.o: $(obj)/%.o FORCE
 	$(call if_changed,x32)
 
-targets += vdsox32.so vdsox32.so.dbg vdsox32.lds $(vobjx32s-y)
-
-$(obj)/vdsox32.o: $(src)/vdsox32.S $(obj)/vdsox32.so
+targets += vdsox32.lds $(vobjx32s-y)
 
 $(obj)/vdsox32.so.dbg: $(src)/vdsox32.lds $(vobjx32s) FORCE
 	$(call if_changed,vdso)
 
 #
 # Build multiple 32-bit vDSO images to choose from at boot time.
 #
-obj-$(VDSO32-y)			+= vdso32-syms.lds
 vdso32.so-$(VDSO32-y)		+= int80
 vdso32.so-$(CONFIG_COMPAT)	+= syscall
 vdso32.so-$(VDSO32-y)		+= sysenter
@@ -138,10 +135,8 @@ VDSO_LDFLAGS_vdso32.lds = -m32 -Wl,-m,elf_i386 -Wl,-soname=linux-gate.so.1
 override obj-dirs = $(dir $(obj)) $(obj)/vdso32/
 
 targets += vdso32/vdso32.lds
-targets += $(vdso32-images) $(vdso32-images:=.dbg)
 targets += vdso32/note.o vdso32/vclock_gettime.o $(vdso32.so-y:%=vdso32/%.o)
-
-extra-y	+= $(vdso32-images)
+targets += vdso32/vclock_gettime.o
 
 $(obj)/vdso32.o: $(vdso32-images:%=$(obj)/%)
 
@@ -166,27 +161,6 @@ $(vdso32-images:%=$(obj)/%.dbg): $(obj)/vdso32-%.so.dbg: FORCE \
 				 $(obj)/vdso32/%.o
 	$(call if_changed,vdso)
 
-# Make vdso32-*-syms.lds from each image, and then make sure they match.
-# The only difference should be that some do not define VDSO32_SYSENTER_RETURN.
-
-targets += vdso32-syms.lds $(vdso32.so-y:%=vdso32-%-syms.lds)
-
-quiet_cmd_vdso32sym = VDSOSYM $@
-define cmd_vdso32sym
-	if LC_ALL=C sort -u $(filter-out FORCE,$^) > $(@D)/.tmp_$(@F) && \
-	   $(foreach H,$(filter-out FORCE,$^),\
-		     if grep -q VDSO32_SYSENTER_RETURN $H; \
-		     then diff -u $(@D)/.tmp_$(@F) $H; \
-		     else sed /VDSO32_SYSENTER_RETURN/d $(@D)/.tmp_$(@F) | \
-			  diff -u - $H; fi &&) : ;\
-	then mv -f $(@D)/.tmp_$(@F) $@; \
-	else rm -f $(@D)/.tmp_$(@F); exit 1; \
-	fi
-endef
-
-$(obj)/vdso32-syms.lds: $(vdso32.so-y:%=$(obj)/vdso32-%-syms.lds) FORCE
-	$(call if_changed,vdso32sym)
-
 #
 # The DSO images are built using a special linker script.
 #
@@ -197,7 +171,7 @@ quiet_cmd_vdso = VDSO    $@
 		 sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@'
 
 VDSO_LDFLAGS = -fPIC -shared $(call cc-ldoption, -Wl$(comma)--hash-style=sysv) \
-		$(LTO_CFLAGS)
+	-Wl,-Bsymbolic $(LTO_CFLAGS)
 GCOV_PROFILE := n
 
 #

arch/x86/vdso/vclock_gettime.c

@@ -154,7 +154,7 @@ notrace static long vdso_fallback_gettime(long clock, struct timespec *ts)
 	asm(
 		"mov %%ebx, %%edx \n"
 		"mov %2, %%ebx \n"
-		"call VDSO32_vsyscall \n"
+		"call __kernel_vsyscall \n"
 		"mov %%edx, %%ebx \n"
 		: "=a" (ret)
 		: "0" (__NR_clock_gettime), "g" (clock), "c" (ts)
@@ -169,7 +169,7 @@ notrace static long vdso_fallback_gtod(struct timeval *tv, struct timezone *tz)
 	asm(
 		"mov %%ebx, %%edx \n"
 		"mov %2, %%ebx \n"
-		"call VDSO32_vsyscall \n"
+		"call __kernel_vsyscall \n"
 		"mov %%edx, %%ebx \n"
 		: "=a" (ret)
 		: "0" (__NR_gettimeofday), "g" (tv), "c" (tz)