SKAS3 for 2.6.24-rc7

RFC PATCH New Ryan Finnie 1 participant 3 messages Jan 13, 2008

Subsystems (7)

Memory management x86 MM core User-mode Linux (UML) +3 more subsystems

lore

Share & tools

lore

Ryan Finnie

author

2008-01-13 08:10 UTC #0 lore raw

RFC PATCH 1/1 SKAS3 for 2.6.24-rc7

Please find attached a SKAS3 patch against Linux 2.6.24-rc7, based on Jeff Dike's 2007-12-08 patch for 2.6.23.

Quoteth Jeff:

Quoted 1 line

2.6.24 is going to be even more interesting, given the x86 merge.

Indeed it is. This patch is not perfect and I need some help. Host i386 and x86_64 build with CONFIG_PROC_MM (I tested i386 and it works, but for x86_64 I just made sure it compiles defconfig). Guest i386 builds correctly and runs fine, but I cannot figure this out when building guest x86_64:

CC arch/um/kernel/ptrace.o
In file included from arch/um/include/skas_ptrace.h:12,
from arch/um/kernel/ptrace.c:13:
arch/um/include/sysdep/skas_ptrace.h:9: error: redefinition of ‘struct ptrace_faultinfo’
arch/um/include/sysdep/skas_ptrace.h:14: error: redefinition of ‘struct ptrace_ldt’
make[1]: *** [arch/um/kernel/ptrace.o] Error 1
make: *** [arch/um/kernel] Error 2

sysdep-i386/skas_ptrace.h and sysdep-x86_64/skas_ptrace.h are nearly identical, so I can't figure out why ptrace_faultinfo and ptrace_ldt are being defined in the first place.

 arch/um/sys-i386/ldt.c           |    2
 arch/x86/Kconfig                 |   20 ++
 arch/x86/ia32/ptrace32.c         |   68 +++++++
 arch/x86/ia32/sys_ia32.c         |   16 +
 arch/x86/kernel/ldt_32.c         |   58 +++---
 arch/x86/kernel/ldt_64.c         |   55 +++--
 arch/x86/kernel/ptrace_32.c      |   61 ++++++
 arch/x86/kernel/ptrace_64.c      |   70 +++++++
 arch/x86/kernel/sys_i386_32.c    |   27 ++
 arch/x86/kernel/sys_x86_64.c     |   15 +
 arch/x86/mm/Makefile_64          |    1
 arch/x86/mm/proc_mm_64.c         |   85 ++++++++
 include/asm-um/compat.h          |    9
 include/asm-um/proc_mm_32.h      |    1
 include/asm-um/proc_mm_64.h      |    1
 include/asm-um/ptrace-x86_64.h   |   11 +
 include/asm-x86/desc_32.h        |    3
 include/asm-x86/desc_64.h        |    3
 include/asm-x86/mmu_context_32.h |   19 +
 include/asm-x86/mmu_context_64.h |   20 +-
 include/asm-x86/proc_mm_32.h     |   18 +
 include/asm-x86/proc_mm_64.h     |   58 ++++++
 include/asm-x86/ptrace-abi.h     |    6
 include/asm-x86/ptrace.h         |   65 ++++++
 include/linux/mm.h               |   19 +
 include/linux/proc_mm.h          |  110 +++++++++++
 mm/Makefile                      |    5
 mm/fremap.c                      |    5
 mm/mmap.c                        |   17 -
 mm/mprotect.c                    |   20 +-
 mm/proc_mm-mod.c                 |   50 +++++
 mm/proc_mm.c                     |  299 +++++++++++++++++++++++++++++++
 32 files changed, 1133 insertions(+), 84 deletions(-)

arch/um/sys-i386/ldt.c +1 -1

diff -ruN linux-2.6.24-rc7.orig/arch/um/sys-i386/ldt.c linux-2.6.24-rc7/arch/um/sys-i386/ldt.c
@@ -6,7 +6,7 @@
 #include "linux/mm.h"
 #include "asm/unistd.h"
 #include "os.h"
-#include "proc_mm.h"
+#include "linux/proc_mm.h"
 #include "skas.h"
 #include "skas_ptrace.h"
 #include "sysdep/tls.h"

arch/x86/ia32/ptrace32.c +68 -0

92 lines not on this page · raw message

arch/x86/ia32/sys_ia32.c +12 -4

diff -ruN linux-2.6.24-rc7.orig/arch/x86/ia32/sys_ia32.c linux-2.6.24-rc7/arch/x86/ia32/sys_ia32.c
@@ -693,11 +693,10 @@
 	return ret;
 }
 
-asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
-	unsigned long prot, unsigned long flags,
+long do32_mmap2(struct mm_struct *mm, unsigned long addr,
+	unsigned long len, unsigned long prot, unsigned long flags,
 	unsigned long fd, unsigned long pgoff)
 {
-	struct mm_struct *mm = current->mm;
 	unsigned long error;
 	struct file * file = NULL;

@@ -709,7 +708,7 @@
 	}
 
 	down_write(&mm->mmap_sem);
-	error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
+	error = __do_mmap_pgoff(mm, file, addr, len, prot, flags, pgoff);
 	up_write(&mm->mmap_sem);
 
 	if (file)

@@ -717,6 +716,15 @@
 	return error;
 }
 
+/* XXX: this wrapper can be probably removed, we can simply use the 64-bit
+ * version.*/
+asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
+	unsigned long prot, unsigned long flags,
+	unsigned long fd, unsigned long pgoff)
+{
+	return do32_mmap2(current->mm, addr, len, prot, flags, fd, pgoff);
+}
+
 asmlinkage long sys32_olduname(struct oldold_utsname __user * name)
 {
 	int err;

arch/x86/Kconfig +20 -0

diff -ruN linux-2.6.24-rc7.orig/arch/x86/Kconfig linux-2.6.24-rc7/arch/x86/Kconfig
@@ -813,6 +813,26 @@
 	  has the cost of more pagetable lookup overhead, and also
 	  consumes more pagetable space per process.
 
+config PROC_MM
+	bool "/proc/mm support"
+	default y
+
+config PROC_MM_DUMPABLE
+	bool "Make UML childs /proc/<pid> completely browsable"
+	default n
+	help
+	  If in doubt, say N.
+
+	  This fiddles with some settings to make sure /proc/<pid> is completely
+	  browsable by who started UML, at the expense of some additional
+	  locking (maybe this could slow down the runned UMLs of a few percents,
+	  I've not tested this).
+
+	  Also, if there is a bug in this feature, there is some little
+	  possibility to do privilege escalation if you have UML installed
+	  setuid (which you shouldn't have done) or if UML changes uid on
+	  startup (which will be a good thing, when enabled) ...
+
 # Common NUMA Features
 config NUMA
 	bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)"

arch/x86/kernel/ldt_32.c +36 ~~-22~~

144 lines not on this page · raw message

arch/x86/kernel/ldt_64.c +32 ~~-23~~

146 lines not on this page · raw message

arch/x86/kernel/ptrace_32.c +61 -0

73 lines not on this page · raw message

arch/x86/kernel/ptrace_64.c +70 -0

82 lines not on this page · raw message

arch/x86/kernel/sys_i386_32.c +21 -6

diff -ruN linux-2.6.24-rc7.orig/arch/x86/kernel/sys_i386_32.c linux-2.6.24-rc7/arch/x86/kernel/sys_i386_32.c
@@ -18,6 +18,7 @@
 #include <linux/file.h>
 #include <linux/utsname.h>
 #include <linux/ipc.h>
+#include <asm/proc_mm_32.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>

@@ -39,13 +40,12 @@
 	return error;
 }
 
-asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
-			  unsigned long prot, unsigned long flags,
-			  unsigned long fd, unsigned long pgoff)
+long do_mmap2(struct mm_struct *mm, unsigned long addr, unsigned long len,
+		unsigned long prot, unsigned long flags, unsigned long fd,
+		unsigned long pgoff)
 {
 	int error = -EBADF;
 	struct file *file = NULL;
-	struct mm_struct *mm = current->mm;
 
 	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
 	if (!(flags & MAP_ANONYMOUS)) {

@@ -55,7 +55,7 @@
 	}
 
 	down_write(&mm->mmap_sem);
-	error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
+	error = __do_mmap_pgoff(mm, file, addr, len, prot, flags, pgoff);
 	up_write(&mm->mmap_sem);
 
 	if (file)

@@ -64,6 +64,18 @@
 	return error;
 }
 
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+       unsigned long prot, unsigned long flags,
+       unsigned long fd, unsigned long pgoff)
+{
+	long ret = do_mmap2(current->mm, addr, len, prot, flags, fd, pgoff);
+
+	/* A tail call would reorder parameters on the stack and they would then
+	 * be restored at the wrong places. */
+	prevent_tail_call(ret);
+	return ret;
+}
+
 /*
  * Perform the select(nd, in, out, ex, tv) and mmap() system
  * calls. Linux/i386 didn't use to be able to handle more than

@@ -92,8 +104,11 @@
 	if (a.offset & ~PAGE_MASK)
 		goto out;
 
-	err = sys_mmap2(a.addr, a.len, a.prot, a.flags,
+	err = do_mmap2(current->mm, a.addr, a.len, a.prot, a.flags,
 			a.fd, a.offset >> PAGE_SHIFT);
+	/* A tail call would reorder parameters on the stack and they would then
+	 * be restored at the wrong places. */
+	prevent_tail_call(err);
 out:
 	return err;
 }

arch/x86/kernel/sys_x86_64.c +11 -4

diff -ruN linux-2.6.24-rc7.orig/arch/x86/kernel/sys_x86_64.c linux-2.6.24-rc7/arch/x86/kernel/sys_x86_64.c
@@ -15,6 +15,7 @@
 
 #include <asm/uaccess.h>
 #include <asm/ia32.h>
+#include <asm/proc_mm_64.h>
 
 /*
  * sys_pipe() is the normal C calling standard for creating

@@ -33,7 +34,7 @@
 	return error;
 }
 
-asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags,
+long do64_mmap(struct mm_struct *mm, unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags,
 	unsigned long fd, unsigned long off)
 {
 	long error;

@@ -51,9 +52,9 @@
 		if (!file)
 			goto out;
 	}
-	down_write(&current->mm->mmap_sem);
-	error = do_mmap_pgoff(file, addr, len, prot, flags, off >> PAGE_SHIFT);
-	up_write(&current->mm->mmap_sem);
+	down_write(&mm->mmap_sem);
+	error = __do_mmap_pgoff(mm, file, addr, len, prot, flags, off >> PAGE_SHIFT);
+	up_write(&mm->mmap_sem);
 
 	if (file)
 		fput(file);

@@ -61,6 +62,12 @@
 	return error;
 }
 
+asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags,
+	unsigned long fd, unsigned long off)
+{
+	return do64_mmap(current->mm, addr, len, prot, flags, fd, off);
+}
+
 static void find_start_end(unsigned long flags, unsigned long *begin,
 			   unsigned long *end)
 {

arch/x86/mm/Makefile_64 +1 -0

diff -ruN linux-2.6.24-rc7.orig/arch/x86/mm/Makefile_64 linux-2.6.24-rc7/arch/x86/mm/Makefile_64
@@ -7,4 +7,5 @@
 obj-$(CONFIG_NUMA) += numa_64.o
 obj-$(CONFIG_K8_NUMA) += k8topology_64.o
 obj-$(CONFIG_ACPI_NUMA) += srat_64.o
+obj-$(CONFIG_PROC_MM) += proc_mm_64.o

arch/x86/mm/proc_mm_64.c +85 -0

85 lines not on this page · raw message

include/asm-um/compat.h +9 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/compat.h linux-2.6.24-rc7/include/asm-um/compat.h
@@ -0,0 +1,9 @@
+#ifndef _ASM_X86_64_COMPAT_H
+#define _ASM_X86_64_COMPAT_H
+
+typedef s32		compat_int_t;
+typedef s32		compat_long_t;
+typedef u32		compat_uint_t;
+typedef u32		compat_ulong_t;
+
+#endif /* _ASM_X86_64_COMPAT_H */

include/asm-um/proc_mm_32.h +1 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/proc_mm_32.h linux-2.6.24-rc7/include/asm-um/proc_mm_32.h
@@ -0,0 +1 @@
+#include <asm/arch/proc_mm_32.h>

include/asm-um/proc_mm_64.h +1 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/proc_mm_64.h linux-2.6.24-rc7/include/asm-um/proc_mm_64.h
@@ -0,0 +1 @@
+#include <asm/arch/proc_mm_64.h>

include/asm-um/ptrace-x86_64.h +11 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/ptrace-x86_64.h linux-2.6.24-rc7/include/asm-um/ptrace-x86_64.h
@@ -78,4 +78,15 @@
 
 extern long arch_prctl(struct task_struct *task, int code,
 		       unsigned long __user *addr);
+
+struct ptrace_faultinfo {
+	int is_write;
+	unsigned long addr;
+};
+
+struct ptrace_ldt {
+	int func;
+  	void *ptr;
+	unsigned long bytecount;
+};
 #endif

include/asm-x86/desc_32.h +3 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/desc_32.h linux-2.6.24-rc7/include/asm-x86/desc_32.h
@@ -216,6 +216,9 @@
 	return base;
 }
 
+extern int __modify_ldt(struct mm_struct * mm, int func, void __user *ptr,
+		      unsigned long bytecount);
+
 #else /* __ASSEMBLY__ */
 
 /*

include/asm-x86/desc_64.h +3 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/desc_64.h linux-2.6.24-rc7/include/asm-x86/desc_64.h
@@ -199,6 +199,9 @@
 
 extern struct desc_ptr idt_descr;
 
+extern int __modify_ldt(struct mm_struct * mm, int func, void __user *ptr,
+		unsigned long bytecount);
+
 #endif /* !__ASSEMBLY__ */
 
 #endif

include/asm-x86/mmu_context_32.h +17 -2

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/mmu_context_32.h linux-2.6.24-rc7/include/asm-x86/mmu_context_32.h
@@ -5,6 +5,7 @@
 #include <asm/atomic.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
+#include <asm/semaphore.h>
 #include <asm/paravirt.h>
 #ifndef CONFIG_PARAVIRT
 #include <asm-generic/mm_hooks.h>

@@ -17,11 +18,22 @@
 
 
 /*
- * Used for LDT copy/destruction.
+ * Used for LDT initialization/destruction. You cannot copy an LDT with
+ * init_new_context, since it thinks you are passing it a new LDT and won't
+ * deallocate its old content.
  */
 int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
 void destroy_context(struct mm_struct *mm);
 
+/* LDT initialization for a clean environment - needed for SKAS.*/
+static inline void init_new_empty_context(struct mm_struct *mm)
+{
+	mutex_init(&mm->context.lock);
+	mm->context.size = 0;
+}
+
+/* LDT copy for SKAS - for the above problem.*/
+int copy_context(struct mm_struct *mm, struct mm_struct *old_mm);
 
 static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {

@@ -40,6 +52,10 @@
 {
 	int cpu = smp_processor_id();
 
+#ifdef CONFIG_SMP
+	prev = per_cpu(cpu_tlbstate, cpu).active_mm;
+#endif
+
 	if (likely(prev != next)) {
 		/* stop flush ipis for the previous mm */
 		cpu_clear(cpu, prev->cpu_vm_mask);

@@ -61,7 +77,6 @@
 #ifdef CONFIG_SMP
 	else {
 		per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_OK;
-		BUG_ON(per_cpu(cpu_tlbstate, cpu).active_mm != next);
 
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
 			/* We were in lazy tlb mode and leave_mm disabled

include/asm-x86/mmu_context_64.h +18 -2

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/mmu_context_64.h linux-2.6.24-rc7/include/asm-x86/mmu_context_64.h
@@ -7,14 +7,29 @@
 #include <asm/pda.h>
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
+#include <asm/semaphore.h>
 #include <asm-generic/mm_hooks.h>
 
 /*
  * possibly do the LDT unload here?
+ * Used for LDT initialization/destruction. You cannot copy an LDT with
+ * init_new_context, since it thinks you are passing it a new LDT and won't
+ * deallocate its old content.
  */
+
 int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
 void destroy_context(struct mm_struct *mm);
 
+/* LDT initialization for a clean environment - needed for SKAS.*/
+static inline void init_new_empty_context(struct mm_struct *mm)
+{
+	mutex_init(&mm->context.lock);
+	mm->context.size = 0;
+}
+
+/* LDT copy for SKAS - for the above problem.*/
+int copy_context(struct mm_struct *mm, struct mm_struct *old_mm);
+
 static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {
 #ifdef CONFIG_SMP

@@ -32,6 +47,9 @@
 			     struct task_struct *tsk)
 {
 	unsigned cpu = smp_processor_id();
+#ifdef CONFIG_SMP
+	prev = read_pda(active_mm);
+#endif
 	if (likely(prev != next)) {
 		/* stop flush ipis for the previous mm */
 		cpu_clear(cpu, prev->cpu_vm_mask);

@@ -48,8 +66,6 @@
 #ifdef CONFIG_SMP
 	else {
 		write_pda(mmu_state, TLBSTATE_OK);
-		if (read_pda(active_mm) != next)
-			out_of_line_bug();
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
 			/* We were in lazy tlb mode and leave_mm disabled 
 			 * tlb flush IPI delivery. We must reload CR3

include/asm-x86/proc_mm_32.h +18 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/proc_mm_32.h linux-2.6.24-rc7/include/asm-x86/proc_mm_32.h
@@ -0,0 +1,18 @@
+#ifndef __ASM_PROC_MM
+#define __ASM_PROC_MM
+
+#include <asm/page.h>
+
+extern long do_mmap2(struct mm_struct *mm, unsigned long addr,
+		unsigned long len, unsigned long prot, unsigned long flags,
+		unsigned long fd, unsigned long pgoff);
+
+static inline long __do_mmap(struct mm_struct *mm, unsigned long addr,
+		     unsigned long len, unsigned long prot,
+		     unsigned long flags, unsigned long fd,
+		     unsigned long off)
+{
+	return do_mmap2(mm, addr, len, prot, flags, fd, off >> PAGE_SHIFT);
+}
+
+#endif /* __ASM_PROC_MM */

include/asm-x86/proc_mm_64.h +58 -0

58 lines not on this page · raw message

include/asm-x86/ptrace-abi.h +6 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/ptrace-abi.h linux-2.6.24-rc7/include/asm-x86/ptrace-abi.h
@@ -65,6 +65,12 @@
 #define PTRACE_GETFPXREGS         18
 #define PTRACE_SETFPXREGS         19
 
+#define PTRACE_FAULTINFO 52
+/* 53 was used for PTRACE_SIGPENDING, don't reuse it. */
+#define PTRACE_LDT 54
+#define PTRACE_SWITCH_MM 55
+#define PTRACE_EX_FAULTINFO	  56
+
 #define PTRACE_OLDSETOPTIONS      21
 
 /* only useful for access 32bit programs / kernels */

include/asm-x86/ptrace.h +65 -0

69 lines not on this page · raw message

include/linux/mm.h +15 -4

diff -ruN linux-2.6.24-rc7.orig/include/linux/mm.h linux-2.6.24-rc7/include/linux/mm.h
@@ -5,6 +5,7 @@
 
 #ifdef __KERNEL__
 
+#include <linux/proc_mm.h>
 #include <linux/gfp.h>
 #include <linux/list.h>
 #include <linux/mmzone.h>

@@ -996,11 +997,18 @@
 
 extern unsigned long get_unmapped_area(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
 
-extern unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,
+extern unsigned long __do_mmap_pgoff(struct mm_struct *mm, struct file *file,
+				   unsigned long addr, unsigned long len,
+				   unsigned long prot, unsigned long flag,
+				   unsigned long pgoff);
+static inline unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,
 	unsigned long len, unsigned long prot,
-	unsigned long flag, unsigned long pgoff);
-extern unsigned long mmap_region(struct file *file, unsigned long addr,
-	unsigned long len, unsigned long flags,
+	unsigned long flag, unsigned long pgoff) {
+	return __do_mmap_pgoff(current->mm, file, addr, len, prot, flag, pgoff);
+}
+
+extern unsigned long mmap_region(struct mm_struct *mm, struct file *file,
+	unsigned long addr, unsigned long len, unsigned long flags,
 	unsigned int vm_flags, unsigned long pgoff,
 	int accountable);

@@ -1019,6 +1027,9 @@
 
 extern int do_munmap(struct mm_struct *, unsigned long, size_t);
 
+extern long do_mprotect(struct mm_struct *mm, unsigned long start,
+			size_t len, unsigned long prot);
+
 extern unsigned long do_brk(unsigned long, unsigned long);
 
 /* filemap.c */

include/linux/proc_mm.h +110 -0

110 lines not on this page · raw message

mm/fremap.c +3 -2

diff -ruN linux-2.6.24-rc7.orig/mm/fremap.c linux-2.6.24-rc7/mm/fremap.c
@@ -192,8 +192,9 @@
 			unsigned long addr;
 
 			flags &= MAP_NONBLOCK;
-			addr = mmap_region(vma->vm_file, start, size,
-					flags, vma->vm_flags, pgoff, 1);
+			addr = mmap_region(current->mm, vma->vm_file, start,
+					   size, flags, vma->vm_flags, pgoff,
+					   1);
 			if (IS_ERR_VALUE(addr)) {
 				err = addr;
 			} else {

mm/Makefile +5 -0

diff -ruN linux-2.6.24-rc7.orig/mm/Makefile linux-2.6.24-rc7/mm/Makefile
@@ -29,5 +29,10 @@
 obj-$(CONFIG_FS_XIP) += filemap_xip.o
 obj-$(CONFIG_MIGRATION) += migrate.o
 obj-$(CONFIG_SMP) += allocpercpu.o
+obj-$(CONFIG_PROC_MM)	+= proc_mm.o
+
+ifeq ($(CONFIG_PROC_MM),y)
+obj-m			+= proc_mm-mod.o
+endif
 obj-$(CONFIG_QUICKLIST) += quicklist.o

mm/mmap.c +8 -9

diff -ruN linux-2.6.24-rc7.orig/mm/mmap.c linux-2.6.24-rc7/mm/mmap.c
@@ -887,12 +887,11 @@
 /*
  * The caller must hold down_write(current->mm->mmap_sem).
  */
-
-unsigned long do_mmap_pgoff(struct file * file, unsigned long addr,
-			unsigned long len, unsigned long prot,
-			unsigned long flags, unsigned long pgoff)
+unsigned long __do_mmap_pgoff(struct mm_struct *mm, struct file * file,
+			    unsigned long addr, unsigned long len,
+			    unsigned long prot, unsigned long flags,
+			    unsigned long pgoff)
 {
-	struct mm_struct * mm = current->mm;
 	struct inode *inode;
 	unsigned int vm_flags;
 	int error;

@@ -1026,10 +1025,10 @@
 	if (error)
 		return error;
 
-	return mmap_region(file, addr, len, flags, vm_flags, pgoff,
+	return mmap_region(mm, file, addr, len, flags, vm_flags, pgoff,
 			   accountable);
 }
-EXPORT_SYMBOL(do_mmap_pgoff);
+EXPORT_SYMBOL(__do_mmap_pgoff);
 
 /*
  * Some shared mappigns will want the pages marked read-only

@@ -1064,12 +1063,12 @@
 }
 
 
-unsigned long mmap_region(struct file *file, unsigned long addr,
+unsigned long mmap_region(struct mm_struct *mm,
+			  struct file *file, unsigned long addr,
 			  unsigned long len, unsigned long flags,
 			  unsigned int vm_flags, unsigned long pgoff,
 			  int accountable)
 {
-	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma, *prev;
 	int correct_wcount = 0;
 	int error;

mm/mprotect.c +15 -5

diff -ruN linux-2.6.24-rc7.orig/mm/mprotect.c linux-2.6.24-rc7/mm/mprotect.c
@@ -211,8 +211,9 @@
 	return error;
 }
 
-asmlinkage long
-sys_mprotect(unsigned long start, size_t len, unsigned long prot)
+long
+do_mprotect(struct mm_struct *mm, unsigned long start, size_t len,
+	     unsigned long prot)
 {
 	unsigned long vm_flags, nstart, end, tmp, reqprot;
 	struct vm_area_struct *vma, *prev;

@@ -242,9 +243,9 @@
 
 	vm_flags = calc_vm_prot_bits(prot);
 
-	down_write(&current->mm->mmap_sem);
+	down_write(&mm->mmap_sem);
 
-	vma = find_vma_prev(current->mm, start, &prev);
+	vma = find_vma_prev(mm, start, &prev);
 	error = -ENOMEM;
 	if (!vma)
 		goto out;

@@ -306,6 +307,15 @@
 		}
 	}
 out:
-	up_write(&current->mm->mmap_sem);
+	up_write(&mm->mmap_sem);
 	return error;
 }
+
+asmlinkage long sys_mprotect(unsigned long start, size_t len, unsigned long prot)
+{
+	long ret = do_mprotect(current->mm, start, len, prot);
+	/* A tail call would reorder parameters on the stack and they would then
+	 * be restored at the wrong places. */
+	prevent_tail_call(ret);
+	return ret;
+}

mm/proc_mm.c +299 -0

299 lines not on this page · raw message

mm/proc_mm-mod.c +50 -0

50 lines not on this page · raw message

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Ryan Finnie

author

2008-01-13 13:03 UTC #1 lore raw

RFC PATCH 1/1 SKAS3 for 2.6.24-rc7 (2nd try)

Quoted 3 lines · Ryan Finnie

sysdep-i386/skas_ptrace.h and sysdep-x86_64/skas_ptrace.h are nearly identical, so I can't figure out why ptrace_faultinfo and ptrace_ldt are being defined in the first place.

Looks like they were added to include/asm-um/ptrace-x86_64.h to solve a disparity in x86_64 that the x86 merge fixed. Removing them let skas_ptrace.h define them, as in i386.

So here we go, attached is an updated version. I also changed CONFIG_PROC_MM_BROWSABLE to depend on CONFIG_PROC_MM. Again, the i386 host and guest build and test properly, but I have only tested building x86_64 host and guest using defconfig. They build correctly, but I haven't test booted them. (I have x86_64 hardware, but don't have a guest userland handy to test with.) Please give it a spin and let me know how it works.

Oh, and let me remind the world: I am not really a C programmer, let alone a kernel programmer. It's quite possible this patch will eat your dog. Hopefully someone like Jeff can chime in Monday and let me know that I did a horrible job porting the patch.

 arch/um/sys-i386/ldt.c           |    2
 arch/x86/Kconfig                 |   21 ++
 arch/x86/ia32/ptrace32.c         |   68 +++++++
 arch/x86/ia32/sys_ia32.c         |   16 +
 arch/x86/kernel/ldt_32.c         |   58 +++---
 arch/x86/kernel/ldt_64.c         |   55 +++--
 arch/x86/kernel/ptrace_32.c      |   61 ++++++
 arch/x86/kernel/ptrace_64.c      |   70 +++++++
 arch/x86/kernel/sys_i386_32.c    |   27 ++
 arch/x86/kernel/sys_x86_64.c     |   15 +
 arch/x86/mm/Makefile_64          |    1
 arch/x86/mm/proc_mm_64.c         |   85 ++++++++
 include/asm-um/compat.h          |    9
 include/asm-um/proc_mm_32.h      |    1
 include/asm-um/proc_mm_64.h      |    1
 include/asm-um/ptrace-x86_64.h   |    1
 include/asm-x86/desc_32.h        |    3
 include/asm-x86/desc_64.h        |    3
 include/asm-x86/mmu_context_32.h |   19 +
 include/asm-x86/mmu_context_64.h |   20 +-
 include/asm-x86/proc_mm_32.h     |   18 +
 include/asm-x86/proc_mm_64.h     |   58 ++++++
 include/asm-x86/ptrace-abi.h     |    6
 include/asm-x86/ptrace.h         |   65 ++++++
 include/linux/mm.h               |   19 +
 include/linux/proc_mm.h          |  110 +++++++++++
 mm/Makefile                      |    5
 mm/fremap.c                      |    5
 mm/mmap.c                        |   17 -
 mm/mprotect.c                    |   20 +-
 mm/proc_mm-mod.c                 |   50 +++++
 mm/proc_mm.c                     |  299 +++++++++++++++++++++++++++++++
 32 files changed, 1124 insertions(+), 84 deletions(-)

arch/um/sys-i386/ldt.c +1 -1

diff -ruN linux-2.6.24-rc7.orig/arch/um/sys-i386/ldt.c linux-2.6.24-rc7/arch/um/sys-i386/ldt.c
@@ -6,7 +6,7 @@
 #include "linux/mm.h"
 #include "asm/unistd.h"
 #include "os.h"
-#include "proc_mm.h"
+#include "linux/proc_mm.h"
 #include "skas.h"
 #include "skas_ptrace.h"
 #include "sysdep/tls.h"

arch/x86/ia32/ptrace32.c +68 -0

92 lines not on this page · raw message

arch/x86/ia32/sys_ia32.c +12 -4

diff -ruN linux-2.6.24-rc7.orig/arch/x86/ia32/sys_ia32.c linux-2.6.24-rc7/arch/x86/ia32/sys_ia32.c
@@ -693,11 +693,10 @@
 	return ret;
 }
 
-asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
-	unsigned long prot, unsigned long flags,
+long do32_mmap2(struct mm_struct *mm, unsigned long addr,
+	unsigned long len, unsigned long prot, unsigned long flags,
 	unsigned long fd, unsigned long pgoff)
 {
-	struct mm_struct *mm = current->mm;
 	unsigned long error;
 	struct file * file = NULL;

@@ -709,7 +708,7 @@
 	}
 
 	down_write(&mm->mmap_sem);
-	error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
+	error = __do_mmap_pgoff(mm, file, addr, len, prot, flags, pgoff);
 	up_write(&mm->mmap_sem);
 
 	if (file)

@@ -717,6 +716,15 @@
 	return error;
 }
 
+/* XXX: this wrapper can be probably removed, we can simply use the 64-bit
+ * version.*/
+asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
+	unsigned long prot, unsigned long flags,
+	unsigned long fd, unsigned long pgoff)
+{
+	return do32_mmap2(current->mm, addr, len, prot, flags, fd, pgoff);
+}
+
 asmlinkage long sys32_olduname(struct oldold_utsname __user * name)
 {
 	int err;

arch/x86/Kconfig +21 -0

diff -ruN linux-2.6.24-rc7.orig/arch/x86/Kconfig linux-2.6.24-rc7/arch/x86/Kconfig
@@ -813,6 +813,27 @@
 	  has the cost of more pagetable lookup overhead, and also
 	  consumes more pagetable space per process.
 
+config PROC_MM
+	bool "/proc/mm support"
+	default y
+
+config PROC_MM_DUMPABLE
+	bool "Make UML childs /proc/<pid> completely browsable"
+	default n
+	depends on PROC_MM
+	help
+	  If in doubt, say N.
+
+	  This fiddles with some settings to make sure /proc/<pid> is completely
+	  browsable by who started UML, at the expense of some additional
+	  locking (maybe this could slow down the runned UMLs of a few percents,
+	  I've not tested this).
+
+	  Also, if there is a bug in this feature, there is some little
+	  possibility to do privilege escalation if you have UML installed
+	  setuid (which you shouldn't have done) or if UML changes uid on
+	  startup (which will be a good thing, when enabled) ...
+
 # Common NUMA Features
 config NUMA
 	bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)"

arch/x86/kernel/ldt_32.c +36 ~~-22~~

144 lines not on this page · raw message

arch/x86/kernel/ldt_64.c +32 ~~-23~~

146 lines not on this page · raw message

arch/x86/kernel/ptrace_32.c +61 -0

73 lines not on this page · raw message

arch/x86/kernel/ptrace_64.c +70 -0

82 lines not on this page · raw message

arch/x86/kernel/sys_i386_32.c +21 -6

diff -ruN linux-2.6.24-rc7.orig/arch/x86/kernel/sys_i386_32.c linux-2.6.24-rc7/arch/x86/kernel/sys_i386_32.c
@@ -18,6 +18,7 @@
 #include <linux/file.h>
 #include <linux/utsname.h>
 #include <linux/ipc.h>
+#include <asm/proc_mm_32.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>

@@ -39,13 +40,12 @@
 	return error;
 }
 
-asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
-			  unsigned long prot, unsigned long flags,
-			  unsigned long fd, unsigned long pgoff)
+long do_mmap2(struct mm_struct *mm, unsigned long addr, unsigned long len,
+		unsigned long prot, unsigned long flags, unsigned long fd,
+		unsigned long pgoff)
 {
 	int error = -EBADF;
 	struct file *file = NULL;
-	struct mm_struct *mm = current->mm;
 
 	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
 	if (!(flags & MAP_ANONYMOUS)) {

@@ -55,7 +55,7 @@
 	}
 
 	down_write(&mm->mmap_sem);
-	error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
+	error = __do_mmap_pgoff(mm, file, addr, len, prot, flags, pgoff);
 	up_write(&mm->mmap_sem);
 
 	if (file)

@@ -64,6 +64,18 @@
 	return error;
 }
 
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+       unsigned long prot, unsigned long flags,
+       unsigned long fd, unsigned long pgoff)
+{
+	long ret = do_mmap2(current->mm, addr, len, prot, flags, fd, pgoff);
+
+	/* A tail call would reorder parameters on the stack and they would then
+	 * be restored at the wrong places. */
+	prevent_tail_call(ret);
+	return ret;
+}
+
 /*
  * Perform the select(nd, in, out, ex, tv) and mmap() system
  * calls. Linux/i386 didn't use to be able to handle more than

@@ -92,8 +104,11 @@
 	if (a.offset & ~PAGE_MASK)
 		goto out;
 
-	err = sys_mmap2(a.addr, a.len, a.prot, a.flags,
+	err = do_mmap2(current->mm, a.addr, a.len, a.prot, a.flags,
 			a.fd, a.offset >> PAGE_SHIFT);
+	/* A tail call would reorder parameters on the stack and they would then
+	 * be restored at the wrong places. */
+	prevent_tail_call(err);
 out:
 	return err;
 }

arch/x86/kernel/sys_x86_64.c +11 -4

diff -ruN linux-2.6.24-rc7.orig/arch/x86/kernel/sys_x86_64.c linux-2.6.24-rc7/arch/x86/kernel/sys_x86_64.c
@@ -15,6 +15,7 @@
 
 #include <asm/uaccess.h>
 #include <asm/ia32.h>
+#include <asm/proc_mm_64.h>
 
 /*
  * sys_pipe() is the normal C calling standard for creating

@@ -33,7 +34,7 @@
 	return error;
 }
 
-asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags,
+long do64_mmap(struct mm_struct *mm, unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags,
 	unsigned long fd, unsigned long off)
 {
 	long error;

@@ -51,9 +52,9 @@
 		if (!file)
 			goto out;
 	}
-	down_write(&current->mm->mmap_sem);
-	error = do_mmap_pgoff(file, addr, len, prot, flags, off >> PAGE_SHIFT);
-	up_write(&current->mm->mmap_sem);
+	down_write(&mm->mmap_sem);
+	error = __do_mmap_pgoff(mm, file, addr, len, prot, flags, off >> PAGE_SHIFT);
+	up_write(&mm->mmap_sem);
 
 	if (file)
 		fput(file);

@@ -61,6 +62,12 @@
 	return error;
 }
 
+asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags,
+	unsigned long fd, unsigned long off)
+{
+	return do64_mmap(current->mm, addr, len, prot, flags, fd, off);
+}
+
 static void find_start_end(unsigned long flags, unsigned long *begin,
 			   unsigned long *end)
 {

arch/x86/mm/Makefile_64 +1 -0

diff -ruN linux-2.6.24-rc7.orig/arch/x86/mm/Makefile_64 linux-2.6.24-rc7/arch/x86/mm/Makefile_64
@@ -7,4 +7,5 @@
 obj-$(CONFIG_NUMA) += numa_64.o
 obj-$(CONFIG_K8_NUMA) += k8topology_64.o
 obj-$(CONFIG_ACPI_NUMA) += srat_64.o
+obj-$(CONFIG_PROC_MM) += proc_mm_64.o

arch/x86/mm/proc_mm_64.c +85 -0

85 lines not on this page · raw message

include/asm-um/compat.h +9 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/compat.h linux-2.6.24-rc7/include/asm-um/compat.h
@@ -0,0 +1,9 @@
+#ifndef _ASM_X86_64_COMPAT_H
+#define _ASM_X86_64_COMPAT_H
+
+typedef s32		compat_int_t;
+typedef s32		compat_long_t;
+typedef u32		compat_uint_t;
+typedef u32		compat_ulong_t;
+
+#endif /* _ASM_X86_64_COMPAT_H */

include/asm-um/proc_mm_32.h +1 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/proc_mm_32.h linux-2.6.24-rc7/include/asm-um/proc_mm_32.h
@@ -0,0 +1 @@
+#include <asm/arch/proc_mm_32.h>

include/asm-um/proc_mm_64.h +1 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/proc_mm_64.h linux-2.6.24-rc7/include/asm-um/proc_mm_64.h
@@ -0,0 +1 @@
+#include <asm/arch/proc_mm_64.h>

include/asm-um/ptrace-x86_64.h +1 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-um/ptrace-x86_64.h linux-2.6.24-rc7/include/asm-um/ptrace-x86_64.h
@@ -78,4 +78,5 @@
 
 extern long arch_prctl(struct task_struct *task, int code,
 		       unsigned long __user *addr);
+
 #endif

include/asm-x86/desc_32.h +3 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/desc_32.h linux-2.6.24-rc7/include/asm-x86/desc_32.h
@@ -216,6 +216,9 @@
 	return base;
 }
 
+extern int __modify_ldt(struct mm_struct * mm, int func, void __user *ptr,
+		      unsigned long bytecount);
+
 #else /* __ASSEMBLY__ */
 
 /*

include/asm-x86/desc_64.h +3 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/desc_64.h linux-2.6.24-rc7/include/asm-x86/desc_64.h
@@ -199,6 +199,9 @@
 
 extern struct desc_ptr idt_descr;
 
+extern int __modify_ldt(struct mm_struct * mm, int func, void __user *ptr,
+		unsigned long bytecount);
+
 #endif /* !__ASSEMBLY__ */
 
 #endif

include/asm-x86/mmu_context_32.h +17 -2

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/mmu_context_32.h linux-2.6.24-rc7/include/asm-x86/mmu_context_32.h
@@ -5,6 +5,7 @@
 #include <asm/atomic.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
+#include <asm/semaphore.h>
 #include <asm/paravirt.h>
 #ifndef CONFIG_PARAVIRT
 #include <asm-generic/mm_hooks.h>

@@ -17,11 +18,22 @@
 
 
 /*
- * Used for LDT copy/destruction.
+ * Used for LDT initialization/destruction. You cannot copy an LDT with
+ * init_new_context, since it thinks you are passing it a new LDT and won't
+ * deallocate its old content.
  */
 int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
 void destroy_context(struct mm_struct *mm);
 
+/* LDT initialization for a clean environment - needed for SKAS.*/
+static inline void init_new_empty_context(struct mm_struct *mm)
+{
+	mutex_init(&mm->context.lock);
+	mm->context.size = 0;
+}
+
+/* LDT copy for SKAS - for the above problem.*/
+int copy_context(struct mm_struct *mm, struct mm_struct *old_mm);
 
 static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {

@@ -40,6 +52,10 @@
 {
 	int cpu = smp_processor_id();
 
+#ifdef CONFIG_SMP
+	prev = per_cpu(cpu_tlbstate, cpu).active_mm;
+#endif
+
 	if (likely(prev != next)) {
 		/* stop flush ipis for the previous mm */
 		cpu_clear(cpu, prev->cpu_vm_mask);

@@ -61,7 +77,6 @@
 #ifdef CONFIG_SMP
 	else {
 		per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_OK;
-		BUG_ON(per_cpu(cpu_tlbstate, cpu).active_mm != next);
 
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
 			/* We were in lazy tlb mode and leave_mm disabled

include/asm-x86/mmu_context_64.h +18 -2

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/mmu_context_64.h linux-2.6.24-rc7/include/asm-x86/mmu_context_64.h
@@ -7,14 +7,29 @@
 #include <asm/pda.h>
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
+#include <asm/semaphore.h>
 #include <asm-generic/mm_hooks.h>
 
 /*
  * possibly do the LDT unload here?
+ * Used for LDT initialization/destruction. You cannot copy an LDT with
+ * init_new_context, since it thinks you are passing it a new LDT and won't
+ * deallocate its old content.
  */
+
 int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
 void destroy_context(struct mm_struct *mm);
 
+/* LDT initialization for a clean environment - needed for SKAS.*/
+static inline void init_new_empty_context(struct mm_struct *mm)
+{
+	mutex_init(&mm->context.lock);
+	mm->context.size = 0;
+}
+
+/* LDT copy for SKAS - for the above problem.*/
+int copy_context(struct mm_struct *mm, struct mm_struct *old_mm);
+
 static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {
 #ifdef CONFIG_SMP

@@ -32,6 +47,9 @@
 			     struct task_struct *tsk)
 {
 	unsigned cpu = smp_processor_id();
+#ifdef CONFIG_SMP
+	prev = read_pda(active_mm);
+#endif
 	if (likely(prev != next)) {
 		/* stop flush ipis for the previous mm */
 		cpu_clear(cpu, prev->cpu_vm_mask);

@@ -48,8 +66,6 @@
 #ifdef CONFIG_SMP
 	else {
 		write_pda(mmu_state, TLBSTATE_OK);
-		if (read_pda(active_mm) != next)
-			out_of_line_bug();
 		if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
 			/* We were in lazy tlb mode and leave_mm disabled 
 			 * tlb flush IPI delivery. We must reload CR3

include/asm-x86/proc_mm_32.h +18 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/proc_mm_32.h linux-2.6.24-rc7/include/asm-x86/proc_mm_32.h
@@ -0,0 +1,18 @@
+#ifndef __ASM_PROC_MM
+#define __ASM_PROC_MM
+
+#include <asm/page.h>
+
+extern long do_mmap2(struct mm_struct *mm, unsigned long addr,
+		unsigned long len, unsigned long prot, unsigned long flags,
+		unsigned long fd, unsigned long pgoff);
+
+static inline long __do_mmap(struct mm_struct *mm, unsigned long addr,
+		     unsigned long len, unsigned long prot,
+		     unsigned long flags, unsigned long fd,
+		     unsigned long off)
+{
+	return do_mmap2(mm, addr, len, prot, flags, fd, off >> PAGE_SHIFT);
+}
+
+#endif /* __ASM_PROC_MM */

include/asm-x86/proc_mm_64.h +58 -0

58 lines not on this page · raw message

include/asm-x86/ptrace-abi.h +6 -0

diff -ruN linux-2.6.24-rc7.orig/include/asm-x86/ptrace-abi.h linux-2.6.24-rc7/include/asm-x86/ptrace-abi.h
@@ -65,6 +65,12 @@
 #define PTRACE_GETFPXREGS         18
 #define PTRACE_SETFPXREGS         19
 
+#define PTRACE_FAULTINFO 52
+/* 53 was used for PTRACE_SIGPENDING, don't reuse it. */
+#define PTRACE_LDT 54
+#define PTRACE_SWITCH_MM 55
+#define PTRACE_EX_FAULTINFO	  56
+
 #define PTRACE_OLDSETOPTIONS      21
 
 /* only useful for access 32bit programs / kernels */

include/asm-x86/ptrace.h +65 -0

69 lines not on this page · raw message

include/linux/mm.h +15 -4

diff -ruN linux-2.6.24-rc7.orig/include/linux/mm.h linux-2.6.24-rc7/include/linux/mm.h
@@ -5,6 +5,7 @@
 
 #ifdef __KERNEL__
 
+#include <linux/proc_mm.h>
 #include <linux/gfp.h>
 #include <linux/list.h>
 #include <linux/mmzone.h>

@@ -996,11 +997,18 @@
 
 extern unsigned long get_unmapped_area(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
 
-extern unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,
+extern unsigned long __do_mmap_pgoff(struct mm_struct *mm, struct file *file,
+				   unsigned long addr, unsigned long len,
+				   unsigned long prot, unsigned long flag,
+				   unsigned long pgoff);
+static inline unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,
 	unsigned long len, unsigned long prot,
-	unsigned long flag, unsigned long pgoff);
-extern unsigned long mmap_region(struct file *file, unsigned long addr,
-	unsigned long len, unsigned long flags,
+	unsigned long flag, unsigned long pgoff) {
+	return __do_mmap_pgoff(current->mm, file, addr, len, prot, flag, pgoff);
+}
+
+extern unsigned long mmap_region(struct mm_struct *mm, struct file *file,
+	unsigned long addr, unsigned long len, unsigned long flags,
 	unsigned int vm_flags, unsigned long pgoff,
 	int accountable);

@@ -1019,6 +1027,9 @@
 
 extern int do_munmap(struct mm_struct *, unsigned long, size_t);
 
+extern long do_mprotect(struct mm_struct *mm, unsigned long start,
+			size_t len, unsigned long prot);
+
 extern unsigned long do_brk(unsigned long, unsigned long);
 
 /* filemap.c */

include/linux/proc_mm.h +110 -0

110 lines not on this page · raw message

mm/fremap.c +3 -2

diff -ruN linux-2.6.24-rc7.orig/mm/fremap.c linux-2.6.24-rc7/mm/fremap.c
@@ -192,8 +192,9 @@
 			unsigned long addr;
 
 			flags &= MAP_NONBLOCK;
-			addr = mmap_region(vma->vm_file, start, size,
-					flags, vma->vm_flags, pgoff, 1);
+			addr = mmap_region(current->mm, vma->vm_file, start,
+					   size, flags, vma->vm_flags, pgoff,
+					   1);
 			if (IS_ERR_VALUE(addr)) {
 				err = addr;
 			} else {

mm/Makefile +5 -0

diff -ruN linux-2.6.24-rc7.orig/mm/Makefile linux-2.6.24-rc7/mm/Makefile
@@ -29,5 +29,10 @@
 obj-$(CONFIG_FS_XIP) += filemap_xip.o
 obj-$(CONFIG_MIGRATION) += migrate.o
 obj-$(CONFIG_SMP) += allocpercpu.o
+obj-$(CONFIG_PROC_MM)	+= proc_mm.o
+
+ifeq ($(CONFIG_PROC_MM),y)
+obj-m			+= proc_mm-mod.o
+endif
 obj-$(CONFIG_QUICKLIST) += quicklist.o

mm/mmap.c +8 -9

diff -ruN linux-2.6.24-rc7.orig/mm/mmap.c linux-2.6.24-rc7/mm/mmap.c
@@ -887,12 +887,11 @@
 /*
  * The caller must hold down_write(current->mm->mmap_sem).
  */
-
-unsigned long do_mmap_pgoff(struct file * file, unsigned long addr,
-			unsigned long len, unsigned long prot,
-			unsigned long flags, unsigned long pgoff)
+unsigned long __do_mmap_pgoff(struct mm_struct *mm, struct file * file,
+			    unsigned long addr, unsigned long len,
+			    unsigned long prot, unsigned long flags,
+			    unsigned long pgoff)
 {
-	struct mm_struct * mm = current->mm;
 	struct inode *inode;
 	unsigned int vm_flags;
 	int error;

@@ -1026,10 +1025,10 @@
 	if (error)
 		return error;
 
-	return mmap_region(file, addr, len, flags, vm_flags, pgoff,
+	return mmap_region(mm, file, addr, len, flags, vm_flags, pgoff,
 			   accountable);
 }
-EXPORT_SYMBOL(do_mmap_pgoff);
+EXPORT_SYMBOL(__do_mmap_pgoff);
 
 /*
  * Some shared mappigns will want the pages marked read-only

@@ -1064,12 +1063,12 @@
 }
 
 
-unsigned long mmap_region(struct file *file, unsigned long addr,
+unsigned long mmap_region(struct mm_struct *mm,
+			  struct file *file, unsigned long addr,
 			  unsigned long len, unsigned long flags,
 			  unsigned int vm_flags, unsigned long pgoff,
 			  int accountable)
 {
-	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma, *prev;
 	int correct_wcount = 0;
 	int error;

mm/mprotect.c +15 -5

diff -ruN linux-2.6.24-rc7.orig/mm/mprotect.c linux-2.6.24-rc7/mm/mprotect.c
@@ -211,8 +211,9 @@
 	return error;
 }
 
-asmlinkage long
-sys_mprotect(unsigned long start, size_t len, unsigned long prot)
+long
+do_mprotect(struct mm_struct *mm, unsigned long start, size_t len,
+	     unsigned long prot)
 {
 	unsigned long vm_flags, nstart, end, tmp, reqprot;
 	struct vm_area_struct *vma, *prev;

@@ -242,9 +243,9 @@
 
 	vm_flags = calc_vm_prot_bits(prot);
 
-	down_write(&current->mm->mmap_sem);
+	down_write(&mm->mmap_sem);
 
-	vma = find_vma_prev(current->mm, start, &prev);
+	vma = find_vma_prev(mm, start, &prev);
 	error = -ENOMEM;
 	if (!vma)
 		goto out;

@@ -306,6 +307,15 @@
 		}
 	}
 out:
-	up_write(&current->mm->mmap_sem);
+	up_write(&mm->mmap_sem);
 	return error;
 }
+
+asmlinkage long sys_mprotect(unsigned long start, size_t len, unsigned long prot)
+{
+	long ret = do_mprotect(current->mm, start, len, prot);
+	/* A tail call would reorder parameters on the stack and they would then
+	 * be restored at the wrong places. */
+	prevent_tail_call(ret);
+	return ret;
+}

mm/proc_mm.c +299 -0

299 lines not on this page · raw message

mm/proc_mm-mod.c +50 -0

50 lines not on this page · raw message

_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Ryan Finnie

author

2008-01-13 13:09 UTC #2 lore raw

Re: [uml-devel] [RFC PATCH] SKAS3 for 2.6.24-rc7 (2nd try)

Oh, and it appears that with a little glue (attached), the SKAS3 and SKAS4 patches can co-exist. I currently have a host running with both SKAS3 and SKAS4 guests running on it. And if you apply SKAS3, SKAS4 and the glue to a guest, it should be able to run on SKAS3 or SKAS4 hosts (with SKAS4 being preferred if the host has both SKAS3 and SKAS4).

Repeating Jeff's warning about SKAS4: "This is very experimental at this point. Don't let it near anything resembling a production system." (Though SKAS4 has been working quite well for me in testing.)

 arch/x86/ia32/ptrace32.c     |    4 ++--
 arch/x86/kernel/ptrace_32.c  |    2 +-
 arch/x86/kernel/ptrace_64.c  |    2 +-
 include/asm-x86/ptrace-abi.h |    2 +-
 include/asm-x86/ptrace.h     |    2 +-
 mm/proc_mm.c                 |    2 +-
 6 files changed, 7 insertions(+), 7 deletions(-)

arch/x86/ia32/ptrace32.c +2 -2

diff -ruN linux-2.6.24-rc7.a/arch/x86/ia32/ptrace32.c linux-2.6.24-rc7.b/arch/x86/ia32/ptrace32.c
@@ -273,7 +273,7 @@
 	case PTRACE_EX_FAULTINFO:
 	case PTRACE_FAULTINFO:
 	case PTRACE_LDT:
-	case PTRACE_SWITCH_MM:
+	case OLD_PTRACE_SWITCH_MM:
 #endif
 		break;

@@ -433,7 +433,7 @@
 		break;
 	}
 
-	case PTRACE_SWITCH_MM: {
+	case OLD_PTRACE_SWITCH_MM: {
 		struct mm_struct *old = child->mm;
 		struct mm_struct *new = proc_mm_get_mm(data);

arch/x86/kernel/ptrace_32.c +1 -1

diff -ruN linux-2.6.24-rc7.a/arch/x86/kernel/ptrace_32.c linux-2.6.24-rc7.b/arch/x86/kernel/ptrace_32.c
@@ -647,7 +647,7 @@
 		break;
 	}
 
-	case PTRACE_SWITCH_MM: {
+	case OLD_PTRACE_SWITCH_MM: {
 		struct mm_struct *old = child->mm;
 		struct mm_struct *new = proc_mm_get_mm(data);

arch/x86/kernel/ptrace_64.c +1 -1

diff -ruN linux-2.6.24-rc7.a/arch/x86/kernel/ptrace_64.c linux-2.6.24-rc7.b/arch/x86/kernel/ptrace_64.c
@@ -601,7 +601,7 @@
 		break;
 	}
 
-	case PTRACE_SWITCH_MM: {
+	case OLD_PTRACE_SWITCH_MM: {
 		struct mm_struct *old = child->mm;
 		struct mm_struct *new = proc_mm_get_mm64(data);

include/asm-x86/ptrace-abi.h +1 -1

diff -ruN linux-2.6.24-rc7.a/include/asm-x86/ptrace-abi.h linux-2.6.24-rc7.b/include/asm-x86/ptrace-abi.h
@@ -68,7 +68,7 @@
 #define PTRACE_FAULTINFO 52
 /* 53 was used for PTRACE_SIGPENDING, don't reuse it. */
 #define PTRACE_LDT 54
-#define PTRACE_SWITCH_MM 55
+#define OLD_PTRACE_SWITCH_MM 55
 #define PTRACE_EX_FAULTINFO	  56
 
 #define PTRACE_OLDSETOPTIONS      21

include/asm-x86/ptrace.h +1 -1

diff -ruN linux-2.6.24-rc7.a/include/asm-x86/ptrace.h linux-2.6.24-rc7.b/include/asm-x86/ptrace.h
@@ -146,7 +146,7 @@
 #define PTRACE_FAULTINFO	  52
 /* 53 was used for PTRACE_SIGPENDING, don't reuse it. */
 #define PTRACE_LDT		  54
-#define PTRACE_SWITCH_MM 	  55
+#define OLD_PTRACE_SWITCH_MM 	  55
 #define PTRACE_EX_FAULTINFO	  56
 
 struct ptrace_faultinfo {

mm/proc_mm.c +1 -1

diff -ruN linux-2.6.24-rc7.a/mm/proc_mm.c linux-2.6.24-rc7.b/mm/proc_mm.c
@@ -32,7 +32,7 @@
 }
 
 /*
- * This is to be used in PTRACE_SWITCH_MM handling. We are going to set
+ * This is to be used in OLD_PTRACE_SWITCH_MM handling. We are going to set
  * child->mm to new, and we must first correctly set new->dumpable.
  * Since we take task_lock of child and it's needed also by the caller, we
  * return with it locked.

_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel