[PATCH 1/2] um: mark rodata read-only and implement _nofault accesses

Benjamin Berg benjamin at sipsolutions.net
Wed Apr 2 23:20:23 PDT 2025


Hi Nathan,

oops, that is a little logic bug in the code. When we are coming from
userspace, we may be running as part of a user task and have an mm. And
then the new "current->pagefault_disabled" logic is skipped. So when
prepend_copy is doing its nofault copy to evade a lock it is crashing
instead of failing gracefully and retrying.

In segv_handler, can you try moving that into a separate "else if"
block just above the "current->mm == NULL" check?

Something like the patch I copied below.

Benjamin

diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index cbe924a0fa87..8a2e68d07de6 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -330,20 +330,20 @@ unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user,
 			panic("Failed to sync kernel TLBs: %d", err);
 		goto out;
 	}
-	else if (current->mm == NULL) {
-		if (current->pagefault_disabled) {
-			if (!mc) {
-				show_regs(container_of(regs, struct pt_regs, regs));
-				panic("Segfault with pagefaults disabled but no mcontext");
-			}
-			if (!current->thread.segv_continue) {
-				show_regs(container_of(regs, struct pt_regs, regs));
-				panic("Segfault without recovery target");
-			}
-			mc_set_rip(mc, current->thread.segv_continue);
-			current->thread.segv_continue = NULL;
-			goto out;
+	else if (current->pagefault_disabled) {
+		if (!mc) {
+			show_regs(container_of(regs, struct pt_regs, regs));
+			panic("Segfault with pagefaults disabled but no mcontext");
 		}
+		if (!current->thread.segv_continue) {
+			show_regs(container_of(regs, struct pt_regs, regs));
+			panic("Segfault without recovery target");
+		}
+		mc_set_rip(mc, current->thread.segv_continue);
+		current->thread.segv_continue = NULL;
+		goto out;
+	}
+	else if (current->mm == NULL) {
 		show_regs(container_of(regs, struct pt_regs, regs));
 		panic("Segfault with no mm");
 	}


On Wed, 2025-04-02 at 15:12 -0700, Nathan Chancellor wrote:
> Hi Benjamin and Johannes,
> 
> On Mon, Feb 10, 2025 at 05:09:25PM +0100, Benjamin Berg wrote:
> > From: Johannes Berg <johannes.berg at intel.com>
> > 
> > Mark read-only data actually read-only (simple mprotect), and
> > to be able to test it also implement _nofault accesses. This
> > works by setting up a new "segv_continue" pointer in current,
> > and then when we hit a segfault we change the signal return
> > context so that we continue at that address. The code using
> > this sets it up so that it jumps to a label and then aborts
> > the access that way, returning -EFAULT.
> > 
> > It's possible to optimize the ___backtrack_faulted() thing by
> > using asm goto (compiler version dependent) and/or gcc's (not
> > sure if clang has it) &&label extension, but at least in one
> > attempt I made the && caused the compiler to not load -EFAULT
> > into the register in case of jumping to the &&label from the
> > fault handler. So leave it like this for now.
> > 
> > Co-developed-by: Benjamin Berg <benjamin.berg at intel.com>
> > Signed-off-by: Johannes Berg <johannes.berg at intel.com>
> > Signed-off-by: Benjamin Berg <benjamin.berg at intel.com>
> > ---
> >  arch/um/Kconfig                          |  1 +
> >  arch/um/include/asm/processor-generic.h  |  2 ++
> >  arch/um/include/asm/uaccess.h            | 20 ++++++++++++-----
> >  arch/um/include/shared/arch.h            |  2 ++
> >  arch/um/include/shared/as-layout.h       |  2 +-
> >  arch/um/include/shared/irq_user.h        |  3 ++-
> >  arch/um/include/shared/kern_util.h       | 12 ++++++----
> >  arch/um/kernel/irq.c                     |  3 ++-
> >  arch/um/kernel/mem.c                     | 10 +++++++++
> >  arch/um/kernel/trap.c                    | 28 +++++++++++++++++++-----
> >  arch/um/os-Linux/signal.c                |  4 ++--
> >  arch/um/os-Linux/skas/process.c          |  8 +++----
> >  arch/x86/um/os-Linux/mcontext.c          | 12 ++++++++++
> >  arch/x86/um/shared/sysdep/faultinfo_32.h | 12 ++++++++++
> >  arch/x86/um/shared/sysdep/faultinfo_64.h | 12 ++++++++++
> >  15 files changed, 108 insertions(+), 23 deletions(-)
> > 
> > diff --git a/arch/um/Kconfig b/arch/um/Kconfig
> > index 18051b1cfce0..79509c7f39de 100644
> > --- a/arch/um/Kconfig
> > +++ b/arch/um/Kconfig
> > @@ -12,6 +12,7 @@ config UML
> >  	select ARCH_HAS_KCOV
> >  	select ARCH_HAS_STRNCPY_FROM_USER
> >  	select ARCH_HAS_STRNLEN_USER
> > +	select ARCH_HAS_STRICT_KERNEL_RWX
> >  	select HAVE_ARCH_AUDITSYSCALL
> >  	select HAVE_ARCH_KASAN if X86_64
> >  	select HAVE_ARCH_KASAN_VMALLOC if HAVE_ARCH_KASAN
> > diff --git a/arch/um/include/asm/processor-generic.h b/arch/um/include/asm/processor-generic.h
> > index 5d6356eafffe..8a789c17acd8 100644
> > --- a/arch/um/include/asm/processor-generic.h
> > +++ b/arch/um/include/asm/processor-generic.h
> > @@ -31,6 +31,8 @@ struct thread_struct {
> >  		} thread;
> >  	} request;
> >  
> > +	void *segv_continue;
> > +
> >  	/* Contains variable sized FP registers */
> >  	struct pt_regs regs;
> >  };
> > diff --git a/arch/um/include/asm/uaccess.h b/arch/um/include/asm/uaccess.h
> > index 1d4b6bbc1b65..3a08f9029a3f 100644
> > --- a/arch/um/include/asm/uaccess.h
> > +++ b/arch/um/include/asm/uaccess.h
> > @@ -9,6 +9,7 @@
> >  
> >  #include <asm/elf.h>
> >  #include <linux/unaligned.h>
> > +#include <sysdep/faultinfo.h>
> >  
> >  #define __under_task_size(addr, size) \
> >  	(((unsigned long) (addr) < TASK_SIZE) && \
> > @@ -44,19 +45,28 @@ static inline int __access_ok(const void __user *ptr, unsigned long size)
> >  		 __access_ok_vsyscall(addr, size));
> >  }
> >  
> > -/* no pagefaults for kernel addresses in um */
> >  #define __get_kernel_nofault(dst, src, type, err_label)			\
> >  do {									\
> > -	*((type *)dst) = get_unaligned((type *)(src));			\
> > -	if (0) /* make sure the label looks used to the compiler */	\
> > +	int __faulted;							\
> > +									\
> > +	___backtrack_faulted(__faulted);				\
> > +	if (__faulted) {						\
> > +		*((type *)dst) = (type) 0;				\
> >  		goto err_label;						\
> > +	}								\
> > +	*((type *)dst) = get_unaligned((type *)(src));			\
> > +	current->thread.segv_continue = NULL;				\
> >  } while (0)
> >  
> >  #define __put_kernel_nofault(dst, src, type, err_label)			\
> >  do {									\
> > -	put_unaligned(*((type *)src), (type *)(dst));			\
> > -	if (0) /* make sure the label looks used to the compiler */	\
> > +	int __faulted;							\
> > +									\
> > +	___backtrack_faulted(__faulted);				\
> > +	if (__faulted)							\
> >  		goto err_label;						\
> > +	put_unaligned(*((type *)src), (type *)(dst));			\
> > +	current->thread.segv_continue = NULL;				\
> >  } while (0)
> >  
> >  #endif
> ...
> > diff --git a/arch/x86/um/shared/sysdep/faultinfo_64.h b/arch/x86/um/shared/sysdep/faultinfo_64.h
> > index ee88f88974ea..26fb4835d3e9 100644
> > --- a/arch/x86/um/shared/sysdep/faultinfo_64.h
> > +++ b/arch/x86/um/shared/sysdep/faultinfo_64.h
> > @@ -29,4 +29,16 @@ struct faultinfo {
> >  
> >  #define PTRACE_FULL_FAULTINFO 1
> >  
> > +#define ___backtrack_faulted(_faulted)					\
> > +	asm volatile (							\
> > +		"mov $0, %0\n"						\
> > +		"movq $__get_kernel_nofault_faulted_%=,%1\n"		\
> > +		"jmp _end_%=\n"						\
> > +		"__get_kernel_nofault_faulted_%=:\n"			\
> > +		"mov $1, %0;"						\
> > +		"_end_%=:"						\
> > +		: "=r" (_faulted),					\
> > +		  "=m" (current->thread.segv_continue) ::		\
> > +	)
> > +
> >  #endif
> 
> I bisected a crash that our CI sees with this change as
> commit d1d7f01f7cd3 ("um: mark rodata read-only and implement _nofault
> accesses"). Unfortunately, I only see this with clang, I do not see it
> with GCC 14.2.0 but I am not sure where I would start trying to figure
> out what is going on here.
> 
>   $ make -skj"$(nproc)" ARCH=um LLVM=1 mrproper defconfig vmlinux
> 
>   # Get rootfs via
>   #   curl -LSs https://github.com/ClangBuiltLinux/boot-utils/releases/download/20241120-044434/x86_64-rootfs.ext4.zst | zstd -d >rootfs.ext4
>   # if necessary
>   $ ./vmlinux ubd0=$PWD/rootfs.ext4
>   Core dump limits :
>           soft - NONE
>           hard - NONE
>   Checking that ptrace can change system call numbers...OK
>   Checking syscall emulation for ptrace...OK
>   Checking environment variables for a tempdir...none found
>   Checking if /dev/shm is on tmpfs...OK
>   Checking PROT_EXEC mmap in /dev/shm...OK
>   Linux version 6.14.0-rc5-00002-gd1d7f01f7cd3 (nathan at n3-xlarge-x86) (ClangBuiltLinux clang version 20.1.2 (https://github.com/llvm/llvm-project.git 58df0ef89dd64126512e4ee27b4ac3fd8ddf6247), ClangBuiltLinux LLD 20.1.2 (https://github.com/llvm/llvm-project.git 58df0ef89dd64126512e4ee27b4ac3fd8ddf6247)) #1 Wed Apr  2 15:07:50 MST 2025
>   ...
>   EXT4-fs (ubda): orphan cleanup on readonly fs
>   EXT4-fs (ubda): mounted filesystem 267f8e03-3e3c-4aec-979b-f4d9b964b716 ro with ordered data mode. Quota mode: none.
>   VFS: Mounted root (ext4 filesystem) readonly on device 98:0.
>   devtmpfs: mounted
>   Run /sbin/init as init process
> 
>   Modules linked in:
>   Pid: 24, comm: mount Not tainted 6.14.0-rc5-00002-gd1d7f01f7cd3
>   RIP: 0033:_end_0+0x38/0x45
>   RSP: 00000000648bfca0  EFLAGS: 00010206
>   RAX: 0000000000000000 RBX: 00000000609c7ff8 RCX: 00000000606350b8
>   RDX: 00000000648bfc5e RSI: 0000000000001000 RDI: 0000000060c0c000
>   RBP: 00000000648bfcc0 R08: 0000000000000000 R09: 0000000060c0c780
>   R10: 0000000000000000 R11: 0000000000000206 R12: 00000000648bfd60
>   R13: 0000000060c0c900 R14: 0000000060c0c9f8 R15: 0000000000000007
>   Kernel panic - not syncing: Kernel mode fault at addr 0x790, ip 0x600c268a
>   CPU: 0 UID: 0 PID: 24 Comm: mount Not tainted 6.14.0-rc5-00002-gd1d7f01f7cd3 #1
>   Stack:
>    600c25e9 00000007 609c7ff8 608c4780
>    648bfcf0 601533ca 601533ab 648bfd60
>    648bfdb0 608c4780 648bfd10 60152ebe
>   Call Trace:
>    [<600c25e9>] ? copy_from_kernel_nofault+0x0/0x64
>    [<601533ca>] prepend_copy+0x1f/0x51
>    [<601533ab>] ? prepend_copy+0x0/0x51
>    [<60152ebe>] prepend+0x60/0x67
>    [<60153379>] prepend_name+0x1f/0x51
>    [<6015335a>] ? prepend_name+0x0/0x51
>    [<60152bad>] prepend_path+0xb9/0x1d8
>    [<6003cb83>] ? um_set_signals+0x0/0x3b
>    [<60152e17>] d_path+0xce/0x115
>    [<601867f5>] proc_pid_readlink+0xa1/0x17d
>    [<6012b182>] vfs_readlink+0xec/0xee
>    [<60138920>] ? touch_atime+0x0/0x162
>    [<60120341>] do_readlinkat+0xbe/0x13a
>    [<6011f91b>] sys_readlinkat+0x10/0x14
>    [<6002caad>] handle_syscall+0x7b/0xaa
>    [<6002ca32>] ? handle_syscall+0x0/0xaa
>    [<6003f687>] userspace+0x289/0x4a5
>    [<6002a8e3>] fork_handler+0x56/0x5d
> 
> Cheers,
> Nathan
> 




More information about the linux-um mailing list