[PATCH] um: Implement probe_kernel_read()

[PATCH] um: Implement probe_kernel_read()

[Richard Weinberger]

UML needs it's own probe_kernel_read() to handle kernel
mode faults correctly.
The implementation uses mincore() on the host side to detect
whether a page is owned by the UML kernel process.

This fixes also a possible crash when sysrq-t is used.
Starting with 3.10 sysrq-t calls probe_kernel_read() to
read details from the kernel workers. As kernel worker are
completely async pointers may turn NULL while reading them.

Cc: <tj@kernel.org>
Cc: <stable@vger.kernel.org> # 3.10.x
Signed-off-by: Richard Weinberger <richard@nod.at>
---
 arch/um/include/shared/os.h |  1 +
 arch/um/kernel/Makefile     |  2 +-
 arch/um/kernel/maccess.c    | 24 +++++++++++++++++++++
 arch/um/os-Linux/process.c  | 52 +++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 78 insertions(+), 1 deletion(-)
 create mode 100644 arch/um/kernel/maccess.c

diff --git a/arch/um/include/shared/os.h b/arch/um/include/shared/os.h
index 95feaa4..c70a234 100644
--- a/arch/um/include/shared/os.h
+++ b/arch/um/include/shared/os.h
@@ -200,6 +200,7 @@ extern int os_unmap_memory(void *addr, int len);
 extern int os_drop_memory(void *addr, int length);
 extern int can_drop_memory(void);
 extern void os_flush_stdout(void);
+extern int os_mincore(void *addr, unsigned long len);
 
 /* execvp.c */
 extern int execvp_noalloc(char *buf, const char *file, char *const argv[]);
diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile
index babe218..d8b78a0 100644
--- a/arch/um/kernel/Makefile
+++ b/arch/um/kernel/Makefile
@@ -13,7 +13,7 @@ clean-files :=
 obj-y = config.o exec.o exitcode.o irq.o ksyms.o mem.o \
 	physmem.o process.o ptrace.o reboot.o sigio.o \
 	signal.o smp.o syscall.o sysrq.o time.o tlb.o trap.o \
-	um_arch.o umid.o skas/
+	um_arch.o umid.o maccess.o skas/
 
 obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o
 obj-$(CONFIG_GPROF)	+= gprof_syms.o
diff --git a/arch/um/kernel/maccess.c b/arch/um/kernel/maccess.c
new file mode 100644
index 0000000..214e2e0
--- /dev/null
+++ b/arch/um/kernel/maccess.c
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2013 Richard Weinberger <richrd@nod.at>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/uaccess.h>
+#include <linux/kernel.h>
+#include <os.h>
+
+long probe_kernel_read(void *dst, const void *src, size_t size)
+{
+	void *psrc = (void *)rounddown((u64)src, PAGE_SIZE);
+
+	if ((u64)src < PAGE_SIZE || size <= 0)
+		return -EFAULT;
+
+	if (os_mincore(psrc, size) <= 0)
+		return -EFAULT;
+
+	return __probe_kernel_read(dst, src, size);
+}
diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c
index b8f34c9..67b9c8f 100644
--- a/arch/um/os-Linux/process.c
+++ b/arch/um/os-Linux/process.c
@@ -4,6 +4,7 @@
  */
 
 #include <stdio.h>
+#include <stdlib.h>
 #include <unistd.h>
 #include <errno.h>
 #include <signal.h>
@@ -232,6 +233,57 @@ out:
 	return ok;
 }
 
+static int os_page_mincore(void *addr)
+{
+	char vec[2];
+	int ret;
+
+	ret = mincore(addr, UM_KERN_PAGE_SIZE, vec);
+	if (ret < 0) {
+		if (errno == ENOMEM || errno == EINVAL)
+			return 0;
+		else
+			return -errno;
+	}
+
+	return vec[0] & 1;
+}
+
+int os_mincore(void *addr, unsigned long len)
+{
+	char *vec;
+	int ret, i;
+
+	if (len <= UM_KERN_PAGE_SIZE)
+		return os_page_mincore(addr);
+
+	vec = calloc(1, (len + UM_KERN_PAGE_SIZE - 1) / UM_KERN_PAGE_SIZE);
+	if (!vec)
+		return -ENOMEM;
+
+	ret = mincore(addr, UM_KERN_PAGE_SIZE, vec);
+	if (ret < 0) {
+		if (errno == ENOMEM || errno == EINVAL)
+			ret = 0;
+		else
+			ret = -errno;
+
+		goto out;
+	}
+
+	for (i = 0; i < ((len + UM_KERN_PAGE_SIZE - 1) / UM_KERN_PAGE_SIZE); i++) {
+		if (!(vec[i] & 1)) {
+			ret = 0;
+			goto out;
+		}
+	}
+
+	ret = 1;
+out:
+	free(vec);
+	return ret;
+}
+
 void init_new_thread_signals(void)
 {
 	set_handler(SIGSEGV);
-- 
1.8.3.1

Re: [PATCH] um: Implement probe_kernel_read()

[Richard Weinberger]

Am 16.08.2013 21:44, schrieb Richard Weinberger:
> UML needs it's own probe_kernel_read() to handle kernel
> mode faults correctly.
> The implementation uses mincore() on the host side to detect
> whether a page is owned by the UML kernel process.
> 
> This fixes also a possible crash when sysrq-t is used.
> Starting with 3.10 sysrq-t calls probe_kernel_read() to
> read details from the kernel workers. As kernel worker are
> completely async pointers may turn NULL while reading them.
> 
> Cc: <tj@kernel.org>
> Cc: <stable@vger.kernel.org> # 3.10.x
> Signed-off-by: Richard Weinberger <richard@nod.at>
> ---
>  arch/um/include/shared/os.h |  1 +
>  arch/um/kernel/Makefile     |  2 +-
>  arch/um/kernel/maccess.c    | 24 +++++++++++++++++++++
>  arch/um/os-Linux/process.c  | 52 +++++++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 78 insertions(+), 1 deletion(-)
>  create mode 100644 arch/um/kernel/maccess.c
> 
> diff --git a/arch/um/include/shared/os.h b/arch/um/include/shared/os.h
> index 95feaa4..c70a234 100644
> --- a/arch/um/include/shared/os.h
> +++ b/arch/um/include/shared/os.h
> @@ -200,6 +200,7 @@ extern int os_unmap_memory(void *addr, int len);
>  extern int os_drop_memory(void *addr, int length);
>  extern int can_drop_memory(void);
>  extern void os_flush_stdout(void);
> +extern int os_mincore(void *addr, unsigned long len);
>  
>  /* execvp.c */
>  extern int execvp_noalloc(char *buf, const char *file, char *const argv[]);
> diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile
> index babe218..d8b78a0 100644
> --- a/arch/um/kernel/Makefile
> +++ b/arch/um/kernel/Makefile
> @@ -13,7 +13,7 @@ clean-files :=
>  obj-y = config.o exec.o exitcode.o irq.o ksyms.o mem.o \
>  	physmem.o process.o ptrace.o reboot.o sigio.o \
>  	signal.o smp.o syscall.o sysrq.o time.o tlb.o trap.o \
> -	um_arch.o umid.o skas/
> +	um_arch.o umid.o maccess.o skas/
>  
>  obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o
>  obj-$(CONFIG_GPROF)	+= gprof_syms.o
> diff --git a/arch/um/kernel/maccess.c b/arch/um/kernel/maccess.c
> new file mode 100644
> index 0000000..214e2e0
> --- /dev/null
> +++ b/arch/um/kernel/maccess.c
> @@ -0,0 +1,24 @@
> +/*
> + * Copyright (C) 2013 Richard Weinberger <richrd@nod.at>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + */
> +
> +#include <linux/uaccess.h>
> +#include <linux/kernel.h>
> +#include <os.h>
> +
> +long probe_kernel_read(void *dst, const void *src, size_t size)
> +{
> +	void *psrc = (void *)rounddown((u64)src, PAGE_SIZE);
> +
> +	if ((u64)src < PAGE_SIZE || size <= 0)
> +		return -EFAULT;
> +

*grrr*, while pressing the send button I realized that it should be s/u64/unsigned long/g.

> +	if (os_mincore(psrc, size) <= 0)
> +		return -EFAULT;
> +
> +	return __probe_kernel_read(dst, src, size);
> +}
> diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c
> index b8f34c9..67b9c8f 100644
> --- a/arch/um/os-Linux/process.c
> +++ b/arch/um/os-Linux/process.c
> @@ -4,6 +4,7 @@
>   */
>  
>  #include <stdio.h>
> +#include <stdlib.h>
>  #include <unistd.h>
>  #include <errno.h>
>  #include <signal.h>
> @@ -232,6 +233,57 @@ out:
>  	return ok;
>  }
>  
> +static int os_page_mincore(void *addr)
> +{
> +	char vec[2];
> +	int ret;
> +
> +	ret = mincore(addr, UM_KERN_PAGE_SIZE, vec);
> +	if (ret < 0) {
> +		if (errno == ENOMEM || errno == EINVAL)
> +			return 0;
> +		else
> +			return -errno;
> +	}
> +
> +	return vec[0] & 1;
> +}
> +
> +int os_mincore(void *addr, unsigned long len)
> +{
> +	char *vec;
> +	int ret, i;
> +
> +	if (len <= UM_KERN_PAGE_SIZE)
> +		return os_page_mincore(addr);
> +
> +	vec = calloc(1, (len + UM_KERN_PAGE_SIZE - 1) / UM_KERN_PAGE_SIZE);
> +	if (!vec)
> +		return -ENOMEM;
> +
> +	ret = mincore(addr, UM_KERN_PAGE_SIZE, vec);
> +	if (ret < 0) {
> +		if (errno == ENOMEM || errno == EINVAL)
> +			ret = 0;
> +		else
> +			ret = -errno;
> +
> +		goto out;
> +	}
> +
> +	for (i = 0; i < ((len + UM_KERN_PAGE_SIZE - 1) / UM_KERN_PAGE_SIZE); i++) {
> +		if (!(vec[i] & 1)) {
> +			ret = 0;
> +			goto out;
> +		}
> +	}
> +
> +	ret = 1;
> +out:
> +	free(vec);
> +	return ret;
> +}
> +
>  void init_new_thread_signals(void)
>  {
>  	set_handler(SIGSEGV);
>