| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <6cf0d59c783fd8989949d48183f886c84aa97825.1498744665.git.naveen.n.rao@linux.vnet.ibm.com>
Date: Thu, 29 Jun 2017 19:38:37 +0530
From: "Naveen N. Rao" <naveen.n.rao@...ux.vnet.ibm.com>
To: Masami Hiramatsu <mhiramat@...nel.org>
Cc: Ingo Molnar <mingo@...nel.org>,
Ananth N Mavinakayanahalli <ananth@...ux.vnet.ibm.com>,
linux-kernel@...r.kernel.org
Subject: [RFC PATCH 1/2] kprobes: Simplify kprobe blacklist management
Currently, the kprobe blacklist file in debugfs (usually
/sys/kernel/debug/kprobes/blacklist) does not capture all address
ranges that are blacklisted. For instance, functions annotated with
__kprobes (between __kprobes_text_start and __kprobes_text_end) are
blacklisted on all architectures. Certain architectures have additional
address ranges that are blacklisted.
These address ranges are currently prevented from being kprobe'd by
arch_within_kprobe_blacklist(). Instead, add the relevant address ranges
to the main kprobe blacklist allowing such address ranges to be
discovered by userspace.
With this patch, on powerpc64le, we see 2 additional entries at the end
of the blacklist:
root@...ntu:/sys/kernel/debug# tail kprobes/blacklist
0xc00000000022d500-0xc00000000022d580 print_type_s64
0xc00000000022d480-0xc00000000022d500 print_type_s32
0xc00000000022d400-0xc00000000022d480 print_type_s16
0xc00000000022d380-0xc00000000022d400 print_type_s8
0xc00000000022d300-0xc00000000022d380 print_type_u64
0xc00000000022d280-0xc00000000022d300 print_type_u32
0xc00000000022d200-0xc00000000022d280 print_type_u16
0xc00000000022d180-0xc00000000022d200 print_type_u8
0xc0000000009ed778-0xc0000000009ed778 __irqentry_text_end
0xc000000000000000-0xc000000000008000 start_first_256B
__irqentry_text_end() happens to be at __kprobes_text_start and hence
figures there. Ditto with start_first_256B(). A subsequent patch helps
provide meaningful names for such address ranges.
Signed-off-by: Naveen N. Rao <naveen.n.rao@...ux.vnet.ibm.com>
---
arch/arm64/kernel/probes/kprobes.c | 31 +++++++++---------
arch/powerpc/kernel/kprobes.c | 10 +++---
arch/x86/kernel/kprobes/core.c | 10 +++---
include/linux/kprobes.h | 2 ++
kernel/kprobes.c | 64 ++++++++++++++++++++++----------------
5 files changed, 66 insertions(+), 51 deletions(-)
diff --git a/arch/arm64/kernel/probes/kprobes.c b/arch/arm64/kernel/probes/kprobes.c
index c5c45942fb6e..7f453a71b5a8 100644
--- a/arch/arm64/kernel/probes/kprobes.c
+++ b/arch/arm64/kernel/probes/kprobes.c
@@ -535,24 +535,27 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
bool arch_within_kprobe_blacklist(unsigned long addr)
{
- if ((addr >= (unsigned long)__kprobes_text_start &&
- addr < (unsigned long)__kprobes_text_end) ||
- (addr >= (unsigned long)__entry_text_start &&
- addr < (unsigned long)__entry_text_end) ||
- (addr >= (unsigned long)__idmap_text_start &&
- addr < (unsigned long)__idmap_text_end) ||
- !!search_exception_tables(addr))
+ if (!!search_exception_tables(addr))
return true;
+ return false;
+}
+
+void __init arch_populate_kprobe_blacklist(void)
+{
+ insert_kprobe_blacklist((unsigned long)__kprobes_text_start,
+ (unsigned long)__kprobes_text_end);
+ insert_kprobe_blacklist((unsigned long)__entry_text_start,
+ (unsigned long)__entry_text_end);
+ insert_kprobe_blacklist((unsigned long)__idmap_text_start,
+ (unsigned long)__idmap_text_end);
+
if (!is_kernel_in_hyp_mode()) {
- if ((addr >= (unsigned long)__hyp_text_start &&
- addr < (unsigned long)__hyp_text_end) ||
- (addr >= (unsigned long)__hyp_idmap_text_start &&
- addr < (unsigned long)__hyp_idmap_text_end))
- return true;
+ insert_kprobe_blacklist((unsigned long)__hyp_text_start,
+ (unsigned long)__hyp_text_end);
+ insert_kprobe_blacklist((unsigned long)__hyp_idmap_text_start,
+ (unsigned long)__hyp_idmap_text_end);
}
-
- return false;
}
void __kprobes __used *trampoline_probe_handler(struct pt_regs *regs)
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index 01addfb0ed0a..f707e42b3482 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -49,12 +49,12 @@ int is_current_kprobe_addr(unsigned long addr)
return (p && (unsigned long)p->addr == addr) ? 1 : 0;
}
-bool arch_within_kprobe_blacklist(unsigned long addr)
+void __init arch_populate_kprobe_blacklist(void)
{
- return (addr >= (unsigned long)__kprobes_text_start &&
- addr < (unsigned long)__kprobes_text_end) ||
- (addr >= (unsigned long)_stext &&
- addr < (unsigned long)__head_end);
+ insert_kprobe_blacklist((unsigned long)__kprobes_text_start,
+ (unsigned long)__kprobes_text_end);
+ insert_kprobe_blacklist((unsigned long)_stext,
+ (unsigned long)__head_end);
}
kprobe_opcode_t *kprobe_lookup_name(const char *name, unsigned int offset)
diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
index 6b877807598b..4a5fc04b3e92 100644
--- a/arch/x86/kernel/kprobes/core.c
+++ b/arch/x86/kernel/kprobes/core.c
@@ -1141,12 +1141,12 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
}
NOKPROBE_SYMBOL(longjmp_break_handler);
-bool arch_within_kprobe_blacklist(unsigned long addr)
+void __init arch_populate_kprobe_blacklist(void)
{
- return (addr >= (unsigned long)__kprobes_text_start &&
- addr < (unsigned long)__kprobes_text_end) ||
- (addr >= (unsigned long)__entry_text_start &&
- addr < (unsigned long)__entry_text_end);
+ insert_kprobe_blacklist((unsigned long)__kprobes_text_start,
+ (unsigned long)__kprobes_text_end);
+ insert_kprobe_blacklist((unsigned long)__entry_text_start,
+ (unsigned long)__entry_text_end);
}
int __init arch_init_kprobes(void)
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 541df0b5b815..6d931ea0604e 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -271,6 +271,8 @@ extern bool arch_function_offset_within_entry(unsigned long offset);
extern bool function_offset_within_entry(kprobe_opcode_t *addr, const char *sym, unsigned long offset);
extern bool within_kprobe_blacklist(unsigned long addr);
+extern void insert_kprobe_blacklist(unsigned long start, unsigned long end);
+extern void arch_populate_kprobe_blacklist(void);
struct kprobe_insn_cache {
struct mutex mutex;
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index adfe3b4cfe05..8ddbee01ce01 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1364,12 +1364,7 @@ static int register_aggr_kprobe(struct kprobe *orig_p, struct kprobe *p)
return ret;
}
-bool __weak arch_within_kprobe_blacklist(unsigned long addr)
-{
- /* The __kprobes marked functions and entry code must not be probed */
- return addr >= (unsigned long)__kprobes_text_start &&
- addr < (unsigned long)__kprobes_text_end;
-}
+bool __weak arch_within_kprobe_blacklist(unsigned long addr) { return false; }
bool within_kprobe_blacklist(unsigned long addr)
{
@@ -2125,12 +2120,40 @@ NOKPROBE_SYMBOL(dump_kprobe);
* since a kprobe need not necessarily be at the beginning
* of a function.
*/
-static int __init populate_kprobe_blacklist(unsigned long *start,
- unsigned long *end)
+
+/* Markers of _kprobe_blacklist section */
+extern unsigned long __start_kprobe_blacklist[];
+extern unsigned long __stop_kprobe_blacklist[];
+
+void __init insert_kprobe_blacklist(unsigned long start, unsigned long end)
{
- unsigned long *iter;
struct kprobe_blacklist_entry *ent;
+
+ ent = kmalloc(sizeof(*ent), GFP_KERNEL);
+ if (!ent) {
+ pr_err_once("kprobes: failed to populate blacklist: %d\n", -ENOMEM);
+ pr_err_once("Please take care of using kprobes.\n");
+ }
+
+ ent->start_addr = start;
+ ent->end_addr = end;
+ INIT_LIST_HEAD(&ent->list);
+ list_add_tail(&ent->list, &kprobe_blacklist);
+}
+
+void __weak __init arch_populate_kprobe_blacklist(void)
+{
+ /* The __kprobes marked functions and entry code must not be probed */
+ insert_kprobe_blacklist((unsigned long)__kprobes_text_start,
+ (unsigned long)__kprobes_text_end);
+}
+
+static void __init populate_kprobe_blacklist(void)
+{
+ unsigned long *iter;
unsigned long entry, offset = 0, size = 0;
+ unsigned long *start = __start_kprobe_blacklist;
+ unsigned long *end = __stop_kprobe_blacklist;
for (iter = start; iter < end; iter++) {
entry = arch_deref_entry_point((void *)*iter);
@@ -2142,15 +2165,11 @@ static int __init populate_kprobe_blacklist(unsigned long *start,
continue;
}
- ent = kmalloc(sizeof(*ent), GFP_KERNEL);
- if (!ent)
- return -ENOMEM;
- ent->start_addr = entry;
- ent->end_addr = entry + size;
- INIT_LIST_HEAD(&ent->list);
- list_add_tail(&ent->list, &kprobe_blacklist);
+ insert_kprobe_blacklist(entry, entry + size);
}
- return 0;
+
+ /* Let architectures add their own entries as well */
+ arch_populate_kprobe_blacklist();
}
/* Module notifier call back, checking kprobes on the module */
@@ -2202,10 +2221,6 @@ static struct notifier_block kprobe_module_nb = {
.priority = 0
};
-/* Markers of _kprobe_blacklist section */
-extern unsigned long __start_kprobe_blacklist[];
-extern unsigned long __stop_kprobe_blacklist[];
-
static int __init init_kprobes(void)
{
int i, err = 0;
@@ -2218,12 +2233,7 @@ static int __init init_kprobes(void)
raw_spin_lock_init(&(kretprobe_table_locks[i].lock));
}
- err = populate_kprobe_blacklist(__start_kprobe_blacklist,
- __stop_kprobe_blacklist);
- if (err) {
- pr_err("kprobes: failed to populate blacklist: %d\n", err);
- pr_err("Please take care of using kprobes.\n");
- }
+ populate_kprobe_blacklist();
if (kretprobe_blacklist_size) {
/* lookup the function address from its name */
--
2.13.1
Powered by blists - more mailing lists