kcov: don't lose track of remote references during softirqs
PATCH 1/1 kcov: don't lose track of remote references during softirqs
In kcov_remote_start()/kcov_remote_stop(), we swap the previous KCOV metadata of the current task into a per-CPU variable. However, the kcov_mode_enabled(mode) check is not sufficient in the case of remote KCOV coverage: current->kcov_mode always remains KCOV_MODE_DISABLED for remote KCOV objects.
If the original task that has invoked the KCOV_REMOTE_ENABLE ioctl happens to get interrupted and kcov_remote_start() is called, it ultimately leads to kcov_remote_stop() NOT restoring the original KCOV reference. So when the task exits, all registered remote KCOV handles remain active forever.
Fix it by introducing a special kcov_mode that is assigned to the task that owns a KCOV remote object. It makes kcov_mode_enabled() return true and yet does not trigger coverage collection in __sanitizer_cov_trace_pc() and write_comp_data().
--- include/linux/kcov.h | 2 ++ kernel/kcov.c | 1 + 2 files changed, 3 insertions(+)
include/linux/kcov.h +2 -0
@@ -21,6 +21,8 @@ enum kcov_mode {KCOV_MODE_TRACE_PC =2 ,/* Collecting comparison operands mode. */ KCOV_MODE_TRACE_CMP =3 , +/* The process owns a KCOV remote reference. */ + KCOV_MODE_REMOTE = 4, };#define KCOV_IN_CTXSW (1 << 30)
kernel/kcov.c +1 -0
@@ -632,6 +632,7 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,return -EINVAL ; kcov->mode = mode; t->kcov = kcov; +WRITE_ONCE (t->kcov_mode,KCOV_MODE_REMOTE ); kcov->t = t; kcov->remote = true; kcov->remote_size = remote_arg->area_size;
Signature · 1 line
2.45.2.505.gda0bf45e8d-goog
Quoted 13 lines · Aleksandr Nogikh
In kcov_remote_start()/kcov_remote_stop(), we swap the previous KCOV metadata of the current task into a per-CPU variable. However, the
kcov_mode_enabled(mode)check is not sufficient in the case of remote KCOV coverage:current->kcov_modealways remainsKCOV_MODE_DISABLEDfor remote KCOV objects.If the original task that has invoked the
KCOV_REMOTE_ENABLEioctl happens to get interrupted andkcov_remote_start()is called, it ultimately leads tokcov_remote_stop()NOT restoring the original KCOV reference. So when the task exits, all registered remote KCOV handles remain active forever.
Quoted 7 lines
Fix it by introducing a special
kcov_modethat is assigned to the task that owns a KCOV remote object. It makeskcov_mode_enabled()return true and yet does not trigger coverage collection in__sanitizer_cov_trace_pc()andwrite_comp_data().Signed-off-by Aleksandr Nogikh <[email protected]>Fixes 5ff3b30ab57d ("kcov: collect coverage from interrupts")
Quoted 33 lines
--- include/linux/kcov.h | 2 ++ kernel/kcov.c | 1 + 2 files changed, 3 insertions(+)include/linux/kcov.h +2
-0@@ -21,6 +21,8 @@ enum kcov_mode {KCOV_MODE_TRACE_PC =2 ,/* Collecting comparison operands mode. */ KCOV_MODE_TRACE_CMP =3 , +/* The process owns a KCOV remote reference. */ + KCOV_MODE_REMOTE = 4, };#define KCOV_IN_CTXSW (1 << 30)kernel/kcov.c +1
-1@@ -632,6 +632,7 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,return -EINVAL ; kcov->mode = mode; t->kcov = kcov; +WRITE_ONCE (t->kcov_mode,KCOV_MODE_REMOTE ); kcov->t = t; kcov->remote = true; kcov->remote_size = remote_arg->area_size; --2.45.2.505.gda0bf45e8d-goog
Quoted 12 lines · On Tue, 11 Jun 2024 15:32:29 +0200 Aleksandr Nogikh <[email protected]> wrote:
In kcov_remote_start()/kcov_remote_stop(), we swap the previous KCOV metadata of the current task into a per-CPU variable. However, the
kcov_mode_enabled(mode)check is not sufficient in the case of remote KCOV coverage:current->kcov_modealways remainsKCOV_MODE_DISABLEDfor remote KCOV objects.If the original task that has invoked the
KCOV_REMOTE_ENABLEioctl happens to get interrupted andkcov_remote_start()is called, it ultimately leads tokcov_remote_stop()NOT restoring the original KCOV reference. So when the task exits, all registered remote KCOV handles remain active forever.
Quoted 4 lines
Fix it by introducing a special
kcov_modethat is assigned to the task that owns a KCOV remote object. It makeskcov_mode_enabled()return true and yet does not trigger coverage collection in__sanitizer_cov_trace_pc()andwrite_comp_data().
What are the userspace visible effects of this bug? I *think* it's just an efficiency thing, but how significant? In other words, should we backport this fix?
Quoted 15 lines · Andrew Morton
Quoted 12 lines · On Tue, 11 Jun 2024 15:32:29 +0200 Aleksandr Nogikh <[email protected]> wrote:
In kcov_remote_start()/kcov_remote_stop(), we swap the previous KCOV metadata of the current task into a per-CPU variable. However, the
kcov_mode_enabled(mode)check is not sufficient in the case of remote KCOV coverage:current->kcov_modealways remainsKCOV_MODE_DISABLEDfor remote KCOV objects.If the original task that has invoked the
KCOV_REMOTE_ENABLEioctl happens to get interrupted andkcov_remote_start()is called, it ultimately leads tokcov_remote_stop()NOT restoring the original KCOV reference. So when the task exits, all registered remote KCOV handles remain active forever.
Quoted 9 lines
Quoted 4 lines
Fix it by introducing a special
kcov_modethat is assigned to the task that owns a KCOV remote object. It makeskcov_mode_enabled()return true and yet does not trigger coverage collection in__sanitizer_cov_trace_pc()andwrite_comp_data().What are the userspace visible effects of this bug? I *think* it's just an efficiency thing, but how significant? In other words, should we backport this fix?
The most uncomfortable effect (at least for syzkaller) is that the bug prevents the reuse of the same /sys/kernel/debug/kcov descriptor. If we obtain it in the parent process and then e.g. drop some capabilities and continuously fork to execute individual programs, at some point current->kcov of the forked process is lost, kcov_task_exit() takes no action, and all KCOV_REMOTE_ENABLE ioctls calls from subsequent forks fail.
And, yes, the efficiency is also affected if we keep on losing remote kcov objects.
a) kcov_remote_map keeps on growing forever.
b) (If I'm not mistaken), we're also not freeing the memory referenced by kcov->area.
I think it would be nice to backport the fix to the stable trees.
Signature · 1 line
Aleksandr
Quoted 51 lines · Aleksandr Nogikh
In kcov_remote_start()/kcov_remote_stop(), we swap the previous KCOV metadata of the current task into a per-CPU variable. However, the
kcov_mode_enabled(mode)check is not sufficient in the case of remote KCOV coverage:current->kcov_modealways remainsKCOV_MODE_DISABLEDfor remote KCOV objects.If the original task that has invoked the
KCOV_REMOTE_ENABLEioctl happens to get interrupted andkcov_remote_start()is called, it ultimately leads tokcov_remote_stop()NOT restoring the original KCOV reference. So when the task exits, all registered remote KCOV handles remain active forever.Fix it by introducing a special
kcov_modethat is assigned to the task that owns a KCOV remote object. It makeskcov_mode_enabled()return true and yet does not trigger coverage collection in__sanitizer_cov_trace_pc()andwrite_comp_data().Signed-off-by Aleksandr Nogikh <[email protected]>Fixes 5ff3b30ab57d ("kcov: collect coverage from interrupts")--- include/linux/kcov.h | 2 ++ kernel/kcov.c | 1 + 2 files changed, 3 insertions(+)include/linux/kcov.h +2
-0@@ -21,6 +21,8 @@ enum kcov_mode {KCOV_MODE_TRACE_PC =2 ,/* Collecting comparison operands mode. */ KCOV_MODE_TRACE_CMP =3 , +/* The process owns a KCOV remote reference. */ + KCOV_MODE_REMOTE = 4, };#define KCOV_IN_CTXSW (1 << 30)kernel/kcov.c +1
-1@@ -632,6 +632,7 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,return -EINVAL ; kcov->mode = mode; t->kcov = kcov; +WRITE_ONCE (t->kcov_mode,KCOV_MODE_REMOTE ); kcov->t = t; kcov->remote = true; kcov->remote_size = remote_arg->area_size; --
Quoted 2 lines
2.45.2.505.gda0bf45e8d-goog
Thank you for fixing this!
Andrey Konovalov
reviewerQuoted 43 lines · Aleksandr Nogikh
In kcov_remote_start()/kcov_remote_stop(), we swap the previous KCOV metadata of the current task into a per-CPU variable. However, the
kcov_mode_enabled(mode)check is not sufficient in the case of remote KCOV coverage:current->kcov_modealways remainsKCOV_MODE_DISABLEDfor remote KCOV objects.If the original task that has invoked the
KCOV_REMOTE_ENABLEioctl happens to get interrupted andkcov_remote_start()is called, it ultimately leads tokcov_remote_stop()NOT restoring the original KCOV reference. So when the task exits, all registered remote KCOV handles remain active forever.Fix it by introducing a special
kcov_modethat is assigned to the task that owns a KCOV remote object. It makeskcov_mode_enabled()return true and yet does not trigger coverage collection in__sanitizer_cov_trace_pc()andwrite_comp_data().Signed-off-by Aleksandr Nogikh <[email protected]>Fixes 5ff3b30ab57d ("kcov: collect coverage from interrupts")--- include/linux/kcov.h | 2 ++ kernel/kcov.c | 1 + 2 files changed, 3 insertions(+)include/linux/kcov.h +2
-0@@ -21,6 +21,8 @@ enum kcov_mode {KCOV_MODE_TRACE_PC =2 ,/* Collecting comparison operands mode. */ KCOV_MODE_TRACE_CMP =3 , +/* The process owns a KCOV remote reference. */ + KCOV_MODE_REMOTE = 4, };#define KCOV_IN_CTXSW (1 << 30)kernel/kcov.c +0
-0
kernel/kcov.c static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd, · line 635
Quoted 4 lines
kernel/kcov.c +1
-0@@ -632,6 +632,7 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,return -EINVAL ; kcov->mode = mode; t->kcov = kcov; +WRITE_ONCE (t->kcov_mode,KCOV_MODE_REMOTE );
Looking at this again, I don't think we need this WRITE_ONCE here, as we have interrupts disabled. But if we do, perhaps it makes sense to add a comment explaining why.
Quoted 6 lines
kernel/kcov.c +0
-1kcov->t = t; kcov->remote = true; kcov->remote_size = remote_arg->area_size; --2.45.2.505.gda0bf45e8d-goog
Quoted 39 lines · Andrey Konovalov
Quoted 37 lines · Aleksandr Nogikh
In kcov_remote_start()/kcov_remote_stop(), we swap the previous KCOV metadata of the current task into a per-CPU variable. However, the
kcov_mode_enabled(mode)check is not sufficient in the case of remote KCOV coverage:current->kcov_modealways remainsKCOV_MODE_DISABLEDfor remote KCOV objects.If the original task that has invoked the
KCOV_REMOTE_ENABLEioctl happens to get interrupted andkcov_remote_start()is called, it ultimately leads tokcov_remote_stop()NOT restoring the original KCOV reference. So when the task exits, all registered remote KCOV handles remain active forever.Fix it by introducing a special
kcov_modethat is assigned to the task that owns a KCOV remote object. It makeskcov_mode_enabled()return true and yet does not trigger coverage collection in__sanitizer_cov_trace_pc()andwrite_comp_data().Signed-off-by Aleksandr Nogikh <[email protected]>Fixes 5ff3b30ab57d ("kcov: collect coverage from interrupts")--- include/linux/kcov.h | 2 ++ kernel/kcov.c | 1 + 2 files changed, 3 insertions(+)include/linux/kcov.h +2
-0@@ -21,6 +21,8 @@ enum kcov_mode {KCOV_MODE_TRACE_PC =2 ,/* Collecting comparison operands mode. */ KCOV_MODE_TRACE_CMP =3 , +/* The process owns a KCOV remote reference. */ + KCOV_MODE_REMOTE = 4, };
Quoted 14 lines
kernel/kcov.c static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd, · line 635
Quoted 10 lines
include/linux/kcov.h +0
-0#define KCOV_IN_CTXSW (1 << 30)kernel/kcov.c +1
-0@@ -632,6 +632,7 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,return -EINVAL ; kcov->mode = mode; t->kcov = kcov; +WRITE_ONCE (t->kcov_mode,KCOV_MODE_REMOTE );Looking at this again, I don't think we need this
WRITE_ONCEhere, as we have interrupts disabled. But if we do, perhaps it makes sense to add a comment explaining why.
Thank you!
I've sent a v2:
https://lore.kernel.org/all/[email protected]/
Quoted 8 lines
Quoted 7 lines
kernel/kcov.c +0
-1kcov->t = t; kcov->remote = true; kcov->remote_size = remote_arg->area_size; --2.45.2.505.gda0bf45e8d-goog