From b617cfc858161140d69cc0b5cc211996b557a1c7 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 29 Apr 2018 15:20:11 +0200 Subject: prctl: Add speculation control prctls Add two new prctls to control aspects of speculation related vulnerabilites and their mitigations to provide finer grained control over performance impacting mitigations. PR_GET_SPECULATION_CTRL returns the state of the speculation misfeature which is selected with arg2 of prctl(2). The return value uses bit 0-2 with the following meaning: Bit Define Description 0 PR_SPEC_PRCTL Mitigation can be controlled per task by PR_SET_SPECULATION_CTRL 1 PR_SPEC_ENABLE The speculation feature is enabled, mitigation is disabled 2 PR_SPEC_DISABLE The speculation feature is disabled, mitigation is enabled If all bits are 0 the CPU is not affected by the speculation misfeature. If PR_SPEC_PRCTL is set, then the per task control of the mitigation is available. If not set, prctl(PR_SET_SPECULATION_CTRL) for the speculation misfeature will fail. PR_SET_SPECULATION_CTRL allows to control the speculation misfeature, which is selected by arg2 of prctl(2) per task. arg3 is used to hand in the control value, i.e. either PR_SPEC_ENABLE or PR_SPEC_DISABLE. The common return values are: EINVAL prctl is not implemented by the architecture or the unused prctl() arguments are not 0 ENODEV arg2 is selecting a not supported speculation misfeature PR_SET_SPECULATION_CTRL has these additional return values: ERANGE arg3 is incorrect, i.e. it's not either PR_SPEC_ENABLE or PR_SPEC_DISABLE ENXIO prctl control of the selected speculation misfeature is disabled The first supported controlable speculation misfeature is PR_SPEC_STORE_BYPASS. Add the define so this can be shared between architectures. Based on an initial patch from Tim Chen and mostly rewritten. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar Reviewed-by: Konrad Rzeszutek Wilk --- kernel/sys.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'kernel/sys.c') diff --git a/kernel/sys.c b/kernel/sys.c index ad692183dfe9..b76dee23bdc9 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -61,6 +61,8 @@ #include #include +#include + #include /* Move somewhere else to avoid recompiling? */ #include @@ -2242,6 +2244,16 @@ static int propagate_has_child_subreaper(struct task_struct *p, void *data) return 1; } +int __weak arch_prctl_spec_ctrl_get(unsigned long which) +{ + return -EINVAL; +} + +int __weak arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl) +{ + return -EINVAL; +} + SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, unsigned long, arg4, unsigned long, arg5) { @@ -2450,6 +2462,16 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_SVE_GET_VL: error = SVE_GET_VL(); break; + case PR_GET_SPECULATION_CTRL: + if (arg3 || arg4 || arg5) + return -EINVAL; + error = arch_prctl_spec_ctrl_get(arg2); + break; + case PR_SET_SPECULATION_CTRL: + if (arg4 || arg5) + return -EINVAL; + error = arch_prctl_spec_ctrl_set(arg2, arg3); + break; default: error = -EINVAL; break; -- cgit v1.3-14-g43fede From 7bbf1373e228840bb0295a2ca26d548ef37f448e Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 1 May 2018 15:19:04 -0700 Subject: nospec: Allow getting/setting on non-current task Adjust arch_prctl_get/set_spec_ctrl() to operate on tasks other than current. This is needed both for /proc/$pid/status queries and for seccomp (since thread-syncing can trigger seccomp in non-current threads). Signed-off-by: Kees Cook Signed-off-by: Thomas Gleixner --- arch/x86/kernel/cpu/bugs.c | 27 ++++++++++++++++----------- include/linux/nospec.h | 7 +++++-- kernel/sys.c | 9 +++++---- 3 files changed, 26 insertions(+), 17 deletions(-) (limited to 'kernel/sys.c') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index fc9187b6fae7..e3afb610f2ad 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -530,31 +530,35 @@ static void ssb_select_mitigation() #undef pr_fmt -static int ssb_prctl_set(unsigned long ctrl) +static int ssb_prctl_set(struct task_struct *task, unsigned long ctrl) { - bool rds = !!test_tsk_thread_flag(current, TIF_RDS); + bool rds = !!test_tsk_thread_flag(task, TIF_RDS); if (ssb_mode != SPEC_STORE_BYPASS_PRCTL) return -ENXIO; if (ctrl == PR_SPEC_ENABLE) - clear_tsk_thread_flag(current, TIF_RDS); + clear_tsk_thread_flag(task, TIF_RDS); else - set_tsk_thread_flag(current, TIF_RDS); + set_tsk_thread_flag(task, TIF_RDS); - if (rds != !!test_tsk_thread_flag(current, TIF_RDS)) + /* + * If being set on non-current task, delay setting the CPU + * mitigation until it is next scheduled. + */ + if (task == current && rds != !!test_tsk_thread_flag(task, TIF_RDS)) speculative_store_bypass_update(); return 0; } -static int ssb_prctl_get(void) +static int ssb_prctl_get(struct task_struct *task) { switch (ssb_mode) { case SPEC_STORE_BYPASS_DISABLE: return PR_SPEC_DISABLE; case SPEC_STORE_BYPASS_PRCTL: - if (test_tsk_thread_flag(current, TIF_RDS)) + if (test_tsk_thread_flag(task, TIF_RDS)) return PR_SPEC_PRCTL | PR_SPEC_DISABLE; return PR_SPEC_PRCTL | PR_SPEC_ENABLE; default: @@ -564,24 +568,25 @@ static int ssb_prctl_get(void) } } -int arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl) +int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, + unsigned long ctrl) { if (ctrl != PR_SPEC_ENABLE && ctrl != PR_SPEC_DISABLE) return -ERANGE; switch (which) { case PR_SPEC_STORE_BYPASS: - return ssb_prctl_set(ctrl); + return ssb_prctl_set(task, ctrl); default: return -ENODEV; } } -int arch_prctl_spec_ctrl_get(unsigned long which) +int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which) { switch (which) { case PR_SPEC_STORE_BYPASS: - return ssb_prctl_get(); + return ssb_prctl_get(task); default: return -ENODEV; } diff --git a/include/linux/nospec.h b/include/linux/nospec.h index 700bb8a4e4ea..a908c954484d 100644 --- a/include/linux/nospec.h +++ b/include/linux/nospec.h @@ -7,6 +7,8 @@ #define _LINUX_NOSPEC_H #include +struct task_struct; + /** * array_index_mask_nospec() - generate a ~0 mask when index < size, 0 otherwise * @index: array element index @@ -57,7 +59,8 @@ static inline unsigned long array_index_mask_nospec(unsigned long index, }) /* Speculation control prctl */ -int arch_prctl_spec_ctrl_get(unsigned long which); -int arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl); +int arch_prctl_spec_ctrl_get(struct task_struct *task, unsigned long which); +int arch_prctl_spec_ctrl_set(struct task_struct *task, unsigned long which, + unsigned long ctrl); #endif /* _LINUX_NOSPEC_H */ diff --git a/kernel/sys.c b/kernel/sys.c index b76dee23bdc9..b0eee418ee0d 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2244,12 +2244,13 @@ static int propagate_has_child_subreaper(struct task_struct *p, void *data) return 1; } -int __weak arch_prctl_spec_ctrl_get(unsigned long which) +int __weak arch_prctl_spec_ctrl_get(struct task_struct *t, unsigned long which) { return -EINVAL; } -int __weak arch_prctl_spec_ctrl_set(unsigned long which, unsigned long ctrl) +int __weak arch_prctl_spec_ctrl_set(struct task_struct *t, unsigned long which, + unsigned long ctrl) { return -EINVAL; } @@ -2465,12 +2466,12 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_GET_SPECULATION_CTRL: if (arg3 || arg4 || arg5) return -EINVAL; - error = arch_prctl_spec_ctrl_get(arg2); + error = arch_prctl_spec_ctrl_get(me, arg2); break; case PR_SET_SPECULATION_CTRL: if (arg4 || arg5) return -EINVAL; - error = arch_prctl_spec_ctrl_set(arg2, arg3); + error = arch_prctl_spec_ctrl_set(me, arg2, arg3); break; default: error = -EINVAL; -- cgit v1.3-14-g43fede From 23d6aef74da86a33fa6bb75f79565e0a16ee97c2 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Fri, 25 May 2018 14:47:57 -0700 Subject: kernel/sys.c: fix potential Spectre v1 issue `resource' can be controlled by user-space, hence leading to a potential exploitation of the Spectre variant 1 vulnerability. This issue was detected with the help of Smatch: kernel/sys.c:1474 __do_compat_sys_old_getrlimit() warn: potential spectre issue 'get_current()->signal->rlim' (local cap) kernel/sys.c:1455 __do_sys_old_getrlimit() warn: potential spectre issue 'get_current()->signal->rlim' (local cap) Fix this by sanitizing *resource* before using it to index current->signal->rlim Notice that given that speculation windows are large, the policy is to kill the speculation on the first load and not worry if it can be completed with a dependent load/store [1]. [1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2 Link: http://lkml.kernel.org/r/20180515030038.GA11822@embeddedor.com Signed-off-by: Gustavo A. R. Silva Reviewed-by: Andrew Morton Cc: Alexei Starovoitov Cc: Dan Williams Cc: Thomas Gleixner Cc: Peter Zijlstra Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sys.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel/sys.c') diff --git a/kernel/sys.c b/kernel/sys.c index b0eee418ee0d..d1b2b8d934bb 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -71,6 +71,9 @@ #include #include +/* Hardening for Spectre-v1 */ +#include + #include "uid16.h" #ifndef SET_UNALIGN_CTL @@ -1453,6 +1456,7 @@ SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, if (resource >= RLIM_NLIMITS) return -EINVAL; + resource = array_index_nospec(resource, RLIM_NLIMITS); task_lock(current->group_leader); x = current->signal->rlim[resource]; task_unlock(current->group_leader); @@ -1472,6 +1476,7 @@ COMPAT_SYSCALL_DEFINE2(old_getrlimit, unsigned int, resource, if (resource >= RLIM_NLIMITS) return -EINVAL; + resource = array_index_nospec(resource, RLIM_NLIMITS); task_lock(current->group_leader); r = current->signal->rlim[resource]; task_unlock(current->group_leader); -- cgit v1.3-14-g43fede