On 7/26/18 4:09 PM, Kees Cook wrote: > On Tue, Jul 24, 2018 at 3:02 PM, Jiri Kosina <jikos@xxxxxxxxxx> wrote: >> On Tue, 24 Jul 2018, Srivatsa S. Bhat wrote: >> >>> However, if you are proposing that you'd like to contribute the enhanced >>> PTI/Spectre (upstream) patches from the SLES 4.4 tree to 4.4 stable, and >>> have them merged instead of this patch series, then I would certainly >>> welcome it! >> >> I'd in principle love us to push everything back to 4.4, but there are a >> few reasons (*) why that's not happening shortly. >> >> Anyway, to point out explicitly what's really needed for those folks >> running 4.4-stable and relying on PTI providing The Real Thing(TM), it's >> either a 4.4-stable port of >> >> http://kernel.suse.com/cgit/kernel-source/plain/patches.suse/x86-entry-64-use-a-per-cpu-trampoline-stack.patch?id=3428a77b02b1ba03e45d8fc352ec350429f57fc7 >> >> or making THREADINFO_GFP imply __GFP_ZERO. > > This is true in Linus's tree now. Should be trivial to backport: > https://git.kernel.org/linus/e01e80634ecdd > Hi Jiri, Kees, Thank you for suggesting the patch! I have attached the (locally tested) 4.4 and 4.9 backports of that patch with this mail. (The mainline commit applies cleanly on 4.14). Greg, could you please consider including them in stable 4.4, 4.9 and 4.14? Thank you very much! Regards, Srivatsa VMware Photon OS
From 7e39d8ccbb0889c03ce6dc0dee0e63d78f37d0a9 Mon Sep 17 00:00:00 2001 From: Kees Cook <keescook@xxxxxxxxxxxx> Date: Fri, 20 Apr 2018 14:55:31 -0700 Subject: [PATCH] fork: unconditionally clear stack on fork commit e01e80634ecdde1dd113ac43b3adad21b47f3957 upstream. One of the classes of kernel stack content leaks[1] is exposing the contents of prior heap or stack contents when a new process stack is allocated. Normally, those stacks are not zeroed, and the old contents remain in place. In the face of stack content exposure flaws, those contents can leak to userspace. Fixing this will make the kernel no longer vulnerable to these flaws, as the stack will be wiped each time a stack is assigned to a new process. There's not a meaningful change in runtime performance; it almost looks like it provides a benefit. Performing back-to-back kernel builds before: Run times: 157.86 157.09 158.90 160.94 160.80 Mean: 159.12 Std Dev: 1.54 and after: Run times: 159.31 157.34 156.71 158.15 160.81 Mean: 158.46 Std Dev: 1.46 Instead of making this a build or runtime config, Andy Lutomirski recommended this just be enabled by default. [1] A noisy search for many kinds of stack content leaks can be seen here: https://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=linux+kernel+stack+leak I did some more with perf and cycle counts on running 100,000 execs of /bin/true. before: Cycles: 218858861551 218853036130 214727610969 227656844122 224980542841 Mean: 221015379122.60 Std Dev: 4662486552.47 after: Cycles: 213868945060 213119275204 211820169456 224426673259 225489986348 Mean: 217745009865.40 Std Dev: 5935559279.99 It continues to look like it's faster, though the deviation is rather wide, but I'm not sure what I could do that would be less noisy. I'm open to ideas! Link: http://lkml.kernel.org/r/20180221021659.GA37073@beast Signed-off-by: Kees Cook <keescook@xxxxxxxxxxxx> Acked-by: Michal Hocko <mhocko@xxxxxxxx> Reviewed-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Cc: Andy Lutomirski <luto@xxxxxxxxxx> Cc: Laura Abbott <labbott@xxxxxxxxxx> Cc: Rasmus Villemoes <rasmus.villemoes@xxxxxxxxx> Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> [ Srivatsa: Backported to 4.4.y ] Signed-off-by: Srivatsa S. Bhat <srivatsa@xxxxxxxxxxxxx> Reviewed-by: Srinidhi Rao <srinidhir@xxxxxxxxxx> --- include/linux/thread_info.h | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h index ff307b5..646891f 100644 --- a/include/linux/thread_info.h +++ b/include/linux/thread_info.h @@ -55,11 +55,7 @@ extern long do_no_restart_syscall(struct restart_block *parm); #ifdef __KERNEL__ -#ifdef CONFIG_DEBUG_STACK_USAGE -# define THREADINFO_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) -#else -# define THREADINFO_GFP (GFP_KERNEL | __GFP_NOTRACK) -#endif +#define THREADINFO_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) /* * flag set/clear/test wrappers -- 2.7.4
From 7debcc6438b4a0bdc9a7b509a751350dad883328 Mon Sep 17 00:00:00 2001 From: Kees Cook <keescook@xxxxxxxxxxxx> Date: Fri, 20 Apr 2018 14:55:31 -0700 Subject: [PATCH] fork: unconditionally clear stack on fork commit e01e80634ecdde1dd113ac43b3adad21b47f3957 upstream. One of the classes of kernel stack content leaks[1] is exposing the contents of prior heap or stack contents when a new process stack is allocated. Normally, those stacks are not zeroed, and the old contents remain in place. In the face of stack content exposure flaws, those contents can leak to userspace. Fixing this will make the kernel no longer vulnerable to these flaws, as the stack will be wiped each time a stack is assigned to a new process. There's not a meaningful change in runtime performance; it almost looks like it provides a benefit. Performing back-to-back kernel builds before: Run times: 157.86 157.09 158.90 160.94 160.80 Mean: 159.12 Std Dev: 1.54 and after: Run times: 159.31 157.34 156.71 158.15 160.81 Mean: 158.46 Std Dev: 1.46 Instead of making this a build or runtime config, Andy Lutomirski recommended this just be enabled by default. [1] A noisy search for many kinds of stack content leaks can be seen here: https://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=linux+kernel+stack+leak I did some more with perf and cycle counts on running 100,000 execs of /bin/true. before: Cycles: 218858861551 218853036130 214727610969 227656844122 224980542841 Mean: 221015379122.60 Std Dev: 4662486552.47 after: Cycles: 213868945060 213119275204 211820169456 224426673259 225489986348 Mean: 217745009865.40 Std Dev: 5935559279.99 It continues to look like it's faster, though the deviation is rather wide, but I'm not sure what I could do that would be less noisy. I'm open to ideas! Link: http://lkml.kernel.org/r/20180221021659.GA37073@beast Signed-off-by: Kees Cook <keescook@xxxxxxxxxxxx> Acked-by: Michal Hocko <mhocko@xxxxxxxx> Reviewed-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Cc: Andy Lutomirski <luto@xxxxxxxxxx> Cc: Laura Abbott <labbott@xxxxxxxxxx> Cc: Rasmus Villemoes <rasmus.villemoes@xxxxxxxxx> Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> [ Srivatsa: Backported to 4.9.y ] Signed-off-by: Srivatsa S. Bhat <srivatsa@xxxxxxxxxxxxx> Reviewed-by: Srinidhi Rao <srinidhir@xxxxxxxxxx> --- include/linux/thread_info.h | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h index 2873baf..5e64367 100644 --- a/include/linux/thread_info.h +++ b/include/linux/thread_info.h @@ -59,12 +59,7 @@ extern long do_no_restart_syscall(struct restart_block *parm); #ifdef __KERNEL__ -#ifdef CONFIG_DEBUG_STACK_USAGE -# define THREADINFO_GFP (GFP_KERNEL_ACCOUNT | __GFP_NOTRACK | \ - __GFP_ZERO) -#else -# define THREADINFO_GFP (GFP_KERNEL_ACCOUNT | __GFP_NOTRACK) -#endif +#define THREADINFO_GFP (GFP_KERNEL_ACCOUNT | __GFP_NOTRACK | __GFP_ZERO) /* * flag set/clear/test wrappers -- 2.7.4
