linux-stable.git/arch, branch v3.10.65

perf/x86/intel/uncore: Make sure only uncore events are collected

2015-01-16T14:59:03+00:00

commit af91568e762d04931dcbdd6bef4655433d8b9418 upstream.

The uncore_collect_events functions assumes that event group
might contain only uncore events which is wrong, because it
might contain any type of events.

This bug leads to uncore framework touching 'not' uncore events,
which could end up all sorts of bugs.

One was triggered by Vince's perf fuzzer, when the uncore code
touched breakpoint event private event space as if it was uncore
event and caused BUG:

   BUG: unable to handle kernel paging request at ffffffff82822068
   IP: [] uncore_assign_events+0x188/0x250
   ...

The code in uncore_assign_events() function was looking for
event->hw.idx data while the event was initialized as a
breakpoint with different members in event->hw union.

This patch forces uncore_collect_events() to collect only uncore
events.

Reported-by: Vince Weaver 
Signed-off-by: Jiri Olsa 
Cc: Arnaldo Carvalho de Melo 
Cc: Frederic Weisbecker 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Stephane Eranian 
Cc: Yan, Zheng 
Link: http://lkml.kernel.org/r/1418243031-20367-2-git-send-email-jolsa@kernel.org
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

ARM: mvebu: disable I/O coherency on non-SMP situations on Armada 370/375/38x/XP

2015-01-16T14:59:02+00:00

commit e55355453600a33bb5ca4f71f2d7214875f3b061 upstream.

Enabling the hardware I/O coherency on Armada 370, Armada 375, Armada
38x and Armada XP requires a certain number of conditions:

 - On Armada 370, the cache policy must be set to write-allocate.

 - On Armada 375, 38x and XP, the cache policy must be set to
   write-allocate, the pages must be mapped with the shareable
   attribute, and the SMP bit must be set

Currently, on Armada XP, when CONFIG_SMP is enabled, those conditions
are met. However, when Armada XP is used in a !CONFIG_SMP kernel, none
of these conditions are met. With Armada 370, the situation is worse:
since the processor is single core, regardless of whether CONFIG_SMP
or !CONFIG_SMP is used, the cache policy will be set to write-back by
the kernel and not write-allocate.

Since solving this problem turns out to be quite complicated, and we
don't want to let users with a mainline kernel known to have
infrequent but existing data corruptions, this commit proposes to
simply disable hardware I/O coherency in situations where it is known
not to work.

And basically, the is_smp() function of the kernel tells us whether it
is OK to enable hardware I/O coherency or not, so this commit slightly
refactors the coherency_type() function to return
COHERENCY_FABRIC_TYPE_NONE when is_smp() is false, or the appropriate
type of the coherency fabric in the other case.

Thanks to this, the I/O coherency fabric will no longer be used at all
in !CONFIG_SMP configurations. It will continue to be used in
CONFIG_SMP configurations on Armada XP, Armada 375 and Armada 38x
(which are multiple cores processors), but will no longer be used on
Armada 370 (which is a single core processor).

In the process, it simplifies the implementation of the
coherency_type() function, and adds a missing call to of_node_put().

Signed-off-by: Thomas Petazzoni 
Fixes: e60304f8cb7bb545e79fe62d9b9762460c254ec2 ("arm: mvebu: Add hardware I/O Coherency support")
Cc:  # v3.8+
Acked-by: Gregory CLEMENT 
Link: https://lkml.kernel.org/r/1415871540-20302-3-git-send-email-thomas.petazzoni@free-electrons.com
Signed-off-by: Jason Cooper 
Signed-off-by: Greg Kroah-Hartman

x86, vdso: Use asm volatile in __getcpu

2015-01-16T14:59:01+00:00

commit 1ddf0b1b11aa8a90cef6706e935fc31c75c406ba upstream.

In Linux 3.18 and below, GCC hoists the lsl instructions in the
pvclock code all the way to the beginning of __vdso_clock_gettime,
slowing the non-paravirt case significantly.  For unknown reasons,
presumably related to the removal of a branch, the performance issue
is gone as of

e76b027e6408 x86,vdso: Use LSL unconditionally for vgetcpu

but I don't trust GCC enough to expect the problem to stay fixed.

There should be no correctness issue, because the __getcpu calls in
__vdso_vlock_gettime were never necessary in the first place.

Note to stable maintainers: In 3.18 and below, depending on
configuration, gcc 4.9.2 generates code like this:

     9c3:       44 0f 03 e8             lsl    %ax,%r13d
     9c7:       45 89 eb                mov    %r13d,%r11d
     9ca:       0f 03 d8                lsl    %ax,%ebx

This patch won't apply as is to any released kernel, but I'll send a
trivial backported version if needed.

[
 Backported by Andy Lutomirski.  Should apply to all affected
 versions.  This fixes a functionality bug as well as a performance
 bug: buggy kernels can infinite loop in __vdso_clock_gettime on
 affected compilers.  See, for exammple:

 https://bugzilla.redhat.com/show_bug.cgi?id=1178975
]

Fixes: 51c19b4f5927 x86: vdso: pvclock gettime support
Cc: Marcelo Tosatti 
Acked-by: Paolo Bonzini 
Signed-off-by: Andy Lutomirski 
Signed-off-by: Greg Kroah-Hartman

x86_64, vdso: Fix the vdso address randomization algorithm

2015-01-16T14:59:01+00:00

commit 394f56fe480140877304d342dec46d50dc823d46 upstream.

The theory behind vdso randomization is that it's mapped at a random
offset above the top of the stack.  To avoid wasting a page of
memory for an extra page table, the vdso isn't supposed to extend
past the lowest PMD into which it can fit.  Other than that, the
address should be a uniformly distributed address that meets all of
the alignment requirements.

The current algorithm is buggy: the vdso has about a 50% probability
of being at the very end of a PMD.  The current algorithm also has a
decent chance of failing outright due to incorrect handling of the
case where the top of the stack is near the top of its PMD.

This fixes the implementation.  The paxtest estimate of vdso
"randomisation" improves from 11 bits to 18 bits.  (Disclaimer: I
don't know what the paxtest code is actually calculating.)

It's worth noting that this algorithm is inherently biased: the vdso
is more likely to end up near the end of its PMD than near the
beginning.  Ideally we would either nix the PMD sharing requirement
or jointly randomize the vdso and the stack to reduce the bias.

In the mean time, this is a considerable improvement with basically
no risk of compatibility issues, since the allowed outputs of the
algorithm are unchanged.

As an easy test, doing this:

for i in `seq 10000`
  do grep -P vdso /proc/self/maps |cut -d- -f1
done |sort |uniq -d

used to produce lots of output (1445 lines on my most recent run).
A tiny subset looks like this:

7fffdfffe000
7fffe01fe000
7fffe05fe000
7fffe07fe000
7fffe09fe000
7fffe0bfe000
7fffe0dfe000

Note the suspicious fe000 endings.  With the fix, I get a much more
palatable 76 repeated addresses.

Reviewed-by: Kees Cook 
Signed-off-by: Andy Lutomirski 
Signed-off-by: Greg Kroah-Hartman

groups: Consolidate the setgroups permission checks

2015-01-08T17:58:16+00:00

commit 7ff4d90b4c24a03666f296c3d4878cd39001e81e upstream.

Today there are 3 instances of setgroups and due to an oversight their
permission checking has diverged.  Add a common function so that
they may all share the same permission checking code.

This corrects the current oversight in the current permission checks
and adds a helper to avoid this in the future.

A user namespace security fix will update this new helper, shortly.

Signed-off-by: "Eric W. Biederman" 
Signed-off-by: Greg Kroah-Hartman

x86/tls: Don't validate lm in set_thread_area() after all

2015-01-08T17:58:15+00:00

commit 3fb2f4237bb452eb4e98f6a5dbd5a445b4fed9d0 upstream.

It turns out that there's a lurking ABI issue.  GCC, when
compiling this in a 32-bit program:

struct user_desc desc = {
	.entry_number    = idx,
	.base_addr       = base,
	.limit           = 0xfffff,
	.seg_32bit       = 1,
	.contents        = 0, /* Data, grow-up */
	.read_exec_only  = 0,
	.limit_in_pages  = 1,
	.seg_not_present = 0,
	.useable         = 0,
};

will leave .lm uninitialized.  This means that anything in the
kernel that reads user_desc.lm for 32-bit tasks is unreliable.

Revert the .lm check in set_thread_area().  The value never did
anything in the first place.

Fixes: 0e58af4e1d21 ("x86/tls: Disallow unusual TLS segments")
Signed-off-by: Andy Lutomirski 
Acked-by: Thomas Gleixner 
Cc: Linus Torvalds 
Link: http://lkml.kernel.org/r/d7875b60e28c512f6a6fc0baf5714d58e7eaadbb.1418856405.git.luto@amacapital.net
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

x86, kvm: Clear paravirt_enabled on KVM guests for espfix32's benefit

2015-01-08T17:58:15+00:00

commit 29fa6825463c97e5157284db80107d1bfac5d77b upstream.

paravirt_enabled has the following effects:

 - Disables the F00F bug workaround warning.  There is no F00F bug
   workaround any more because Linux's standard IDT handling already
   works around the F00F bug, but the warning still exists.  This
   is only cosmetic, and, in any event, there is no such thing as
   KVM on a CPU with the F00F bug.

 - Disables 32-bit APM BIOS detection.  On a KVM paravirt system,
   there should be no APM BIOS anyway.

 - Disables tboot.  I think that the tboot code should check the
   CPUID hypervisor bit directly if it matters.

 - paravirt_enabled disables espfix32.  espfix32 should *not* be
   disabled under KVM paravirt.

The last point is the purpose of this patch.  It fixes a leak of the
high 16 bits of the kernel stack address on 32-bit KVM paravirt
guests.  Fixes CVE-2014-8134.

Suggested-by: Konrad Rzeszutek Wilk 
Signed-off-by: Andy Lutomirski 
Signed-off-by: Paolo Bonzini 
Signed-off-by: Greg Kroah-Hartman

x86_64, switch_to(): Load TLS descriptors before switching DS and ES

2015-01-08T17:58:15+00:00

commit f647d7c155f069c1a068030255c300663516420e upstream.

Otherwise, if buggy user code points DS or ES into the TLS
array, they would be corrupted after a context switch.

This also significantly improves the comments and documents some
gotchas in the code.

Before this patch, the both tests below failed.  With this
patch, the es test passes, although the gsbase test still fails.

 ----- begin es test -----

/*
 * Copyright (c) 2014 Andy Lutomirski
 * GPL v2
 */

static unsigned short GDT3(int idx)
{
	return (idx << 3) | 3;
}

static int create_tls(int idx, unsigned int base)
{
	struct user_desc desc = {
		.entry_number    = idx,
		.base_addr       = base,
		.limit           = 0xfffff,
		.seg_32bit       = 1,
		.contents        = 0, /* Data, grow-up */
		.read_exec_only  = 0,
		.limit_in_pages  = 1,
		.seg_not_present = 0,
		.useable         = 0,
	};

	if (syscall(SYS_set_thread_area, &desc) != 0)
		err(1, "set_thread_area");

	return desc.entry_number;
}

int main()
{
	int idx = create_tls(-1, 0);
	printf("Allocated GDT index %d\n", idx);

	unsigned short orig_es;
	asm volatile ("mov %%es,%0" : "=rm" (orig_es));

	int errors = 0;
	int total = 1000;
	for (int i = 0; i < total; i++) {
		asm volatile ("mov %0,%%es" : : "rm" (GDT3(idx)));
		usleep(100);

		unsigned short es;
		asm volatile ("mov %%es,%0" : "=rm" (es));
		asm volatile ("mov %0,%%es" : : "rm" (orig_es));
		if (es != GDT3(idx)) {
			if (errors == 0)
				printf("[FAIL]\tES changed from 0x%hx to 0x%hx\n",
				       GDT3(idx), es);
			errors++;
		}
	}

	if (errors) {
		printf("[FAIL]\tES was corrupted %d/%d times\n", errors, total);
		return 1;
	} else {
		printf("[OK]\tES was preserved\n");
		return 0;
	}
}

 ----- end es test -----

 ----- begin gsbase test -----

/*
 * gsbase.c, a gsbase test
 * Copyright (c) 2014 Andy Lutomirski
 * GPL v2
 */

static unsigned char *testptr, *testptr2;

static unsigned char read_gs_testvals(void)
{
	unsigned char ret;
	asm volatile ("movb %%gs:%1, %0" : "=r" (ret) : "m" (*testptr));
	return ret;
}

int main()
{
	int errors = 0;

	testptr = mmap((void *)0x200000000UL, 1, PROT_READ | PROT_WRITE,
		       MAP_PRIVATE | MAP_FIXED | MAP_ANONYMOUS, -1, 0);
	if (testptr == MAP_FAILED)
		err(1, "mmap");

	testptr2 = mmap((void *)0x300000000UL, 1, PROT_READ | PROT_WRITE,
		       MAP_PRIVATE | MAP_FIXED | MAP_ANONYMOUS, -1, 0);
	if (testptr2 == MAP_FAILED)
		err(1, "mmap");

	*testptr = 0;
	*testptr2 = 1;

	if (syscall(SYS_arch_prctl, ARCH_SET_GS,
		    (unsigned long)testptr2 - (unsigned long)testptr) != 0)
		err(1, "ARCH_SET_GS");

	usleep(100);

	if (read_gs_testvals() == 1) {
		printf("[OK]\tARCH_SET_GS worked\n");
	} else {
		printf("[FAIL]\tARCH_SET_GS failed\n");
		errors++;
	}

	asm volatile ("mov %0,%%gs" : : "r" (0));

	if (read_gs_testvals() == 0) {
		printf("[OK]\tWriting 0 to gs worked\n");
	} else {
		printf("[FAIL]\tWriting 0 to gs failed\n");
		errors++;
	}

	usleep(100);

	if (read_gs_testvals() == 0) {
		printf("[OK]\tgsbase is still zero\n");
	} else {
		printf("[FAIL]\tgsbase was corrupted\n");
		errors++;
	}

	return errors == 0 ? 0 : 1;
}

 ----- end gsbase test -----

Signed-off-by: Andy Lutomirski 
Cc: Andi Kleen 
Cc: Linus Torvalds 
Link: http://lkml.kernel.org/r/509d27c9fec78217691c3dad91cec87e1006b34a.1418075657.git.luto@amacapital.net
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

x86/tls: Disallow unusual TLS segments

2015-01-08T17:58:15+00:00

commit 0e58af4e1d2166e9e33375a0f121e4867010d4f8 upstream.

Users have no business installing custom code segments into the
GDT, and segments that are not present but are otherwise valid
are a historical source of interesting attacks.

For completeness, block attempts to set the L bit.  (Prior to
this patch, the L bit would have been silently dropped.)

This is an ABI break.  I've checked glibc, musl, and Wine, and
none of them look like they'll have any trouble.

Note to stable maintainers: this is a hardening patch that fixes
no known bugs.  Given the possibility of ABI issues, this
probably shouldn't be backported quickly.

Signed-off-by: Andy Lutomirski 
Acked-by: H. Peter Anvin 
Cc: Konrad Rzeszutek Wilk 
Cc: Linus Torvalds 
Cc: Willy Tarreau 
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

x86/tls: Validate TLS entries to protect espfix

2015-01-08T17:58:15+00:00

commit 41bdc78544b8a93a9c6814b8bbbfef966272abbe upstream.

Installing a 16-bit RW data segment into the GDT defeats espfix.
AFAICT this will not affect glibc, Wine, or dosemu at all.

Signed-off-by: Andy Lutomirski 
Acked-by: H. Peter Anvin 
Cc: Konrad Rzeszutek Wilk 
Cc: Linus Torvalds 
Cc: Willy Tarreau 
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman