Searched +hist:6 +hist:b12de69 (Results 1 - 1 of 1) sorted by relevance

H A DMakefilediff 6f5b41a2 Mon Aug 02 12:39:08 MDT 2021 Nick Desaulniers <> Makefile: move initial clang flag handling into scripts/Makefile.clang

With some of the changes we'd like to make to CROSS_COMPILE, the initial
block of clang flag handling which controls things like the target triple,
whether or not to use the integrated assembler and how to find GAS,
and erroring on unknown warnings is becoming unwieldy. Move it into its
own file under scripts/.

Reviewed-by: Nathan Chancellor <>
Signed-off-by: Nick Desaulniers <>
Signed-off-by: Masahiro Yamada <>
diff 5f6e0fe0 Wed Jun 09 20:03:31 MDT 2021 Masahiro Yamada <> parisc: Fix compile failure when building 64-bit kernel natively

Commit 23243c1ace9f ("arch: use cross_compiling to check whether it is
a cross build or not") broke 64-bit parisc builds on 32-bit parisc

Helge mentioned:
- 64-bit parisc userspace is not supported yet [1]
- hppa gcc does not support "-m64" flag [2]

That means, parisc developers working on a 32-bit parisc machine need
to use hppa64-linux-gnu-gcc (cross compiler) for building the 64-bit
parisc kernel.

After the offending commit, gcc is used in such a case because
both $(SRCARCH) and $(SUBARCH) are 'parisc', hence cross_compiling is

A correct way is to introduce ARCH=parisc64 because building the 64-bit
parisc kernel on a 32-bit parisc system is not exactly a native build,
but rather a semi-cross build.


Fixes: 23243c1ace9f ("arch: use cross_compiling to check whether it is a cross build or not")
Signed-off-by: Masahiro Yamada <>
Reported-by: Meelis Roos <>
Tested-by: Meelis Roos <>
Cc: <> # v5.13+
Signed-off-by: Helge Deller <>
diff 6efb943b Sun May 09 15:17:44 MDT 2021 Linus Torvalds <> Linux 5.13-rc1
diff 6e0839fd Sat Mar 13 23:15:50 MST 2021 Masahiro Yamada <> kbuild: replace sed with $(subst ) or $(patsubst )

For simple text replacement, it is better to use a built-in function
instead of sed if possible. You can save one process forking.

I do not mean to replace all sed invocations because GNU Make itself
does not support regular expression (unless you use guile).

I just replaced simple ones.

Signed-off-by: Masahiro Yamada <>
diff 39218ff4 Thu Apr 01 17:23:44 MDT 2021 Kees Cook <> stack: Optionally randomize kernel stack offset each syscall

This provides the ability for architectures to enable kernel stack base
address offset randomization. This feature is controlled by the boot
param "randomize_kstack_offset=on/off", with its default value set by

This feature is based on the original idea from the last public release
of PaX's RANDKSTACK feature:
All the credit for the original idea goes to the PaX team. Note that
the design and implementation of this upstream randomize_kstack_offset
feature differs greatly from the RANDKSTACK feature (see below).

Reasoning for the feature:

This feature aims to make harder the various stack-based attacks that
rely on deterministic stack structure. We have had many such attacks in
past (just to name few):

As Linux kernel stack protections have been constantly improving
(vmap-based stack allocation with guard pages, removal of thread_info,
STACKLEAK), attackers have had to find new ways for their exploits
to work. They have done so, continuing to rely on the kernel's stack
determinism, in situations where VMAP_STACK and THREAD_INFO_IN_TASK_STRUCT
were not relevant. For example, the following recent attacks would have
been hampered if the stack offset was non-deterministic between syscalls:
(page 70: targeting the pt_regs copy with linear stack overflow)
(leaked stack address from one syscall as a target during next syscall)

The main idea is that since the stack offset is randomized on each system
call, it is harder for an attack to reliably land in any particular place
on the thread stack, even with address exposures, as the stack base will
change on the next syscall. Also, since randomization is performed after
placing pt_regs, the ptrace-based approach[1] to discover the randomized
offset during a long-running syscall should not be possible.

Design description:

During most of the kernel's execution, it runs on the "thread stack",
which is pretty deterministic in its structure: it is fixed in size,
and on every entry from userspace to kernel on a syscall the thread
stack starts construction from an address fetched from the per-cpu
cpu_current_top_of_stack variable. The first element to be pushed to the
thread stack is the pt_regs struct that stores all required CPU registers
and syscall parameters. Finally the specific syscall function is called,
with the stack being used as the kernel executes the resulting request.

The goal of randomize_kstack_offset feature is to add a random offset
after the pt_regs has been pushed to the stack and before the rest of the
thread stack is used during the syscall processing, and to change it every
time a process issues a syscall. The source of randomness is currently
architecture-defined (but x86 is using the low byte of rdtsc()). Future
improvements for different entropy sources is possible, but out of scope
for this patch. Further more, to add more unpredictability, new offsets
are chosen at the end of syscalls (the timing of which should be less
easy to measure from userspace than at syscall entry time), and stored
in a per-CPU variable, so that the life of the value does not stay
explicitly tied to a single task.

As suggested by Andy Lutomirski, the offset is added using alloca()
and an empty asm() statement with an output constraint, since it avoids
changes to assembly syscall entry code, to the unwinder, and provides
correct stack alignment as defined by the compiler.

In order to make this available by default with zero performance impact
for those that don't want it, it is boot-time selectable with static
branches. This way, if the overhead is not wanted, it can just be
left turned off with no performance impact.

The generated assembly for x86_64 with GCC looks like this:

ffffffff81003977: 65 8b 05 02 ea 00 7f mov %gs:0x7f00ea02(%rip),%eax
# 12380 <kstack_offset>
ffffffff8100397e: 25 ff 03 00 00 and $0x3ff,%eax
ffffffff81003983: 48 83 c0 0f add $0xf,%rax
ffffffff81003987: 25 f8 07 00 00 and $0x7f8,%eax
ffffffff8100398c: 48 29 c4 sub %rax,%rsp
ffffffff8100398f: 48 8d 44 24 0f lea 0xf(%rsp),%rax
ffffffff81003994: 48 83 e0 f0 and $0xfffffffffffffff0,%rax

As a result of the above stack alignment, this patch introduces about
5 bits of randomness after pt_regs is spilled to the thread stack on
x86_64, and 6 bits on x86_32 (since its has 1 fewer bit required for
stack alignment). The amount of entropy could be adjusted based on how
much of the stack space we wish to trade for security.

My measure of syscall performance overhead (on x86_64):

lmbench: /usr/lib/lmbench/bin/x86_64-linux-gnu/lat_syscall -N 10000 null
randomize_kstack_offset=y Simple syscall: 0.7082 microseconds
randomize_kstack_offset=n Simple syscall: 0.7016 microseconds

So, roughly 0.9% overhead growth for a no-op syscall, which is very
manageable. And for people that don't want this, it's off by default.

There are two gotchas with using the alloca() trick. First,
compilers that have Stack Clash protection (-fstack-clash-protection)
enabled by default (e.g. Ubuntu[3]) add pagesize stack probes to
any dynamic stack allocations. While the randomization offset is
always less than a page, the resulting assembly would still contain
(unreachable!) probing routines, bloating the resulting assembly. To
avoid this, -fno-stack-clash-protection is unconditionally added to
the kernel Makefile since this is the only dynamic stack allocation in
the kernel (now that VLAs have been removed) and it is provably safe
from Stack Clash style attacks.

The second gotcha with alloca() is a negative interaction with
-fstack-protector*, in that it sees the alloca() as an array allocation,
which triggers the unconditional addition of the stack canary function
pre/post-amble which slows down syscalls regardless of the static
branch. In order to avoid adding this unneeded check and its associated
performance impact, architectures need to carefully remove uses of
-fstack-protector-strong (or -fstack-protector) in the compilation units
that use the add_random_kstack() macro and to audit the resulting stack
mitigation coverage (to make sure no desired coverage disappears). No
change is visible for this on x86 because the stack protector is already
unconditionally disabled for the compilation unit, but the change is
required on arm64. There is, unfortunately, no attribute that can be
used to disable stack protector for specific functions.

Comparison to PaX RANDKSTACK feature:

The RANDKSTACK feature randomizes the location of the stack start
(cpu_current_top_of_stack), i.e. including the location of pt_regs
structure itself on the stack. Initially this patch followed the same
approach, but during the recent discussions[2], it has been determined
to be of a little value since, if ptrace functionality is available for
an attacker, they can use PTRACE_PEEKUSR/PTRACE_POKEUSR to read/write
different offsets in the pt_regs struct, observe the cache behavior of
the pt_regs accesses, and figure out the random stack offset. Another
difference is that the random offset is stored in a per-cpu variable,
rather than having it be per-thread. As a result, these implementations
differ a fair bit in their implementation details and results, though
obviously the intent is similar.


Co-developed-by: Elena Reshetova <>
Signed-off-by: Elena Reshetova <>
Signed-off-by: Kees Cook <>
Signed-off-by: Thomas Gleixner <>
Reviewed-by: Thomas Gleixner <>
diff 98cd6f52 Fri Feb 05 01:22:20 MST 2021 Nick Desaulniers <> Kconfig: allow explicit opt in to DWARF v5

DWARF v5 is the latest standard of the DWARF debug info format. GCC 11
will change the implicit default DWARF version, if left unspecified, to

Allow users of Clang and older versions of GCC that have not changed the
implicit default DWARF version to DWARF v5 to opt in. This can help
testing consumers of DWARF debug info in preparation of v5 becoming more
widespread, as well as result in significant binary size savings of the
pre-stripped vmlinux image.

DWARF5 wins significantly in terms of size when mixed with compression

363M vmlinux.clang12.dwarf5.compressed
434M vmlinux.clang12.dwarf4.compressed
439M vmlinux.clang12.dwarf2.compressed
457M vmlinux.clang12.dwarf5
536M vmlinux.clang12.dwarf4
548M vmlinux.clang12.dwarf2

515M vmlinux.gcc10.2.dwarf5.compressed
599M vmlinux.gcc10.2.dwarf4.compressed
624M vmlinux.gcc10.2.dwarf2.compressed
630M vmlinux.gcc10.2.dwarf5
765M vmlinux.gcc10.2.dwarf4
809M vmlinux.gcc10.2.dwarf2

Though the quality of debug info is harder to quantify; size is not a
proxy for quality.

Jakub notes:
One thing is GCC DWARF-5 support, that is whether the compiler will
support -gdwarf-5 flag, and that support should be there from GCC 7

All [GCC] 5.1 - 6.x did was start accepting -gdwarf-5 as experimental
option that enabled some small DWARF subset (initially only a few
DW_LANG_* codes newly added to DWARF5 drafts). Only GCC 7 (released
after DWARF 5 has been finalized) started emitting DWARF5 section
headers and got most of the DWARF5 changes in...

Another separate thing is whether the assembler does support
the -gdwarf-5 option (i.e. if you can compile assembler files
with -Wa,-gdwarf-5) ... That option is about whether the assembler
will emit DWARF5 or DWARF2 .debug_line. It is fine to compile C sources
with -gdwarf-5 and use DWARF2 .debug_line for assembler files if as
doesn't support it.

Version check GCC so that we don't need to worry about the difference in
command line args between GNU readelf and llvm-readelf/llvm-dwarfdump to
validate the DWARF Version in the assembler feature detection script.

Most issues with clang produced assembler were fixed in binutils 2.35.1,
but 2.35.2 fixed issues related to requiring the flag -Wa,-gdwarf-5
explicitly. The added shell script test checks for the latter, and is
only required when using clang without its integrated assembler, though
we use for clang regardless as we do not yet have a way to query the
assembler from Kconfig.

Disabled for now if CONFIG_DEBUG_INFO_BTF is set; pahole doesn't yet
recognize the new additions to the DWARF debug info.

This only modifies the DWARF version emitted by the compiler, not the

The DWARF version of a binary can be validated with:
$ llvm-dwarfdump <object file> | head -n 4 | grep version
$ readelf --debug-dump=info <object file> 2>/dev/null | grep Version

Parts of the tree don't reuse DEBUG_CFLAGS as they should; such cleanup
is left as a follow up.

Reported-by: Sedat Dilek <>
Suggested-by: Arvind Sankar <>
Suggested-by: Caroline Tice <>
Suggested-by: Fangrui Song <>
Suggested-by: Jakub Jelinek <>
Suggested-by: Masahiro Yamada <>
Suggested-by: Nathan Chancellor <>
Signed-off-by: Nick Desaulniers <>
Tested-by: Sedat Dilek <> # LLVM/Clang v12.0.0-rc1 x86-64
Signed-off-by: Masahiro Yamada <>
diff 0dd77e95 Tue Jan 19 21:04:03 MST 2021 Masahiro Yamada <> kbuild: stop removing stale <linux/version.h> file

Revert commit 223c24a7dba9 ("kbuild: Automatically remove stale
<linux/version.h> file").

It was more than 6 years ago. I do not expect anybody to start
git-bisect for such a big window.

Signed-off-by: Masahiro Yamada <>
diff 6ee1d745 Sun Jan 24 17:47:14 MST 2021 Linus Torvalds <> Linux 5.11-rc5
diff 4c207c50 Tue Oct 13 17:47:37 MDT 2020 Nick Desaulniers <> Revert "kbuild: disable clang's default use of -fmerge-all-constants"

This reverts commit 87e0d4f0f37fb0c8c4aeeac46fff5e957738df79.

-fno-merge-all-constants has been the default since clang-6; the minimum
supported version of clang in the kernel is clang-10 (10.0.1).

Suggested-by: Nathan Chancellor <>
Signed-off-by: Nick Desaulniers <>
Signed-off-by: Andrew Morton <>
Tested-by: Sedat Dilek <>
Reviewed-by: Fangrui Song <>
Reviewed-by: Nathan Chancellor <>
Reviewed-by: Sedat Dilek <>
Reviewed-by: Kees Cook <>
Cc: Andrey Konovalov <>
Cc: Marco Elver <>
Cc: Miguel Ojeda <>
Cc: Alexei Starovoitov <>
Cc: Daniel Borkmann <>
Cc: Masahiro Yamada <>
Cc: Vincenzo Frascino <>
Cc: Will Deacon <>
Signed-off-by: Linus Torvalds <>
diff 6ad7cbc0 Sat Aug 22 08:56:18 MDT 2020 Nathan Huckleberry <> Makefile: Add clang-tidy and static analyzer support to makefile

This patch adds clang-tidy and the clang static-analyzer as make
targets. The goal of this patch is to make static analysis tools
usable and extendable by any developer or researcher who is familiar
with basic c++.

The current static analysis tools require intimate knowledge of the
internal workings of the static analysis. Clang-tidy and the clang
static analyzers expose an easy to use api and allow users unfamiliar
with clang to write new checks with relative ease.


Clang-tidy is an easily extendable 'linter' that runs on the AST.
Clang-tidy checks are easy to write and understand. A check consists of
two parts, a matcher and a checker. The matcher is created using a
domain specific language that acts on the AST
( When AST
nodes are found by the matcher a callback is made to the checker. The
checker can then execute additional checks and issue warnings.

Here is an example clang-tidy check to report functions that have calls
to local_irq_disable without calls to local_irq_enable and vice-versa.
Functions flagged with __attribute((annotation("ignore_irq_balancing")))
are ignored for analysis. (

===Clang static analyzer===

The clang static analyzer is a more powerful static analysis tool that
uses symbolic execution to find bugs. Currently there is a check that
looks for potential security bugs from invalid uses of kmalloc and
kfree. There are several more general purpose checks that are useful for
the kernel.

The clang static analyzer is well documented and designed to be

The main draw of the clang tools is how accessible they are. The clang
documentation is very nice and these tools are built specifically to be
easily extendable by any developer. They provide an accessible method of
bug-finding and research to people who are not overly familiar with the
kernel codebase.

Signed-off-by: Nathan Huckleberry <>
Reviewed-by: Nick Desaulniers <>
Tested-by: Nick Desaulniers <>
Tested-by: Lukas Bulwahn <>
Signed-off-by: Masahiro Yamada <>

Completed in 276 milliseconds