linux.git/arch/arc/kernel, branch v5.13

ARC: fix CONFIG_HARDENED_USERCOPY

2021-06-11T00:37:00+00:00

Currently enabling this triggers a warning

| usercopy: Kernel memory overwrite attempt detected to kernel text (offset 155633, size 11)!
| usercopy: BUG: failure at mm/usercopy.c:99/usercopy_abort()!
|
|gcc generated __builtin_trap
|Path: /bin/busybox
|CPU: 0 PID: 84 Comm: init Not tainted 5.4.22
|
|[ECR ]: 0x00090005 => gcc generated __builtin_trap
|[EFA ]: 0x9024fcaa
|[BLINK ]: usercopy_abort+0x8a/0x8c
|[ERET ]: memfd_fcntl+0x0/0x470
|[STAT32]: 0x80080802 : IE K
|...
|...
|Stack Trace:
| memfd_fcntl+0x0/0x470
| usercopy_abort+0x8a/0x8c
| __check_object_size+0x10e/0x138
| copy_strings+0x1f4/0x38c
| __do_execve_file+0x352/0x848
| EV_Trap+0xcc/0xd0

The issue is triggered by an allocation in "init reclaimed" region.
ARC _stext emcompasses the init region (for historical reasons we wanted
the init.text to be under .text as well). This however trips up
__check_object_size()->check_kernel_text_object() which treats this as
object bleeding into kernel text.

Fix that by rezoning _stext to start from regular kernel .text and leave
out .init altogether.

Fixes: https://github.com/foss-for-synopsys-dwc-arc-processors/linux/issues/15
Reported-by: Evgeniy Didin 
Reviewed-by: Kees Cook 
Signed-off-by: Vineet Gupta

ARCv2: save ABI registers across signal handling

2021-06-11T00:21:38+00:00

ARCv2 has some configuration dependent registers (r30, r58, r59) which
could be targetted by the compiler. To keep the ABI stable, these were
unconditionally part of the glibc ABI
(sysdeps/unix/sysv/linux/arc/sys/ucontext.h:mcontext_t) however we
missed populating them (by saving/restoring them across signal
handling).

This patch fixes the issue by
 - adding arcv2 ABI regs to kernel struct sigcontext
 - populating them during signal handling

Change to struct sigcontext might seem like a glibc ABI change (although
it primarily uses ucontext_t:mcontext_t) but the fact is
 - it has only been extended (existing fields are not touched)
 - the old sigcontext was ABI incomplete to begin with anyways

Fixes: https://github.com/foss-for-synopsys-dwc-arc-processors/linux/issues/53
Cc: 
Tested-by: kernel test robot 
Reported-by: Vladimir Isaev 
Signed-off-by: Vineet Gupta

ARC: entry: fix off-by-one error in syscall number validation

2021-05-10T19:38:59+00:00

We have NR_syscall syscalls from [0 .. NR_syscall-1].
However the check for invalid syscall number is "> NR_syscall" as
opposed to >=. This off-by-one error erronesously allows "NR_syscall"
to be treated as valid syscall causeing out-of-bounds access into
syscall-call table ensuing a crash (holes within syscall table have a
invalid-entry handler but this is beyond the array implementing the
table).

This problem showed up on v5.6 kernel when testing glibc 2.33 (v5.10
kernel capable, includng faccessat2 syscall 439). The v5.6 kernel has
NR_syscalls=439 (0 to 438). Due to the bug, 439 passed by glibc was
not handled as -ENOSYS but processed leading to a crash.

Link: https://github.com/foss-for-synopsys-dwc-arc-processors/linux/issues/48
Reported-by: Shahab Vahedi 
Cc: 
Signed-off-by: Vineet Gupta

ARC: kgdb: add 'fallthrough' to prevent a warning

2021-05-10T19:38:59+00:00

Use the 'fallthrough' macro to document that this switch case
does indeed fall through to the next case.

../arch/arc/kernel/kgdb.c: In function 'kgdb_arch_handle_exception':
../arch/arc/kernel/kgdb.c:141:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
  141 |   if (kgdb_hex2long(&ptr, &addr))
      |      ^
../arch/arc/kernel/kgdb.c:144:2: note: here
  144 |  case 'D':
      |  ^~~~

Cc: linux-snps-arc@lists.infradead.org
Signed-off-by: Randy Dunlap 
Signed-off-by: Vineet Gupta

arc: Fix typos/spellos

2021-05-10T19:38:59+00:00

s/commiting/committing/
s/defintion/definition/
s/gaurantees/guarantees/
s/interrpted/interrupted/
s/interrutps/interrupts/
s/succeded/succeeded/
s/unconditonally/unconditionally/

Reviewed-by: Christian Brauner 
Acked-by: Randy Dunlap 
Signed-off-by: Bhaskar Chowdhury 
Signed-off-by: Vineet Gupta

ARC: treewide: avoid the pointer addition with NULL pointer

2021-03-22T17:05:43+00:00

Signed-off-by: dean.yang_cp 
Acked-by: Vineet Gupta 
Signed-off-by: Vineet Gupta

arc: kernel: Return -EFAULT if copy_to_user() fails

2021-03-22T17:05:43+00:00

The copy_to_user() function returns the number of bytes remaining to be
copied, but we want to return -EFAULT if the copy doesn't complete.

Signed-off-by: Wang Qing 
Signed-off-by: Vineet Gupta

arch: setup PF_IO_WORKER threads like PF_KTHREAD

2021-02-22T00:25:22+00:00

PF_IO_WORKER are kernel threads too, but they aren't PF_KTHREAD in the
sense that we don't assign ->set_child_tid with our own structure. Just
ensure that every arch sets up the PF_IO_WORKER threads like kthreads
in the arch implementation of copy_thread().

Signed-off-by: Jens Axboe

ARC: unbork 5.11 bootup: fix snafu in _TIF_NOTIFY_SIGNAL handling

2021-01-07T22:04:42+00:00

Linux 5.11.rcX was failing to boot on ARC HSDK board. Turns out we have
a couple of issues, this being the first one, and I'm to blame as I
didn't pay attention during review.

TIF_NOTIFY_SIGNAL support requires checking multiple TIF_* bits in
kernel return code path. Old code only needed to check a single bit so
BBIT0  worked. New code needs to check multiple bits so
AND  instruction. So needs to use bit mask variant _TIF_SIGPENDING

Cc: Jens Axboe 
Fixes: 53855e12588743ea128 ("arc: add support for TIF_NOTIFY_SIGNAL")
Link: https://github.com/foss-for-synopsys-dwc-arc-processors/linux/issues/34
Signed-off-by: Vineet Gupta

Merge tag 'tif-task_work.arch-2020-12-14' of git://git.kernel.dk/linux-block

2020-12-16T20:33:35+00:00

Pull TIF_NOTIFY_SIGNAL updates from Jens Axboe:
 "This sits on top of of the core entry/exit and x86 entry branch from
  the tip tree, which contains the generic and x86 parts of this work.

  Here we convert the rest of the archs to support TIF_NOTIFY_SIGNAL.

  With that done, we can get rid of JOBCTL_TASK_WORK from task_work and
  signal.c, and also remove a deadlock work-around in io_uring around
  knowing that signal based task_work waking is invoked with the sighand
  wait queue head lock.

  The motivation for this work is to decouple signal notify based
  task_work, of which io_uring is a heavy user of, from sighand. The
  sighand lock becomes a huge contention point, particularly for
  threaded workloads where it's shared between threads. Even outside of
  threaded applications it's slower than it needs to be.

  Roman Gershman  reported that his networked
  workload dropped from 1.6M QPS at 80% CPU to 1.0M QPS at 100% CPU
  after io_uring was changed to use TIF_NOTIFY_SIGNAL. The time was all
  spent hammering on the sighand lock, showing 57% of the CPU time there
  [1].

  There are further cleanups possible on top of this. One example is
  TIF_PATCH_PENDING, where a patch already exists to use
  TIF_NOTIFY_SIGNAL instead. Hopefully this will also lead to more
  consolidation, but the work stands on its own as well"

[1] https://github.com/axboe/liburing/issues/215

* tag 'tif-task_work.arch-2020-12-14' of git://git.kernel.dk/linux-block: (28 commits)
  io_uring: remove 'twa_signal_ok' deadlock work-around
  kernel: remove checking for TIF_NOTIFY_SIGNAL
  signal: kill JOBCTL_TASK_WORK
  io_uring: JOBCTL_TASK_WORK is no longer used by task_work
  task_work: remove legacy TWA_SIGNAL path
  sparc: add support for TIF_NOTIFY_SIGNAL
  riscv: add support for TIF_NOTIFY_SIGNAL
  nds32: add support for TIF_NOTIFY_SIGNAL
  ia64: add support for TIF_NOTIFY_SIGNAL
  h8300: add support for TIF_NOTIFY_SIGNAL
  c6x: add support for TIF_NOTIFY_SIGNAL
  alpha: add support for TIF_NOTIFY_SIGNAL
  xtensa: add support for TIF_NOTIFY_SIGNAL
  arm: add support for TIF_NOTIFY_SIGNAL
  microblaze: add support for TIF_NOTIFY_SIGNAL
  hexagon: add support for TIF_NOTIFY_SIGNAL
  csky: add support for TIF_NOTIFY_SIGNAL
  openrisc: add support for TIF_NOTIFY_SIGNAL
  sh: add support for TIF_NOTIFY_SIGNAL
  um: add support for TIF_NOTIFY_SIGNAL
  ...