Skip to content
  1. May 11, 2015
  3. Apr 27, 2015
  5. Apr 24, 2015
    • Linus Torvalds's avatar
      x86: fix special __probe_kernel_write() tail zeroing case · d869844b
      Linus Torvalds authored 
      Commit cae2a173

 ("x86: clean up/fix 'copy_in_user()' tail zeroing")
fixed the failure case tail zeroing of one special case of the x86-64
generic user-copy routine, namely when used for the user-to-user case
("copy_in_user()").

But in the process it broke an even more unusual case: using the user
copy routine for kernel-to-kernel copying.

Now, normally kernel-kernel copies are obviously done using memcpy(),
but we have a couple of special cases when we use the user-copy
functions.  One is when we pass a kernel buffer to a regular user-buffer
routine, using set_fs(KERNEL_DS).  That's a "normal" case, and continued
to work fine, because it never takes any faults (with the possible
exception of a silent and successful vmalloc fault).

But Jan Beulich pointed out another, very unusual, special case: when we
use the user-copy routines not because it's a path that expects a user
pointer, but for a couple of ftrace/kgdb cases that want to do a kernel
copy, but do so using "unsafe" buffers, and use the user-copy routine to
gracefully handle faults.  IOW, for probe_kernel_write().

And that broke for the case of a faulting kernel destination, because we
saw the kernel destination and wanted to try to clear the tail of the
buffer.  Which doesn't work, since that's what faults.

This only triggers for things like kgdb and ftrace users (eg trying
setting a breakpoint on read-only memory), but it's definitely a bug.
The fix is to not compare against the kernel address start (TASK_SIZE),
but instead use the same limits "access_ok()" uses.

Reported-and-tested-by: default avatarJan Beulich <jbeulich@suse.com>
Cc: stable@vger.kernel.org # 4.0
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      d869844b
    • Ard Biesheuvel's avatar
      crypto: x86/sha512_ssse3 - fixup for asm function prototype change · 00425bb1
      Ard Biesheuvel authored 
      Patch e68410eb ("crypto: x86/sha512_ssse3 - move SHA-384/512
SSSE3 implementation to base layer") changed the prototypes of the
core asm SHA-512 implementations so that they are compatible with
the prototype used by the base layer.

However, in one instance, the register that was used for passing the
input buffer was reused as a scratch register later on in the code,
and since the input buffer param changed places with the digest param
-which needs to be written back before the function returns- this
resulted in the scratch register to be dereferenced in a memory write
operation, causing a GPF.

Fix this by changing the scratch register to use the same register as
the input buffer param again.

Fixes: e68410eb

 ("crypto: x86/sha512_ssse3 - move SHA-384/512 SSSE3 implementation to base layer")
Reported-By: default avatarBobby Powers <bobbypowers@gmail.com>
Tested-By: default avatarBobby Powers <bobbypowers@gmail.com>
Signed-off-by: default avatarArd Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: default avatarHerbert Xu <herbert@gondor.apana.org.au>
      00425bb1
    • Ley Foon Tan's avatar
      nios2: rework cache · 1a70db49
      Ley Foon Tan authored 
      

- flush dcache before flush instruction cache
- remork update_mmu_cache and flush_dcache_page
- add shmparam.h

Signed-off-by: default avatarLey Foon Tan <lftan@altera.com>
      1a70db49
    • Ezequiel Garcia's avatar
      nios2: Add types.h header required for __u32 type · 2009337e
      Ezequiel Garcia authored 
      

Reported by the header checker (CONFIG_HEADERS_CHECK=y):

  CHECK   usr/include/asm/ (31 files)
./usr/include/asm/ptrace.h:77: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: default avatarEzequiel Garcia <ezequiel@vanguardiasur.com.ar>
Acked-by: default avatarLey Foon Tan <lftan@altera.com>
      2009337e
    • Steven Miao's avatar
      eth: bf609 eth clock: add pclk clock for stmmac driver probe · d91e14b3
      Steven Miao authored 
      

Signed-off-by: default avatarSteven Miao <realmz6@gmail.com>
      d91e14b3
    • Chen Gang's avatar
      blackfin: Wire up missing syscalls · 4f650a59
      Chen Gang authored 
      

The related syscalls are below which may cause samples/kdbus building
break in next-20150401 tree, the related information and error:

    CALL    scripts/checksyscalls.sh
  <stdin>:1223:2: warning: #warning syscall kcmp not implemented [-Wcpp]
  <stdin>:1226:2: warning: #warning syscall finit_module not implemented [-Wcpp]
  <stdin>:1229:2: warning: #warning syscall sched_setattr not implemented [-Wcpp]
  <stdin>:1232:2: warning: #warning syscall sched_getattr not implemented [-Wcpp]
  <stdin>:1235:2: warning: #warning syscall renameat2 not implemented [-Wcpp]
  <stdin>:1238:2: warning: #warning syscall seccomp not implemented [-Wcpp]
  <stdin>:1241:2: warning: #warning syscall getrandom not implemented [-Wcpp]
  <stdin>:1244:2: warning: #warning syscall memfd_create not implemented [-Wcpp]
  <stdin>:1247:2: warning: #warning syscall bpf not implemented [-Wcpp]
  <stdin>:1250:2: warning: #warning syscall execveat not implemented [-Wcpp]
  [...]
    HOSTCC  samples/kdbus/kdbus-workers
  samples/kdbus/kdbus-workers.c: In function ‘prime_new’:
  samples/kdbus/kdbus-workers.c:930:18: error: ‘__NR_memfd_create’ undeclared (first use in this function)
    p->fd = syscall(__NR_memfd_create, "prime-area", MFD_CLOEXEC);
                    ^
  samples/kdbus/kdbus-workers.c:930:18: note: each undeclared identifier is reported only once for each function it appears in

Signed-off-by: default avatarChen Gang <gang.chen.5i5j@gmail.com>
      4f650a59
  7. Apr 23, 2015
  9. Apr 22, 2015
  11. Apr 21, 2015
    • Nathan Lynch's avatar
      ARM: 8344/1: VDSO: honor CONFIG_VDSO in Makefile · f80f6531
      Nathan Lynch authored 
      

When CONFIG_VDSO=n, the build normally does not enter arch/arm/vdso/
because arch/arm/Makefile does not add it to core-y.

However, if the user runs 'make arch/arm/vdso/' the VDSO targets will
get visited.  This is because the VDSO Makefile itself does not
consider the value of CONFIG_VDSO.

It is arguably better and more consistent behavior to generate an
empty built-in.o when CONFIG_VDSO=n and the user attempts to build
arch/arm/vdso/.  It's nicer because it doesn't try to build things
that Kconfig dependencies are there to prevent (e.g. the dependency on
AEABI), and it's less confusing than building objects that won't be
used in the final image.

Signed-off-by: default avatarNathan Lynch <nathan_lynch@mentor.com>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      f80f6531
    • Nathan Lynch's avatar
      ARM: 8343/1: VDSO: add build artifacts to .gitignore · 2b507a2d
      Nathan Lynch authored 
      

vdsomunge and vdso.so.raw are outputs that don't get matched by the
normal ignore rules.

Signed-off-by: default avatarNathan Lynch <nathan_lynch@mentor.com>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      2b507a2d
    • Russell King's avatar
      ARM: Fix nommu booting · 0a9024e8
      Russell King authored 
      Commit bf35706f ("ARM: 8314/1: replace PROCINFO embedded branch with
relative offset") broke booting on nommu platforms as it didn't update
the nommu boot code.  This patch fixes that oversight.

Fixes: bf35706f

 ("ARM: 8314/1: replace PROCINFO embedded branch with relative offset")
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      0a9024e8
    • Paul Mackerras's avatar
      KVM: PPC: Book3S HV: Use msgsnd for signalling threads on POWER8 · 66feed61
      Paul Mackerras authored 
      

This uses msgsnd where possible for signalling other threads within
the same core on POWER8 systems, rather than IPIs through the XICS
interrupt controller.  This includes waking secondary threads to run
the guest, the interrupts generated by the virtual XICS, and the
interrupts to bring the other threads out of the guest when exiting.

Aggregated statistics from debugfs across vcpus for a guest with 32
vcpus, 8 threads/vcore, running on a POWER8, show this before the
change:

 rm_entry:     3387.6ns (228 - 86600, 1008969 samples)
  rm_exit:     4561.5ns (12 - 3477452, 1009402 samples)
  rm_intr:     1660.0ns (12 - 553050, 3600051 samples)

and this after the change:

 rm_entry:     3060.1ns (212 - 65138, 953873 samples)
  rm_exit:     4244.1ns (12 - 9693408, 954331 samples)
  rm_intr:     1342.3ns (12 - 1104718, 3405326 samples)

for a test of booting Fedora 20 big-endian to the login prompt.

The time taken for a H_PROD hcall (which is handled in the host
kernel) went down from about 35 microseconds to about 16 microseconds
with this change.

The noinline added to kvmppc_run_core turned out to be necessary for
good performance, at least with gcc 4.9.2 as packaged with Fedora 21
and a little-endian POWER8 host.

Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
Signed-off-by: default avatarAlexander Graf <agraf@suse.de>
      66feed61
    • Paul Mackerras's avatar
      KVM: PPC: Book3S HV: Translate kvmhv_commence_exit to C · eddb60fb
      Paul Mackerras authored 
      

This replaces the assembler code for kvmhv_commence_exit() with C code
in book3s_hv_builtin.c.  It also moves the IPI sending code that was
in book3s_hv_rm_xics.c into a new kvmhv_rm_send_ipi() function so it
can be used by kvmhv_commence_exit() as well as icp_rm_set_vcpu_irq().

Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
Signed-off-by: default avatarAlexander Graf <agraf@suse.de>
      eddb60fb
    • Paul Mackerras's avatar
      KVM: PPC: Book3S HV: Streamline guest entry and exit · 6af27c84
      Paul Mackerras authored 
      

On entry to the guest, secondary threads now wait for the primary to
switch the MMU after loading up most of their state, rather than before.
This means that the secondary threads get into the guest sooner, in the
common case where the secondary threads get to kvmppc_hv_entry before
the primary thread.

On exit, the first thread out increments the exit count and interrupts
the other threads (to get them out of the guest) before saving most
of its state, rather than after.  That means that the other threads
exit sooner and means that the first thread doesn't spend so much
time waiting for the other threads at the point where the MMU gets
switched back to the host.

This pulls out the code that increments the exit count and interrupts
other threads into a separate function, kvmhv_commence_exit().
This also makes sure that r12 and vcpu->arch.trap are set correctly
in some corner cases.

Statistics from /sys/kernel/debug/kvm/vm*/vcpu*/timings show the
improvement.  Aggregating across vcpus for a guest with 32 vcpus,
8 threads/vcore, running on a POWER8, gives this before the change:

 rm_entry:     avg 4537.3ns (222 - 48444, 1068878 samples)
  rm_exit:     avg 4787.6ns (152 - 165490, 1010717 samples)
  rm_intr:     avg 1673.6ns (12 - 341304, 3818691 samples)

and this after the change:

 rm_entry:     avg 3427.7ns (232 - 68150, 1118921 samples)
  rm_exit:     avg 4716.0ns (12 - 150720, 1119477 samples)
  rm_intr:     avg 1614.8ns (12 - 522436, 3850432 samples)

showing a substantial reduction in the time spent per guest entry in
the real-mode guest entry code, and smaller reductions in the real
mode guest exit and interrupt handling times.  (The test was to start
the guest and boot Fedora 20 big-endian to the login prompt.)

Signed-off-by: default avatarPaul Mackerras <paulus@samba.org>
Signed-off-by: default avatarAlexander Graf <agraf@suse.de>
      6af27c84