linux-stable.git/arch/parisc, branch v4.19.76

parisc: Fix build of compressed kernel even with debug enabled

2019-08-06T17:06:55+00:00

commit 3fe6c873af2f2247544debdbe51ec29f690a2ccf upstream.

With debug info enabled (CONFIG_DEBUG_INFO=y) the resulting vmlinux may get
that huge that we need to increase the start addresss for the decompression
text section otherwise one will face a linker error.

Reported-by: Sven Schnelle 
Tested-by: Sven Schnelle 
Cc: stable@vger.kernel.org # v4.14+
Signed-off-by: Helge Deller 
Signed-off-by: Greg Kroah-Hartman

parisc: Fix kernel panic due invalid values in IAOQ0 or IAOQ1

2019-07-26T07:14:29+00:00

commit 10835c854685393a921b68f529bf740fa7c9984d upstream.

On parisc the privilege level of a process is stored in the lowest two bits of
the instruction pointers (IAOQ0 and IAOQ1). On Linux we use privilege level 0
for the kernel and privilege level 3 for user-space. So userspace should not be
allowed to modify IAOQ0 or IAOQ1 of a ptraced process to change it's privilege
level to e.g. 0 to try to gain kernel privileges.

This patch prevents such modifications by always setting the two lowest bits to
one (which relates to privilege level 3 for user-space) if IAOQ0 or IAOQ1 are
modified via ptrace calls in the native and compat ptrace paths.

Link: https://bugs.gentoo.org/481768
Reported-by: Jeroen Roovers 
Cc: 
Tested-by: Rolf Eike Beer 
Signed-off-by: Helge Deller 
Signed-off-by: Greg Kroah-Hartman

parisc: Ensure userspace privilege for ptraced processes in regset functions

2019-07-26T07:14:29+00:00

commit 34c32fc603311a72cb558e5e337555434f64c27b upstream.

On parisc the privilege level of a process is stored in the lowest two bits of
the instruction pointers (IAOQ0 and IAOQ1). On Linux we use privilege level 0
for the kernel and privilege level 3 for user-space. So userspace should not be
allowed to modify IAOQ0 or IAOQ1 of a ptraced process to change it's privilege
level to e.g. 0 to try to gain kernel privileges.

This patch prevents such modifications in the regset support functions by
always setting the two lowest bits to one (which relates to privilege level 3
for user-space) if IAOQ0 or IAOQ1 are modified via ptrace regset calls.

Link: https://bugs.gentoo.org/481768
Cc:  # v4.7+
Tested-by: Rolf Eike Beer 
Signed-off-by: Helge Deller 
Signed-off-by: Greg Kroah-Hartman

parisc: Fix compiler warnings in float emulation code

2019-06-25T03:35:55+00:00

[ Upstream commit 6b98d9134e14f5ef4bcf64b27eedf484ed19a1ec ]

Avoid such compiler warnings:
arch/parisc/math-emu/cnv_float.h:71:27: warning: ‘<<’ in boolean context, did you mean ‘<’ ? [-Wint-in-bool-context]
     ((Dintp1(dint_valueA) << 33 - SGL_EXP_LENGTH) || Dintp2(dint_valueB))
arch/parisc/math-emu/fcnvxf.c:257:6: note: in expansion of macro ‘Dint_isinexact_to_sgl’
  if (Dint_isinexact_to_sgl(srcp1,srcp2)) {

Signed-off-by: Helge Deller 
Signed-off-by: Sasha Levin

parisc: Rename LEVEL to PA_ASM_LEVEL to avoid name clash with DRBD code

2019-05-25T16:23:24+00:00

commit 1829dda0e87f4462782ca81be474c7890efe31ce upstream.

LEVEL is a very common word, and now after many years it suddenly
clashed with another LEVEL define in the DRBD code.
Rename it to PA_ASM_LEVEL instead.

Reported-by: kbuild test robot 
Signed-off-by: Helge Deller 
Cc: 
Signed-off-by: Greg Kroah-Hartman

parisc: Use PA_ASM_LEVEL in boot code

2019-05-25T16:23:23+00:00

commit bdca5d64ee92abeacd6dada0bc6f6f8e6350dd67 upstream.

The LEVEL define clashed with the DRBD code.

Reported-by: kbuild test robot 
Signed-off-by: Helge Deller 
Cc:  # v4.14+
Signed-off-by: Greg Kroah-Hartman

parisc: Export running_on_qemu symbol for modules

2019-05-25T16:23:23+00:00

commit 3e1120f4b57bc12437048494ab56648edaa5b57d upstream.

Signed-off-by: Helge Deller 
CC: stable@vger.kernel.org # v4.9+
Signed-off-by: Greg Kroah-Hartman

parisc: also set iaoq_b in instruction_pointer_set()

2019-04-17T06:38:49+00:00

commit f324fa58327791b2696628b31480e7e21c745706 upstream.

When setting the instruction pointer on PA-RISC we also need
to set the back of the instruction queue to the new offset, otherwise
we will execute on instruction from the new location, and jumping
back to the old location stored in iaoq_b.

Signed-off-by: Sven Schnelle 
Signed-off-by: Helge Deller 
Fixes: 75ebedf1d263 ("parisc: Add HAVE_REGS_AND_STACK_ACCESS_API feature")
Cc: stable@vger.kernel.org # 4.19+
Signed-off-by: Greg Kroah-Hartman

parisc: regs_return_value() should return gpr28

2019-04-17T06:38:49+00:00

commit 45efd871bf0a47648f119d1b41467f70484de5bc upstream.

While working on kretprobes for PA-RISC I was wondering while the
kprobes sanity test always fails on kretprobes. This is caused by
returning gpr20 instead of gpr28.

Signed-off-by: Sven Schnelle 
Signed-off-by: Helge Deller 
Cc: stable@vger.kernel.org # 4.14+
Signed-off-by: Greg Kroah-Hartman

parisc: Detect QEMU earlier in boot process

2019-04-17T06:38:49+00:00

commit d006e95b5561f708d0385e9677ffe2c46f2ae345 upstream.

While adding LASI support to QEMU, I noticed that the QEMU detection in
the kernel happens much too late. For example, when a LASI chip is found
by the kernel, it registers the LASI LED driver as well.  But when we
run on QEMU it makes sense to avoid spending unnecessary CPU cycles, so
we need to access the running_on_QEMU flag earlier than before.

This patch now makes the QEMU detection the fist task of the Linux
kernel by moving it to where the kernel enters the C-coding.

Fixes: 310d82784fb4 ("parisc: qemu idle sleep support")
Signed-off-by: Helge Deller 
Cc: stable@vger.kernel.org # v4.14+
Signed-off-by: Greg Kroah-Hartman