| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This cleans up the module support that was commited earlier to work with
what's actually emitted from our GCC port as it lands upstream. Most of
the work here is adding new relocations to the kernel.
There's some limitations on module loading imposed by the kernel:
* The kernel doesn't support linker relaxation, which is necessary to
support R_RISCV_ALIGN. In order to get reliable module building
you're going to need to a GCC that supports the new '-mno-relax',
which IIRC isn't going to be out until 8.1.0. It's somewhat unlikely
that R_RISCV_ALIGN will appear in a module even without '-mno-relax'
support, so issues shouldn't be common.
* There is no large code model for RISC-V, which means modules must be
loaded within a 32-bit signed offset of the kernel. We don't
currently have any mechanism for ensuring this memory remains free or
moving pages around, so issues here might be common.
I fixed a singcle merge conflict in arch/riscv/kernel/Makefile.
|
|
These fixes fall into three categories
* The definiton of __smp_{store_release,load_acquire}, which allow us to
emit a full fence when unnecessary.
* Fixes to avoid relying on the behavior of "*.aqrl" atomics, as those
are specified in the currently released RISC-V memory model draft in
a way that makes them useless for Linux. This might change in the
future, but now the code matches the memory model spec as it's written
so at least we're getting closer to something sane. The actual fix is
to delete the RISC-V specific atomics and drop back to generic
versions that use the new fences from above.
* Cleanups to our atomic macros, which are mostly non-functional
changes.
Unfortunately I haven't given these as thorough of a testing as I
probably should have, but I've poked through the code and they seem
generally OK.
|
|
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Just fail on align type. Kernel modules loader didn't do relax
like linker, it is difficult to remove or migrate the code,
but the remnant nop instructions harm the performaace of module.
We expect the building module with the no-relax option.
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
HI20 and LO12_I/LO12_S relocate the absolute address, the range of
offset must in 32-bit.
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
For CALL_PLT, emit the plt entry only when offset is more than 32-bit.
For PCREL_LO12, it uses the location of corresponding HI20 to
get the address of external symbol. It should check the HI20 type
is the PCREL_HI20 or GOT_HI20, because sometime the location will
have two or more relocation types.
For example:
0: 00000797 auipc a5,0x0
0: R_RISCV_ALIGN *ABS*
0: R_RISCV_GOT_HI20 SYMBOL
4: 0007b783 ld a5,0(a5) # 0 <SYMBOL>
4: R_RISCV_PCREL_LO12_I .L0
4: R_RISCV_RELAX *ABS*
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Separate the function symbol address from .plt to .got.plt section.
The original plt entry has trampoline code with symbol address,
there is a 32-bit padding bwtween jar instruction and symbol address.
Extract the symbol address to .got.plt to reduce the module size.
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
The address of external symbols will locate more than 32-bit offset
in 64-bit kernel with sv39 or sv48 virtual addressing.
Module loader emits the GOT and PLT entries for data symbols and
function symbols respectively.
The PLT entry is a trampoline code for jumping to the 64-bit
real address. The GOT entry is just the data symbol address.
Signed-off-by: Zong Li <zong@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Atomics present the same issue with locking: release and acquire
variants need to be strengthened to meet the constraints defined
by the Linux-kernel memory consistency model [1].
Atomics present a further issue: implementations of atomics such
as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs,
which do not give full-ordering with .aqrl; for example, current
implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test
below to end up with the state indicated in the "exists" clause.
In order to "synchronize" LKMM and RISC-V's implementation, this
commit strengthens the implementations of the atomics operations
by replacing .rl and .aq with the use of ("lightweigth") fences,
and by replacing .aqrl LR/SC pairs in sequences such as:
0: lr.w.aqrl %0, %addr
bne %0, %old, 1f
...
sc.w.aqrl %1, %new, %addr
bnez %1, 0b
1:
with sequences of the form:
0: lr.w %0, %addr
bne %0, %old, 1f
...
sc.w.rl %1, %new, %addr /* SC-release */
bnez %1, 0b
fence rw, rw /* "full" fence */
1:
following Daniel's suggestion.
These modifications were validated with simulation of the RISC-V
memory consistency model.
C lr-sc-aqrl-pair-vs-full-barrier
{}
P0(int *x, int *y, atomic_t *u)
{
int r0;
int r1;
WRITE_ONCE(*x, 1);
r0 = atomic_cmpxchg(u, 0, 1);
r1 = READ_ONCE(*y);
}
P1(int *x, int *y, atomic_t *v)
{
int r0;
int r1;
WRITE_ONCE(*y, 1);
r0 = atomic_cmpxchg(v, 0, 1);
r1 = READ_ONCE(*x);
}
exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)
[1] https://marc.info/?l=linux-kernel&m=151930201102853&w=2
https://groups.google.com/a/groups.riscv.org/forum/#!topic/isa-dev/hKywNHBkAXM
https://marc.info/?l=linux-kernel&m=151633436614259&w=2
Suggested-by: Daniel Lustig <dlustig@nvidia.com>
Signed-off-by: Andrea Parri <parri.andrea@gmail.com>
Cc: Palmer Dabbelt <palmer@sifive.com>
Cc: Albert Ou <albert@sifive.com>
Cc: Daniel Lustig <dlustig@nvidia.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Jade Alglave <j.alglave@ucl.ac.uk>
Cc: Luc Maranget <luc.maranget@inria.fr>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Akira Yokosawa <akiyks@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-riscv@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Current implementations map locking operations using .rl and .aq
annotations. However, this mapping is unsound w.r.t. the kernel
memory consistency model (LKMM) [1]:
Referring to the "unlock-lock-read-ordering" test reported below,
Daniel wrote:
"I think an RCpc interpretation of .aq and .rl would in fact
allow the two normal loads in P1 to be reordered [...]
The intuition would be that the amoswap.w.aq can forward from
the amoswap.w.rl while that's still in the store buffer, and
then the lw x3,0(x4) can also perform while the amoswap.w.rl
is still in the store buffer, all before the l1 x1,0(x2)
executes. That's not forbidden unless the amoswaps are RCsc,
unless I'm missing something.
Likewise even if the unlock()/lock() is between two stores.
A control dependency might originate from the load part of
the amoswap.w.aq, but there still would have to be something
to ensure that this load part in fact performs after the store
part of the amoswap.w.rl performs globally, and that's not
automatic under RCpc."
Simulation of the RISC-V memory consistency model confirmed this
expectation.
In order to "synchronize" LKMM and RISC-V's implementation, this
commit strengthens the implementations of the locking operations
by replacing .rl and .aq with the use of ("lightweigth") fences,
resp., "fence rw, w" and "fence r , rw".
C unlock-lock-read-ordering
{}
/* s initially owned by P1 */
P0(int *x, int *y)
{
WRITE_ONCE(*x, 1);
smp_wmb();
WRITE_ONCE(*y, 1);
}
P1(int *x, int *y, spinlock_t *s)
{
int r0;
int r1;
r0 = READ_ONCE(*y);
spin_unlock(s);
spin_lock(s);
r1 = READ_ONCE(*x);
}
exists (1:r0=1 /\ 1:r1=0)
[1] https://marc.info/?l=linux-kernel&m=151930201102853&w=2
https://groups.google.com/a/groups.riscv.org/forum/#!topic/isa-dev/hKywNHBkAXM
https://marc.info/?l=linux-kernel&m=151633436614259&w=2
Signed-off-by: Andrea Parri <parri.andrea@gmail.com>
Cc: Palmer Dabbelt <palmer@sifive.com>
Cc: Albert Ou <albert@sifive.com>
Cc: Daniel Lustig <dlustig@nvidia.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Jade Alglave <j.alglave@ucl.ac.uk>
Cc: Luc Maranget <luc.maranget@inria.fr>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Akira Yokosawa <akiyks@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-riscv@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Introduce __smp_{store_release,load_acquire}, and rely on the generic
definitions for smp_{store_release,load_acquire}. This avoids the use
of full ("rw,rw") fences on SMP.
Signed-off-by: Andrea Parri <parri.andrea@gmail.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
In walk_stackframe, the pc now receives the address from calling
ftrace_graph_ret_addr instead of manual calculation.
Note that the original calculation,
pc = frame->ra - 4
is buggy when the instruction at the return address happened to be a
compressed inst. But since it is not a critical part of ftrace, it is
ignored for now to ease the review process.
Cc: Greentime Hu <greentime@andestech.com>
Signed-off-by: Alan Kao <alankao@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Cc: Greentime Hu <greentime@andestech.com>
Signed-off-by: Alan Kao <alankao@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Cc: Greentime Hu <greentime@andestech.com>
Signed-off-by: Alan Kao <alankao@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Once the function_graph tracer is enabled, a filtered function has the
following call sequence:
* ftracer_caller ==> on/off by ftrace_make_call/ftrace_make_nop
* ftrace_graph_caller
* ftrace_graph_call ==> on/off by ftrace_en/disable_ftrace_graph_caller
* prepare_ftrace_return
Considering the following DYNAMIC_FTRACE_WITH_REGS feature, it would be
more extendable to have a ftrace_graph_caller function, instead of
calling prepare_ftrace_return directly in ftrace_caller.
Cc: Greentime Hu <greentime@andestech.com>
Signed-off-by: Alan Kao <alankao@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
We now have dynamic ftrace with the following added items:
* ftrace_make_call, ftrace_make_nop (in kernel/ftrace.c)
The two functions turn each recorded call site of filtered functions
into a call to ftrace_caller or nops
* ftracce_update_ftrace_func (in kernel/ftrace.c)
turns the nops at ftrace_call into a call to a generic entry for
function tracers.
* ftrace_caller (in kernel/mcount-dyn.S)
The entry where each _mcount call sites calls to once they are
filtered to be traced.
Also, this patch fixes the semantic problems in mcount.S, which will be
treated as only a reference implementation once we have the dynamic
ftrace.
Cc: Greentime Hu <greentime@andestech.com>
Signed-off-by: Alan Kao <alankao@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
Now recordmcount.pl recognizes RISC-V object files. For the mechanism to
work, we have to disable the linker relaxation.
Cc: Greentime Hu <greentime@andestech.com>
Signed-off-by: Alan Kao <alankao@andestech.com>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"Two fixlets"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/hwbp: Simplify the perf-hwbp code, fix documentation
perf/x86/intel: Fix linear IP of PEBS real_ip on Haswell and later CPUs
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
"Two UV platform fixes, and a kbuild fix"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/platform/UV: Fix critical UV MMR address error
x86/platform/uv/BAU: Add APIC idt entry
x86/purgatory: Avoid creating stray .<pid>.d files, remove -MD from KBUILD_CFLAGS
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 PTI fixes from Ingo Molnar:
"Two fixes: a relatively simple objtool fix that makes Clang built
kernels work with ORC debug info, plus an alternatives macro fix"
* 'x86-pti-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/alternatives: Fixup alternative_call_2
objtool: Add Clang support
|
|
Pull KVM fixes from Radim Krčmář:
"PPC:
- Fix a bug causing occasional machine check exceptions on POWER8
hosts (introduced in 4.16-rc1)
x86:
- Fix a guest crashing regression with nested VMX and restricted
guest (introduced in 4.16-rc1)
- Fix dependency check for pv tlb flush (the wrong dependency that
effectively disabled the feature was added in 4.16-rc4, the
original feature in 4.16-rc1, so it got decent testing)"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: x86: Fix pv tlb flush dependencies
KVM: nVMX: sync vmcs02 segment regs prior to vmx_set_cr0
KVM: PPC: Book3S HV: Fix duplication of host SLB entries
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"Some more powerpc fixes for 4.16. Apologies if this is a bit big at
rc7, but they're all reasonably important fixes. None are actually for
new code, so they aren't indicative of 4.16 being in bad shape from
our point of view.
- Fix missing AT_BASE_PLATFORM (in auxv) when we're using a new
firmware interface for describing CPU features.
- Fix lost pending interrupts due to a race in our interrupt
soft-masking code.
- A workaround for a nest MMU bug with TLB invalidations on Power9.
- A workaround for broadcast TLB invalidations on Power9.
- Fix a bug in our instruction SLB miss handler, when handling bad
addresses (eg. >= TASK_SIZE), which could corrupt non-volatile user
GPRs.
Thanks to: Aneesh Kumar K.V, Balbir Singh, Benjamin Herrenschmidt,
Nicholas Piggin"
* tag 'powerpc-4.16-6' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/64s: Fix i-side SLB miss bad address handler saving nonvolatile GPRs
powerpc/mm: Fixup tlbie vs store ordering issue on POWER9
powerpc/mm/radix: Move the functions that does the actual tlbie closer
powerpc/mm/radix: Remove unused code
powerpc/mm: Workaround Nest MMU bug with TLB invalidations
powerpc/mm: Add tracking of the number of coprocessors using a context
powerpc/64s: Fix lost pending interrupt due to race causing lost update to irq_happened
powerpc/64s: Fix NULL AT_BASE_PLATFORM when using DT CPU features
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC fixes from Arnd Bergmann:
"Here are are a couple of last-minute fixes for 4.16, mostly for
regressions. As usual, the majory are device tree changes:
- USB 3 support on rk3399 didn't work and is being reverted for now
- One fix for an old suspend/resume bug on rk3399
- A few regulator related fixes on Banana Pi M2, and on imx7d-sdb
- A boot regression fix for all Aspeed SoCs failing to find their
memory
- One more dtc warning fix
The other changes are:
- A few updates to the MAINTAINERS file
- A revert for an incorrect orion5x cleanup
- Two power management fixes for OMAP"
* tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
ARM: OMAP: Fix SRAM W+X mapping
ARM: dts: aspeed: Add default memory node
mailmap: Update email address for Gregory CLEMENT
ARM: davinci: fix the GPIO lookup for omapl138-hawk
MAINTAINERS: Update Tegra IOMMU maintainer
ARM: dts: imx7d-sdb: Fix regulator-usb-otg2-vbus node name
ARM: ux500: Fix PMU IRQ regression
ARM: dts: rockchip: Add missing #sound-dai-cells on rk3288
Revert "arm64: dts: rockchip: add usb3-phy otg-port support for rk3399"
arm64: dts: rockchip: Fix rk3399-gru-* s2r (pinctrl hogs, wifi reset)
ARM: OMAP: Fix dmtimer init for omap1
MAINTAINERS: update email address for Maxime Ripard
ARM: dts: sun6i: a31s: bpi-m2: add missing regulators
ARM: dts: sun6i: a31s: bpi-m2: improve pmic properties
|
|
A critical error was found testing the fixed UV4 HUB in that an MMR address
was found to be incorrect. This causes the virtual address space for
accessing the MMIOH1 region to be allocated with the incorrect size.
Fixes: 673aa20c55a1 ("x86/platform/UV: Update uv_mmrs.h to prepare for UV4A fixes")
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Dimitri Sivanich <dimitri.sivanich@hpe.com>
Cc: Russ Anderson <russ.anderson@hpe.com>
Cc: Andrew Banman <andrew.banman@hpe.com>
Link: https://lkml.kernel.org/r/20180328174011.041801248@stormcage.americas.sgi.com
|
|
PV TLB FLUSH can only be turned on when steal time is enabled.
The condition got reversed during conflict resolution.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Fixes: 4f2f61fc5071 ("KVM: X86: Avoid traversing all the cpus for pv tlb flush when steal time is disabled")
[Rebased on top of kvm/master and reworded the commit message. - Radim]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
|
|
BAU uses the old alloc_initr_gate90 method to setup its interrupt. This
fails silently as the BAU vector is in the range of APIC vectors that are
registered to the spurious interrupt handler. As a consequence BAU
broadcasts are not handled, and the broadcast source CPU hangs.
Update BAU to use new idt structure.
Fixes: dc20b2d52653 ("x86/idt: Move interrupt gate initialization to IDT code")
Signed-off-by: Andrew Banman <abanman@hpe.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Mike Travis <mike.travis@hpe.com>
Cc: Dimitri Sivanich <sivanich@hpe.com>
Cc: Russ Anderson <rja@hpe.com>
Cc: stable@vger.kernel.org
Cc: "H. Peter Anvin" <hpa@zytor.com>
Link: https://lkml.kernel.org/r/1522188546-196177-1-git-send-email-abanman@hpe.com
|
|
Pull ARM fixes from Russell King:
"A small number of small fixes for ARM, mostly for some build issues.
One fix for a regression caused by the cpu hotplug conversion from a
few kernel versions ago"
* 'fixes' of git://git.armlinux.org.uk/~rmk/linux-arm:
ARM: 8750/1: deflate_xip_data.sh: minor fixes
ARM: 8748/1: mm: Define vdso_start, vdso_end as array
ARM: 8747/1: make CONFIG_DEBUG_WX depend on MMU
ARM: 8746/1: vfp: Go back to clearing vfp_current_hw_state[]
|
|
ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into fixes
Pull "Allwinner Fixes for 4.16" from Maxime Ripard:
The first and second patches fix the regulator support for the Bananapi M2
board.
The last one updates my email address in MAINTAINERS.
* tag 'sunxi-fixes-for-4.16' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
MAINTAINERS: update email address for Maxime Ripard
ARM: dts: sun6i: a31s: bpi-m2: add missing regulators
ARM: dts: sun6i: a31s: bpi-m2: improve pmic properties
|
|
ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into fixes
Pull "Two fixes for omap variants for v4.16-rc cycle" from Tony Lindgren:
Fix insecure W+X mapping warning for SRAM for omaps that
don't yet use drivers/misc/*sram*.c code. An earlier attempt
at fixing this turned out to cause problems with PM on omap3,
this version works with PM on omap3.
Also fix dmtimer probe for omap16xx devices that was noticed
with the pending dmtimer move to drivers. It seems this has
been broken for a while and is a non-critical for booting.
It is needed for PM on omap16xx though.
* tag 'omap-for-v4.16/sram-fix-signed' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
ARM: OMAP: Fix SRAM W+X mapping
ARM: OMAP: Fix dmtimer init for omap1
|
|
The following pattern fails to compile while the same pattern
with alternative_call() does:
if (...)
alternative_call_2(...);
else
alternative_call_2(...);
as it expands into
if (...)
{
}; <===
else
{
};
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20180114120504.GA11368@avx2
|
|
this patch fix a bug in how the pebs->real_ip is handled in the PEBS
handler. real_ip only exists in Haswell and later processor. It is
actually the eventing IP, i.e., where the event occurred. As opposed
to the pebs->ip which is the PEBS interrupt IP which is always off
by one.
The problem is that the real_ip just like the IP needs to be fixed up
because PEBS does not record all the machine state registers, and
in particular the code segement (cs). This is why we have the set_linear_ip()
function. The problem was that set_linear_ip() was only used on the pebs->ip
and not the pebs->real_ip.
We have profiles which ran into invalid callstacks because of this.
Here is an example:
..... 0: ffffffffffffff80 recent entry, marker kernel v
..... 1: 000000000040044d <= user address in kernel space!
..... 2: fffffffffffffe00 marker enter user v
..... 3: 000000000040044d
..... 4: 00000000004004b6 oldest entry
Debugging output in get_perf_callchain():
[ 857.769909] CALLCHAIN: CPU8 ip=40044d regs->cs=10 user_mode(regs)=0
The problem is that the kernel entry in 1: points to a user level
address. How can that be?
The reason is that with PEBS sampling the instruction that caused the event
to occur and the instruction where the CPU was when the interrupt was posted
may be far apart. And sometime during that time window, the privilege level may
change. This happens, for instance, when the PEBS sample is taken close to a
kernel entry point. Here PEBS, eventing IP (real_ip) captured a user level
instruction. But by the time the PMU interrupt fired, the processor had already
entered kernel space. This is why the debug output shows a user address with
user_mode() false.
The problem comes from PEBS not recording the code segment (cs) register.
The register is used in x86_64 to determine if executing in kernel vs user
space. This is okay because the kernel has a software workaround called
set_linear_ip(). But the issue in setup_pebs_sample_data() is that
set_linear_ip() is never called on the real_ip value when it is available
(Haswell and later) and precise_ip > 1.
This patch fixes this problem and eliminates the callchain discrepancy.
The patch restructures the code around set_linear_ip() to minimize the number
of times the IP has to be set.
Signed-off-by: Stephane Eranian <eranian@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: kan.liang@intel.com
Link: http://lkml.kernel.org/r/1521788507-10231-1-git-send-email-eranian@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
The SLB bad address handler's trap number fixup does not preserve the
low bit that indicates nonvolatile GPRs have not been saved. This
leads save_nvgprs to skip saving them, and subsequent functions and
return from interrupt will think they are saved.
This causes kernel branch-to-garbage debugging to not have correct
registers, can also cause userspace to have its registers clobbered
after a segfault.
Fixes: f0f558b131db ("powerpc/mm: Preserve CFAR value on SLB miss caused by access to bogus address")
Cc: stable@vger.kernel.org # v4.9+
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 and PTI fixes from Ingo Molnar:
"Misc fixes:
- fix EFI pagetables freeing
- fix vsyscall pagetable setting on Xen PV guests
- remove ancient CONFIG_X86_PPRO_FENCE=y - x86 is TSO again
- fix two binutils (ld) development version related incompatibilities
- clean up breakpoint handling
- fix an x86 self-test"
* 'x86-pti-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/entry/64: Don't use IST entry for #BP stack
x86/efi: Free efi_pgd with free_pages()
x86/vsyscall/64: Use proper accessor to update P4D entry
x86/cpu: Remove the CONFIG_X86_PPRO_FENCE=y quirk
x86/boot/64: Verify alignment of the LOAD segment
x86/build/64: Force the linker to use 2MB page size
selftests/x86/ptrace_syscall: Fix for yet more glibc interference
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"Misc kernel side fixes.
Generic:
- cgroup events counting fix
x86:
- Intel PMU truncated-parameter fix
- RDPMC fix
- API naming fix/rename
- uncore driver big-hardware PCI enumeration fix
- uncore driver filter constraint fix"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/cgroup: Fix child event counting bug
perf/x86/intel/uncore: Fix multi-domain PCI CHA enumeration bug on Skylake servers
perf/x86/intel: Rename confusing 'freerunning PEBS' API and implementation to 'large PEBS'
perf/x86/intel/uncore: Add missing filter constraint for SKX CHA event
perf/x86/intel: Don't accidentally clear high bits in bdw_limit_period()
perf/x86/intel: Disable userspace RDPMC usage for large PEBS
|
|
KBUILD_CFLAGS
The kernel build system already takes care of generating the dependency
files. Having the additional -MD in KBUILD_CFLAGS leads to stray
.<pid>.d files in the build directory when we call the cc-option macro.
Signed-off-by: Sven Wegener <sven.wegener@stealer.net>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthias Kaehlcke <mka@chromium.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vivek Goyal <vgoyal@redhat.com>
Link: http://lkml.kernel.org/r/alpine.LNX.2.21.1803242219380.30139@titan.int.lan.stealer.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Send nm complaints about broken pipe (when sed exits early) to /dev/null.
All errors should be printed to stderr.
Don't trap on normal exit so the trap can return an error code.
Signed-off-by: Nicolas Pitre <nico@linaro.org>
Tested-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
|
|
Define vdso_start, vdso_end as array to avoid compile-time analysis error
for the case of built with CONFIG_FORTIFY_SOURCE.
and, since vdso_start, vdso_end are used in vdso.c only,
move extern-declaration from vdso.h to vdso.c.
If kernel is built with CONFIG_FORTIFY_SOURCE,
compile-time error happens at this code.
- if (memcmp(&vdso_start, "177ELF", 4))
The size of "&vdso_start" is recognized as 1 byte, but n is 4,
So that compile-time error is reported.
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Jinbum Park <jinb.park7@gmail.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
|
|
Without CONFIG_MMU, this results in a build failure:
./arch/arm/include/asm/memory.h:92:23: error: initializer element is not constant
#define VECTORS_BASE vectors_base
arch/arm/mm/dump.c:32:4: note: in expansion of macro 'VECTORS_BASE'
{ VECTORS_BASE, "Vectors" },
arch/arm/mm/dump.c:71:11: error: 'L_PTE_USER' undeclared here (not in a function); did you mean 'VTIME_USER'?
.mask = L_PTE_USER,
^~~~~~~~~~
Obviously the feature only makes sense with an MMU, so let's add the
dependency here.
Fixes: a8e53c151fe7 ("ARM: 8737/1: mm: dump: add checking for writable and executable")
Acked-by: Laura Abbott <labbott@redhat.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
|
|
Commit 384b38b66947 ("ARM: 7873/1: vfp: clear vfp_current_hw_state
for dying cpu") fixed the cpu dying notifier by clearing
vfp_current_hw_state[]. However commit e5b61bafe704 ("arm: Convert VFP
hotplug notifiers to state machine") incorrectly used the original
vfp_force_reload() function in the cpu dying notifier.
Fix it by going back to clearing vfp_current_hw_state[].
Fixes: e5b61bafe704 ("arm: Convert VFP hotplug notifiers to state machine")
Cc: linux-stable <stable@vger.kernel.org>
Reported-by: Kohji Okuno <okuno.kohji@jp.panasonic.com>
Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
|
|
There's nothing IST-worthy about #BP/int3. We don't allow kprobes
in the small handful of places in the kernel that run at CPL0 with
an invalid stack, and 32-bit kernels have used normal interrupt
gates for #BP forever.
Furthermore, we don't allow kprobes in places that have usergs while
in kernel mode, so "paranoid" is also unnecessary.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
|
|
The efi_pgd is allocated as PGD_ALLOCATION_ORDER pages and therefore must
also be freed as PGD_ALLOCATION_ORDER pages with free_pages().
Fixes: d9e9a6418065 ("x86/mm/pti: Allocate a separate user PGD")
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-efi@vger.kernel.org
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/1521746333-19593-1-git-send-email-longman@redhat.com
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips
Pull MIPS fixes from James Hogan:
"Another miscellaneous pile of MIPS fixes for 4.16:
- lantiq: fixes for clocks and Amazon SE (4.14)
- ralink: fix booting on MT7621 (4.5)
- ralink: fix halt (3.9)"
* tag 'mips_fixes_4.16_5' of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips:
MIPS: ralink: Fix booting on MT7621
MIPS: ralink: Remove ralink_halt()
MIPS: lantiq: ase: Enable MFD_SYSCON
MIPS: lantiq: Enable AHB Bus for USB
MIPS: lantiq: Fix Danube USB clock
|
|
Segment registers must be synchronized prior to any code that may
trigger a call to emulation_required()/guest_state_valid(), e.g.
vmx_set_cr0(). Because preparing vmcs02 writes segmentation fields
directly, i.e. doesn't use vmx_set_segment(), emulation_required
will not be re-evaluated when synchronizing the segment registers,
which can result in L0 incorrectly starting emulation of L2.
Fixes: 8665c3f97320 ("KVM: nVMX: initialize descriptor cache fields in prepare_vmcs02_full")
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
[Move all of prepare_vmcs02_full earlier, not just segment registers. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc into kvm-master
PPC KVM fix -
Fix a bug causing occasional machine check exceptions on POWER8 hosts,
introduced in 4.16-rc1.
|
|
On POWER9, under some circumstances, a broadcast TLB invalidation
might complete before all previous stores have drained, potentially
allowing stale stores from becoming visible after the invalidation.
This works around it by doubling up those TLB invalidations which was
verified by HW to be sufficient to close the risk window.
This will be documented in a yet-to-be-published errata.
Fixes: 1a472c9dba6b ("powerpc/mm/radix: Add tlbflush routines")
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
[mpe: Enable the feature in the DT CPU features code for all Power9,
rename the feature to CPU_FTR_P9_TLBIE_BUG per benh.]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
|
