Fix this link error:
ERROR: modpost: "rcu_idle_enter" [drivers/acpi/processor.ko] undefined!
ERROR: modpost: "rcu_idle_exit" [drivers/acpi/processor.ko] undefined!
when CONFIG_ACPI_PROCESSOR is built as module. PeterZ says that in light
of ARM needing those soon too, they should simply be exported.
Fixes: 1fecfdbb7a ("ACPI: processor: Take over RCU-idle for C3-BM idle")
Reported-by: Sven Joachim <[email protected]>
Suggested-by: Peter Zijlstra <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Paul E. McKenney <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
Commit
b972fdba86 ("EDAC/ghes: Fix NULL pointer dereference in ghes_edac_register()")
didn't clear all the information from the scanned system and, more
specifically, left ghes_hw.num_dimms to its previous value. On a
second load (CONFIG_DEBUG_TEST_DRIVER_REMOVE=y), the driver would use
the leftover num_dimms value which is not 0 and thus the 0 check in
enumerate_dimms() will get bypassed and it would go directly to the
pointer deref:
d = &hw->dimms[hw->num_dimms];
which is, of course, NULL:
#PF: supervisor write access in kernel mode
#PF: error_code(0x0002) - not-present page
PGD 0 P4D 0
Oops: 0002 [#1] PREEMPT SMP
CPU: 7 PID: 1 Comm: swapper/0 Not tainted 5.9.0-rc4+ #7
Hardware name: GIGABYTE MZ01-CE1-00/MZ01-CE1-00, BIOS F02 08/29/2018
RIP: 0010:enumerate_dimms.cold+0x7b/0x375
Reset the whole ghes_hw on driver unregister so that no stale values are
used on a second system scan.
Fixes: b972fdba86 ("EDAC/ghes: Fix NULL pointer dereference in ghes_edac_register()")
Cc: Shiju Jose <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Add functionality to disable writing to MSRs from userspace. Writes can
still be allowed by supplying the allow_writes=on module parameter. The
kernel will be tainted so that it shows in oopses.
Having unfettered access to all MSRs on a system is and has always been
a disaster waiting to happen. Think performance counter MSRs, MSRs with
sticky or locked bits, MSRs making major system changes like loading
microcode, MTRRs, PAT configuration, TSC counter, security mitigations
MSRs, you name it.
This also destroys all the kernel's caching of MSR values for
performance, as the recent case with MSR_AMD64_LS_CFG showed.
Another example is writing MSRs by mistake by simply typing the wrong
MSR address. System freezes have been experienced that way.
In general, poking at MSRs under the kernel's feet is a bad bad idea.
So log writing to MSRs by default. Longer term, such writes will be
disabled by default.
If userspace still wants to do that, then proper interfaces should be
defined which are under the kernel's control and accesses to those MSRs
can be synchronized and sanitized properly.
[ Fix sparse warnings. ]
Reported-by: kernel test robot <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Tested-by: Sean Christopherson <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
1068ed4547 ("x86/msr: Lift AMD family 0x15 power-specific MSRs")
moved the three F15h power MSRs to the architectural list but that was
wrong as they belong in the family 0x15 list. That also caused:
In file included from trace/beauty/tracepoints/x86_msr.c:10:
perf/trace/beauty/generated/x86_arch_MSRs_array.c:292:45: error: initialized field overwritten [-Werror=override-init]
292 | [0xc0010280 - x86_AMD_V_KVM_MSRs_offset] = "F15H_PTSC",
| ^~~~~~~~~~~
perf/trace/beauty/generated/x86_arch_MSRs_array.c:292:45: note: (near initialization for 'x86_AMD_V_KVM_MSRs[640]')
due to MSR_F15H_PTSC ending up being defined twice. Move them where they
belong and drop the duplicate.
Also, drop the respective tools/ changes of the msr-index.h copy the
above commit added because perf tool developers prefer to go through
those changes themselves in order to figure out whether changes to the
kernel headers would need additional handling in perf.
Fixes: 1068ed4547 ("x86/msr: Lift AMD family 0x15 power-specific MSRs")
Reported-by: Stephen Rothwell <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Acked-by: Arnaldo Carvalho de Melo <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Commit:
da92110dfd ("EDAC, amd64_edac: Extend scrub rate support to F15hM60h")
added support for F15h, model 0x60 CPUs but in doing so, missed to read
back SCRCTRL PCI config register on F15h CPUs which are *not* model
0x60. Add that read so that doing
$ cat /sys/devices/system/edac/mc/mc0/sdram_scrub_rate
can show the previously set DRAM scrub rate.
Fixes: da92110dfd ("EDAC, amd64_edac: Extend scrub rate support to F15hM60h")
Reported-by: Anders Andersson <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: <[email protected]> #v4.4..
Link: https://lkml.kernel.org/r/CAKkunMbNWppx_i6xSdDHLseA2QQmGJqj_crY=NF-GZML5np4Vw@mail.gmail.com
Merge the test whether the CPU supports STIBP into the test which
determines whether STIBP is required. Thus try to simplify what is
already an insane logic.
Remove a superfluous newline in a comment, while at it.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Anthony Steinhauser <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Change the hardware scanning and figuring out how many DIMMs a machine
has to a single, one-time thing which happens once on driver init. After
that scanning completes, struct ghes_hw_desc contains a representation
of the hardware which the driver can then use for later initialization.
Then, copy the DIMM information into the respective EDAC core
representation of those.
Get rid of ghes_edac_dimm_fill and use a struct dimm_info array
directly.
This way, hw detection and further driver initialization is nicely
and logically split. Further additions should all be added to
ghes_scan_system() and the hw representation extended as needed.
There should be no functionality change resulting from this patch.
Signed-off-by: Borislav Petkov <[email protected]>
... into the global msr-index.h header because they're used in multiple
compilation units. Sort the MSR list a bit. Update the msr-index.h copy
in tools.
No functional changes.
Signed-off-by: Borislav Petkov <[email protected]>
Acked-by: Guenter Roeck <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
The commit
c84cb3735f ("x86/apic: Move TSC deadline timer debug printk")
removed the message which said that the deadline timer was enabled.
It added a pr_debug() message which is issued when deadline timer
validation succeeds.
Well, issued only when CONFIG_DYNAMIC_DEBUG is enabled - otherwise
pr_debug() calls get optimized away if DEBUG is not defined in the
compilation unit.
Therefore, make the above message pr_info() so that it is visible in
dmesg.
Signed-off-by: Borislav Petkov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Pull EFI fixes from Ard Biesheuvel:
"- fix EFI framebuffer earlycon for wide fonts
- avoid filling screen_info with garbage if the EFI framebuffer is not
available
- fix a potential host tool build error due to a symbol clash on x86
- work around a EFI firmware bug regarding the binary format of the TPM
final events table
- fix a missing memory free by reworking the E820 table sizing routine to
not do the allocation in the first place
- add CPER parsing for firmware errors"
... or the odyssey of trying to disable the stack protector for the
function which generates the stack canary value.
The whole story started with Sergei reporting a boot crash with a kernel
built with gcc-10:
Kernel panic — not syncing: stack-protector: Kernel stack is corrupted in: start_secondary
CPU: 1 PID: 0 Comm: swapper/1 Not tainted 5.6.0-rc5—00235—gfffb08b37df9 #139
Hardware name: Gigabyte Technology Co., Ltd. To be filled by O.E.M./H77M—D3H, BIOS F12 11/14/2013
Call Trace:
dump_stack
panic
? start_secondary
__stack_chk_fail
start_secondary
secondary_startup_64
-—-[ end Kernel panic — not syncing: stack—protector: Kernel stack is corrupted in: start_secondary
This happens because gcc-10 tail-call optimizes the last function call
in start_secondary() - cpu_startup_entry() - and thus emits a stack
canary check which fails because the canary value changes after the
boot_init_stack_canary() call.
To fix that, the initial attempt was to mark the one function which
generates the stack canary with:
__attribute__((optimize("-fno-stack-protector"))) ... start_secondary(void *unused)
however, using the optimize attribute doesn't work cumulatively
as the attribute does not add to but rather replaces previously
supplied optimization options - roughly all -fxxx options.
The key one among them being -fno-omit-frame-pointer and thus leading to
not present frame pointer - frame pointer which the kernel needs.
The next attempt to prevent compilers from tail-call optimizing
the last function call cpu_startup_entry(), shy of carving out
start_secondary() into a separate compilation unit and building it with
-fno-stack-protector, was to add an empty asm("").
This current solution was short and sweet, and reportedly, is supported
by both compilers but we didn't get very far this time: future (LTO?)
optimization passes could potentially eliminate this, which leads us
to the third attempt: having an actual memory barrier there which the
compiler cannot ignore or move around etc.
That should hold for a long time, but hey we said that about the other
two solutions too so...
Reported-by: Sergei Trofimovich <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Tested-by: Kalle Valo <[email protected]>
Cc: <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
... and get rid of the function pointers which would spit out the
microcode revision based on the CPU stepping.
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Mark Gross <mgross.linux.intel.com>
Cc: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
... to fix these build warnings:
In file included from ./arch/x86/include/asm/uv/uv_hub.h:22,
from drivers/misc/sgi-gru/grukdump.c:16:
./arch/x86/include/asm/uv/uv.h:39:21: warning: ‘struct flush_tlb_info’ declared \
inside parameter list will not be visible outside of this definition or declaration
39 | const struct flush_tlb_info *info);
| ^~~~~~~~~~~~~~
In file included from ./arch/x86/include/asm/uv/uv_hub.h:22,
from drivers/misc/sgi-gru/grutlbpurge.c:28:
./arch/x86/include/asm/uv/uv.h:39:21: warning: ‘struct flush_tlb_info’ declared \
inside parameter list will not be visible outside of this definition or declaration
39 | const struct flush_tlb_info *info);
| ^~~~~~~~~~~~~~
...
after
bfe3d8f631 ("x86/tlb: Restrict access to tlbstate")
restricted access to tlbstate.
Signed-off-by: Borislav Petkov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Fix:
arch/x86/kvm/svm/sev.c: In function ‘sev_pin_memory’:
arch/x86/kvm/svm/sev.c:360:3: error: implicit declaration of function ‘release_pages’;\
did you mean ‘reclaim_pages’? [-Werror=implicit-function-declaration]
360 | release_pages(pages, npinned);
| ^~~~~~~~~~~~~
| reclaim_pages
because svm.c includes pagemap.h but the carved out sev.c needs it too.
Triggered by a randconfig build.
Fixes: eaf78265a4 ("KVM: SVM: Move SEV code to separate file")
Signed-off-by: Borislav Petkov <[email protected]>
Message-Id: <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
Work around this warning:
kernel/sched/cputime.c: In function ‘kcpustat_field’:
kernel/sched/cputime.c:1007:6: warning: ‘val’ may be used uninitialized in this function [-Wmaybe-uninitialized]
because GCC can't see that val is used only when err is 0.
Acked-by: Peter Zijlstra <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
The severity grading code returns IN_KERNEL_RECOV error context for
errors which have happened in kernel space but from which the kernel can
recover. Whether the recovery can happen is determined by the exception
table entry having as handler ex_handler_fault() and which has been
declared at build time using _ASM_EXTABLE_FAULT().
IN_KERNEL_RECOV is used in mce_severity_intel() to lookup the
corresponding error severity in the severities table.
However, the mapping back from error severity to whether the error is
IN_KERNEL_RECOV is ambiguous and in the very paranoid case - which
might not be possible right now - but be better safe than sorry later,
an exception fixup could be attempted for another MCE whose address
is in the exception table and has the proper severity. Which would be
unfortunate, to say the least.
Therefore, mark such MCEs explicitly as MCE_IN_KERNEL_RECOV so that the
recovery attempt is done only for them.
Document the whole handling, while at it, as it is not trivial.
Reported-by: Thomas Gleixner <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Tested-by: Tony Luck <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Make the doublefault exception handler unconditional on 32-bit. Yes,
it is important to be able to catch #DF exceptions instead of silent
reboots. Yes, the code size increase is worth every byte. And one less
CONFIG symbol is just the cherry on top.
No functional changes.
Signed-off-by: Borislav Petkov <[email protected]>
Acked-by: Andy Lutomirski <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
The currently minimum-supported binutils version 2.21 has the problem of
promoting symbols which are defined outside of a section into absolute.
According to Arvind:
binutils-2.21 and -2.22. An x86-64 defconfig will fail with
Invalid absolute R_X86_64_32S relocation: _etext
and after fixing that one, with
Invalid absolute R_X86_64_32S relocation: __end_of_kernel_reserve
Those two versions of binutils have a bug when it comes to handling
symbols defined outside of a section and binutils 2.23 has the proper
fix, see: https://sourceware.org/legacy-ml/binutils/2012-06/msg00155.html
Therefore, up to the fixed version directly, skipping the broken ones.
Currently shipping distros already have the fixed binutils version so
there should be no breakage resulting from this.
For more details about the whole thing, see the thread in Link.
Signed-off-by: Borislav Petkov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Acked-by: Kees Cook <[email protected]>
Acked-by: Jason A. Donenfeld <[email protected]>
Acked-by: Nick Desaulniers <[email protected]>
Acked-by: Ingo Molnar <[email protected]>
Signed-off-by: Masahiro Yamada <[email protected]>
In order to use efi_mem_type(), one needs CONFIG_EFI enabled. Otherwise
that function is undefined. Use IS_ENABLED() to check and avoid the
ifdeffery as the compiler optimizes away the following unreachable code
then.
Fixes: 985e537a40 ("x86/ioremap: Map EFI runtime services data as encrypted for SEV")
Reported-by: Randy Dunlap <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Tested-by: Randy Dunlap <[email protected]>
Cc: Tom Lendacky <[email protected]>
Cc: <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
threshold_create_bank() creates a bank descriptor per MCA error
thresholding counter which can be controlled over sysfs. It publishes
the pointer to that bank in a per-CPU variable and then goes on to
create additional thresholding blocks if the bank has such.
However, that creation of additional blocks in
allocate_threshold_blocks() can fail, leading to a use-after-free
through the per-CPU pointer.
Therefore, publish that pointer only after all blocks have been setup
successfully.
Fixes: 019f34fccf ("x86, MCE, AMD: Move shared bank to node descriptor")
Reported-by: Saar Amar <[email protected]>
Reported-by: Dan Carpenter <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
... and fold its function body into its single call site.
No functional changes:
# arch/x86/kernel/cpu/amd.o:
text data bss dec hex filename
5994 385 1 6380 18ec amd.o.before
5994 385 1 6380 18ec amd.o.after
md5:
99ec6daa095b502297884e949c520f90 amd.o.before.asm
99ec6daa095b502297884e949c520f90 amd.o.after.asm
Signed-off-by: Borislav Petkov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
On machines which do not populate all nodes with DIMMs, the driver
doesn't initialize an instance there. However, the instance removal
remove_one_instance() path will warn unconditionally, which is wrong.
Remove the WARN_ON() even if the warning is innocent because it causes a
splat in dmesg.
Signed-off-by: Borislav Petkov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Hoist the user_mode() case up because it is less code and can be dealt
with up-front like the other special cases UMIP and vm86.
This saves an indentation level for the kernel-mode #GP case and allows
to "unfold" the code more so that it is more readable.
No functional changes.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Jann Horn <[email protected]>
Cc: [email protected]
There was a stray tab in the help text of the aforementioned config
option which showed like this:
The "print fmt" of the trace events will show the enum/sizeof names
instead of their values. This can cause problems for user space tools
...
in menuconfig. Remove it and end a sentence with a fullstop.
No functional changes.
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Borislav Petkov <[email protected]>
Signed-off-by: Steven Rostedt (VMware) <[email protected]>
This message keeps flooding dmesg on boxes where ECC is disabled or the
DIMMs do not support ECC but the module gets auto-probed. What's even
worse is that autoprobing happens on every CPU due to the CPU-family
matching the driver does and uevent being generated for each CPU device.
What is more, this message is becoming even more useless on newer
systems where forcing ECC is not recommended and it should be done in
the BIOS so the BIOS can do all the necessary work, i.e., just setting a
bit in an MSR is not enough anymore.
So get rid of it.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Yazen Ghannam <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
History lesson courtesy of Steve:
"When ftrace first was introduced to the kernel, it used gcc's
mcount profiling mechanism. The mcount mechanism would add a call to
"mcount" at the start of every function but after the stack frame was
set up. Later, in gcc 4.6, gcc introduced -mfentry, that would create a
call to "__fentry__" instead of "mcount", before the stack frame was
set up. In order to handle both cases, ftrace defined a macro
"function_hook" that would be either "mcount" or "__fentry__" depending
on which one was being used.
The Linux kernel no longer supports the "mcount" method, thus there's
no reason to keep the "function_hook" define around. Simply use
"__fentry__", as there is no ambiguity to the name anymore."
Drop it everywhere.
Signed-off-by: Borislav Petkov <[email protected]>
Acked-by: Jiri Slaby <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Cc: Josh Poimboeuf <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Peter Zijlstra <[email protected]>
Cc: "Steven Rostedt (VMware)" <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: [email protected]
Link: http://lkml.kernel.org/r/[email protected]
The dummy variable is the high part of the microcode revision MSR which
is defined as reserved. Mark it unused so that W=1 builds don't trigger
the above warning.
No functional changes.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fix
arch/x86/kernel/apic/probe_32.c: In function ‘default_setup_apic_routing’:
arch/x86/kernel/apic/probe_32.c:146:7: warning: this statement may fall through [-Wimplicit-fallthrough=]
if (!APIC_XAPIC(version)) {
^
arch/x86/kernel/apic/probe_32.c:151:3: note: here
case X86_VENDOR_HYGON:
^~~~
for 32-bit builds.
Signed-off-by: Borislav Petkov <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
... into a separate function for better readability. Split out from a
patch from Fenghua Yu <[email protected]> to keep the mechanical,
sole code movement separate for easy review.
No functional changes.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Fenghua Yu <[email protected]>
Cc: [email protected]
Adric Blake reported the following warning during suspend-resume:
Enabling non-boot CPUs ...
x86: Booting SMP configuration:
smpboot: Booting Node 0 Processor 1 APIC 0x2
unchecked MSR access error: WRMSR to 0x10f (tried to write 0x0000000000000000) \
at rIP: 0xffffffff8d267924 (native_write_msr+0x4/0x20)
Call Trace:
intel_set_tfa
intel_pmu_cpu_starting
? x86_pmu_dead_cpu
x86_pmu_starting_cpu
cpuhp_invoke_callback
? _raw_spin_lock_irqsave
notify_cpu_starting
start_secondary
secondary_startup_64
microcode: sig=0x806ea, pf=0x80, revision=0x96
microcode: updated to revision 0xb4, date = 2019-04-01
CPU1 is up
The MSR in question is MSR_TFA_RTM_FORCE_ABORT and that MSR is emulated
by microcode. The log above shows that the microcode loader callback
happens after the PMU restoration, leading to the conjecture that
because the microcode hasn't been updated yet, that MSR is not present
yet, leading to the #GP.
Add a microcode loader-specific hotplug vector which comes before
the PERF vectors and thus executes earlier and makes sure the MSR is
present.
Fixes: 400816f60c ("perf/x86/intel: Implement support for TSX Force Abort")
Reported-by: Adric Blake <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: <[email protected]>
Cc: [email protected]
Link: https://bugzilla.kernel.org/show_bug.cgi?id=203637
When dumping the array elements, print them in the following format:
[ PFN | generation in binary | count ]
to be perfectly clear what all those sections are.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: linux-edac <[email protected]>
... which is the better, more-fitting name anyway.
Tony:
- make action_threshold u64 due to debugfs accessors expecting u64.
- rename the remaining: s/count_threshold/action_threshold/g
Co-developed-by: Tony Luck <[email protected]>
Signed-off-by: Tony Luck <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: linux-edac <[email protected]>
When the value requested doesn't match the allowed (min,max) range,
the @data buffer should not be modified with the invalid value because
reading "decay_interval" shows it otherwise as if the previous write
succeeded.
Move the data write after the check.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: linux-edac <[email protected]>
The count_threshold should be checked unconditionally, after insertion
too, so that a count_threshold value of 1 can cause an immediate
offlining. I.e., offline the page on the *first* error encountered.
Add comments to make it clear what cec_add_elem() does, while at it.
Reported-by: WANG Chao <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
When inserting random PFNs for debugging the CEC through
(debugfs)/ras/cec/pfn, depending on the return value of pfn_set(),
multiple values get inserted per a single write.
That is because simple_attr_write() interprets a retval of 0 as
success and claims the whole input. However, pfn_set() returns the
cec_add_elem() value, which, if > 0 and smaller than the whole input
length, makes glibc continue issuing the write syscall until there's
input left:
pfn_set
simple_attr_write
debugfs_attr_write
full_proxy_write
vfs_write
ksys_write
do_syscall_64
entry_SYSCALL_64_after_hwframe
leading to those repeated calls.
Return 0 to fix that.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: linux-edac <[email protected]>
Switch to using Donald Knuth's binary search algorithm (The Art of
Computer Programming, vol. 3, section 6.2.1). This should've been done
from the very beginning but the author must've been smoking something
very potent at the time.
The problem with the current one was that it would return the wrong
element index in certain situations:
https://lkml.kernel.org/r/CAM_iQpVd02zkVJ846cj-Fg1yUNuz6tY5q1Vpj4LrXmE06dPYYg@mail.gmail.com
and the noodling code after the loop was fishy at best.
So switch to using Knuth's binary search. The final result is much
cleaner and straightforward.
Fixes: 011d826111 ("RAS: Add a Corrected Errors Collector")
Reported-by: Cong Wang <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: linux-edac <[email protected]>
Cc: <[email protected]>
After commit:
672ff6cff8 ("KVM: x86: Raise #GP when guest vCPU do not support PMU")
my AMD guests started #GPing like this:
general protection fault: 0000 [#1] PREEMPT SMP
CPU: 1 PID: 4355 Comm: bash Not tainted 5.1.0-rc6+ #3
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
RIP: 0010:x86_perf_event_update+0x3b/0xa0
with Code: pointing to RDPMC. It is RDPMC because the guest has the
hardware watchdog CONFIG_HARDLOCKUP_DETECTOR_PERF enabled which uses
perf. Instrumenting kvm_pmu_rdpmc() some, showed that it fails due to:
if (!pmu->version)
return 1;
which the above commit added. Since AMD's PMU leaves the version at 0,
that causes the #GP injection into the guest.
Set pmu->version arbitrarily to 1 and move it above the non-applicable
struct kvm_pmu members.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Janakarajan Natarajan <[email protected]>
Cc: [email protected]
Cc: Liran Alon <[email protected]>
Cc: Mihai Carabas <[email protected]>
Cc: Paolo Bonzini <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: "Radim Krčmář" <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Tom Lendacky <[email protected]>
Cc: [email protected]
Cc: [email protected]
Fixes: 672ff6cff8 ("KVM: x86: Raise #GP when guest vCPU do not support PMU")
Signed-off-by: Paolo Bonzini <[email protected]>
The hardware configuration register has some useful bits which can be
used by guests. Implement McStatusWrEn which can be used by guests when
injecting MCEs with the in-kernel mce-inject module.
For that, we need to set bit 18 - McStatusWrEn - first, before writing
the MCi_STATUS registers (otherwise we #GP).
Add the required machinery to do so.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Jim Mattson <[email protected]>
Cc: Joerg Roedel <[email protected]>
Cc: KVM <[email protected]>
Cc: Paolo Bonzini <[email protected]>
Cc: Radim Krčmář <[email protected]>
Cc: Sean Christopherson <[email protected]>
Cc: Tom Lendacky <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: Yazen Ghannam <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
It is not absolutely clear from the docs how the cleanup path after
device_add() should look like so spell it out explicitly.
No functional changes, just documentation.
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Rafael J. Wysocki <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The original intention to move RDSP parsing very early, before KASLR
does its ranges selection, was to accommodate movable memory regions
machines (CONFIG_MEMORY_HOTREMOVE) to still be able to do memory
hotplug.
However, that broke kexec'ing a kernel on EFI machines because depending
on where the EFI systab was mapped, on at least one machine it isn't
present in the kexec mapping of the second kernel, leading to a triple
fault in the early code.
Fixing this properly requires significantly involved surgery and we
cannot allow ourselves to do that, that close to the merge window.
So disable the RSDP parsing code temporarily until it is fixed properly
in the next release cycle.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Ard Biesheuvel <[email protected]>
Cc: Baoquan He <[email protected]>
Cc: Chao Fan <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: [email protected]
Cc: Ingo Molnar <[email protected]>
Cc: Juergen Gross <[email protected]>
Cc: [email protected]
Cc: Kees Cook <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: [email protected]
Cc: Thomas Gleixner <[email protected]>
Cc: Tom Lendacky <[email protected]>
Cc: x86-ml <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
So we are going to be staring at those in the next years, let's make
them more succinct. In particular:
- change "address = " to "address: "
- "-privileged" reads funny. It should be simply "kernel" or "user"
- "from kernel code" reads funny too. "kernel mode" or "user mode" is
more natural.
An actual example says more than 1000 words, of course:
[ 0.248370] BUG: kernel NULL pointer dereference, address: 00000000000005b8
[ 0.249120] #PF: supervisor write access in kernel mode
[ 0.249717] #PF: error_code(0x0002) - not-present page
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
This is the ancient loading interface which requires special tools to be
used. The bigger problem with it is that it is as inadequate for proper
loading of microcode as the late microcode loading interface is because
it happens just as late.
iucode_tool's manpage already warns people that it is deprecated.
Deprecate it and turn it off by default along with a big fat warning in
the Kconfig help. It will go away sometime in the future.
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
Commit
2613f36ed9 ("x86/microcode: Attempt late loading only when new microcode is present")
added the new define UCODE_NEW to denote that an update should happen
only when newer microcode (than installed on the system) has been found.
But it missed adjusting that for the old /dev/cpu/microcode loading
interface. Fix it.
Fixes: 2613f36ed9 ("x86/microcode: Attempt late loading only when new microcode is present")
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Cc: Jann Horn <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Using static_cpu_has() is pointless on those paths, convert them to the
boot_cpu_has() variant.
No functional changes.
Signed-off-by: Borislav Petkov <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
This driver is Intel-only so loading on anything which is not Intel is
pointless. Prevent it from doing so.
While at it, correct the "not supported" print statement to say CPU
"model" which is what that test does.
Fixes: 076b862c7e (cpufreq: intel_pstate: Add reasons for failure and debug messages)
Suggested-by: Erwan Velu <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Thomas Renninger <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
__pure is used to make gcc do Common Subexpression Elimination (CSE)
and thus save subsequent invocations of a function which does a complex
computation (without side effects). As a simple example:
bool a = _static_cpu_has(x);
bool b = _static_cpu_has(x);
gets turned into
bool a = _static_cpu_has(x);
bool b = a;
However, gcc doesn't do CSE with asm()s when those get inlined - like it
is done with _static_cpu_has() - because, for example, the t_yes/t_no
labels are different for each inlined function body and thus cannot be
detected as equivalent anymore for the CSE heuristic to hit.
However, this all is beside the point because best it should be avoided
to have more than one call to _static_cpu_has(X) in the same function
due to the fact that each such call is an alternatives patch site and it
is simply pointless.
Therefore, drop the __pure attribute as it is not doing anything.
Reported-by: Nadav Amit <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
Calling this function has been wrong for a while now:
* Can't call schedule_work() in #MC context.
* mce_notify_irq() either.
* None of that noodling is needed anymore - all it needs to do is kick
the IRQ work which would self-IPI so that once the #MC handler is done,
the work queue will run and process queued MCE records.
So remove it.
Reported-by: Peter Zijlstra <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Tony Luck <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
dump_thread32() in aout_core_dump() does not clear the user32 structure
allocated on the stack as the first thing on function entry.
As a result, the dump.u_comm, dump.u_ar0 and dump.signal which get
assigned before the clearing, get overwritten.
Rename that function to fill_dump() to make it clear what it does and
call it first thing.
This was caught while staring at a patch by Derek Robson
<[email protected]>.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Derek Robson <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Michael Matz <[email protected]>
Cc: [email protected]
Cc: <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
When building randconfigs, one of the failures is:
ld: arch/x86/boot/compressed/kaslr.o: in function `choose_random_location':
kaslr.c:(.text+0xbf7): undefined reference to `count_immovable_mem_regions'
ld: kaslr.c:(.text+0xcbe): undefined reference to `immovable_mem'
make[2]: *** [arch/x86/boot/compressed/vmlinux] Error 1
because CONFIG_ACPI is not enabled in this particular .config but
CONFIG_MEMORY_HOTREMOVE is and count_immovable_mem_regions() is
unresolvable because it is defined in compressed/acpi.c which is the
compilation unit that depends on CONFIG_ACPI.
Add CONFIG_ACPI to the explicit dependencies for MEMORY_HOTREMOVE.
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Chao Fan <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
ac09c5f43c ("x86/boot: Build the command line parsing code unconditionally")
enabled building the command line parsing code unconditionally but it
forgot to remove the respective ifdeffery around the prototypes in the
misc.h header, leading to
arch/x86/boot/compressed/acpi.c: In function ‘get_acpi_rsdp’:
arch/x86/boot/compressed/acpi.c:37:8: warning: implicit declaration of function \
‘cmdline_find_option’ [-Wimplicit-function-declaration]
ret = cmdline_find_option("acpi_rsdp", val, MAX_ADDR_LEN);
^~~~~~~~~~~~~~~~~~~
for configs where neither CONFIG_EARLY_PRINTK nor CONFIG_RANDOMIZE_BASE
was defined.
Drop the ifdeffery in the header too.
Fixes: ac09c5f43c ("x86/boot: Build the command line parsing code unconditionally")
Reported-by: kbuild test robot <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Cc: Chao Fan <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/5c51daf0.83pQEkvDZILqoSYW%[email protected]
Link: https://lkml.kernel.org/r/[email protected]
... to pick up dependent change:
00ae831dfe ("x86/cpu: Add Atom Tremont (Jacobsville)")
introducing the model number define which will be needed by the new
i10nm_edac driver for 10nm Intel Atoms.
Signed-off-by: Borislav Petkov <[email protected]>
... instead of twice in the code. In any case, CR0 ends up being read
once anyway:
1. The CONFIG_MATH_EMULATION case does so and exits.
2. The normal case does it once too.
However, read it on function entry instead to make the code even simpler
to follow.
No functional changes.
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Andy Lutomirski <[email protected]>
Acked-by: Sebastian Andrzej Siewior <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]
Borislav Petkov