IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
Pull core x86 updates from Ingo Molnar:
- The biggest change is the rework of the percpu code, to support the
'Named Address Spaces' GCC feature, by Uros Bizjak:
- This allows C code to access GS and FS segment relative memory
via variables declared with such attributes, which allows the
compiler to better optimize those accesses than the previous
inline assembly code.
- The series also includes a number of micro-optimizations for
various percpu access methods, plus a number of cleanups of %gs
accesses in assembly code.
- These changes have been exposed to linux-next testing for the
last ~5 months, with no known regressions in this area.
- Fix/clean up __switch_to()'s broken but accidentally working handling
of FPU switching - which also generates better code
- Propagate more RIP-relative addressing in assembly code, to generate
slightly better code
- Rework the CPU mitigations Kconfig space to be less idiosyncratic, to
make it easier for distros to follow & maintain these options
- Rework the x86 idle code to cure RCU violations and to clean up the
logic
- Clean up the vDSO Makefile logic
- Misc cleanups and fixes
* tag 'x86-core-2024-03-11' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (52 commits)
x86/idle: Select idle routine only once
x86/idle: Let prefer_mwait_c1_over_halt() return bool
x86/idle: Cleanup idle_setup()
x86/idle: Clean up idle selection
x86/idle: Sanitize X86_BUG_AMD_E400 handling
sched/idle: Conditionally handle tick broadcast in default_idle_call()
x86: Increase brk randomness entropy for 64-bit systems
x86/vdso: Move vDSO to mmap region
x86/vdso/kbuild: Group non-standard build attributes and primary object file rules together
x86/vdso: Fix rethunk patching for vdso-image-{32,64}.o
x86/retpoline: Ensure default return thunk isn't used at runtime
x86/vdso: Use CONFIG_COMPAT_32 to specify vdso32
x86/vdso: Use $(addprefix ) instead of $(foreach )
x86/vdso: Simplify obj-y addition
x86/vdso: Consolidate targets and clean-files
x86/bugs: Rename CONFIG_RETHUNK => CONFIG_MITIGATION_RETHUNK
x86/bugs: Rename CONFIG_CPU_SRSO => CONFIG_MITIGATION_SRSO
x86/bugs: Rename CONFIG_CPU_IBRS_ENTRY => CONFIG_MITIGATION_IBRS_ENTRY
x86/bugs: Rename CONFIG_CPU_UNRET_ENTRY => CONFIG_MITIGATION_UNRET_ENTRY
x86/bugs: Rename CONFIG_SLS => CONFIG_MITIGATION_SLS
...
Pull x86 entry update from Thomas Gleixner:
"A single update for the x86 entry code:
The current CR3 handling for kernel page table isolation in the
paranoid return paths which are relevant for #NMI, #MCE, #VC, #DB and
#DF is unconditionally writing CR3 with the value retrieved on
exception entry.
In the vast majority of cases when returning to the kernel this is a
pointless exercise because CR3 was not modified on exception entry.
The only situation where this is necessary is when the exception
interrupts a entry from user before switching to kernel CR3 or
interrupts an exit to user after switching back to user CR3.
As CR3 writes can be expensive on some systems this becomes measurable
overhead with high frequency #NMIs such as perf.
Avoid this overhead by checking the CR3 value, which was saved on
entry, and write it back to CR3 only when it is a user CR3"
* tag 'x86-entry-2024-03-11' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/entry: Avoid redundant CR3 write on paranoid returns
The CR3 restore happens in:
1. #NMI return.
2. paranoid_exit() (i.e. #MCE, #VC, #DB and #DF return)
Contrary to the implication in commit 21e9445911 ("x86/mm: Optimize
RESTORE_CR3"), the kernel never modifies CR3 in any of these exceptions,
except for switching from user to kernel pagetables under PTI. That
means that most of the time when returning from an exception that
interrupted the kernel no CR3 restore is necessary. Writing CR3 is
expensive on some machines.
Most of the time because the interrupt might have come during kernel entry
before the user to kernel CR3 switch or the during exit after the kernel to
user switch. In the former case skipping the restore would be correct, but
definitely not for the latter.
So check the saved CR3 value and restore it only, if it is a user CR3.
Give the macro a new name to clarify its usage, and remove a comment that
was describing the original behaviour along with the not longer needed jump
label.
Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
Signed-off-by: Brendan Jackman <jackmanb@google.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20240108113950.360438-1-jackmanb@google.com
[Rewrote commit message; responded to review comments]
Change-Id: I6e56978c4753fb943a7897ff101f519514fa0827
The code in common_interrupt_return() does a bunch of unconditional
work that is really only needed on PTI kernels. Specifically it
unconditionally copies the IRET frame back onto the entry stack,
swizzles onto the entry stack and does IRET from there.
However, without PTI we can simply IRET from whatever stack we're on.
ivb-ep, mitigations=off, gettid-1m:
PRE:
140,118,538 cycles:k ( +- 0.01% )
236,692,878 instructions:k # 1.69 insn per cycle ( +- 0.00% )
POST:
140,026,608 cycles:k ( +- 0.01% )
236,696,176 instructions:k # 1.69 insn per cycle ( +- 0.00% )
(this is with --repeat 100 and the run-to-run variance is bigger than
the difference shown)
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20231120143626.638107480@infradead.org
Do fine-grained Kconfig for all the various retbleed parts.
NOTE: if your compiler doesn't support return thunks this will
silently 'upgrade' your mitigation to IBPB, you might not like this.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Implement Kernel IBRS - currently the only known option to mitigate RSB
underflow speculation issues on Skylake hardware.
Note: since IBRS_ENTER requires fuller context established than
UNTRAIN_RET, it must be placed after it. However, since UNTRAIN_RET
itself implies a RET, it must come after IBRS_ENTER. This means
IBRS_ENTER needs to also move UNTRAIN_RET.
Note 2: KERNEL_IBRS is sub-optimal for XenPV.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Yes, r11 and rcx have been restored previously, but since they're being
popped anyway (into rsi) might as well pop them into their own regs --
setting them to the value they already are.
Less magical code.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20220506121631.365070674@infradead.org
Since the upper regs don't exist for ia32 code, preserving them
doesn't hurt and it simplifies the code.
This doesn't add any attack surface that would not already be
available through INT80.
Notably:
- 32bit SYSENTER: didn't clear si, dx, cx.
- 32bit SYSCALL, INT80: *do* clear si since the C functions don't
take a second argument.
- 64bit: didn't clear si since the C functions take a second
argument; except the error_entry path might have only one argument,
so clearing si was missing here.
32b SYSENTER should be clearing all those 3 registers, nothing uses them
and selftests pass.
Unconditionally clear rsi since it simplifies code.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20220506121631.293889636@infradead.org
PUSH_AND_CLEAR_REGS, as the name implies, performs two functions:
pushing registers and clearing registers. They don't necessarily have
to be performed in immediate sequence, although all current users
do. Split it into two macros for the case where that isn't desired;
the FRED enabling patchset will eventually make use of this.
Signed-off-by: H. Peter Anvin (Intel) <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20210510185316.3307264-6-hpa@zytor.com
inst.h was included in calling.h solely to instantiate the RDPID macro.
The usage of RDPID was removed in
6a3ea3e68b ("x86/entry/64: Do not use RDPID in paranoid entry to accomodate KVM")
so remove the include.
Fixes: 6a3ea3e68b ("x86/entry/64: Do not use RDPID in paranoid entry to accomodate KVM")
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20200827171735.93825-1-ubizjak@gmail.com
KVM has an optmization to avoid expensive MRS read/writes on
VMENTER/EXIT. It caches the MSR values and restores them either when
leaving the run loop, on preemption or when going out to user space.
The affected MSRs are not required for kernel context operations. This
changed with the recently introduced mechanism to handle FSGSBASE in the
paranoid entry code which has to retrieve the kernel GSBASE value by
accessing per CPU memory. The mechanism needs to retrieve the CPU number
and uses either LSL or RDPID if the processor supports it.
Unfortunately RDPID uses MSR_TSC_AUX which is in the list of cached and
lazily restored MSRs, which means between the point where the guest value
is written and the point of restore, MSR_TSC_AUX contains a random number.
If an NMI or any other exception which uses the paranoid entry path happens
in such a context, then RDPID returns the random guest MSR_TSC_AUX value.
As a consequence this reads from the wrong memory location to retrieve the
kernel GSBASE value. Kernel GS is used to for all regular this_cpu_*()
operations. If the GSBASE in the exception handler points to the per CPU
memory of a different CPU then this has the obvious consequences of data
corruption and crashes.
As the paranoid entry path is the only place which accesses MSR_TSX_AUX
(via RDPID) and the fallback via LSL is not significantly slower, remove
the RDPID alternative from the entry path and always use LSL.
The alternative would be to write MSR_TSC_AUX on every VMENTER and VMEXIT
which would be inflicting massive overhead on that code path.
[ tglx: Rewrote changelog ]
Fixes: eaad981291 ("x86/entry/64: Introduce the FIND_PERCPU_BASE macro")
Reported-by: Tom Lendacky <thomas.lendacky@amd.com>
Debugged-by: Tom Lendacky <thomas.lendacky@amd.com>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20200821105229.18938-1-pbonzini@redhat.com
Without FSGSBASE, user space cannot change GSBASE other than through a
PRCTL. The kernel enforces that the user space GSBASE value is postive as
negative values are used for detecting the kernel space GSBASE value in the
paranoid entry code.
If FSGSBASE is enabled, user space can set arbitrary GSBASE values without
kernel intervention, including negative ones, which breaks the paranoid
entry assumptions.
To avoid this, paranoid entry needs to unconditionally save the current
GSBASE value independent of the interrupted context, retrieve and write the
kernel GSBASE and unconditionally restore the saved value on exit. The
restore happens either in paranoid_exit or in the special exit path of the
NMI low level code.
All other entry code pathes which use unconditional SWAPGS are not affected
as they do not depend on the actual content.
[ tglx: Massaged changelogs and comments ]
Suggested-by: H. Peter Anvin <hpa@zytor.com>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Chang S. Bae <chang.seok.bae@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/1557309753-24073-13-git-send-email-chang.seok.bae@intel.com
Link: https://lkml.kernel.org/r/20200528201402.1708239-12-sashal@kernel.org
Device interrupts which go through do_IRQ() or the spurious interrupt
handler have their separate entry code on 64 bit for no good reason.
Both 32 and 64 bit transport the vector number through ORIG_[RE]AX in
pt_regs. Further the vector number is forced to fit into an u8 and is
complemented and offset by 0x80 so it's in the signed character
range. Otherwise GAS would expand the pushq to a 5 byte instruction for any
vector > 0x7F.
Treat the vector number like an error code and hand it to the C function as
argument. This allows to get rid of the extra entry code in a later step.
Simplify the error code push magic by implementing the pushq imm8 via a
'.byte 0x6a, vector' sequence so GAS is not able to screw it up. As the
pushq imm8 is sign extending the resulting error code needs to be truncated
to 8 bits in C code.
Originally-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Link: https://lore.kernel.org/r/20200521202118.796915981@linutronix.de
The PUSH_AND_CLEAR_REGS macro zeroes each register immediately after
pushing it. If an NMI or exception hits after a register is cleared,
but before the UNWIND_HINT_REGS annotation, the ORC unwinder will
wrongly think the previous value of the register was zero. This can
confuse the unwinding process and cause it to exit early.
Because ORC is simpler than DWARF, there are a limited number of unwind
annotation states, so it's not possible to add an individual unwind hint
after each push/clear combination. Instead, the register clearing
instructions need to be consolidated and moved to after the
UNWIND_HINT_REGS annotation.
Fixes: 3f01daecd5 ("x86/entry/64: Introduce the PUSH_AND_CLEAN_REGS macro")
Reviewed-by: Miroslav Benes <mbenes@suse.cz>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Dave Jones <dsj@fb.com>
Cc: Jann Horn <jannh@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Link: https://lore.kernel.org/r/68fd3d0bc92ae2d62ff7879d15d3684217d51f08.1587808742.git.jpoimboe@redhat.com
Spectre v1 isn't only about array bounds checks. It can affect any
conditional checks. The kernel entry code interrupt, exception, and NMI
handlers all have conditional swapgs checks. Those may be problematic in
the context of Spectre v1, as kernel code can speculatively run with a user
GS.
For example:
if (coming from user space)
swapgs
mov %gs:<percpu_offset>, %reg
mov (%reg), %reg1
When coming from user space, the CPU can speculatively skip the swapgs, and
then do a speculative percpu load using the user GS value. So the user can
speculatively force a read of any kernel value. If a gadget exists which
uses the percpu value as an address in another load/store, then the
contents of the kernel value may become visible via an L1 side channel
attack.
A similar attack exists when coming from kernel space. The CPU can
speculatively do the swapgs, causing the user GS to get used for the rest
of the speculative window.
The mitigation is similar to a traditional Spectre v1 mitigation, except:
a) index masking isn't possible; because the index (percpu offset)
isn't user-controlled; and
b) an lfence is needed in both the "from user" swapgs path and the
"from kernel" non-swapgs path (because of the two attacks described
above).
The user entry swapgs paths already have SWITCH_TO_KERNEL_CR3, which has a
CR3 write when PTI is enabled. Since CR3 writes are serializing, the
lfences can be skipped in those cases.
On the other hand, the kernel entry swapgs paths don't depend on PTI.
To avoid unnecessary lfences for the user entry case, create two separate
features for alternative patching:
X86_FEATURE_FENCE_SWAPGS_USER
X86_FEATURE_FENCE_SWAPGS_KERNEL
Use these features in entry code to patch in lfences where needed.
The features aren't enabled yet, so there's no functional change.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
In preparation for wider use, move the ENCODE_FRAME_POINTER macros to
a common header and provide inline asm versions.
These macros are used to encode a pt_regs frame for the unwinder; see
unwind_frame.c:decode_frame_pointer().
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently, CONFIG_JUMP_LABEL just means "I _want_ to use jump label".
The jump label is controlled by HAVE_JUMP_LABEL, which is defined
like this:
#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_JUMP_LABEL)
# define HAVE_JUMP_LABEL
#endif
We can improve this by testing 'asm goto' support in Kconfig, then
make JUMP_LABEL depend on CC_HAS_ASM_GOTO.
Ugly #ifdef HAVE_JUMP_LABEL will go away, and CONFIG_JUMP_LABEL will
match to the real kernel capability.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
This reverts commit 5bdcd510c2.
The macro based workarounds for GCC's inlining bugs caused regressions: distcc
and other distro build setups broke, and the fixes are not easy nor will they
solve regressions on already existing installations.
So we are reverting this patch and the 8 followup patches.
What makes this revert easier is that GCC9 will likely include the new 'asm inline'
syntax that makes inlining of assembly blocks a lot more robust.
This is a superior method to any macro based hackeries - and might even be
backported to GCC8, which would make all modern distros get the inlining
fixes as well.
Many thanks to Masahiro Yamada and others for helping sort out these problems.
Reported-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Borislav Petkov <bp@alien8.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Juergen Gross <jgross@suse.com>
Cc: Richard Biener <rguenther@suse.de>
Cc: Kees Cook <keescook@chromium.org>
Cc: Segher Boessenkool <segher@kernel.crashing.org>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Nadav Amit <namit@vmware.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull stackleak gcc plugin from Kees Cook:
"Please pull this new GCC plugin, stackleak, for v4.20-rc1. This plugin
was ported from grsecurity by Alexander Popov. It provides efficient
stack content poisoning at syscall exit. This creates a defense
against at least two classes of flaws:
- Uninitialized stack usage. (We continue to work on improving the
compiler to do this in other ways: e.g. unconditional zero init was
proposed to GCC and Clang, and more plugin work has started too).
- Stack content exposure. By greatly reducing the lifetime of valid
stack contents, exposures via either direct read bugs or unknown
cache side-channels become much more difficult to exploit. This
complements the existing buddy and heap poisoning options, but
provides the coverage for stacks.
The x86 hooks are included in this series (which have been reviewed by
Ingo, Dave Hansen, and Thomas Gleixner). The arm64 hooks have already
been merged through the arm64 tree (written by Laura Abbott and
reviewed by Mark Rutland and Will Deacon).
With VLAs having been removed this release, there is no need for
alloca() protection, so it has been removed from the plugin"
* tag 'stackleak-v4.20-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
arm64: Drop unneeded stackleak_check_alloca()
stackleak: Allow runtime disabling of kernel stack erasing
doc: self-protection: Add information about STACKLEAK feature
fs/proc: Show STACKLEAK metrics in the /proc file system
lkdtm: Add a test for STACKLEAK
gcc-plugins: Add STACKLEAK plugin for tracking the kernel stack
x86/entry: Add STACKLEAK erasing the kernel stack at the end of syscalls
The STACKLEAK feature (initially developed by PaX Team) has the following
benefits:
1. Reduces the information that can be revealed through kernel stack leak
bugs. The idea of erasing the thread stack at the end of syscalls is
similar to CONFIG_PAGE_POISONING and memzero_explicit() in kernel
crypto, which all comply with FDP_RIP.2 (Full Residual Information
Protection) of the Common Criteria standard.
2. Blocks some uninitialized stack variable attacks (e.g. CVE-2017-17712,
CVE-2010-2963). That kind of bugs should be killed by improving C
compilers in future, which might take a long time.
This commit introduces the code filling the used part of the kernel
stack with a poison value before returning to userspace. Full
STACKLEAK feature also contains the gcc plugin which comes in a
separate commit.
The STACKLEAK feature is ported from grsecurity/PaX. More information at:
https://grsecurity.net/https://pax.grsecurity.net/
This code is modified from Brad Spengler/PaX Team's code in the last
public patch of grsecurity/PaX based on our understanding of the code.
Changes or omissions from the original code are ours and don't reflect
the original grsecurity/PaX code.
Performance impact:
Hardware: Intel Core i7-4770, 16 GB RAM
Test #1: building the Linux kernel on a single core
0.91% slowdown
Test #2: hackbench -s 4096 -l 2000 -g 15 -f 25 -P
4.2% slowdown
So the STACKLEAK description in Kconfig includes: "The tradeoff is the
performance impact: on a single CPU system kernel compilation sees a 1%
slowdown, other systems and workloads may vary and you are advised to
test this feature on your expected workload before deploying it".
Signed-off-by: Alexander Popov <alex.popov@linux.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Previously, error_entry() and paranoid_entry() saved the GP registers
onto stack space previously allocated by its callers. Combine these two
steps in the callers, and use the generic PUSH_AND_CLEAR_REGS macro
for that.
This adds a significant amount ot text size. However, Ingo Molnar points
out that:
"these numbers also _very_ significantly over-represent the
extra footprint. The assumptions that resulted in
us compressing the IRQ entry code have changed very
significantly with the new x86 IRQ allocation code we
introduced in the last year:
- IRQ vectors are usually populated in tightly clustered
groups.
With our new vector allocator code the typical per CPU
allocation percentage on x86 systems is ~3 device vectors
and ~10 fixed vectors out of ~220 vectors - i.e. a very
low ~6% utilization (!). [...]
The days where we allocated a lot of vectors on every
CPU and the compression of the IRQ entry code text
mattered are over.
- Another issue is that only a small minority of vectors
is frequent enough to actually matter to cache utilization
in practice: 3-4 key IPIs and 1-2 device IRQs at most - and
those vectors tend to be tightly clustered as well into about
two groups, and are probably already on 2-3 cache lines in
practice.
For the common case of 'cache cold' IRQs it's the depth of
the call chain and the fragmentation of the resulting I$
that should be the main performance limit - not the overall
size of it.
- The CPU side cost of IRQ delivery is still very expensive
even in the best, most cached case, as in 'over a thousand
cycles'. So much stuff is done that maybe contemporary x86
IRQ entry microcode already prefetches the IDT entry and its
expected call target address."[*]
[*] http://lkml.kernel.org/r/20180208094710.qnjixhm6hybebdv7@gmail.com
The "testb $3, CS(%rsp)" instruction in the idtentry macro does not need
modification. Previously, %rsp was manually decreased by 15*8; with
this patch, %rsp is decreased by 15 pushq instructions.
[jpoimboe@redhat.com: unwind hint improvements]
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dan.j.williams@intel.com
Link: http://lkml.kernel.org/r/20180211104949.12992-7-linux@dominikbrodowski.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The switch to the user space page tables in the low level ASM code sets
unconditionally bit 12 and bit 11 of CR3. Bit 12 is switching the base
address of the page directory to the user part, bit 11 is switching the
PCID to the PCID associated with the user page tables.
This fails on a machine which lacks PCID support because bit 11 is set in
CR3. Bit 11 is reserved when PCID is inactive.
While the Intel SDM claims that the reserved bits are ignored when PCID is
disabled, the AMD APM states that they should be cleared.
This went unnoticed as the AMD APM was not checked when the code was
developed and reviewed and test systems with Intel CPUs never failed to
boot. The report is against a Centos 6 host where the guest fails to boot,
so it's not yet clear whether this is a virt issue or can happen on real
hardware too, but thats irrelevant as the AMD APM clearly ask for clearing
the reserved bits.
Make sure that on non PCID machines bit 11 is not set by the page table
switching code.
Andy suggested to rename the related bits and masks so they are clearly
describing what they should be used for, which is done as well for clarity.
That split could have been done with alternatives but the macro hell is
horrible and ugly. This can be done on top if someone cares to remove the
extra orq. For now it's a straight forward fix.
Fixes: 6fd166aae7 ("x86/mm: Use/Fix PCID to optimize user/kernel switches")
Reported-by: Laura Abbott <labbott@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: stable <stable@vger.kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Willy Tarreau <w@1wt.eu>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1801140009150.2371@nanos
