[PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block

linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed

* [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
@ 2022-08-17 18:54 Nathan Chancellor
  2022-08-17 22:08 ` Nick Desaulniers
  2022-08-18  7:49 ` Peter Zijlstra
  0 siblings, 2 replies; 8+ messages in thread
From: Nathan Chancellor @ 2022-08-17 18:54 UTC (permalink / raw)
  To: Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, x86
  Cc: Peter Zijlstra, Nick Desaulniers, linux-kernel, llvm, Nathan Chancellor

LLVM 16 will have support for this flag so move it out of the GCC-only
block to allow LLVM builds to take advantage of it.

Link: https://github.com/ClangBuiltLinux/linux/issues/1665
Link: https://github.com/llvm/llvm-project/commit/6f867f9102838ebe314c1f3661fdf95700386e5a
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
---

I was not sure if this information is relevant for the commit message
but I can boot without any issues on my test machines (two Intel and one
AMD).

 arch/x86/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 7854685c5f25..987da87c7778 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -14,13 +14,13 @@ endif
 
 ifdef CONFIG_CC_IS_GCC
 RETPOLINE_CFLAGS	:= $(call cc-option,-mindirect-branch=thunk-extern -mindirect-branch-register)
-RETPOLINE_CFLAGS	+= $(call cc-option,-mindirect-branch-cs-prefix)
 RETPOLINE_VDSO_CFLAGS	:= $(call cc-option,-mindirect-branch=thunk-inline -mindirect-branch-register)
 endif
 ifdef CONFIG_CC_IS_CLANG
 RETPOLINE_CFLAGS	:= -mretpoline-external-thunk
 RETPOLINE_VDSO_CFLAGS	:= -mretpoline
 endif
+RETPOLINE_CFLAGS	+= $(call cc-option,-mindirect-branch-cs-prefix)
 
 ifdef CONFIG_RETHUNK
 RETHUNK_CFLAGS		:= -mfunction-return=thunk-extern

base-commit: 568035b01cfb107af8d2e4bd2fb9aea22cf5b868
-- 
2.37.2


^ permalink raw reply related	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
  2022-08-17 18:54 [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block Nathan Chancellor
@ 2022-08-17 22:08 ` Nick Desaulniers
  2022-08-18  7:45   ` Peter Zijlstra
  2022-08-18  7:49 ` Peter Zijlstra
  1 sibling, 1 reply; 8+ messages in thread
From: Nick Desaulniers @ 2022-08-17 22:08 UTC (permalink / raw)
  To: Nathan Chancellor, Peter Zijlstra
  Cc: Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, x86,
	linux-kernel, llvm

On Wed, Aug 17, 2022 at 11:56 AM Nathan Chancellor <nathan@kernel.org> wrote:
>
> LLVM 16 will have support for this flag so move it out of the GCC-only
> block to allow LLVM builds to take advantage of it.
>
> Link: https://github.com/ClangBuiltLinux/linux/issues/1665
> Link: https://github.com/llvm/llvm-project/commit/6f867f9102838ebe314c1f3661fdf95700386e5a
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>

Thanks for the patch, I also boot tested this (in virtual guests for
both 64b and 32b).

Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>

Peter, is there anything else special about these prefixes needed to
make use of them?

> ---
>
> I was not sure if this information is relevant for the commit message
> but I can boot without any issues on my test machines (two Intel and one
> AMD).
>
>  arch/x86/Makefile | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/x86/Makefile b/arch/x86/Makefile
> index 7854685c5f25..987da87c7778 100644
> --- a/arch/x86/Makefile
> +++ b/arch/x86/Makefile
> @@ -14,13 +14,13 @@ endif
>
>  ifdef CONFIG_CC_IS_GCC
>  RETPOLINE_CFLAGS       := $(call cc-option,-mindirect-branch=thunk-extern -mindirect-branch-register)
> -RETPOLINE_CFLAGS       += $(call cc-option,-mindirect-branch-cs-prefix)
>  RETPOLINE_VDSO_CFLAGS  := $(call cc-option,-mindirect-branch=thunk-inline -mindirect-branch-register)
>  endif
>  ifdef CONFIG_CC_IS_CLANG
>  RETPOLINE_CFLAGS       := -mretpoline-external-thunk
>  RETPOLINE_VDSO_CFLAGS  := -mretpoline
>  endif
> +RETPOLINE_CFLAGS       += $(call cc-option,-mindirect-branch-cs-prefix)
>
>  ifdef CONFIG_RETHUNK
>  RETHUNK_CFLAGS         := -mfunction-return=thunk-extern
>
> base-commit: 568035b01cfb107af8d2e4bd2fb9aea22cf5b868
> --
> 2.37.2
>
>


-- 
Thanks,
~Nick Desaulniers

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
  2022-08-17 22:08 ` Nick Desaulniers
@ 2022-08-18  7:45   ` Peter Zijlstra
  2022-08-18  8:00     ` Peter Zijlstra
                       ` (2 more replies)
  0 siblings, 3 replies; 8+ messages in thread
From: Peter Zijlstra @ 2022-08-18  7:45 UTC (permalink / raw)
  To: Nick Desaulniers
  Cc: Nathan Chancellor, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	Dave Hansen, x86, linux-kernel, llvm

On Wed, Aug 17, 2022 at 03:08:01PM -0700, Nick Desaulniers wrote:
> On Wed, Aug 17, 2022 at 11:56 AM Nathan Chancellor <nathan@kernel.org> wrote:
> >
> > LLVM 16 will have support for this flag so move it out of the GCC-only
> > block to allow LLVM builds to take advantage of it.
> >
> > Link: https://github.com/ClangBuiltLinux/linux/issues/1665
> > Link: https://github.com/llvm/llvm-project/commit/6f867f9102838ebe314c1f3661fdf95700386e5a
> > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> 
> Thanks for the patch, I also boot tested this (in virtual guests for
> both 64b and 32b).
> 
> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
> Tested-by: Nick Desaulniers <ndesaulniers@google.com>
> 
> Peter, is there anything else special about these prefixes needed to
> make use of them?

The thing to do is boot with (warning, *very* verbose):

	"spectre_v2=retpoline,lfence debug-alternative"

and observe that the retpoline sites all replace:

	"cs call __x86_indirect_thunk_r11" (6 bytes)

with:

	"lfence; jmp *%r11" (6 bytes)


This being clang, you'll ofcourse still have a few weird:

	"Jcc.d32 __x86_indirect_thunk_r11"

sites that will not be able to be replaced, but there's nothing I can do
about that other than to continue to encourage y'all to stop emitting
them ;-)

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
  2022-08-18  7:45   ` Peter Zijlstra
@ 2022-08-18  8:00     ` Peter Zijlstra
  2022-09-30 18:10     ` Nick Desaulniers
  2022-10-04 21:27     ` Nick Desaulniers
  2 siblings, 0 replies; 8+ messages in thread
From: Peter Zijlstra @ 2022-08-18  8:00 UTC (permalink / raw)
  To: Nick Desaulniers
  Cc: Nathan Chancellor, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	Dave Hansen, x86, linux-kernel, llvm

On Thu, Aug 18, 2022 at 09:45:39AM +0200, Peter Zijlstra wrote:

> The thing to do is boot with (warning, *very* verbose):
> 
> 	"spectre_v2=retpoline,lfence debug-alternative"
> 
> and observe that the retpoline sites all replace:
> 
> 	"cs call __x86_indirect_thunk_r11" (6 bytes)
> 
> with:
> 
> 	"lfence; jmp *%r11" (6 bytes)

		^^ call, ofc, unless the original was jmp then jmp...

/me goes get more morning juice in an attempt to wake up.

> 
> 
> This being clang, you'll ofcourse still have a few weird:
> 
> 	"Jcc.d32 __x86_indirect_thunk_r11"
> 
> sites that will not be able to be replaced, but there's nothing I can do
> about that other than to continue to encourage y'all to stop emitting
> them ;-)

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
  2022-08-18  7:45   ` Peter Zijlstra
  2022-08-18  8:00     ` Peter Zijlstra
@ 2022-09-30 18:10     ` Nick Desaulniers
  2022-10-04 21:27     ` Nick Desaulniers
  2 siblings, 0 replies; 8+ messages in thread
From: Nick Desaulniers @ 2022-09-30 18:10 UTC (permalink / raw)
  To: Peter Zijlstra
  Cc: Nathan Chancellor, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	Dave Hansen, x86, linux-kernel, llvm, Moreira, Joao

On Thu, Aug 18, 2022 at 12:45 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> This being clang, you'll ofcourse still have a few weird:
>
>         "Jcc.d32 __x86_indirect_thunk_r11"
>
> sites that will not be able to be replaced, but there's nothing I can do
> about that other than to continue to encourage y'all to stop emitting
> them ;-)

Joao has a patch up for this:
https://reviews.llvm.org/D134915
-- 
Thanks,
~Nick Desaulniers

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
  2022-08-18  7:45   ` Peter Zijlstra
  2022-08-18  8:00     ` Peter Zijlstra
  2022-09-30 18:10     ` Nick Desaulniers
@ 2022-10-04 21:27     ` Nick Desaulniers
  2022-10-06 17:56       ` Nick Desaulniers
  2 siblings, 1 reply; 8+ messages in thread
From: Nick Desaulniers @ 2022-10-04 21:27 UTC (permalink / raw)
  To: Peter Zijlstra, Borislav Petkov
  Cc: Nathan Chancellor, Thomas Gleixner, Ingo Molnar, Dave Hansen,
	x86, linux-kernel, llvm, Moreira, Joao

[-- Attachment #1: Type: text/plain, Size: 2806 bytes --]

On Thu, Aug 18, 2022 at 12:45 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Wed, Aug 17, 2022 at 03:08:01PM -0700, Nick Desaulniers wrote:
> > On Wed, Aug 17, 2022 at 11:56 AM Nathan Chancellor <nathan@kernel.org> wrote:
> > >
> > > LLVM 16 will have support for this flag so move it out of the GCC-only
> > > block to allow LLVM builds to take advantage of it.
> > >
> > > Link: https://github.com/ClangBuiltLinux/linux/issues/1665
> > > Link: https://github.com/llvm/llvm-project/commit/6f867f9102838ebe314c1f3661fdf95700386e5a
> > > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> >
> > Thanks for the patch, I also boot tested this (in virtual guests for
> > both 64b and 32b).
> >
> > Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
> > Tested-by: Nick Desaulniers <ndesaulniers@google.com>
> >
> > Peter, is there anything else special about these prefixes needed to
> > make use of them?
>
> The thing to do is boot with (warning, *very* verbose):
>
>         "spectre_v2=retpoline,lfence debug-alternative"

ah, debug-alternative also needs loglevel=8. Was wondering why I
wasn't getting anything new in the output...

No kidding on the verbosity. For a defconfig, booting in QEMU and
redirecting stdout to a file produced a 95257 line file.

$ grep "SMP alternatives: retpoline" /tmp/log.txt | wc -l
18623
$ du -h /tmp/log.txt
11M /tmp/log.txt
$ grep "SMP alternatives: retpoline" /tmp/log.txt | tr -s ' '
...
[ 56.736261] SMP alternatives: retpoline at:
pm_check_save_msr+0x30/0x80 (ffffffffaa34b320) len: 6 to:
__x86_indirect_thunk_r11+0x0/0x20
$ grep "SMP alternatives: retpoline" /tmp/log.txt | tr -s ' ' | cut -d
' ' -f 10 | sort -u
5
6
# Curious about the 5, looks like there's only 2 cases:
$ grep "SMP alternatives: retpoline" /tmp/log.txt | grep 'len: 5'
[    0.147261] SMP alternatives: retpoline at: ret_from_fork+0x1d/0x30
(ffffffffa9601dbd) len: 5 to: __x86_indirect_thunk_rbx+0x0/0x20
[    2.274261] SMP alternatives: retpoline at: __efi_call+0x23/0x30
(ffffffffa9671663) len: 5 to: __x86_indirect_thunk_rdi+0x0/0x20


Is there anything else I can do to help verify Nathan and Joao's
patches?  If not, would you mind Ack'ing this so the x86 maintainers
can pick it up?


>
> and observe that the retpoline sites all replace:
>
>         "cs call __x86_indirect_thunk_r11" (6 bytes)
>
> with:
>
>         "lfence; jmp *%r11" (6 bytes)
>
>
> This being clang, you'll ofcourse still have a few weird:
>
>         "Jcc.d32 __x86_indirect_thunk_r11"
>
> sites that will not be able to be replaced, but there's nothing I can do
> about that other than to continue to encourage y'all to stop emitting
> them ;-)

Noting that Joao's patch was applied in my tests.
https://reviews.llvm.org/D134915
I plan to land Joao's patch imminently.
-- 
Thanks,
~Nick Desaulniers

[-- Attachment #2: log.txt --]
[-- Type: text/plain, Size: 24985 bytes --]


^[[01;32mQEMU location: ^[[0m/usr/bin^[[0m

^[[01;32mQEMU version: ^[[0mQEMU emulator version 7.0.0 (Debian 1:7.0+dfsg-7)
^[[0m
$ timeout --foreground 3m stdbuf -oL -eL qemu-system-x86_64 -d unimp,guest_errors -kernel /android0/kernel-all/arch/x86_64/boot/bzImage -append "console=ttyS0 earlycon=uart8250,io,0x3f8 spectre_v2=retpoline,lfence debug-alternative" -cpu host -enable-kvm -smp 64 -display none -initrd /android1/boot-utils/images/x86_64/rootfs.cpio -m 512m -nodefaults -no-reboot -serial mon:stdio
[    0.000000] Linux version 6.0.0-rc7-00182-g3801f027b5b5 (ndesaulniers@ndesaulniers-desktop.svl.corp.google.com) (Nick Desaulniers clang version 16.0.0 (git@github.com:llvm/llvm-project.git 4b104c2ec016b3722190cdd67d23de6adc552f9e), LLD 16.0.0) #2 SMP PREEMPT_DYNAMIC Tue Oct 4 14:05:52 PDT 2022
[    0.000000] Command line: console=ttyS0 earlycon=uart8250,io,0x3f8 spectre_v2=retpoline,lfence debug-alternative
[    0.000000] x86/fpu: Supporting XSAVE feature 0x001: 'x87 floating point registers'
[    0.000000] x86/fpu: Supporting XSAVE feature 0x002: 'SSE registers'
[    0.000000] x86/fpu: Supporting XSAVE feature 0x004: 'AVX registers'
[    0.000000] x86/fpu: xstate_offset[2]:  576, xstate_sizes[2]:  256
[    0.000000] x86/fpu: Enabled xstate features 0x7, context size is 832 bytes, using 'compacted' format.
[    0.000000] signal: max sigframe size: 1776
[    0.000000] BIOS-provided physical RAM map:
[    0.000000] BIOS-e820: [mem 0x0000000000000000-0x000000000009fbff] usable
[    0.000000] BIOS-e820: [mem 0x000000000009fc00-0x000000000009ffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000000f0000-0x00000000000fffff] reserved
[    0.000000] BIOS-e820: [mem 0x0000000000100000-0x000000001ffdffff] usable
[    0.000000] BIOS-e820: [mem 0x000000001ffe0000-0x000000001fffffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000feffc000-0x00000000feffffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000fffc0000-0x00000000ffffffff] reserved
[    0.000000] earlycon: uart8250 at I/O port 0x3f8 (options '')
[    0.000000] printk: bootconsole [uart8250] enabled
[    0.000000] NX (Execute Disable) protection: active
[    0.000000] SMBIOS 2.8 present.
[    0.000000] DMI: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.0-debian-1.16.0-4 04/01/2014
[    0.000000] Hypervisor detected: KVM
[    0.000000] kvm-clock: Using msrs 4b564d01 and 4b564d00
[    0.000000] kvm-clock: using sched offset of 212626233 cycles
[    0.000586] clocksource: kvm-clock: mask: 0xffffffffffffffff max_cycles: 0x1cd42e4dffb, max_idle_ns: 881590591483 ns
[    0.002414] tsc: Detected 2694.936 MHz processor
[    0.003227] last_pfn = 0x1ffe0 max_arch_pfn = 0x400000000
[    0.003877] x86/PAT: Configuration [0-7]: WB  WC  UC- UC  WB  WP  UC- WT  
[    0.006441] found SMP MP-table at [mem 0x000f5ba0-0x000f5baf]
[    0.007108] Using GB pages for direct mapping
[    0.007713] RAMDISK: [mem 0x1fb0d000-0x1ffdffff]
[    0.008275] ACPI: Early table checksum verification disabled
[    0.008923] ACPI: RSDP 0x00000000000F59D0 000014 (v00 BOCHS )
[    0.009572] ACPI: RSDT 0x000000001FFE2FBE 000034 (v01 BOCHS  BXPC     00000001 BXPC 00000001)
[    0.010533] ACPI: FACP 0x000000001FFE2C7A 000074 (v01 BOCHS  BXPC     00000001 BXPC 00000001)
[    0.011491] ACPI: DSDT 0x000000001FFE0040 002C3A (v01 BOCHS  BXPC     00000001 BXPC 00000001)
[    0.012445] ACPI: FACS 0x000000001FFE0000 000040
[    0.012976] ACPI: APIC 0x000000001FFE2CEE 000270 (v01 BOCHS  BXPC     00000001 BXPC 00000001)
[    0.013929] ACPI: HPET 0x000000001FFE2F5E 000038 (v01 BOCHS  BXPC     00000001 BXPC 00000001)
[    0.014880] ACPI: WAET 0x000000001FFE2F96 000028 (v01 BOCHS  BXPC     00000001 BXPC 00000001)
[    0.015830] ACPI: Reserving FACP table memory at [mem 0x1ffe2c7a-0x1ffe2ced]
[    0.016615] ACPI: Reserving DSDT table memory at [mem 0x1ffe0040-0x1ffe2c79]
[    0.017439] ACPI: Reserving FACS table memory at [mem 0x1ffe0000-0x1ffe003f]
[    0.018229] ACPI: Reserving APIC table memory at [mem 0x1ffe2cee-0x1ffe2f5d]
[    0.019019] ACPI: Reserving HPET table memory at [mem 0x1ffe2f5e-0x1ffe2f95]
[    0.019822] ACPI: Reserving WAET table memory at [mem 0x1ffe2f96-0x1ffe2fbd]
[    0.020814] No NUMA configuration found
[    0.021245] Faking a node at [mem 0x0000000000000000-0x000000001ffdffff]
[    0.021995] NODE_DATA(0) allocated [mem 0x1fb09000-0x1fb0cfff]
[    0.022657] Zone ranges:
[    0.022940]   DMA      [mem 0x0000000000001000-0x0000000000ffffff]
[    0.023632]   DMA32    [mem 0x0000000001000000-0x000000001ffdffff]
[    0.024324]   Normal   empty
[    0.024647] Movable zone start for each node
[    0.025136] Early memory node ranges
[    0.025534]   node   0: [mem 0x0000000000001000-0x000000000009efff]
[    0.026238]   node   0: [mem 0x0000000000100000-0x000000001ffdffff]
[    0.026946] Initmem setup node 0 [mem 0x0000000000001000-0x000000001ffdffff]
[    0.027893] On node 0, zone DMA: 1 pages in unavailable ranges
[    0.027911] On node 0, zone DMA: 97 pages in unavailable ranges
[    0.029539] On node 0, zone DMA32: 32 pages in unavailable ranges
[    0.030423] ACPI: PM-Timer IO Port: 0x608
[    0.031561] ACPI: LAPIC_NMI (acpi_id[0xff] dfl dfl lint[0x1])
[    0.032226] IOAPIC[0]: apic_id 0, version 17, address 0xfec00000, GSI 0-23
[    0.033012] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
[    0.033722] ACPI: INT_SRC_OVR (bus 0 bus_irq 5 global_irq 5 high level)
[    0.034462] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level)
[    0.035201] ACPI: INT_SRC_OVR (bus 0 bus_irq 10 global_irq 10 high level)
[    0.035960] ACPI: INT_SRC_OVR (bus 0 bus_irq 11 global_irq 11 high level)
[    0.036719] ACPI: Using ACPI (MADT) for SMP configuration information
[    0.037453] ACPI: HPET id: 0x8086a201 base: 0xfed00000
[    0.038033] TSC deadline timer available
[    0.038475] smpboot: Allowing 64 CPUs, 0 hotplug CPUs
[    0.039060] kvm-guest: KVM setup pv remote TLB flush
[    0.039623] kvm-guest: setup PV sched yield
[    0.040097] PM: hibernation: Registered nosave memory: [mem 0x00000000-0x00000fff]
[    0.040951] PM: hibernation: Registered nosave memory: [mem 0x0009f000-0x0009ffff]
[    0.041797] PM: hibernation: Registered nosave memory: [mem 0x000a0000-0x000effff]
[    0.042645] PM: hibernation: Registered nosave memory: [mem 0x000f0000-0x000fffff]
[    0.043492] [mem 0x20000000-0xfeffbfff] available for PCI devices
[    0.044172] Booting paravirtualized kernel on KVM
[    0.044695] clocksource: refined-jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 1910969940391419 ns
[    0.050262] setup_percpu: NR_CPUS:64 nr_cpumask_bits:64 nr_cpu_ids:64 nr_node_ids:1
[    0.053360] percpu: Embedded 52 pages/cpu s175464 r8192 d29336 u262144
[    0.054187] Fallback order for Node 0: 0 
[    0.054661] Built 1 zonelists, mobility grouping on.  Total pages: 128736
[    0.055446] Policy zone: DMA32
[    0.055804] Kernel command line: console=ttyS0 earlycon=uart8250,io,0x3f8 spectre_v2=retpoline,lfence debug-alternative
[    0.057090] Unknown kernel command line parameters "spectre_v2=retpoline,lfence", will be passed to user space.
[    0.058248] printk: log_buf_len individual max cpu contribution: 4096 bytes
[    0.059047] printk: log_buf_len total cpu_extra contributions: 258048 bytes
[    0.059844] printk: log_buf_len min size: 262144 bytes
[    0.060767] printk: log_buf_len: 524288 bytes
[    0.061271] printk: early log buf free: 255272(97%)
[    0.061849] Dentry cache hash table entries: 65536 (order: 7, 524288 bytes, linear)
[    0.062887] Inode-cache hash table entries: 32768 (order: 6, 262144 bytes, linear)
[    0.063892] mem auto-init: stack:off, heap alloc:off, heap free:off
[    0.065541] Memory: 458384K/523768K available (16396K kernel code, 2538K rwdata, 5072K rodata, 1540K init, 1272K bss, 65124K reserved, 0K cma-reserved)
[    0.067269] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=64, Nodes=1
[    0.068933] Dynamic Preempt: voluntary
[    0.069602] rcu: Preemptible hierarchical RCU implementation.
[    0.070269] rcu: 	RCU event tracing is enabled.
[    0.070796] 	Trampoline variant of Tasks RCU enabled.
[    0.071389] rcu: RCU calculated value of scheduler-enlistment delay is 100 jiffies.
[    0.073182] NR_IRQS: 4352, nr_irqs: 936, preallocated irqs: 16
[    0.074075] rcu: srcu_init: Setting srcu_struct sizes based on contention.
[    0.074900] random: crng init done
[    0.075396] Console: colour *CGA 80x25
[    0.075853] printk: console [ttyS0] enabled
[    0.075853] printk: console [ttyS0] enabled
[    0.076831] printk: bootconsole [uart8250] disabled
[    0.076831] printk: bootconsole [uart8250] disabled
[    0.077979] ACPI: Core revision 20220331
[    0.078546] clocksource: hpet: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 19112604467 ns
[    0.079687] APIC: Switch to symmetric I/O mode setup
[    0.080295] Switched APIC routing to physical flat.
[    0.080898] kvm-guest: setup PV IPIs
[    0.082251] ..TIMER: vector=0x30 apic1=0 pin1=2 apic2=-1 pin2=-1
[    0.083376] tsc: Marking TSC unstable due to TSCs unsynchronized
[    0.084480] Calibrating delay loop (skipped) preset value.. 5389.87 BogoMIPS (lpj=2694936)
[    0.085479] pid_max: default: 65536 minimum: 512
[    0.086517] LSM: Security Framework initializing
[    0.087490] SELinux:  Initializing.
[    0.088155] Mount-cache hash table entries: 1024 (order: 1, 8192 bytes, linear)
[    0.088480] Mountpoint-cache hash table entries: 1024 (order: 1, 8192 bytes, linear)
[    0.089791] x86/cpu: User Mode Instruction Prevention (UMIP) activated
[    0.091557] Last level iTLB entries: 4KB 512, 2MB 255, 4MB 127
[    0.092479] Last level dTLB entries: 4KB 512, 2MB 255, 4MB 127, 1GB 0
[    0.093482] Spectre V1 : Mitigation: usercopy/swapgs barriers and __user pointer sanitization
[    0.094480] Spectre V2 : retpoline,lfence selected on command line.
[    0.095479] Spectre V2 : WARNING: LFENCE mitigation is not recommended for this CPU, data leaks possible!
[    0.096479] Spectre V2 : Mitigation: LFENCE
[    0.097479] Spectre V2 : Spectre v2 / SpectreRSB mitigation: Filling RSB on context switch
[    0.098480] Spectre V2 : Spectre v2 / SpectreRSB : Filling RSB on VMEXIT
[    0.099479] Spectre V2 : Enabling Speculation Barrier for firmware calls
[    0.100479] RETBleed: Mitigation: untrained return thunk
[    0.101481] Spectre V2 : mitigation: Enabling conditional Indirect Branch Prediction Barrier
[    0.102480] Speculative Store Bypass: Mitigation: Speculative Store Bypass disabled via prctl
[    0.370312] Freeing SMP alternatives memory: 52K
[    0.370575] smpboot: CPU0: AMD Ryzen Threadripper PRO 3995WX 64-Cores (family: 0x17, model: 0x31, stepping: 0x0)
[    0.371575] cblist_init_generic: Setting adjustable number of callback queues.
[    0.372481] cblist_init_generic: Setting shift to 6 and lim to 1.
[    0.373491] Performance Events: Fam17h+ core perfctr, AMD PMU driver.
[    0.374483] ... version:                0
[    0.375194] ... bit width:              48
[    0.375481] ... generic registers:      6
[    0.376189] ... value mask:             0000ffffffffffff
[    0.376481] ... max period:             00007fffffffffff
[    0.377425] ... fixed-purpose events:   0
[    0.377481] ... event mask:             000000000000003f
[    0.378512] rcu: Hierarchical SRCU implementation.
[    0.379360] rcu: 	Max phase no-delay instances is 400.
[    0.379998] smp: Bringing up secondary CPUs ...
[    0.380540] x86: Booting SMP configuration:
[    0.381287] .... node  #0, CPUs:        #1  #2  #3  #4  #5  #6  #7  #8  #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20 #21 #22 #23 #24 #25 #26 #27 #28 #29 #30 #31 #32 #33 #34 #35 #36 #37 #38 #39 #40 #41 #42 #43 #44 #45 #46 #47 #48 #49 #50 #51 #52 #53 #54 #55 #56 #57 #58 #59 #60 #61 #62 #63
[    0.437796] smp: Brought up 1 node, 64 CPUs
[    0.439238] smpboot: Max logical packages: 1
[    0.439483] smpboot: Total of 64 processors activated (344951.80 BogoMIPS)
[    0.442193] devtmpfs: initialized
[    0.442626] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 1911260446275000 ns
[    0.443489] futex hash table entries: 16384 (order: 8, 1048576 bytes, linear)
[    0.444781] PM: RTC time: 21:07:22, date: 2022-10-04
[    0.445700] NET: Registered PF_NETLINK/PF_ROUTE protocol family
[    0.446532] audit: initializing netlink subsys (disabled)
[    0.447532] audit: type=2000 audit(1664917642.765:1): state=initialized audit_enabled=0 res=1
[    0.447547] thermal_sys: Registered thermal governor 'step_wise'
[    0.448482] thermal_sys: Registered thermal governor 'user_space'
[    0.449517] cpuidle: using governor menu
[    0.453738] PCI: Using configuration type 1 for base access
[    0.454482] PCI: Using configuration type 1 for extended access
[    0.459162] kprobes: kprobe jump-optimization is enabled. All kprobes are optimized if possible.
[    0.459558] HugeTLB: registered 2.00 MiB page size, pre-allocated 0 pages
[    0.460485] HugeTLB: 28 KiB vmemmap can be freed for a 2.00 MiB page
[    0.461637] ACPI: Added _OSI(Module Device)
[    0.462506] ACPI: Added _OSI(Processor Device)
[    0.463484] ACPI: Added _OSI(3.0 _SCP Extensions)
[    0.464483] ACPI: Added _OSI(Processor Aggregator Device)
[    0.465482] ACPI: Added _OSI(Linux-Dell-Video)
[    0.466305] ACPI: Added _OSI(Linux-Lenovo-NV-HDMI-Audio)
[    0.466500] ACPI: Added _OSI(Linux-HPI-Hybrid-Graphics)
[    0.468166] ACPI: 1 ACPI AML tables successfully acquired and loaded
[    0.469092] ACPI: Interpreter enabled
[    0.469492] ACPI: PM: (supports S0 S3 S4 S5)
[    0.470264] ACPI: Using IOAPIC for interrupt routing
[    0.470495] PCI: Using host bridge windows from ACPI; if necessary, use "pci=nocrs" and report a bug
[    0.471481] PCI: Using E820 reservations for host bridge windows
[    0.472598] ACPI: Enabled 2 GPEs in block 00 to 0F
[    0.476492] ACPI: PCI Root Bridge [PCI0] (domain 0000 [bus 00-ff])
[    0.477483] acpi PNP0A03:00: _OSC: OS supports [ExtendedConfig ASPM ClockPM Segments MSI HPX-Type3]
[    0.478560] PCI host bridge to bus 0000:00
[    0.479321] pci_bus 0000:00: root bus resource [io  0x0000-0x0cf7 window]
[    0.479481] pci_bus 0000:00: root bus resource [io  0x0d00-0xffff window]
[    0.480481] pci_bus 0000:00: root bus resource [mem 0x000a0000-0x000bffff window]
[    0.481481] pci_bus 0000:00: root bus resource [mem 0x20000000-0xfebfffff window]
[    0.482481] pci_bus 0000:00: root bus resource [mem 0x100000000-0x17fffffff window]
[    0.483481] pci_bus 0000:00: root bus resource [bus 00-ff]
[    0.484582] pci 0000:00:00.0: [8086:1237] type 00 class 0x060000
[    0.486001] pci 0000:00:01.0: [8086:7000] type 00 class 0x060100
[    0.486932] pci 0000:00:01.1: [8086:7010] type 00 class 0x010180
[    0.488615] pci 0000:00:01.1: reg 0x20: [io  0xc000-0xc00f]
[    0.489974] pci 0000:00:01.1: legacy IDE quirk: reg 0x10: [io  0x01f0-0x01f7]
[    0.490481] pci 0000:00:01.1: legacy IDE quirk: reg 0x14: [io  0x03f6]
[    0.491481] pci 0000:00:01.1: legacy IDE quirk: reg 0x18: [io  0x0170-0x0177]
[    0.492481] pci 0000:00:01.1: legacy IDE quirk: reg 0x1c: [io  0x0376]
[    0.493580] pci 0000:00:01.3: [8086:7113] type 00 class 0x068000
[    0.494835] pci 0000:00:01.3: quirk: [io  0x0600-0x063f] claimed by PIIX4 ACPI
[    0.495487] pci 0000:00:01.3: quirk: [io  0x0700-0x070f] claimed by PIIX4 SMB
[    0.496895] ACPI: PCI: Interrupt link LNKA configured for IRQ 10
[    0.497616] ACPI: PCI: Interrupt link LNKB configured for IRQ 10
[    0.498599] ACPI: PCI: Interrupt link LNKC configured for IRQ 11
[    0.499594] ACPI: PCI: Interrupt link LNKD configured for IRQ 11
[    0.500566] ACPI: PCI: Interrupt link LNKS configured for IRQ 9
[    0.502967] iommu: Default domain type: Translated 
[    0.503481] iommu: DMA domain TLB invalidation policy: lazy mode 
[    0.504518] SCSI subsystem initialized
[    0.505260] ACPI: bus type USB registered
[    0.505492] usbcore: registered new interface driver usbfs
[    0.506486] usbcore: registered new interface driver hub
[    0.507487] usbcore: registered new device driver usb
[    0.508420] pps_core: LinuxPPS API ver. 1 registered
[    0.508481] pps_core: Software ver. 5.3.6 - Copyright 2005-2007 Rodolfo Giometti <giometti@linux.it>
[    0.509483] PTP clock support registered
[    0.510203] Advanced Linux Sound Architecture Driver Initialized.
[    0.510566] NetLabel: Initializing
[    0.511174] NetLabel:  domain hash size = 128
[    0.511481] NetLabel:  protocols = UNLABELED CIPSOv4 CALIPSO
[    0.512489] NetLabel:  unlabeled traffic allowed by default
[    0.513508] PCI: Using ACPI for IRQ routing
[    0.514318] vgaarb: loaded
[    0.514540] hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
[    0.515427] hpet0: 3 comparators, 64-bit 100.000000 MHz counter
[    0.518952] clocksource: Switched to clocksource kvm-clock
[    0.519645] VFS: Disk quotas dquot_6.6.0
[    0.520417] VFS: Dquot-cache hash table entries: 512 (order 0, 4096 bytes)
[    0.521743] pnp: PnP ACPI init
[    0.522474] pnp: PnP ACPI: found 5 devices
[    0.531089] clocksource: acpi_pm: mask: 0xffffff max_cycles: 0xffffff, max_idle_ns: 2085701024 ns
[    0.532688] NET: Registered PF_INET protocol family
[    0.533583] IP idents hash table entries: 8192 (order: 4, 65536 bytes, linear)
[    0.535014] tcp_listen_portaddr_hash hash table entries: 256 (order: 0, 4096 bytes, linear)
[    0.536505] Table-perturb hash table entries: 65536 (order: 6, 262144 bytes, linear)
[    0.537891] TCP established hash table entries: 4096 (order: 3, 32768 bytes, linear)
[    0.539284] TCP bind hash table entries: 4096 (order: 4, 65536 bytes, linear)
[    0.540572] TCP: Hash tables configured (established 4096 bind 4096)
[    0.541772] UDP hash table entries: 256 (order: 1, 8192 bytes, linear)
[    0.542952] UDP-Lite hash table entries: 256 (order: 1, 8192 bytes, linear)
[    0.544256] NET: Registered PF_UNIX/PF_LOCAL protocol family
[    0.545564] RPC: Registered named UNIX socket transport module.
[    0.546640] RPC: Registered udp transport module.
[    0.547496] RPC: Registered tcp transport module.
[    0.548339] RPC: Registered tcp NFSv4.1 backchannel transport module.
[    0.549850] pci_bus 0000:00: resource 4 [io  0x0000-0x0cf7 window]
[    0.550977] pci_bus 0000:00: resource 5 [io  0x0d00-0xffff window]
[    0.552093] pci_bus 0000:00: resource 6 [mem 0x000a0000-0x000bffff window]
[    0.553326] pci_bus 0000:00: resource 7 [mem 0x20000000-0xfebfffff window]
[    0.554566] pci_bus 0000:00: resource 8 [mem 0x100000000-0x17fffffff window]
[    0.555851] pci 0000:00:01.0: PIIX3: Enabling Passive Release
[    0.556904] pci 0000:00:00.0: Limiting direct PCI/PCI transfers
[    0.558003] PCI: CLS 0 bytes, default 64
[    0.558783] Unpacking initramfs...
[    0.562846] Freeing initrd memory: 4940K
[    0.576849] Initialise system trusted keyrings
[    0.577801] workingset: timestamp_bits=56 max_order=17 bucket_order=0
[    0.579938] NFS: Registering the id_resolver key type
[    0.580868] Key type id_resolver registered
[    0.581636] Key type id_legacy registered
[    0.582408] 9p: Installing v9fs 9p2000 file system support
[    0.586897] Key type asymmetric registered
[    0.587663] Asymmetric key parser 'x509' registered
[    0.588554] Block layer SCSI generic (bsg) driver version 0.4 loaded (major 251)
[    0.589904] io scheduler mq-deadline registered
[    0.590730] io scheduler kyber registered
[    0.595421] input: Power Button as /devices/LNXSYSTM:00/LNXPWRBN:00/input/input0
[    0.596814] ACPI: button: Power Button [PWRF]
[    0.598997] Serial: 8250/16550 driver, 4 ports, IRQ sharing enabled
[    0.600262] 00:03: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A
[    0.602055] Non-volatile memory driver v1.3
[    0.602829] Linux agpgart interface v0.103
[    0.603604] ACPI: bus type drm_connector registered
[    0.616441] loop: module loaded
[    0.617699] scsi host0: ata_piix
[    0.618552] scsi host1: ata_piix
[    0.619151] ata1: PATA max MWDMA2 cmd 0x1f0 ctl 0x3f6 bmdma 0xc000 irq 14
[    0.620367] ata2: PATA max MWDMA2 cmd 0x170 ctl 0x376 bmdma 0xc008 irq 15
[    0.621703] e100: Intel(R) PRO/100 Network Driver
[    0.622763] e100: Copyright(c) 1999-2006 Intel Corporation
[    0.623904] e1000: Intel(R) PRO/1000 Network Driver
[    0.624798] e1000: Copyright (c) 1999-2006 Intel Corporation.
[    0.625848] e1000e: Intel(R) PRO/1000 Network Driver
[    0.626746] e1000e: Copyright(c) 1999 - 2015 Intel Corporation.
[    0.627816] sky2: driver version 1.30
[    0.628581] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
[    0.629773] ehci-pci: EHCI PCI platform driver
[    0.630584] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver
[    0.631701] ohci-pci: OHCI PCI platform driver
[    0.632512] uhci_hcd: USB Universal Host Controller Interface driver
[    0.633687] usbcore: registered new interface driver usblp
[    0.634686] usbcore: registered new interface driver usb-storage
[    0.635788] i8042: PNP: PS/2 Controller [PNP0303:KBD,PNP0f13:MOU] at 0x60,0x64 irq 1,12
[    0.638026] serio: i8042 KBD port at 0x60,0x64 irq 1
[    0.638941] serio: i8042 AUX port at 0x60,0x64 irq 12
[    0.640153] rtc_cmos 00:04: RTC can wake from S4
[    0.641733] input: AT Translated Set 2 keyboard as /devices/platform/i8042/serio0/input/input1
[    0.642559] rtc_cmos 00:04: registered as rtc0
[    0.645921] rtc_cmos 00:04: alarms up to one day, y3k, 242 bytes nvram, hpet irqs
[    0.649860] device-mapper: ioctl: 4.47.0-ioctl (2022-07-28) initialised: dm-devel@redhat.com
[    0.652752] hid: raw HID events driver (C) Jiri Kosina
[    0.654526] usbcore: registered new interface driver usbhid
[    0.655539] usbhid: USB HID core driver
[    0.656561] Initializing XFRM netlink socket
[    0.657532] NET: Registered PF_INET6 protocol family
[    0.658846] Segment Routing with IPv6
[    0.659538] In-situ OAM (IOAM) with IPv6
[    0.660272] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver
[    0.661440] NET: Registered PF_PACKET protocol family
[    0.662395] 9pnet: Installing 9P2000 support
[    0.663196] Key type dns_resolver registered
[    0.670887] IPI shorthand broadcast: enabled
[    0.671720] sched_clock: Marking stable (665092110, 6605125)->(723079606, -51382371)
[    0.673571] registered taskstats version 1
[    0.674338] Loading compiled-in X.509 certificates
[    0.675423] cryptomgr_test (366) used greatest stack depth: 15728 bytes left
[    0.677936] cryptomgr_test (380) used greatest stack depth: 15568 bytes left
[    0.680677] PM:   Magic number: 10:617:145
[    0.681475] ata_port ata2: hash matches
[    0.682187]  ata2: hash matches
[    0.682816] acpi device:16: hash matches
[    0.683554] printk: console [netcon0] enabled
[    0.684341] netconsole: network logging started
[    0.685826] cfg80211: Loading compiled-in X.509 certificates for regulatory database
[    0.687907] modprobe (389) used greatest stack depth: 14272 bytes left
[    0.688964] cfg80211: Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
[    0.691434] platform regulatory.0: Direct firmware load for regulatory.db failed with error -2
[    0.693029] cfg80211: failed to load regulatory.db
[    0.693937] ALSA device list:
[    0.694496]   No soundcards found.
[    0.777773] Freeing unused kernel image (initmem) memory: 1540K
[    0.779535] Write protecting the kernel read-only data: 24576k
[    0.782175] Freeing unused kernel image (text/rodata gap) memory: 2032K
[    0.784549] Freeing unused kernel image (rodata/data gap) memory: 1072K
[    0.786428] Run /init as init process
Starting syslogd: [    0.797052] start-stop-daem (408) used greatest stack depth: 14184 bytes left
OK
Starting klogd: OK
Running sysctl: [    0.803599] S02sysctl (416) used greatest stack depth: 14048 bytes left
OK
Saving random seed: OK
Starting network: [    0.812740] ip (433) used greatest stack depth: 12920 bytes left
OK
Linux version 6.0.0-rc7-00182-g3801f027b5b5 (ndesaulniers@ndesaulniers-desktop.svl.corp.google.com) (Nick Desaulniers clang version 16.0.0 (git@github.com:llvm/llvm-project.git 4b104c2ec016b3722190cdd67d23de6adc552f9e), LLD 16.0.0) #2 SMP PREEMPT_DYNAMIC Tue Oct 4 14:05:52 PDT 2022
Stopping network: OK
Saving random seed: OK
Stopping klogd: OK
Stopping syslogd: OK
umount: devtmpfs busy - remounted read-only
umount: can't unmount /: Invalid argument
\rThe system is going down NOW!
\rSent SIGTERM to all processes
[    1.275742] input: ImExPS/2 Generic Explorer Mouse as /devices/platform/i8042/serio1/input/input3
\rSent SIGKILL to all processes
\rRequesting system poweroff
[    2.851951] ACPI: PM: Preparing to enter system sleep state S5
[    2.853012] reboot: Power down

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
  2022-10-04 21:27     ` Nick Desaulniers
@ 2022-10-06 17:56       ` Nick Desaulniers
  0 siblings, 0 replies; 8+ messages in thread
From: Nick Desaulniers @ 2022-10-06 17:56 UTC (permalink / raw)
  To: Peter Zijlstra, Borislav Petkov
  Cc: Nathan Chancellor, Thomas Gleixner, Ingo Molnar, Dave Hansen,
	x86, linux-kernel, llvm, Moreira, Joao

On Tue, Oct 4, 2022 at 2:27 PM Nick Desaulniers <ndesaulniers@google.com> wrote:
>
> On Thu, Aug 18, 2022 at 12:45 AM Peter Zijlstra <peterz@infradead.org> wrote:
> >
> > On Wed, Aug 17, 2022 at 03:08:01PM -0700, Nick Desaulniers wrote:
> > > On Wed, Aug 17, 2022 at 11:56 AM Nathan Chancellor <nathan@kernel.org> wrote:
> > > >
> > > > LLVM 16 will have support for this flag so move it out of the GCC-only
> > > > block to allow LLVM builds to take advantage of it.
> > > >
> > > > Link: https://github.com/ClangBuiltLinux/linux/issues/1665
> > > > Link: https://github.com/llvm/llvm-project/commit/6f867f9102838ebe314c1f3661fdf95700386e5a
> > > > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > >
> > > Thanks for the patch, I also boot tested this (in virtual guests for
> > > both 64b and 32b).
> > >
> > > Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
> > > Tested-by: Nick Desaulniers <ndesaulniers@google.com>
> > >
> > > Peter, is there anything else special about these prefixes needed to
> > > make use of them?
> >
> > The thing to do is boot with (warning, *very* verbose):
> >
> >         "spectre_v2=retpoline,lfence debug-alternative"
>
> ah, debug-alternative also needs loglevel=8. Was wondering why I
> wasn't getting anything new in the output...
>
> No kidding on the verbosity. For a defconfig, booting in QEMU and
> redirecting stdout to a file produced a 95257 line file.
>
> $ grep "SMP alternatives: retpoline" /tmp/log.txt | wc -l
> 18623
> $ du -h /tmp/log.txt
> 11M /tmp/log.txt
> $ grep "SMP alternatives: retpoline" /tmp/log.txt | tr -s ' '
> ...
> [ 56.736261] SMP alternatives: retpoline at:
> pm_check_save_msr+0x30/0x80 (ffffffffaa34b320) len: 6 to:
> __x86_indirect_thunk_r11+0x0/0x20
> $ grep "SMP alternatives: retpoline" /tmp/log.txt | tr -s ' ' | cut -d
> ' ' -f 10 | sort -u
> 5
> 6
> # Curious about the 5, looks like there's only 2 cases:
> $ grep "SMP alternatives: retpoline" /tmp/log.txt | grep 'len: 5'
> [    0.147261] SMP alternatives: retpoline at: ret_from_fork+0x1d/0x30
> (ffffffffa9601dbd) len: 5 to: __x86_indirect_thunk_rbx+0x0/0x20
> [    2.274261] SMP alternatives: retpoline at: __efi_call+0x23/0x30
> (ffffffffa9671663) len: 5 to: __x86_indirect_thunk_rdi+0x0/0x20

For the typical len 6 case:
[    0.710345] SMP alternatives: ffffffffa2b6d1f2: orig: 2e e8 88 5f 09 ff
[    0.711345] SMP alternatives: ffffffffa2b6d1f2: repl: 0f ae e8 41 ff d3
[    0.712345] SMP alternatives: retpoline at: setup_arch+0x4b0/0x58c
(ffffffffa2b6d2ef) len: 6 to: __x86_indirect_thunk_r11+0x0/0x20

2e is the cs prefix, so it looks like this is working.

https://github.com/llvm/llvm-project/issues/58201
:^)

> Is there anything else I can do to help verify Nathan and Joao's
> patches?  If not, would you mind Ack'ing this so the x86 maintainers
> can pick it up?

I see it's already been Acked, so NVM. This is good to go.

> Noting that Joao's patch was applied in my tests.
> https://reviews.llvm.org/D134915
> I plan to land Joao's patch imminently.

Merging.
-- 
Thanks,
~Nick Desaulniers

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block
  2022-08-17 18:54 [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block Nathan Chancellor
  2022-08-17 22:08 ` Nick Desaulniers
@ 2022-08-18  7:49 ` Peter Zijlstra
  1 sibling, 0 replies; 8+ messages in thread
From: Peter Zijlstra @ 2022-08-18  7:49 UTC (permalink / raw)
  To: Nathan Chancellor
  Cc: Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, x86,
	Nick Desaulniers, linux-kernel, llvm

On Wed, Aug 17, 2022 at 11:54:11AM -0700, Nathan Chancellor wrote:
> LLVM 16 will have support for this flag so move it out of the GCC-only
> block to allow LLVM builds to take advantage of it.
> 
> Link: https://github.com/ClangBuiltLinux/linux/issues/1665
> Link: https://github.com/llvm/llvm-project/commit/6f867f9102838ebe314c1f3661fdf95700386e5a
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>

Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

> ---
> 
> I was not sure if this information is relevant for the commit message
> but I can boot without any issues on my test machines (two Intel and one
> AMD).
> 
>  arch/x86/Makefile | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/Makefile b/arch/x86/Makefile
> index 7854685c5f25..987da87c7778 100644
> --- a/arch/x86/Makefile
> +++ b/arch/x86/Makefile
> @@ -14,13 +14,13 @@ endif
>  
>  ifdef CONFIG_CC_IS_GCC
>  RETPOLINE_CFLAGS	:= $(call cc-option,-mindirect-branch=thunk-extern -mindirect-branch-register)
> -RETPOLINE_CFLAGS	+= $(call cc-option,-mindirect-branch-cs-prefix)
>  RETPOLINE_VDSO_CFLAGS	:= $(call cc-option,-mindirect-branch=thunk-inline -mindirect-branch-register)
>  endif
>  ifdef CONFIG_CC_IS_CLANG
>  RETPOLINE_CFLAGS	:= -mretpoline-external-thunk
>  RETPOLINE_VDSO_CFLAGS	:= -mretpoline
>  endif
> +RETPOLINE_CFLAGS	+= $(call cc-option,-mindirect-branch-cs-prefix)

Thanks for using the same option!

^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2022-10-06 17:57 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-08-17 18:54 [PATCH] x86/build: Move '-mindirect-branch-cs-prefix' out of GCC-only block Nathan Chancellor
2022-08-17 22:08 ` Nick Desaulniers
2022-08-18  7:45   ` Peter Zijlstra
2022-08-18  8:00     ` Peter Zijlstra
2022-09-30 18:10     ` Nick Desaulniers
2022-10-04 21:27     ` Nick Desaulniers
2022-10-06 17:56       ` Nick Desaulniers
2022-08-18  7:49 ` Peter Zijlstra

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).