mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-05 07:26:48 +07:00
e4a84be6f0
The Intel(R) Xeon Phi(TM) Processor x200 Family (codename: Knights Landing) has an erratum where a processor thread setting the Accessed or Dirty bits may not do so atomically against its checks for the Present bit. This may cause a thread (which is about to page fault) to set A and/or D, even though the Present bit had already been atomically cleared. These bits are truly "stray". In the case of the Dirty bit, the thread associated with the stray set was *not* allowed to write to the page. This means that we do not have to launder the bit(s); we can simply ignore them. If the PTE is used for storing a swap index or a NUMA migration index, the A bit could be misinterpreted as part of the swap type. The stray bits being set cause a software-cleared PTE to be interpreted as a swap entry. In some cases (like when the swap index ends up being for a non-existent swapfile), the kernel detects the stray value and WARN()s about it, but there is no guarantee that the kernel can always detect it. When we have 64-bit PTEs (64-bit mode or 32-bit PAE), we were able to move the swap PTE format around to avoid these troublesome bits. But, 32-bit non-PAE is tight on bits. So, disallow it from running on this hardware. I can't imagine anyone wanting to run 32-bit non-highmem kernels on this hardware, but disallowing them from running entirely is surely the safe thing to do. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Dave Hansen <dave@sr71.net> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Luis R. Rodriguez <mcgrof@suse.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Toshi Kani <toshi.kani@hp.com> Cc: dave.hansen@intel.com Cc: linux-mm@kvack.org Cc: mhocko@suse.com Link: http://lkml.kernel.org/r/20160708001914.D0B50110@viggo.jf.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
221 lines
5.8 KiB
C
221 lines
5.8 KiB
C
/* -*- linux-c -*- ------------------------------------------------------- *
|
|
*
|
|
* Copyright (C) 1991, 1992 Linus Torvalds
|
|
* Copyright 2007 rPath, Inc. - All Rights Reserved
|
|
*
|
|
* This file is part of the Linux kernel, and is made available under
|
|
* the terms of the GNU General Public License version 2.
|
|
*
|
|
* ----------------------------------------------------------------------- */
|
|
|
|
/*
|
|
* Check for obligatory CPU features and abort if the features are not
|
|
* present. This code should be compilable as 16-, 32- or 64-bit
|
|
* code, so be very careful with types and inline assembly.
|
|
*
|
|
* This code should not contain any messages; that requires an
|
|
* additional wrapper.
|
|
*
|
|
* As written, this code is not safe for inclusion into the kernel
|
|
* proper (after FPU initialization, in particular).
|
|
*/
|
|
|
|
#ifdef _SETUP
|
|
# include "boot.h"
|
|
#endif
|
|
#include <linux/types.h>
|
|
#include <asm/intel-family.h>
|
|
#include <asm/processor-flags.h>
|
|
#include <asm/required-features.h>
|
|
#include <asm/msr-index.h>
|
|
#include "string.h"
|
|
|
|
static u32 err_flags[NCAPINTS];
|
|
|
|
static const int req_level = CONFIG_X86_MINIMUM_CPU_FAMILY;
|
|
|
|
static const u32 req_flags[NCAPINTS] =
|
|
{
|
|
REQUIRED_MASK0,
|
|
REQUIRED_MASK1,
|
|
0, /* REQUIRED_MASK2 not implemented in this file */
|
|
0, /* REQUIRED_MASK3 not implemented in this file */
|
|
REQUIRED_MASK4,
|
|
0, /* REQUIRED_MASK5 not implemented in this file */
|
|
REQUIRED_MASK6,
|
|
0, /* REQUIRED_MASK7 not implemented in this file */
|
|
};
|
|
|
|
#define A32(a, b, c, d) (((d) << 24)+((c) << 16)+((b) << 8)+(a))
|
|
|
|
static int is_amd(void)
|
|
{
|
|
return cpu_vendor[0] == A32('A', 'u', 't', 'h') &&
|
|
cpu_vendor[1] == A32('e', 'n', 't', 'i') &&
|
|
cpu_vendor[2] == A32('c', 'A', 'M', 'D');
|
|
}
|
|
|
|
static int is_centaur(void)
|
|
{
|
|
return cpu_vendor[0] == A32('C', 'e', 'n', 't') &&
|
|
cpu_vendor[1] == A32('a', 'u', 'r', 'H') &&
|
|
cpu_vendor[2] == A32('a', 'u', 'l', 's');
|
|
}
|
|
|
|
static int is_transmeta(void)
|
|
{
|
|
return cpu_vendor[0] == A32('G', 'e', 'n', 'u') &&
|
|
cpu_vendor[1] == A32('i', 'n', 'e', 'T') &&
|
|
cpu_vendor[2] == A32('M', 'x', '8', '6');
|
|
}
|
|
|
|
static int is_intel(void)
|
|
{
|
|
return cpu_vendor[0] == A32('G', 'e', 'n', 'u') &&
|
|
cpu_vendor[1] == A32('i', 'n', 'e', 'I') &&
|
|
cpu_vendor[2] == A32('n', 't', 'e', 'l');
|
|
}
|
|
|
|
/* Returns a bitmask of which words we have error bits in */
|
|
static int check_cpuflags(void)
|
|
{
|
|
u32 err;
|
|
int i;
|
|
|
|
err = 0;
|
|
for (i = 0; i < NCAPINTS; i++) {
|
|
err_flags[i] = req_flags[i] & ~cpu.flags[i];
|
|
if (err_flags[i])
|
|
err |= 1 << i;
|
|
}
|
|
|
|
return err;
|
|
}
|
|
|
|
/*
|
|
* Returns -1 on error.
|
|
*
|
|
* *cpu_level is set to the current CPU level; *req_level to the required
|
|
* level. x86-64 is considered level 64 for this purpose.
|
|
*
|
|
* *err_flags_ptr is set to the flags error array if there are flags missing.
|
|
*/
|
|
int check_cpu(int *cpu_level_ptr, int *req_level_ptr, u32 **err_flags_ptr)
|
|
{
|
|
int err;
|
|
|
|
memset(&cpu.flags, 0, sizeof cpu.flags);
|
|
cpu.level = 3;
|
|
|
|
if (has_eflag(X86_EFLAGS_AC))
|
|
cpu.level = 4;
|
|
|
|
get_cpuflags();
|
|
err = check_cpuflags();
|
|
|
|
if (test_bit(X86_FEATURE_LM, cpu.flags))
|
|
cpu.level = 64;
|
|
|
|
if (err == 0x01 &&
|
|
!(err_flags[0] &
|
|
~((1 << X86_FEATURE_XMM)|(1 << X86_FEATURE_XMM2))) &&
|
|
is_amd()) {
|
|
/* If this is an AMD and we're only missing SSE+SSE2, try to
|
|
turn them on */
|
|
|
|
u32 ecx = MSR_K7_HWCR;
|
|
u32 eax, edx;
|
|
|
|
asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
|
|
eax &= ~(1 << 15);
|
|
asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
|
|
|
|
get_cpuflags(); /* Make sure it really did something */
|
|
err = check_cpuflags();
|
|
} else if (err == 0x01 &&
|
|
!(err_flags[0] & ~(1 << X86_FEATURE_CX8)) &&
|
|
is_centaur() && cpu.model >= 6) {
|
|
/* If this is a VIA C3, we might have to enable CX8
|
|
explicitly */
|
|
|
|
u32 ecx = MSR_VIA_FCR;
|
|
u32 eax, edx;
|
|
|
|
asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
|
|
eax |= (1<<1)|(1<<7);
|
|
asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
|
|
|
|
set_bit(X86_FEATURE_CX8, cpu.flags);
|
|
err = check_cpuflags();
|
|
} else if (err == 0x01 && is_transmeta()) {
|
|
/* Transmeta might have masked feature bits in word 0 */
|
|
|
|
u32 ecx = 0x80860004;
|
|
u32 eax, edx;
|
|
u32 level = 1;
|
|
|
|
asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx));
|
|
asm("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx));
|
|
asm("cpuid"
|
|
: "+a" (level), "=d" (cpu.flags[0])
|
|
: : "ecx", "ebx");
|
|
asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx));
|
|
|
|
err = check_cpuflags();
|
|
} else if (err == 0x01 &&
|
|
!(err_flags[0] & ~(1 << X86_FEATURE_PAE)) &&
|
|
is_intel() && cpu.level == 6 &&
|
|
(cpu.model == 9 || cpu.model == 13)) {
|
|
/* PAE is disabled on this Pentium M but can be forced */
|
|
if (cmdline_find_option_bool("forcepae")) {
|
|
puts("WARNING: Forcing PAE in CPU flags\n");
|
|
set_bit(X86_FEATURE_PAE, cpu.flags);
|
|
err = check_cpuflags();
|
|
}
|
|
else {
|
|
puts("WARNING: PAE disabled. Use parameter 'forcepae' to enable at your own risk!\n");
|
|
}
|
|
}
|
|
if (!err)
|
|
err = check_knl_erratum();
|
|
|
|
if (err_flags_ptr)
|
|
*err_flags_ptr = err ? err_flags : NULL;
|
|
if (cpu_level_ptr)
|
|
*cpu_level_ptr = cpu.level;
|
|
if (req_level_ptr)
|
|
*req_level_ptr = req_level;
|
|
|
|
return (cpu.level < req_level || err) ? -1 : 0;
|
|
}
|
|
|
|
int check_knl_erratum(void)
|
|
{
|
|
/*
|
|
* First check for the affected model/family:
|
|
*/
|
|
if (!is_intel() ||
|
|
cpu.family != 6 ||
|
|
cpu.model != INTEL_FAM6_XEON_PHI_KNL)
|
|
return 0;
|
|
|
|
/*
|
|
* This erratum affects the Accessed/Dirty bits, and can
|
|
* cause stray bits to be set in !Present PTEs. We have
|
|
* enough bits in our 64-bit PTEs (which we have on real
|
|
* 64-bit mode or PAE) to avoid using these troublesome
|
|
* bits. But, we do not have enough space in our 32-bit
|
|
* PTEs. So, refuse to run on 32-bit non-PAE kernels.
|
|
*/
|
|
if (IS_ENABLED(CONFIG_X86_64) || IS_ENABLED(CONFIG_X86_PAE))
|
|
return 0;
|
|
|
|
puts("This 32-bit kernel can not run on this Xeon Phi x200\n"
|
|
"processor due to a processor erratum. Use a 64-bit\n"
|
|
"kernel, or enable PAE in this 32-bit kernel.\n\n");
|
|
|
|
return -1;
|
|
}
|
|
|
|
|