public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Borislav Petkov <bp@alien8.de>
To: Ingo Molnar <mingo@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>
Subject: [PATCH 04/20] x86/mm/pat: Use 7th PAT MSR slot for Write-Through PAT type
Date: Thu,  4 Jun 2015 18:55:12 +0200	[thread overview]
Message-ID: <1433436928-31903-5-git-send-email-bp@alien8.de> (raw)
In-Reply-To: <1433436928-31903-1-git-send-email-bp@alien8.de>

From: Toshi Kani <toshi.kani@hp.com>

Assign Write-Through type to the PA7 slot in the PAT MSR when the
processor is not affected by PAT errata. The PA7 slot is chosen to
improve robustness in the presence of errata that might cause the high
PAT bit to be ignored. This way a buggy PA7 slot access will hit the PA3
slot, which is UC, so at worst we lose performance without causing a
correctness issue.

The following Intel processors are affected by the PAT errata.

  Errata               CPUID
  ----------------------------------------------------
  Pentium 2, A52       family 0x6, model 0x5
  Pentium 3, E27       family 0x6, model 0x7, 0x8
  Pentium 3 Xenon, G26 family 0x6, model 0x7, 0x8, 0xa
  Pentium M, Y26       family 0x6, model 0x9
  Pentium M 90nm, X9   family 0x6, model 0xd
  Pentium 4, N46       family 0xf, model 0x0

Instead of making sharp boundary checks, we remain conservative and
exclude all Pentium 2, 3, M and 4 family processors. For those,
_PAGE_CACHE_MODE_WT is redirected to UC- per the default setup in
__cachemode2pte_tbl[].

Signed-off-by: Toshi Kani <toshi.kani@hp.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: arnd@arndb.de
Cc: Elliott@hp.com
Cc: hch@lst.de
Cc: hmh@hmh.eng.br
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: jgross@suse.com
Cc: konrad.wilk@oracle.com
Cc: linux-mm <linux-mm@kvack.org>
Cc: linux-nvdimm@lists.01.org
Cc: Luis R. Rodriguez <mcgrof@suse.com>
Cc: stefan.bader@canonical.com
Cc: x86-ml <x86@kernel.org>
Cc: yigal@plexistor.com
Link: https://lkml.kernel.org/r/1433187393-22688-2-git-send-email-toshi.kani@hp.com
Signed-off-by: Borislav Petkov <bp@suse.de>
---
 arch/x86/mm/pat.c | 59 ++++++++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 50 insertions(+), 9 deletions(-)

diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
index f89e460c55a8..59ab1a0fe21b 100644
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -235,6 +235,7 @@ static void pat_ap_init(u64 pat)
 void pat_init(void)
 {
 	u64 pat;
+	struct cpuinfo_x86 *c = &boot_cpu_data;
 
 	if (!pat_enabled()) {
 		/*
@@ -244,7 +245,7 @@ void pat_init(void)
 		 * has PAT but the "nopat" boot option has been specified. This
 		 * emulated PAT table is used when MSR_IA32_CR_PAT returns 0.
 		 *
-		 * PTE encoding used:
+		 * PTE encoding:
 		 *
 		 *       PCD
 		 *       |PWT  PAT
@@ -259,21 +260,61 @@ void pat_init(void)
 		 */
 		pat = PAT(0, WB) | PAT(1, WT) | PAT(2, UC_MINUS) | PAT(3, UC) |
 		      PAT(4, WB) | PAT(5, WT) | PAT(6, UC_MINUS) | PAT(7, UC);
-	} else {
+
+	} else if ((c->x86_vendor == X86_VENDOR_INTEL) &&
+		   (((c->x86 == 0x6) && (c->x86_model <= 0xd)) ||
+		    ((c->x86 == 0xf) && (c->x86_model <= 0x6)))) {
 		/*
-		 * PTE encoding used in Linux:
+		 * PAT support with the lower four entries. Intel Pentium 2,
+		 * 3, M, and 4 are affected by PAT errata, which makes the
+		 * upper four entries unusable. To be on the safe side, we don't
+		 * use those.
+		 *
+		 *  PTE encoding:
 		 *      PAT
 		 *      |PCD
-		 *      ||PWT
-		 *      |||
-		 *      000 WB          _PAGE_CACHE_WB
-		 *      001 WC          _PAGE_CACHE_WC
-		 *      010 UC-         _PAGE_CACHE_UC_MINUS
-		 *      011 UC          _PAGE_CACHE_UC
+		 *      ||PWT  PAT
+		 *      |||    slot
+		 *      000    0    WB : _PAGE_CACHE_MODE_WB
+		 *      001    1    WC : _PAGE_CACHE_MODE_WC
+		 *      010    2    UC-: _PAGE_CACHE_MODE_UC_MINUS
+		 *      011    3    UC : _PAGE_CACHE_MODE_UC
 		 * PAT bit unused
+		 *
+		 * NOTE: When WT or WP is used, it is redirected to UC- per
+		 * the default setup in __cachemode2pte_tbl[].
 		 */
 		pat = PAT(0, WB) | PAT(1, WC) | PAT(2, UC_MINUS) | PAT(3, UC) |
 		      PAT(4, WB) | PAT(5, WC) | PAT(6, UC_MINUS) | PAT(7, UC);
+	} else {
+		/*
+		 * Full PAT support.  We put WT in slot 7 to improve
+		 * robustness in the presence of errata that might cause
+		 * the high PAT bit to be ignored.  This way, a buggy slot 7
+		 * access will hit slot 3, and slot 3 is UC, so at worst
+		 * we lose performance without causing a correctness issue.
+		 * Pentium 4 erratum N46 is an example for such an erratum,
+		 * although we try not to use PAT at all on affected CPUs.
+		 *
+		 *  PTE encoding:
+		 *      PAT
+		 *      |PCD
+		 *      ||PWT  PAT
+		 *      |||    slot
+		 *      000    0    WB : _PAGE_CACHE_MODE_WB
+		 *      001    1    WC : _PAGE_CACHE_MODE_WC
+		 *      010    2    UC-: _PAGE_CACHE_MODE_UC_MINUS
+		 *      011    3    UC : _PAGE_CACHE_MODE_UC
+		 *      100    4    WB : Reserved
+		 *      101    5    WC : Reserved
+		 *      110    6    UC-: Reserved
+		 *      111    7    WT : _PAGE_CACHE_MODE_WT
+		 *
+		 * The reserved slots are unused, but mapped to their
+		 * corresponding types in the presence of PAT errata.
+		 */
+		pat = PAT(0, WB) | PAT(1, WC) | PAT(2, UC_MINUS) | PAT(3, UC) |
+		      PAT(4, WB) | PAT(5, WC) | PAT(6, UC_MINUS) | PAT(7, WT);
 	}
 
 	if (!boot_cpu_done) {
-- 
2.3.5


  parent reply	other threads:[~2015-06-04 16:55 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-04 16:55 [PATCH 00/20] tip queue 2015-06-04 Borislav Petkov
2015-06-04 16:55 ` [PATCH 01/20] x86/mm/pat: Untangle pat_init() Borislav Petkov
2015-06-07 17:39   ` [tip:x86/mm] " tip-bot for Borislav Petkov
2015-06-04 16:55 ` [PATCH 02/20] x86/mm/pat: Emulate PAT when it is disabled Borislav Petkov
2015-06-07 17:39   ` [tip:x86/mm] " tip-bot for Borislav Petkov
2015-06-04 16:55 ` [PATCH 03/20] x86/mm/pat: Remove pat_enabled() checks Borislav Petkov
2015-06-07 17:40   ` [tip:x86/mm] " tip-bot for Borislav Petkov
2015-06-04 16:55 ` Borislav Petkov [this message]
2015-06-04 16:55 ` [PATCH 05/20] x86/mm/pat: Change reserve_memtype() for Write-Through type Borislav Petkov
2015-06-07 17:40   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 06/20] x86/mm: Teach is_new_memtype_allowed() about " Borislav Petkov
2015-06-07 17:41   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 07/20] x86/mm, asm-generic: Add ioremap_wt() for creating Write-Through mappings Borislav Petkov
2015-06-07 17:41   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 08/20] arch/*/io.h: Add ioremap_wt() to all architectures Borislav Petkov
2015-06-07 17:41   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 09/20] video/fbdev, asm/io.h: Remove ioremap_writethrough() Borislav Petkov
2015-06-07 17:41   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 10/20] x86/mm/pat: Add pgprot_writethrough() Borislav Petkov
2015-06-07 17:42   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 11/20] x86/mm/pat: Extend set_page_memtype() to support Write-Through type Borislav Petkov
2015-06-07 17:42   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 12/20] x86/mm: Add set_memory_wt() for " Borislav Petkov
2015-06-07 17:42   ` [tip:x86/mm] x86/mm/pat: " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 13/20] drivers/block/pmem: Map NVDIMM in Write-Through mode Borislav Petkov
2015-06-07 17:43   ` [tip:x86/mm] " tip-bot for Toshi Kani
2015-06-04 16:55 ` [PATCH 14/20] x86/mce: Add Local MCE definitions Borislav Petkov
2015-06-07 17:43   ` [tip:x86/core] " tip-bot for Ashok Raj
2015-06-04 16:55 ` [PATCH 15/20] x86/mce: Add infrastructure to support Local MCE Borislav Petkov
2015-06-07 17:43   ` [tip:x86/core] " tip-bot for Ashok Raj
2015-06-04 16:55 ` [PATCH 16/20] x86/mce: Handle Local MCE events Borislav Petkov
2015-06-07 17:44   ` [tip:x86/core] " tip-bot for Ashok Raj
2015-06-04 16:55 ` [PATCH 17/20] x86: Kill CONFIG_X86_HT Borislav Petkov
2015-06-07 17:44   ` [tip:x86/core] " tip-bot for Borislav Petkov
2015-06-04 16:55 ` [PATCH 18/20] x86/uapi: Do not export <asm/msr-index.h> as part of the user API headers Borislav Petkov
2015-06-07 17:44   ` [tip:x86/core] " tip-bot for Borislav Petkov
2015-06-04 16:55 ` [PATCH 19/20] x86/microcode: Disable builtin microcode loading on 32-bit for now Borislav Petkov
2015-06-07 17:45   ` [tip:x86/microcode] " tip-bot for Borislav Petkov
2015-06-04 16:55 ` [PATCH 20/20] x86/microcode: Correct variables type Borislav Petkov
2015-06-07 17:45   ` [tip:x86/microcode] x86/microcode: Correct CPU family related variable types tip-bot for Andy Shevchenko

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1433436928-31903-5-git-send-email-bp@alien8.de \
    --to=bp@alien8.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox