Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1333117 > unrolled thread
| Started by | Dave Hansen <dave@sr71.net> |
|---|---|
| First post | 2016-02-12 22:10 +0100 |
| Last post | 2016-02-16 10:30 +0100 |
| Articles | 20 on this page of 43 — 7 participants |
Back to article view | Back to linux.kernel
[PATCH 00/33] x86: Memory Protection Keys (v10) Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[PATCH 18/33] x86, mm: simplify get_user_pages() PTE bit handling Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] x86/mm/gup: Simplify get_user_pages() PTE bit handling tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:30 +0100
[PATCH 23/33] x86, pkeys: dump PKRU with other kernel registers Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] x86/mm/pkeys: Dump PKRU with other kernel registers tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:30 +0100
[PATCH 30/33] x86, fpu: allow setting of XSAVE state Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] x86/fpu: Allow setting of XSAVE state tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:30 +0100
[PATCH 03/33] mm, gup: switch callers of get_user_pages() to not pass tsk/mm Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] mm/gup: Switch all callers of get_user_pages() to not pass tsk/mm tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:20 +0100
[PATCH 19/33] x86, pkeys: check VMAs and PTEs for protection keys Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] mm/gup, x86/mm/pkeys: Check VMAs and PTEs for protection keys tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:30 +0100
[PATCH 26/33] x86, pkeys: actually enable Memory Protection Keys in CPU Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] x86/mm/pkeys: Actually enable Memory Protection Keys in the CPU tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:30 +0100
[PATCH 07/33] x86, pkeys: define new CR4 bit Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] x86/cpu, x86/mm/pkeys: Define new CR4 bit tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:20 +0100
[PATCH 13/33] x86, pkeys: pass VMA down in to fault signal generation code Dave Hansen <dave@sr71.net> - 2016-02-12 22:10 +0100
[tip:mm/pkeys] x86/mm/pkeys: Pass VMA down in to fault signal generation code tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:30 +0100
[PATCH 11/33] x86, pkeys: store protection in high VMA flags Dave Hansen <dave@sr71.net> - 2016-02-12 22:20 +0100
[tip:mm/pkeys] mm/core, x86/mm/pkeys: Store protection bits in high VMA flags tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:30 +0100
[PATCH 10/33] x86, pkeys: new page fault error code bit: PF_PK Dave Hansen <dave@sr71.net> - 2016-02-12 22:20 +0100
[tip:mm/pkeys] x86/mm/pkeys: Add new 'PF_PK' page fault error code bit tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:20 +0100
[PATCH 09/33] x86, pkeys: PTE bits for storing protection key Dave Hansen <dave@sr71.net> - 2016-02-12 22:20 +0100
[tip:mm/pkeys] x86/mm/pkeys: Add PTE bits for storing protection key tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:20 +0100
[PATCH 06/33] x86, pkeys: cpuid bit definition Dave Hansen <dave@sr71.net> - 2016-02-12 22:20 +0100
[tip:mm/pkeys] x86/cpufeature, x86/mm/pkeys: Add protection keys related CPUID definitions tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:20 +0100
[PATCH 02/33] mm: overload get_user_pages() functions Dave Hansen <dave@sr71.net> - 2016-02-12 22:20 +0100
Re: [PATCH 02/33] mm: overload get_user_pages() functions Ingo Molnar <mingo@kernel.org> - 2016-02-16 09:40 +0100
Re: [PATCH 02/33] mm: overload get_user_pages() functions Dave Hansen <dave@sr71.net> - 2016-02-17 19:20 +0100
[tip:mm/pkeys] mm/gup: Overload get_user_pages() functions tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:20 +0100
[PATCH 01/33] mm: introduce get_user_pages_remote() Dave Hansen <dave@sr71.net> - 2016-02-12 22:20 +0100
Re: [PATCH 01/33] mm: introduce get_user_pages_remote() Balbir Singh <bsingharora@gmail.com> - 2016-02-15 07:10 +0100
Re: [PATCH 01/33] mm: introduce get_user_pages_remote() Dave Hansen <dave@sr71.net> - 2016-02-15 17:30 +0100
Re: [PATCH 01/33] mm: introduce get_user_pages_remote() Srikar Dronamraju <srikar@linux.vnet.ibm.com> - 2016-02-15 07:20 +0100
[tip:x86/pkeys] mm/gup: Introduce get_user_pages_remote() tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-16 13:20 +0100
Re: [tip:x86/pkeys] mm/gup: Introduce get_user_pages_remote() Konstantin Khlebnikov <koct9i@gmail.com> - 2016-02-20 07:30 +0100
[PATCH 05/33] x86, pkeys: Add Kconfig option Dave Hansen <dave@sr71.net> - 2016-02-12 22:20 +0100
[tip:mm/pkeys] x86/mm/pkeys: Add Kconfig option tip-bot for Dave Hansen <tipbot@zytor.com> - 2016-02-18 21:20 +0100
[PATCH] x86/mm/pkeys: Do not enable them by default Borislav Petkov <bp@alien8.de> - 2016-02-19 12:30 +0100
Re: [PATCH] x86/mm/pkeys: Do not enable them by default Dave Hansen <dave@sr71.net> - 2016-02-19 18:20 +0100
Re: [PATCH] x86/mm/pkeys: Do not enable them by default Borislav Petkov <bp@alien8.de> - 2016-02-19 18:30 +0100
Re: [PATCH] x86/mm/pkeys: Do not enable them by default Dave Hansen <dave@sr71.net> - 2016-02-19 18:50 +0100
Re: [PATCH] x86/mm/pkeys: Do not enable them by default Borislav Petkov <bp@alien8.de> - 2016-02-19 19:40 +0100
Re: [PATCH 00/33] x86: Memory Protection Keys (v10) Ingo Molnar <mingo@kernel.org> - 2016-02-16 10:30 +0100
Page 2 of 3 — ← Prev page 1 [2] 3 Next page →
| From | tip-bot for Dave Hansen <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-18 21:20 +0100 |
| Subject | [tip:mm/pkeys] x86/mm/pkeys: Add new 'PF_PK' page fault error code bit |
| Message-ID | <r3Ddo-2wj-35@gated-at.bofh.it> |
| In reply to | #1333129 |
Commit-ID: b3ecd51559ae7a8f40b10443773b9cd0e6a50f5e
Gitweb: http://git.kernel.org/tip/b3ecd51559ae7a8f40b10443773b9cd0e6a50f5e
Author: Dave Hansen <dave.hansen@linux.intel.com>
AuthorDate: Fri, 12 Feb 2016 13:02:07 -0800
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Thu, 18 Feb 2016 09:31:50 +0100
x86/mm/pkeys: Add new 'PF_PK' page fault error code bit
Note: "PK" is how the Intel SDM refers to this bit, so we also
use that nomenclature.
This only defines the bit, it does not plumb it anywhere to be
handled.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: linux-mm@kvack.org
Link: http://lkml.kernel.org/r/20160212210207.DA7B43E6@viggo.jf.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/mm/fault.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index eef44d9..9f72f9c 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -33,6 +33,7 @@
* bit 2 == 0: kernel-mode access 1: user-mode access
* bit 3 == 1: use of reserved bit detected
* bit 4 == 1: fault was an instruction fetch
+ * bit 5 == 1: protection keys block access
*/
enum x86_pf_error_code {
@@ -41,6 +42,7 @@ enum x86_pf_error_code {
PF_USER = 1 << 2,
PF_RSVD = 1 << 3,
PF_INSTR = 1 << 4,
+ PF_PK = 1 << 5,
};
/*
@@ -916,6 +918,12 @@ static int spurious_fault_check(unsigned long error_code, pte_t *pte)
if ((error_code & PF_INSTR) && !pte_exec(*pte))
return 0;
+ /*
+ * Note: We do not do lazy flushing on protection key
+ * changes, so no spurious fault will ever set PF_PK.
+ */
+ if ((error_code & PF_PK))
+ return 1;
return 1;
}
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-12 22:20 +0100 |
| Subject | [PATCH 09/33] x86, pkeys: PTE bits for storing protection key |
| Message-ID | <r1tia-3WT-7@gated-at.bofh.it> |
| In reply to | #1333117 |
From: Dave Hansen <dave.hansen@linux.intel.com> Previous documentation has referred to these 4 bits as "ignored". That means that software could have made use of them. But, as far as I know, the kernel never used them. They are still ignored when protection keys is not enabled, so they could theoretically still get used for software purposes. We also implement "empty" versions so that code that references to them can be optimized away by the compiler when the config option is not enabled. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> --- b/arch/x86/include/asm/pgtable_types.h | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-) diff -puN arch/x86/include/asm/pgtable_types.h~pkeys-04-ptebits arch/x86/include/asm/pgtable_types.h --- a/arch/x86/include/asm/pgtable_types.h~pkeys-04-ptebits 2016-02-12 10:44:17.672312467 -0800 +++ b/arch/x86/include/asm/pgtable_types.h 2016-02-12 10:44:17.675312604 -0800 @@ -20,13 +20,18 @@ #define _PAGE_BIT_SOFTW2 10 /* " */ #define _PAGE_BIT_SOFTW3 11 /* " */ #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ +#define _PAGE_BIT_SOFTW4 58 /* available for programmer */ +#define _PAGE_BIT_PKEY_BIT0 59 /* Protection Keys, bit 1/4 */ +#define _PAGE_BIT_PKEY_BIT1 60 /* Protection Keys, bit 2/4 */ +#define _PAGE_BIT_PKEY_BIT2 61 /* Protection Keys, bit 3/4 */ +#define _PAGE_BIT_PKEY_BIT3 62 /* Protection Keys, bit 4/4 */ +#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ + #define _PAGE_BIT_SPECIAL _PAGE_BIT_SOFTW1 #define _PAGE_BIT_CPA_TEST _PAGE_BIT_SOFTW1 #define _PAGE_BIT_HIDDEN _PAGE_BIT_SOFTW3 /* hidden by kmemcheck */ #define _PAGE_BIT_SOFT_DIRTY _PAGE_BIT_SOFTW3 /* software dirty tracking */ -#define _PAGE_BIT_SOFTW4 58 /* available for programmer */ -#define _PAGE_BIT_DEVMAP _PAGE_BIT_SOFTW4 -#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ +#define _PAGE_BIT_DEVMAP _PAGE_BIT_SOFTW4 /* If _PAGE_BIT_PRESENT is clear, we use these: */ /* - if the user mapped it with PROT_NONE; pte_present gives true */ @@ -47,6 +52,17 @@ #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) #define _PAGE_SPECIAL (_AT(pteval_t, 1) << _PAGE_BIT_SPECIAL) #define _PAGE_CPA_TEST (_AT(pteval_t, 1) << _PAGE_BIT_CPA_TEST) +#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS +#define _PAGE_PKEY_BIT0 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT0) +#define _PAGE_PKEY_BIT1 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT1) +#define _PAGE_PKEY_BIT2 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT2) +#define _PAGE_PKEY_BIT3 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT3) +#else +#define _PAGE_PKEY_BIT0 (_AT(pteval_t, 0)) +#define _PAGE_PKEY_BIT1 (_AT(pteval_t, 0)) +#define _PAGE_PKEY_BIT2 (_AT(pteval_t, 0)) +#define _PAGE_PKEY_BIT3 (_AT(pteval_t, 0)) +#endif #define __HAVE_ARCH_PTE_SPECIAL #ifdef CONFIG_KMEMCHECK _
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Dave Hansen <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-18 21:20 +0100 |
| Subject | [tip:mm/pkeys] x86/mm/pkeys: Add PTE bits for storing protection key |
| Message-ID | <r3Ddo-2wj-19@gated-at.bofh.it> |
| In reply to | #1333130 |
Commit-ID: 5c1d90f51027e197e1299ab1235a2fed78910905 Gitweb: http://git.kernel.org/tip/5c1d90f51027e197e1299ab1235a2fed78910905 Author: Dave Hansen <dave.hansen@linux.intel.com> AuthorDate: Fri, 12 Feb 2016 13:02:05 -0800 Committer: Ingo Molnar <mingo@kernel.org> CommitDate: Thu, 18 Feb 2016 09:31:44 +0100 x86/mm/pkeys: Add PTE bits for storing protection key Previous documentation has referred to these 4 bits as "ignored". That means that software could have made use of them. But, as far as I know, the kernel never used them. They are still ignored when protection keys is not enabled, so they could theoretically still get used for software purposes. We also implement "empty" versions so that code that references to them can be optimized away by the compiler when the config option is not enabled. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> 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: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20160212210205.81E33ED6@viggo.jf.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org> --- arch/x86/include/asm/pgtable_types.h | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index 4432ab7..cae10ba 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -20,13 +20,18 @@ #define _PAGE_BIT_SOFTW2 10 /* " */ #define _PAGE_BIT_SOFTW3 11 /* " */ #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ +#define _PAGE_BIT_SOFTW4 58 /* available for programmer */ +#define _PAGE_BIT_PKEY_BIT0 59 /* Protection Keys, bit 1/4 */ +#define _PAGE_BIT_PKEY_BIT1 60 /* Protection Keys, bit 2/4 */ +#define _PAGE_BIT_PKEY_BIT2 61 /* Protection Keys, bit 3/4 */ +#define _PAGE_BIT_PKEY_BIT3 62 /* Protection Keys, bit 4/4 */ +#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ + #define _PAGE_BIT_SPECIAL _PAGE_BIT_SOFTW1 #define _PAGE_BIT_CPA_TEST _PAGE_BIT_SOFTW1 #define _PAGE_BIT_HIDDEN _PAGE_BIT_SOFTW3 /* hidden by kmemcheck */ #define _PAGE_BIT_SOFT_DIRTY _PAGE_BIT_SOFTW3 /* software dirty tracking */ -#define _PAGE_BIT_SOFTW4 58 /* available for programmer */ -#define _PAGE_BIT_DEVMAP _PAGE_BIT_SOFTW4 -#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ +#define _PAGE_BIT_DEVMAP _PAGE_BIT_SOFTW4 /* If _PAGE_BIT_PRESENT is clear, we use these: */ /* - if the user mapped it with PROT_NONE; pte_present gives true */ @@ -47,6 +52,17 @@ #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) #define _PAGE_SPECIAL (_AT(pteval_t, 1) << _PAGE_BIT_SPECIAL) #define _PAGE_CPA_TEST (_AT(pteval_t, 1) << _PAGE_BIT_CPA_TEST) +#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS +#define _PAGE_PKEY_BIT0 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT0) +#define _PAGE_PKEY_BIT1 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT1) +#define _PAGE_PKEY_BIT2 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT2) +#define _PAGE_PKEY_BIT3 (_AT(pteval_t, 1) << _PAGE_BIT_PKEY_BIT3) +#else +#define _PAGE_PKEY_BIT0 (_AT(pteval_t, 0)) +#define _PAGE_PKEY_BIT1 (_AT(pteval_t, 0)) +#define _PAGE_PKEY_BIT2 (_AT(pteval_t, 0)) +#define _PAGE_PKEY_BIT3 (_AT(pteval_t, 0)) +#endif #define __HAVE_ARCH_PTE_SPECIAL #ifdef CONFIG_KMEMCHECK
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-12 22:20 +0100 |
| Subject | [PATCH 06/33] x86, pkeys: cpuid bit definition |
| Message-ID | <r1tia-3WT-11@gated-at.bofh.it> |
| In reply to | #1333117 |
From: Dave Hansen <dave.hansen@linux.intel.com> There are two CPUID bits for protection keys. One is for whether the CPU contains the feature, and the other will appear set once the OS enables protection keys. Specifically: Bit 04: OSPKE. If 1, OS has set CR4.PKE to enable Protection keys (and the RDPKRU/WRPKRU instructions) This is because userspace can not see CR4 contents, but it can see CPUID contents. X86_FEATURE_PKU is referred to as "PKU" in the hardware documentation: CPUID.(EAX=07H,ECX=0H):ECX.PKU [bit 3] X86_FEATURE_OSPKE is "OSPKU": CPUID.(EAX=07H,ECX=0H):ECX.OSPKE [bit 4] These are the first CPU features which need to look at the ECX word in CPUID leaf 0x7, so this patch also includes fetching that word in to the cpuinfo->x86_capability[] array. Add it to the disabled-features mask when its config option is off. Even though we are not using it here, we also extend the REQUIRED_MASK_BIT_SET() macro to keep it mirroring the DISABLED_MASK_BIT_SET() version. This means that in almost all code, you should use: cpu_has(c, X86_FEATURE_PKU) and *not* the CONFIG option. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> --- b/arch/x86/include/asm/cpufeature.h | 61 +++++++++++++++++++---------- b/arch/x86/include/asm/disabled-features.h | 15 +++++++ b/arch/x86/include/asm/required-features.h | 7 +++ b/arch/x86/kernel/cpu/common.c | 1 4 files changed, 63 insertions(+), 21 deletions(-) diff -puN arch/x86/include/asm/cpufeature.h~pkeys-01-cpuid arch/x86/include/asm/cpufeature.h --- a/arch/x86/include/asm/cpufeature.h~pkeys-01-cpuid 2016-02-12 10:44:16.315250432 -0800 +++ b/arch/x86/include/asm/cpufeature.h 2016-02-12 10:44:16.323250798 -0800 @@ -12,7 +12,7 @@ #include <asm/disabled-features.h> #endif -#define NCAPINTS 16 /* N 32-bit words worth of info */ +#define NCAPINTS 17 /* N 32-bit words worth of info */ #define NBUGINTS 1 /* N 32-bit bug flags */ /* @@ -269,6 +269,10 @@ #define X86_FEATURE_PAUSEFILTER (15*32+10) /* filtered pause intercept */ #define X86_FEATURE_PFTHRESHOLD (15*32+12) /* pause filter threshold */ +/* Intel-defined CPU features, CPUID level 0x00000007:0 (ecx), word 16 */ +#define X86_FEATURE_PKU (16*32+ 3) /* Protection Keys for Userspace */ +#define X86_FEATURE_OSPKE (16*32+ 4) /* OS Protection Keys Enable */ + /* * BUG word(s) */ @@ -307,6 +311,7 @@ enum cpuid_leafs CPUID_8000_0008_EBX, CPUID_6_EAX, CPUID_8000_000A_EDX, + CPUID_7_ECX, }; #ifdef CONFIG_X86_FEATURE_NAMES @@ -329,28 +334,42 @@ extern const char * const x86_bug_flags[ test_bit(bit, (unsigned long *)((c)->x86_capability)) #define REQUIRED_MASK_BIT_SET(bit) \ - ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0)) || \ - (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1)) || \ - (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2)) || \ - (((bit)>>5)==3 && (1UL<<((bit)&31) & REQUIRED_MASK3)) || \ - (((bit)>>5)==4 && (1UL<<((bit)&31) & REQUIRED_MASK4)) || \ - (((bit)>>5)==5 && (1UL<<((bit)&31) & REQUIRED_MASK5)) || \ - (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) || \ - (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) || \ - (((bit)>>5)==8 && (1UL<<((bit)&31) & REQUIRED_MASK8)) || \ - (((bit)>>5)==9 && (1UL<<((bit)&31) & REQUIRED_MASK9)) ) + ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0 )) || \ + (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1 )) || \ + (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2 )) || \ + (((bit)>>5)==3 && (1UL<<((bit)&31) & REQUIRED_MASK3 )) || \ + (((bit)>>5)==4 && (1UL<<((bit)&31) & REQUIRED_MASK4 )) || \ + (((bit)>>5)==5 && (1UL<<((bit)&31) & REQUIRED_MASK5 )) || \ + (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6 )) || \ + (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7 )) || \ + (((bit)>>5)==8 && (1UL<<((bit)&31) & REQUIRED_MASK8 )) || \ + (((bit)>>5)==9 && (1UL<<((bit)&31) & REQUIRED_MASK9 )) || \ + (((bit)>>5)==10 && (1UL<<((bit)&31) & REQUIRED_MASK10)) || \ + (((bit)>>5)==11 && (1UL<<((bit)&31) & REQUIRED_MASK11)) || \ + (((bit)>>5)==12 && (1UL<<((bit)&31) & REQUIRED_MASK12)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & REQUIRED_MASK13)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & REQUIRED_MASK14)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & REQUIRED_MASK15)) || \ + (((bit)>>5)==14 && (1UL<<((bit)&31) & REQUIRED_MASK16)) ) #define DISABLED_MASK_BIT_SET(bit) \ - ( (((bit)>>5)==0 && (1UL<<((bit)&31) & DISABLED_MASK0)) || \ - (((bit)>>5)==1 && (1UL<<((bit)&31) & DISABLED_MASK1)) || \ - (((bit)>>5)==2 && (1UL<<((bit)&31) & DISABLED_MASK2)) || \ - (((bit)>>5)==3 && (1UL<<((bit)&31) & DISABLED_MASK3)) || \ - (((bit)>>5)==4 && (1UL<<((bit)&31) & DISABLED_MASK4)) || \ - (((bit)>>5)==5 && (1UL<<((bit)&31) & DISABLED_MASK5)) || \ - (((bit)>>5)==6 && (1UL<<((bit)&31) & DISABLED_MASK6)) || \ - (((bit)>>5)==7 && (1UL<<((bit)&31) & DISABLED_MASK7)) || \ - (((bit)>>5)==8 && (1UL<<((bit)&31) & DISABLED_MASK8)) || \ - (((bit)>>5)==9 && (1UL<<((bit)&31) & DISABLED_MASK9)) ) + ( (((bit)>>5)==0 && (1UL<<((bit)&31) & DISABLED_MASK0 )) || \ + (((bit)>>5)==1 && (1UL<<((bit)&31) & DISABLED_MASK1 )) || \ + (((bit)>>5)==2 && (1UL<<((bit)&31) & DISABLED_MASK2 )) || \ + (((bit)>>5)==3 && (1UL<<((bit)&31) & DISABLED_MASK3 )) || \ + (((bit)>>5)==4 && (1UL<<((bit)&31) & DISABLED_MASK4 )) || \ + (((bit)>>5)==5 && (1UL<<((bit)&31) & DISABLED_MASK5 )) || \ + (((bit)>>5)==6 && (1UL<<((bit)&31) & DISABLED_MASK6 )) || \ + (((bit)>>5)==7 && (1UL<<((bit)&31) & DISABLED_MASK7 )) || \ + (((bit)>>5)==8 && (1UL<<((bit)&31) & DISABLED_MASK8 )) || \ + (((bit)>>5)==9 && (1UL<<((bit)&31) & DISABLED_MASK9 )) || \ + (((bit)>>5)==10 && (1UL<<((bit)&31) & DISABLED_MASK10)) || \ + (((bit)>>5)==11 && (1UL<<((bit)&31) & DISABLED_MASK11)) || \ + (((bit)>>5)==12 && (1UL<<((bit)&31) & DISABLED_MASK12)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & DISABLED_MASK13)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & DISABLED_MASK14)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & DISABLED_MASK15)) || \ + (((bit)>>5)==14 && (1UL<<((bit)&31) & DISABLED_MASK16)) ) #define cpu_has(c, bit) \ (__builtin_constant_p(bit) && REQUIRED_MASK_BIT_SET(bit) ? 1 : \ diff -puN arch/x86/include/asm/disabled-features.h~pkeys-01-cpuid arch/x86/include/asm/disabled-features.h --- a/arch/x86/include/asm/disabled-features.h~pkeys-01-cpuid 2016-02-12 10:44:16.317250524 -0800 +++ b/arch/x86/include/asm/disabled-features.h 2016-02-12 10:44:16.323250798 -0800 @@ -28,6 +28,14 @@ # define DISABLE_CENTAUR_MCR 0 #endif /* CONFIG_X86_64 */ +#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS +# define DISABLE_PKU (1<<(X86_FEATURE_PKU)) +# define DISABLE_OSPKE (1<<(X86_FEATURE_OSPKE)) +#else +# define DISABLE_PKU 0 +# define DISABLE_OSPKE 0 +#endif /* CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS */ + /* * Make sure to add features to the correct mask */ @@ -41,5 +49,12 @@ #define DISABLED_MASK7 0 #define DISABLED_MASK8 0 #define DISABLED_MASK9 (DISABLE_MPX) +#define DISABLED_MASK10 0 +#define DISABLED_MASK11 0 +#define DISABLED_MASK12 0 +#define DISABLED_MASK13 0 +#define DISABLED_MASK14 0 +#define DISABLED_MASK15 0 +#define DISABLED_MASK16 (DISABLE_PKU|DISABLE_OSPKE) #endif /* _ASM_X86_DISABLED_FEATURES_H */ diff -puN arch/x86/include/asm/required-features.h~pkeys-01-cpuid arch/x86/include/asm/required-features.h --- a/arch/x86/include/asm/required-features.h~pkeys-01-cpuid 2016-02-12 10:44:16.318250570 -0800 +++ b/arch/x86/include/asm/required-features.h 2016-02-12 10:44:16.324250844 -0800 @@ -92,5 +92,12 @@ #define REQUIRED_MASK7 0 #define REQUIRED_MASK8 0 #define REQUIRED_MASK9 0 +#define REQUIRED_MASK10 0 +#define REQUIRED_MASK11 0 +#define REQUIRED_MASK12 0 +#define REQUIRED_MASK13 0 +#define REQUIRED_MASK14 0 +#define REQUIRED_MASK15 0 +#define REQUIRED_MASK16 0 #endif /* _ASM_X86_REQUIRED_FEATURES_H */ diff -puN arch/x86/kernel/cpu/common.c~pkeys-01-cpuid arch/x86/kernel/cpu/common.c --- a/arch/x86/kernel/cpu/common.c~pkeys-01-cpuid 2016-02-12 10:44:16.320250661 -0800 +++ b/arch/x86/kernel/cpu/common.c 2016-02-12 10:44:16.324250844 -0800 @@ -611,6 +611,7 @@ void get_cpu_cap(struct cpuinfo_x86 *c) c->x86_capability[CPUID_7_0_EBX] = ebx; c->x86_capability[CPUID_6_EAX] = cpuid_eax(0x00000006); + c->x86_capability[CPUID_7_ECX] = ecx; } /* Extended state features: level 0x0000000d */ _
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Dave Hansen <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-18 21:20 +0100 |
| Subject | [tip:mm/pkeys] x86/cpufeature, x86/mm/pkeys: Add protection keys related CPUID definitions |
| Message-ID | <r3Ddn-2wj-15@gated-at.bofh.it> |
| In reply to | #1333131 |
Commit-ID: dfb4a70f20c5b3880da56ee4c9484bdb4e8f1e65 Gitweb: http://git.kernel.org/tip/dfb4a70f20c5b3880da56ee4c9484bdb4e8f1e65 Author: Dave Hansen <dave.hansen@linux.intel.com> AuthorDate: Fri, 12 Feb 2016 13:02:01 -0800 Committer: Ingo Molnar <mingo@kernel.org> CommitDate: Tue, 16 Feb 2016 10:11:13 +0100 x86/cpufeature, x86/mm/pkeys: Add protection keys related CPUID definitions There are two CPUID bits for protection keys. One is for whether the CPU contains the feature, and the other will appear set once the OS enables protection keys. Specifically: Bit 04: OSPKE. If 1, OS has set CR4.PKE to enable Protection keys (and the RDPKRU/WRPKRU instructions) This is because userspace can not see CR4 contents, but it can see CPUID contents. X86_FEATURE_PKU is referred to as "PKU" in the hardware documentation: CPUID.(EAX=07H,ECX=0H):ECX.PKU [bit 3] X86_FEATURE_OSPKE is "OSPKU": CPUID.(EAX=07H,ECX=0H):ECX.OSPKE [bit 4] These are the first CPU features which need to look at the ECX word in CPUID leaf 0x7, so this patch also includes fetching that word in to the cpuinfo->x86_capability[] array. Add it to the disabled-features mask when its config option is off. Even though we are not using it here, we also extend the REQUIRED_MASK_BIT_SET() macro to keep it mirroring the DISABLED_MASK_BIT_SET() version. This means that in almost all code, you should use: cpu_has(c, X86_FEATURE_PKU) and *not* the CONFIG option. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> 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: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20160212210201.7714C250@viggo.jf.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org> --- arch/x86/include/asm/cpufeature.h | 59 +++++++++++++++++++++----------- arch/x86/include/asm/cpufeatures.h | 2 +- arch/x86/include/asm/disabled-features.h | 15 ++++++++ arch/x86/include/asm/required-features.h | 7 ++++ arch/x86/kernel/cpu/common.c | 1 + 5 files changed, 63 insertions(+), 21 deletions(-) diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 68e4e82..50e292a 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -26,6 +26,7 @@ enum cpuid_leafs CPUID_8000_0008_EBX, CPUID_6_EAX, CPUID_8000_000A_EDX, + CPUID_7_ECX, }; #ifdef CONFIG_X86_FEATURE_NAMES @@ -48,28 +49,42 @@ extern const char * const x86_bug_flags[NBUGINTS*32]; test_bit(bit, (unsigned long *)((c)->x86_capability)) #define REQUIRED_MASK_BIT_SET(bit) \ - ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0)) || \ - (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1)) || \ - (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2)) || \ - (((bit)>>5)==3 && (1UL<<((bit)&31) & REQUIRED_MASK3)) || \ - (((bit)>>5)==4 && (1UL<<((bit)&31) & REQUIRED_MASK4)) || \ - (((bit)>>5)==5 && (1UL<<((bit)&31) & REQUIRED_MASK5)) || \ - (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) || \ - (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) || \ - (((bit)>>5)==8 && (1UL<<((bit)&31) & REQUIRED_MASK8)) || \ - (((bit)>>5)==9 && (1UL<<((bit)&31) & REQUIRED_MASK9)) ) + ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0 )) || \ + (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1 )) || \ + (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2 )) || \ + (((bit)>>5)==3 && (1UL<<((bit)&31) & REQUIRED_MASK3 )) || \ + (((bit)>>5)==4 && (1UL<<((bit)&31) & REQUIRED_MASK4 )) || \ + (((bit)>>5)==5 && (1UL<<((bit)&31) & REQUIRED_MASK5 )) || \ + (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6 )) || \ + (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7 )) || \ + (((bit)>>5)==8 && (1UL<<((bit)&31) & REQUIRED_MASK8 )) || \ + (((bit)>>5)==9 && (1UL<<((bit)&31) & REQUIRED_MASK9 )) || \ + (((bit)>>5)==10 && (1UL<<((bit)&31) & REQUIRED_MASK10)) || \ + (((bit)>>5)==11 && (1UL<<((bit)&31) & REQUIRED_MASK11)) || \ + (((bit)>>5)==12 && (1UL<<((bit)&31) & REQUIRED_MASK12)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & REQUIRED_MASK13)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & REQUIRED_MASK14)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & REQUIRED_MASK15)) || \ + (((bit)>>5)==14 && (1UL<<((bit)&31) & REQUIRED_MASK16)) ) #define DISABLED_MASK_BIT_SET(bit) \ - ( (((bit)>>5)==0 && (1UL<<((bit)&31) & DISABLED_MASK0)) || \ - (((bit)>>5)==1 && (1UL<<((bit)&31) & DISABLED_MASK1)) || \ - (((bit)>>5)==2 && (1UL<<((bit)&31) & DISABLED_MASK2)) || \ - (((bit)>>5)==3 && (1UL<<((bit)&31) & DISABLED_MASK3)) || \ - (((bit)>>5)==4 && (1UL<<((bit)&31) & DISABLED_MASK4)) || \ - (((bit)>>5)==5 && (1UL<<((bit)&31) & DISABLED_MASK5)) || \ - (((bit)>>5)==6 && (1UL<<((bit)&31) & DISABLED_MASK6)) || \ - (((bit)>>5)==7 && (1UL<<((bit)&31) & DISABLED_MASK7)) || \ - (((bit)>>5)==8 && (1UL<<((bit)&31) & DISABLED_MASK8)) || \ - (((bit)>>5)==9 && (1UL<<((bit)&31) & DISABLED_MASK9)) ) + ( (((bit)>>5)==0 && (1UL<<((bit)&31) & DISABLED_MASK0 )) || \ + (((bit)>>5)==1 && (1UL<<((bit)&31) & DISABLED_MASK1 )) || \ + (((bit)>>5)==2 && (1UL<<((bit)&31) & DISABLED_MASK2 )) || \ + (((bit)>>5)==3 && (1UL<<((bit)&31) & DISABLED_MASK3 )) || \ + (((bit)>>5)==4 && (1UL<<((bit)&31) & DISABLED_MASK4 )) || \ + (((bit)>>5)==5 && (1UL<<((bit)&31) & DISABLED_MASK5 )) || \ + (((bit)>>5)==6 && (1UL<<((bit)&31) & DISABLED_MASK6 )) || \ + (((bit)>>5)==7 && (1UL<<((bit)&31) & DISABLED_MASK7 )) || \ + (((bit)>>5)==8 && (1UL<<((bit)&31) & DISABLED_MASK8 )) || \ + (((bit)>>5)==9 && (1UL<<((bit)&31) & DISABLED_MASK9 )) || \ + (((bit)>>5)==10 && (1UL<<((bit)&31) & DISABLED_MASK10)) || \ + (((bit)>>5)==11 && (1UL<<((bit)&31) & DISABLED_MASK11)) || \ + (((bit)>>5)==12 && (1UL<<((bit)&31) & DISABLED_MASK12)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & DISABLED_MASK13)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & DISABLED_MASK14)) || \ + (((bit)>>5)==13 && (1UL<<((bit)&31) & DISABLED_MASK15)) || \ + (((bit)>>5)==14 && (1UL<<((bit)&31) & DISABLED_MASK16)) ) #define cpu_has(c, bit) \ (__builtin_constant_p(bit) && REQUIRED_MASK_BIT_SET(bit) ? 1 : \ @@ -79,6 +94,10 @@ extern const char * const x86_bug_flags[NBUGINTS*32]; (__builtin_constant_p(bit) && REQUIRED_MASK_BIT_SET(bit) ? 1 : \ x86_this_cpu_test_bit(bit, (unsigned long *)&cpu_info.x86_capability)) +/* Intel-defined CPU features, CPUID level 0x00000007:0 (ecx), word 16 */ +#define X86_FEATURE_PKU (16*32+ 3) /* Protection Keys for Userspace */ +#define X86_FEATURE_OSPKE (16*32+ 4) /* OS Protection Keys Enable */ + /* * This macro is for detection of features which need kernel * infrastructure to be used. It may *not* directly test the CPU diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 0ceb6ad..cbb2c56 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -12,7 +12,7 @@ /* * Defines x86 CPU feature bits */ -#define NCAPINTS 16 /* N 32-bit words worth of info */ +#define NCAPINTS 17 /* N 32-bit words worth of info */ #define NBUGINTS 1 /* N 32-bit bug flags */ /* diff --git a/arch/x86/include/asm/disabled-features.h b/arch/x86/include/asm/disabled-features.h index f226df0..39343be 100644 --- a/arch/x86/include/asm/disabled-features.h +++ b/arch/x86/include/asm/disabled-features.h @@ -28,6 +28,14 @@ # define DISABLE_CENTAUR_MCR 0 #endif /* CONFIG_X86_64 */ +#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS +# define DISABLE_PKU (1<<(X86_FEATURE_PKU)) +# define DISABLE_OSPKE (1<<(X86_FEATURE_OSPKE)) +#else +# define DISABLE_PKU 0 +# define DISABLE_OSPKE 0 +#endif /* CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS */ + /* * Make sure to add features to the correct mask */ @@ -41,5 +49,12 @@ #define DISABLED_MASK7 0 #define DISABLED_MASK8 0 #define DISABLED_MASK9 (DISABLE_MPX) +#define DISABLED_MASK10 0 +#define DISABLED_MASK11 0 +#define DISABLED_MASK12 0 +#define DISABLED_MASK13 0 +#define DISABLED_MASK14 0 +#define DISABLED_MASK15 0 +#define DISABLED_MASK16 (DISABLE_PKU|DISABLE_OSPKE) #endif /* _ASM_X86_DISABLED_FEATURES_H */ diff --git a/arch/x86/include/asm/required-features.h b/arch/x86/include/asm/required-features.h index 5c6e4fb..4916144 100644 --- a/arch/x86/include/asm/required-features.h +++ b/arch/x86/include/asm/required-features.h @@ -92,5 +92,12 @@ #define REQUIRED_MASK7 0 #define REQUIRED_MASK8 0 #define REQUIRED_MASK9 0 +#define REQUIRED_MASK10 0 +#define REQUIRED_MASK11 0 +#define REQUIRED_MASK12 0 +#define REQUIRED_MASK13 0 +#define REQUIRED_MASK14 0 +#define REQUIRED_MASK15 0 +#define REQUIRED_MASK16 0 #endif /* _ASM_X86_REQUIRED_FEATURES_H */ diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index f4bb2c4..a719ad7 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -627,6 +627,7 @@ void get_cpu_cap(struct cpuinfo_x86 *c) c->x86_capability[CPUID_7_0_EBX] = ebx; c->x86_capability[CPUID_6_EAX] = cpuid_eax(0x00000006); + c->x86_capability[CPUID_7_ECX] = ecx; } /* Extended state features: level 0x0000000d */
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-12 22:20 +0100 |
| Subject | [PATCH 02/33] mm: overload get_user_pages() functions |
| Message-ID | <r1tib-3WT-13@gated-at.bofh.it> |
| In reply to | #1333117 |
From: Dave Hansen <dave.hansen@linux.intel.com>
The concept here was a suggestion from Ingo. The implementation
horrors are all mine.
This allows get_user_pages(), get_user_pages_unlocked(), and
get_user_pages_locked() to be called with or without the
leading tsk/mm arguments. We will give a compile-time warning
about the old style being __deprecated and we will also
WARN_ON() if the non-remote version is used for a remote-style
access.
Doing this, folks will get nice warnings and will not break the
build. This should be nice for -next and will hopefully let
developers fix up their own code instead of maintainers needing
to do it at merge time.
The way we do this is hideous. It uses the __VA_ARGS__ macro
functionality to call different functions based on the number
of arguments passed to the macro.
There's an additional hack to ensure that our EXPORT_SYMBOL()
of the deprecated symbols doesn't trigger a warning.
We should be able to remove this mess as soon as -rc1 hits in
the release after this is merged.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: jack@suse.cz
---
b/include/linux/mm.h | 74 ++++++++++++++++++++++++++++++++++++++++++++-------
b/mm/gup.c | 62 +++++++++++++++++++++++++++++++++---------
b/mm/nommu.c | 64 +++++++++++++++++++++++++++++++-------------
b/mm/util.c | 4 --
4 files changed, 158 insertions(+), 46 deletions(-)
diff -puN include/linux/mm.h~gup-arg-helpers include/linux/mm.h
--- a/include/linux/mm.h~gup-arg-helpers 2016-02-12 10:44:13.845137517 -0800
+++ b/include/linux/mm.h 2016-02-12 10:44:13.854137929 -0800
@@ -1229,24 +1229,78 @@ long get_user_pages_remote(struct task_s
unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
struct vm_area_struct **vmas);
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
- int write, int force, struct page **pages,
- struct vm_area_struct **vmas);
-long get_user_pages_locked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
- int write, int force, struct page **pages,
- int *locked);
+long get_user_pages6(unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas);
+long get_user_pages_locked6(unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages, int *locked);
long __get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
unsigned int gup_flags);
-long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_unlocked5(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages);
int get_user_pages_fast(unsigned long start, int nr_pages, int write,
struct page **pages);
+/* suppress warnings from use in EXPORT_SYMBOL() */
+#ifndef __DISABLE_GUP_DEPRECATED
+#define __gup_deprecated __deprecated
+#else
+#define __gup_deprecated
+#endif
+/*
+ * These macros provide backward-compatibility with the old
+ * get_user_pages() variants which took tsk/mm. These
+ * functions/macros provide both compile-time __deprecated so we
+ * can catch old-style use and not break the build. The actual
+ * functions also have WARN_ON()s to let us know at runtime if
+ * the get_user_pages() should have been the "remote" variant.
+ *
+ * These are hideous, but temporary.
+ *
+ * If you run into one of these __deprecated warnings, look
+ * at how you are calling get_user_pages(). If you are calling
+ * it with current/current->mm as the first two arguments,
+ * simply remove those arguments. The behavior will be the same
+ * as it is now. If you are calling it on another task, use
+ * get_user_pages_remote() instead.
+ *
+ * Any questions? Ask Dave Hansen <dave@sr71.net>
+ */
+long
+__gup_deprecated
+get_user_pages8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas);
+#define GUP_MACRO(_1, _2, _3, _4, _5, _6, _7, _8, get_user_pages, ...) \
+ get_user_pages
+#define get_user_pages(...) GUP_MACRO(__VA_ARGS__, \
+ get_user_pages8, x, \
+ get_user_pages6, x, x, x, x, x)(__VA_ARGS__)
+
+__gup_deprecated
+long get_user_pages_locked8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ int *locked);
+#define GUPL_MACRO(_1, _2, _3, _4, _5, _6, _7, _8, get_user_pages_locked, ...) \
+ get_user_pages_locked
+#define get_user_pages_locked(...) GUPL_MACRO(__VA_ARGS__, \
+ get_user_pages_locked8, x, \
+ get_user_pages_locked6, x, x, x, x)(__VA_ARGS__)
+
+__gup_deprecated
+long get_user_pages_unlocked7(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages);
+#define GUPU_MACRO(_1, _2, _3, _4, _5, _6, _7, get_user_pages_unlocked, ...) \
+ get_user_pages_unlocked
+#define get_user_pages_unlocked(...) GUPU_MACRO(__VA_ARGS__, \
+ get_user_pages_unlocked7, x, \
+ get_user_pages_unlocked5, x, x, x, x)(__VA_ARGS__)
+
/* Container for pinned pfns / pages */
struct frame_vector {
unsigned int nr_allocated; /* Number of frames we have space for */
diff -puN mm/gup.c~gup-arg-helpers mm/gup.c
--- a/mm/gup.c~gup-arg-helpers 2016-02-12 10:44:13.847137609 -0800
+++ b/mm/gup.c 2016-02-12 10:44:13.855137974 -0800
@@ -1,3 +1,4 @@
+#define __DISABLE_GUP_DEPRECATED 1
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/err.h>
@@ -807,15 +808,15 @@ static __always_inline long __get_user_p
* if (locked)
* up_read(&mm->mmap_sem);
*/
-long get_user_pages_locked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_locked6(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
int *locked)
{
- return __get_user_pages_locked(tsk, mm, start, nr_pages, write, force,
- pages, NULL, locked, true, FOLL_TOUCH);
+ return __get_user_pages_locked(current, current->mm, start, nr_pages,
+ write, force, pages, NULL, locked, true,
+ FOLL_TOUCH);
}
-EXPORT_SYMBOL(get_user_pages_locked);
+EXPORT_SYMBOL(get_user_pages_locked6);
/*
* Same as get_user_pages_unlocked(...., FOLL_TOUCH) but it allows to
@@ -860,14 +861,13 @@ EXPORT_SYMBOL(__get_user_pages_unlocked)
* or if "force" shall be set to 1 (get_user_pages_fast misses the
* "force" parameter).
*/
-long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_unlocked5(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages)
{
- return __get_user_pages_unlocked(tsk, mm, start, nr_pages, write,
- force, pages, FOLL_TOUCH);
+ return __get_user_pages_unlocked(current, current->mm, start, nr_pages,
+ write, force, pages, FOLL_TOUCH);
}
-EXPORT_SYMBOL(get_user_pages_unlocked);
+EXPORT_SYMBOL(get_user_pages_unlocked5);
/*
* get_user_pages_remote() - pin user pages in memory
@@ -939,16 +939,15 @@ EXPORT_SYMBOL(get_user_pages_remote);
* This is the same as get_user_pages_remote() for the time
* being.
*/
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages6(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
struct vm_area_struct **vmas)
{
- return __get_user_pages_locked(tsk, mm, start, nr_pages,
+ return __get_user_pages_locked(current, current->mm, start, nr_pages,
write, force, pages, vmas, NULL, false,
FOLL_TOUCH);
}
-EXPORT_SYMBOL(get_user_pages);
+EXPORT_SYMBOL(get_user_pages6);
/**
* populate_vma_page_range() - populate a range of pages in the vma.
@@ -1484,3 +1483,38 @@ int get_user_pages_fast(unsigned long st
}
#endif /* CONFIG_HAVE_GENERIC_RCU_GUP */
+
+long get_user_pages8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
+{
+ WARN_ONCE(tsk != current, "get_user_pages() called on remote task");
+ WARN_ONCE(mm != current->mm, "get_user_pages() called on remote mm");
+
+ return get_user_pages6(start, nr_pages, write, force, pages, vmas);
+}
+EXPORT_SYMBOL(get_user_pages8);
+
+long get_user_pages_locked8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages, int *locked)
+{
+ WARN_ONCE(tsk != current, "get_user_pages_locked() called on remote task");
+ WARN_ONCE(mm != current->mm, "get_user_pages_locked() called on remote mm");
+
+ return get_user_pages_locked6(start, nr_pages, write, force, pages, locked);
+}
+EXPORT_SYMBOL(get_user_pages_locked8);
+
+long get_user_pages_unlocked7(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages)
+{
+ WARN_ONCE(tsk != current, "get_user_pages_unlocked() called on remote task");
+ WARN_ONCE(mm != current->mm, "get_user_pages_unlocked() called on remote mm");
+
+ return get_user_pages_unlocked5(start, nr_pages, write, force, pages);
+}
+EXPORT_SYMBOL(get_user_pages_unlocked7);
+
diff -puN mm/nommu.c~gup-arg-helpers mm/nommu.c
--- a/mm/nommu.c~gup-arg-helpers 2016-02-12 10:44:13.848137654 -0800
+++ b/mm/nommu.c 2016-02-12 10:44:13.856138020 -0800
@@ -15,6 +15,8 @@
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#define __DISABLE_GUP_DEPRECATED
+
#include <linux/export.h>
#include <linux/mm.h>
#include <linux/vmacache.h>
@@ -182,8 +184,7 @@ finish_or_fault:
* slab page or a secondary page from a compound page
* - don't permit access to VMAs that don't support it, such as I/O mappings
*/
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages6(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
struct vm_area_struct **vmas)
{
@@ -194,20 +195,18 @@ long get_user_pages(struct task_struct *
if (force)
flags |= FOLL_FORCE;
- return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas,
- NULL);
+ return __get_user_pages(current, current->mm, start, nr_pages, flags,
+ pages, vmas, NULL);
}
-EXPORT_SYMBOL(get_user_pages);
+EXPORT_SYMBOL(get_user_pages6);
-long get_user_pages_locked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
- int write, int force, struct page **pages,
- int *locked)
+long get_user_pages_locked6(unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ int *locked)
{
- return get_user_pages(tsk, mm, start, nr_pages, write, force,
- pages, NULL);
+ return get_user_pages6(start, nr_pages, write, force, pages, NULL);
}
-EXPORT_SYMBOL(get_user_pages_locked);
+EXPORT_SYMBOL(get_user_pages_locked6);
long __get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
@@ -216,21 +215,20 @@ long __get_user_pages_unlocked(struct ta
{
long ret;
down_read(&mm->mmap_sem);
- ret = get_user_pages(tsk, mm, start, nr_pages, write, force,
- pages, NULL);
+ ret = __get_user_pages(tsk, mm, start, nr_pages, gup_flags, pages,
+ NULL, NULL);
up_read(&mm->mmap_sem);
return ret;
}
EXPORT_SYMBOL(__get_user_pages_unlocked);
-long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_unlocked5(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages)
{
- return __get_user_pages_unlocked(tsk, mm, start, nr_pages, write,
- force, pages, 0);
+ return __get_user_pages_unlocked(current, current->mm, start, nr_pages,
+ write, force, pages, 0);
}
-EXPORT_SYMBOL(get_user_pages_unlocked);
+EXPORT_SYMBOL(get_user_pages_unlocked5);
/**
* follow_pfn - look up PFN at a user virtual address
@@ -2108,3 +2106,31 @@ static int __meminit init_admin_reserve(
return 0;
}
subsys_initcall(init_admin_reserve);
+
+long get_user_pages8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
+{
+ return get_user_pages6(start, nr_pages, write, force, pages, vmas);
+}
+EXPORT_SYMBOL(get_user_pages8);
+
+long get_user_pages_locked8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ int *locked)
+{
+ return get_user_pages_locked6(start, nr_pages, write,
+ force, pages, locked);
+}
+EXPORT_SYMBOL(get_user_pages_locked8);
+
+long get_user_pages_unlocked7(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages)
+{
+ return get_user_pages_unlocked5(start, nr_pages, write, force, pages);
+}
+EXPORT_SYMBOL(get_user_pages_unlocked7);
+
diff -puN mm/util.c~gup-arg-helpers mm/util.c
--- a/mm/util.c~gup-arg-helpers 2016-02-12 10:44:13.850137746 -0800
+++ b/mm/util.c 2016-02-12 10:44:13.856138020 -0800
@@ -283,9 +283,7 @@ EXPORT_SYMBOL_GPL(__get_user_pages_fast)
int __weak get_user_pages_fast(unsigned long start,
int nr_pages, int write, struct page **pages)
{
- struct mm_struct *mm = current->mm;
- return get_user_pages_unlocked(current, mm, start, nr_pages,
- write, 0, pages);
+ return get_user_pages_unlocked(start, nr_pages, write, 0, pages);
}
EXPORT_SYMBOL_GPL(get_user_pages_fast);
_
[toc] | [prev] | [next] | [standalone]
| From | Ingo Molnar <mingo@kernel.org> |
|---|---|
| Date | 2016-02-16 09:40 +0100 |
| Subject | Re: [PATCH 02/33] mm: overload get_user_pages() functions |
| Message-ID | <r2JkR-5nT-1@gated-at.bofh.it> |
| In reply to | #1333132 |
* Dave Hansen <dave@sr71.net> wrote: > > From: Dave Hansen <dave.hansen@linux.intel.com> > > The concept here was a suggestion from Ingo. The implementation > horrors are all mine. > > This allows get_user_pages(), get_user_pages_unlocked(), and > get_user_pages_locked() to be called with or without the > leading tsk/mm arguments. We will give a compile-time warning > about the old style being __deprecated and we will also > WARN_ON() if the non-remote version is used for a remote-style > access. So at minimum this should be WARN_ON_ONCE(), to make it easier to recover some meaningful kernel log from such incidents. But: > Doing this, folks will get nice warnings and will not break the > build. This should be nice for -next and will hopefully let > developers fix up their own code instead of maintainers needing > to do it at merge time. > > The way we do this is hideous. It uses the __VA_ARGS__ macro > functionality to call different functions based on the number > of arguments passed to the macro. > > There's an additional hack to ensure that our EXPORT_SYMBOL() > of the deprecated symbols doesn't trigger a warning. > > We should be able to remove this mess as soon as -rc1 hits in > the release after this is merged. So when I suggested this then it looked a _lot_ cleanear to me, in my head! OTOH this, if factored out a bit perhaps, could be the basis for a useful technical model to do 'phased in, -next invariant' prototype migrations in the future, especially when it involves lots of subsystems. Strictly only in cases where -rc1 will truly get rid of the __VA_ARGS__ hackery - which we'd do in this case. Nevertheless I'd love to have a high level buy-in from either Linus or Andrew that we can do it this way, as the hackery looks very hideous... The alternative would be to allow the -next churn and to allow the occasional (fairly trivial but tester-disruptive) build breakage. Thanks, Ingo
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-17 19:20 +0100 |
| Subject | Re: [PATCH 02/33] mm: overload get_user_pages() functions |
| Message-ID | <r3eRH-1IR-1@gated-at.bofh.it> |
| In reply to | #1335141 |
On 02/16/2016 12:36 AM, Ingo Molnar wrote: >> > From: Dave Hansen <dave.hansen@linux.intel.com> >> > >> > The concept here was a suggestion from Ingo. The implementation >> > horrors are all mine. >> > >> > This allows get_user_pages(), get_user_pages_unlocked(), and >> > get_user_pages_locked() to be called with or without the >> > leading tsk/mm arguments. We will give a compile-time warning >> > about the old style being __deprecated and we will also >> > WARN_ON() if the non-remote version is used for a remote-style >> > access. > So at minimum this should be WARN_ON_ONCE(), to make it easier to recover some > meaningful kernel log from such incidents. I went to go fix this in the code but realized that I coded it up as WARN_ONCE(). The description was just imprecise. So I won't be sending a code fix for this.
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Dave Hansen <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-18 21:20 +0100 |
| Subject | [tip:mm/pkeys] mm/gup: Overload get_user_pages() functions |
| Message-ID | <r3Ddp-2wj-49@gated-at.bofh.it> |
| In reply to | #1333132 |
Commit-ID: cde70140fed8429acf7a14e2e2cbd3e329036653
Gitweb: http://git.kernel.org/tip/cde70140fed8429acf7a14e2e2cbd3e329036653
Author: Dave Hansen <dave.hansen@linux.intel.com>
AuthorDate: Fri, 12 Feb 2016 13:01:55 -0800
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Tue, 16 Feb 2016 10:11:12 +0100
mm/gup: Overload get_user_pages() functions
The concept here was a suggestion from Ingo. The implementation
horrors are all mine.
This allows get_user_pages(), get_user_pages_unlocked(), and
get_user_pages_locked() to be called with or without the
leading tsk/mm arguments. We will give a compile-time warning
about the old style being __deprecated and we will also
WARN_ON() if the non-remote version is used for a remote-style
access.
Doing this, folks will get nice warnings and will not break the
build. This should be nice for -next and will hopefully let
developers fix up their own code instead of maintainers needing
to do it at merge time.
The way we do this is hideous. It uses the __VA_ARGS__ macro
functionality to call different functions based on the number
of arguments passed to the macro.
There's an additional hack to ensure that our EXPORT_SYMBOL()
of the deprecated symbols doesn't trigger a warning.
We should be able to remove this mess as soon as -rc1 hits in
the release after this is merged.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Alexander Kuleshov <kuleshovmail@gmail.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave@sr71.net>
Cc: Dominik Dingel <dingel@linux.vnet.ibm.com>
Cc: Geliang Tang <geliangtang@163.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Leon Romanovsky <leon@leon.nu>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: Mateusz Guzik <mguzik@redhat.com>
Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Xie XiuQi <xiexiuqi@huawei.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
Link: http://lkml.kernel.org/r/20160212210155.73222EE1@viggo.jf.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
include/linux/mm.h | 74 ++++++++++++++++++++++++++++++++++++++++++++++--------
mm/gup.c | 62 ++++++++++++++++++++++++++++++++++-----------
mm/nommu.c | 64 ++++++++++++++++++++++++++++++++--------------
mm/util.c | 4 +--
4 files changed, 158 insertions(+), 46 deletions(-)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index faf3b70..4c73178 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1229,24 +1229,78 @@ long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
struct vm_area_struct **vmas);
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
- int write, int force, struct page **pages,
- struct vm_area_struct **vmas);
-long get_user_pages_locked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
- int write, int force, struct page **pages,
- int *locked);
+long get_user_pages6(unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas);
+long get_user_pages_locked6(unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages, int *locked);
long __get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
unsigned int gup_flags);
-long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_unlocked5(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages);
int get_user_pages_fast(unsigned long start, int nr_pages, int write,
struct page **pages);
+/* suppress warnings from use in EXPORT_SYMBOL() */
+#ifndef __DISABLE_GUP_DEPRECATED
+#define __gup_deprecated __deprecated
+#else
+#define __gup_deprecated
+#endif
+/*
+ * These macros provide backward-compatibility with the old
+ * get_user_pages() variants which took tsk/mm. These
+ * functions/macros provide both compile-time __deprecated so we
+ * can catch old-style use and not break the build. The actual
+ * functions also have WARN_ON()s to let us know at runtime if
+ * the get_user_pages() should have been the "remote" variant.
+ *
+ * These are hideous, but temporary.
+ *
+ * If you run into one of these __deprecated warnings, look
+ * at how you are calling get_user_pages(). If you are calling
+ * it with current/current->mm as the first two arguments,
+ * simply remove those arguments. The behavior will be the same
+ * as it is now. If you are calling it on another task, use
+ * get_user_pages_remote() instead.
+ *
+ * Any questions? Ask Dave Hansen <dave@sr71.net>
+ */
+long
+__gup_deprecated
+get_user_pages8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas);
+#define GUP_MACRO(_1, _2, _3, _4, _5, _6, _7, _8, get_user_pages, ...) \
+ get_user_pages
+#define get_user_pages(...) GUP_MACRO(__VA_ARGS__, \
+ get_user_pages8, x, \
+ get_user_pages6, x, x, x, x, x)(__VA_ARGS__)
+
+__gup_deprecated
+long get_user_pages_locked8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ int *locked);
+#define GUPL_MACRO(_1, _2, _3, _4, _5, _6, _7, _8, get_user_pages_locked, ...) \
+ get_user_pages_locked
+#define get_user_pages_locked(...) GUPL_MACRO(__VA_ARGS__, \
+ get_user_pages_locked8, x, \
+ get_user_pages_locked6, x, x, x, x)(__VA_ARGS__)
+
+__gup_deprecated
+long get_user_pages_unlocked7(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages);
+#define GUPU_MACRO(_1, _2, _3, _4, _5, _6, _7, get_user_pages_unlocked, ...) \
+ get_user_pages_unlocked
+#define get_user_pages_unlocked(...) GUPU_MACRO(__VA_ARGS__, \
+ get_user_pages_unlocked7, x, \
+ get_user_pages_unlocked5, x, x, x, x)(__VA_ARGS__)
+
/* Container for pinned pfns / pages */
struct frame_vector {
unsigned int nr_allocated; /* Number of frames we have space for */
diff --git a/mm/gup.c b/mm/gup.c
index 36ca850..8a035e0 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -1,3 +1,4 @@
+#define __DISABLE_GUP_DEPRECATED 1
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/err.h>
@@ -807,15 +808,15 @@ static __always_inline long __get_user_pages_locked(struct task_struct *tsk,
* if (locked)
* up_read(&mm->mmap_sem);
*/
-long get_user_pages_locked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_locked6(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
int *locked)
{
- return __get_user_pages_locked(tsk, mm, start, nr_pages, write, force,
- pages, NULL, locked, true, FOLL_TOUCH);
+ return __get_user_pages_locked(current, current->mm, start, nr_pages,
+ write, force, pages, NULL, locked, true,
+ FOLL_TOUCH);
}
-EXPORT_SYMBOL(get_user_pages_locked);
+EXPORT_SYMBOL(get_user_pages_locked6);
/*
* Same as get_user_pages_unlocked(...., FOLL_TOUCH) but it allows to
@@ -860,14 +861,13 @@ EXPORT_SYMBOL(__get_user_pages_unlocked);
* or if "force" shall be set to 1 (get_user_pages_fast misses the
* "force" parameter).
*/
-long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_unlocked5(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages)
{
- return __get_user_pages_unlocked(tsk, mm, start, nr_pages, write,
- force, pages, FOLL_TOUCH);
+ return __get_user_pages_unlocked(current, current->mm, start, nr_pages,
+ write, force, pages, FOLL_TOUCH);
}
-EXPORT_SYMBOL(get_user_pages_unlocked);
+EXPORT_SYMBOL(get_user_pages_unlocked5);
/*
* get_user_pages_remote() - pin user pages in memory
@@ -939,16 +939,15 @@ EXPORT_SYMBOL(get_user_pages_remote);
* This is the same as get_user_pages_remote() for the time
* being.
*/
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages6(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
struct vm_area_struct **vmas)
{
- return __get_user_pages_locked(tsk, mm, start, nr_pages,
+ return __get_user_pages_locked(current, current->mm, start, nr_pages,
write, force, pages, vmas, NULL, false,
FOLL_TOUCH);
}
-EXPORT_SYMBOL(get_user_pages);
+EXPORT_SYMBOL(get_user_pages6);
/**
* populate_vma_page_range() - populate a range of pages in the vma.
@@ -1484,3 +1483,38 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write,
}
#endif /* CONFIG_HAVE_GENERIC_RCU_GUP */
+
+long get_user_pages8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
+{
+ WARN_ONCE(tsk != current, "get_user_pages() called on remote task");
+ WARN_ONCE(mm != current->mm, "get_user_pages() called on remote mm");
+
+ return get_user_pages6(start, nr_pages, write, force, pages, vmas);
+}
+EXPORT_SYMBOL(get_user_pages8);
+
+long get_user_pages_locked8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages, int *locked)
+{
+ WARN_ONCE(tsk != current, "get_user_pages_locked() called on remote task");
+ WARN_ONCE(mm != current->mm, "get_user_pages_locked() called on remote mm");
+
+ return get_user_pages_locked6(start, nr_pages, write, force, pages, locked);
+}
+EXPORT_SYMBOL(get_user_pages_locked8);
+
+long get_user_pages_unlocked7(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages)
+{
+ WARN_ONCE(tsk != current, "get_user_pages_unlocked() called on remote task");
+ WARN_ONCE(mm != current->mm, "get_user_pages_unlocked() called on remote mm");
+
+ return get_user_pages_unlocked5(start, nr_pages, write, force, pages);
+}
+EXPORT_SYMBOL(get_user_pages_unlocked7);
+
diff --git a/mm/nommu.c b/mm/nommu.c
index fbf6f0f1..b64d04d 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -15,6 +15,8 @@
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#define __DISABLE_GUP_DEPRECATED
+
#include <linux/export.h>
#include <linux/mm.h>
#include <linux/vmacache.h>
@@ -182,8 +184,7 @@ finish_or_fault:
* slab page or a secondary page from a compound page
* - don't permit access to VMAs that don't support it, such as I/O mappings
*/
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages6(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
struct vm_area_struct **vmas)
{
@@ -194,20 +195,18 @@ long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
if (force)
flags |= FOLL_FORCE;
- return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas,
- NULL);
+ return __get_user_pages(current, current->mm, start, nr_pages, flags,
+ pages, vmas, NULL);
}
-EXPORT_SYMBOL(get_user_pages);
+EXPORT_SYMBOL(get_user_pages6);
-long get_user_pages_locked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
- int write, int force, struct page **pages,
- int *locked)
+long get_user_pages_locked6(unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ int *locked)
{
- return get_user_pages(tsk, mm, start, nr_pages, write, force,
- pages, NULL);
+ return get_user_pages6(start, nr_pages, write, force, pages, NULL);
}
-EXPORT_SYMBOL(get_user_pages_locked);
+EXPORT_SYMBOL(get_user_pages_locked6);
long __get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
@@ -216,21 +215,20 @@ long __get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
{
long ret;
down_read(&mm->mmap_sem);
- ret = get_user_pages(tsk, mm, start, nr_pages, write, force,
- pages, NULL);
+ ret = __get_user_pages(tsk, mm, start, nr_pages, gup_flags, pages,
+ NULL, NULL);
up_read(&mm->mmap_sem);
return ret;
}
EXPORT_SYMBOL(__get_user_pages_unlocked);
-long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages,
+long get_user_pages_unlocked5(unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages)
{
- return __get_user_pages_unlocked(tsk, mm, start, nr_pages, write,
- force, pages, 0);
+ return __get_user_pages_unlocked(current, current->mm, start, nr_pages,
+ write, force, pages, 0);
}
-EXPORT_SYMBOL(get_user_pages_unlocked);
+EXPORT_SYMBOL(get_user_pages_unlocked5);
/**
* follow_pfn - look up PFN at a user virtual address
@@ -2108,3 +2106,31 @@ static int __meminit init_admin_reserve(void)
return 0;
}
subsys_initcall(init_admin_reserve);
+
+long get_user_pages8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
+{
+ return get_user_pages6(start, nr_pages, write, force, pages, vmas);
+}
+EXPORT_SYMBOL(get_user_pages8);
+
+long get_user_pages_locked8(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ int *locked)
+{
+ return get_user_pages_locked6(start, nr_pages, write,
+ force, pages, locked);
+}
+EXPORT_SYMBOL(get_user_pages_locked8);
+
+long get_user_pages_unlocked7(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages)
+{
+ return get_user_pages_unlocked5(start, nr_pages, write, force, pages);
+}
+EXPORT_SYMBOL(get_user_pages_unlocked7);
+
diff --git a/mm/util.c b/mm/util.c
index 4fb14ca..1e60116 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -283,9 +283,7 @@ EXPORT_SYMBOL_GPL(__get_user_pages_fast);
int __weak get_user_pages_fast(unsigned long start,
int nr_pages, int write, struct page **pages)
{
- struct mm_struct *mm = current->mm;
- return get_user_pages_unlocked(current, mm, start, nr_pages,
- write, 0, pages);
+ return get_user_pages_unlocked(start, nr_pages, write, 0, pages);
}
EXPORT_SYMBOL_GPL(get_user_pages_fast);
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-12 22:20 +0100 |
| Subject | [PATCH 01/33] mm: introduce get_user_pages_remote() |
| Message-ID | <r1tib-3WT-15@gated-at.bofh.it> |
| In reply to | #1333117 |
From: Dave Hansen <dave.hansen@linux.intel.com>
For protection keys, we need to understand whether protections
should be enforced in software or not. In general, we enforce
protections when working on our own task, but not when on others.
We call these "current" and "remote" operations.
This patch introduces a new get_user_pages() variant:
get_user_pages_remote()
Which is a replacement for when get_user_pages() is called on
non-current tsk/mm.
We also introduce a new gup flag: FOLL_REMOTE which can be used
for the "__" gup variants to get this new behavior.
The uprobes is_trap_at_addr() location holds mmap_sem and
calls get_user_pages(current->mm) on an instruction address. This
makes it a pretty unique gup caller. Being an instruction access
and also really originating from the kernel (vs. the app), I opted
to consider this a 'remote' access where protection keys will not
be enforced.
Without protection keys, this patch should not change any behavior.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: jack@suse.cz
---
b/drivers/gpu/drm/etnaviv/etnaviv_gem.c | 6 +++---
b/drivers/gpu/drm/i915/i915_gem_userptr.c | 10 +++++-----
b/drivers/infiniband/core/umem_odp.c | 8 ++++----
b/fs/exec.c | 8 ++++++--
b/include/linux/mm.h | 5 +++++
b/kernel/events/uprobes.c | 10 ++++++++--
b/mm/gup.c | 27 ++++++++++++++++++++++-----
b/mm/memory.c | 2 +-
b/mm/process_vm_access.c | 11 ++++++++---
b/security/tomoyo/domain.c | 9 ++++++++-
b/virt/kvm/async_pf.c | 8 +++++++-
11 files changed, 77 insertions(+), 27 deletions(-)
diff -puN drivers/gpu/drm/etnaviv/etnaviv_gem.c~introduce-get_user_pages_remote drivers/gpu/drm/etnaviv/etnaviv_gem.c
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.170106660 -0800
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c 2016-02-12 10:44:13.190107574 -0800
@@ -753,9 +753,9 @@ static struct page **etnaviv_gem_userptr
down_read(&mm->mmap_sem);
while (pinned < npages) {
- ret = get_user_pages(task, mm, ptr, npages - pinned,
- !etnaviv_obj->userptr.ro, 0,
- pvec + pinned, NULL);
+ ret = get_user_pages_remote(task, mm, ptr, npages - pinned,
+ !etnaviv_obj->userptr.ro, 0,
+ pvec + pinned, NULL);
if (ret < 0)
break;
diff -puN drivers/gpu/drm/i915/i915_gem_userptr.c~introduce-get_user_pages_remote drivers/gpu/drm/i915/i915_gem_userptr.c
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.171106705 -0800
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c 2016-02-12 10:44:13.190107574 -0800
@@ -584,11 +584,11 @@ __i915_gem_userptr_get_pages_worker(stru
down_read(&mm->mmap_sem);
while (pinned < npages) {
- ret = get_user_pages(work->task, mm,
- obj->userptr.ptr + pinned * PAGE_SIZE,
- npages - pinned,
- !obj->userptr.read_only, 0,
- pvec + pinned, NULL);
+ ret = get_user_pages_remote(work->task, mm,
+ obj->userptr.ptr + pinned * PAGE_SIZE,
+ npages - pinned,
+ !obj->userptr.read_only, 0,
+ pvec + pinned, NULL);
if (ret < 0)
break;
diff -puN drivers/infiniband/core/umem_odp.c~introduce-get_user_pages_remote drivers/infiniband/core/umem_odp.c
--- a/drivers/infiniband/core/umem_odp.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.173106797 -0800
+++ b/drivers/infiniband/core/umem_odp.c 2016-02-12 10:44:13.191107620 -0800
@@ -572,10 +572,10 @@ int ib_umem_odp_map_dma_pages(struct ib_
* complex (and doesn't gain us much performance in most use
* cases).
*/
- npages = get_user_pages(owning_process, owning_mm, user_virt,
- gup_num_pages,
- access_mask & ODP_WRITE_ALLOWED_BIT, 0,
- local_page_list, NULL);
+ npages = get_user_pages_remote(owning_process, owning_mm,
+ user_virt, gup_num_pages,
+ access_mask & ODP_WRITE_ALLOWED_BIT,
+ 0, local_page_list, NULL);
up_read(&owning_mm->mmap_sem);
if (npages < 0)
diff -puN fs/exec.c~introduce-get_user_pages_remote fs/exec.c
--- a/fs/exec.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.175106888 -0800
+++ b/fs/exec.c 2016-02-12 10:44:13.192107666 -0800
@@ -198,8 +198,12 @@ static struct page *get_arg_page(struct
return NULL;
}
#endif
- ret = get_user_pages(current, bprm->mm, pos,
- 1, write, 1, &page, NULL);
+ /*
+ * We are doing an exec(). 'current' is the process
+ * doing the exec and bprm->mm is the new process's mm.
+ */
+ ret = get_user_pages_remote(current, bprm->mm, pos, 1, write,
+ 1, &page, NULL);
if (ret <= 0)
return NULL;
diff -puN include/linux/mm.h~introduce-get_user_pages_remote include/linux/mm.h
--- a/include/linux/mm.h~introduce-get_user_pages_remote 2016-02-12 10:44:13.176106934 -0800
+++ b/include/linux/mm.h 2016-02-12 10:44:13.192107666 -0800
@@ -1225,6 +1225,10 @@ long __get_user_pages(struct task_struct
unsigned long start, unsigned long nr_pages,
unsigned int foll_flags, struct page **pages,
struct vm_area_struct **vmas, int *nonblocking);
+long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas);
long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
@@ -2168,6 +2172,7 @@ static inline struct page *follow_page(s
#define FOLL_MIGRATION 0x400 /* wait for page to replace migration entry */
#define FOLL_TRIED 0x800 /* a retry, previous pass started an IO */
#define FOLL_MLOCK 0x1000 /* lock present pages */
+#define FOLL_REMOTE 0x2000 /* we are working on non-current tsk/mm */
typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
void *data);
diff -puN kernel/events/uprobes.c~introduce-get_user_pages_remote kernel/events/uprobes.c
--- a/kernel/events/uprobes.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.178107026 -0800
+++ b/kernel/events/uprobes.c 2016-02-12 10:44:13.193107711 -0800
@@ -299,7 +299,7 @@ int uprobe_write_opcode(struct mm_struct
retry:
/* Read the page with vaddr into memory */
- ret = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma);
+ ret = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma);
if (ret <= 0)
return ret;
@@ -1700,7 +1700,13 @@ static int is_trap_at_addr(struct mm_str
if (likely(result == 0))
goto out;
- result = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
+ /*
+ * The NULL 'tsk' here ensures that any faults that occur here
+ * will not be accounted to the task. 'mm' *is* current->mm,
+ * but we treat this as a 'remote' access since it is
+ * essentially a kernel access to the memory.
+ */
+ result = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
if (result < 0)
return result;
diff -puN mm/gup.c~introduce-get_user_pages_remote mm/gup.c
--- a/mm/gup.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.180107117 -0800
+++ b/mm/gup.c 2016-02-12 10:44:13.194107757 -0800
@@ -870,7 +870,7 @@ long get_user_pages_unlocked(struct task
EXPORT_SYMBOL(get_user_pages_unlocked);
/*
- * get_user_pages() - pin user pages in memory
+ * get_user_pages_remote() - pin user pages in memory
* @tsk: the task_struct to use for page fault accounting, or
* NULL if faults are not to be recorded.
* @mm: mm_struct of target mm
@@ -924,12 +924,29 @@ EXPORT_SYMBOL(get_user_pages_unlocked);
* should use get_user_pages because it cannot pass
* FAULT_FLAG_ALLOW_RETRY to handle_mm_fault.
*/
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages, int write,
- int force, struct page **pages, struct vm_area_struct **vmas)
+long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
{
return __get_user_pages_locked(tsk, mm, start, nr_pages, write, force,
- pages, vmas, NULL, false, FOLL_TOUCH);
+ pages, vmas, NULL, false,
+ FOLL_TOUCH | FOLL_REMOTE);
+}
+EXPORT_SYMBOL(get_user_pages_remote);
+
+/*
+ * This is the same as get_user_pages_remote() for the time
+ * being.
+ */
+long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
+{
+ return __get_user_pages_locked(tsk, mm, start, nr_pages,
+ write, force, pages, vmas, NULL, false,
+ FOLL_TOUCH);
}
EXPORT_SYMBOL(get_user_pages);
diff -puN mm/memory.c~introduce-get_user_pages_remote mm/memory.c
--- a/mm/memory.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.182107208 -0800
+++ b/mm/memory.c 2016-02-12 10:44:13.195107803 -0800
@@ -3664,7 +3664,7 @@ static int __access_remote_vm(struct tas
void *maddr;
struct page *page = NULL;
- ret = get_user_pages(tsk, mm, addr, 1,
+ ret = get_user_pages_remote(tsk, mm, addr, 1,
write, 1, &page, &vma);
if (ret <= 0) {
#ifndef CONFIG_HAVE_IOREMAP_PROT
diff -puN mm/process_vm_access.c~introduce-get_user_pages_remote mm/process_vm_access.c
--- a/mm/process_vm_access.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.183107254 -0800
+++ b/mm/process_vm_access.c 2016-02-12 10:44:13.195107803 -0800
@@ -98,9 +98,14 @@ static int process_vm_rw_single_vec(unsi
int pages = min(nr_pages, max_pages_per_loop);
size_t bytes;
- /* Get the pages we're interested in */
- pages = get_user_pages_unlocked(task, mm, pa, pages,
- vm_write, 0, process_pages);
+ /*
+ * Get the pages we're interested in. We must
+ * add FOLL_REMOTE because task/mm might not
+ * current/current->mm
+ */
+ pages = __get_user_pages_unlocked(task, mm, pa, pages,
+ vm_write, 0, process_pages,
+ FOLL_REMOTE);
if (pages <= 0)
return -EFAULT;
diff -puN security/tomoyo/domain.c~introduce-get_user_pages_remote security/tomoyo/domain.c
--- a/security/tomoyo/domain.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.185107346 -0800
+++ b/security/tomoyo/domain.c 2016-02-12 10:44:13.196107848 -0800
@@ -874,7 +874,14 @@ bool tomoyo_dump_page(struct linux_binpr
}
/* Same with get_arg_page(bprm, pos, 0) in fs/exec.c */
#ifdef CONFIG_MMU
- if (get_user_pages(current, bprm->mm, pos, 1, 0, 1, &page, NULL) <= 0)
+ /*
+ * This is called at execve() time in order to dig around
+ * in the argv/environment of the new proceess
+ * (represented by bprm). 'current' is the process doing
+ * the execve().
+ */
+ if (get_user_pages_remote(current, bprm->mm, pos, 1,
+ 0, 1, &page, NULL) <= 0)
return false;
#else
page = bprm->page[pos / PAGE_SIZE];
diff -puN virt/kvm/async_pf.c~introduce-get_user_pages_remote virt/kvm/async_pf.c
--- a/virt/kvm/async_pf.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.187107437 -0800
+++ b/virt/kvm/async_pf.c 2016-02-12 10:44:13.196107848 -0800
@@ -79,7 +79,13 @@ static void async_pf_execute(struct work
might_sleep();
- get_user_pages_unlocked(NULL, mm, addr, 1, 1, 0, NULL);
+ /*
+ * This work is run asynchromously to the task which owns
+ * mm and might be done in another context, so we must
+ * use FOLL_REMOTE.
+ */
+ __get_user_pages_unlocked(NULL, mm, addr, 1, 1, 0, NULL, FOLL_REMOTE);
+
kvm_async_page_present_sync(vcpu, apf);
spin_lock(&vcpu->async_pf.lock);
_
[toc] | [prev] | [next] | [standalone]
| From | Balbir Singh <bsingharora@gmail.com> |
|---|---|
| Date | 2016-02-15 07:10 +0100 |
| Subject | Re: [PATCH 01/33] mm: introduce get_user_pages_remote() |
| Message-ID | <r2kw9-5tY-1@gated-at.bofh.it> |
| In reply to | #1333135 |
On Fri, 2016-02-12 at 13:01 -0800, Dave Hansen wrote: > From: Dave Hansen <dave.hansen@linux.intel.com> > > For protection keys, we need to understand whether protections > should be enforced in software or not. In general, we enforce > protections when working on our own task, but not when on others. > We call these "current" and "remote" operations. > > This patch introduces a new get_user_pages() variant: > > get_user_pages_remote() > > Which is a replacement for when get_user_pages() is called on > non-current tsk/mm. > In summary then get_user_pages_remote() do not enforce protections? Balbir Singh.
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-15 17:30 +0100 |
| Subject | Re: [PATCH 01/33] mm: introduce get_user_pages_remote() |
| Message-ID | <r2uc9-3qG-3@gated-at.bofh.it> |
| In reply to | #1334172 |
On 02/14/2016 10:09 PM, Balbir Singh wrote: >> > For protection keys, we need to understand whether protections >> > should be enforced in software or not. In general, we enforce >> > protections when working on our own task, but not when on others. >> > We call these "current" and "remote" operations. >> > >> > This patch introduces a new get_user_pages() variant: >> > >> > get_user_pages_remote() >> > >> > Which is a replacement for when get_user_pages() is called on >> > non-current tsk/mm. >> > > In summary then get_user_pages_remote() do not enforce protections? Yes, exactly.
[toc] | [prev] | [next] | [standalone]
| From | Srikar Dronamraju <srikar@linux.vnet.ibm.com> |
|---|---|
| Date | 2016-02-15 07:20 +0100 |
| Subject | Re: [PATCH 01/33] mm: introduce get_user_pages_remote() |
| Message-ID | <r2kFQ-5xp-13@gated-at.bofh.it> |
| In reply to | #1333135 |
> diff -puN kernel/events/uprobes.c~introduce-get_user_pages_remote kernel/events/uprobes.c > --- a/kernel/events/uprobes.c~introduce-get_user_pages_remote 2016-02-12 10:44:13.178107026 -0800 > +++ b/kernel/events/uprobes.c 2016-02-12 10:44:13.193107711 -0800 > @@ -299,7 +299,7 @@ int uprobe_write_opcode(struct mm_struct > > retry: > /* Read the page with vaddr into memory */ > - ret = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma); > + ret = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma); > if (ret <= 0) > return ret; > > @@ -1700,7 +1700,13 @@ static int is_trap_at_addr(struct mm_str > if (likely(result == 0)) > goto out; > > - result = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &page, NULL); > + /* > + * The NULL 'tsk' here ensures that any faults that occur here > + * will not be accounted to the task. 'mm' *is* current->mm, > + * but we treat this as a 'remote' access since it is > + * essentially a kernel access to the memory. > + */ > + result = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &page, NULL); > if (result < 0) > return result; > Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com> -- Thanks and Regards Srikar Dronamraju
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Dave Hansen <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-16 13:20 +0100 |
| Subject | [tip:x86/pkeys] mm/gup: Introduce get_user_pages_remote() |
| Message-ID | <r2MLN-7HT-25@gated-at.bofh.it> |
| In reply to | #1333135 |
Commit-ID: 1e9877902dc7e11d2be038371c6fbf2dfcd469d7
Gitweb: http://git.kernel.org/tip/1e9877902dc7e11d2be038371c6fbf2dfcd469d7
Author: Dave Hansen <dave.hansen@linux.intel.com>
AuthorDate: Fri, 12 Feb 2016 13:01:54 -0800
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Tue, 16 Feb 2016 10:04:09 +0100
mm/gup: Introduce get_user_pages_remote()
For protection keys, we need to understand whether protections
should be enforced in software or not. In general, we enforce
protections when working on our own task, but not when on others.
We call these "current" and "remote" operations.
This patch introduces a new get_user_pages() variant:
get_user_pages_remote()
Which is a replacement for when get_user_pages() is called on
non-current tsk/mm.
We also introduce a new gup flag: FOLL_REMOTE which can be used
for the "__" gup variants to get this new behavior.
The uprobes is_trap_at_addr() location holds mmap_sem and
calls get_user_pages(current->mm) on an instruction address. This
makes it a pretty unique gup caller. Being an instruction access
and also really originating from the kernel (vs. the app), I opted
to consider this a 'remote' access where protection keys will not
be enforced.
Without protection keys, this patch should not change any behavior.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
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: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: jack@suse.cz
Cc: linux-mm@kvack.org
Link: http://lkml.kernel.org/r/20160212210154.3F0E51EA@viggo.jf.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
drivers/gpu/drm/etnaviv/etnaviv_gem.c | 6 +++---
drivers/gpu/drm/i915/i915_gem_userptr.c | 10 +++++-----
drivers/infiniband/core/umem_odp.c | 8 ++++----
fs/exec.c | 8 ++++++--
include/linux/mm.h | 5 +++++
kernel/events/uprobes.c | 10 ++++++++--
mm/gup.c | 27 ++++++++++++++++++++++-----
mm/memory.c | 2 +-
mm/process_vm_access.c | 11 ++++++++---
security/tomoyo/domain.c | 9 ++++++++-
virt/kvm/async_pf.c | 8 +++++++-
11 files changed, 77 insertions(+), 27 deletions(-)
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
index 4b519e4..97d4457 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
@@ -753,9 +753,9 @@ static struct page **etnaviv_gem_userptr_do_get_pages(
down_read(&mm->mmap_sem);
while (pinned < npages) {
- ret = get_user_pages(task, mm, ptr, npages - pinned,
- !etnaviv_obj->userptr.ro, 0,
- pvec + pinned, NULL);
+ ret = get_user_pages_remote(task, mm, ptr, npages - pinned,
+ !etnaviv_obj->userptr.ro, 0,
+ pvec + pinned, NULL);
if (ret < 0)
break;
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 59e45b3..90dbf81 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -584,11 +584,11 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
down_read(&mm->mmap_sem);
while (pinned < npages) {
- ret = get_user_pages(work->task, mm,
- obj->userptr.ptr + pinned * PAGE_SIZE,
- npages - pinned,
- !obj->userptr.read_only, 0,
- pvec + pinned, NULL);
+ ret = get_user_pages_remote(work->task, mm,
+ obj->userptr.ptr + pinned * PAGE_SIZE,
+ npages - pinned,
+ !obj->userptr.read_only, 0,
+ pvec + pinned, NULL);
if (ret < 0)
break;
diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
index e69bf26..75077a0 100644
--- a/drivers/infiniband/core/umem_odp.c
+++ b/drivers/infiniband/core/umem_odp.c
@@ -572,10 +572,10 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt,
* complex (and doesn't gain us much performance in most use
* cases).
*/
- npages = get_user_pages(owning_process, owning_mm, user_virt,
- gup_num_pages,
- access_mask & ODP_WRITE_ALLOWED_BIT, 0,
- local_page_list, NULL);
+ npages = get_user_pages_remote(owning_process, owning_mm,
+ user_virt, gup_num_pages,
+ access_mask & ODP_WRITE_ALLOWED_BIT,
+ 0, local_page_list, NULL);
up_read(&owning_mm->mmap_sem);
if (npages < 0)
diff --git a/fs/exec.c b/fs/exec.c
index dcd4ac7..d885b98 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -198,8 +198,12 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
return NULL;
}
#endif
- ret = get_user_pages(current, bprm->mm, pos,
- 1, write, 1, &page, NULL);
+ /*
+ * We are doing an exec(). 'current' is the process
+ * doing the exec and bprm->mm is the new process's mm.
+ */
+ ret = get_user_pages_remote(current, bprm->mm, pos, 1, write,
+ 1, &page, NULL);
if (ret <= 0)
return NULL;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index b1d4b8c..faf3b70 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1225,6 +1225,10 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
unsigned int foll_flags, struct page **pages,
struct vm_area_struct **vmas, int *nonblocking);
+long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas);
long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
unsigned long start, unsigned long nr_pages,
int write, int force, struct page **pages,
@@ -2170,6 +2174,7 @@ static inline struct page *follow_page(struct vm_area_struct *vma,
#define FOLL_MIGRATION 0x400 /* wait for page to replace migration entry */
#define FOLL_TRIED 0x800 /* a retry, previous pass started an IO */
#define FOLL_MLOCK 0x1000 /* lock present pages */
+#define FOLL_REMOTE 0x2000 /* we are working on non-current tsk/mm */
typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
void *data);
diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 0167679..8eef5f5 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -299,7 +299,7 @@ int uprobe_write_opcode(struct mm_struct *mm, unsigned long vaddr,
retry:
/* Read the page with vaddr into memory */
- ret = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma);
+ ret = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma);
if (ret <= 0)
return ret;
@@ -1700,7 +1700,13 @@ static int is_trap_at_addr(struct mm_struct *mm, unsigned long vaddr)
if (likely(result == 0))
goto out;
- result = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
+ /*
+ * The NULL 'tsk' here ensures that any faults that occur here
+ * will not be accounted to the task. 'mm' *is* current->mm,
+ * but we treat this as a 'remote' access since it is
+ * essentially a kernel access to the memory.
+ */
+ result = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
if (result < 0)
return result;
diff --git a/mm/gup.c b/mm/gup.c
index 7bf19ff..36ca850 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -870,7 +870,7 @@ long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
EXPORT_SYMBOL(get_user_pages_unlocked);
/*
- * get_user_pages() - pin user pages in memory
+ * get_user_pages_remote() - pin user pages in memory
* @tsk: the task_struct to use for page fault accounting, or
* NULL if faults are not to be recorded.
* @mm: mm_struct of target mm
@@ -924,12 +924,29 @@ EXPORT_SYMBOL(get_user_pages_unlocked);
* should use get_user_pages because it cannot pass
* FAULT_FLAG_ALLOW_RETRY to handle_mm_fault.
*/
-long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
- unsigned long start, unsigned long nr_pages, int write,
- int force, struct page **pages, struct vm_area_struct **vmas)
+long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
{
return __get_user_pages_locked(tsk, mm, start, nr_pages, write, force,
- pages, vmas, NULL, false, FOLL_TOUCH);
+ pages, vmas, NULL, false,
+ FOLL_TOUCH | FOLL_REMOTE);
+}
+EXPORT_SYMBOL(get_user_pages_remote);
+
+/*
+ * This is the same as get_user_pages_remote() for the time
+ * being.
+ */
+long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
+ unsigned long start, unsigned long nr_pages,
+ int write, int force, struct page **pages,
+ struct vm_area_struct **vmas)
+{
+ return __get_user_pages_locked(tsk, mm, start, nr_pages,
+ write, force, pages, vmas, NULL, false,
+ FOLL_TOUCH);
}
EXPORT_SYMBOL(get_user_pages);
diff --git a/mm/memory.c b/mm/memory.c
index 38090ca..8bfbad0 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3685,7 +3685,7 @@ static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
void *maddr;
struct page *page = NULL;
- ret = get_user_pages(tsk, mm, addr, 1,
+ ret = get_user_pages_remote(tsk, mm, addr, 1,
write, 1, &page, &vma);
if (ret <= 0) {
#ifndef CONFIG_HAVE_IOREMAP_PROT
diff --git a/mm/process_vm_access.c b/mm/process_vm_access.c
index 5d453e5..07514d4 100644
--- a/mm/process_vm_access.c
+++ b/mm/process_vm_access.c
@@ -98,9 +98,14 @@ static int process_vm_rw_single_vec(unsigned long addr,
int pages = min(nr_pages, max_pages_per_loop);
size_t bytes;
- /* Get the pages we're interested in */
- pages = get_user_pages_unlocked(task, mm, pa, pages,
- vm_write, 0, process_pages);
+ /*
+ * Get the pages we're interested in. We must
+ * add FOLL_REMOTE because task/mm might not
+ * current/current->mm
+ */
+ pages = __get_user_pages_unlocked(task, mm, pa, pages,
+ vm_write, 0, process_pages,
+ FOLL_REMOTE);
if (pages <= 0)
return -EFAULT;
diff --git a/security/tomoyo/domain.c b/security/tomoyo/domain.c
index 3865145..ade7c6c 100644
--- a/security/tomoyo/domain.c
+++ b/security/tomoyo/domain.c
@@ -874,7 +874,14 @@ bool tomoyo_dump_page(struct linux_binprm *bprm, unsigned long pos,
}
/* Same with get_arg_page(bprm, pos, 0) in fs/exec.c */
#ifdef CONFIG_MMU
- if (get_user_pages(current, bprm->mm, pos, 1, 0, 1, &page, NULL) <= 0)
+ /*
+ * This is called at execve() time in order to dig around
+ * in the argv/environment of the new proceess
+ * (represented by bprm). 'current' is the process doing
+ * the execve().
+ */
+ if (get_user_pages_remote(current, bprm->mm, pos, 1,
+ 0, 1, &page, NULL) <= 0)
return false;
#else
page = bprm->page[pos / PAGE_SIZE];
diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
index 3531599..d604e87 100644
--- a/virt/kvm/async_pf.c
+++ b/virt/kvm/async_pf.c
@@ -79,7 +79,13 @@ static void async_pf_execute(struct work_struct *work)
might_sleep();
- get_user_pages_unlocked(NULL, mm, addr, 1, 1, 0, NULL);
+ /*
+ * This work is run asynchromously to the task which owns
+ * mm and might be done in another context, so we must
+ * use FOLL_REMOTE.
+ */
+ __get_user_pages_unlocked(NULL, mm, addr, 1, 1, 0, NULL, FOLL_REMOTE);
+
kvm_async_page_present_sync(vcpu, apf);
spin_lock(&vcpu->async_pf.lock);
[toc] | [prev] | [next] | [standalone]
| From | Konstantin Khlebnikov <koct9i@gmail.com> |
|---|---|
| Date | 2016-02-20 07:30 +0100 |
| Subject | Re: [tip:x86/pkeys] mm/gup: Introduce get_user_pages_remote() |
| Message-ID | <r49dg-MC-7@gated-at.bofh.it> |
| In reply to | #1335322 |
On Tue, Feb 16, 2016 at 3:14 PM, tip-bot for Dave Hansen
<tipbot@zytor.com> wrote:
> Commit-ID: 1e9877902dc7e11d2be038371c6fbf2dfcd469d7
> Gitweb: http://git.kernel.org/tip/1e9877902dc7e11d2be038371c6fbf2dfcd469d7
> Author: Dave Hansen <dave.hansen@linux.intel.com>
> AuthorDate: Fri, 12 Feb 2016 13:01:54 -0800
> Committer: Ingo Molnar <mingo@kernel.org>
> CommitDate: Tue, 16 Feb 2016 10:04:09 +0100
>
> mm/gup: Introduce get_user_pages_remote()
>
> For protection keys, we need to understand whether protections
> should be enforced in software or not. In general, we enforce
> protections when working on our own task, but not when on others.
> We call these "current" and "remote" operations.
>
> This patch introduces a new get_user_pages() variant:
>
> get_user_pages_remote()
>
> Which is a replacement for when get_user_pages() is called on
> non-current tsk/mm.
As I see task-struct argument could be NULL as well as in old api.
They only usage for it is updating task->maj_flt/min_flt.
May be just remove arg and always account major/minor faults into
current task - currently counters are plain unsigned long, so remote
access could corrupt them.
>
> We also introduce a new gup flag: FOLL_REMOTE which can be used
> for the "__" gup variants to get this new behavior.
>
> The uprobes is_trap_at_addr() location holds mmap_sem and
> calls get_user_pages(current->mm) on an instruction address. This
> makes it a pretty unique gup caller. Being an instruction access
> and also really originating from the kernel (vs. the app), I opted
> to consider this a 'remote' access where protection keys will not
> be enforced.
>
> Without protection keys, this patch should not change any behavior.
>
> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
> Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
> Cc: Andrea Arcangeli <aarcange@redhat.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Andy Lutomirski <luto@amacapital.net>
> 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: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> Cc: Linus Torvalds <torvalds@linux-foundation.org>
> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Rik van Riel <riel@redhat.com>
> Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
> Cc: Vlastimil Babka <vbabka@suse.cz>
> Cc: jack@suse.cz
> Cc: linux-mm@kvack.org
> Link: http://lkml.kernel.org/r/20160212210154.3F0E51EA@viggo.jf.intel.com
> Signed-off-by: Ingo Molnar <mingo@kernel.org>
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gem.c | 6 +++---
> drivers/gpu/drm/i915/i915_gem_userptr.c | 10 +++++-----
> drivers/infiniband/core/umem_odp.c | 8 ++++----
> fs/exec.c | 8 ++++++--
> include/linux/mm.h | 5 +++++
> kernel/events/uprobes.c | 10 ++++++++--
> mm/gup.c | 27 ++++++++++++++++++++++-----
> mm/memory.c | 2 +-
> mm/process_vm_access.c | 11 ++++++++---
> security/tomoyo/domain.c | 9 ++++++++-
> virt/kvm/async_pf.c | 8 +++++++-
> 11 files changed, 77 insertions(+), 27 deletions(-)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
> index 4b519e4..97d4457 100644
> --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c
> +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
> @@ -753,9 +753,9 @@ static struct page **etnaviv_gem_userptr_do_get_pages(
>
> down_read(&mm->mmap_sem);
> while (pinned < npages) {
> - ret = get_user_pages(task, mm, ptr, npages - pinned,
> - !etnaviv_obj->userptr.ro, 0,
> - pvec + pinned, NULL);
> + ret = get_user_pages_remote(task, mm, ptr, npages - pinned,
> + !etnaviv_obj->userptr.ro, 0,
> + pvec + pinned, NULL);
> if (ret < 0)
> break;
>
> diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
> index 59e45b3..90dbf81 100644
> --- a/drivers/gpu/drm/i915/i915_gem_userptr.c
> +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
> @@ -584,11 +584,11 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
>
> down_read(&mm->mmap_sem);
> while (pinned < npages) {
> - ret = get_user_pages(work->task, mm,
> - obj->userptr.ptr + pinned * PAGE_SIZE,
> - npages - pinned,
> - !obj->userptr.read_only, 0,
> - pvec + pinned, NULL);
> + ret = get_user_pages_remote(work->task, mm,
> + obj->userptr.ptr + pinned * PAGE_SIZE,
> + npages - pinned,
> + !obj->userptr.read_only, 0,
> + pvec + pinned, NULL);
> if (ret < 0)
> break;
>
> diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
> index e69bf26..75077a0 100644
> --- a/drivers/infiniband/core/umem_odp.c
> +++ b/drivers/infiniband/core/umem_odp.c
> @@ -572,10 +572,10 @@ int ib_umem_odp_map_dma_pages(struct ib_umem *umem, u64 user_virt, u64 bcnt,
> * complex (and doesn't gain us much performance in most use
> * cases).
> */
> - npages = get_user_pages(owning_process, owning_mm, user_virt,
> - gup_num_pages,
> - access_mask & ODP_WRITE_ALLOWED_BIT, 0,
> - local_page_list, NULL);
> + npages = get_user_pages_remote(owning_process, owning_mm,
> + user_virt, gup_num_pages,
> + access_mask & ODP_WRITE_ALLOWED_BIT,
> + 0, local_page_list, NULL);
> up_read(&owning_mm->mmap_sem);
>
> if (npages < 0)
> diff --git a/fs/exec.c b/fs/exec.c
> index dcd4ac7..d885b98 100644
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -198,8 +198,12 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
> return NULL;
> }
> #endif
> - ret = get_user_pages(current, bprm->mm, pos,
> - 1, write, 1, &page, NULL);
> + /*
> + * We are doing an exec(). 'current' is the process
> + * doing the exec and bprm->mm is the new process's mm.
> + */
> + ret = get_user_pages_remote(current, bprm->mm, pos, 1, write,
> + 1, &page, NULL);
> if (ret <= 0)
> return NULL;
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index b1d4b8c..faf3b70 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1225,6 +1225,10 @@ long __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
> unsigned long start, unsigned long nr_pages,
> unsigned int foll_flags, struct page **pages,
> struct vm_area_struct **vmas, int *nonblocking);
> +long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
> + unsigned long start, unsigned long nr_pages,
> + int write, int force, struct page **pages,
> + struct vm_area_struct **vmas);
> long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
> unsigned long start, unsigned long nr_pages,
> int write, int force, struct page **pages,
> @@ -2170,6 +2174,7 @@ static inline struct page *follow_page(struct vm_area_struct *vma,
> #define FOLL_MIGRATION 0x400 /* wait for page to replace migration entry */
> #define FOLL_TRIED 0x800 /* a retry, previous pass started an IO */
> #define FOLL_MLOCK 0x1000 /* lock present pages */
> +#define FOLL_REMOTE 0x2000 /* we are working on non-current tsk/mm */
>
> typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr,
> void *data);
> diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
> index 0167679..8eef5f5 100644
> --- a/kernel/events/uprobes.c
> +++ b/kernel/events/uprobes.c
> @@ -299,7 +299,7 @@ int uprobe_write_opcode(struct mm_struct *mm, unsigned long vaddr,
>
> retry:
> /* Read the page with vaddr into memory */
> - ret = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma);
> + ret = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma);
> if (ret <= 0)
> return ret;
>
> @@ -1700,7 +1700,13 @@ static int is_trap_at_addr(struct mm_struct *mm, unsigned long vaddr)
> if (likely(result == 0))
> goto out;
>
> - result = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
> + /*
> + * The NULL 'tsk' here ensures that any faults that occur here
> + * will not be accounted to the task. 'mm' *is* current->mm,
> + * but we treat this as a 'remote' access since it is
> + * essentially a kernel access to the memory.
> + */
> + result = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
> if (result < 0)
> return result;
>
> diff --git a/mm/gup.c b/mm/gup.c
> index 7bf19ff..36ca850 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -870,7 +870,7 @@ long get_user_pages_unlocked(struct task_struct *tsk, struct mm_struct *mm,
> EXPORT_SYMBOL(get_user_pages_unlocked);
>
> /*
> - * get_user_pages() - pin user pages in memory
> + * get_user_pages_remote() - pin user pages in memory
> * @tsk: the task_struct to use for page fault accounting, or
> * NULL if faults are not to be recorded.
> * @mm: mm_struct of target mm
> @@ -924,12 +924,29 @@ EXPORT_SYMBOL(get_user_pages_unlocked);
> * should use get_user_pages because it cannot pass
> * FAULT_FLAG_ALLOW_RETRY to handle_mm_fault.
> */
> -long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
> - unsigned long start, unsigned long nr_pages, int write,
> - int force, struct page **pages, struct vm_area_struct **vmas)
> +long get_user_pages_remote(struct task_struct *tsk, struct mm_struct *mm,
> + unsigned long start, unsigned long nr_pages,
> + int write, int force, struct page **pages,
> + struct vm_area_struct **vmas)
> {
> return __get_user_pages_locked(tsk, mm, start, nr_pages, write, force,
> - pages, vmas, NULL, false, FOLL_TOUCH);
> + pages, vmas, NULL, false,
> + FOLL_TOUCH | FOLL_REMOTE);
> +}
> +EXPORT_SYMBOL(get_user_pages_remote);
> +
> +/*
> + * This is the same as get_user_pages_remote() for the time
> + * being.
> + */
> +long get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
> + unsigned long start, unsigned long nr_pages,
> + int write, int force, struct page **pages,
> + struct vm_area_struct **vmas)
> +{
> + return __get_user_pages_locked(tsk, mm, start, nr_pages,
> + write, force, pages, vmas, NULL, false,
> + FOLL_TOUCH);
> }
> EXPORT_SYMBOL(get_user_pages);
>
> diff --git a/mm/memory.c b/mm/memory.c
> index 38090ca..8bfbad0 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -3685,7 +3685,7 @@ static int __access_remote_vm(struct task_struct *tsk, struct mm_struct *mm,
> void *maddr;
> struct page *page = NULL;
>
> - ret = get_user_pages(tsk, mm, addr, 1,
> + ret = get_user_pages_remote(tsk, mm, addr, 1,
> write, 1, &page, &vma);
> if (ret <= 0) {
> #ifndef CONFIG_HAVE_IOREMAP_PROT
> diff --git a/mm/process_vm_access.c b/mm/process_vm_access.c
> index 5d453e5..07514d4 100644
> --- a/mm/process_vm_access.c
> +++ b/mm/process_vm_access.c
> @@ -98,9 +98,14 @@ static int process_vm_rw_single_vec(unsigned long addr,
> int pages = min(nr_pages, max_pages_per_loop);
> size_t bytes;
>
> - /* Get the pages we're interested in */
> - pages = get_user_pages_unlocked(task, mm, pa, pages,
> - vm_write, 0, process_pages);
> + /*
> + * Get the pages we're interested in. We must
> + * add FOLL_REMOTE because task/mm might not
> + * current/current->mm
> + */
> + pages = __get_user_pages_unlocked(task, mm, pa, pages,
> + vm_write, 0, process_pages,
> + FOLL_REMOTE);
> if (pages <= 0)
> return -EFAULT;
>
> diff --git a/security/tomoyo/domain.c b/security/tomoyo/domain.c
> index 3865145..ade7c6c 100644
> --- a/security/tomoyo/domain.c
> +++ b/security/tomoyo/domain.c
> @@ -874,7 +874,14 @@ bool tomoyo_dump_page(struct linux_binprm *bprm, unsigned long pos,
> }
> /* Same with get_arg_page(bprm, pos, 0) in fs/exec.c */
> #ifdef CONFIG_MMU
> - if (get_user_pages(current, bprm->mm, pos, 1, 0, 1, &page, NULL) <= 0)
> + /*
> + * This is called at execve() time in order to dig around
> + * in the argv/environment of the new proceess
> + * (represented by bprm). 'current' is the process doing
> + * the execve().
> + */
> + if (get_user_pages_remote(current, bprm->mm, pos, 1,
> + 0, 1, &page, NULL) <= 0)
> return false;
> #else
> page = bprm->page[pos / PAGE_SIZE];
> diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
> index 3531599..d604e87 100644
> --- a/virt/kvm/async_pf.c
> +++ b/virt/kvm/async_pf.c
> @@ -79,7 +79,13 @@ static void async_pf_execute(struct work_struct *work)
>
> might_sleep();
>
> - get_user_pages_unlocked(NULL, mm, addr, 1, 1, 0, NULL);
> + /*
> + * This work is run asynchromously to the task which owns
> + * mm and might be done in another context, so we must
> + * use FOLL_REMOTE.
> + */
> + __get_user_pages_unlocked(NULL, mm, addr, 1, 1, 0, NULL, FOLL_REMOTE);
> +
> kvm_async_page_present_sync(vcpu, apf);
>
> spin_lock(&vcpu->async_pf.lock);
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-12 22:20 +0100 |
| Subject | [PATCH 05/33] x86, pkeys: Add Kconfig option |
| Message-ID | <r1tib-3WT-25@gated-at.bofh.it> |
| In reply to | #1333117 |
From: Dave Hansen <dave.hansen@linux.intel.com> I don't have a strong opinion on whether we need a Kconfig prompt or not. Protection Keys has relatively little code associated with it, and it is not a heavyweight feature to keep enabled. However, I can imagine that folks would still appreciate being able to disable it. Note that, with disabled-features.h, the checks in the code for protection keys are always the same: cpu_has(c, X86_FEATURE_PKU) With the config option disabled, this essentially turns into an #ifdef. We will hide the prompt for now. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> --- b/arch/x86/Kconfig | 4 ++++ 1 file changed, 4 insertions(+) diff -puN arch/x86/Kconfig~pkeys-01-kconfig arch/x86/Kconfig --- a/arch/x86/Kconfig~pkeys-01-kconfig 2016-02-12 10:44:15.909231872 -0800 +++ b/arch/x86/Kconfig 2016-02-12 10:44:15.913232055 -0800 @@ -1713,6 +1713,10 @@ config X86_INTEL_MPX If unsure, say N. +config X86_INTEL_MEMORY_PROTECTION_KEYS + def_bool y + depends on CPU_SUP_INTEL && X86_64 + config EFI bool "EFI runtime service support" depends on ACPI _
[toc] | [prev] | [next] | [standalone]
| From | tip-bot for Dave Hansen <tipbot@zytor.com> |
|---|---|
| Date | 2016-02-18 21:20 +0100 |
| Subject | [tip:mm/pkeys] x86/mm/pkeys: Add Kconfig option |
| Message-ID | <r3Ddo-2wj-33@gated-at.bofh.it> |
| In reply to | #1333137 |
Commit-ID: 35e97790f5f1e5cf2b5522c55e3e31d5c81bd226 Gitweb: http://git.kernel.org/tip/35e97790f5f1e5cf2b5522c55e3e31d5c81bd226 Author: Dave Hansen <dave.hansen@linux.intel.com> AuthorDate: Fri, 12 Feb 2016 13:02:00 -0800 Committer: Ingo Molnar <mingo@kernel.org> CommitDate: Tue, 16 Feb 2016 10:11:13 +0100 x86/mm/pkeys: Add Kconfig option I don't have a strong opinion on whether we need a Kconfig prompt or not. Protection Keys has relatively little code associated with it, and it is not a heavyweight feature to keep enabled. However, I can imagine that folks would still appreciate being able to disable it. Note that, with disabled-features.h, the checks in the code for protection keys are always the same: cpu_has(c, X86_FEATURE_PKU) With the config option disabled, this essentially turns into an We will hide the prompt for now. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Lutomirski <luto@amacapital.net> 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: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20160212210200.DB7055E8@viggo.jf.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org> --- arch/x86/Kconfig | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index ab2ed53..3632cdd 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1714,6 +1714,10 @@ config X86_INTEL_MPX If unsure, say N. +config X86_INTEL_MEMORY_PROTECTION_KEYS + def_bool y + depends on CPU_SUP_INTEL && X86_64 + config EFI bool "EFI runtime service support" depends on ACPI
[toc] | [prev] | [next] | [standalone]
| From | Borislav Petkov <bp@alien8.de> |
|---|---|
| Date | 2016-02-19 12:30 +0100 |
| Subject | [PATCH] x86/mm/pkeys: Do not enable them by default |
| Message-ID | <r3Rq2-4mn-21@gated-at.bofh.it> |
| In reply to | #1337681 |
On Thu, Feb 18, 2016 at 12:16:47PM -0800, tip-bot for Dave Hansen wrote:
> Commit-ID: 35e97790f5f1e5cf2b5522c55e3e31d5c81bd226
> Gitweb: http://git.kernel.org/tip/35e97790f5f1e5cf2b5522c55e3e31d5c81bd226
> Author: Dave Hansen <dave.hansen@linux.intel.com>
> AuthorDate: Fri, 12 Feb 2016 13:02:00 -0800
> Committer: Ingo Molnar <mingo@kernel.org>
> CommitDate: Tue, 16 Feb 2016 10:11:13 +0100
>
> x86/mm/pkeys: Add Kconfig option
>
> I don't have a strong opinion on whether we need a Kconfig prompt
> or not. Protection Keys has relatively little code associated
> with it, and it is not a heavyweight feature to keep enabled.
> However, I can imagine that folks would still appreciate being
> able to disable it.
>
> Note that, with disabled-features.h, the checks in the code
> for protection keys are always the same:
>
> cpu_has(c, X86_FEATURE_PKU)
>
> With the config option disabled, this essentially turns into an
whoops, something is missing here. An "#ifdef."
...
> arch/x86/Kconfig | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
> index ab2ed53..3632cdd 100644
> --- a/arch/x86/Kconfig
> +++ b/arch/x86/Kconfig
> @@ -1714,6 +1714,10 @@ config X86_INTEL_MPX
>
> If unsure, say N.
>
> +config X86_INTEL_MEMORY_PROTECTION_KEYS
> + def_bool y
This is not necessary.
---
From: Borislav Petkov <bp@suse.de>
Date: Fri, 19 Feb 2016 12:19:50 +0100
Subject: [PATCH] x86/mm/pkeys: Do not enable them by default
No need to default to y.
Signed-off-by: Borislav Petkov <bp@suse.de>
---
arch/x86/Kconfig | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index d10826d2cb5e..109bc46ccb60 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1719,8 +1719,7 @@ config X86_INTEL_MPX
If unsure, say N.
config X86_INTEL_MEMORY_PROTECTION_KEYS
- prompt "Intel Memory Protection Keys"
- def_bool y
+ bool "Intel Memory Protection Keys"
# Note: only available in 64-bit mode
depends on CPU_SUP_INTEL && X86_64
---help---
--
2.3.5
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
[toc] | [prev] | [next] | [standalone]
| From | Dave Hansen <dave@sr71.net> |
|---|---|
| Date | 2016-02-19 18:20 +0100 |
| Subject | Re: [PATCH] x86/mm/pkeys: Do not enable them by default |
| Message-ID | <r3WSK-8qu-11@gated-at.bofh.it> |
| In reply to | #1338063 |
On 02/19/2016 03:27 AM, Borislav Petkov wrote: > config X86_INTEL_MEMORY_PROTECTION_KEYS > - prompt "Intel Memory Protection Keys" > - def_bool y > + bool "Intel Memory Protection Keys" > # Note: only available in 64-bit mode > depends on CPU_SUP_INTEL && X86_64 > ---help--- Hi Borislav, I'd really prefer this be left on by default. This is a feature that I expect to be widely enabled in distribution kernels.
[toc] | [prev] | [next] | [standalone]
| From | Borislav Petkov <bp@alien8.de> |
|---|---|
| Date | 2016-02-19 18:30 +0100 |
| Subject | Re: [PATCH] x86/mm/pkeys: Do not enable them by default |
| Message-ID | <r3X2q-8uR-13@gated-at.bofh.it> |
| In reply to | #1338362 |
On Fri, Feb 19, 2016 at 09:11:03AM -0800, Dave Hansen wrote:
> I'd really prefer this be left on by default. This is a feature that I
> expect to be widely enabled in distribution kernels.
Distribution kernels can enable it without defaulting to y here. Also,
this code doesn't need to be built on the majority of x86 boxes out
there because they don't have the hw support.
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
[toc] | [prev] | [next] | [standalone]
Page 2 of 3 — ← Prev page 1 [2] 3 Next page →
Back to top | Article view | linux.kernel
csiph-web