* [PATCH 0/4] [v3] x86 disabled features macros
@ 2014-07-02 16:29 Dave Hansen
2014-07-02 16:29 ` [PATCH 1/4] x86: axe the lightly-used cpu_has_pae Dave Hansen
` (3 more replies)
0 siblings, 4 replies; 9+ messages in thread
From: Dave Hansen @ 2014-07-02 16:29 UTC (permalink / raw)
To: bp; +Cc: x86, hpa, qiaowei.ren, linux-kernel, Dave Hansen
Note: the fourth patch in the series would not hurt if it got
pulled in to a tree now, but it does not make any sense until we
get MPX merged. We plan to roll it in to the MPX patch set.
Changes from v2:
* also axe PAE macro
* fix typo in comment
* put K6_MTRR in the right DISABLED_MASK leaf
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH 1/4] x86: axe the lightly-used cpu_has_pae
2014-07-02 16:29 [PATCH 0/4] [v3] x86 disabled features macros Dave Hansen
@ 2014-07-02 16:29 ` Dave Hansen
2014-07-02 19:19 ` Borislav Petkov
2014-07-02 16:29 ` [PATCH 2/4] x86: introduce disabled-features Dave Hansen
` (2 subsequent siblings)
3 siblings, 1 reply; 9+ messages in thread
From: Dave Hansen @ 2014-07-02 16:29 UTC (permalink / raw)
To: bp; +Cc: x86, hpa, qiaowei.ren, linux-kernel, Dave Hansen
From: Dave Hansen <dave.hansen@linux.intel.com>
cpu_has_pae is only referenced in one place: the X86_32 kexec
code (in a file not even built on 64-bit). It hardly warrants
its own macro, or the trouble we go to ensuring that it can't
be called in X86_64 code.
Axe the macro and replace it with a direct cpu feature check.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
---
b/arch/x86/include/asm/cpufeature.h | 4 ----
b/arch/x86/kernel/machine_kexec_32.c | 3 ++-
2 files changed, 2 insertions(+), 5 deletions(-)
diff -puN arch/x86/include/asm/cpufeature.h~x86-axe-cpu_has_pae arch/x86/include/asm/cpufeature.h
--- a/arch/x86/include/asm/cpufeature.h~x86-axe-cpu_has_pae 2014-07-02 09:16:20.852254443 -0700
+++ b/arch/x86/include/asm/cpufeature.h 2014-07-02 09:16:20.857254668 -0700
@@ -286,7 +286,6 @@ extern const char * const x86_power_flag
#define cpu_has_de boot_cpu_has(X86_FEATURE_DE)
#define cpu_has_pse boot_cpu_has(X86_FEATURE_PSE)
#define cpu_has_tsc boot_cpu_has(X86_FEATURE_TSC)
-#define cpu_has_pae boot_cpu_has(X86_FEATURE_PAE)
#define cpu_has_pge boot_cpu_has(X86_FEATURE_PGE)
#define cpu_has_apic boot_cpu_has(X86_FEATURE_APIC)
#define cpu_has_sep boot_cpu_has(X86_FEATURE_SEP)
@@ -344,9 +343,6 @@ extern const char * const x86_power_flag
#undef cpu_has_vme
#define cpu_has_vme 0
-#undef cpu_has_pae
-#define cpu_has_pae ___BUG___
-
#undef cpu_has_mp
#define cpu_has_mp 1
diff -puN arch/x86/kernel/machine_kexec_32.c~x86-axe-cpu_has_pae arch/x86/kernel/machine_kexec_32.c
--- a/arch/x86/kernel/machine_kexec_32.c~x86-axe-cpu_has_pae 2014-07-02 09:16:20.853254488 -0700
+++ b/arch/x86/kernel/machine_kexec_32.c 2014-07-02 09:16:20.857254668 -0700
@@ -247,7 +247,8 @@ void machine_kexec(struct kimage *image)
/* now call it */
image->start = relocate_kernel_ptr((unsigned long)image->head,
(unsigned long)page_list,
- image->start, cpu_has_pae,
+ image->start,
+ boot_cpu_has(X86_FEATURE_PAE),
image->preserve_context);
#ifdef CONFIG_KEXEC_JUMP
_
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH 2/4] x86: introduce disabled-features
2014-07-02 16:29 [PATCH 0/4] [v3] x86 disabled features macros Dave Hansen
2014-07-02 16:29 ` [PATCH 1/4] x86: axe the lightly-used cpu_has_pae Dave Hansen
@ 2014-07-02 16:29 ` Dave Hansen
2014-07-02 19:27 ` Borislav Petkov
2014-07-02 16:29 ` [PATCH 3/4] x86: add more disabled features Dave Hansen
2014-07-02 16:29 ` [PATCH 4/4] x86: add MPX to disaabled features Dave Hansen
3 siblings, 1 reply; 9+ messages in thread
From: Dave Hansen @ 2014-07-02 16:29 UTC (permalink / raw)
To: bp; +Cc: x86, hpa, qiaowei.ren, linux-kernel, Dave Hansen
From: Dave Hansen <dave.hansen@linux.intel.com>
I believe the REQUIRED_MASK aproach was taken so that it was
easier to consult in assembly (arch/x86/kernel/verify_cpu.S).
DISABLED_MASK does not have the same restriction, but I
implemented it the same way for consistency.
We have a REQUIRED_MASK... which does two things:
1. Keeps a list of cpuid bits to check in very early boot and
refuse to boot if those are not present.
2. Consulted during cpu_has() checks, which allows us to
optimize out things at compile-time. In other words, if we
*KNOW* we will not boot with the feature off, then we can
safely assume that it will be present forever.
But, we don't have a similar mechanism for CPU features which
may be present but that we know we will not use. We simply
use our existing mechanisms to repeatedly check the status of
the bit at runtime (well, the alternatives patching helps here
but it does not provide compile-time optimization).
Adding a feature to disabled-features.h allows the bit to be
checked via a new macro: cpu_feature_enabled(). Note that
for features in DISABLED_MASK, checks with this macro have
all of the benefits of an #ifdef. Before, we would have done
this in a header:
#ifdef CONFIG_X86_INTEL_MPX
#define cpu_has_mpx cpu_has(X86_FEATURE_MPX)
#else
#define cpu_has_mpx 0
#endif
and this in the code:
if (cpu_has_mpx)
do_some_mpx_thing();
Now, just add your feature to DISABLED_MASK and you can do this
everywhere, and get the same benefits you would have from
#ifdefs:
if (cpu_feature_enabled(X86_FEATURE_MPX))
do_some_mpx_thing();
We need a new function and *not* a modification to cpu_has()
because there are cases where we actually need to check the CPU
itself, despite what features the kernel supports. The best
example of this is a hypervisor which has no control over what
features its guests are using and where the guest does not depend
on the host for support.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
---
b/arch/x86/boot/mkcpustr.c | 1 +
b/arch/x86/include/asm/cpufeature.h | 28 ++++++++++++++++++++++++++++
b/arch/x86/include/asm/disabled-features.h | 27 +++++++++++++++++++++++++++
3 files changed, 56 insertions(+)
diff -puN arch/x86/boot/mkcpustr.c~x86-disabled_features arch/x86/boot/mkcpustr.c
--- a/arch/x86/boot/mkcpustr.c~x86-disabled_features 2014-07-02 09:16:21.137267243 -0700
+++ b/arch/x86/boot/mkcpustr.c 2014-07-02 09:16:21.142267468 -0700
@@ -16,6 +16,7 @@
#include <stdio.h>
#include "../include/asm/required-features.h"
+#include "../include/asm/disabled-features.h"
#include "../include/asm/cpufeature.h"
#include "../kernel/cpu/capflags.c"
diff -puN arch/x86/include/asm/cpufeature.h~x86-disabled_features arch/x86/include/asm/cpufeature.h
--- a/arch/x86/include/asm/cpufeature.h~x86-disabled_features 2014-07-02 09:16:21.139267333 -0700
+++ b/arch/x86/include/asm/cpufeature.h 2014-07-02 09:16:21.143267514 -0700
@@ -8,6 +8,10 @@
#include <asm/required-features.h>
#endif
+#ifndef _ASM_X86_DISABLED_FEATURES_H
+#include <asm/disabled-features.h>
+#endif
+
#define NCAPINTS 10 /* N 32-bit words worth of info */
#define NBUGINTS 1 /* N 32-bit bug flags */
@@ -260,6 +264,18 @@ extern const char * const x86_power_flag
(((bit)>>5)==8 && (1UL<<((bit)&31) & REQUIRED_MASK8)) || \
(((bit)>>5)==9 && (1UL<<((bit)&31) & REQUIRED_MASK9)) )
+#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)) )
+
#define cpu_has(c, bit) \
(__builtin_constant_p(bit) && REQUIRED_MASK_BIT_SET(bit) ? 1 : \
test_cpu_cap(c, bit))
@@ -268,6 +284,18 @@ extern const char * const x86_power_flag
(__builtin_constant_p(bit) && REQUIRED_MASK_BIT_SET(bit) ? 1 : \
x86_this_cpu_test_bit(bit, (unsigned long *)&cpu_info.x86_capability))
+/*
+ * This macro is for detection of features which need kernel
+ * infrastructure to be used. It may *not* directly test the CPU
+ * itself. Use the cpu_has() family if you want true runtime
+ * testing of CPU features, like in hypervisor code where you are
+ * supporting a possible guest feature where host support for it
+ * is not relevant.
+ */
+#define cpu_feature_enabled(bit) \
+ (__builtin_constant_p(bit) && DISABLED_MASK_BIT_SET(bit) ? 0 : \
+ cpu_has(&boot_cpu_data, bit))
+
#define boot_cpu_has(bit) cpu_has(&boot_cpu_data, bit)
#define set_cpu_cap(c, bit) set_bit(bit, (unsigned long *)((c)->x86_capability))
diff -puN /dev/null arch/x86/include/asm/disabled-features.h
--- /dev/null 2014-04-10 11:28:14.066815724 -0700
+++ b/arch/x86/include/asm/disabled-features.h 2014-07-02 09:16:21.143267514 -0700
@@ -0,0 +1,27 @@
+#ifndef _ASM_X86_DISABLED_FEATURES_H
+#define _ASM_X86_DISABLED_FEATURES_H
+
+/* These features, although they might be available in a CPU
+ * will not be used because the compile options to support
+ * them are not present.
+ *
+ * This code allows them to be checked and disabled at
+ * compile time without an explicit #ifdef. Use
+ * cpu_feature_enabled().
+ */
+
+/*
+ * Make sure to add features to the correct mask
+ */
+#define DISABLED_MASK0 0
+#define DISABLED_MASK1 0
+#define DISABLED_MASK2 0
+#define DISABLED_MASK3 0
+#define DISABLED_MASK4 0
+#define DISABLED_MASK5 0
+#define DISABLED_MASK6 0
+#define DISABLED_MASK7 0
+#define DISABLED_MASK8 0
+#define DISABLED_MASK9 0
+
+#endif /* _ASM_X86_DISABLED_FEATURES_H */
_
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH 3/4] x86: add more disabled features
2014-07-02 16:29 [PATCH 0/4] [v3] x86 disabled features macros Dave Hansen
2014-07-02 16:29 ` [PATCH 1/4] x86: axe the lightly-used cpu_has_pae Dave Hansen
2014-07-02 16:29 ` [PATCH 2/4] x86: introduce disabled-features Dave Hansen
@ 2014-07-02 16:29 ` Dave Hansen
2014-07-02 19:34 ` Borislav Petkov
2014-07-02 16:29 ` [PATCH 4/4] x86: add MPX to disaabled features Dave Hansen
3 siblings, 1 reply; 9+ messages in thread
From: Dave Hansen @ 2014-07-02 16:29 UTC (permalink / raw)
To: bp; +Cc: x86, hpa, qiaowei.ren, linux-kernel, Dave Hansen
From: Dave Hansen <dave.hansen@linux.intel.com>
There are a few other features than MPX that we can make
assumptions about at compile-time based on compile options.
Add them to disabled-features.h and check them with
cpu_feature_enabled().
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
---
b/arch/x86/include/asm/cpufeature.h | 16 ----------------
b/arch/x86/include/asm/disabled-features.h | 16 ++++++++++++++--
b/arch/x86/kernel/cpu/common.c | 2 +-
b/arch/x86/kernel/cpu/mtrr/main.c | 6 +++---
4 files changed, 18 insertions(+), 22 deletions(-)
diff -puN arch/x86/include/asm/cpufeature.h~x86-disabled_features-addmore arch/x86/include/asm/cpufeature.h
--- a/arch/x86/include/asm/cpufeature.h~x86-disabled_features-addmore 2014-07-02 09:16:21.437280769 -0700
+++ b/arch/x86/include/asm/cpufeature.h 2014-07-02 09:16:21.446281174 -0700
@@ -310,7 +310,6 @@ extern const char * const x86_power_flag
} while (0)
#define cpu_has_fpu boot_cpu_has(X86_FEATURE_FPU)
-#define cpu_has_vme boot_cpu_has(X86_FEATURE_VME)
#define cpu_has_de boot_cpu_has(X86_FEATURE_DE)
#define cpu_has_pse boot_cpu_has(X86_FEATURE_PSE)
#define cpu_has_tsc boot_cpu_has(X86_FEATURE_TSC)
@@ -330,9 +329,6 @@ extern const char * const x86_power_flag
#define cpu_has_ht boot_cpu_has(X86_FEATURE_HT)
#define cpu_has_mp boot_cpu_has(X86_FEATURE_MP)
#define cpu_has_nx boot_cpu_has(X86_FEATURE_NX)
-#define cpu_has_k6_mtrr boot_cpu_has(X86_FEATURE_K6_MTRR)
-#define cpu_has_cyrix_arr boot_cpu_has(X86_FEATURE_CYRIX_ARR)
-#define cpu_has_centaur_mcr boot_cpu_has(X86_FEATURE_CENTAUR_MCR)
#define cpu_has_xstore boot_cpu_has(X86_FEATURE_XSTORE)
#define cpu_has_xstore_enabled boot_cpu_has(X86_FEATURE_XSTORE_EN)
#define cpu_has_xcrypt boot_cpu_has(X86_FEATURE_XCRYPT)
@@ -368,21 +364,9 @@ extern const char * const x86_power_flag
#ifdef CONFIG_X86_64
-#undef cpu_has_vme
-#define cpu_has_vme 0
-
#undef cpu_has_mp
#define cpu_has_mp 1
-#undef cpu_has_k6_mtrr
-#define cpu_has_k6_mtrr 0
-
-#undef cpu_has_cyrix_arr
-#define cpu_has_cyrix_arr 0
-
-#undef cpu_has_centaur_mcr
-#define cpu_has_centaur_mcr 0
-
#endif /* CONFIG_X86_64 */
#if __GNUC__ >= 4
diff -puN arch/x86/include/asm/disabled-features.h~x86-disabled_features-addmore arch/x86/include/asm/disabled-features.h
--- a/arch/x86/include/asm/disabled-features.h~x86-disabled_features-addmore 2014-07-02 09:16:21.439280859 -0700
+++ b/arch/x86/include/asm/disabled-features.h 2014-07-02 09:16:21.447281218 -0700
@@ -10,13 +10,25 @@
* cpu_feature_enabled().
*/
+#ifdef CONFIG_X86_64
+# define DISABLE_VME (1<<(X86_FEATURE_VME & 31))
+# define DISABLE_K6_MTRR (1<<(X86_FEATURE_K6_MTRR & 31))
+# define DISABLE_CYRIX_ARR (1<<(X86_FEATURE_CYRIX_ARR & 31))
+# define DISABLE_CENTAUR_MCR (1<<(X86_FEATURE_CENTAUR_MCR & 31))
+#else
+# define DISABLE_VME 0
+# define DISABLE_K6_MTRR 0
+# define DISABLE_CYRIX_ARR 0
+# define DISABLE_CENTAUR_MCR 0
+#endif /* CONFIG_X86_64 */
+
/*
* Make sure to add features to the correct mask
*/
-#define DISABLED_MASK0 0
+#define DISABLED_MASK0 (DISABLE_VME)
#define DISABLED_MASK1 0
#define DISABLED_MASK2 0
-#define DISABLED_MASK3 0
+#define DISABLED_MASK3 (DISABLE_CYRIX_ARR|DISABLE_CENTAUR_MCR|DISABLE_K6_MTRR)
#define DISABLED_MASK4 0
#define DISABLED_MASK5 0
#define DISABLED_MASK6 0
diff -puN arch/x86/kernel/cpu/common.c~x86-disabled_features-addmore arch/x86/kernel/cpu/common.c
--- a/arch/x86/kernel/cpu/common.c~x86-disabled_features-addmore 2014-07-02 09:16:21.442280992 -0700
+++ b/arch/x86/kernel/cpu/common.c 2014-07-02 09:16:21.447281218 -0700
@@ -1346,7 +1346,7 @@ void cpu_init(void)
printk(KERN_INFO "Initializing CPU#%d\n", cpu);
- if (cpu_has_vme || cpu_has_tsc || cpu_has_de)
+ if (cpu_feature_enabled(X86_FEATURE_VME) || cpu_has_tsc || cpu_has_de)
clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
load_current_idt();
diff -puN arch/x86/kernel/cpu/mtrr/main.c~x86-disabled_features-addmore arch/x86/kernel/cpu/mtrr/main.c
--- a/arch/x86/kernel/cpu/mtrr/main.c~x86-disabled_features-addmore 2014-07-02 09:16:21.442280992 -0700
+++ b/arch/x86/kernel/cpu/mtrr/main.c 2014-07-02 09:16:21.448281262 -0700
@@ -707,7 +707,7 @@ void __init mtrr_bp_init(void)
} else {
switch (boot_cpu_data.x86_vendor) {
case X86_VENDOR_AMD:
- if (cpu_has_k6_mtrr) {
+ if (cpu_feature_enabled(X86_FEATURE_K6_MTRR)) {
/* Pre-Athlon (K6) AMD CPU MTRRs */
mtrr_if = mtrr_ops[X86_VENDOR_AMD];
size_or_mask = SIZE_OR_MASK_BITS(32);
@@ -715,14 +715,14 @@ void __init mtrr_bp_init(void)
}
break;
case X86_VENDOR_CENTAUR:
- if (cpu_has_centaur_mcr) {
+ if (cpu_feature_enabled(X86_FEATURE_CENTAUR_MCR)) {
mtrr_if = mtrr_ops[X86_VENDOR_CENTAUR];
size_or_mask = SIZE_OR_MASK_BITS(32);
size_and_mask = 0;
}
break;
case X86_VENDOR_CYRIX:
- if (cpu_has_cyrix_arr) {
+ if (cpu_feature_enabled(X86_FEATURE_CYRIX_ARR)) {
mtrr_if = mtrr_ops[X86_VENDOR_CYRIX];
size_or_mask = SIZE_OR_MASK_BITS(32);
size_and_mask = 0;
_
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH 4/4] x86: add MPX to disaabled features
2014-07-02 16:29 [PATCH 0/4] [v3] x86 disabled features macros Dave Hansen
` (2 preceding siblings ...)
2014-07-02 16:29 ` [PATCH 3/4] x86: add more disabled features Dave Hansen
@ 2014-07-02 16:29 ` Dave Hansen
2014-07-02 19:43 ` Borislav Petkov
3 siblings, 1 reply; 9+ messages in thread
From: Dave Hansen @ 2014-07-02 16:29 UTC (permalink / raw)
To: bp; +Cc: x86, hpa, qiaowei.ren, linux-kernel, Dave Hansen
From: Dave Hansen <dave.hansen@linux.intel.com>
This will go in to the MPX patch set, but I'm posting it here so
folk can see how it will work since it was the original
motivation for this work.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
---
b/arch/x86/include/asm/disabled-features.h | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff -puN arch/x86/include/asm/disabled-features.h~x86-diabled-features-mpx arch/x86/include/asm/disabled-features.h
--- a/arch/x86/include/asm/disabled-features.h~x86-diabled-features-mpx 2014-07-02 09:16:34.368863016 -0700
+++ b/arch/x86/include/asm/disabled-features.h 2014-07-02 09:16:34.371863152 -0700
@@ -10,6 +10,12 @@
* cpu_feature_enabled().
*/
+#ifdef CONFIG_X86_INTEL_MPX
+# define DISABLE_MPX 0
+#else
+# define DISABLE_MPX (1<<(X86_FEATURE_MPX & 31))
+#endif
+
#ifdef CONFIG_X86_64
# define DISABLE_VME (1<<(X86_FEATURE_VME & 31))
# define DISABLE_K6_MTRR (1<<(X86_FEATURE_K6_MTRR & 31))
@@ -34,6 +40,6 @@
#define DISABLED_MASK6 0
#define DISABLED_MASK7 0
#define DISABLED_MASK8 0
-#define DISABLED_MASK9 0
+#define DISABLED_MASK9 (DISABLE_MPX)
#endif /* _ASM_X86_DISABLED_FEATURES_H */
_
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 1/4] x86: axe the lightly-used cpu_has_pae
2014-07-02 16:29 ` [PATCH 1/4] x86: axe the lightly-used cpu_has_pae Dave Hansen
@ 2014-07-02 19:19 ` Borislav Petkov
0 siblings, 0 replies; 9+ messages in thread
From: Borislav Petkov @ 2014-07-02 19:19 UTC (permalink / raw)
To: Dave Hansen; +Cc: x86, hpa, qiaowei.ren, linux-kernel
On Wed, Jul 02, 2014 at 09:29:26AM -0700, Dave Hansen wrote:
>
> From: Dave Hansen <dave.hansen@linux.intel.com>
>
> cpu_has_pae is only referenced in one place: the X86_32 kexec
> code (in a file not even built on 64-bit). It hardly warrants
> its own macro, or the trouble we go to ensuring that it can't
> be called in X86_64 code.
>
> Axe the macro and replace it with a direct cpu feature check.
>
> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Borislav Petkov <bp@suse.de>
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 2/4] x86: introduce disabled-features
2014-07-02 16:29 ` [PATCH 2/4] x86: introduce disabled-features Dave Hansen
@ 2014-07-02 19:27 ` Borislav Petkov
0 siblings, 0 replies; 9+ messages in thread
From: Borislav Petkov @ 2014-07-02 19:27 UTC (permalink / raw)
To: Dave Hansen; +Cc: x86, hpa, qiaowei.ren, linux-kernel
On Wed, Jul 02, 2014 at 09:29:28AM -0700, Dave Hansen wrote:
>
> From: Dave Hansen <dave.hansen@linux.intel.com>
>
> I believe the REQUIRED_MASK aproach was taken so that it was
> easier to consult in assembly (arch/x86/kernel/verify_cpu.S).
> DISABLED_MASK does not have the same restriction, but I
> implemented it the same way for consistency.
>
> We have a REQUIRED_MASK... which does two things:
> 1. Keeps a list of cpuid bits to check in very early boot and
> refuse to boot if those are not present.
> 2. Consulted during cpu_has() checks, which allows us to
> optimize out things at compile-time. In other words, if we
> *KNOW* we will not boot with the feature off, then we can
> safely assume that it will be present forever.
>
> But, we don't have a similar mechanism for CPU features which
> may be present but that we know we will not use. We simply
> use our existing mechanisms to repeatedly check the status of
> the bit at runtime (well, the alternatives patching helps here
> but it does not provide compile-time optimization).
>
> Adding a feature to disabled-features.h allows the bit to be
> checked via a new macro: cpu_feature_enabled(). Note that
> for features in DISABLED_MASK, checks with this macro have
> all of the benefits of an #ifdef. Before, we would have done
> this in a header:
>
> #ifdef CONFIG_X86_INTEL_MPX
> #define cpu_has_mpx cpu_has(X86_FEATURE_MPX)
> #else
> #define cpu_has_mpx 0
> #endif
>
> and this in the code:
>
> if (cpu_has_mpx)
> do_some_mpx_thing();
>
> Now, just add your feature to DISABLED_MASK and you can do this
> everywhere, and get the same benefits you would have from
> #ifdefs:
>
> if (cpu_feature_enabled(X86_FEATURE_MPX))
> do_some_mpx_thing();
>
> We need a new function and *not* a modification to cpu_has()
> because there are cases where we actually need to check the CPU
> itself, despite what features the kernel supports. The best
> example of this is a hypervisor which has no control over what
> features its guests are using and where the guest does not depend
> on the host for support.
>
> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Borislav Petkov <bp@suse.de>
...
> +/*
> + * Make sure to add features to the correct mask
Hehe, good. :-)
> + */
> +#define DISABLED_MASK0 0
> +#define DISABLED_MASK1 0
> +#define DISABLED_MASK2 0
> +#define DISABLED_MASK3 0
> +#define DISABLED_MASK4 0
> +#define DISABLED_MASK5 0
> +#define DISABLED_MASK6 0
> +#define DISABLED_MASK7 0
> +#define DISABLED_MASK8 0
> +#define DISABLED_MASK9 0
> +
> +#endif /* _ASM_X86_DISABLED_FEATURES_H */
> _
>
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 3/4] x86: add more disabled features
2014-07-02 16:29 ` [PATCH 3/4] x86: add more disabled features Dave Hansen
@ 2014-07-02 19:34 ` Borislav Petkov
0 siblings, 0 replies; 9+ messages in thread
From: Borislav Petkov @ 2014-07-02 19:34 UTC (permalink / raw)
To: Dave Hansen; +Cc: x86, hpa, qiaowei.ren, linux-kernel
On Wed, Jul 02, 2014 at 09:29:29AM -0700, Dave Hansen wrote:
>
> From: Dave Hansen <dave.hansen@linux.intel.com>
>
> There are a few other features than MPX that we can make
> assumptions about at compile-time based on compile options.
> Add them to disabled-features.h and check them with
> cpu_feature_enabled().
>
> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Borislav Petkov <bp@suse.de>
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 4/4] x86: add MPX to disaabled features
2014-07-02 16:29 ` [PATCH 4/4] x86: add MPX to disaabled features Dave Hansen
@ 2014-07-02 19:43 ` Borislav Petkov
0 siblings, 0 replies; 9+ messages in thread
From: Borislav Petkov @ 2014-07-02 19:43 UTC (permalink / raw)
To: Dave Hansen; +Cc: x86, hpa, qiaowei.ren, linux-kernel
On Wed, Jul 02, 2014 at 09:29:30AM -0700, Dave Hansen wrote:
>
> From: Dave Hansen <dave.hansen@linux.intel.com>
>
> This will go in to the MPX patch set, but I'm posting it here so
> folk can see how it will work since it was the original
> motivation for this work.
>
> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Borislav Petkov <bp@suse.de>
--
Regards/Gruss,
Boris.
Sent from a fat crate under my desk. Formatting is fine.
--
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2014-07-02 19:43 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-07-02 16:29 [PATCH 0/4] [v3] x86 disabled features macros Dave Hansen
2014-07-02 16:29 ` [PATCH 1/4] x86: axe the lightly-used cpu_has_pae Dave Hansen
2014-07-02 19:19 ` Borislav Petkov
2014-07-02 16:29 ` [PATCH 2/4] x86: introduce disabled-features Dave Hansen
2014-07-02 19:27 ` Borislav Petkov
2014-07-02 16:29 ` [PATCH 3/4] x86: add more disabled features Dave Hansen
2014-07-02 19:34 ` Borislav Petkov
2014-07-02 16:29 ` [PATCH 4/4] x86: add MPX to disaabled features Dave Hansen
2014-07-02 19:43 ` Borislav Petkov
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox