linux-kernel - [PATCH] x86: make FPU state alignment enforcement and checking consistent

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Wed, 16 Oct 2013 12:41:46 +0100
From:	"Jan Beulich" <JBeulich@...e.com>
To:	<mingo@...e.hu>, <tglx@...utronix.de>, <hpa@...or.com>
Cc:	<linux-kernel@...r.kernel.org>
Subject: [PATCH] x86: make FPU state alignment enforcement and checking
 consistent

There's no point in validating only 16-byte alignment when XSAVE is in
use - 64-byte alignment is required here.

There's also no point in enforcing 64-btye alignment when XSAVE is not
in use - 16-byte alignment suffices here.

Also restrict the visibility of task_xstate_cachep by moving it to the
correct internal header (instead of it being declared in one included
by [almost] everyone).

Signed-off-by: Jan Beulich <jbeulich@...e.com>
---
 arch/x86/include/asm/fpu-internal.h |    5 ++++-
 arch/x86/include/asm/processor.h    |    1 -
 arch/x86/kernel/process.c           |    6 ++++--
 3 files changed, 8 insertions(+), 4 deletions(-)

--- 3.12-rc5/arch/x86/include/asm/fpu-internal.h
+++ 3.12-rc5-x86-xstate-alignment/arch/x86/include/asm/fpu-internal.h
@@ -38,6 +38,7 @@ int ia32_setup_frame(int sig, struct ksi
 # define ia32_setup_rt_frame	__setup_rt_frame
 #endif
 
+extern struct kmem_cache *task_xstate_cachep;
 extern unsigned int mxcsr_feature_mask;
 extern void fpu_init(void);
 extern void eager_fpu_init(void);
@@ -573,7 +574,9 @@ static inline int fpu_alloc(struct fpu *
 	fpu->state = kmem_cache_alloc(task_xstate_cachep, GFP_KERNEL);
 	if (!fpu->state)
 		return -ENOMEM;
-	WARN_ON((unsigned long)fpu->state & 15);
+	WARN_ON((unsigned long)fpu->state & (use_xsave() ?
+					     __alignof(fpu->state->xsave) - 1 :
+					     __alignof(fpu->state->fxsave) - 1));
 	return 0;
 }
 
--- 3.12-rc5/arch/x86/include/asm/processor.h
+++ 3.12-rc5-x86-xstate-alignment/arch/x86/include/asm/processor.h
@@ -436,7 +436,6 @@ DECLARE_PER_CPU_ALIGNED(struct stack_can
 
 extern unsigned int xstate_size;
 extern void free_thread_xstate(struct task_struct *);
-extern struct kmem_cache *task_xstate_cachep;
 
 struct perf_event;
 
--- 3.12-rc5/arch/x86/kernel/process.c
+++ 3.12-rc5-x86-xstate-alignment/arch/x86/kernel/process.c
@@ -87,11 +87,13 @@ void arch_release_task_struct(struct tas
 	free_thread_xstate(tsk);
 }
 
-void arch_task_cache_init(void)
+void __init arch_task_cache_init(void)
 {
         task_xstate_cachep =
         	kmem_cache_create("task_xstate", xstate_size,
-				  __alignof__(union thread_xstate),
+				  boot_cpu_has(X86_FEATURE_XSAVE) ?
+				  __alignof(struct xsave_struct) :
+				  __alignof(struct i387_fxsave_struct),
 				  SLAB_PANIC | SLAB_NOTRACK, NULL);
 }
 



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/