Merge branch 'x86-fpu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 FPU updates from Ingo Molnar:
 "This tree contains two main changes:

   - The big FPU code rewrite: wide reaching cleanups and reorganization
     that pulls all the FPU code together into a clean base in
     arch/x86/fpu/.

     The resulting code is leaner and faster, and much easier to
     understand.  This enables future work to further simplify the FPU
     code (such as removing lazy FPU restores).

     By its nature these changes have a substantial regression risk: FPU
     code related bugs are long lived, because races are often subtle
     and bugs mask as user-space failures that are difficult to track
     back to kernel side backs.  I'm aware of no unfixed (or even
     suspected) FPU related regression so far.

   - MPX support rework/fixes.  As this is still not a released CPU
     feature, there were some buglets in the code - should be much more
     robust now (Dave Hansen)"

* 'x86-fpu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (250 commits)
  x86/fpu: Fix double-increment in setup_xstate_features()
  x86/mpx: Allow 32-bit binaries on 64-bit kernels again
  x86/mpx: Do not count MPX VMAs as neighbors when unmapping
  x86/mpx: Rewrite the unmap code
  x86/mpx: Support 32-bit binaries on 64-bit kernels
  x86/mpx: Use 32-bit-only cmpxchg() for 32-bit apps
  x86/mpx: Introduce new 'directory entry' to 'addr' helper function
  x86/mpx: Add temporary variable to reduce masking
  x86: Make is_64bit_mm() widely available
  x86/mpx: Trace allocation of new bounds tables
  x86/mpx: Trace the attempts to find bounds tables
  x86/mpx: Trace entry to bounds exception paths
  x86/mpx: Trace #BR exceptions
  x86/mpx: Introduce a boot-time disable flag
  x86/mpx: Restrict the mmap() size check to bounds tables
  x86/mpx: Remove redundant MPX_BNDCFG_ADDR_MASK
  x86/mpx: Clean up the code by not passing a task pointer around when unnecessary
  x86/mpx: Use the new get_xsave_field_ptr()API
  x86/fpu/xstate: Wrap get_xsave_addr() to make it safer
  x86/fpu/xstate: Fix up bad get_xsave_addr() assumptions
  ...

4 files changed, 9 insertions, 9 deletions

diff --git a/drivers/char/hw_random/via-rng.c b/drivers/char/hw_random/via-rng.c
index a3bebef..0c98a9d 100644
--- a/drivers/char/hw_random/via-rng.c
+++ b/drivers/char/hw_random/via-rng.c
@@ -33,7 +33,7 @@
 #include <asm/io.h>
 #include <asm/msr.h>
 #include <asm/cpufeature.h>
-#include <asm/i387.h>
+#include <asm/fpu/api.h>
 
 
 
diff --git a/drivers/crypto/padlock-aes.c b/drivers/crypto/padlock-aes.c
index c178ed8..da2d677 100644
--- a/drivers/crypto/padlock-aes.c
+++ b/drivers/crypto/padlock-aes.c
@@ -22,7 +22,7 @@
 #include <asm/cpu_device_id.h>
 #include <asm/byteorder.h>
 #include <asm/processor.h>
-#include <asm/i387.h>
+#include <asm/fpu/api.h>
 
 /*
  * Number of data blocks actually fetched for each xcrypt insn.
diff --git a/drivers/crypto/padlock-sha.c b/drivers/crypto/padlock-sha.c
index 95f7d27..4e154c9 100644
--- a/drivers/crypto/padlock-sha.c
+++ b/drivers/crypto/padlock-sha.c
@@ -23,7 +23,7 @@
 #include <linux/kernel.h>
 #include <linux/scatterlist.h>
 #include <asm/cpu_device_id.h>
-#include <asm/i387.h>
+#include <asm/fpu/api.h>
 
 struct padlock_sha_desc {
 	struct shash_desc fallback;
diff --git a/drivers/lguest/x86/core.c b/drivers/lguest/x86/core.c
index 30f2aef..6a4cd77 100644
--- a/drivers/lguest/x86/core.c
+++ b/drivers/lguest/x86/core.c
@@ -46,7 +46,7 @@
 #include <asm/setup.h>
 #include <asm/lguest.h>
 #include <asm/uaccess.h>
-#include <asm/i387.h>
+#include <asm/fpu/internal.h>
 #include <asm/tlbflush.h>
 #include "../lg.h"
 
@@ -251,7 +251,7 @@ void lguest_arch_run_guest(struct lg_cpu *cpu)
 	 * we set it now, so we can trap and pass that trap to the Guest if it
 	 * uses the FPU.
 	 */
-	if (cpu->ts && user_has_fpu())
+	if (cpu->ts && fpregs_active())
 		stts();
 
 	/*
@@ -283,7 +283,7 @@ void lguest_arch_run_guest(struct lg_cpu *cpu)
 		wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
 
 	/* Clear the host TS bit if it was set above. */
-	if (cpu->ts && user_has_fpu())
+	if (cpu->ts && fpregs_active())
 		clts();
 
 	/*
@@ -297,12 +297,12 @@ void lguest_arch_run_guest(struct lg_cpu *cpu)
 	/*
 	 * Similarly, if we took a trap because the Guest used the FPU,
 	 * we have to restore the FPU it expects to see.
-	 * math_state_restore() may sleep and we may even move off to
+	 * fpu__restore() may sleep and we may even move off to
 	 * a different CPU. So all the critical stuff should be done
 	 * before this.
 	 */
-	else if (cpu->regs->trapnum == 7 && !user_has_fpu())
-		math_state_restore();
+	else if (cpu->regs->trapnum == 7 && !fpregs_active())
+		fpu__restore(&current->thread.fpu);
 }
 
 /*H:130