[GIT pull] x86 updates for 3.11

[GIT pull] x86 updates for 3.11

[Thomas Gleixner]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 4881 bytes --]

Linus,

please pull the latest x86-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86-urgent-for-linus

   * Guarantee IDT page alignment
   * Revert an UEFI patch which caused a regression
   * Add a missing check for EFI runtime services
   * Fix for an x86 erratum where read access to nonexisting MSRs does
     not result in a fault.

Thanks,

	tglx

------------------>
H. Peter Anvin (1):
      x86, suspend: Handle CPUs which fail to #GP on RDMSR

Kees Cook (1):
      x86: Make sure IDT is page aligned

Matt Fleming (2):
      efivars: check for EFI_RUNTIME_SERVICES
      Revert "UEFI: Don't pass boot services regions to SetVirtualAddressMap()"

Xiong Zhou (1):
      x86/platform/ce4100: Add header file for reboot type


 arch/x86/kernel/acpi/sleep.c      |   18 ++++++++++++++++--
 arch/x86/kernel/head_64.S         |    4 ----
 arch/x86/kernel/traps.c           |    7 ++-----
 arch/x86/platform/ce4100/ce4100.c |    1 +
 arch/x86/platform/efi/efi.c       |    7 -------
 drivers/firmware/efi/efivars.c    |    3 +++
 6 files changed, 22 insertions(+), 18 deletions(-)

diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index 2a34aaf..3312010 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -48,9 +48,20 @@ int x86_acpi_suspend_lowlevel(void)
 #ifndef CONFIG_64BIT
 	native_store_gdt((struct desc_ptr *)&header->pmode_gdt);
 
+	/*
+	 * We have to check that we can write back the value, and not
+	 * just read it.  At least on 90 nm Pentium M (Family 6, Model
+	 * 13), reading an invalid MSR is not guaranteed to trap, see
+	 * Erratum X4 in "Intel Pentium M Processor on 90 nm Process
+	 * with 2-MB L2 Cache and Intel® Processor A100 and A110 on 90
+	 * nm process with 512-KB L2 Cache Specification Update".
+	 */
 	if (!rdmsr_safe(MSR_EFER,
 			&header->pmode_efer_low,
-			&header->pmode_efer_high))
+			&header->pmode_efer_high) &&
+	    !wrmsr_safe(MSR_EFER,
+			header->pmode_efer_low,
+			header->pmode_efer_high))
 		header->pmode_behavior |= (1 << WAKEUP_BEHAVIOR_RESTORE_EFER);
 #endif /* !CONFIG_64BIT */
 
@@ -61,7 +72,10 @@ int x86_acpi_suspend_lowlevel(void)
 	}
 	if (!rdmsr_safe(MSR_IA32_MISC_ENABLE,
 			&header->pmode_misc_en_low,
-			&header->pmode_misc_en_high))
+			&header->pmode_misc_en_high) &&
+	    !wrmsr_safe(MSR_IA32_MISC_ENABLE,
+			header->pmode_misc_en_low,
+			header->pmode_misc_en_high))
 		header->pmode_behavior |=
 			(1 << WAKEUP_BEHAVIOR_RESTORE_MISC_ENABLE);
 	header->realmode_flags = acpi_realmode_flags;
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 5e4d8a8..317b8cc 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -514,10 +514,6 @@ ENTRY(phys_base)
 	
 	.section .bss, "aw", @nobits
 	.align L1_CACHE_BYTES
-ENTRY(idt_table)
-	.skip IDT_ENTRIES * 16
-
-	.align L1_CACHE_BYTES
 ENTRY(debug_idt_table)
 	.skip IDT_ENTRIES * 16
 
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index b0865e8..0952614 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -68,13 +68,10 @@
 #include <asm/setup.h>
 
 asmlinkage int system_call(void);
+#endif
 
-/*
- * The IDT has to be page-aligned to simplify the Pentium
- * F0 0F bug workaround.
- */
+/* The IDT has to be page-aligned to keep it aligned with its fixmap. */
 gate_desc idt_table[NR_VECTORS] __page_aligned_data = { { { { 0, 0 } } }, };
-#endif
 
 DECLARE_BITMAP(used_vectors, NR_VECTORS);
 EXPORT_SYMBOL_GPL(used_vectors);
diff --git a/arch/x86/platform/ce4100/ce4100.c b/arch/x86/platform/ce4100/ce4100.c
index f8ab494..9962015 100644
--- a/arch/x86/platform/ce4100/ce4100.c
+++ b/arch/x86/platform/ce4100/ce4100.c
@@ -14,6 +14,7 @@
 #include <linux/module.h>
 #include <linux/serial_reg.h>
 #include <linux/serial_8250.h>
+#include <linux/reboot.h>
 
 #include <asm/ce4100.h>
 #include <asm/prom.h>
diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
index c8d5577..90f6ed1 100644
--- a/arch/x86/platform/efi/efi.c
+++ b/arch/x86/platform/efi/efi.c
@@ -931,13 +931,6 @@ void __init efi_enter_virtual_mode(void)
 			va = efi_ioremap(md->phys_addr, size,
 					 md->type, md->attribute);
 
-		if (!(md->attribute & EFI_MEMORY_RUNTIME)) {
-			if (!va)
-				pr_err("ioremap of 0x%llX failed!\n",
-				       (unsigned long long)md->phys_addr);
-			continue;
-		}
-
 		md->virt_addr = (u64) (unsigned long) va;
 
 		if (!va) {
diff --git a/drivers/firmware/efi/efivars.c b/drivers/firmware/efi/efivars.c
index 8bd1bb6..8a7432a 100644
--- a/drivers/firmware/efi/efivars.c
+++ b/drivers/firmware/efi/efivars.c
@@ -583,6 +583,9 @@ int efivars_sysfs_init(void)
 	struct kobject *parent_kobj = efivars_kobject();
 	int error = 0;
 
+	if (!efi_enabled(EFI_RUNTIME_SERVICES))
+		return -ENODEV;
+
 	/* No efivars has been registered yet */
 	if (!parent_kobj)
 		return 0;

Re: [GIT pull] x86 updates for 3.11

[Linus Torvalds]

On Sat, Jul 13, 2013 at 4:21 AM, Thomas Gleixner <tglx@linutronix.de> wrote:
>
>    * Guarantee IDT page alignment

What the F*CK, guys?

This piece-of-shit commit is marked for stable, but you clearly never
even test-compiled it, did you?

Because on x86-64 (the which is the only place where the patch
matters), I don't see how you could have avoided this honking huge
warning otherwise:

  arch/x86/kernel/traps.c:74:1: warning: braces around scalar
initializer [enabled by default]
   gate_desc idt_table[NR_VECTORS] __page_aligned_data = { { { { 0, 0 } } }, };
   ^
  arch/x86/kernel/traps.c:74:1: warning: (near initialization for
‘idt_table[0].offset_low’) [enabled by default]
  arch/x86/kernel/traps.c:74:1: warning: braces around scalar
initializer [enabled by default]
  arch/x86/kernel/traps.c:74:1: warning: (near initialization for
‘idt_table[0].offset_low’) [enabled by default]
  arch/x86/kernel/traps.c:74:1: warning: excess elements in scalar
initializer [enabled by default]
  arch/x86/kernel/traps.c:74:1: warning: (near initialization for
‘idt_table[0].offset_low’) [enabled by default]

and I don't think this is compiler-specific, because that code is
crap. The declaration for gate_desc is very very different for 32-bit
and 64-bit x86 for whatever braindamaged reasons.

Seriously, WTF? I made the mistake of doing multiple merges
back-to-back with the intention of not doing a full allmodconfig build
in between them, and now I have to undo them all because this pull
request was full of unbelievable shit.

And why the hell was this marked for stable even *IF* it hadn't been
complete and utter tripe? It even has a comment in the commit message
about how this probably doesn't matter. So it's doubly crap: it's
*wrong*, and it didn't actually fix anything to begin with.

There aren't enough swear-words in the English language, so now I'll
have to call you perkeleen vittupää just to express my disgust and
frustration with this crap.

            Linus

Re: [GIT pull] x86 updates for 3.11

[H. Peter Anvin]

Fail on me.  I got rushed and sloppy.  I really need to automate looking
for warnings pre-commit and not rely on Fengguang's robot.

	-hpa

Re: [GIT pull] x86 updates for 3.11

[Thomas Gleixner]

On Sat, 13 Jul 2013, H. Peter Anvin wrote:

> Fail on me.  I got rushed and sloppy.  I really need to automate looking
> for warnings pre-commit and not rely on Fengguang's robot.

/me too. I took it for granted that this was tested by Ingos machinery
w/o noticing that it went in just when Ingo left for a long
weekend. So in my own rush to get out for the weekend I left out my
usual smoke tests on that branch. Sorry about that.

Thanks,

	tglx

Re: [GIT pull] x86 updates for 3.11

[Kees Cook]

On Mon, Jul 15, 2013 at 1:45 PM, Thomas Gleixner <tglx@linutronix.de> wrote:
> On Sat, 13 Jul 2013, H. Peter Anvin wrote:
>
>> Fail on me.  I got rushed and sloppy.  I really need to automate looking
>> for warnings pre-commit and not rely on Fengguang's robot.
>
> /me too. I took it for granted that this was tested by Ingos machinery
> w/o noticing that it went in just when Ingo left for a long
> weekend. So in my own rush to get out for the weekend I left out my
> usual smoke tests on that branch. Sorry about that.

I apologize; I was rushed as well. I've changed my build and test
procedures to more loudly present warnings.

-Kees

--
Kees Cook
Chrome OS Security