x86/fpu: Do not BUG_ON() in early FPU code
authorDave Hansen <dave.hansen@linux.intel.com>
Wed, 20 Jul 2016 19:45:51 +0000 (12:45 -0700)
committerIngo Molnar <mingo@kernel.org>
Thu, 21 Jul 2016 16:18:45 +0000 (18:18 +0200)
I don't think it is really possible to have a system where CPUID
enumerates support for XSAVE but that it does not have FP/SSE
(they are "legacy" features and always present).

But, I did manage to hit this case in qemu when I enabled its
somewhat shaky XSAVE support.  The bummer is that the FPU is set
up before we parse the command-line or have *any* console support
including earlyprintk.  That turned what should have been an easy
thing to debug in to a bit more of an odyssey.

So a BUG() here is worthless.  All it does it guarantee that
if/when we hit this case we have an empty console.  So, remove
the BUG() and try to limp along by disabling XSAVE and trying to
continue.  Add a comment on why we are doing this, and also add
a common "out_disable" path for leaving fpu__init_system_xstate().

Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
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: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Casasnovas <quentin.casasnovas@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160720194551.63BB2B58@viggo.jf.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
arch/x86/kernel/fpu/xstate.c

index 3169bca..680049a 100644 (file)
@@ -714,8 +714,13 @@ void __init fpu__init_system_xstate(void)
        xfeatures_mask = eax + ((u64)edx << 32);
 
        if ((xfeatures_mask & XFEATURE_MASK_FPSSE) != XFEATURE_MASK_FPSSE) {
+               /*
+                * This indicates that something really unexpected happened
+                * with the enumeration.  Disable XSAVE and try to continue
+                * booting without it.  This is too early to BUG().
+                */
                pr_err("x86/fpu: FP/SSE not present amongst the CPU's xstate features: 0x%llx.\n", xfeatures_mask);
-               BUG();
+               goto out_disable;
        }
 
        xfeatures_mask &= fpu__get_supported_xfeatures_mask();
@@ -723,11 +728,8 @@ void __init fpu__init_system_xstate(void)
        /* Enable xstate instructions to be able to continue with initialization: */
        fpu__init_cpu_xstate();
        err = init_xstate_size();
-       if (err) {
-               /* something went wrong, boot without any XSAVE support */
-               fpu__init_disable_system_xstate();
-               return;
-       }
+       if (err)
+               goto out_disable;
 
        /*
         * Update info used for ptrace frames; use standard-format size and no
@@ -744,6 +746,11 @@ void __init fpu__init_system_xstate(void)
                xfeatures_mask,
                fpu_kernel_xstate_size,
                boot_cpu_has(X86_FEATURE_XSAVES) ? "compacted" : "standard");
+       return;
+
+out_disable:
+       /* something went wrong, try to boot without any XSAVE support */
+       fpu__init_disable_system_xstate();
 }
 
 /*