Fail gracefuly when attempting to load 64-bit kFreeBSD on IA32 CPU
PATCH 1/1 Fail gracefuly when attempting to load 64-bit kFreeBSD on IA32 CPU
Signature · 6 lines
-- Robert Millan The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and how) you may access your data; but nobody's threatening your freedom: we still allow you to remove your data and not access it at all."
2009-10-07 Robert Millan <[email protected]>
Fail gracefuly when attempting to load 64-bit kFreeBSD on IA32 CPU.
* include/grub/i386/cpuid.h: New file.
* commands/i386/cpuid.c: Include <grub/i386/cpuid.h>'. (has_longmode): Rename to ... (grub_cpuid_has_longmode): ... this. Update all users. Remove unsigned' attribute.
* loader/i386/bsd.c: Include `<grub/i386/cpuid.h>'. (grub_bsd_load_elf): Fail if load of 64-bit kernel was requested on a CPU that doesn't implement AMD64 instruction set.
grub/i386/cpuid.h +19 -0
@@ -0,0 +1,19 @@ +/* +* GRUB -- GRand Unified Bootloader +* Copyright (C) 2009 Free Software Foundation, Inc. +* +* GRUB is free software: you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation, either version 3 of the License, or +* (at your option) any later version. +* +* GRUB is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with GRUB. If not, see <http://www.gnu.org/licenses/>. +*/ + +extern unsigned char grub_cpuid_has_longmode;
i386/cpuid.c +6 -5
@@ -1,7 +1,7 @@/* cpuid.c - test for CPU features */ /* * GRUB -- GRand Unified Bootloader -* Copyright (C) 2006, 2007 Free Software Foundation, Inc. +* Copyright (C) 2006, 2007, 2009 Free Software Foundation, Inc. * Based on gcc/gcc/config/i386/driver-i386.c * * GRUB is free software: you can redistribute it and/or modify
@@ -24,6 +24,7 @@#include <grub/env.h> #include <grub/command.h> #include <grub/extcmd.h> +#include <grub/i386/cpuid.h> #define cpuid (num,a,b,c,d) \ asm volatile ("xchgl %%ebx, %1; cpuid; xchgl %%ebx, %1" \
@@ -38,14 +39,14 @@ static const struct grub_arg_option options[] =#define bit_LM (1 << 29) -static unsigned char has_longmode =0 ; +unsigned char grub_cpuid_has_longmode =0 ;static grub_err_t grub_cmd_cpuid (grub_extcmd_t cmd __attribute__ ((unused)),int argc __attribute__ ((unused)),char **args __attribute__ ((unused))) { -return has_longmode ?GRUB_ERR_NONE +return grub_cpuid_has_longmode ?GRUB_ERR_NONE :grub_error (GRUB_ERR_TEST_FAILURE ,"false" ); }
@@ -55,7 +56,7 @@ GRUB_MOD_INIT(cpuid) {#ifdef __x86_64__/* grub-emu */ - has_longmode =1 ; + grub_cpuid_has_longmode =1 ;#else unsigned int eax, ebx, ecx, edx;unsigned int max_level;
@@ -82,7 +83,7 @@ GRUB_MOD_INIT(cpuid) goto done;cpuid (0x80000001 , eax, ebx, ecx, edx); - has_longmode = !!(edx & bit_LM); + grub_cpuid_has_longmode = !!(edx & bit_LM); done:#endif
i386/bsd.c +4 -0
@@ -19,6 +19,7 @@#include <grub/loader.h> #include <grub/cpu/loader.h> #include <grub/cpu/bsd.h> +#include <grub/i386/cpuid.h> #include <grub/machine/init.h> #include <grub/machine/memory.h> #include <grub/memory.h>
@@ -871,6 +872,9 @@ grub_bsd_load_elf (grub_elf_t elf) { is_64bit =1 ; +if (! grub_cpuid_has_longmode) +return grub_error (GRUB_ERR_BAD_OS ,"Your CPU does not implement AMD64 architecture." ); +/* FreeBSD has 64-bit entry point. */ if (kernel_type ==KERNEL_TYPE_FREEBSD ) {
Am Mittwoch, den 07.10.2009, 23:13 +0200 schrieb Robert Millan:
Quoted 2 lines
(
grub_cpuid_has_longmode): ... this. Update all users. Remove `unsigned' attribute.
I think that should be static' not unsigned':
-static unsigned char has_longmode = 0;
+unsigned char grub_cpuid_has_longmode = 0;
But don't we need to add now an `insmod cpuid' in grub-mkconfig so that this actually works / is useful?
Or does the cpuid command get loaded automatically somewhere already?
Signature · 2 lines
Felix Zielcke Proud Debian Maintainer and GNU GRUB developer
Quoted 12 lines · Felix Zielcke
Am Mittwoch, den 07.10.2009, 23:13 +0200 schrieb Robert Millan:
Quoted 3 lines
(
grub_cpuid_has_longmode): ... this. Update all users. Remove `unsigned' attribute.I think that should be
static' notunsigned':i386/cpuid.c +1
-1-static unsigned char has_longmode =0 ; +unsigned char grub_cpuid_has_longmode =0 ;
Quoted 5 lines
But don't we need to add now an `insmod cpuid' in grub-mkconfig so that this actually works / is useful?
Or does the cpuid command get loaded automatically somewhere already?
It gets loaded because of symbol dependency of bsd.mod on
grub_cpuid_has_longmode
Signature · 3 lines
Regards Vladimir 'phcoder' Serbinenko Personal git repository: http://repo.or.cz/w/grub2/phcoder.git
i386/cpuid.c static const struct grub_arg_option options[] = · line 42
Quoted 9 lines · Felix Zielcke
Am Mittwoch, den 07.10.2009, 23:13 +0200 schrieb Robert Millan:
Quoted 2 lines
(
grub_cpuid_has_longmode): ... this. Update all users. Remove `unsigned' attribute.I think that should be
static' notunsigned':i386/cpuid.c +1
-1-static unsigned char has_longmode =0 ; +unsigned char grub_cpuid_has_longmode =0 ;
Oh, right.
Signature · 5 lines
Robert Millan The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and how) you may access your data; but nobody's threatening your freedom: we still allow you to remove your data and not access it at all."
grub/i386/cpuid.h
Quoted 4 lines
Index:
include/grub/i386/cpuid.h
===================================================================grub/i386/cpuid.h +0
-0
This file doesn't follow convention of having an ifdef around the file
Quoted 1 line
+extern unsigned char
grub_cpuid_has_longmode;
I don't like it being declared as a variable: user may inadvertently assign a value to it. I would be more confortable with a function
Other than that patch looks good
Signature · 3 lines
Regards Vladimir 'phcoder' Serbinenko Personal git repository: http://repo.or.cz/w/grub2/phcoder.git
Hi,
I didn't notice this mail untill a few minutes ago (was still in fetch queue), so my first patch was already checked in.
Quoted 6 lines · Vladimir 'phcoder' Serbinenko
grub/i386/cpuid.h
Quoted 4 lines
Index:
include/grub/i386/cpuid.h
===================================================================grub/i386/cpuid.h +0
-0This file doesn't follow convention of having an ifdef around the file
Thanks, I've added it now.
Quoted 2 lines
Quoted 1 line
+extern unsigned char
grub_cpuid_has_longmode;I don't like it being declared as a variable: user may inadvertently assign a value to it. I would be more confortable with a function
I tend to prefer a (const) function too, but this required some restructuring in cpuid.c so for 1.97 I opted for keeping the changes minimal.
Signature · 5 lines
Robert Millan The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and how) you may access your data; but nobody's threatening your freedom: we still allow you to remove your data and not access it at all."
Quoted 18 lines · Robert Millan
Hi,
I didn't notice this mail untill a few minutes ago (was still in fetch queue), so my first patch was already checked in.
Quoted 7 lines · Vladimir 'phcoder' Serbinenko
grub/i386/cpuid.h
Quoted 5 lines
Index:
include/grub/i386/cpuid.h
===================================================================grub/i386/cpuid.h +0
-0This file doesn't follow convention of having an ifdef around the file
Thanks, I've added it now.
Quoted 9 lines
Quoted 4 lines
Quoted 2 lines
+extern unsigned char
grub_cpuid_has_longmode;I don't like it being declared as a variable: user may inadvertently assign a value to it. I would be more confortable with a function
I tend to prefer a (const) function too, but this required some restructuring in
cpuid.cso for 1.97 I opted for keeping the changes minimal.
Why would these changes be intrusive? It's just a function returning a static value or did I miss something?
Signature · 3 lines
Regards Vladimir 'phcoder' Serbinenko Personal git repository: http://repo.or.cz/w/grub2/phcoder.git