retpoline/module: Warn for missing retpoline in module
Subsystems (3)
PATCH v2 1/1 retpoline/module: Warn for missing retpoline in module
From: Andi Kleen <[email protected]>
There's a risk that a kernel that has full retpoline mitigations becomes vulnerable when a module gets loaded that hasn't been compiled with the right compiler or the right option.
We cannot fix it, but should at least warn the user when that happens.
When the a module hasn't been compiled with a retpoline aware compiler, print a warning and change the SPECTRE_V2 mitigation mode to show the system is vulnerable now.
For modules it is checked at compile time, however it cannot check assembler or other non compiled objects used in the module link.
v2: Change warning message
v3: Port to latest tree
v4: Remove tainting
--- arch/x86/include/asm/processor.h | 4 ++++ arch/x86/kernel/cpu/bugs.c | 12 ++++++++++++ kernel/module.c | 8 +++++++- scripts/mod/modpost.c | 9 +++++++++ 4 files changed, 32 insertions(+), 1 deletion(-)
arch/x86/include/asm/processor.h +4 -0
@@ -970,4 +970,8 @@ bool xen_set_default_idle(void);void stop_this_cpu (void *dummy);void df_debug (struct pt_regs *regs,long error_code); + +void disable_retpoline (void ); +bool retpoline_enabled (void ); +#endif /* _ASM_X86_PROCESSOR_H */
arch/x86/kernel/cpu/bugs.c +12 -0
@@ -93,6 +93,18 @@ static const char *spectre_v2_strings[] = {static enum spectre_v2_mitigation spectre_v2_enabled =SPECTRE_V2_NONE ; +/* A module has been loaded. Disable reporting that we're good. */ +void disable_retpoline (void ) +{ + spectre_v2_enabled =SPECTRE_V2_NONE ; +pr_err ("system may be vunerable to spectre\n" ); +} + +bool retpoline_enabled (void ) +{ +return spectre_v2_enabled !=SPECTRE_V2_NONE ; +} +static void __init spec2_print_if_insecure (const char *reason) {if (boot_cpu_has_bug (X86_BUG_SPECTRE_V2 ))
kernel/module.c +7 -1
@@ -3020,7 +3020,13 @@ static int check_modinfo(struct module *mod, struct load_info *info, int flags) mod->name);add_taint_module (mod,TAINT_OOT_MODULE ,LOCKDEP_STILL_OK ); } - +#ifdef RETPOLINE +if (retpoline_enabled () && !get_modinfo (info,"retpoline" )) { +pr_warn ("%s: loading module not compiled with retpoline compiler.\n" , + mod->name); +disable_retpoline (); + } +#endif if (get_modinfo (info,"staging" )) {add_taint_module (mod,TAINT_CRAP ,LOCKDEP_STILL_OK ); pr_warn("%s: module is from the staging directory, the quality "
scripts/mod/modpost.c +9 -0
@@ -2165,6 +2165,14 @@ static void add_intree_flag(struct buffer *b, int is_intree)buf_printf (b,"\nMODULE_INFO(intree, \"Y\");\n" ); } +/* Cannot check for assembler */ +static void add_retpoline (struct buffer *b) +{ +buf_printf (b,"\n#ifdef RETPOLINE\n" ); +buf_printf (b,"MODULE_INFO(retpoline, \"Y\");\n" ); +buf_printf (b,"#endif\n" ); +} +static void add_staging_flag (struct buffer *b,const char *name) {static const char *staging_dir ="drivers/staging" ;
@@ -2506,6 +2514,7 @@ int main(int argc, char **argv) err |=check_modname_len (mod);add_header (&buf, mod);add_intree_flag (&buf, !external_module); +add_retpoline (&buf);add_staging_flag (&buf, mod->name); err |=add_versions (&buf, mod);add_depends (&buf, mod, modules);
Signature · 1 line
2.14.3
arch/x86/include/asm/processor.h bool xen_set_default_idle(void); · line 975
Quoted 6 lines · Andi Kleen
From: Andi Kleen <[email protected]>
voidstop_this_cpu(void *dummy);
voiddf_debug(struct pt_regs *regs, long error_code);arch/x86/include/asm/processor.h +3
-0+ +void disable_retpoline (void ); +bool retpoline_enabled (void );
Can you please use a consistent name space? retpoline_ ... or such?
arch/x86/kernel/cpu/bugs.c static const char *spectre_v2_strings[] = { · line 99
Quoted 4 lines
arch/x86/kernel/cpu/bugs.c +4
-0+/* A module has been loaded. Disable reporting that we're good. */ +void disable_retpoline (void ) +{ + spectre_v2_enabled =SPECTRE_V2_NONE ;
I really don't like fiddling with that variable. That's just hackery. The variable reflects the actual enabled mitigation state of the kernel proper.
arch/x86/kernel/cpu/bugs.c static const char *spectre_v2_strings[] = { · line 106
Quoted 7 lines
arch/x86/kernel/cpu/bugs.c +7
-0+pr_err ("system may be vunerable to spectre\n" ); +} + +bool retpoline_enabled (void ) +{ +return spectre_v2_enabled !=SPECTRE_V2_NONE ; +}
That'll break once we get other mitigation variants.
kernel/module.c static int check_modinfo(struct module *mod, struct load_info *info, int flags) · line 3023
Quoted 5 lines
kernel/module.c +0
-1@@ -3020,7 +3020,13 @@ static int check_modinfo(struct module *mod, struct load_info *info, int flags) mod->name);add_taint_module (mod,TAINT_OOT_MODULE ,LOCKDEP_STILL_OK ); } -
These newlines are there to separate stuff for readability sake.
kernel/module.c static int check_modinfo(struct module *mod, struct load_info *info, int flags) · line 3029
Quoted 7 lines
kernel/module.c +7
-0+#ifdef RETPOLINE +if (retpoline_enabled () && !get_modinfo (info,"retpoline" )) { +pr_warn ("%s: loading module not compiled with retpoline compiler.\n" , + mod->name); +disable_retpoline (); + } +#endif
This really can be done in a cleaner way.
in linux/module.h
#ifdef RETPOLINE
extern bool retpoline_module_ok(bool has_retpoline); #else
static inline bool retpoline_module_ok(bool has_retpoline)
{
return true;
}#endif
static void check_modinfo_retpoline(mod, info)
{
if (retpoline_module_ok(get_modinfo(info, "retpoline")))
return;
pr_warn("%s: loading module not compiled with retpoline compiler.\n",
mod->name);
}That only needs one function and that one can take care of setting a variable in the spectre code which then influences the sysfs output.
And that output should not be "Vulnerable" like you force with the hack above. It actually should tell WHY it is vulnerable despite having had protection in place before the module was loaded.
Thanks,
tglx
Thanks.
I just sent a v3 that changes the VERMAGIC only, based on Greg's earlier feedback.
It has the drawbacks that it:
- refuses loading instead of warns
- doesn't stop refusing when the feature is runtime disabled
But it's much simpler, just a few lines of ifdef.
We can either go with the v3, or rework this one into a v4?
-Andi
Quoted 8 lines
I just sent a v3 that changes the VERMAGIC only, based on Greg's earlier feedback.
It has the drawbacks that it:
- refuses loading instead of warns
- doesn't stop refusing when the feature is runtime disabledBut it's much simpler, just a few lines of ifdef.
I think simple is good at this point
Quoted 5 lines · Andi Kleen
Thanks.
I just sent a v3 that changes the VERMAGIC only, based on Greg's earlier feedback.
Quoted 7 lines
It has the drawbacks that it:
- refuses loading instead of warns
- doesn't stop refusing when the feature is runtime disabledBut it's much simpler, just a few lines of ifdef.
We can either go with the v3, or rework this one into a v4?
V3 is fine. Not loading is the right thing to do :)
Thanks,
tglx