[PATCH gnupg v5] Disable CPU speculation-related misfeatures

[Jacob Bachmeyer]

On 7/6/25 15:03, Guido Trentalancia via Gnupg-devel wrote:
> [...]
> diff -pru a/common/init.c b/common/init.c
> --- a/common/init.c	2025-05-25 15:43:45.871984100 +0200
> +++ b/common/init.c	2025-07-06 18:24:18.564538076 +0200
> [...]
> @@ -132,6 +136,29 @@ writestring_via_estream (int mode, const
>   void
>   early_system_init (void)
>   {
> +#if defined(__linux__)
> +
> +/* Disable CPU speculation-related misfeatures which are in
> + * fact vulnerabilities causing data leaks: see the kernel
> + * documentation: Documentation/userspace-api/spec_ctrl.rst
> + *
> + * - Speculative Store Bypass
> + * - Indirect Branch Speculation
> + * - Flush L1D Cache on context switch out of the task
> + */
> +#ifdef PR_SPEC_STORE_BYPASS
> +  prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_STORE_BYPASS, PR_SPEC_FORCE_DISABLE, 0, 0);
> +#endif
> +
> +#ifdef PR_SPEC_INDIRECT_BRANCH
> +  prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_INDIRECT_BRANCH, PR_SPEC_FORCE_DISABLE, 0, 0);
> +#endif
> +
> +#if defined(ENABLE_L1D_CACHE_FLUSH) && defined(PR_SPEC_L1D_FLUSH)
> +  prctl(PR_SET_SPECULATION_CTRL, PR_SPEC_L1D_FLUSH, PR_SPEC_ENABLE, 0, 0);
My understanding is that this prctl(2) call raises SIGBUS unless the 
kernel was given a boot option to enable this workaround.  If a library 
is going to do this behind the main program's back, that library *needs* 
to be prepared to catch a SIGBUS at this point.
> +#endif
> +
> +#endif /* __linux__ */
>   }
>   
>   
> diff -pru a/configure.ac b/configure.ac
> --- a/configure.ac	2025-07-06 18:01:54.128546282 +0200
> +++ b/configure.ac	2025-07-06 21:56:51.219048292 +0200
> @@ -244,6 +244,16 @@ AC_ARG_ENABLE(selinux-support,
>   AC_MSG_RESULT($selinux_support)
>   
>   
> +# Fix security vulnerability CVE-2020-0550 by enabling
> +# Level 1 Data Cache flushing on context switch.
> +AC_MSG_CHECKING([whether Level 1 Data Cache is flushed on context switch])

This message is worded very badly:  the configure script is *not* 
checking whether context switches flush L1; it is checking whether the 
configuration option to *request* that context switches flush L1 has 
been given.

A better way to word this would be "[whether L1 data cache should be 
flushed on context switch]".

> +AC_ARG_ENABLE(l1d-cache-flushing,
> +              AS_HELP_STRING([--enable-l1d-cache-flushing],
> +                             [enable L1D cache flushing]),
> +              l1d_cache_flushing=$enableval, l1d_cache_flushing=no)
> +AC_MSG_RESULT($l1d_cache_flushing)
> +
> +
>   AC_MSG_CHECKING([whether to allocate extra secure memory])
>   AC_ARG_ENABLE(large-secmem,
>                 AS_HELP_STRING([--enable-large-secmem],
> @@ -1313,6 +1323,15 @@ fi
>   
>   
>   #
> +# Level 1 Data Cache flushing on context switch (CVE-2020-0550)
> +#
> +if test "$l1d_cache_flushing" = yes ; then
> +  AC_DEFINE(ENABLE_L1D_CACHE_FLUSH,1,
> +          [Define to enable Layer 1 Data Cache flushing])

Again, this *enables* nothing; it causes the program to *request* the L1 
cache be flushed on context switch.

> +fi
> +
> +
> +#
>   # Checks for header files.
>   #
>   AC_MSG_NOTICE([checking for header files])
> @@ -1322,6 +1341,13 @@ AC_CHECK_HEADERS([unistd.h langinfo.h te
>                     ucred.h sys/ucred.h sys/sysmacros.h sys/mkdev.h])
>   
>   
> +# See whether libc supports the prctl()
> +case "${host}" in
> +    *-*-linux*)
> +        AC_CHECK_HEADERS([sys/prctl.h])
> +        ;;
> +esac
> +
>   #
>   # Checks for typedefs, structures, and compiler characteristics.
>   #
>
[...]

Overall, I am still unconvinced of the appropriateness of this patch.  
As I understand, these leaks only matter in multi-tenant systems, or if 
Mallory otherwise has access to your machine, in which case you have 
bigger problems if your private key is on such a system.


-- Jacob