linux-kernel - Re: [PATCH] scripts/kallsyms: skip ppc compiler stub *.long_branch.* / *.plt

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 22 Sep 2020 13:43:23 -0700
From:   Guenter Roeck <linux@...ck-us.net>
To:     Masahiro Yamada <masahiroy@...nel.org>
Cc:     linux-kbuild@...r.kernel.org,
        Geert Uytterhoeven <geert@...ux-m68k.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] scripts/kallsyms: skip ppc compiler stub *.long_branch.*
 / *.plt_branch.*

On Wed, Sep 23, 2020 at 02:48:56AM +0900, Masahiro Yamada wrote:
> PowerPC allmodconfig often fails to build as follows:
> 
>     LD      .tmp_vmlinux.kallsyms1
>     KSYM    .tmp_vmlinux.kallsyms1.o
>     LD      .tmp_vmlinux.kallsyms2
>     KSYM    .tmp_vmlinux.kallsyms2.o
>     LD      .tmp_vmlinux.kallsyms3
>     KSYM    .tmp_vmlinux.kallsyms3.o
>     LD      vmlinux
>     SORTTAB vmlinux
>     SYSMAP  System.map
>   Inconsistent kallsyms data
>   Try make KALLSYMS_EXTRA_PASS=1 as a workaround
>   make[2]: *** [../Makefile:1162: vmlinux] Error 1
> 
> Setting KALLSYMS_EXTRA_PASS=1 does not help.
> 
> This is caused by the compiler inserting stubs such as *.long_branch.*
> and *.plt_branch.*
> 
>   $ powerpc-linux-nm -n .tmp_vmlinux.kallsyms2
>    [ snip ]
>   c00000000210c000 T __init_begin
>   c00000000210c000 T _sinittext
>   c00000000210c010 t 00000075.plt_branch.da9:19
>   c00000000210c020 t 00000075.plt_branch.1677:5
>   c00000000210c030 t 00000075.long_branch.memmove
>   c00000000210c034 t 00000075.plt_branch.9e0:5
>   c00000000210c044 t 00000075.plt_branch.free_initrd_mem
>     ...
> 
> Actually, the problem mentioned in scripts/link-vmlinux.sh comments;
> "In theory it's possible this results in even more stubs, but unlikely"
> is happening here, and ends up with another kallsyms step required.
> 
> scripts/kallsyms.c already ignores various compiler stubs. Let's do
> similar to make kallsysms for PowerPC always succeed in 2 steps.
> 
> Reported-by: Guenter Roeck <linux@...ck-us.net>
> Signed-off-by: Masahiro Yamada <masahiroy@...nel.org>

Tested-by: Guenter Roeck <linux@...ck-us.net>

Thanks,
Guenter

> ---
> 
>  scripts/kallsyms.c | 16 +++++++++++++++-
>  1 file changed, 15 insertions(+), 1 deletion(-)
> 
> diff --git a/scripts/kallsyms.c b/scripts/kallsyms.c
> index 0096cd965332..306b9b38150f 100644
> --- a/scripts/kallsyms.c
> +++ b/scripts/kallsyms.c
> @@ -120,17 +120,25 @@ static bool is_ignored_symbol(const char *name, char type)
>  		NULL
>  	};
>  
> +	static const char * const ignored_matches[] = {
> +		".long_branch.",	/* ppc stub */
> +		".plt_branch.",		/* ppc stub */
> +		NULL
> +	};
> +
>  	const char * const *p;
>  
> -	/* Exclude symbols which vary between passes. */
> +	/* ignore symbol names that exactly match to a particular string. */
>  	for (p = ignored_symbols; *p; p++)
>  		if (!strcmp(name, *p))
>  			return true;
>  
> +	/* ignore symbol names that start with a particular string. */
>  	for (p = ignored_prefixes; *p; p++)
>  		if (!strncmp(name, *p, strlen(*p)))
>  			return true;
>  
> +	/* ignore symbol names that end with a particular string. */
>  	for (p = ignored_suffixes; *p; p++) {
>  		int l = strlen(name) - strlen(*p);
>  
> @@ -138,6 +146,12 @@ static bool is_ignored_symbol(const char *name, char type)
>  			return true;
>  	}
>  
> +	/* ignore symbol names that contain a particular string. */
> +	for (p = ignored_matches; *p; p++) {
> +		if (strstr(name, *p))
> +			return true;
> +	}
> +
>  	if (type == 'U' || type == 'u')
>  		return true;
>  	/* exclude debugging symbols */
> -- 
> 2.25.1
>