public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: libc-alpha@sourceware.org, Szabolcs Nagy <szabolcs.nagy@arm.com>
Subject: Re: [PATCH v4 08/10] csu: Move static pie self relocation later [BZ #27072]
Date: Tue, 19 Jan 2021 11:07:05 -0300	[thread overview]
Message-ID: <1ba70d1b-08f8-6a5d-ecf4-45200744c9d8@linaro.org> (raw)
In-Reply-To: <4224b7c0428492696fe6d6c01739adcf69fc677d.1610986541.git.szabolcs.nagy@arm.com>



On 18/01/2021 13:25, Szabolcs Nagy via Libc-alpha wrote:
> IFUNC resolvers may depend on tunables and cpu feature setup so
> move static pie self relocation after those.
> 
> It is hard to guarantee that the ealy startup code does not rely
> on relocations so this is a bit fragile. It would be more robust
> to handle RELATIVE relocs early and only IRELATIVE relocs later,
> but the current relocation processing code cannot do that.
> 
> The early startup code before relocation processing includes
> 
>   _dl_aux_init (auxvec);
>   __libc_init_secure ();
>   __tunables_init (__environ);
>   ARCH_INIT_CPU_FEATURES ();
> 
> These are simple enough that RELATIVE relocs can be avoided.
> 
> __ehdr_start may require RELATIVE relocation so it was moved
> later, fortunately ehdr and phdr are not used in the early code.
> 
> Fixes bug 27072.

LGTM, thanks.

Reviewed-by: Adhemerval Zanella  <adhemerval.zanella@linaro.org>

> ---
>  csu/libc-start.c | 44 +++++++++++++++++++++++++-------------------
>  1 file changed, 25 insertions(+), 19 deletions(-)
> 
> diff --git a/csu/libc-start.c b/csu/libc-start.c
> index 1e90dcb0a7..c2b59431a3 100644
> --- a/csu/libc-start.c
> +++ b/csu/libc-start.c
> @@ -146,8 +146,6 @@ LIBC_START_MAIN (int (*main) (int, char **, char ** MAIN_AUXVEC_DECL),
>    int result;
>  
>  #ifndef SHARED
> -  _dl_relocate_static_pie ();
> -
>    char **ev = &argv[argc + 1];
>  
>    __environ = ev;
Ok.


> @@ -169,24 +167,7 @@ LIBC_START_MAIN (int (*main) (int, char **, char ** MAIN_AUXVEC_DECL),
>    }
>  #  endif
>    _dl_aux_init (auxvec);
> -  if (GL(dl_phdr) == NULL)
>  # endif
> -    {
> -      /* Starting from binutils-2.23, the linker will define the
> -         magic symbol __ehdr_start to point to our own ELF header
> -         if it is visible in a segment that also includes the phdrs.
> -         So we can set up _dl_phdr and _dl_phnum even without any
> -         information from auxv.  */
> -
> -      extern const ElfW(Ehdr) __ehdr_start
> -	__attribute__ ((weak, visibility ("hidden")));
> -      if (&__ehdr_start != NULL)
> -        {
> -          assert (__ehdr_start.e_phentsize == sizeof *GL(dl_phdr));
> -          GL(dl_phdr) = (const void *) &__ehdr_start + __ehdr_start.e_phoff;
> -          GL(dl_phnum) = __ehdr_start.e_phnum;
> -        }
> -    }
>  
>    /* Initialize very early so that tunables can use it.  */
>    __libc_init_secure ();

Ok.

> @@ -195,6 +176,11 @@ LIBC_START_MAIN (int (*main) (int, char **, char ** MAIN_AUXVEC_DECL),
>  
>    ARCH_INIT_CPU_FEATURES ();
>  
> +  /* Do static pie self relocation after tunables and cpu features
> +     are setup for ifunc resolvers. Before this point relocations
> +     must be avoided.  */
> +  _dl_relocate_static_pie ();
> +
>    /* Perform IREL{,A} relocations.  */
>    ARCH_SETUP_IREL ();
>  

Ok.

> @@ -206,6 +192,26 @@ LIBC_START_MAIN (int (*main) (int, char **, char ** MAIN_AUXVEC_DECL),
>       hwcap and platform fields available in the TCB.  */
>    ARCH_APPLY_IREL ();
>  
> +# ifdef HAVE_AUX_VECTOR
> +  if (GL(dl_phdr) == NULL)
> +# endif
> +    {
> +      /* Starting from binutils-2.23, the linker will define the
> +         magic symbol __ehdr_start to point to our own ELF header
> +         if it is visible in a segment that also includes the phdrs.
> +         So we can set up _dl_phdr and _dl_phnum even without any
> +         information from auxv.  */
> +
> +      extern const ElfW(Ehdr) __ehdr_start
> +	__attribute__ ((weak, visibility ("hidden")));
> +      if (&__ehdr_start != NULL)
> +        {
> +          assert (__ehdr_start.e_phentsize == sizeof *GL(dl_phdr));
> +          GL(dl_phdr) = (const void *) &__ehdr_start + __ehdr_start.e_phoff;
> +          GL(dl_phnum) = __ehdr_start.e_phnum;
> +        }
> +    }
> +
>    /* Set up the stack checker's canary.  */
>    uintptr_t stack_chk_guard = _dl_setup_stack_chk_guard (_dl_random);
>  # ifdef THREAD_SET_STACK_GUARD
> 

Ok.

  reply	other threads:[~2021-01-19 14:07 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-18 16:22 [PATCH v4 00/10] fix ifunc with static pie " Szabolcs Nagy
2021-01-18 16:23 ` [PATCH v4 01/10] configure: Require PI_STATIC_AND_HIDDEN for static pie Szabolcs Nagy
2021-01-18 16:23 ` [PATCH v4 02/10] libmvec: Add extra-test-objs to test-extras Szabolcs Nagy
2021-01-18 20:04   ` Adhemerval Zanella
2021-01-18 16:23 ` [PATCH v4 03/10] elf: Make the tunable struct definition internal only Szabolcs Nagy
2021-01-18 16:24 ` [PATCH v4 04/10] elf: Avoid RELATIVE relocs in __tunables_init Szabolcs Nagy
2021-01-18 16:24 ` [PATCH v4 05/10] Use hidden visibility for early static PIE code Szabolcs Nagy
2021-01-18 21:49   ` Adhemerval Zanella
2021-01-18 16:24 ` [PATCH v4 06/10] elf: Avoid RELATIVE relocation for _dl_sysinfo Szabolcs Nagy
2021-01-19 13:51   ` Adhemerval Zanella
2021-01-19 14:25     ` V2 " H.J. Lu
2021-01-19 14:35       ` Adhemerval Zanella
2021-01-18 16:25 ` [PATCH v4 07/10] Use <startup.h> in __libc_init_secure Szabolcs Nagy
2021-01-19 13:56   ` Adhemerval Zanella
2021-01-18 16:25 ` [PATCH v4 08/10] csu: Move static pie self relocation later [BZ #27072] Szabolcs Nagy
2021-01-19 14:07   ` Adhemerval Zanella [this message]
2021-01-19 14:35     ` Szabolcs Nagy
2021-01-19 14:36       ` Adhemerval Zanella
2021-01-19 14:48         ` H.J. Lu
2021-01-19 15:24           ` Szabolcs Nagy
2021-01-19 15:32             ` H.J. Lu
2021-01-19 16:47               ` H.J. Lu
2021-01-19 17:03                 ` Szabolcs Nagy
2021-01-19 17:10                   ` H.J. Lu
2021-01-19 17:25                     ` Fāng-ruì Sòng
2021-01-19 17:33                       ` H.J. Lu
2021-01-19 17:38                         ` Fāng-ruì Sòng
2021-01-19 17:38                     ` Szabolcs Nagy
2021-01-19 17:42                       ` H.J. Lu
2021-01-19 17:47                         ` Szabolcs Nagy
2021-01-19 17:53                           ` H.J. Lu
2021-01-19 17:59                             ` H.J. Lu
2021-01-18 16:25 ` [PATCH v4 09/10] x86: Check ifunc resolver with CPU_FEATURE_USABLE " Szabolcs Nagy
2021-01-19 14:11   ` Adhemerval Zanella
2021-01-19 14:37     ` V2 " H.J. Lu
2021-01-21 16:33       ` H.J. Lu
2021-01-18 16:26 ` [PATCH v4 10/10] Make libc symbols hidden in static PIE Szabolcs Nagy
2021-01-18 21:37 ` [PATCH v4 00/10] fix ifunc with static pie [BZ #27072] Adhemerval Zanella
2021-01-19 18:25   ` Szabolcs Nagy
2021-01-19 19:41     ` H.J. Lu
2021-01-19 20:16       ` Adhemerval Zanella
2021-01-19 21:38         ` H.J. Lu
2021-01-20 11:29           ` Adhemerval Zanella
2021-01-20 12:38             ` Szabolcs Nagy
2021-01-20 12:49               ` H.J. Lu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1ba70d1b-08f8-6a5d-ecf4-45200744c9d8@linaro.org \
    --to=adhemerval.zanella@linaro.org \
    --cc=libc-alpha@sourceware.org \
    --cc=szabolcs.nagy@arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).