public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Lukasz Majewski <lukma@denx.de>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>,
	Fangrui Song <maskray@google.com>,
	Florian Weimer <fweimer@redhat.com>
Cc: Joseph Myers <joseph@codesourcery.com>,
	Carlos O'Donell <carlos@redhat.com>,
	Szabolcs Nagy <szabolcs.nagy@arm.com>,
	Andreas Schwab <schwab@linux-m68k.org>,
	libc-alpha <libc-alpha@sourceware.org>,
	Lukasz Majewski <lukma@denx.de>
Subject: [PATCH v2] dl: Use "adr" assembler command to get proper load address on ARM
Date: Fri, 15 Oct 2021 09:54:17 +0200	[thread overview]
Message-ID: <20211015075417.29931-1-lukma@denx.de> (raw)
In-Reply-To: <20210907131616.23472-1-lukma@denx.de>

This change is a partial revert of commit
bca0f5cbc9257c13322b99e55235c4f21ba0bd82
"arm: Simplify elf_machine_{load_address,dynamic}" which imposed usage
of __ehdr_start linker variable to get the address of loaded program.

The elf_machine_load_address() function is declared in the
sysdeps/arm/dl-machine.h header. It is called from (very early)
_dl_start() entry point for the program. It shall return the load
address of the dynamic linker program.

With this revert the 'adr' assembler instruction is used instead of a
place holder:

arm-poky-linux-gnueabi-objdump -t ld-linux-armhf.so.3 | grep ehdr
00000000 l       .note.gnu.build-id     00000000      __ehdr_start

which is pre-set by binutils.

The problem starts when one runs 'prelink' on the rootfs created with
for example OE/Yocto.
Then the _ehdr_start stays as 0x0, but the ELF header's sections have
different addresses - for example 0x41000000 instead of the originally
set 0x0.

This is crucial when /sbin/init is executed. Value set in __ehdr_start
symbol is not updated. This causes the program to crash very early
when ld-linux-armhf.so.3's _dl_start is executed, as calculated offset
for loader relocation is going to hit the kernel space (0xf7xxyyyy).

It looks like the correct way to obtain the _dl_start offset on ARM is
to use assembler instruction 'adr' at execution time (so the prelink
assigned offset is taken into consideration) instead of __ehdr_start.

With this patch we only modify the elf_machine_load_address() function,
as it is called very early, before the ld-linux-armhf.so.3 is performing
relocation (also its own one).

HW:
Hardware name:
	- ARM-Versatile Express (Run with QEMU)
	- Beagle Bone Black

Build Environment: OE/Yocto -> poky
SHA1: 1e2e9a84d6dd81d7f6dd69c0d119d0149d10ade1

Fixes: BZ #28293
---
 sysdeps/arm/dl-machine.h | 28 +++++++++++++++++++++++++---
 1 file changed, 25 insertions(+), 3 deletions(-)

diff --git a/sysdeps/arm/dl-machine.h b/sysdeps/arm/dl-machine.h
index dfa05eee44..d6e5f1d5ec 100644
--- a/sysdeps/arm/dl-machine.h
+++ b/sysdeps/arm/dl-machine.h
@@ -39,11 +39,33 @@ elf_machine_matches_host (const Elf32_Ehdr *ehdr)
 }
 
 /* Return the run-time load address of the shared object.  */
-static inline ElfW(Addr) __attribute__ ((unused))
+static inline Elf32_Addr __attribute__ ((unused))
 elf_machine_load_address (void)
 {
-  extern const ElfW(Ehdr) __ehdr_start attribute_hidden;
-  return (ElfW(Addr)) &__ehdr_start;
+  Elf32_Addr pcrel_addr;
+#ifdef SHARED
+  extern Elf32_Addr __dl_start (void *) asm ("_dl_start");
+  Elf32_Addr got_addr = (Elf32_Addr) &__dl_start;
+  asm ("adr %0, _dl_start" : "=r" (pcrel_addr));
+#else
+  extern Elf32_Addr __dl_relocate_static_pie (void *)
+    asm ("_dl_relocate_static_pie") attribute_hidden;
+  Elf32_Addr got_addr = (Elf32_Addr) &__dl_relocate_static_pie;
+  asm ("adr %0, _dl_relocate_static_pie" : "=r" (pcrel_addr));
+#endif
+#ifdef __thumb__
+  /* Clear the low bit of the function address.
+
+     NOTE: got_addr is from GOT table whose lsb is always set by linker if it's
+     Thumb function address.  PCREL_ADDR comes from PC-relative calculation
+     which will finish during assembling.  GAS assembler before the fix for
+     PR gas/21458 was not setting the lsb but does after that.  Always do the
+     strip for both, so the code works with various combinations of glibc and
+     Binutils.  */
+  got_addr &= ~(Elf32_Addr) 1;
+  pcrel_addr &= ~(Elf32_Addr) 1;
+#endif
+  return pcrel_addr - got_addr;
 }
 
 /* Return the link-time address of _DYNAMIC.  */
-- 
2.20.1


  parent reply	other threads:[~2021-10-15  7:54 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-07 13:16 [PATCH] dl: Use "adr" assembler command to get proper load address Lukasz Majewski
2021-09-07 16:49 ` Fangrui Song
2021-09-07 17:32   ` Lukasz Majewski
2021-09-07 17:44     ` Fangrui Song
2021-09-08 15:05       ` Lukasz Majewski
2021-09-08 17:41         ` Fāng-ruì Sòng
2021-09-08 19:19         ` Adhemerval Zanella
2021-09-08 20:34           ` Lukasz Majewski
2021-09-09  7:18             ` Lukasz Majewski
2021-09-09  9:49               ` Lukasz Majewski
2021-09-10 10:10                 ` Lukasz Majewski
2021-09-17  8:29                   ` Lukasz Majewski
2021-09-17 13:27                     ` Joseph Myers
2021-09-17 16:17                       ` Andreas Schwab
2021-09-26 19:58                       ` Lukasz Majewski
2021-09-27 16:00                         ` Joseph Myers
2021-10-05  7:45       ` Lukasz Majewski
2021-10-06  7:57         ` Fangrui Song
2021-10-06  9:03           ` Lukasz Majewski
2021-10-06 11:43             ` Lukasz Majewski
2021-10-06 12:55               ` Szabolcs Nagy
2021-10-07  9:19                 ` Lukasz Majewski
2021-10-07 10:00                   ` Lukasz Majewski
2021-10-07 14:15                     ` Szabolcs Nagy
2021-10-07 14:58                       ` Lukasz Majewski
2021-10-07 14:16                     ` Adhemerval Zanella
2021-10-07 14:29                       ` H.J. Lu
2021-10-07 15:57                         ` Szabolcs Nagy
2021-10-07 16:22                           ` H.J. Lu
2021-10-07 16:53                             ` Adhemerval Zanella
2021-10-07 17:05                               ` H.J. Lu
2021-10-07 17:24                               ` Fāng-ruì Sòng
2021-10-08  9:15                                 ` Szabolcs Nagy
2021-10-11  8:56                         ` Lukasz Majewski
2021-10-11 10:18                           ` Szabolcs Nagy
2021-10-11 11:47                             ` Lukasz Majewski
2021-10-11 12:01                               ` H.J. Lu
2021-10-11 13:10                                 ` Lukasz Majewski
2021-10-11 13:22                                   ` H.J. Lu
2021-10-11 14:31                                     ` Lukasz Majewski
2021-10-11 13:34                                 ` Adhemerval Zanella
2021-10-11 12:48                               ` Szabolcs Nagy
2021-10-15  7:54 ` Lukasz Majewski [this message]
2021-10-15 12:09   ` [PATCH v2] dl: Use "adr" assembler command to get proper load address on ARM Szabolcs Nagy
2021-10-15 12:21     ` H.J. Lu
2021-10-15 12:59       ` Lukasz Majewski
2021-10-15 23:53         ` Fāng-ruì Sòng
2021-10-18 11:08           ` Szabolcs Nagy
2021-10-18 11:35             ` Florian Weimer
2021-10-19 12:03               ` Lukasz Majewski
2021-10-25 10:18               ` Lukasz Majewski
2021-10-25 10:25                 ` Florian Weimer
2021-10-25 10:53                   ` Lukasz Majewski
2021-10-25 13:34                     ` Szabolcs Nagy
2021-10-25 14:04                       ` Lukasz Majewski
2021-10-25 15:09                         ` Szabolcs Nagy
2021-10-25 17:26                           ` Joseph Myers
2021-10-26 13:52                             ` Lukasz Majewski
2021-10-26 20:55                               ` Joseph Myers
2021-10-27  9:38                                 ` Szabolcs Nagy
2021-10-25 18:25                           ` Lukasz Majewski
2021-10-15 13:59     ` Lukasz Majewski

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=20211015075417.29931-1-lukma@denx.de \
    --to=lukma@denx.de \
    --cc=adhemerval.zanella@linaro.org \
    --cc=carlos@redhat.com \
    --cc=fweimer@redhat.com \
    --cc=joseph@codesourcery.com \
    --cc=libc-alpha@sourceware.org \
    --cc=maskray@google.com \
    --cc=schwab@linux-m68k.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).