public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
* [PATCH v3] hppa: Fix bind-now audit (BZ #28857)
@ 2022-02-07 14:15 Adhemerval Zanella
  2022-02-07 15:28 ` John David Anglin
  2022-02-07 15:46 ` Andreas Schwab
  0 siblings, 2 replies; 4+ messages in thread
From: Adhemerval Zanella @ 2022-02-07 14:15 UTC (permalink / raw)
  To: libc-alpha, John David Anglin

On hppa, a function pointer returned by la_symbind is actually a function
descriptor has the plabel bit set (bit 30).  This must be cleared to get
the actual address of the descriptor.  If the descriptor has been bound,
the first word of the descriptor is the physical address of theA function,
otherwise, the first word of the descriptor points to a trampoline in the
PLT.

This patch also adds a workaround on tests because on hppa (and it seems
to be the only ABI I have see it), some shared library adds a dynamic PLT
relocation to am empty symbol name:

$ readelf -r elf/tst-audit25mod1.so
[...]
Relocation section '.rela.plt' at offset 0x464 contains 6 entries:
 Offset     Info    Type            Sym.Value  Sym. Name + Addend
00002008  00000081 R_PARISC_IPLT                508
[...]

It breaks some assumptions on the test, where a symbol with an empty
name ("") is passed on la_symbind.

Checked on x86_64-linux-gnu and hppa-linux-gnu.
---
v3: Remove _dl_lookup_address usage.
---
 elf/Makefile                | 2 +-
 elf/dl-audit.c              | 3 ++-
 elf/tst-auditmod24a.c       | 4 +++-
 elf/tst-auditmod24d.c       | 4 +++-
 elf/tst-auditmod25.c        | 2 +-
 sysdeps/hppa/dl-lookupcfg.h | 8 +++++---
 6 files changed, 15 insertions(+), 8 deletions(-)

diff --git a/elf/Makefile b/elf/Makefile
index 5bdf0a383d..7372cb191c 100644
--- a/elf/Makefile
+++ b/elf/Makefile
@@ -2210,7 +2210,7 @@ $(objpfx)tst-audit24c.out: $(objpfx)tst-auditmod24c.so
 $(objpfx)tst-audit24c: $(objpfx)tst-audit24amod1.so \
 		       $(objpfx)tst-audit24amod2.so
 tst-audit24c-ENV = LD_BIND_NOW=1 LD_AUDIT=$(objpfx)tst-auditmod24c.so
-LDFLAGS-tst-audit24b = -Wl,-z,lazy
+LDFLAGS-tst-audit24c = -Wl,-z,lazy
 
 $(objpfx)tst-audit24d.out: $(objpfx)tst-auditmod24d.so
 $(objpfx)tst-audit24d: $(objpfx)tst-audit24dmod1.so \
diff --git a/elf/dl-audit.c b/elf/dl-audit.c
index 794bfd45cd..efc0492474 100644
--- a/elf/dl-audit.c
+++ b/elf/dl-audit.c
@@ -257,7 +257,8 @@ _dl_audit_symbind (struct link_map *l, struct reloc_result *reloc_result,
       reloc_result->flags = flags;
     }
 
-  DL_FIXUP_BINDNOW_RELOC (value, new_value, sym.st_value);
+  if (flags & LA_SYMB_ALTVALUE)
+    DL_FIXUP_BINDNOW_RELOC (value, new_value, sym.st_value);
 }
 
 void
diff --git a/elf/tst-auditmod24a.c b/elf/tst-auditmod24a.c
index d8e88f3984..3075dfae2f 100644
--- a/elf/tst-auditmod24a.c
+++ b/elf/tst-auditmod24a.c
@@ -110,5 +110,7 @@ la_symbind32 (Elf32_Sym *sym, unsigned int ndx,
       return sym->st_value;
     }
 
-  abort ();
+  if (symname[0] != '\0')
+    abort ();
+  return sym->st_value;
 }
diff --git a/elf/tst-auditmod24d.c b/elf/tst-auditmod24d.c
index 8c803ecc0a..badc6be451 100644
--- a/elf/tst-auditmod24d.c
+++ b/elf/tst-auditmod24d.c
@@ -116,5 +116,7 @@ la_symbind32 (Elf32_Sym *sym, unsigned int ndx,
 	}
     }
 
-  abort ();
+  if (symname[0] != '\0')
+    abort ();
+  return sym->st_value;
 }
diff --git a/elf/tst-auditmod25.c b/elf/tst-auditmod25.c
index 526f5c54bc..20640a8daf 100644
--- a/elf/tst-auditmod25.c
+++ b/elf/tst-auditmod25.c
@@ -72,7 +72,7 @@ la_symbind32 (Elf32_Sym *sym, unsigned int ndx,
 	      unsigned int *flags, const char *symname)
 #endif
 {
-  if (*refcook != -1 && *defcook != -1)
+  if (*refcook != -1 && *defcook != -1 && symname[0] != '\0')
     fprintf (stderr, "la_symbind: %s %u\n", symname,
 	     *flags & (LA_SYMB_NOPLTENTER | LA_SYMB_NOPLTEXIT) ? 1 : 0);
   return sym->st_value;
diff --git a/sysdeps/hppa/dl-lookupcfg.h b/sysdeps/hppa/dl-lookupcfg.h
index 8da2412fea..535177df66 100644
--- a/sysdeps/hppa/dl-lookupcfg.h
+++ b/sysdeps/hppa/dl-lookupcfg.h
@@ -79,7 +79,9 @@ void attribute_hidden _dl_unmap (struct link_map *map);
 /* Extract the code address from a fixup value */
 #define DL_FIXUP_VALUE_CODE_ADDR(value) ((value).ip)
 #define DL_FIXUP_VALUE_ADDR(value) ((uintptr_t) &(value))
-#define DL_FIXUP_ADDR_VALUE(addr) (*(struct fdesc *) (addr))
+/* Clear the plabel bit to get the actual address of the descriptor.  */
+#define DL_FIXUP_ADDR_VALUE(addr) \
+  (*(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (addr) & ~2))
 #define DL_FIXUP_BINDNOW_ADDR_VALUE(addr) (addr)
-#define DL_FIXUP_BINDNOW_RELOC(value, new_value, st_value) \
-  (*value) = *(struct fdesc *) (st_value)
+#define DL_FIXUP_BINDNOW_RELOC(value, new_value, st_value)	\
+  *(value) = *(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (new_value) & ~2);
-- 
2.32.0


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v3] hppa: Fix bind-now audit (BZ #28857)
  2022-02-07 14:15 [PATCH v3] hppa: Fix bind-now audit (BZ #28857) Adhemerval Zanella
@ 2022-02-07 15:28 ` John David Anglin
  2022-02-07 15:46 ` Andreas Schwab
  1 sibling, 0 replies; 4+ messages in thread
From: John David Anglin @ 2022-02-07 15:28 UTC (permalink / raw)
  To: Adhemerval Zanella, libc-alpha, John David Anglin

On 2022-02-07 9:15 a.m., Adhemerval Zanella wrote:
> diff --git a/sysdeps/hppa/dl-lookupcfg.h b/sysdeps/hppa/dl-lookupcfg.h
> index 8da2412fea..535177df66 100644
> --- a/sysdeps/hppa/dl-lookupcfg.h
> +++ b/sysdeps/hppa/dl-lookupcfg.h
> @@ -79,7 +79,9 @@ void attribute_hidden _dl_unmap (struct link_map *map);
>   /* Extract the code address from a fixup value */
>   #define DL_FIXUP_VALUE_CODE_ADDR(value) ((value).ip)
>   #define DL_FIXUP_VALUE_ADDR(value) ((uintptr_t) &(value))
> -#define DL_FIXUP_ADDR_VALUE(addr) (*(struct fdesc *) (addr))
> +/* Clear the plabel bit to get the actual address of the descriptor.  */
> +#define DL_FIXUP_ADDR_VALUE(addr) \
> +  (*(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (addr) & ~2))
>   #define DL_FIXUP_BINDNOW_ADDR_VALUE(addr) (addr)
> -#define DL_FIXUP_BINDNOW_RELOC(value, new_value, st_value) \
> -  (*value) = *(struct fdesc *) (st_value)
> +#define DL_FIXUP_BINDNOW_RELOC(value, new_value, st_value)	\
> +  *(value) = *(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (new_value) & ~2);
This is okay.

Dave

-- 
John David Anglin  dave.anglin@bell.net


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v3] hppa: Fix bind-now audit (BZ #28857)
  2022-02-07 14:15 [PATCH v3] hppa: Fix bind-now audit (BZ #28857) Adhemerval Zanella
  2022-02-07 15:28 ` John David Anglin
@ 2022-02-07 15:46 ` Andreas Schwab
  2022-02-07 17:13   ` John David Anglin
  1 sibling, 1 reply; 4+ messages in thread
From: Andreas Schwab @ 2022-02-07 15:46 UTC (permalink / raw)
  To: Adhemerval Zanella via Libc-alpha; +Cc: John David Anglin, Adhemerval Zanella

On Feb 07 2022, Adhemerval Zanella via Libc-alpha wrote:

> +/* Clear the plabel bit to get the actual address of the descriptor.  */
> +#define DL_FIXUP_ADDR_VALUE(addr) \
> +  (*(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (addr) & ~2))

Why not uintptr_t?

> +#define DL_FIXUP_BINDNOW_RELOC(value, new_value, st_value)	\
> +  *(value) = *(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (new_value) & ~2);

Likewise.  Why is there a semicolon?

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510  2552 DF73 E780 A9DA AEC1
"And now for something completely different."

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v3] hppa: Fix bind-now audit (BZ #28857)
  2022-02-07 15:46 ` Andreas Schwab
@ 2022-02-07 17:13   ` John David Anglin
  0 siblings, 0 replies; 4+ messages in thread
From: John David Anglin @ 2022-02-07 17:13 UTC (permalink / raw)
  To: Andreas Schwab, Adhemerval Zanella via Libc-alpha
  Cc: John David Anglin, Adhemerval Zanella

On 2022-02-07 10:46 a.m., Andreas Schwab wrote:
> On Feb 07 2022, Adhemerval Zanella via Libc-alpha wrote:
>
>> +/* Clear the plabel bit to get the actual address of the descriptor.  */
>> +#define DL_FIXUP_ADDR_VALUE(addr) \
>> +  (*(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (addr) & ~2))
> Why not uintptr_t?
>
>> +#define DL_FIXUP_BINDNOW_RELOC(value, new_value, st_value)	\
>> +  *(value) = *(DL_FIXUP_VALUE_TYPE *) ((unsigned int) (new_value) & ~2);
> Likewise.  Why is there a semicolon?
Good suggestions.

-- 
John David Anglin  dave.anglin@bell.net


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2022-02-07 17:13 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-02-07 14:15 [PATCH v3] hppa: Fix bind-now audit (BZ #28857) Adhemerval Zanella
2022-02-07 15:28 ` John David Anglin
2022-02-07 15:46 ` Andreas Schwab
2022-02-07 17:13   ` John David Anglin

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).