On 4/25/2025 4:37 AM, Steven Rostedt wrote:
> From: Josh Poimboeuf <[email protected]>
> 
> Enable sframe generation in the VDSO library so kernel and user space
> can unwind through it.
> 
> Signed-off-by: Josh Poimboeuf <[email protected]>
> Signed-off-by: Steven Rostedt (Google) <[email protected]>

> diff --git a/arch/x86/entry/vdso/vdso-layout.lds.S 
> b/arch/x86/entry/vdso/vdso-layout.lds.S

> @@ -87,6 +88,7 @@ SECTIONS
>   * Very old versions of ld do not recognize this name token; use the 
> constant.
>   */
>  #define PT_GNU_EH_FRAME      0x6474e550
> +#define PT_GNU_SFRAME        0x6474e554
>  
>  /*
>   * We must supply the ELF program headers explicitly to get just one
> @@ -98,4 +100,5 @@ PHDRS
>       dynamic         PT_DYNAMIC      FLAGS(4);               /* PF_R */
>       note            PT_NOTE         FLAGS(4);               /* PF_R */
>       eh_frame_hdr    PT_GNU_EH_FRAME;

On s390 I found that the respective s390-change needs to be guarded to
prevent the vDSO from erroneously getting generated with a bogus
GNU_SFRAME program table entry, if CONFIG_AS_SFRAME is not enabled:

$ readelf -Wl arch/s390/kernel/vdso64/vdso64.so
...
Program Headers:
  Type           Offset   VirtAddr           PhysAddr           FileSiz  MemSiz 
  Flg Align
  LOAD           0x000000 0x0000000000000000 0x0000000000000000 0x001508 
0x001508 R E 0x1000
  DYNAMIC        0x001190 0x0000000000001190 0x0000000000001190 0x000100 
0x000100 R   0x8
  NOTE           0x000420 0x0000000000000420 0x0000000000000420 0x00003c 
0x00003c R   0x4
  GNU_EH_FRAME   0x000000 0x0000000000000000 0x0000000000000000 0x000000 
0x000000     0x8
  GNU_SFRAME     0x0014f0 0x00000000000014f0 0x00000000000014f0 0x000018 
0x000018 RW  0x8
...

$ xxd arch/s390/kernel/vdso64/vdso64.so
...
000014f0: 0000 0000 0000 1190 0000 0000 0000 0000  ................
00001500: 0000 0000 0000 0000 4743 433a 2028 5562  ........GCC: (Ub
00001510: 756e 7475 2031 342e 322e 302d 3139 7562  untu 14.2.0-19ub
00001520: 756e 7475 3229 2031 342e 322e 3000 0000  untu2) 14.2.0...
...

Following would be the guard for x86 (same as the one used below in
arch/x86/include/asm/dwarf2.h):

#if defined(__x86_64__) && defined(CONFIG_AS_SFRAME)

> +     sframe          PT_GNU_SFRAME;

#endif

>  }

> diff --git a/arch/x86/include/asm/dwarf2.h b/arch/x86/include/asm/dwarf2.h

> @@ -12,8 +12,11 @@
>        * For the vDSO, emit both runtime unwind information and debug
>        * symbols for the .dbg file.
>        */
> -
> +#if defined(__x86_64__) && defined(CONFIG_AS_SFRAME)
> +     .cfi_sections .eh_frame, .debug_frame, .sframe
> +#else
>       .cfi_sections .eh_frame, .debug_frame
> +#endif
>  
>  #define CFI_STARTPROC                .cfi_startproc
>  #define CFI_ENDPROC          .cfi_endproc

Regards,
Jens
-- 
Jens Remus
Linux on Z Development (D3303)
+49-7031-16-1128 Office
[email protected]

IBM

IBM Deutschland Research & Development GmbH; Vorsitzender des Aufsichtsrats: 
Wolfgang Wendt; Geschäftsführung: David Faller; Sitz der Gesellschaft: 
Böblingen; Registergericht: Amtsgericht Stuttgart, HRB 243294
IBM Data Privacy Statement: https://www.ibm.com/privacy/


Reply via email to