[v2] pc-bios/s390-ccw: Fixes and improvements for start.S

[PATCH v2 4/4] pc-bios/s390-ccw: Don't use __bss_start with the "larl" instruction

Posted by Thomas Huth 2 years, 7 months ago

start.S currently cannot be compiled with Clang 16 and binutils 2.40:

 ld: start.o(.text+0x8): misaligned symbol `__bss_start' (0xc1e5) for
     relocation R_390_PC32DBL

According to the built-in linker script of ld, the symbol __bss_start
can actually point *before* the .bss section and does not need to have
any alignment, so in certain situations (like when using the internal
assembler of Clang), the __bss_start symbol can indeed be unaligned
and thus it is not suitable for being used with the "larl" instruction
that needs an address that is at least aligned to halfwords.
The problem went unnoticed so far since binutils <= 2.39 did not
check the alignment, but starting with binutils 2.40, such unaligned
addresses are now refused.

Fix it by using the real start address of the .bss section instead.

Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2216662
Reported-by: Miroslav Rezanina <mrezanin@redhat.com>
Suggested-by: Nick Clifton <nickc@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 pc-bios/s390-ccw/start.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pc-bios/s390-ccw/start.S b/pc-bios/s390-ccw/start.S
index 111dea261b..a63c4e3ff2 100644
--- a/pc-bios/s390-ccw/start.S
+++ b/pc-bios/s390-ccw/start.S
@@ -18,7 +18,7 @@ _start:
     larl    %r15,stack + STACK_SIZE - 160   /* Set up stack */
 
     /* clear bss */
-    larl    %r2,__bss_start
+    larl    %r2,.bss
     larl    %r3,_end
     slgr    %r3,%r2    /* get sizeof bss */
     ltgr    %r3,%r3    /* bss empty? */
-- 
2.39.3

Re: [PATCH v2 4/4] pc-bios/s390-ccw: Don't use __bss_start with the "larl" instruction

Posted by Eric Farman 2 years, 7 months ago

On Tue, 2023-06-27 at 09:47 +0200, Thomas Huth wrote:
> start.S currently cannot be compiled with Clang 16 and binutils 2.40:
> 
>  ld: start.o(.text+0x8): misaligned symbol `__bss_start' (0xc1e5) for
>      relocation R_390_PC32DBL
> 
> According to the built-in linker script of ld, the symbol __bss_start
> can actually point *before* the .bss section and does not need to
> have
> any alignment, so in certain situations (like when using the internal
> assembler of Clang), the __bss_start symbol can indeed be unaligned
> and thus it is not suitable for being used with the "larl"
> instruction
> that needs an address that is at least aligned to halfwords.
> The problem went unnoticed so far since binutils <= 2.39 did not
> check the alignment, but starting with binutils 2.40, such unaligned
> addresses are now refused.
> 
> Fix it by using the real start address of the .bss section instead.
> 
> Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2216662
> Reported-by: Miroslav Rezanina <mrezanin@redhat.com>
> Suggested-by: Nick Clifton <nickc@redhat.com>
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
>  pc-bios/s390-ccw/start.S | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Reviewed-by: Eric Farman <farman@linux.ibm.com>

Re: [PATCH v2 4/4] pc-bios/s390-ccw: Don't use __bss_start with the "larl" instruction

Posted by Claudio Imbrenda 2 years, 7 months ago

On Tue, 27 Jun 2023 09:47:03 +0200
Thomas Huth <thuth@redhat.com> wrote:

> start.S currently cannot be compiled with Clang 16 and binutils 2.40:
> 
>  ld: start.o(.text+0x8): misaligned symbol `__bss_start' (0xc1e5) for
>      relocation R_390_PC32DBL
> 
> According to the built-in linker script of ld, the symbol __bss_start
> can actually point *before* the .bss section and does not need to have
> any alignment, so in certain situations (like when using the internal
> assembler of Clang), the __bss_start symbol can indeed be unaligned
> and thus it is not suitable for being used with the "larl" instruction
> that needs an address that is at least aligned to halfwords.
> The problem went unnoticed so far since binutils <= 2.39 did not
> check the alignment, but starting with binutils 2.40, such unaligned
> addresses are now refused.
> 
> Fix it by using the real start address of the .bss section instead.
> 
> Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2216662
> Reported-by: Miroslav Rezanina <mrezanin@redhat.com>
> Suggested-by: Nick Clifton <nickc@redhat.com>
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
>  pc-bios/s390-ccw/start.S | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/pc-bios/s390-ccw/start.S b/pc-bios/s390-ccw/start.S
> index 111dea261b..a63c4e3ff2 100644
> --- a/pc-bios/s390-ccw/start.S
> +++ b/pc-bios/s390-ccw/start.S
> @@ -18,7 +18,7 @@ _start:
>      larl    %r15,stack + STACK_SIZE - 160   /* Set up stack */
>  
>      /* clear bss */
> -    larl    %r2,__bss_start
> +    larl    %r2,.bss
>      larl    %r3,_end

since we are here, do you have guarantees that _end is always correctly
aligned?

if so:

Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>

>      slgr    %r3,%r2    /* get sizeof bss */
>      ltgr    %r3,%r3    /* bss empty? */

Re: [PATCH v2 4/4] pc-bios/s390-ccw: Don't use __bss_start with the "larl" instruction

Posted by Thomas Huth 2 years, 7 months ago

On 27/06/2023 11.29, Claudio Imbrenda wrote:
> On Tue, 27 Jun 2023 09:47:03 +0200
> Thomas Huth <thuth@redhat.com> wrote:
> 
>> start.S currently cannot be compiled with Clang 16 and binutils 2.40:
>>
>>   ld: start.o(.text+0x8): misaligned symbol `__bss_start' (0xc1e5) for
>>       relocation R_390_PC32DBL
>>
>> According to the built-in linker script of ld, the symbol __bss_start
>> can actually point *before* the .bss section and does not need to have
>> any alignment, so in certain situations (like when using the internal
>> assembler of Clang), the __bss_start symbol can indeed be unaligned
>> and thus it is not suitable for being used with the "larl" instruction
>> that needs an address that is at least aligned to halfwords.
>> The problem went unnoticed so far since binutils <= 2.39 did not
>> check the alignment, but starting with binutils 2.40, such unaligned
>> addresses are now refused.
>>
>> Fix it by using the real start address of the .bss section instead.
>>
>> Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2216662
>> Reported-by: Miroslav Rezanina <mrezanin@redhat.com>
>> Suggested-by: Nick Clifton <nickc@redhat.com>
>> Signed-off-by: Thomas Huth <thuth@redhat.com>
>> ---
>>   pc-bios/s390-ccw/start.S | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/pc-bios/s390-ccw/start.S b/pc-bios/s390-ccw/start.S
>> index 111dea261b..a63c4e3ff2 100644
>> --- a/pc-bios/s390-ccw/start.S
>> +++ b/pc-bios/s390-ccw/start.S
>> @@ -18,7 +18,7 @@ _start:
>>       larl    %r15,stack + STACK_SIZE - 160   /* Set up stack */
>>   
>>       /* clear bss */
>> -    larl    %r2,__bss_start
>> +    larl    %r2,.bss
>>       larl    %r3,_end
> 
> since we are here, do you have guarantees that _end is always correctly
> aligned?

Yes, you can see the internal linker script by running
"ld --verbose", and there I get:

   ...
   .data1          : { *(.data1) }
   _edata = .; PROVIDE (edata = .);
   . = .;
   __bss_start = .;
   .bss            :
   {
    *(.dynbss)
    *(.bss .bss.* .gnu.linkonce.b.*)
    *(COMMON)
    /* Align here to ensure that the .bss section occupies space up to
       _end.  Align after .bss to ensure correct alignment even if the
       .bss section disappears because there are no input sections.
       FIXME: Why do we need it? When there is no .bss section, we do not
       pad the .data section.  */
    . = ALIGN(. != 0 ? 64 / 8 : 1);
   }
   . = ALIGN(64 / 8);
   . = SEGMENT_START("ldata-segment", .);
   . = ALIGN(64 / 8);
   _end = .; PROVIDE (end = .);
   . = DATA_SEGMENT_END (.);
   ...

As you can see, there is no alignment in front of
__bss_start, but there is alignment in front of
__end.

> if so:
> 
> Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>

Thanks!

  Thomas