Re: Help generating LELTP0 with SPE

Hi!

On Thu, Nov 13, 2025 at 10:34:01AM -0800, Anirudh Mathukumilli wrote:
> Very concretely, I compiled libstdc++ with GCC version 8 (which still had
> support for powerpcpse) and found that the destructor thunks were in different
> sections:
> ```
> .LLSDACSE1350:
>         .section        .text._ZNSt9strstreamD1Ev
>         .size   _ZNSt9strstreamD1Ev, .-_ZNSt9strstreamD1Ev
>         .set    .LTHUNK4,_ZNSt9strstreamD1Ev
>         .section        .text._ZTv0_n12_NSt9strstreamD1Ev,"ax",@progbits
>         .align 2
>         .globl _ZTv0_n12_NSt9strstreamD1Ev
>         .type   _ZTv0_n12_NSt9strstreamD1Ev, @function
> _ZTv0_n12_NSt9strstreamD1Ev:
> .LFB1430:
> .LM389:
>         .cfi_startproc
>         lwz 12,0(3)
>         lwz 12,-12(12)
>         add 3,3,12
>         b .LTHUNK4@local
>         .cfi_endproc
> .LFE1430:
>         .size   _ZTv0_n12_NSt9strstreamD1Ev, .-_ZTv0_n12_NSt9strstreamD1Ev
>         .set    .LTHUNK5,_ZNSt9strstreamD1Ev
>         .section        .text._ZThn8_NSt9strstreamD1Ev,"ax",@progbits
>         .align 2
>         .globl _ZThn8_NSt9strstreamD1Ev
>         .type   _ZThn8_NSt9strstreamD1Ev, @function
> _ZThn8_NSt9strstreamD1Ev:
> .LFB1431:
> .LM390:
>         .cfi_startproc
>         addi 3,3,-8
>         b .LTHUNK5@local
>         .cfi_endproc
> ```
> Now, here's what my port generates:
> ```
> .LLSDACSE1355:
>         .section        .text._ZNSt9strstreamD1Ev
>         .size   _ZNSt9strstreamD1Ev, .-_ZNSt9strstreamD1Ev
>         .set    .LTHUNK4,_ZNSt9strstreamD1Ev
>         .section        .text._ZTv0_n12_NSt9strstreamD1Ev,"ax",@progbits
> .LFB1428:
> .LM674:
>         .cfi_startproc
>         lwz 12,0(3)
>         lwz 12,-12(12)
>         add 3,3,12
>         b .LTHUNK4@local
>         .cfi_endproc
> .LFE1428:
>         .set    .LTHUNK5,_ZNSt9strstreamD1Ev
>         .section        .text._ZThn8_NSt9strstreamD1Ev,"ax",@progbits
> .LFB1429:
> .LM675:
>         .cfi_startproc
>         addi 3,3,-8
>         b .LTHUNK5@local
>         .cfi_endproc
> .LFE1429:
>         .section        .text._ZNSt9strstreamD0Ev,"ax",@progbits
>         .align 2
>         .globl _ZNSt9strstreamD0Ev
>         .type   _ZNSt9strstreamD0Ev, @function
> ```
> I might be missing something, but both seem to put the destructor thunks in
> different sections - so why was GCC 8 able to short circuit and generate static
> numbers while the latest release cannot?

Where do you see the difference between two label addresses here?

>         powerpcspe-* was deprecated in GCC 8, and was removed from GCC 9.  That
>         was in 2018.
> 
>     I’m aware of this - I actually started my work by reverting your patch to
>     delete it! 

You did not get any of the improvements since then (the ones that are
implemented in target code).  That was the big reason to delete this
port btw: it is a significant cost to maintain it, for people unrelated
to the port!

>         > Looking at strstream.s, I noticed that the .LELTP0 label looks like
>         this:
>         > ```
>         > .LELTP0:
>         > ...
>         >         .uleb128 .LM46-.LM45

If those two labels are in different sections, the assembler will give
an error already.  So what emits this piece of assembler code?


Segher