On Thu, Jun 26, 2025 at 1:56 PM H.J. Lu <hjl.to...@gmail.com> wrote:
>
> On Thu, Jun 26, 2025 at 1:24 PM Hongtao Liu <crazy...@gmail.com> wrote:
> >
> > On Thu, Jun 26, 2025 at 6:20 AM H.J. Lu <hjl.to...@gmail.com> wrote:
> > >
> > > For tcpsock_test.go in libgo tests,
> > >
> > > commit aba3b9d3a48a0703fd565f7c5f0caf604f59970b
> > > Author: H.J. Lu <hjl.to...@gmail.com>
> > > Date: Fri May 9 07:17:07 2025 +0800
> > >
> > > x86: Extend the remove_redundant_vector pass
> > >
> > > added an instruction:
> > >
> > > (insn 501 101 102 21 (set (reg:V2DI 234)
> > > (vec_duplicate:V2DI (reg:DI 111 [ _46 ])))
> > > "tcpsock_test.go":691:12 disc
> > > rim 1 -1
> > > (nil))
> > >
> > > after
> > >
> > > (insn 101 100 501 21 (set (reg:DI 111 [ _46 ])
> > > (mem:DI (reg/f:DI 110 [ _45 ]) [5 *_45+0 S8 A64]))
> > > "tcpsock_test.go":691
> > > :12 discrim 1 99 {*movdi_internal}
> > > (expr_list:REG_DEAD (reg/f:DI 110 [ _45 ])
> > > (expr_list:REG_EH_REGION (const_int 1 [0x1])
> > > (nil))))
> > >
> > > which resulted in
> > >
> > > (insn 101 100 501 21 (set (reg:DI 111 [ _46 ])
> > > (mem:DI (reg/f:DI 110 [ _45 ]) [5 *_45+0 S8 A64]))
> > > "tcpsock_test.go":691
> > > :12 discrim 1 99 {*movdi_internal}
> > > (expr_list:REG_DEAD (reg/f:DI 110 [ _45 ])
> > > (expr_list:REG_EH_REGION (const_int 1 [0x1])
> > > (nil))))
> > > (insn 501 101 102 21 (set (reg:V2DI 234)
> > > (vec_duplicate:V2DI (reg:DI 111 [ _46 ])))
> > > "tcpsock_test.go":691:12 disc
> > > rim 1 -1
> > > (nil))
> > >
> > > and caused:
> > >
> > > tcpsock_test.go: In function 'net.TestTCPBig..func2':
> > > tcpsock_test.go:684:28: error: in basic block 21:
> > > 684 | go func() {
> > > | ^
> > > tcpsock_test.go:684:28: error: flow control insn inside a basic block
> > > (insn 101 100 501 21 (set (reg:DI 111 [ _46 ])
> > > (mem:DI (reg/f:DI 110 [ _45 ]) [5 *_45+0 S8 A64]))
> > > "tcpsock_test.go":691
> > > :12 discrim 1 99 {*movdi_internal}
> > > (expr_list:REG_DEAD (reg/f:DI 110 [ _45 ])
> > > (expr_list:REG_EH_REGION (const_int 1 [0x1])
> > > (nil))))
> > > during RTL pass: rrvl
> > > tcpsock_test.go:684:28: internal compiler error: in rtl_verify_bb_insns,
> > > at cfgr
> > > tl.cc:2834
> > >
> > > Copy the REG_EH_REGION note to the newly added instruction and split the
> > > block after the previous instruction.
> > >
> > > PR target/120816
> > > * config/i386/i386-features.cc (remove_redundant_vector_load):
> > > Handle REG_EH_REGION note in DEF_INSN.
> > >
> > > OK for master
> > + if (note)
> > + {
> > + control_flow_insns.safe_push (load->def_insn);
> > + add_reg_note (insn, REG_EH_REGION,
> > + XEXP (note, 0));
> > + }
> > I think we don't need to add_reg_not to vec_duplicate insn.
>
> I tried this:
>
> diff --git a/gcc/config/i386/i386-features.cc
> b/gcc/config/i386/i386-features.cc
> index 054f8d5ddc8..30e314b999f 100644
> --- a/gcc/config/i386/i386-features.cc
> +++ b/gcc/config/i386/i386-features.cc
> @@ -3857,8 +3857,6 @@ remove_redundant_vector_load (void)
> if (note)
> {
> control_flow_insns.safe_push (load->def_insn);
> - add_reg_note (insn, REG_EH_REGION,
> - XEXP (note, 0));
> }
> }
>
> and got
>
> tcpsock_test.go: In function ‘net.TestTCPBig..func2’:
> tcpsock_test.go:684:28: error: missing REG_EH_REGION note at the end of bb 71
> 684 | go func() {
> | ^
> during RTL pass: rrvl
> tcpsock_test.go:684:28: internal compiler error: verify_flow_info failed
> 0x2c6522b internal_error(char const*, ...)
> /export/gnu/import/git/gitlab/x86-gcc-test/gcc/diagnostic-global-context.cc:517
> 0x65ddf5 verify_flow_info()
> /export/gnu/import/git/gitlab/x86-gcc-test/gcc/cfghooks.cc:287
> 0xc9612e execute_function_todo
> /export/gnu/import/git/gitlab/x86-gcc-test/gcc/passes.cc:2115
> 0xc94d66 do_per_function
> /export/gnu/import/git/gitlab/x86-gcc-test/gcc/passes.cc:1703
> 0xc962a1 execute_todo
> /export/gnu/import/git/gitlab/x86-gcc-test/gcc/passes.cc:2155
> Please submit a full bug report, with preprocessed source (by using
> -freport-bug).
> Please include the complete backtrace with any bug report.
> See <https://gcc.gnu.org/bugs/> for instructions.
>
> > Other LGTM.
> > I'm wondering if emit_insn_after can be extended to handle this
>
> It is a possibility.
Ok for the original patch.
>
> > situation. It looks like a general issue to me, at least we met it
> > several times.
> >
> > >
> > > --
> > > H.J.
> >
> >
> >
> > --
> > BR,
> > Hongtao
>
>
>
> --
> H.J.
--
BR,
Hongtao
