Re: [PATCH] RISC-V: Fix testcase failed when default -mcmodel=medany

[Lehua Ding]

Committed V2 patch, thank you so much.




------------------ Original ------------------
From:                                                                           
                                             "Robin Dapp"                       
                                                             
<gcc-patches@gcc.gnu.org&gt;;
Date:&nbsp;Wed, Jul 19, 2023 02:03 PM
To:&nbsp;"Lehua 
Ding"<lehua.d...@rivai.ai&gt;;"gcc-patches"<gcc-patches@gcc.gnu.org&gt;;
Cc:&nbsp;"rdapp.gcc"<rdapp....@gmail.com&gt;;"juzhe.zhong"<juzhe.zh...@rivai.ai&gt;;"kito.cheng"<kito.ch...@gmail.com&gt;;"palmer"<pal...@rivosinc.com&gt;;"jeffreyalaw"<jeffreya...@gmail.com&gt;;
Subject:&nbsp;Re: [PATCH] RISC-V: Fix testcase failed when default 
-mcmodel=medany



Hi Lehua,

&gt; I think you are rigth, I would like to remove the `-mcmodel=medany` option 
and
&gt; relax assert from `__riscv_save/restore_4` to `__riscv_save/restore_(3|4)` 
to let
&gt; this testcase not brittle on any -mcmodel.&nbsp; Then I'm also going to 
add another
&gt; testcase (I dont known how to run -march=rv32imafc and -march=rv64imafc on
&gt; the same testcase) that uses -march=rv64imafc.

Thanks, ideally we'd have a common include that contains the code
but I don't think we need to stress on that now.&nbsp; At a later point,
we'd ideally have no rv32/rv64 ever in the options (apart from
specific bugs) and just run the test suite for both so only one file
would be needed.&nbsp; But we don't have that yet, so LGTM.

Regards
&nbsp;Robin