From: Nelson Chu <nelson@rivosinc.com>
To: Tsukasa OI <research_trasio@irq.a4lg.com>
Cc: Kito Cheng <kito.cheng@sifive.com>,
Palmer Dabbelt <palmer@dabbelt.com>,
binutils@sourceware.org
Subject: Re: [PATCH] RISC-V: Allow merging 'H' extension
Date: Mon, 28 Nov 2022 09:35:41 +0800 [thread overview]
Message-ID: <CAPpQWtBHUoyfuF31B4ndS8_+M_E7RdWqPn7P+VbaUbpxOTDQ4g@mail.gmail.com> (raw)
In-Reply-To: <c55d94343b1e04c73baf4db9d8b22111528f1b40.1669432329.git.research_trasio@irq.a4lg.com>
On Sat, Nov 26, 2022 at 11:14 AM Tsukasa OI
<research_trasio@irq.a4lg.com> wrote:
>
> From: Tsukasa OI <research_trasio@irq.a4lg.com>
>
> Because riscv_merge_std_ext function did not merge the 'H' extension, linked
> executables lacked 'H' extension when multiple objects are linked.
>
> This issue is found while building OpenSBI with 'H' extension (resulting
> ELF files did not contain "h1p0" in "Tag_RISCV_arch" even if *all* linked
> object files contained it).
>
> This commit adds 'h' to standard_exts variable to merge 'H' extension.
>
> bfd/ChangeLog:
>
> * elfnn-riscv.c (riscv_merge_std_ext): Add 'H' extension merging.
>
> ld/ChangeLog:
>
> * testsuite/ld-riscv-elf/attr-merge-arch-02.d: Test merging of 'H'
> and 'Zicsr' extensions.
> * testsuite/ld-riscv-elf/attr-merge-arch-02a.s: Likewise.
Thanks, we missed this, so it looks good except that I think we don't
need to add/update test cases to test this trivial fix.
Nelson
> ---
> bfd/elfnn-riscv.c | 2 +-
> ld/testsuite/ld-riscv-elf/attr-merge-arch-02.d | 2 +-
> ld/testsuite/ld-riscv-elf/attr-merge-arch-02a.s | 2 +-
> 3 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/bfd/elfnn-riscv.c b/bfd/elfnn-riscv.c
> index a2d85dbe9396..a83c8ad2695e 100644
> --- a/bfd/elfnn-riscv.c
> +++ b/bfd/elfnn-riscv.c
> @@ -3427,7 +3427,7 @@ riscv_merge_std_ext (bfd *ibfd,
> struct riscv_subset_t **pin,
> struct riscv_subset_t **pout)
> {
> - const char *standard_exts = "mafdqlcbjtpvn";
> + const char *standard_exts = "mafdqlcbjtpvnh";
> const char *p;
> struct riscv_subset_t *in = *pin;
> struct riscv_subset_t *out = *pout;
> diff --git a/ld/testsuite/ld-riscv-elf/attr-merge-arch-02.d b/ld/testsuite/ld-riscv-elf/attr-merge-arch-02.d
> index 381ef850d974..d1ed8667303c 100644
> --- a/ld/testsuite/ld-riscv-elf/attr-merge-arch-02.d
> +++ b/ld/testsuite/ld-riscv-elf/attr-merge-arch-02.d
> @@ -6,4 +6,4 @@
>
> Attribute Section: riscv
> File Attributes
> - Tag_RISCV_arch: "rv32i2p1_a2p0"
> + Tag_RISCV_arch: "rv32i2p1_a2p0_h1p0_zicsr2p0"
> diff --git a/ld/testsuite/ld-riscv-elf/attr-merge-arch-02a.s b/ld/testsuite/ld-riscv-elf/attr-merge-arch-02a.s
> index 4593241c0249..696e83975346 100644
> --- a/ld/testsuite/ld-riscv-elf/attr-merge-arch-02a.s
> +++ b/ld/testsuite/ld-riscv-elf/attr-merge-arch-02a.s
> @@ -1 +1 @@
> - .attribute arch, "rv32i2p1_a2p0"
> + .attribute arch, "rv32i2p1_a2p0_h1p0_zicsr2p0"
>
> base-commit: 66e7b0f4d9c88781c2b7589f86e31a7d153ed60c
> --
> 2.38.1
>
next prev parent reply other threads:[~2022-11-28 1:35 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-26 3:13 Tsukasa OI
2022-11-28 1:35 ` Nelson Chu [this message]
2022-11-28 2:26 ` Tsukasa OI
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAPpQWtBHUoyfuF31B4ndS8_+M_E7RdWqPn7P+VbaUbpxOTDQ4g@mail.gmail.com \
--to=nelson@rivosinc.com \
--cc=binutils@sourceware.org \
--cc=kito.cheng@sifive.com \
--cc=palmer@dabbelt.com \
--cc=research_trasio@irq.a4lg.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).