From: Deepak Gupta <debug@rivosinc.com>
To: Alistair Francis <alistair23@gmail.com>
Cc: qemu-riscv@nongnu.org, qemu-devel@nongnu.org, palmer@dabbelt.com,
alistair.francis@wdc.com, bmeng.cn@gmail.com,
liwei1518@gmail.com, dbarboza@ventanamicro.com,
zhiwei_liu@linux.alibaba.com, Ved Shanbhogue <ved@rivosinc.com>
Subject: Re: [PATCH 2/2] target/riscv: fixes a bug against `ssamoswap` behavior in M-mode
Date: Wed, 5 Mar 2025 22:13:18 -0800 [thread overview]
Message-ID: <Z8k8/rhi/COfv/vj@debug.ba.rivosinc.com> (raw)
In-Reply-To: <CAKmqyKMip2t-QyEvFFvnsMkXV1Wokt0dXAqpe4MAzyCHpUm--A@mail.gmail.com>
On Thu, Mar 06, 2025 at 03:29:00PM +1000, Alistair Francis wrote:
>On Tue, Feb 18, 2025 at 12:57 PM Deepak Gupta <debug@rivosinc.com> wrote:
>>
>> Commit f06bfe3dc38c ("target/riscv: implement zicfiss instructions") adds
>> `ssamoswap` instruction. `ssamoswap` takes the code-point from existing
>> reserved encoding (and not a zimop like other shadow stack instructions).
>> If shadow stack is not enabled (via xenvcfg.SSE), then `ssamoswap` must
>> result in an illegal instruction exception. However there is a slightly
>> modified behavior for M-mode.
>>
>> Shadow stack are not available in M-mode and all shadow stack instructions
>> in M-mode exhibit zimop behavior. However, `ssamoswap` can still succeed
>> if MPRV=1 and MPP is non-zero (see section 2.7 of zicfiss specification).
>> This patch corrects that behavior for `ssamoswap`.
>
>Section "22.2.3. Shadow Stack Memory Protection " of the latest priv
>spec [1] seems to say: "When the effective privilege mode is M, any
>memory access by an SSAMOSWAP.W/D
>instruction will result in a store/AMO access-fault exception."
Hmm I didn't look at priv spec. Let me fix this one.
>
>1: https://github.com/riscv/riscv-isa-manual/releases/tag/riscv-isa-release-9cfaf37-2025-03-06
>
>Alistair
>
>>
>> Fixes: f06bfe3dc38c ("target/riscv: implement zicfiss instructions")
>>
>> Reported-by: Ved Shanbhogue <ved@rivosinc.com>
>> Signed-off-by: Deepak Gupta <debug@rivosinc.com>
>> ---
>> target/riscv/insn_trans/trans_rvzicfiss.c.inc | 13 +++++++++++--
>> 1 file changed, 11 insertions(+), 2 deletions(-)
>>
>> diff --git a/target/riscv/insn_trans/trans_rvzicfiss.c.inc b/target/riscv/insn_trans/trans_rvzicfiss.c.inc
>> index e3ebc4977c..ec016cd70f 100644
>> --- a/target/riscv/insn_trans/trans_rvzicfiss.c.inc
>> +++ b/target/riscv/insn_trans/trans_rvzicfiss.c.inc
>> @@ -15,6 +15,13 @@
>> * You should have received a copy of the GNU General Public License along with
>> * this program. If not, see <http://www.gnu.org/licenses/>.
>> */
>> +
>> + #define REQUIRE_ZICFISS(ctx) do { \
>> + if (!ctx->cfg_ptr->ext_zicfiss) { \
>> + return false; \
>> + } \
>> +} while (0)
>> +
>> static bool trans_sspopchk(DisasContext *ctx, arg_sspopchk *a)
>> {
>> if (!ctx->bcfi_enabled) {
>> @@ -77,7 +84,8 @@ static bool trans_ssrdp(DisasContext *ctx, arg_ssrdp *a)
>> static bool trans_ssamoswap_w(DisasContext *ctx, arg_amoswap_w *a)
>> {
>> REQUIRE_A_OR_ZAAMO(ctx);
>> - if (!ctx->bcfi_enabled) {
>> + REQUIRE_ZICFISS(ctx);
>> + if ((ctx->priv != PRV_M) && !ctx->bcfi_enabled) {
>> return false;
>> }
>>
>> @@ -97,7 +105,8 @@ static bool trans_ssamoswap_d(DisasContext *ctx, arg_amoswap_w *a)
>> {
>> REQUIRE_64BIT(ctx);
>> REQUIRE_A_OR_ZAAMO(ctx);
>> - if (!ctx->bcfi_enabled) {
>> + REQUIRE_ZICFISS(ctx);
>> + if ((ctx->priv != PRV_M) && !ctx->bcfi_enabled) {
>> return false;
>> }
>>
>> --
>> 2.34.1
>>
>>
next prev parent reply other threads:[~2025-03-06 6:13 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-18 2:54 [PATCH 1/2] target/riscv: fix access permission checks for CSR_SSP Deepak Gupta
2025-02-18 2:54 ` [PATCH 2/2] target/riscv: fixes a bug against `ssamoswap` behavior in M-mode Deepak Gupta
2025-03-06 5:29 ` Alistair Francis
2025-03-06 6:13 ` Deepak Gupta [this message]
2025-03-06 6:22 ` Alistair Francis
2025-03-06 6:30 ` Deepak Gupta
2025-03-06 5:20 ` [PATCH 1/2] target/riscv: fix access permission checks for CSR_SSP Alistair Francis
2025-03-06 6:12 ` Deepak Gupta
2025-03-06 6:20 ` Alistair Francis
2025-03-06 6:22 ` Deepak Gupta
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=Z8k8/rhi/COfv/vj@debug.ba.rivosinc.com \
--to=debug@rivosinc.com \
--cc=alistair.francis@wdc.com \
--cc=alistair23@gmail.com \
--cc=bmeng.cn@gmail.com \
--cc=dbarboza@ventanamicro.com \
--cc=liwei1518@gmail.com \
--cc=palmer@dabbelt.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
--cc=ved@rivosinc.com \
--cc=zhiwei_liu@linux.alibaba.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 NNTP newsgroup(s).