From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: qemu-arm@nongnu.org
Subject: [PATCH] target/arm: Do not use gen_mte_checkN in trans_STGP
Date: Thu, 27 Jul 2023 09:26:21 -0700 [thread overview]
Message-ID: <20230727162621.445400-1-richard.henderson@linaro.org> (raw)
STGP writes to tag memory, it does not check it.
This happened to work because we wrote tag memory first
so that the check always succeeded.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
target/arm/tcg/translate-a64.c | 41 +++++++++++++---------------------
1 file changed, 15 insertions(+), 26 deletions(-)
diff --git a/target/arm/tcg/translate-a64.c b/target/arm/tcg/translate-a64.c
index 5fa1257d32..dfd18e19ca 100644
--- a/target/arm/tcg/translate-a64.c
+++ b/target/arm/tcg/translate-a64.c
@@ -3020,37 +3020,17 @@ static bool trans_STGP(DisasContext *s, arg_ldstpair *a)
tcg_gen_addi_i64(dirty_addr, dirty_addr, offset);
}
- if (!s->ata) {
- /*
- * TODO: We could rely on the stores below, at least for
- * system mode, if we arrange to add MO_ALIGN_16.
- */
- gen_helper_stg_stub(cpu_env, dirty_addr);
- } else if (tb_cflags(s->base.tb) & CF_PARALLEL) {
- gen_helper_stg_parallel(cpu_env, dirty_addr, dirty_addr);
- } else {
- gen_helper_stg(cpu_env, dirty_addr, dirty_addr);
- }
-
- mop = finalize_memop(s, MO_64);
- clean_addr = gen_mte_checkN(s, dirty_addr, true, false, 2 << MO_64, mop);
-
+ clean_addr = clean_data_tbi(s, dirty_addr);
tcg_rt = cpu_reg(s, a->rt);
tcg_rt2 = cpu_reg(s, a->rt2);
/*
- * STGP is defined as two 8-byte memory operations and one tag operation.
- * We implement it as one single 16-byte memory operation for convenience.
- * Rebuild mop as for STP.
- * TODO: The atomicity with LSE2 is stronger than required.
- * Need a form of MO_ATOM_WITHIN16_PAIR that never requires
- * 16-byte atomicity.
+ * STGP is defined as two 8-byte memory operations, aligned to TAG_GRANULE,
+ * and one tag operation. We implement it as one single aligned 16-byte
+ * memory operation for convenience. Note that the alignment ensures
+ * MO_ATOM_IFALIGN_PAIR produces 8-byte atomicity for the memory store.
*/
- mop = MO_128;
- if (s->align_mem) {
- mop |= MO_ALIGN_8;
- }
- mop = finalize_memop_pair(s, mop);
+ mop = MO_128 | MO_ALIGN | MO_ATOM_IFALIGN_PAIR;
tmp = tcg_temp_new_i128();
if (s->be_data == MO_LE) {
@@ -3060,6 +3040,15 @@ static bool trans_STGP(DisasContext *s, arg_ldstpair *a)
}
tcg_gen_qemu_st_i128(tmp, clean_addr, get_mem_index(s), mop);
+ /* Perform the tag store, if tag access enabled. */
+ if (s->ata) {
+ if (tb_cflags(s->base.tb) & CF_PARALLEL) {
+ gen_helper_stg_parallel(cpu_env, dirty_addr, dirty_addr);
+ } else {
+ gen_helper_stg(cpu_env, dirty_addr, dirty_addr);
+ }
+ }
+
op_addr_ldstpair_post(s, a, dirty_addr, offset);
return true;
}
--
2.34.1
next reply other threads:[~2023-07-27 18:07 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-27 16:26 Richard Henderson [this message]
2023-07-28 13:17 ` [PATCH] target/arm: Do not use gen_mte_checkN in trans_STGP Peter Maydell
2023-07-28 13:30 ` Richard Henderson
2023-08-03 13:10 ` Peter Maydell
2023-08-03 14:42 ` Richard Henderson
2023-09-01 12:36 ` Peter Maydell
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=20230727162621.445400-1-richard.henderson@linaro.org \
--to=richard.henderson@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
/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).