From: Nicholas Piggin <npiggin@gmail.com>
To: qemu-ppc@nongnu.org
Cc: Nicholas Piggin <npiggin@gmail.com>,
qemu-devel@nongnu.org,
Daniel Henrique Barboza <danielhb413@gmail.com>,
"Chinmay Rath" <rathc@linux.ibm.com>
Subject: [PATCH 3/3] target/ppc: Add ISA v3.1 variants of sync instruction
Date: Wed, 1 May 2024 23:04:34 +1000 [thread overview]
Message-ID: <20240501130435.941189-4-npiggin@gmail.com> (raw)
In-Reply-To: <20240501130435.941189-1-npiggin@gmail.com>
POWER10 adds a new field to sync for store-store syncs, and some
new variants of the existing syncs that include persistent memory.
Implement the store-store syncs and plwsync/phwsync.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
target/ppc/insn32.decode | 6 ++--
target/ppc/translate/misc-impl.c.inc | 41 ++++++++++++++++++++--------
2 files changed, 32 insertions(+), 15 deletions(-)
diff --git a/target/ppc/insn32.decode b/target/ppc/insn32.decode
index 6b89804b15..a180380750 100644
--- a/target/ppc/insn32.decode
+++ b/target/ppc/insn32.decode
@@ -1001,7 +1001,7 @@ MSGSYNC 011111 ----- ----- ----- 1101110110 -
# Memory Barrier Instructions
-&X_sync l
-@X_sync ...... ... l:2 ..... ..... .......... . &X_sync
-SYNC 011111 --- .. ----- ----- 1001010110 - @X_sync
+&X_sync l sc
+@X_sync ...... .. l:3 ... sc:2 ..... .......... . &X_sync
+SYNC 011111 -- ... --- .. ----- 1001010110 - @X_sync
EIEIO 011111 ----- ----- ----- 1101010110 -
diff --git a/target/ppc/translate/misc-impl.c.inc b/target/ppc/translate/misc-impl.c.inc
index 9226467f81..3467b49d0d 100644
--- a/target/ppc/translate/misc-impl.c.inc
+++ b/target/ppc/translate/misc-impl.c.inc
@@ -25,6 +25,7 @@ static bool trans_SYNC(DisasContext *ctx, arg_X_sync *a)
{
TCGBar bar = TCG_MO_ALL;
uint32_t l = a->l;
+ uint32_t sc = a->sc;
/*
* BookE uses the msync mnemonic. This means hwsync, except in the
@@ -46,20 +47,36 @@ static bool trans_SYNC(DisasContext *ctx, arg_X_sync *a)
gen_inval_exception(ctx, POWERPC_EXCP_INVAL_INVAL);
}
- if ((l == 1) && (ctx->insns_flags2 & PPC2_MEM_LWSYNC)) {
- bar = TCG_MO_LD_LD | TCG_MO_LD_ST | TCG_MO_ST_ST;
- }
-
/*
- * We may need to check for a pending TLB flush.
- *
- * We do this on ptesync (l == 2) on ppc64 and any sync on ppc32.
- *
- * Additionally, this can only happen in kernel mode however so
- * check MSR_PR as well.
+ * In ISA v3.1, the L field grew one bit. Mask that out to ignore it in
+ * older processors. It also added the SC field, zero this to ignore
+ * it too.
*/
- if (((l == 2) || !(ctx->insns_flags & PPC_64B)) && !ctx->pr) {
- gen_check_tlb_flush(ctx, true);
+ if (!(ctx->insns_flags2 & PPC2_ISA310)) {
+ l &= 0x3;
+ sc = 0;
+ }
+
+ if (sc) {
+ /* Store syncs [stsync, stcisync, stncisync]. These ignore L. */
+ bar = TCG_MO_ST_ST;
+ } else {
+ if (((l == 1) && (ctx->insns_flags2 & PPC2_MEM_LWSYNC)) || (l == 5)) {
+ /* lwsync, or plwsync on POWER10 and later */
+ bar = TCG_MO_LD_LD | TCG_MO_LD_ST | TCG_MO_ST_ST;
+ }
+
+ /*
+ * We may need to check for a pending TLB flush.
+ *
+ * We do this on ptesync (l == 2) on ppc64 and any sync on ppc32.
+ *
+ * Additionally, this can only happen in kernel mode however so
+ * check MSR_PR as well.
+ */
+ if (((l == 2) || !(ctx->insns_flags & PPC_64B)) && !ctx->pr) {
+ gen_check_tlb_flush(ctx, true);
+ }
}
tcg_gen_mb(bar | TCG_BAR_SC);
--
2.43.0
next prev parent reply other threads:[~2024-05-01 13:06 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-01 13:04 [PATCH 0/3] target/ppc: Fixes and updates for sync instructions Nicholas Piggin
2024-05-01 13:04 ` [PATCH 1/3] target/ppc: Move sync instructions to decodetree Nicholas Piggin
2024-05-07 6:41 ` Chinmay Rath
2024-05-01 13:04 ` [PATCH 2/3] target/ppc: Fix embedded memory barriers Nicholas Piggin
2024-05-07 7:24 ` Chinmay Rath
2024-05-01 13:04 ` Nicholas Piggin [this message]
2024-05-07 7:09 ` [PATCH 3/3] target/ppc: Add ISA v3.1 variants of sync instruction Chinmay Rath
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=20240501130435.941189-4-npiggin@gmail.com \
--to=npiggin@gmail.com \
--cc=danielhb413@gmail.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=rathc@linux.ibm.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.