From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: qemu-arm@nongnu.org
Subject: [PATCH v4 14/16] accel/tcg: Add x86_64 load_atom_extract_al16_or_al8
Date: Thu, 25 May 2023 17:23:32 -0700 [thread overview]
Message-ID: <20230526002334.1760495-15-richard.henderson@linaro.org> (raw)
In-Reply-To: <20230526002334.1760495-1-richard.henderson@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
| 50 +++++++++++++++++++
1 file changed, 50 insertions(+)
create mode 100644 host/include/x86_64/host/load-extract-al16-al8.h
--git a/host/include/x86_64/host/load-extract-al16-al8.h b/host/include/x86_64/host/load-extract-al16-al8.h
new file mode 100644
index 0000000000..31b6fe8c45
--- /dev/null
+++ b/host/include/x86_64/host/load-extract-al16-al8.h
@@ -0,0 +1,50 @@
+/*
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ * Atomic extract 64 from 128-bit, x86_64 version.
+ *
+ * Copyright (C) 2023 Linaro, Ltd.
+ */
+
+#ifndef X86_64_LOAD_EXTRACT_AL16_AL8_H
+#define X86_64_LOAD_EXTRACT_AL16_AL8_H
+
+#ifdef CONFIG_INT128_TYPE
+#include "host/cpuinfo.h"
+
+/**
+ * load_atom_extract_al16_or_al8:
+ * @pv: host address
+ * @s: object size in bytes, @s <= 8.
+ *
+ * Load @s bytes from @pv, when pv % s != 0. If [p, p+s-1] does not
+ * cross an 16-byte boundary then the access must be 16-byte atomic,
+ * otherwise the access must be 8-byte atomic.
+ */
+static inline uint64_t ATTRIBUTE_ATOMIC128_OPT
+load_atom_extract_al16_or_al8(void *pv, int s)
+{
+ uintptr_t pi = (uintptr_t)pv;
+ __int128_t *ptr_align = (__int128_t *)(pi & ~7);
+ int shr = (pi & 7) * 8;
+ Int128Alias r;
+
+ /*
+ * ptr_align % 16 is now only 0 or 8.
+ * If the host supports atomic loads with VMOVDQU, then always use that,
+ * making the branch highly predictable. Otherwise we must use VMOVDQA
+ * when ptr_align % 16 == 0 for 16-byte atomicity.
+ */
+ if ((cpuinfo & CPUINFO_ATOMIC_VMOVDQU) || (pi & 8)) {
+ asm("vmovdqu %1, %0" : "=x" (r.i) : "m" (*ptr_align));
+ } else {
+ asm("vmovdqa %1, %0" : "=x" (r.i) : "m" (*ptr_align));
+ }
+ return int128_getlo(int128_urshift(r.s, shr));
+}
+#else
+/* Fallback definition that must be optimized away, or error. */
+uint64_t QEMU_ERROR("unsupported atomic")
+ load_atom_extract_al16_or_al8(void *pv, int s);
+#endif
+
+#endif /* X86_64_LOAD_EXTRACT_AL16_AL8_H */
--
2.34.1
next prev parent reply other threads:[~2023-05-26 0:25 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-26 0:23 [PATCH v4 00/16] tcg: Improvements to atomic128 Richard Henderson
2023-05-26 0:23 ` [PATCH v4 01/16] tcg: Fix register move type in tcg_out_ld_helper_ret Richard Henderson
2023-05-30 13:36 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 02/16] accel/tcg: Fix check for page writeability in load_atomic16_or_exit Richard Henderson
2023-05-30 13:44 ` Peter Maydell
2023-05-30 13:58 ` Richard Henderson
2023-05-30 14:06 ` Peter Maydell
2023-05-30 14:29 ` Richard Henderson
2023-05-30 14:48 ` Peter Maydell
2023-05-30 15:09 ` Richard Henderson
2023-05-30 15:18 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 03/16] meson: Split test for __int128_t type from __int128_t arithmetic Richard Henderson
2023-05-30 13:47 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 04/16] qemu/atomic128: Add x86_64 atomic128-ldst.h Richard Henderson
2023-05-26 0:23 ` [PATCH v4 05/16] tcg/i386: Support 128-bit load/store Richard Henderson
2023-05-30 15:02 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 06/16] tcg/aarch64: Rename temporaries Richard Henderson
2023-05-26 0:23 ` [PATCH v4 07/16] tcg/aarch64: Reserve TCG_REG_TMP1, TCG_REG_TMP2 Richard Henderson
2023-05-30 13:52 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 08/16] tcg/aarch64: Simplify constraints on qemu_ld/st Richard Henderson
2023-05-30 13:55 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 09/16] tcg/aarch64: Support 128-bit load/store Richard Henderson
2023-05-26 0:23 ` [PATCH v4 10/16] tcg/ppc: " Richard Henderson
2023-05-26 0:23 ` [PATCH v4 11/16] tcg/s390x: " Richard Henderson
2023-05-26 0:23 ` [PATCH v4 12/16] accel/tcg: Extract load_atom_extract_al16_or_al8 to host header Richard Henderson
2023-05-30 13:58 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 13/16] accel/tcg: Extract store_atom_insert_al16 " Richard Henderson
2023-05-30 13:59 ` Peter Maydell
2023-05-26 0:23 ` Richard Henderson [this message]
2023-05-30 14:01 ` [PATCH v4 14/16] accel/tcg: Add x86_64 load_atom_extract_al16_or_al8 Peter Maydell
2023-05-26 0:23 ` [PATCH v4 15/16] accel/tcg: Add aarch64 lse2 load_atom_extract_al16_or_al8 Richard Henderson
2023-05-30 14:02 ` Peter Maydell
2023-05-26 0:23 ` [PATCH v4 16/16] accel/tcg: Add aarch64 store_atom_insert_al16 Richard Henderson
2023-05-30 14:04 ` 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=20230526002334.1760495-15-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).