From: Richard Henderson <rth@twiddle.net>
To: qemu-devel@nongnu.org
Cc: peter.maydell@linaro.org
Subject: [Qemu-devel] [PULL 01/19] target-tilegx: Tidy simd_helper.c
Date: Wed, 7 Oct 2015 20:32:59 +1100 [thread overview]
Message-ID: <1444210397-20679-2-git-send-email-rth@twiddle.net> (raw)
In-Reply-To: <1444210397-20679-1-git-send-email-rth@twiddle.net>
Using the V1 macro when we want to replicate a byte across
the 8 elements of the word. Using deposit and extract for
manipulating specific elements.
Signed-off-by: Richard Henderson <rth@twiddle.net>
---
target-tilegx/simd_helper.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/target-tilegx/simd_helper.c b/target-tilegx/simd_helper.c
index b931929..f573f9b 100644
--- a/target-tilegx/simd_helper.c
+++ b/target-tilegx/simd_helper.c
@@ -23,12 +23,16 @@
#include "exec/helper-proto.h"
+/* Broadcast a value to all elements of a vector. */
+#define V1(X) (((X) & 0xff) * 0x0101010101010101ull)
+
+
uint64_t helper_v1shl(uint64_t a, uint64_t b)
{
uint64_t m;
b &= 7;
- m = 0x0101010101010101ULL * (0xff >> b);
+ m = V1(0xff >> b);
return (a & m) << b;
}
@@ -37,7 +41,7 @@ uint64_t helper_v1shru(uint64_t a, uint64_t b)
uint64_t m;
b &= 7;
- m = 0x0101010101010101ULL * ((0xff << b) & 0xff);
+ m = V1(0xff << b);
return (a & m) >> b;
}
@@ -48,8 +52,7 @@ uint64_t helper_v1shrs(uint64_t a, uint64_t b)
b &= 7;
for (i = 0; i < 64; i += 8) {
- int64_t ae = (int8_t)(a >> i);
- r |= ((ae >> b) & 0xff) << i;
+ r = deposit64(r, i, 8, sextract64(a, i + b, 8 - b));
}
return r;
}
--
2.4.3
next prev parent reply other threads:[~2015-10-07 9:33 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-07 9:32 [Qemu-devel] [PULL 00/19] Collected tilegx patches Richard Henderson
2015-10-07 9:32 ` Richard Henderson [this message]
2015-10-07 9:33 ` [Qemu-devel] [PULL 02/19] target-tilegx: Implement v*shl, v*shru, and v*shrs instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 03/19] target-tilegx: Implement v*add and v*sub instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 04/19] target-tilegx: Implement v1multu instruction Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 05/19] target-tilegx: Implement crc instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 06/19] target-tilegx: Implement table index instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 07/19] target-tilegx: Implement complex multiply instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 08/19] target-tilegx: Let x1 pipe process bpt instruction only Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 09/19] linux-user/syscall_defs.h: Sync the latest si_code from Linux kernel Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 11/19] target-tilegx: Decode ill pseudo-instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 12/19] target-tilegx: Use TILEGX_EXCP_SIGNAL instead of TILEGX_EXCP_SEGV Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 13/19] target-tilegx: Fix a typo for mnemonic about "ld_add" Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 14/19] target-tilegx: Handle nofault prefetch instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 15/19] target-tilegx: Implement v2sh* instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 16/19] target-tilegx: Implement v?int_* instructions Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 17/19] target-tilegx: Implement v2mults instruction Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 18/19] target-tilegx: Use TILEGX_EXCP_OPCODE_UNKNOWN and TILEGX_EXCP_OPCODE_UNIMPLEMENTED correctly Richard Henderson
2015-10-07 9:33 ` [Qemu-devel] [PULL 19/19] target-tilegx: Support iret instruction and related special registers Richard Henderson
2015-10-08 13:18 ` [Qemu-devel] [PULL 00/19] Collected tilegx patches 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=1444210397-20679-2-git-send-email-rth@twiddle.net \
--to=rth@twiddle.net \
--cc=peter.maydell@linaro.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).