qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Cc: Christophe Lyon <christophe.lyon@st.com>, patches@linaro.org
Subject: [Qemu-devel] [PATCH 07/10] target-arm: fix decoding of Neon 64 bit shifts.
Date: Tue, 15 Feb 2011 13:44:47 +0000	[thread overview]
Message-ID: <1297777490-5323-8-git-send-email-peter.maydell@linaro.org> (raw)
In-Reply-To: <1297777490-5323-1-git-send-email-peter.maydell@linaro.org>

From: Christophe Lyon <christophe.lyon@st.com>

Fix decoding of 64 bits variants of VSHRN, VRSHRN, VQSHRN, VQSHRUN,
VQRSHRN, VQRSHRUN, taking into account whether inputs are unsigned
or not.

Signed-off-by: Christophe Lyon <christophe.lyon@st.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
---
 target-arm/translate.c |   45 ++++++++++++++++++++++++++++++---------------
 1 files changed, 30 insertions(+), 15 deletions(-)

diff --git a/target-arm/translate.c b/target-arm/translate.c
index bd26b1b..a02b20f 100644
--- a/target-arm/translate.c
+++ b/target-arm/translate.c
@@ -4815,6 +4815,8 @@ static int disas_neon_data_insn(CPUState * env, DisasContext *s, uint32_t insn)
             } else if (op < 10) {
                 /* Shift by immediate and narrow:
                    VSHRN, VRSHRN, VQSHRN, VQRSHRN.  */
+                int input_unsigned = (op == 8) ? !u : u;
+
                 shift = shift - (1 << (size + 3));
                 size++;
                 switch (size) {
@@ -4841,33 +4843,46 @@ static int disas_neon_data_insn(CPUState * env, DisasContext *s, uint32_t insn)
                     if (size == 3) {
                         neon_load_reg64(cpu_V0, rm + pass);
                         if (q) {
-                          if (u)
-                            gen_helper_neon_rshl_u64(cpu_V0, cpu_V0, tmp64);
-                          else
-                            gen_helper_neon_rshl_s64(cpu_V0, cpu_V0, tmp64);
+                            if (input_unsigned) {
+                                gen_helper_neon_rshl_u64(cpu_V0, cpu_V0,
+                                                         tmp64);
+                            } else {
+                                gen_helper_neon_rshl_s64(cpu_V0, cpu_V0,
+                                                         tmp64);
+                            }
                         } else {
-                          if (u)
-                            gen_helper_neon_shl_u64(cpu_V0, cpu_V0, tmp64);
-                          else
-                            gen_helper_neon_shl_s64(cpu_V0, cpu_V0, tmp64);
+                            if (input_unsigned) {
+                                gen_helper_neon_shl_u64(cpu_V0, cpu_V0,
+                                                        tmp64);
+                            } else {
+                                gen_helper_neon_shl_s64(cpu_V0, cpu_V0,
+                                                        tmp64);
+                            }
                         }
                     } else {
                         tmp = neon_load_reg(rm + pass, 0);
-                        gen_neon_shift_narrow(size, tmp, tmp2, q, u);
+                        gen_neon_shift_narrow(size, tmp, tmp2, q,
+                                              input_unsigned);
                         tmp3 = neon_load_reg(rm + pass, 1);
-                        gen_neon_shift_narrow(size, tmp3, tmp2, q, u);
+                        gen_neon_shift_narrow(size, tmp3, tmp2, q,
+                                              input_unsigned);
                         tcg_gen_concat_i32_i64(cpu_V0, tmp, tmp3);
                         dead_tmp(tmp);
                         dead_tmp(tmp3);
                     }
                     tmp = new_tmp();
-                    if (op == 8 && !u) {
-                        gen_neon_narrow(size - 1, tmp, cpu_V0);
+                    if (op == 8) {
+                        if (u) { /* VQSHRUN / VQRSHRUN */
+                            gen_neon_unarrow_sats(size - 1, tmp, cpu_V0);
+                        } else { /* VSHRN / VRSHRN */
+                            gen_neon_narrow(size - 1, tmp, cpu_V0);
+                        }
                     } else {
-                        if (op == 8)
-                            gen_neon_narrow_sats(size - 1, tmp, cpu_V0);
-                        else
+                        if (u) { /* VQSHRN / VQRSHRN */
                             gen_neon_narrow_satu(size - 1, tmp, cpu_V0);
+                        } else { /* VQSHRN / VQRSHRN */
+                            gen_neon_narrow_sats(size - 1, tmp, cpu_V0);
+                        }
                     }
                     neon_store_reg(rd, pass, tmp);
                 } /* for pass */
-- 
1.7.1

  parent reply	other threads:[~2011-02-15 13:45 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-15 13:44 [Qemu-devel] [PATCH 00/10] Fix Neon shift instructions Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 01/10] target-arm: Fix rounding constant addition for Neon shifts Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 02/10] target-arm: Fix signed VRSHL by large shift counts Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 03/10] target-arm: Fix unsigned VRSHL.s8 and .s16 right shifts by type width Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 04/10] target-arm: fix unsigned 64 bit right shifts Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 05/10] target-arm: Fix saturated values for Neon " Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 06/10] target-arm: fix Neon VQSHRN and VSHRN Peter Maydell
2011-02-15 13:44 ` Peter Maydell [this message]
2011-02-15 13:44 ` [Qemu-devel] [PATCH 08/10] target-arm: Fix signed VQRSHL by large shift counts Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 09/10] target-arm: Fix unsigned " Peter Maydell
2011-02-15 13:44 ` [Qemu-devel] [PATCH 10/10] target-arm: Fix shift by immediate and narrow where src, dest overlap Peter Maydell
2011-02-20 16:52   ` Aurelien Jarno
2011-02-15 16:35 ` [Qemu-devel] Re: [PATCH 00/10] Fix Neon shift instructions Christophe Lyon
2011-02-20 16:52 ` [Qemu-devel] " Aurelien Jarno

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=1297777490-5323-8-git-send-email-peter.maydell@linaro.org \
    --to=peter.maydell@linaro.org \
    --cc=christophe.lyon@st.com \
    --cc=patches@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).