All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nathan Froyd <froydnj@codesourcery.com>
To: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 20/40] Add vs{l,r} instructions.
Date: Wed, 7 Jan 2009 13:25:02 -0800	[thread overview]
Message-ID: <20090107212502.GE28711@codesourcery.com> (raw)
In-Reply-To: <20090105192730.GK18974@volta.aurel32.net>

On Mon, Jan 05, 2009 at 08:27:30PM +0100, Aurelien Jarno wrote:
> On Mon, Jan 05, 2009 at 10:29:56AM -0800, Nathan Froyd wrote:
> > FWIW, doing the check has the nice property of delivering the same
> > results as real hardware.  If you're using a comparison program like
> > ppctester, reducing those spurious failures is a win.  (There's already
> > spurious failures for div instructions in corner cases.)
> 
> Then I am fine with that. Could you please just add a comment that it is
> not necessary according to the spec, but done to match real hardware?

Done thusly.

-Nathan

>From aa19aa5e31aa9eec36a2cdfb2e1ed806555eda96 Mon Sep 17 00:00:00 2001
From: Nathan Froyd <froydnj@codesourcery.com>
Date: Thu, 11 Dec 2008 12:02:03 -0800
Subject: [PATCH 4/5] Add vs{l,r} instructions.

Add explanatory comment about why we check equality of shift counts.

Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
---
 target-ppc/helper.h    |    2 ++
 target-ppc/op_helper.c |   39 +++++++++++++++++++++++++++++++++++++++
 target-ppc/translate.c |    2 ++
 3 files changed, 43 insertions(+), 0 deletions(-)

diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index 009f516..e15d6b0 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -185,6 +185,8 @@ DEF_HELPER_3(vsubuws, void, avr, avr, avr)
 DEF_HELPER_3(vrlb, void, avr, avr, avr)
 DEF_HELPER_3(vrlh, void, avr, avr, avr)
 DEF_HELPER_3(vrlw, void, avr, avr, avr)
+DEF_HELPER_3(vsl, void, avr, avr, avr)
+DEF_HELPER_3(vsr, void, avr, avr, avr)
 DEF_HELPER_4(vsldoi, void, avr, avr, avr, i32)
 DEF_HELPER_3(vspltb, void, avr, avr, i32)
 DEF_HELPER_3(vsplth, void, avr, avr, i32)
diff --git a/target-ppc/op_helper.c b/target-ppc/op_helper.c
index ef19bde..81abd63 100644
--- a/target-ppc/op_helper.c
+++ b/target-ppc/op_helper.c
@@ -2491,6 +2491,45 @@ void helper_vsel (ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b, ppc_avr_t *c)
     r->u64[1] = (a->u64[1] & ~c->u64[1]) | (b->u64[1] & c->u64[1]);
 }
 
+#if defined(WORDS_BIGENDIAN)
+#define LEFT 0
+#define RIGHT 1
+#else
+#define LEFT 1
+#define RIGHT 0
+#endif
+/* The specification says that the results are undefined if all of the
+ * shift counts are not identical.  We check to make sure that they are
+ * to conform to what real hardware appears to do.  */
+#define VSHIFT(suffix, leftp)                                           \
+    void helper_vs##suffix (ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)   \
+    {                                                                   \
+        int shift = b->u8[LO_IDX*0x15] & 0x7;                           \
+        int doit = 1;                                                   \
+        int i;                                                          \
+        for (i = 0; i < ARRAY_SIZE(r->u8); i++) {                       \
+            doit = doit && ((b->u8[i] & 0x7) == shift);                 \
+        }                                                               \
+        if (doit) {                                                     \
+            if (shift == 0) {                                           \
+                *r = *a;                                                \
+            } else if (leftp) {                                         \
+                uint64_t carry = a->u64[LO_IDX] >> (64 - shift);        \
+                r->u64[HI_IDX] = (a->u64[HI_IDX] << shift) | carry;     \
+                r->u64[LO_IDX] = a->u64[LO_IDX] << shift;               \
+            } else {                                                    \
+                uint64_t carry = a->u64[HI_IDX] << (64 - shift);        \
+                r->u64[LO_IDX] = (a->u64[LO_IDX] >> shift) | carry;     \
+                r->u64[HI_IDX] = a->u64[HI_IDX] >> shift;               \
+            }                                                           \
+        }                                                               \
+    }
+VSHIFT(l, LEFT)
+VSHIFT(r, RIGHT)
+#undef VSHIFT
+#undef LEFT
+#undef RIGHT
+
 #define VSL(suffix, element)                                            \
     void helper_vsl##suffix (ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)  \
     {                                                                   \
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index e92156f..65a80f2 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -6356,6 +6356,8 @@ GEN_VXFORM(vsubsws, 0, 30);
 GEN_VXFORM(vrlb, 2, 0);
 GEN_VXFORM(vrlh, 2, 1);
 GEN_VXFORM(vrlw, 2, 2);
+GEN_VXFORM(vsl, 2, 7);
+GEN_VXFORM(vsr, 2, 11);
 GEN_VXFORM(vpkuhum, 7, 0);
 GEN_VXFORM(vpkuwum, 7, 1);
 GEN_VXFORM(vpkuhus, 7, 2);
-- 
1.6.0.5

  reply	other threads:[~2009-01-07 21:25 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-12-31  3:09 [Qemu-devel] [PATCH] target-ppc: add integer Altivec instructions, take 2 Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 01/40] Fix TCG error in gen_avr_ptr Nathan Froyd
2009-01-03 13:31   ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 02/40] Add helper macros for later patches Nathan Froyd
2009-01-03 13:32   ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 03/40] Add GEN_VXFORM macro for subsequent instructions Nathan Froyd
2009-01-03 13:32   ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 04/40] Add v{add,sub}u{b,h,w}m instructions Nathan Froyd
2009-01-03 13:33   ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 05/40] Add vavg{s,u}{b,h,w} instructions Nathan Froyd
2009-01-03 13:33   ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 06/40] Add v{min, max}{s, u}{b, h, w} instructions Nathan Froyd
2009-01-03 13:33   ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 07/40] Add GEN_VXRFORM{, 1} macros for subsequent instructions Nathan Froyd
2009-01-03 13:58   ` Aurelien Jarno
2009-01-07 21:22     ` Nathan Froyd
2009-01-08 18:55       ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 08/40] Add vcmpequ{b, h, w} and vcmpgt{s, u}{b, h, w} instructions Nathan Froyd
2009-01-03 14:02   ` Aurelien Jarno
2009-01-07 21:23     ` Nathan Froyd
2009-01-08 18:56       ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 09/40] Add vscr access macros Nathan Froyd
2009-01-03 14:04   ` Aurelien Jarno
2008-12-31  3:09 ` [Qemu-devel] [PATCH 10/40] Add vmrg{l,h}{b,h,w} instructions Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 11/40] Add vmul{e,o}{s,u}{b,h} instructions Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 12/40] Add vsr{,a}{b,h,w} instructions Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 13/40] Add vsl{b,h,w} instructions Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 14/40] Add vs{l,r}o instructions Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 15/40] Add v{add,sub}cuw instructions Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 16/40] Add lvs{l,r} instructions Nathan Froyd
2008-12-31  3:09 ` [Qemu-devel] [PATCH 17/40] Add m{f,t}vscr instructions Nathan Froyd
2009-01-03 19:43   ` Aurelien Jarno
2008-12-31  3:10 ` [Qemu-devel] [PATCH 18/40] Add v{add, sub}{s, u}{b, h, w}s instructions Nathan Froyd
2009-01-03 21:13   ` Aurelien Jarno
2009-01-07 21:24     ` Nathan Froyd
2009-01-08 23:19       ` Aurelien Jarno
2008-12-31  3:10 ` [Qemu-devel] [PATCH 19/40] Add vrl{b,h,w} instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 20/40] Add vs{l,r} instructions Nathan Froyd
2009-01-03 21:26   ` Aurelien Jarno
2009-01-05 18:29     ` Nathan Froyd
2009-01-05 19:27       ` Aurelien Jarno
2009-01-07 21:25         ` Nathan Froyd [this message]
2009-01-08 18:56           ` Aurelien Jarno
2008-12-31  3:10 ` [Qemu-devel] [PATCH 21/40] Add vsldoi instruction Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 22/40] Add GEN_VXFORM_SIMM macro for subsequent instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 23/40] Add vspltis{b,h,w} instructions Nathan Froyd
2009-01-03 21:43   ` Aurelien Jarno
2009-01-07 21:25     ` Nathan Froyd
2009-01-08 18:57       ` Aurelien Jarno
2008-12-31  3:10 ` [Qemu-devel] [PATCH 24/40] Add GEN_VXFORM_UIMM macro for subsequent instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 25/40] Add vsplt{b,h,w} instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 26/40] Add GEN_VXFORM_NOA macro for subsequent instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 27/40] Add vupk{h,l}px instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 28/40] Add vupk{h,l}s{b,h} instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 29/40] Add GEN_VAFORM_PAIRED macro for subsequent instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 30/40] Add vmsum{u,m}bm instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 31/40] Add vsel and vperm instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 32/40] Add saturating arithmetic conversion functions for subsequent instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 33/40] Add vpks{h, w}{s, u}s, vpku{h, w}us, and vpku{h, w}um instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 34/40] Add vpkpx instruction Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 35/40] Add vmh{,r}addshs instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 36/40] Add vmsumuh{m,s} instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 37/40] Add vmsumsh{m,s} instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 38/40] Add vmladduhm instruction Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 39/40] Add {l,st}ve{b,h,w}x instructions Nathan Froyd
2008-12-31  3:10 ` [Qemu-devel] [PATCH 40/40] Add vsumsws, vsum2sws, and vsum4{sbs, shs, ubs} instructions Nathan Froyd
2009-01-04 22:20 ` [Qemu-devel] [PATCH] target-ppc: add integer Altivec instructions, take 2 Aurelien Jarno
2009-01-04 22:54   ` 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=20090107212502.GE28711@codesourcery.com \
    --to=froydnj@codesourcery.com \
    --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 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.