From: joserz@linux.vnet.ibm.com
To: David Gibson <david@gibson.dropbear.id.au>
Cc: qemu-ppc@nongnu.org, qemu-devel@nongnu.org,
nikunj@linux.vnet.ibm.com, bharata@linux.vnet.ibm.com
Subject: Re: [Qemu-devel] [PATCH] 2/4] target-ppc: Implement bcdctn. instruction
Date: Thu, 27 Oct 2016 14:13:13 -0200 [thread overview]
Message-ID: <20161027161313.GB19890@pacoca> (raw)
In-Reply-To: <20161027012017.GB19918@umbus.fritz.box>
On Thu, Oct 27, 2016 at 12:20:17PM +1100, David Gibson wrote:
> On Wed, Oct 26, 2016 at 11:18:56AM -0200, Jose Ricardo Ziviani wrote:
> > bcdctn. converts from BCD to National numeric format. National format
> > uses a byte to represent a digit where the most significant nibble is
> > always 0x3 and the least sign. nibbles is the digit itself.
> >
> > Signed-off-by: Jose Ricardo Ziviani <joserz@linux.vnet.ibm.com>
> > ---
> > target-ppc/helper.h | 1 +
> > target-ppc/int_helper.c | 46 +++++++++++++++++++++++++++++++++++++
> > target-ppc/translate/vmx-impl.inc.c | 24 ++++++++++++++++++-
> > 3 files changed, 70 insertions(+), 1 deletion(-)
> >
> > diff --git a/target-ppc/helper.h b/target-ppc/helper.h
> > index d30ec60..92eaaf0 100644
> > --- a/target-ppc/helper.h
> > +++ b/target-ppc/helper.h
> > @@ -370,6 +370,7 @@ DEF_HELPER_4(vpermxor, void, avr, avr, avr, avr)
> > DEF_HELPER_4(bcdadd, i32, avr, avr, avr, i32)
> > DEF_HELPER_4(bcdsub, i32, avr, avr, avr, i32)
> > DEF_HELPER_3(bcdcfn, i32, avr, avr, i32)
> > +DEF_HELPER_2(bcdctn, i32, avr, avr)
> >
> > DEF_HELPER_2(xsadddp, void, env, i32)
> > DEF_HELPER_2(xssubdp, void, env, i32)
> > diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
> > index 494c74e..cffe82c 100644
> > --- a/target-ppc/int_helper.c
> > +++ b/target-ppc/int_helper.c
> > @@ -2494,6 +2494,17 @@ static uint8_t get_national_digit(ppc_avr_t *reg, int n)
> > #endif
> > }
> >
> > +static void set_national_digit(ppc_avr_t *reg, uint8_t val, int n)
> > +{
> > +#if defined(HOST_WORDS_BIGENDIAN)
> > + reg->u16[8 - n] &= 0;
> > + reg->u16[8 - n] |= val;
>
> The &= always sets the value to 0, so you might as well just use a
> plain assignment in place of the &=, |=.
OK! I'll fix it in v2.
>
> > +#else
> > + reg->u16[n] &= 0;
> > + reg->u16[n] |= val;
> > +#endif
> > +}
> > +
> > static int bcd_cmp_mag(ppc_avr_t *a, ppc_avr_t *b)
> > {
> > int i;
> > @@ -2667,6 +2678,41 @@ uint32_t helper_bcdcfn(ppc_avr_t *r, ppc_avr_t *b, uint32_t ps)
> > return cr;
> > }
> >
> > +uint32_t helper_bcdctn(ppc_avr_t *r, ppc_avr_t *b)
> > +{
> > + int i;
> > + int cr = 0;
> > + int invalid = 0;
> > + int sgnb = bcd_get_sgn(b);
> > + ppc_avr_t ret = { .u64 = { 0, 0 } };
> > +
> > + int eq_flag = (b->u64[HI_IDX] == 0) && ((b->u64[LO_IDX] >> 4) == 0);
> > + int ox_flag = (b->u64[HI_IDX] != 0) || ((b->u64[LO_IDX] >> 8) != 0);
>
> This looks wrong. You're shifing the low half right 8 bits == 2
> nybbles == 1 digit + sign. So this will set the overflow flag if your
> input is a number of >1 digit. I think you want >>32, so it only sets
> overflow if the input exceeds 7 decimal digits + sign.
OK! I'll fix it in v2.
>
> > + for (i = 1; i < 8; i++) {
> > + set_national_digit(&ret, 0x30 + bcd_get_digit(b, i, &invalid), i);
> > + }
> > + set_national_digit(&ret, (sgnb == -1) ? NATIONAL_NEG : NATIONAL_PLUS, 0);
> > +
> > + if (!eq_flag) {
> > + cr = (sgnb == -1) ? 1 << CRF_LT : 1 << CRF_GT;
> > + } else {
> > + cr = 1 << CRF_EQ;
> > + }
> > +
> > + if (ox_flag) {
> > + cr |= 1 << CRF_SO;
> > + }
> > +
> > + if (unlikely(invalid)) {
> > + cr = 1 << CRF_SO;
> > + }
> > +
> > + *r = ret;
> > +
> > + return cr;
> > +}
> > +
> > void helper_vsbox(ppc_avr_t *r, ppc_avr_t *a)
> > {
> > int i;
> > diff --git a/target-ppc/translate/vmx-impl.inc.c b/target-ppc/translate/vmx-impl.inc.c
> > index 2abdcac..4364881 100644
> > --- a/target-ppc/translate/vmx-impl.inc.c
> > +++ b/target-ppc/translate/vmx-impl.inc.c
> > @@ -894,9 +894,29 @@ static void gen_##op(DisasContext *ctx) \
> > tcg_temp_free_i32(ps); \
> > }
> >
> > +#define GEN_BCD3(op) \
> > +static void gen_##op(DisasContext *ctx) \
> > +{ \
> > + TCGv_ptr rb, rd; \
> > + \
> > + if (unlikely(!ctx->altivec_enabled)) { \
> > + gen_exception(ctx, POWERPC_EXCP_VPU); \
> > + return; \
> > + } \
> > + \
> > + rb = gen_avr_ptr(rB(ctx->opcode)); \
> > + rd = gen_avr_ptr(rD(ctx->opcode)); \
> > + \
> > + gen_helper_##op(cpu_crf[6], rd, rb); \
> > + \
> > + tcg_temp_free_ptr(rb); \
> > + tcg_temp_free_ptr(rd); \
> > +}
> > GEN_BCD(bcdadd)
> > GEN_BCD(bcdsub)
> > GEN_BCD2(bcdcfn)
> > +GEN_BCD3(bcdctn)
> >
> > static void gen_xpnd04_1(DisasContext *ctx)
> > {
> > @@ -908,7 +928,8 @@ static void gen_xpnd04_1(DisasContext *ctx)
> > case 4:
> > break; /* bcdctz. */
> > case 5:
> > - break; /* bcdctn. */
> > + gen_bcdctn(ctx);
> > + break;
> > case 6:
> > break; /* bcdcfz. */
> > case 7:
>
> Uh.. doesn't adding bcdctn to this make it identical to gen_xpnd04_2?
Actually this opcode is invalid for bcdctn as per ISA3.0.
Do you think that I should throw something like a SIGILL for this particular case?
>
> > @@ -1024,3 +1045,4 @@ GEN_VXFORM_DUAL(vsldoi, PPC_ALTIVEC, PPC_NONE,
> > #undef GEN_VAFORM_PAIRED
> >
> > #undef GEN_BCD2
> > +#undef GEN_BCD3
>
> --
> David Gibson | I'll have my music baroque, and my code
> david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
> | _way_ _around_!
> http://www.ozlabs.org/~dgibson
next prev parent reply other threads:[~2016-10-27 16:13 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-10-26 13:18 [Qemu-devel] [PATCH] 0/4] POWER9 TCG enablements - BCD functions part I Jose Ricardo Ziviani
2016-10-26 13:18 ` [Qemu-devel] [PATCH] 1/4] target-ppc: Implement bcdcfn. instruction Jose Ricardo Ziviani
2016-10-27 1:05 ` David Gibson
2016-10-27 15:29 ` joserz
2016-10-26 13:18 ` [Qemu-devel] [PATCH] 2/4] target-ppc: Implement bcdctn. instruction Jose Ricardo Ziviani
2016-10-27 1:20 ` David Gibson
2016-10-27 16:13 ` joserz [this message]
2016-10-31 0:16 ` David Gibson
2016-10-26 13:18 ` [Qemu-devel] [PATCH] 3/4] target-ppc: Implement bcdcfz. instruction Jose Ricardo Ziviani
2016-10-27 1:35 ` David Gibson
2016-10-27 18:02 ` joserz
2016-10-26 13:18 ` [Qemu-devel] [PATCH] 4/4] target-ppc: Implement bcdctz. instruction Jose Ricardo Ziviani
2016-10-27 1:47 ` David Gibson
2016-10-27 18:29 ` joserz
2016-10-27 20:08 ` [Qemu-devel] [Qemu-ppc] " joserz
2016-10-31 2:50 ` David Gibson
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=20161027161313.GB19890@pacoca \
--to=joserz@linux.vnet.ibm.com \
--cc=bharata@linux.vnet.ibm.com \
--cc=david@gibson.dropbear.id.au \
--cc=nikunj@linux.vnet.ibm.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@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).