All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: Rajalakshmi Srinivasaraghavan <raji@linux.vnet.ibm.com>
Cc: qemu-ppc@nongnu.org, rth@twiddle.net, qemu-devel@nongnu.org,
	nikunj@linux.vnet.ibm.com, benh@kernel.crashing.org
Subject: Re: [Qemu-devel] [PATCH v3 2/5] target-ppc: add vector extract instructions
Date: Tue, 16 Aug 2016 14:21:35 +1000	[thread overview]
Message-ID: <20160816042135.GG14530@voom.fritz.box> (raw)
In-Reply-To: <1470901008-3284-3-git-send-email-raji@linux.vnet.ibm.com>

[-- Attachment #1: Type: text/plain, Size: 6773 bytes --]

On Thu, Aug 11, 2016 at 01:06:45PM +0530, Rajalakshmi Srinivasaraghavan wrote:
> The following vector extract instructions are added from ISA 3.0.
> 
> vextractub - Vector Extract Unsigned Byte
> vextractuh - Vector Extract Unsigned Halfword
> vextractuw - Vector Extract Unsigned Word
> vextractd - Vector Extract Unsigned Doubleword
> 
> Signed-off-by: Rajalakshmi Srinivasaraghavan <raji@linux.vnet.ibm.com>
> ---
>  target-ppc/helper.h             |    4 ++++
>  target-ppc/int_helper.c         |   26 ++++++++++++++++++++++++++
>  target-ppc/translate/vmx-impl.c |   10 ++++++++++
>  target-ppc/translate/vmx-ops.c  |   10 +++++++---
>  4 files changed, 47 insertions(+), 3 deletions(-)
> 
> diff --git a/target-ppc/helper.h b/target-ppc/helper.h
> index 0923779..59e7b88 100644
> --- a/target-ppc/helper.h
> +++ b/target-ppc/helper.h
> @@ -250,6 +250,10 @@ DEF_HELPER_2(vspltisw, void, avr, i32)
>  DEF_HELPER_3(vspltb, void, avr, avr, i32)
>  DEF_HELPER_3(vsplth, void, avr, avr, i32)
>  DEF_HELPER_3(vspltw, void, avr, avr, i32)
> +DEF_HELPER_3(vextractub, void, avr, avr, i32)
> +DEF_HELPER_3(vextractuh, void, avr, avr, i32)
> +DEF_HELPER_3(vextractuw, void, avr, avr, i32)
> +DEF_HELPER_3(vextractd, void, avr, avr, i32)
>  DEF_HELPER_3(vinsertb, void, avr, avr, i32)
>  DEF_HELPER_3(vinserth, void, avr, avr, i32)
>  DEF_HELPER_3(vinsertw, void, avr, avr, i32)
> diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
> index 3f8e439..a917bd5 100644
> --- a/target-ppc/int_helper.c
> +++ b/target-ppc/int_helper.c
> @@ -1819,6 +1819,32 @@ VINSERT(h, u16, 3)
>  VINSERT(w, u32, 1)
>  VINSERT(d, u64, 0)
>  #undef VINSERT
> +#if defined(HOST_WORDS_BIGENDIAN)
> +#define VEXTRACT(suffix, element, index)                                     \
> +    void helper_vextract##suffix(ppc_avr_t *r, ppc_avr_t *b, uint32_t splat) \

splat is not a good parameter name here, since the element is being
extracted, rather than splatted.

> +    {                                                                        \
> +        uint32_t s = sizeof(r->element[0]) * index;                          \
> +        ppc_avr_t result = { .u64 = { 0, 0 } };                              \
> +        memcpy(&result.element[index], &b->u8[splat],                        \
> +               sizeof(result.element[0]));                                   \
> +        *r = result;                                                         \
> +    }
> +#else
> +#define VEXTRACT(suffix, element, index)                                     \
> +    void helper_vextract##suffix(ppc_avr_t *r, ppc_avr_t *b, uint32_t splat) \
> +    {                                                                        \
> +        ppc_avr_t result = { .u64 = { 0, 0 } };                              \
> +        uint32_t s = (16 - splat) - sizeof(r->element[0]);                   \
> +        uint32_t d = (ARRAY_SIZE(r->element) - index) - 1;
> \

Same comments on the index value as for vinsert*.

> +        memcpy(&result.element[d], &b->u8[s], sizeof(result.element[0]));    \
> +        *r = result;                                                         \
> +    }
> +#endif
> +VEXTRACT(ub, u8, 7)
> +VEXTRACT(uh, u16, 3)
> +VEXTRACT(uw, u32, 1)
> +VEXTRACT(d, u64, 0)
> +#undef VEXTRACT
>  
>  #define VSPLTI(suffix, element, splat_type)                     \
>      void helper_vspltis##suffix(ppc_avr_t *r, uint32_t splat)   \
> diff --git a/target-ppc/translate/vmx-impl.c b/target-ppc/translate/vmx-impl.c
> index f6a97ac..766a645 100644
> --- a/target-ppc/translate/vmx-impl.c
> +++ b/target-ppc/translate/vmx-impl.c
> @@ -648,6 +648,10 @@ static void glue(gen_, name)(DisasContext *ctx)                         \
>  GEN_VXFORM_UIMM(vspltb, 6, 8);
>  GEN_VXFORM_UIMM(vsplth, 6, 9);
>  GEN_VXFORM_UIMM(vspltw, 6, 10);
> +GEN_VXFORM_UIMM_SPLAT(vextractub, 6, 8, 15);
> +GEN_VXFORM_UIMM_SPLAT(vextractuh, 6, 9, 14);
> +GEN_VXFORM_UIMM_SPLAT(vextractuw, 6, 10, 12);
> +GEN_VXFORM_UIMM_SPLAT(vextractd, 6, 11, 8);
>  GEN_VXFORM_UIMM_SPLAT(vinsertb, 6, 12, 15);
>  GEN_VXFORM_UIMM_SPLAT(vinserth, 6, 13, 14);
>  GEN_VXFORM_UIMM_SPLAT(vinsertw, 6, 14, 12);
> @@ -656,6 +660,12 @@ GEN_VXFORM_UIMM_ENV(vcfux, 5, 12);
>  GEN_VXFORM_UIMM_ENV(vcfsx, 5, 13);
>  GEN_VXFORM_UIMM_ENV(vctuxs, 5, 14);
>  GEN_VXFORM_UIMM_ENV(vctsxs, 5, 15);
> +GEN_VXFORM_DUAL(vspltb, PPC_NONE, PPC2_ALTIVEC_207,
> +                      vextractub, PPC_NONE, PPC2_ISA300);
> +GEN_VXFORM_DUAL(vsplth, PPC_NONE, PPC2_ALTIVEC_207,
> +                      vextractuh, PPC_NONE, PPC2_ISA300);
> +GEN_VXFORM_DUAL(vspltw, PPC_NONE, PPC2_ALTIVEC_207,
> +                      vextractuw, PPC_NONE, PPC2_ISA300);
>  GEN_VXFORM_DUAL(vspltisb, PPC_NONE, PPC2_ALTIVEC_207,
>                        vinsertb, PPC_NONE, PPC2_ISA300);
>  GEN_VXFORM_DUAL(vspltish, PPC_NONE, PPC2_ALTIVEC_207,
> diff --git a/target-ppc/translate/vmx-ops.c b/target-ppc/translate/vmx-ops.c
> index ca69e56..aafe70b 100644
> --- a/target-ppc/translate/vmx-ops.c
> +++ b/target-ppc/translate/vmx-ops.c
> @@ -197,6 +197,13 @@ GEN_VXRFORM_DUAL(vcmpbfp, vcmpgtsd, 3, 15, PPC_ALTIVEC, PPC_NONE)
>  #define GEN_VXFORM_DUAL_INV(name0, name1, opc2, opc3, inval0, inval1, type) \
>  GEN_OPCODE_DUAL(name0##_##name1, 0x04, opc2, opc3, inval0, inval1, type, \
>                                                                 PPC_NONE)
> +GEN_VXFORM_DUAL_INV(vspltb, vextractub, 6, 8, 0x00000000, 0x100000,
> +                                               PPC2_ALTIVEC_207),
> +GEN_VXFORM_DUAL_INV(vsplth, vextractuh, 6, 9, 0x00000000, 0x100000,
> +                                               PPC2_ALTIVEC_207),
> +GEN_VXFORM_DUAL_INV(vspltw, vextractuw, 6, 10, 0x00000000, 0x100000,
> +                                               PPC2_ALTIVEC_207),
> +GEN_VXFORM_300_EXT(vextractd, 6, 11, 0x100000),
>  GEN_VXFORM_DUAL_INV(vspltisb, vinsertb, 6, 12, 0x00000000, 0x100000,
>                                                 PPC2_ALTIVEC_207),
>  GEN_VXFORM_DUAL_INV(vspltish, vinserth, 6, 13, 0x00000000, 0x100000,
> @@ -226,9 +233,6 @@ GEN_VXFORM_NOA(vrfiz, 5, 9),
>  
>  #define GEN_VXFORM_UIMM(name, opc2, opc3)                               \
>      GEN_HANDLER(name, 0x04, opc2, opc3, 0x00000000, PPC_ALTIVEC)
> -GEN_VXFORM_UIMM(vspltb, 6, 8),
> -GEN_VXFORM_UIMM(vsplth, 6, 9),
> -GEN_VXFORM_UIMM(vspltw, 6, 10),
>  GEN_VXFORM_UIMM(vcfux, 5, 12),
>  GEN_VXFORM_UIMM(vcfsx, 5, 13),
>  GEN_VXFORM_UIMM(vctuxs, 5, 14),

-- 
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

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

  reply	other threads:[~2016-08-16  4:45 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-11  7:36 [Qemu-devel] [PATCH v3 0/5] POWER9 TCG enablement - part3 Rajalakshmi Srinivasaraghavan
2016-08-11  7:36 ` [Qemu-devel] [PATCH v3 1/5] target-ppc: add vector insert instructions Rajalakshmi Srinivasaraghavan
2016-08-16  4:18   ` David Gibson
2016-08-19  5:46     ` Rajalakshmi Srinivasaraghavan
2016-08-23 15:08       ` David Gibson
2016-08-11  7:36 ` [Qemu-devel] [PATCH v3 2/5] target-ppc: add vector extract instructions Rajalakshmi Srinivasaraghavan
2016-08-16  4:21   ` David Gibson [this message]
2016-08-11  7:36 ` [Qemu-devel] [PATCH v3 3/5] target-ppc: add vector count trailing zeros instructions Rajalakshmi Srinivasaraghavan
2016-08-16  4:46   ` David Gibson
2016-08-11  7:36 ` [Qemu-devel] [PATCH v3 4/5] target-ppc: add vector bit permute doubleword instruction Rajalakshmi Srinivasaraghavan
2016-08-16  4:33   ` David Gibson
2016-08-11  7:36 ` [Qemu-devel] [PATCH v3 5/5] target-ppc: add vector permute right indexed instruction Rajalakshmi Srinivasaraghavan
2016-08-16  4:45   ` David Gibson
2016-08-16 10:14     ` Rajalakshmi Srinivasaraghavan

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=20160816042135.GG14530@voom.fritz.box \
    --to=david@gibson.dropbear.id.au \
    --cc=benh@kernel.crashing.org \
    --cc=nikunj@linux.vnet.ibm.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    --cc=raji@linux.vnet.ibm.com \
    --cc=rth@twiddle.net \
    /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.