From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: qemu-arm@nongnu.org
Subject: Re: [PATCH 00/61] target/arm: Implement FEAT_SME2
Date: Mon, 24 Feb 2025 12:35:26 -0800 [thread overview]
Message-ID: <1824fbe1-33a1-40db-8e9a-afc00f13d2b4@linaro.org> (raw)
In-Reply-To: <9e0853e9-b590-4a84-8214-fbb543bd4359@linaro.org>
On 2/24/25 12:27, Richard Henderson wrote:
> On 2/6/25 11:56, Richard Henderson wrote:
>> Based-on:20250201164012.1660228-1-peter.maydell@linaro.org
>> ("[PATCH v2 00/69] target/arm: FEAT_AFP and FEAT_RPRES")
>>
>> This implements the Scalar Matrix Extensions, version 2, plus two
>> trivial extensions for float16 and bfloat16.
>>
>> This hasn't been tested much at all; I need to either get FVP up and
>> running for RISU comparison, or write some stand-alone test cases.
>> But in the meantime this could use some eyes.
>>
>> SME2 is the first vector-like extension we've had that has dynamic
>> indexing of registers: ZArray[(rv + offset) % svl], where RV is a
>> general register. So the first thing I do is extend TCG's gvec
>> support to handle TCGv_ptr base + offset addressing. I only changed
>> enough to handle what I needed within SME2; changing it all would be
>> a big job, and it would (at least for the moment) remain unused.
>>
>> Still to-do are few more extensions for SME2p1.
>
> Ho hum. I've missed the entire set of counted predicate insns.
> Hooray for Arm sample code!
Oh, duh -- they're in SME2p1, which is I guess not as optional as I assumed.
r~
prev parent reply other threads:[~2025-02-24 20:36 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-06 19:56 [PATCH 00/61] target/arm: Implement FEAT_SME2 Richard Henderson
2025-02-06 19:56 ` [PATCH 01/61] tcg: Add dbase argument to do_dup_store Richard Henderson
2025-02-06 19:56 ` [PATCH 02/61] tcg: Add dbase argument to do_dup Richard Henderson
2025-02-06 19:56 ` [PATCH 03/61] tcg: Add dbase argument to expand_clr Richard Henderson
2025-02-06 19:56 ` [PATCH 04/61] tcg: Add base arguments to check_overlap_[234] Richard Henderson
2025-02-06 19:56 ` [PATCH 05/61] tcg: Split out tcg_gen_gvec_2_var Richard Henderson
2025-02-06 19:56 ` [PATCH 06/61] tcg: Split out tcg_gen_gvec_3_var Richard Henderson
2025-02-06 19:56 ` [PATCH 07/61] tcg: Split out tcg_gen_gvec_mov_var Richard Henderson
2025-02-06 19:56 ` [PATCH 08/61] tcg: Split out tcg_gen_gvec_{add,sub}_var Richard Henderson
2025-02-06 19:56 ` [PATCH 09/61] target/arm: Introduce FPST_ZA, FPST_ZA_F16 Richard Henderson
2025-02-06 19:56 ` [PATCH 10/61] target/arm: Use FPST_ZA for sme_fmopa_[hsd] Richard Henderson
2025-02-06 19:56 ` [PATCH 11/61] target/arm: Rename zarray to za_state.za Richard Henderson
2025-02-06 19:56 ` [PATCH 12/61] target/arm: Add isar_feature_aa64_sme2* Richard Henderson
2025-02-06 19:56 ` [PATCH 13/61] target/arm: Add ZT0 Richard Henderson
2025-02-06 19:56 ` [PATCH 14/61] target/arm: Add zt0_excp_el to DisasContext Richard Henderson
2025-02-06 19:56 ` [PATCH 15/61] target/arm: Implement SME2 ZERO ZT0 Richard Henderson
2025-02-06 19:56 ` [PATCH 16/61] target/arm: Implement SME2 LDR/STR ZT0 Richard Henderson
2025-02-06 19:56 ` [PATCH 17/61] target/arm: Implement SME2 MOVT Richard Henderson
2025-02-06 19:56 ` [PATCH 18/61] target/arm: Split get_tile_rowcol argument tile_index Richard Henderson
2025-02-06 19:56 ` [PATCH 19/61] target/arm: Rename MOVA for translate Richard Henderson
2025-02-06 19:56 ` [PATCH 20/61] target/arm: Implement SME2 MOVA to/from tile, multiple registers Richard Henderson
2025-02-06 19:56 ` [PATCH 21/61] target/arm: Split out get_zarray Richard Henderson
2025-02-06 19:56 ` [PATCH 22/61] target/arm: Implement SME2 MOVA to/from array, multiple registers Richard Henderson
2025-02-06 19:56 ` [PATCH 23/61] target/arm: Implement SME2 BMOPA Richard Henderson
2025-02-06 19:56 ` [PATCH 24/61] target/arm: Implement SME2 SMOPS, UMOPS (2-way) Richard Henderson
2025-02-06 19:56 ` [PATCH 25/61] target/arm: Introduce gen_gvec_sve2_sqdmulh Richard Henderson
2025-02-06 19:56 ` [PATCH 26/61] target/arm: Implement SME2 Multiple and Single SVE Destructive Richard Henderson
2025-02-06 19:56 ` [PATCH 27/61] target/arm: Implement SME2 Multiple Vectors " Richard Henderson
2025-02-06 19:56 ` [PATCH 28/61] target/arm: Implement SME2 ADD/SUB (array results, multiple and single vector) Richard Henderson
2025-02-06 19:56 ` [PATCH 29/61] target/arm: Implement SME2 ADD/SUB (array results, multiple vectors) Richard Henderson
2025-02-06 19:56 ` [PATCH 30/61] target/arm: Pass ZA to helper_sve2_fmlal_zz[zx]w_s Richard Henderson
2025-02-06 19:56 ` [PATCH 31/61] target/arm: Implement SME2 FMLAL, BFMLAL Richard Henderson
2025-02-06 19:56 ` [PATCH 32/61] target/arm: Implement SME2 FDOT Richard Henderson
2025-02-06 19:56 ` [PATCH 33/61] target/arm: Implement SME2 BFDOT Richard Henderson
2025-02-06 19:56 ` [PATCH 34/61] target/arm: Implement SME2 FVDOT, BFVDOT Richard Henderson
2025-02-06 19:56 ` [PATCH 35/61] target/arm: Rename helper_gvec_*dot_[bh] to *_4[bh] Richard Henderson
2025-02-06 19:56 ` [PATCH 36/61] target/arm: Remove helper_gvec_sudot_idx_4b Richard Henderson
2025-02-06 19:56 ` [PATCH 37/61] target/arm: Implemement SME2 SDOT, UDOT, USDOT, SUDOT Richard Henderson
2025-02-06 19:56 ` [PATCH 38/61] target/arm: Implement SME2 SVDOT, UVDOT, SUVDOT, USVDOT Richard Henderson
2025-02-06 19:56 ` [PATCH 39/61] target/arm: Implement SME2 SMLAL, SMLSL, UMLAL, UMLSL Richard Henderson
2025-02-06 19:56 ` [PATCH 40/61] target/arm: Implement SME2 SMLALL, SMLSLL, UMLALL, UMLSLL Richard Henderson
2025-02-06 19:56 ` [PATCH 41/61] target/arm: Rename gvec_fml[as]_[hs] with _nf_ infix Richard Henderson
2025-02-06 19:56 ` [PATCH 42/61] target/arm: Implement SME2 FMLA, FMLS Richard Henderson
2025-02-06 19:56 ` [PATCH 43/61] target/arm: Implement SME2 BFMLA, BFMLS Richard Henderson
2025-02-06 19:56 ` [PATCH 44/61] target/arm: Implement SME2 FADD, FSUB, BFADD, BFSUB Richard Henderson
2025-02-06 19:56 ` [PATCH 45/61] target/arm: Remove CPUARMState.vfp.scratch Richard Henderson
2025-02-06 19:57 ` [PATCH 46/61] target/arm: Implement SME2 BFCVT, BFCVTN, FCVT, FCVTN Richard Henderson
2025-02-06 19:57 ` [PATCH 47/61] target/arm: Implement SME2 FCVT (widening), FCVTL Richard Henderson
2025-02-06 19:57 ` [PATCH 48/61] target/arm: Implement SME2 FCVTZS, FCVTZU Richard Henderson
2025-02-06 19:57 ` [PATCH 49/61] target/arm: Implement SME2 SCVTF, UCVTF Richard Henderson
2025-02-06 19:57 ` [PATCH 50/61] target/arm: Implement SME2 FRINTN, FRINTP, FRINTM, FRINTA Richard Henderson
2025-02-06 19:57 ` [PATCH 51/61] target/arm: Introduce do_[us]sat_[bhs] macros Richard Henderson
2025-02-06 19:57 ` [PATCH 52/61] target/arm: Use do_[us]sat_[bhs] in sve_helper.c Richard Henderson
2025-02-06 19:57 ` [PATCH 53/61] target/arm: Implement SME2 SQCVT, UQCVT, SQCVTU Richard Henderson
2025-02-06 19:57 ` [PATCH 54/61] target/arm: Implement SME2 SUNPK, UUNPK Richard Henderson
2025-02-06 19:57 ` [PATCH 55/61] target/arm: Implement SME2 ZIP, UZP (four registers) Richard Henderson
2025-02-06 19:57 ` [PATCH 56/61] target/arm: Move do_urshr, do_srshr to vec_internal.h Richard Henderson
2025-02-06 19:57 ` [PATCH 57/61] target/arm: Implement SME2 SQRSHR, UQRSHR, SQRSHRN Richard Henderson
2025-02-06 19:57 ` [PATCH 58/61] target/arm: Implement SME2 ZIP, UZP (two registers) Richard Henderson
2025-02-06 19:57 ` [PATCH 59/61] target/arm: Implement SME2 FCLAMP, SCLAMP, UCLAMP Richard Henderson
2025-02-06 19:57 ` [PATCH 60/61] target/arm: Implement SME2 SEL Richard Henderson
2025-02-06 19:57 ` [PATCH 61/61] target/arm: Enable FEAT_SME2, FEAT_SME_F16F16, FEAT_SVE_B16B16 on -cpu max Richard Henderson
2025-02-24 20:27 ` [PATCH 00/61] target/arm: Implement FEAT_SME2 Richard Henderson
2025-02-24 20:35 ` Richard Henderson [this message]
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=1824fbe1-33a1-40db-8e9a-afc00f13d2b4@linaro.org \
--to=richard.henderson@linaro.org \
--cc=qemu-arm@nongnu.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).