From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.io!.POSTED.blaine.gmane.org!not-for-mail From: Yuan Fu Newsgroups: gmane.emacs.devel Subject: Re: How to add pseudo vector types Date: Thu, 22 Jul 2021 13:47:20 -0400 Message-ID: References: <83h7gw6pyj.fsf@gnu.org> <45EBF16A-C953-42C7-97D1-3A2BFEF7DD01@gmail.com> <83y2a764oy.fsf@gnu.org> <83v95b60fn.fsf@gnu.org> <00DD5BFE-D14E-449A-9319-E7B725DEBFB3@gmail.com> <83r1fz5xr9.fsf@gnu.org> <1AAB1BCC-362B-4249-B785-4E0530E15C60@gmail.com> <83czri67h0.fsf@gnu.org> <46BBFF88-76C3-4818-8805-5437409BEA93@gmail.com> <83wnpq46uk.fsf@gnu.org> <533BD53B-4E85-4E9E-B46A-346A5BBAD0F5@gmail.com> <258CB68D-1CC1-42C8-BDCD-2A8A8099B783@gmail.com> <1a776770-50b7-93cd-6591-c9a5b3a56eb8@gmail.com> <8335s64v10.fsf@gnu.org> <5380C92B-6C15-4490-A1E0-1C3132DBB16A@gmail.com> <83k0li2shw.fsf@gnu.org> Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.60.0.2.21\)) Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Injection-Info: ciao.gmane.io; posting-host="blaine.gmane.org:116.202.254.214"; logging-data="28926"; mail-complaints-to="usenet@ciao.gmane.io" Cc: =?utf-8?Q?Cl=C3=A9ment_Pit-Claudel?= , Stefan Monnier , emacs-devel@gnu.org To: Eli Zaretskii Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane-mx.org@gnu.org Thu Jul 22 19:49:02 2021 Return-path: Envelope-to: ged-emacs-devel@m.gmane-mx.org Original-Received: from lists.gnu.org ([209.51.188.17]) by ciao.gmane.io with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1m6cof-0007Je-G4 for ged-emacs-devel@m.gmane-mx.org; Thu, 22 Jul 2021 19:49:01 +0200 Original-Received: from localhost ([::1]:45202 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1m6coe-0007ve-Ef for ged-emacs-devel@m.gmane-mx.org; Thu, 22 Jul 2021 13:49:00 -0400 Original-Received: from eggs.gnu.org ([2001:470:142:3::10]:58852) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1m6cn8-0005jN-LI for emacs-devel@gnu.org; Thu, 22 Jul 2021 13:47:27 -0400 Original-Received: from mail-qt1-x834.google.com ([2607:f8b0:4864:20::834]:33461) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1m6cn6-0006ue-4E; Thu, 22 Jul 2021 13:47:26 -0400 Original-Received: by mail-qt1-x834.google.com with SMTP id q15so4825342qtp.0; Thu, 22 Jul 2021 10:47:23 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=85QjhDgh/wXkJGmMrYVH8RFuF4Pd9HBlxsnLxHxujSY=; b=i3gBEshH878lZeruLNiMT1VzbvfmhMaeKAm8Rw4CQJPZBHWEoT+U1gBoTDRYkhz6iY kyAkyjLSLTo3ub5i0E6rbMVGVbZSVeWpLrApCZ/N43WxWUWyELY+RNJHo4QLqtG4q7z1 mZOjV9A9gvar/WIiCjn4tZpSqvxGZpyIFgFaf0mu8CUK/Jmp1B/7hW8+z8fNsb2UHK3p XBvHXL4vZyXamo2HCV+R7kfH6wOwgvZDrBl/ouaHsrg9bMDZvZLJ9YYg2WyRArYzcfyx Ood+75nVIVZuFgGvB8V5riBkzRiB5Nl2h5y/uCy1MwHTk+j8Y0YqF5xmfQq6RQyR8K8s 74Zw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=85QjhDgh/wXkJGmMrYVH8RFuF4Pd9HBlxsnLxHxujSY=; b=PZ+shtX4a4SZtR0ieYFlU8SHrJfQGhEb/o9fY6iA7wqvUH2htwzFvKzIkNM0kyl1lo wIO8k4FzKHo8UlokO6PY5te4ZLYas/f+rp4gZCya4cUiJ7tzvRxGpiZ3U6sRnV1XFF1F fq2ax2FHuG8v2ZUmvlY43HlvglsrbltjxN1L33gQkdIG84Uu8S8NKdGfarve5hNaKCDw SKxTGMKafGwdxtlf8yhd43bxs/GJjUFM2RGWxcxMp0ImXhp9GYdIdH7E/0srklFZ0KUW pIOCe+/aB2aV/quAW/mPAbfpDViGJtrMgvT8/pksAGpHiUrmrZrCiDmsf97dG/BfZHNT cv1w== X-Gm-Message-State: AOAM531c3EdyTd++6RBa4wDJcarp+AIAaJaMZSEhHDM3dmgWHqnxF54J d834yBRJxMTdpkfHUPuQVZHfhn+RIoDa9Q== X-Google-Smtp-Source: ABdhPJz46K2MkxYkjICkPLYwzMms5g/qQkm3UH+KQ+KvwviGVaLRzfHl712pvIHDSF21f5e/3O8KvQ== X-Received: by 2002:ac8:544:: with SMTP id c4mr702677qth.299.1626976042214; Thu, 22 Jul 2021 10:47:22 -0700 (PDT) Original-Received: from 2603-7080-0302-635e-501b-d938-5abe-a221.res6.spectrum.com (2603-7080-0302-635e-501b-d938-5abe-a221.res6.spectrum.com. [2603:7080:302:635e:501b:d938:5abe:a221]) by smtp.gmail.com with ESMTPSA id m65sm2394754qkc.87.2021.07.22.10.47.21 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Thu, 22 Jul 2021 10:47:21 -0700 (PDT) In-Reply-To: <83k0li2shw.fsf@gnu.org> X-Mailer: Apple Mail (2.3654.60.0.2.21) Received-SPF: pass client-ip=2607:f8b0:4864:20::834; envelope-from=casouri@gmail.com; helo=mail-qt1-x834.google.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane-mx.org@gnu.org Original-Sender: "Emacs-devel" Xref: news.gmane.io gmane.emacs.devel:271472 Archived-At: > On Jul 22, 2021, at 1:00 PM, Eli Zaretskii wrote: >=20 >> From: Yuan Fu >> Date: Thu, 22 Jul 2021 09:47:45 -0400 >> Cc: Stefan Monnier , >> Cl=C3=A9ment Pit-Claudel , >> emacs-devel@gnu.org >>=20 >> Yes, I meant to discuss this. The problem with respecting narrowing = is that, a user can freely narrow and widen arbitrarily, and Emacs needs = to translate them into insertion & deletion of the buffer text for = tree-sitter, every time a user narrows or widens the buffer. Plus, if = tree-sitter respects narrowing, it could happen where a user narrows the = buffer, the font-locking changes and is not correct anymore. Maybe = that=E2=80=99s not the user want. Also, if someone narrows and widens = often, maybe narrow to a function for better focus, tree-sitter needs to = constantly re-parse most of the buffer. These are not significant = disadvantages, but what do we get from respecting narrowing that = justifies code complexity and these small annoyances? >=20 > But that's how the current font-lock and indentation work: they never > look beyond the narrowing limits. So why should the TS-based features > behave differently? >=20 > As for temporary narrowing: if we record the changes, but don't send > them to TS until we actually need re-parsing, then we could eliminate > the temporary narrowing when we report the changes to TS, leaving only > the narrowing that exists at the time of the re-parse. At least for > fontifications, that time is redisplay time, and users do expect to > see the text fontified according to the current narrowing. >=20 >>>> *bytes_read =3D (uint32_t) len; >>>=20 >>> Is using uint32_t the restriction of tree-sitter? Doesn't it = support >>> reading more than 2 gigabytes? >>=20 >> I=E2=80=99m not sure why it asks for uint32 specifically, but = that=E2=80=99s what it asks for its api. I don=E2=80=99t think you are = supposed to use tree-sitter on files of size of gigabytes, because the = author mentioned that tree-sitter uses over 10x as much memory as the = size of the source file [1]. On files larger than a couple of megabytes, = I think we better turn off tree-sitter. Normally those files are not = regular source files, anyway, and we don=E2=80=99t need a parse tree for = a log. >=20 > I don't necessarily agree with the "not regular source files" part. > For example, JSON files can be quite large. And there are also log > files, which are even larger -- did no one adapt TS to fontifying > those yet? There is a JSON parser, but I don=E2=80=99t think there is one for log = files. >=20 > More generally: is the problem real? If you make a file that is 1000 > copies of xdisp.c, and then submit it to TS, do you really get 10GB of > memory consumption? This is something that is good to know up front, > so we'd know what to expect down the road. Yes. I concatenated 100 xdisp.c together, and parsed them with my simple = C program. It used 1.8 G. I didn=E2=80=99t test for 1000 together, but I = think the trend is linear. time -l ./main-large-c 16.48 real 15.32 user 0.81 sys 1883959296 maximum resident set size 0 average shared memory size 0 average unshared data size 0 average unshared stack size 459951 page reclaims 22 page faults 0 swaps 0 block input operations 0 block output operations 0 messages sent 0 messages received 0 signals received 6 voluntary context switches 1653 involuntary context switches 107310143182 instructions retired 58561420060 cycles elapsed 1883095040 peak memory footprint >=20 >> That leads to another point. I suspect the memory limit will come = before the speed limit, i.e., as the file size increases, the memory = consumption will become unacceptable before the speed does. So it is = possible that we want to outright disable tree-sitter for larger files, = then we don=E2=80=99t need to do much to improve the responsiveness of = tree-sitter on large files. And we might want to delete the parse tree = if a buffer has been idle for a while. Of course, that=E2=80=99s just my = superstition, we=E2=80=99ll see once we can measure the performance. >=20 > See above: IMO, we should benchmark both the CPU and memory > performance of TS for such large files, before we decide on the course > of action. That=E2=80=99s my thought, too. I should have reserved my suspicion = until I have benchmark measurements. >=20 >>>> +DEFUN ("tree-sitter-node-type", >>>> + Ftree_sitter_node_type, Stree_sitter_node_type, 1, 1, 0, >>>> + doc: /* Return the NODE's type as a symbol. */) >>>> + (Lisp_Object node) >>>> +{ >>>> + CHECK_TS_NODE (node); >>>> + TSNode ts_node =3D XTS_NODE (node)->node; >>>> + const char *type =3D ts_node_type(ts_node); >>>> + return intern_c_string (type); >>>=20 >>> Why do we need to intern the string each time? can't we store the >>> interned symbol there, instead of a C string, in the first place? >>=20 >> I=E2=80=99m not sure what do you mean by =E2=80=9Cstore the interned = symbol there=E2=80=9D, where do I store the interned symbol? >=20 > In the struct that ts_node_type accesses, instead of the 'char *' > string you store there now. The struct that ts_node_type accesses is a TSNode, which is defined by = tree-sitter. ts_node_type is an API provided by tree-sitter, I=E2=80=99m = just exposing it to lisp. I could return strings instead of symbols, but = I thought symbols might be more appropriate and more convenient for = users of this function.=20 >> (BTW, If you see something wrong, that=E2=80=99s probably because I = don=E2=80=99t know the right way to do it, and grepping only got me that = far.) >=20 > Do what? feel free to ask questions when you aren't sure how to > accomplish something on the C level. Thanks. Is below the correct way to set a buffer-local variable? (I=E2=80=99= m setting tree-sitter-parser-list.) struct buffer *old_buffer =3D current_buffer; set_buffer_internal (XBUFFER (buffer)); Fset (Qtree_sitter_parser_list, Fcons (lisp_parser, Fsymbol_value (Qtree_sitter_parser_list))); set_buffer_internal (old_buffer); Also, we don=E2=80=99t call change hooks in replace_range_2, why? Should = I update tree-sitter trees in that function, or should I not? Yuan=