From nobody Thu Oct 2 19:30:13 2025 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 194DE30148D; Fri, 12 Sep 2025 11:46:35 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757677595; cv=none; b=ZIgTHaYJ/Yb6jJscFPi5Wc+kAaxRzUPDeMM5n5FMDikVqaEmmDu92nhuaJ01sGxQFSHgN2PwY8SFwe8Dcy1oyOhYGBGbMG0gIRSjfEPyagItX4kWGRpcFTsHUGU+PVYX92SMq56uqIXfbrmVMTl9uE9mnY82vzWKrcEmI4UiMXc= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757677595; c=relaxed/simple; bh=Oh6mMFUZNaaDz2lQnoLmBkYKDyszlEG8OhnySt8zzKc=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: Content-Type:MIME-Version; b=ETEvOBpHKMLSwNE1fORcQvBRPNaFOO0vRb96B5uZlqJQDSwIuuDZxQSo+idj98cn+1rK0kyPVQHI/mqkgZ7KpfMzWsh5aRq4GV++KGKxEXLgJeqsv9G4gJEPuTQxV1P1JjwnqGrwpSx7CECsDEiNlHWglLt2QjbzcvlzsbKBuhY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=i+LiSyzw; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="i+LiSyzw" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 89821C113D0; Fri, 12 Sep 2025 11:46:34 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1757677594; bh=Oh6mMFUZNaaDz2lQnoLmBkYKDyszlEG8OhnySt8zzKc=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=i+LiSyzwKblcwydcePDwkp1aMviaFVLTpJPkjpyxppVjV6dU/T0wx0BdBXHHqfefv XFGyInrgHuMpurQQwEqC5i1p4HgacVVtlIb2LP9ek+c4ogk+ey3EbD6RadRjvxBWKg 2j6cY6iR2edbQYu5rkMHSCOx1esQiwzbSQ8wYSgYS1HIb1g/w/MgtgdG+BVgRwWs0S ni55hdvddVHSJWpj1jQCv/OFpDmT7he/cMGaZGLcyFH8m39m8KuOotMxlxDNKDv5px cMYfkT322JN9Ss33nb5enB+GXxaESlnvrQS8hp7XE+CMFOaWt9+wTHp/lzYJzKeUw6 MZeL4ZDRwb6Xg== Received: from mchehab by mail.kernel.org with local (Exim 4.98.2) (envelope-from ) id 1ux2Ea-00000008Rsn-3Xx7; Fri, 12 Sep 2025 13:46:32 +0200 From: Mauro Carvalho Chehab To: Linux Doc Mailing List Cc: Mauro Carvalho Chehab , Jonathan Corbet , Mauro Carvalho Chehab , linux-kernel@vger.kernel.org Subject: [PATCH v5 14/18] tools/docs: sphinx-build-wrapper: allow building PDF files in parallel Date: Fri, 12 Sep 2025 13:46:21 +0200 Message-ID: <357091cecbbce5663fd50f532cde5eb322c41cae.1757677427.git.mchehab+huawei@kernel.org> X-Mailer: git-send-email 2.51.0 In-Reply-To: References: Content-Type: text/plain; charset="utf-8" Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Sender: Mauro Carvalho Chehab Use POSIX jobserver when available or -j to run PDF builds in parallel, restoring pdf build performance. Yet, running it when debugging troubles is a bad idea, so, when calling directly via command line, except if "-j" is splicitly requested, it will serialize the build. With such change, a PDF doc builds now takes around 5 minutes on a Ryzen 9 machine with 32 cpu threads: # Explicitly paralelize both Sphinx and LaTeX pdf builds $ make cleandocs; time scripts/sphinx-build-wrapper pdfdocs -j 33 real 5m17.901s user 15m1.499s sys 2m31.482s # Use POSIX jobserver to paralelize both sphinx-build and LaTeX $ make cleandocs; time make pdfdocs real 5m22.369s user 15m9.076s sys 2m31.419s # Serializes PDF build, while keeping Sphinx parallelized. # it is equivalent of passing -jauto via command line $ make cleandocs; time scripts/sphinx-build-wrapper pdfdocs real 11m20.901s user 13m2.910s sys 1m44.553s Signed-off-by: Mauro Carvalho Chehab --- tools/docs/sphinx-build-wrapper | 158 +++++++++++++++++++++++--------- 1 file changed, 117 insertions(+), 41 deletions(-) diff --git a/tools/docs/sphinx-build-wrapper b/tools/docs/sphinx-build-wrap= per index 465896e628fd..d66bb337ba1f 100755 --- a/tools/docs/sphinx-build-wrapper +++ b/tools/docs/sphinx-build-wrapper @@ -52,6 +52,8 @@ import shutil import subprocess import sys =20 +from concurrent import futures + from lib.python_version import PythonVersion =20 LIB_DIR =3D "../../scripts/lib" @@ -314,6 +316,82 @@ class SphinxBuilder: except (OSError, IOError) as e: print(f"Warning: Failed to copy CSS: {e}", file=3Dsys.stderr) =20 + def build_pdf_file(self, latex_cmd, from_dir, path): + """Builds a single pdf file using latex_cmd""" + try: + subprocess.run(latex_cmd + [path], + cwd=3Dfrom_dir, check=3DTrue) + + return True + except subprocess.CalledProcessError: + return False + + def pdf_parallel_build(self, tex_suffix, latex_cmd, tex_files, n_jobs): + """Build PDF files in parallel if possible""" + builds =3D {} + build_failed =3D False + max_len =3D 0 + has_tex =3D False + + # + # LaTeX PDF error code is almost useless for us: + # any warning makes it non-zero. For kernel doc builds it always r= eturn + # non-zero even when build succeeds. So, let's do the best next th= ing: + # Ignore build errors. At the end, check if all PDF files were bui= lt, + # printing a summary with the built ones and returning 0 if all of + # them were actually built. + # + with futures.ThreadPoolExecutor(max_workers=3Dn_jobs) as executor: + jobs =3D {} + + for from_dir, pdf_dir, entry in tex_files: + name =3D entry.name + + if not name.endswith(tex_suffix): + continue + + name =3D name[:-len(tex_suffix)] + + max_len =3D max(max_len, len(name)) + + has_tex =3D True + + future =3D executor.submit(self.build_pdf_file, latex_cmd, + from_dir, entry.path) + jobs[future] =3D (from_dir, pdf_dir, name) + + for future in futures.as_completed(jobs): + from_dir, pdf_dir, name =3D jobs[future] + + pdf_name =3D name + ".pdf" + pdf_from =3D os.path.join(from_dir, pdf_name) + + try: + success =3D future.result() + + if success and os.path.exists(pdf_from): + pdf_to =3D os.path.join(pdf_dir, pdf_name) + + os.rename(pdf_from, pdf_to) + builds[name] =3D os.path.relpath(pdf_to, self.buil= ddir) + else: + builds[name] =3D "FAILED" + build_failed =3D True + except futures.Error as e: + builds[name] =3D f"FAILED ({repr(e)})" + build_failed =3D True + + # + # Handle case where no .tex files were found + # + if not has_tex: + name =3D "Sphinx LaTeX builder" + max_len =3D max(max_len, len(name)) + builds[name] =3D "FAILED (no .tex file was generated)" + build_failed =3D True + + return builds, build_failed, max_len + def handle_pdf(self, output_dirs): """ Extra steps for PDF output. @@ -324,7 +402,9 @@ class SphinxBuilder: """ builds =3D {} max_len =3D 0 + tex_suffix =3D ".tex" =20 + tex_files =3D [] for from_dir in output_dirs: pdf_dir =3D os.path.join(from_dir, "../pdf") os.makedirs(pdf_dir, exist_ok=3DTrue) @@ -336,55 +416,51 @@ class SphinxBuilder: =20 latex_cmd.extend(shlex.split(self.latexopts)) =20 - tex_suffix =3D ".tex" - - # - # Process each .tex file - # - - has_tex =3D False - build_failed =3D False + # Get a list of tex files to process with os.scandir(from_dir) as it: for entry in it: - if not entry.name.endswith(tex_suffix): - continue + if entry.name.endswith(tex_suffix): + tex_files.append((from_dir, pdf_dir, entry)) =20 - name =3D entry.name[:-len(tex_suffix)] - has_tex =3D True + # + # When using make, this won't be used, as the number of jobs comes + # from POSIX jobserver. So, this covers the case where build comes + # from command line. On such case, serialize by default, except if + # the user explicitly sets the number of jobs. + # + n_jobs =3D 1 =20 - # - # LaTeX PDF error code is almost useless for us: - # any warning makes it non-zero. For kernel doc builds= it - # always return non-zero even when build succeeds. - # So, let's do the best next thing: check if all PDF - # files were built. If they're, print a summary and - # return 0 at the end of this function - # - try: - subprocess.run(latex_cmd + [entry.path], - cwd=3Dfrom_dir, check=3DTrue) - except subprocess.CalledProcessError: - pass + # n_jobs is either an integer or "auto". Only use it if it is a nu= mber + if self.n_jobs: + try: + n_jobs =3D int(self.n_jobs) + except ValueError: + pass =20 - pdf_name =3D name + ".pdf" - pdf_from =3D os.path.join(from_dir, pdf_name) - pdf_to =3D os.path.join(pdf_dir, pdf_name) + # + # When using make, jobserver.claim is the number of jobs that were + # used with "-j" and that aren't used by other make targets + # + with JobserverExec() as jobserver: + n_jobs =3D 1 =20 - if os.path.exists(pdf_from): - os.rename(pdf_from, pdf_to) - builds[name] =3D os.path.relpath(pdf_to, self.buil= ddir) - else: - builds[name] =3D "FAILED" - build_failed =3D True + # + # Handle the case when a parameter is passed via command line, + # using it as default, if jobserver doesn't claim anything + # + if self.n_jobs: + try: + n_jobs =3D int(self.n_jobs) + except ValueError: + pass =20 - name =3D entry.name.removesuffix(".tex") - max_len =3D max(max_len, len(name)) + if jobserver.claim: + n_jobs =3D jobserver.claim =20 - if not has_tex: - name =3D os.path.basename(from_dir) - max_len =3D max(max_len, len(name)) - builds[name] =3D "FAILED (no .tex)" - build_failed =3D True + builds, build_failed, max_len =3D self.pdf_parallel_build(tex_= suffix, + latex_= cmd, + tex_fi= les, + n_jobs) =20 msg =3D "Summary" msg +=3D "\n" + "=3D" * len(msg) --=20 2.51.0