qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Philippe Mathieu-Daudé" <philmd@redhat.com>
To: Aleksandar Markovic <aleksandar.m.mail@gmail.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>,
	Cornelia Huck <cohuck@redhat.com>,
	Laurent Vivier <laurent@vivier.eu>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"qemu-s390x@nongnu.org" <qemu-s390x@nongnu.org>,
	Alistair Francis <Alistair.Francis@wdc.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Richard Henderson <rth@twiddle.net>
Subject: Re: [PATCH v2 00/12] user-mode: Prune build dependencies (part 1)
Date: Mon, 16 Mar 2020 08:45:10 +0100	[thread overview]
Message-ID: <d73a1e52-06d4-e57e-43b9-30c5c03d71d0@redhat.com> (raw)
In-Reply-To: <CAL1e-=jBg6OVe+Qgg5SOoJ+S4x70vQF_rttE2_VTddhR4O5n0w@mail.gmail.com>

On 3/16/20 1:16 AM, Aleksandar Markovic wrote:
> 
> 
> On Monday, March 16, 2020, Philippe Mathieu-Daudé <philmd@redhat.com 
> <mailto:philmd@redhat.com>> wrote:
> 
>     This is the first part of a series reducing user-mode
>     dependencies. By stripping out unused code, the build
>     and testing time is reduced (as is space used by objects).
> 
>     Part 1:
>     - reduce user-mode object list
>     - remove some migration code from user-mode
>     - remove cpu_get_crash_info()
> 
> 
> What is the purpose of dividing into parts? What is the content of other 
> parts, and when do you plan to submit those? A series is usually a 
> stand-alone and a complete logical unit - why did you decide to submit 
> "parts" separately (just curious)?

Big series are hard to digest and scare reviewers. Peter told me twice 
his rule of thumb is to split a series if it gets bigger than 20 patches 
(and a patch if it modify more than 200 lines). He also recently said he 
skipped review of a ~32 patches series of mine which was too big. I 
don't want other reviewers to do that neither, so I try to split <=20.

Each series could be applied apart, except the last patch from the 3rd 
part (qapi: Restrict code generated for user-mode) which is the one 
really cutting down user-mode code by avoiding pulling in system-mode 
symbols.

First part is generic, second part is QAPI-related, and third part 
concerns hw/core/qdev-properties.c. Each part is covered by different 
maintainers.

> 
> Does this series affect executables' size, or cut build times only?

Both. It will saves us CI testing time, save time to distributions 
packaging linux-user-only builds, produce smaller binaries.

> 
> Thanks,
> Aleksandar
> 
>     Since v1:
>     - Addressed Laurent/Richard review comments
>     - Removed 'exec: Drop redundant #ifdeffery'
>     - Removed 'target: Restrict write_elfXX_note() handlers to system-mode'
> 
>     v1:
>     https://www.mail-archive.com/qemu-devel@nongnu.org/msg688456.html
>     <https://www.mail-archive.com/qemu-devel@nongnu.org/msg688456.html>
> 
>     Philippe Mathieu-Daudé (12):
>        Makefile: Only build virtiofsd if system-mode is enabled
>        configure: Avoid building TCG when not needed
>        tests/Makefile: Only display TCG-related tests when TCG is available
>        tests/Makefile: Restrict some softmmu-only tests
>        util/Makefile: Reduce the user-mode object list
>        stubs/Makefile: Reduce the user-mode object list
>        target/riscv/cpu: Restrict CPU migration to system-mode
>        exec: Assert CPU migration is not used on user-only build
>        arch_init: Remove unused 'qapi-commands-misc.h' include
>        target/i386: Restrict CpuClass::get_crash_info() to system-mode
>        target/s390x: Restrict CpuClass::get_crash_info() to system-mode
>        hw/core: Restrict CpuClass::get_crash_info() to system-mode
> 
>       configure              |  4 +++
>       Makefile               |  2 +-
>       include/hw/core/cpu.h  |  7 ++++-
>       arch_init.c            |  1 -
>       exec.c                 |  4 ++-
>       hw/core/cpu.c          |  2 ++
>       target/i386/cpu.c      |  6 ++++-
>       target/riscv/cpu.c     |  6 +++--
>       target/s390x/cpu.c     | 12 ++++-----
>       stubs/Makefile.objs    | 52 +++++++++++++++++++++----------------
>       tests/Makefile.include | 18 +++++++------
>       util/Makefile.objs     | 59 +++++++++++++++++++++++++++---------------
>       12 files changed, 108 insertions(+), 65 deletions(-)
> 
>     -- 
>     2.21.1
> 
> 



      reply	other threads:[~2020-03-16  7:51 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-15 23:57 [PATCH v2 00/12] user-mode: Prune build dependencies (part 1) Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 01/12] Makefile: Only build virtiofsd if system-mode is enabled Philippe Mathieu-Daudé
2020-03-16 18:57   ` Richard Henderson
2020-03-19 11:13   ` Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 02/12] configure: Avoid building TCG when not needed Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 03/12] tests/Makefile: Only display TCG-related tests when TCG is available Philippe Mathieu-Daudé
2020-03-19 11:15   ` Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 04/12] tests/Makefile: Restrict some softmmu-only tests Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 05/12] util/Makefile: Reduce the user-mode object list Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 06/12] stubs/Makefile: " Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 07/12] target/riscv/cpu: Restrict CPU migration to system-mode Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 08/12] exec: Assert CPU migration is not used on user-only build Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 09/12] arch_init: Remove unused 'qapi-commands-misc.h' include Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 10/12] target/i386: Restrict CpuClass::get_crash_info() to system-mode Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 11/12] target/s390x: " Philippe Mathieu-Daudé
2020-03-15 23:57 ` [PATCH v2 12/12] hw/core: " Philippe Mathieu-Daudé
2020-03-16  0:16 ` [PATCH v2 00/12] user-mode: Prune build dependencies (part 1) Aleksandar Markovic
2020-03-16  7:45   ` Philippe Mathieu-Daudé [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=d73a1e52-06d4-e57e-43b9-30c5c03d71d0@redhat.com \
    --to=philmd@redhat.com \
    --cc=Alistair.Francis@wdc.com \
    --cc=aleksandar.m.mail@gmail.com \
    --cc=cohuck@redhat.com \
    --cc=ehabkost@redhat.com \
    --cc=laurent@vivier.eu \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-s390x@nongnu.org \
    --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 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).