From: Glauber Costa <gcosta@redhat.com>
To: linux-kernel@vger.kernel.org
Cc: akpm@linux-foundation.org, tglx@linutronix.de, mingo@elte.hu,
hugh@veritas.com
Subject: [PATCH 0/15] Improve x86 smpboot integration
Date: Mon, 9 Jun 2008 11:16:43 -0300 [thread overview]
Message-ID: <1213021018-14159-1-git-send-email-gcosta@redhat.com> (raw)
Ingo,
First of all, sorry for being away for so long ;-)
Here it goes a series of improvements for the x86 smpboot integration.
The final goal is the same: To reduce the difference between architectures,
just this time I do a little bit more ;-)
Basically, this one reduces greatly the ifdef count:
[root@t60 linux-2.6-x86]# git-show 674ce:arch/x86/kernel/smpboot.c \
| grep -e "^#ifdef \+CONFIG_X86_\(32\|64\)" | wc -l
18
[root@t60 linux-2.6-x86]# git-show HEAD:arch/x86/kernel/smpboot.c \
| grep -e "^#ifdef \+CONFIG_X86_\(32\|64\)" | wc -l
8
The remaining ones are _mainly_ (not all) due to the fact that x86_64 uses the pda,
while i386 goes with normal per-cpu data. They are things like:
per_cpu(current_task, cpu) = c_idle.idle; vs cpu_pda(cpu)->pcurrent = c_idle.idle;
Also, there is the low mappings piece of code Hugh detected. To that, I intend
to also find a common base between them. Just I'm not doing it in this series, because
it deserves special attention.
As usual, this series was compiled tested in a whole bunch of different configs ( ~ 10
for each architecture), including all i386 variants. Boot tested in all my hardware.
The final diffstat is:
arch/x86/kernel/head_32.S | 2
arch/x86/kernel/head_64.S | 48 +---------------
arch/x86/kernel/io_apic_32.c | 2
arch/x86/kernel/setup64.c | 1
arch/x86/kernel/smpboot.c | 97 ++++------------------------------
b/arch/x86/kernel/acpi/sleep.c | 2
b/arch/x86/kernel/apic_32.c | 6 --
b/arch/x86/kernel/apic_64.c | 10 +++
b/arch/x86/kernel/head_32.S | 6 +-
b/arch/x86/kernel/head_64.S | 5 +
b/arch/x86/kernel/io_apic_32.c | 5 +
b/arch/x86/kernel/io_apic_64.c | 9 ++-
b/arch/x86/kernel/process_32.c | 16 +++++
b/arch/x86/kernel/setup64.c | 5 -
b/arch/x86/kernel/setup_64.c | 27 +++++++++
b/arch/x86/kernel/smpboot.c | 8 --
b/arch/x86/kernel/traps_32.c | 3 -
b/arch/x86/kernel/x8664_ksyms_64.c | 5 -
b/arch/x86/mach-voyager/voyager_smp.c | 2
b/include/asm-x86/desc.h | 24 +++-----
b/include/asm-x86/hw_irq.h | 3 -
b/include/asm-x86/numa_32.h | 1
b/include/asm-x86/segment.h | 24 ++++----
b/include/asm-x86/smp.h | 2
include/asm-x86/hw_irq.h | 4 -
25 files changed, 122 insertions(+), 195 deletions(-)
next reply other threads:[~2008-06-09 14:19 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-06-09 14:16 Glauber Costa [this message]
2008-06-09 14:16 ` [PATCH 01/15] x86: use stack_start in x86_64 Glauber Costa
2008-06-09 14:16 ` [PATCH 02/15] x86: don't use gdt_page openly Glauber Costa
2008-06-09 14:16 ` [PATCH 03/15] x86: remove early_gdt_descr reference Glauber Costa
2008-06-09 14:16 ` [PATCH 04/15] x86: move x86_64 gdt closer to i386 Glauber Costa
2008-06-09 14:16 ` [PATCH 05/15] x86: use initial_code for i386 Glauber Costa
2008-06-09 14:16 ` [PATCH 06/15] x86: boot secondary cpus through initial_code Glauber Costa
2008-06-09 14:16 ` [PATCH 07/15] x86: clearing io_apic harmless for x86_64 Glauber Costa
2008-06-09 14:16 ` [PATCH 08/15] x86: remove ifdef from stepping Glauber Costa
2008-06-09 14:16 ` [PATCH 09/15] x86: change __setup_vector_irq with setup_vector_irq Glauber Costa
2008-06-09 14:16 ` [PATCH 10/15] x86: provide connect_bsp_APIC for x86_64 Glauber Costa
2008-06-09 14:16 ` [PATCH 11/15] x86: move enabling of io_apic to prepare_cpus Glauber Costa
2008-06-09 14:16 ` [PATCH 12/15] x86: change naming to match x86_64 Glauber Costa
2008-06-09 14:16 ` [PATCH 13/15] x86: remove cpu from maps Glauber Costa
2008-06-09 14:16 ` [PATCH 14/15] x86: move cpu_exit_clear to process_32.c Glauber Costa
2008-06-09 14:16 ` [PATCH 15/15] x86: take load_sp0 out of smpboot.c Glauber Costa
2008-06-09 15:23 ` [PATCH 11/15] x86: move enabling of io_apic to prepare_cpus Maciej W. Rozycki
2008-06-09 15:52 ` Glauber Costa
2008-06-09 19:44 ` Glauber Costa
2008-06-09 20:12 ` Maciej W. Rozycki
2008-06-09 20:53 ` Yinghai Lu
2008-06-09 21:00 ` Maciej W. Rozycki
2008-06-10 2:46 ` Maciej W. Rozycki
2008-06-10 5:08 ` Yinghai Lu
2008-06-10 13:00 ` Glauber Costa
2008-06-10 13:30 ` Maciej W. Rozycki
2008-06-10 19:09 ` Yinghai Lu
2008-06-10 19:36 ` Maciej W. Rozycki
2008-06-10 19:49 ` Yinghai Lu
2008-06-11 0:29 ` Maciej W. Rozycki
2008-06-11 2:32 ` Yinghai Lu
2008-06-11 12:57 ` Maciej W. Rozycki
2008-06-09 21:02 ` Glauber Costa
2008-06-09 15:23 ` [PATCH 03/15] x86: remove early_gdt_descr reference James Bottomley
2008-06-09 15:49 ` Glauber Costa
2008-06-09 17:20 ` James Bottomley
2008-06-09 17:23 ` Glauber Costa
2008-06-09 17:40 ` James Bottomley
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=1213021018-14159-1-git-send-email-gcosta@redhat.com \
--to=gcosta@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=hugh@veritas.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=tglx@linutronix.de \
/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
all inboxes | Powered by JetHome®