From: Roman Zippel <zippel@linux-m68k.org>
To: Christoph Hellwig <hch@infradead.org>
Cc: William Lee Irwin III <wli@holomorphy.com>,
linux-kernel@vger.kernel.org, Andrew Morton <akpm@osdl.org>
Subject: Re: [12/12] fix thread_info.h ignoring __HAVE_THREAD_FUNCTIONS
Date: Mon, 14 Jun 2004 12:19:12 +0200 (CEST) [thread overview]
Message-ID: <Pine.LNX.4.58.0406141032210.10292@scrub.local> (raw)
In-Reply-To: <20040614081639.GI7162@infradead.org>
Hi,
On Mon, 14 Jun 2004, Christoph Hellwig wrote:
> On Sun, Jun 13, 2004 at 05:48:55PM -0700, William Lee Irwin III wrote:
> > * Check __HAVE_THREAD_FUNCTIONS in include/linux/thread_info.h (m68k)
> > This fixes the build on m68k; its thread_info functions need to be used.
>
> I don't like this one a lot and prefer to discuss it with the m68k folks
> first. Given they didn't sent it to Linus themselves I guess they're not
> completely proud of it ;-)
That's out of the "should we just blindly copy everything from i386?"
department. These thread info functions used char fields for a short
while, which is actually preferable over bitfields on a lot of archs.
The various thread flags have different usage, often the current thread is
the only one accessing it, but all of them right now use possibly
expensive bit field functions.
I'm thinking about reverting this one and just copy what everyone else is
doing, as the benefit is probably not that big for m68k.
There is another pending change in this department: current_thread_info().
For all nonbroken archs which have proper thread register it would
actually be beneficial, to keep the task structure and thread info
together and access them via the thread register, but a certain arch
and include dependencies forces everyone to derive the thread info pointer
from the stack pointer.
The most important change here is to separate task_struct out of sched.h
and I'd really like to get this change in, even if it has to wait for 2.7.
bye, Roman
next prev parent reply other threads:[~2004-06-14 10:19 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-06-14 0:31 [0/12] Debian bugfixes William Lee Irwin III
2004-06-14 0:33 ` [1/12] don't dereference netdev->name before register_netdev() William Lee Irwin III
2004-06-14 0:34 ` [2/12] lower priority of "too many keys" msg in atkbd.c William Lee Irwin III
2004-06-14 0:36 ` [3/12] remove irda usage of isa_virt_to_bus() William Lee Irwin III
2004-06-14 0:37 ` [4/12] unregister driver if probing fails in sb_card.c William Lee Irwin III
2004-06-14 0:38 ` [5/12] Ignore errors from tw_setfeature in 3w-xxxx.c William Lee Irwin III
2004-06-14 0:39 ` [6/12] fix advansys.c highmem bugs William Lee Irwin III
2004-06-14 0:40 ` [7/12] Handle NO_SENSE in sd_rw_intr in sd.c William Lee Irwin III
2004-06-14 0:41 ` [8/12] fake inquiry for Sony Clie PEG-TJ25 in unusual_devs.h William Lee Irwin III
2004-06-14 0:43 ` [9/12] fix duplicate environment variables passed to init William Lee Irwin III
2004-06-14 0:45 ` [10/12] fix handling of '/' embedded in filenames in isofs William Lee Irwin III
2004-06-14 0:47 ` [11/12] fix isofs ignoring noexec and mode mount options William Lee Irwin III
2004-06-14 0:48 ` [12/12] fix thread_info.h ignoring __HAVE_THREAD_FUNCTIONS William Lee Irwin III
2004-06-14 8:16 ` Christoph Hellwig
2004-06-14 8:34 ` William Lee Irwin III
2004-06-14 10:19 ` Roman Zippel [this message]
2004-06-14 10:23 ` Christoph Hellwig
2004-06-14 10:47 ` Roman Zippel
2004-06-14 8:15 ` [10/12] fix handling of '/' embedded in filenames in isofs Christoph Hellwig
2004-06-14 8:14 ` [8/12] fake inquiry for Sony Clie PEG-TJ25 in unusual_devs.h Christoph Hellwig
2004-06-14 8:14 ` [7/12] Handle NO_SENSE in sd_rw_intr in sd.c Christoph Hellwig
2004-06-14 4:26 ` [6/12] fix advansys.c highmem bugs Andrew Morton
2004-06-14 8:11 ` William Lee Irwin III
2004-06-14 8:13 ` Christoph Hellwig
2004-06-14 8:12 ` [5/12] Ignore errors from tw_setfeature in 3w-xxxx.c Christoph Hellwig
2004-06-14 8:11 ` [3/12] remove irda usage of isa_virt_to_bus() Christoph Hellwig
2004-06-14 8:11 ` [2/12] lower priority of "too many keys" msg in atkbd.c Christoph Hellwig
2004-06-14 4:14 ` [1/12] don't dereference netdev->name before register_netdev() Andrew Morton
2004-06-14 4:27 ` Jeff Garzik
2004-06-14 5:26 ` Herbert Xu
2004-06-14 8:21 ` William Lee Irwin III
2004-06-14 8:10 ` Christoph Hellwig
2004-06-14 8:22 ` William Lee Irwin III
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=Pine.LNX.4.58.0406141032210.10292@scrub.local \
--to=zippel@linux-m68k.org \
--cc=akpm@osdl.org \
--cc=hch@infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=wli@holomorphy.com \
/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®