From: "Chris Li" <christ.li@gmail.com>
To: Thomas Schmid <Thomas.Schmid@br-automation.com>
Cc: linux-sparse@vger.kernel.org
Subject: Re: Listing special marked symbols - attribute section
Date: Thu, 3 May 2007 15:37:17 -0700 [thread overview]
Message-ID: <70318cbf0705031537o4afd1bf8v39635e38c32548c8@mail.gmail.com> (raw)
In-Reply-To: <OFD5E73F1F.17F8F44A-ONC12572CF.003EC61A-C12572D0.00319C0B@br-automation.com>
On 5/3/07, Thomas Schmid <Thomas.Schmid@br-automation.com> wrote:
> In 0.3 are several extensions to parse attributes, but it seems that
> <sectionname> isn't stored anywhere.
> Is it possible to get <sectionname>?
>
> If there is no way to get <sectionname> in the actual implementation, i
> think of an extension in struct ctype.
> Any comments?
> Would this be useful for someone else?
Yes, it would be useful. The easiest way is just increase the ctype to include
a pointer to section name. The biggest objection is from the fact that
symbol/ctype
is the most common data structure in the sparse. If we increase the size
of ctype, there is a huge impact of the sparse memory usage.
I did try to use some kind of extend the attributes. The idea is have
the most common
attribute embed in ctype. The not common ones will store in a extend
attribute structure.
The extend attribute structure itself is flat. The initial
implementation will only contain
address space and context. I found myself keep breaking the existing
address space
attributes.
The attribute propagation not very clear in sparse. Some times we use assignment
and some times the attribute is inherited from implicit memcpy and
some time we strip
the attribute out. I wish to have a more unify way to define how
attribute are created
and propagate through the ctype, maybe a table or some thing.
Chris
next prev parent reply other threads:[~2007-05-03 22:37 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-05-03 9:01 Listing special marked symbols - attribute section Thomas Schmid
2007-05-03 22:37 ` Chris Li [this message]
2007-05-04 13:08 ` Antwort: " Thomas Schmid
2007-05-04 18:29 ` Christopher Li
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=70318cbf0705031537o4afd1bf8v39635e38c32548c8@mail.gmail.com \
--to=christ.li@gmail.com \
--cc=Thomas.Schmid@br-automation.com \
--cc=linux-sparse@vger.kernel.org \
/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).