git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Marco Costalba" <mcostalba@gmail.com>
To: "Linus Torvalds" <torvalds@osdl.org>
Cc: "Git Mailing List" <git@vger.kernel.org>,
	"Junio C Hamano" <junkio@cox.net>,
	"Alex Riesen" <raa.lkml@gmail.com>,
	"Shawn Pearce" <spearce@spearce.org>
Subject: Re: [RFC \ WISH] Add -o option to git-rev-list
Date: Sun, 10 Dec 2006 23:05:36 +0100	[thread overview]
Message-ID: <e5bfff550612101405h507cc877k87491264d11b0aa8@mail.gmail.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0612101213520.12500@woody.osdl.org>

On 12/10/06, Linus Torvalds <torvalds@osdl.org> wrote:
>

> data into ASCII etc. So temp-files are almost never a better solution than
> keeping things in memory (unless you use those temp-files to truly
> _share_ data between processes, ie you do a shared mmap and they can
> re-use the same pages actively in a way they couldn't otherwise).
>

Ok. Perhaps I'm doing something wrong but the following code it's
always 10% slower then the temporary file one (4.7s against 4.3s for
linux tree)

bool DataLoader::start(const QStringList& args, const QString& workDir) {

	QDir::setCurrent(workDir);

	_file = popen(args.join(" ").ascii(), "r");
	if (!_file)
		return false;
	
	loadTime.start();
	guiUpdateTimer.start(10, true); // will call on_timeout() in 10ms
	return true;
}

void DataLoader::on_timeout() {

	if (canceling)
		deleteLater();

// 	int fd = fileno(_file);    // read() case

	ssize_t len = 0;
	while (1) {

		QByteArray* ba = new QByteArray(FILE_BLOCK_SIZE); // 64KB

// 		len = read(fd, ba->data(), ba->size());        // read() case

		len = fread(ba->data(), 1, ba->size(), _file); // fread() case
		if (len <= 0) {
			delete ba;
			break;

		} else if (len < (ssize_t)ba->size()) // very rare, 4 out of 40000
on Linux tree
			ba->resize(len);

		loadedBytes += len;
		fh->rowData.append(ba); // fh->rowData it's a pointer's list
		parseSingleBuffer(ba);

		// avoid reading small chunks if data producer is still running
		if (len < FILE_BLOCK_SIZE)
			break;
	}

// 	if (len == 0) {         // read() case
	if (feof(_file)) {      // fread() case
	
		emit loaded(fh, loadedBytes, loadTime.elapsed(), true, "", ""); //
show some stats

		pclose(_file);
		_file = NULL;
		deleteLater();
	} else
		guiUpdateTimer.start(100, true); // next read in 100ms	
}

Uncomment 'read() case' lines and comment out the 'fread case()' ones
and you have a way slooooower code, about 10 times slower!



  reply	other threads:[~2006-12-10 22:05 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-12-10 11:38 [RFC \ WISH] Add -o option to git-rev-list Marco Costalba
2006-12-10 14:54 ` Alex Riesen
2006-12-10 18:16 ` Linus Torvalds
2006-12-10 19:51   ` Marco Costalba
2006-12-10 20:00     ` globs in partial checkout? Michael S. Tsirkin
2006-12-10 20:13       ` Linus Torvalds
2006-12-10 21:07         ` Michael S. Tsirkin
2006-12-10 20:08     ` [RFC \ WISH] Add -o option to git-rev-list Linus Torvalds
2006-12-10 20:19       ` Linus Torvalds
2006-12-10 22:05         ` Marco Costalba [this message]
2006-12-10 22:09           ` Marco Costalba
2006-12-10 22:16           ` Linus Torvalds
2006-12-10 22:35             ` Marco Costalba
2006-12-10 22:53               ` Linus Torvalds
2006-12-11  0:15                 ` Marco Costalba
2006-12-11  0:51                   ` Linus Torvalds
2006-12-11  7:17                     ` Marco Costalba
2006-12-11 10:00                       ` Alex Riesen
2006-12-11 16:59                       ` Linus Torvalds
2006-12-11 17:07                         ` Linus Torvalds
2006-12-11 17:39                           ` Marco Costalba
2006-12-11 18:15                             ` Linus Torvalds
2006-12-11 18:59                               ` Marco Costalba
2006-12-11 19:25                                 ` Linus Torvalds
2006-12-11 20:28                                 ` Josef Weidendorfer
2006-12-11 20:40                                   ` Linus Torvalds
2006-12-11 20:54                                     ` Josef Weidendorfer
2006-12-11 21:14                                       ` Linus Torvalds
2006-12-15 18:45                                         ` Marco Costalba
2006-12-15 19:20                                           ` Linus Torvalds
2006-12-15 20:41                                             ` Marco Costalba
2006-12-15 21:04                                               ` Marco Costalba
2006-12-11  9:26                   ` Josef Weidendorfer
2006-12-11 12:52                     ` Marco Costalba
2006-12-11 13:28                       ` Josef Weidendorfer
2006-12-11 17:28                         ` Marco Costalba
2006-12-11 11:39     ` Andreas Ericsson
2006-12-11 12:59       ` Marco Costalba
2006-12-11 13:40         ` Andreas Ericsson

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=e5bfff550612101405h507cc877k87491264d11b0aa8@mail.gmail.com \
    --to=mcostalba@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=junkio@cox.net \
    --cc=raa.lkml@gmail.com \
    --cc=spearce@spearce.org \
    --cc=torvalds@osdl.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).