qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@gmail.com>
To: Liu Ping Fan <qemulist@gmail.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>,
	Jan Kiszka <jan.kiszka@siemens.com>,
	qemu-devel@nongnu.org, mdroth <mdroth@linux.vnet.ibm.com>,
	Anthony Liguori <anthony@codemonkey.ws>,
	Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [Qemu-devel] [RFC PATCH v4 01/15] util: introduce gsource event abstration
Date: Thu, 18 Apr 2013 16:01:11 +0200	[thread overview]
Message-ID: <20130418140111.GA22842@stefanha-thinkpad.redhat.com> (raw)
In-Reply-To: <1366187964-14265-2-git-send-email-qemulist@gmail.com>

On Wed, Apr 17, 2013 at 04:39:10PM +0800, Liu Ping Fan wrote:
> +static gboolean prepare(GSource *src, gint *time)
> +{
> +    EventGSource *nsrc = (EventGSource *)src;
> +    int events = 0;
> +
> +    if (!nsrc->readable && !nsrc->writable) {
> +        return false;
> +    }
> +    if (nsrc->readable && nsrc->readable(nsrc->opaque)) {
> +        events |= G_IO_IN;
> +    }
> +    if ((nsrc->writable) && nsrc->writable(nsrc->opaque)) {
> +        events |= G_IO_OUT;
> +    }

G_IO_ERR, G_IO_HUP, G_IO_PRI?

Here is the select(2) to GCondition mapping:
rfds -> G_IO_IN | G_IO_HUP | G_IO_ERR
wfds -> G_IO_OUT | G_IO_ERR
xfds -> G_IO_PRI

In other words, we're missing events by just using G_IO_IN and G_IO_OUT.
Whether that matters depends on EventGSource users.  For sockets it can
matter.

> +void event_source_release(EventGSource *src)
> +{
> +    g_source_destroy(&src->source);

Leaks src.

> +}
> +
> +GPollFD *events_source_get_gfd(EventsGSource *src, int fd)

events_source_add_fd() seems like a better name since this function
always allocates a new GPollFD, it never "gets" an existing one.

> +{
> +    GPollFD *retfd;
> +    unsigned long idx;
> +
> +    idx = find_first_zero_bit(src->alloc_bmp, src->bmp_sz);
> +    if (idx == src->bmp_sz) {
> +        //idx = src->bmp_sz;

Commented out line.

> +void events_source_close_gfd(EventsGSource *src, GPollFD *pollfd)

"close" usually means close(2).  I suggest "remove" instead.

> +EventsGSource *events_source_new(GSourceFuncs *funcs, GSourceFunc dispatch_cb, void *opaque)
> +{
> +    EventsGSource *src = (EventsGSource *)g_source_new(funcs, sizeof(EventsGSource));
> +
> +    /* 8bits size at initial */
> +    src->bmp_sz = 8;
> +    src->alloc_bmp = g_malloc0(src->bmp_sz >> 3);

This is unportable.  alloc_bmp is unsigned long, you are allocating just
one byte!

Please drop the bitmap approach and use a doubly-linked list or another
glib container type of your choice.  It needs 3 operations: add, remove,
and iterate.

> +/* multi fd drive gsource*/
> +typedef struct EventsGSource {
> +    GSource source;
> +    /* 8 for initial, stand for 8 pollfds */
> +    unsigned int bmp_sz;
> +    unsigned long *alloc_bmp;
> +    GPollFD *pollfds;
> +    void *opaque;
> +} EventsGSource;
> +
> +EventsGSource *events_source_new(GSourceFuncs *funcs, GSourceFunc dispatch_cb, void *opaque);
> +void events_source_release(EventsGSource *src);
> +gboolean events_source_check(GSource *src);
> +gboolean events_source_dispatch(GSource *src, GSourceFunc cb, gpointer data);
> +GPollFD *events_source_get_gfd(EventsGSource *src, int fd);
> +void events_source_close_gfd(EventsGSource *src, GPollFD *pollfd);

Why are check/dispatch public?  Perhaps events_source_new() just needs a
prepare() argument instead of exposing GSourceFuncs.

  reply	other threads:[~2013-04-18 14:01 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-17  8:39 [Qemu-devel] [RFC PATCH v4 00/15] port network layer onto glib Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 01/15] util: introduce gsource event abstration Liu Ping Fan
2013-04-18 14:01   ` Stefan Hajnoczi [this message]
2013-04-19  6:52     ` liu ping fan
2013-04-19 11:59       ` Stefan Hajnoczi
2013-04-22  7:50         ` liu ping fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 02/15] net: introduce bind_ctx to NetClientInfo Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 03/15] net: port tap onto GSource Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 04/15] net: resolve race of tap backend and its peer Liu Ping Fan
2013-04-18 14:11   ` Stefan Hajnoczi
2013-04-19  5:43     ` liu ping fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 05/15] net: port vde onto GSource Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 06/15] net: port socket to GSource Liu Ping Fan
2013-04-18 14:34   ` Stefan Hajnoczi
2013-04-19  5:58     ` liu ping fan
2013-04-19 12:03       ` Stefan Hajnoczi
2013-04-22  7:52         ` liu ping fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 07/15] net: port tap-win32 onto GSource Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 08/15] net: hub use lock to protect ports list Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 09/15] net: introduce lock to protect NetQueue Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 10/15] net: introduce lock to protect NetClientState's peer's access Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 11/15] net: make netclient re-entrant with refcnt Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 12/15] slirp: make timeout local Liu Ping Fan
2013-04-18 14:22   ` Paolo Bonzini
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 13/15] slirp: make slirp event dispatch based on slirp instance, not global Liu Ping Fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 14/15] slirp: handle race condition Liu Ping Fan
2013-04-18  7:13   ` Jan Kiszka
2013-04-19  0:18     ` liu ping fan
2013-04-19  8:21       ` Jan Kiszka
2013-04-22  5:55         ` liu ping fan
2013-04-23  7:20           ` liu ping fan
2013-04-17  8:39 ` [Qemu-devel] [RFC PATCH v4 15/15] slirp: use lock to protect the slirp_instances Liu Ping Fan
2013-04-18  7:20   ` Jan Kiszka
2013-04-18 14:16     ` Paolo Bonzini
2013-04-19  6:13       ` liu ping fan

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=20130418140111.GA22842@stefanha-thinkpad.redhat.com \
    --to=stefanha@gmail.com \
    --cc=anthony@codemonkey.ws \
    --cc=jan.kiszka@siemens.com \
    --cc=mdroth@linux.vnet.ibm.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemulist@gmail.com \
    --cc=stefanha@redhat.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;
as well as URLs for NNTP newsgroup(s).