All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Leonardo E. Reiter" <lreiter@win4lin.com>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] [PATCH] Add support to enable dynamic removable block devices
Date: Sun, 16 Apr 2006 13:21:05 -0400	[thread overview]
Message-ID: <44427D01.7060804@win4lin.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 1020 bytes --]

Hello,

attached is a patch that modifies the block driver infrastructure to 
support optional "dynamic" removable block devices.  For example, today, 
you can attach a CDROM block device, but it has to be manually attached 
and ejected from the monitor, even if this block device points to a real 
physical driver (i.e. /dev/cdrom on Linux), which itself supports really 
removable media.

The patch does not change any current functionality.  But, it can be 
used to build a block driver which has dynamic state for the media, even 
though the block device itself is attached statically.  For example, 
this will enable [in the future - new block driver needed] attaching a 
real CDROM device, and having the ability to actually eject the media 
and insert new media without having to use the monitor.

Regards,

Leo Reiter

-- 
Leonardo E. Reiter
Vice President of Product Development, CTO

Win4Lin, Inc.
Virtual Computing from Desktop to Data Center
Main: +1 512 339 7979
Fax: +1 512 532 6501
http://www.win4lin.com

[-- Attachment #2: qemu-blk-rmv.patch --]
[-- Type: text/x-patch, Size: 2704 bytes --]

Index: block.c
===================================================================
RCS file: /cvsroot/qemu/qemu/block.c,v
retrieving revision 1.25
diff -a -u -r1.25 block.c
--- block.c	18 Dec 2005 18:28:15 -0000	1.25
+++ block.c	16 Apr 2006 17:14:59 -0000
@@ -446,13 +446,21 @@
     return 0;
 }
 
+/* this function is needed to avoid checking for/calling dynamic methods for 
+ * removable devices if the device is not designated as removable */
+static inline int bdrv_can_write(BlockDriverState *bs)
+{
+    if (!bs->removable)
+        return (bs->inserted) && (!bs->read_only);
+    else
+        return (bdrv_is_inserted(bs)) && (!bdrv_is_read_only(bs));
+}
+
 /* return -1 if error */
 int bdrv_write(BlockDriverState *bs, int64_t sector_num, 
                const uint8_t *buf, int nb_sectors)
 {
-    if (!bs->inserted)
-        return -1;
-    if (bs->read_only)
+    if (!bdrv_can_write(bs))
         return -1;
     if (sector_num == 0 && bs->boot_sector_enabled && nb_sectors > 0) {
         memcpy(bs->boot_sector_data, buf, 512);   
@@ -520,11 +528,15 @@
 
 int bdrv_is_read_only(BlockDriverState *bs)
 {
+    if (bs->drv && bs->drv->bdrv_is_read_only)
+        bs->read_only = bs->drv->bdrv_is_read_only(bs);
     return bs->read_only;
 }
 
 int bdrv_is_inserted(BlockDriverState *bs)
 {
+    if (bs->drv && bs->drv->bdrv_is_inserted)
+        bs->inserted = bs->drv->bdrv_is_inserted(bs);
     return bs->inserted;
 }
 
@@ -535,6 +547,13 @@
 
 void bdrv_set_locked(BlockDriverState *bs, int locked)
 {
+    /* XXX: bs->drv->bdrv_set_locked() has no need to set bs->locked; if this 
+     * function gets called from the disk controller, it has to always set the 
+     * locked state to whatever the disk controller says, regardless of whether 
+     * or not this is implemented by the block driver itself or whether or not
+     * the block driver succeeds in setting the locked state. */
+    if (bs->drv && bs->drv->bdrv_set_locked)
+        bs->drv->bdrv_set_locked(bs, locked);
     bs->locked = locked;
 }
 
Index: block_int.h
===================================================================
RCS file: /cvsroot/qemu/qemu/block_int.h,v
retrieving revision 1.4
diff -a -u -r1.4 block_int.h
--- block_int.h	18 Dec 2005 18:28:15 -0000	1.4
+++ block_int.h	16 Apr 2006 17:14:59 -0000
@@ -40,6 +40,9 @@
                              int nb_sectors, int *pnum);
     int (*bdrv_set_key)(BlockDriverState *bs, const char *key);
     int (*bdrv_make_empty)(BlockDriverState *bs);
+    int (*bdrv_is_read_only)(BlockDriverState *bs);
+    int (*bdrv_is_inserted)(BlockDriverState *bs);
+    void (*bdrv_set_locked)(BlockDriverState *bs, int locked);
     struct BlockDriver *next;
 };
 

                 reply	other threads:[~2006-04-16 17:21 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=44427D01.7060804@win4lin.com \
    --to=lreiter@win4lin.com \
    --cc=qemu-devel@nongnu.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.