|
From: | Avi Kivity |
Subject: | Re: [Qemu-devel] QEMU/KVM SCSI lock up |
Date: | Thu, 03 Apr 2008 14:49:02 +0300 |
User-agent: | Thunderbird 2.0.0.12 (X11/20080226) |
Matteo Frigo wrote:
Avi Kivity <address@hidden> writes:Will that not issue the read for the wrong sector?The patch that I submitted issues the read for the correct sector: It reads (r->sector - n) instead of r->sector. I apologize if my explanation of the patch was confusing on this point.
I misread the patch, sorry.
I think the correct fix is to move r->sector and r->sector_count adjustment into scsi_read_complete() and scsi_write_complete().This will work as long as you can guarantee that scsi_read_complete() is called before the recursive call to scsi_read_data().
Yes, scsi_read_complete() calls s->completion, which triggers the recursion, by my reading.
-- error compiling committee.c: too many arguments to function
[Prev in Thread] | Current Thread | [Next in Thread] |