From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:39867) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Smimx-0003bE-NR for qemu-devel@nongnu.org; Thu, 05 Jul 2012 05:56:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Smimq-00070h-Sq for qemu-devel@nongnu.org; Thu, 05 Jul 2012 05:55:55 -0400 Received: from mx1.redhat.com ([209.132.183.28]:3775) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Smimq-00070a-Ka for qemu-devel@nongnu.org; Thu, 05 Jul 2012 05:55:48 -0400 Message-ID: <4FF5649E.4070606@redhat.com> Date: Thu, 05 Jul 2012 11:55:42 +0200 From: Kevin Wolf MIME-Version: 1.0 References: <1340984094-5451-1-git-send-email-armbru@redhat.com> <1340984094-5451-4-git-send-email-armbru@redhat.com> <4FF45E87.8060100@redhat.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 03/32] vvfat: Fix partition table List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Markus Armbruster Cc: Blue Swirl , pbonzini@redhat.com, aliguori@us.ibm.com, qemu-devel@nongnu.org, stefanha@linux.vnet.ibm.com Am 05.07.2012 11:23, schrieb Markus Armbruster: > Kevin Wolf writes: > >> Am 29.06.2012 22:33, schrieb Blue Swirl: >>> On Fri, Jun 29, 2012 at 3:34 PM, Markus Armbruster wrote: >>>> Unless parameter ":floppy:" is given, vvfat creates a virtual image >>>> with DOS MBR defining a single partition which holds the FAT file >>>> system. The size of the virtual image depends on the width of the >>>> FAT: 32 MiB (CHS 64, 16, 63) for 12 bit FAT, 504 MiB (CHS 1024, 16, >>>> 63) for 16 and 32 bit FAT, leaving (64*16-1)*63 = 64449 and >>>> (1024*16-1)*64 = 1032129 sectors for the partition. >>>> >>>> However, it screws up the end of the partition in the MBR: >>>> >>>> FAT width param. start CHS end CHS start LBA size >>>> :32: 0,1,1 1023,14,63 63 1032065 >>>> :16: 0,1,1 1023,14,55 63 1032057 >>>> :12: 0,1,1 63,14,55 63 64377 >>>> >>>> The actual FAT file system nevertheless assumes the partition has >>>> 1032129 or 64449 sectors. Oops. >>>> >>>> Signed-off-by: Markus Armbruster >>>> --- >>>> block/vvfat.c | 7 ++++--- >>>> 1 files changed, 4 insertions(+), 3 deletions(-) >>>> >>>> diff --git a/block/vvfat.c b/block/vvfat.c >>>> index 0fd3367..62745b5 100644 >>>> --- a/block/vvfat.c >>>> +++ b/block/vvfat.c >>>> @@ -394,11 +394,12 @@ static void init_mbr(BDRVVVFATState* s) >>>> >>>> /* LBA is used when partition is outside the CHS geometry */ >>>> lba = sector2CHS(s->bs, &partition->start_CHS, s->first_sectors_number-1); >>>> - lba|= sector2CHS(s->bs, &partition->end_CHS, s->sector_count); >>>> + lba |= sector2CHS(s->bs, &partition->end_CHS, s->bs->total_sectors - 1); >>>> >>>> /*LBA partitions are identified only by start/length_sector_long not by CHS*/ >>>> - partition->start_sector_long =cpu_to_le32(s->first_sectors_number-1); >>>> - partition->length_sector_long=cpu_to_le32(s->sector_count - s->first_sectors_number+1); >>>> + partition->start_sector_long = cpu_to_le32(s->first_sectors_number-1); >>> >>> Spaces around '-'. Thanks for fixing the other cases, BTW. >> >> For compensation there's an extra space before the '='. > > The original lined up the two '='. I preserved that. Not that I care > for it. Want me to drop the extra space? Ah, didn't notice that. I don't mind then. >>>> + partition->length_sector_long = cpu_to_le32(s->bs->total_sectors >>>> + - s->first_sectors_number + 1); >> >> Just wondering... This should be the same as s->sector_count, right? > > Hmm. vvfat_open() assigns: > > s->sector_count = cyls * heads * secs - (s->first_sectors_number - 1); > bs->total_sectors = cyls * heads * secs; > > But it then changes it minds and does: > > s->sector_count = s->faked_sectors + s->sectors_per_cluster*s->cluster_count; Which probably means that they differ if some sub-cluster sized space is left unused at the end of the disk. It's not useful to have this space included in the partition, but it doesn't hurt either. So I suppose either way is fine. >> This whole geometry calculation code in vvfat is way too convoluted to >> understand when you haven't looked at it for some months... > > Yes, it is. I was briefly tempted to replace it, but then decided to > limit my footprint in this buck-crazy block driver to minimal bug fixes. Heh, quite understandable. Kevin