From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:47326) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XifnA-0000DY-S1 for qemu-devel@nongnu.org; Mon, 27 Oct 2014 04:36:52 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Xifn5-0007x5-5E for qemu-devel@nongnu.org; Mon, 27 Oct 2014 04:36:44 -0400 Received: from mx-v6.kamp.de ([2a02:248:0:51::16]:60656 helo=mx01.kamp.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Xifn4-0007x1-Ru for qemu-devel@nongnu.org; Mon, 27 Oct 2014 04:36:39 -0400 Message-ID: <544E0412.6000405@kamp.de> Date: Mon, 27 Oct 2014 09:36:34 +0100 From: Peter Lieven MIME-Version: 1.0 References: <1414253919-3044-1-git-send-email-pl@kamp.de> <1414253919-3044-4-git-send-email-pl@kamp.de> <544E0216.2050400@redhat.com> In-Reply-To: <544E0216.2050400@redhat.com> Content-Type: text/plain; charset=iso-8859-15; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCHv5 3/6] block/iscsi: set max_transfer_length List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Max Reitz , qemu-devel@nongnu.org Cc: kwolf@redhat.com, famz@redhat.com, benoit@irqsave.net, ronniesahlberg@gmail.com, armbru@redhat.com, stefanha@redhat.com On 27.10.2014 09:28, Max Reitz wrote: > On 2014-10-25 at 18:18, Peter Lieven wrote: >> Copy the max_xfer_len from the BlockLimits VPD or use the >> maximum value fitting in the CDB. >> >> Signed-off-by: Peter Lieven >> --- >> block/iscsi.c | 17 +++++++++++++++-- >> 1 file changed, 15 insertions(+), 2 deletions(-) >> >> diff --git a/block/iscsi.c b/block/iscsi.c >> index 233f462..1ae4add 100644 >> --- a/block/iscsi.c >> +++ b/block/iscsi.c >> @@ -297,6 +297,11 @@ static int64_t sector_lun2qemu(int64_t sector, IscsiLun *iscsilun) >> return sector * iscsilun->block_size / BDRV_SECTOR_SIZE; >> } >> +static int nb_sectors_lun2qemu(int64_t sector, IscsiLun *iscsilun) >> +{ >> + return MIN(sector_lun2qemu(sector, iscsilun), INT_MAX); >> +} >> + >> static int64_t sector_qemu2lun(int64_t sector, IscsiLun *iscsilun) >> { >> return sector * BDRV_SECTOR_SIZE / iscsilun->block_size; >> @@ -1449,10 +1454,18 @@ static void iscsi_close(BlockDriverState *bs) >> static void iscsi_refresh_limits(BlockDriverState *bs, Error **errp) >> { >> - IscsiLun *iscsilun = bs->opaque; >> - >> /* We don't actually refresh here, but just return data queried in >> * iscsi_open(): iscsi targets don't change their limits. */ >> + >> + IscsiLun *iscsilun = bs->opaque; >> + uint32_t max_xfer_len = iscsilun->use_16_for_rw ? 0xffffffff : 0xffff; >> + >> + if (iscsilun->bl.max_xfer_len) { >> + max_xfer_len = MIN(max_xfer_len, iscsilun->bl.max_xfer_len); >> + } >> + >> + bs->bl.max_transfer_length = nb_sectors_lun2qemu(max_xfer_len, iscsilun); >> + >> if (iscsilun->lbp.lbpu) { >> if (iscsilun->bl.max_unmap < 0xffffffff) { >> bs->bl.max_discard = sector_lun2qemu(iscsilun->bl.max_unmap, > > Hm, seems strange to have a function called nb_sectors_lun2qemu() not only convert values, but also cap the result. But anyway: Would you give it another name? Peter