[GIT PULL] First round of SCSI updates for the 3.5+ merge window

The most important feature of this patch set is the new async infrastructure that makes sure async_synchronize_full() synchronizes all domains and allows us to remove all the hacks (like having scsi_complete_async_scans() in the device base code) and means that the async infrastructure will "just w

Re: scsi_requeue_command and dropping reference to sdev_gendev

On Mon, 2012-07-23 at 21:13 -0700, Ani Sinha wrote: > Hi : > > I am looking at the latest upstream code and I see a potential issue : > > - scsi_requeue_command() drops reference to scsi_device->sdev_gendev in > scsi_unprep_request->scsi_put_command()->__scsi_put_command(). This can > trigger the

Re: [PATCH] scsi_lib: add NULL check to scsi_setup_blk_pc_cmnd

On Mon, 2012-07-23 at 15:24 -0400, Jörn Engel wrote: > On Mon, 23 July 2012 23:45:55 +0400, James Bottomley wrote: > > On Mon, 2012-07-23 at 13:32 -0400, Jörn Engel wrote: > > > At least two slightly different paths can lead to a NULL pointer > > > dereference in scsi_prep_state_check and have been

scsi_requeue_command and dropping reference to sdev_gendev

Hi : I am looking at the latest upstream code and I see a potential issue : - scsi_requeue_command() drops reference to scsi_device->sdev_gendev in scsi_unprep_request->scsi_put_command()->__scsi_put_command(). This can trigger the cleanup routine scsi_device_dev_release_usercontext() that will d

[PATCH] tcm_vhost: Expose ABI version via VHOST_SCSI_GET_ABI_VERSION

From: Nicholas Bellinger As requested by Anthony, here is a patch against target-pending/for-next-merge to expose an ABI version to userspace via a new VHOST_SCSI_GET_ABI_VERSION ioctl operation. As mentioned in the comment, ABI Rev 0 is for pre 2012 out-of-tree code, and ABI Rev 1 (the current

Re: [PATCH 5/5] block: genhd: add an interface to set disk's poll interval

On Mon, Jul 23, 2012 at 12:43:34PM -0600, Betty Dall wrote: > Hi Aaron, Hi, > > On Mon, 2012-07-23 at 14:49 +0800, Aaron Lu wrote: > > Set the ODD's in kernel poll interval to 2s for the user in case the > > user is using an old distro on which udev will not set the system wide > > block paramet

Re: [PATCH 4/5] scsi: pm: use runtime resume callback if available

Hi, On Mon, Jul 23, 2012 at 06:36:12PM +0400, Sergei Shtylyov wrote: > Hello. > > On 07/23/2012 10:49 AM, Aaron Lu wrote: > > > When runtime resume a scsi device, if the device's driver has > > implemented runtime resume callback, use that instead of the resume > > callback. > > > sr driver nee

Re: [PATCH] scsi_lib: add NULL check to scsi_setup_blk_pc_cmnd

On Mon, 23 July 2012 23:45:55 +0400, James Bottomley wrote: > On Mon, 2012-07-23 at 13:32 -0400, Jörn Engel wrote: > > At least two slightly different paths can lead to a NULL pointer > > dereference in scsi_prep_state_check and have been hit in practice. > > Have you checked this with the patches

Re: [PATCH] scsi_lib: add NULL check to scsi_setup_blk_pc_cmnd

On Mon, 2012-07-23 at 13:32 -0400, Jörn Engel wrote: > At least two slightly different paths can lead to a NULL pointer > dereference in scsi_prep_state_check and have been hit in practice. Have you checked this with the patches in scsi-misc? There's a series of patches in there that alters the w

[PATCH] scsi_lib: add NULL check to scsi_setup_blk_pc_cmnd

At least two slightly different paths can lead to a NULL pointer dereference in scsi_prep_state_check and have been hit in practice. 1. Call Trace: [] scsi_setup_blk_pc_cmnd+0x2b/0x170 [] sd_prep_fn+0x568/0xdd0 [] blk_peek_request+0xb4/0x240 [] scsi_request_fn+0x43e/0x4a0 [] __blk_run_queue+0x

Re: [PATCH 5/5] block: genhd: add an interface to set disk's poll interval

Hi Aaron, On Mon, 2012-07-23 at 14:49 +0800, Aaron Lu wrote: > Set the ODD's in kernel poll interval to 2s for the user in case the > user is using an old distro on which udev will not set the system wide > block parameter events_dfl_poll_msecs. Why did you pick 2 seconds? > > Signed-off-by: Aa

Re: 'Device not ready' issue on mpt2sas since 3.1.10

Hello, On Mon, Jul 23, 2012 at 01:14:00AM +0200, Matthias Prager wrote: > as it turns out my workaround (setting allow_restart=1) isn't all that > useful after all. There are no more i/o errors because the drive just > never goes to standby mode anymore (at least 'hdparm -y /dev/sda' does > not se

Re: [PATCH 2/5] scsi: sr: runtime pm when ODD is open/closed

On Mon, 23 Jul 2012, Aaron Lu wrote: > The ODD can either be runtime resumed by the user or by a software > request. And for the latter part, we only support runtime resume the ODD > when the eject request is received. We did this in sr's block ioctl > function, this looks ugly. > > Change this b

Re: [PATCH 1/5] scsi: sr: fix for sr suspend and resume

On Mon, 23 Jul 2012, Aaron Lu wrote: > In sr_suspend, we do not need to do anything if it is not a runtime pm > request, so just return by checking the PM_EVENT_AUTO flag. > And in sr_resume, only reset the suspend_count back to 1 if the ODD is > waken up by the user, or the usage count of the scs

Re: [PATCH 4/5] scsi: pm: use runtime resume callback if available

Hello. On 07/23/2012 10:49 AM, Aaron Lu wrote: > When runtime resume a scsi device, if the device's driver has > implemented runtime resume callback, use that instead of the resume > callback. > sr driver needs this to properly do different things for system resume > and runtime resume. > Signe

[PATCH 3/5] scsi: sr: block events when runtime suspended

When the ODD is runtime suspended, there is no need to poll it for events, so block events poll for it and unblock when resumed. Signed-off-by: Aaron Lu --- block/genhd.c | 2 ++ drivers/scsi/sr.c | 7 --- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/block/genhd.c b/blo