Re: mpt irq timeout problem after reboot - only if non-verbose booting !?!

[Harald Schmalzbauer]

 schrieb John Baldwin am 17.10.2012 19:19 (localtime):
 Are you using any RAID volumes?  The only shutdown handler in mpt that looks
 like it might want interrupts to work is mpt_raid_shutdown().  It needs to use
 polled I/O instead of disabling interrupts I think.  Try this:

 Index: mpt_raid.c
 ===
 --- mpt_raid.c(revision 241641)
 +++ mpt_raid.c(working copy)
 @@ -115,7 +115,7 @@ static timeout_t mpt_raid_timer;
  static void mpt_enable_vol(struct mpt_softc *mpt,
  struct mpt_raid_volume *mpt_vol, int enable);
  #endif
 -static void mpt_verify_mwce(struct mpt_softc *, struct mpt_raid_volume *);
 +static void mpt_verify_mwce(struct mpt_softc *, struct mpt_raid_volume *, 
 int);
  static void mpt_adjust_queue_depth(struct mpt_softc *, struct 
 mpt_raid_volume *,
  struct cam_path *);
  #if __FreeBSD_version  50
 @@ -135,7 +135,7 @@ static void mpt_disk_prt(struct mpt_softc *mpt, st
  static int mpt_issue_raid_req(struct mpt_softc *mpt,
  struct mpt_raid_volume *vol, struct mpt_raid_disk *disk, request_t *req,
  u_int Action, uint32_t ActionDataWord, bus_addr_t addr, bus_size_t len,
 -int write, int wait);
 +int write, int wait, int sleep_ok);
  
  static int mpt_refresh_raid_data(struct mpt_softc *mpt);
  static void mpt_schedule_raid_refresh(struct mpt_softc *mpt);
 @@ -517,7 +517,7 @@ mpt_raid_shutdown(struct mpt_softc *mpt)
  
   mpt-raid_mwce_setting = MPT_RAID_MWCE_OFF;
   RAID_VOL_FOREACH(mpt, mpt_vol) {
 - mpt_verify_mwce(mpt, mpt_vol);
 + mpt_verify_mwce(mpt, mpt_vol, FALSE);
   }
  }
  
 @@ -592,7 +592,7 @@ static int
  mpt_issue_raid_req(struct mpt_softc *mpt, struct mpt_raid_volume *vol,
  struct mpt_raid_disk *disk, request_t *req, u_int Action,
  uint32_t ActionDataWord, bus_addr_t addr, bus_size_t len,
 -int write, int wait)
 +int write, int wait, int sleep_ok)
  {
   MSG_RAID_ACTION_REQUEST *rap;
   SGE_SIMPLE32 *se;
 @@ -623,7 +623,7 @@ mpt_issue_raid_req(struct mpt_softc *mpt, struct m
  
   if (wait) {
   return (mpt_wait_req(mpt, req, REQ_STATE_DONE, REQ_STATE_DONE,
 -  /*sleep_ok*/FALSE, /*time_ms*/2000));
 +  sleep_ok, /*time_ms*/2000));
   } else {
   return (0);
   }
 @@ -763,7 +763,7 @@ mpt_raid_quiesce_disk(struct mpt_softc *mpt, struc
   MPI_RAID_ACTION_QUIESCE_PHYS_IO,
   /*ActionData*/0, /*addr*/0,
   /*len*/0, /*write*/FALSE,
 - /*wait*/FALSE);
 + /*wait*/FALSE, /*sleep_ok*/FALSE);
   if (rv != 0)
   return (CAM_REQ_CMP_ERR);
  
 @@ -882,7 +882,7 @@ mpt_enable_vol(struct mpt_softc *mpt, struct mpt_r
   enable ? MPI_RAID_ACTION_ENABLE_VOLUME
  : MPI_RAID_ACTION_DISABLE_VOLUME,
   /*data*/0, /*addr*/0, /*len*/0,
 - /*write*/FALSE, /*wait*/TRUE);
 + /*write*/FALSE, /*wait*/TRUE, /*sleep_ok*/TRUE);
   if (rv == ETIMEDOUT) {
   mpt_vol_prt(mpt, mpt_vol, mpt_enable_vol: 
   %s Volume Timed-out\n,
 @@ -903,7 +903,8 @@ mpt_enable_vol(struct mpt_softc *mpt, struct mpt_r
  #endif
  
  static void
 -mpt_verify_mwce(struct mpt_softc *mpt, struct mpt_raid_volume *mpt_vol)
 +mpt_verify_mwce(struct mpt_softc *mpt, struct mpt_raid_volume *mpt_vol,
 +int sleep_ok)
  {
   request_t *req;
   struct mpt_raid_action_result *ar;
 @@ -950,7 +951,7 @@ static void
   return;
   }
  
 - req = mpt_get_request(mpt, /*sleep_ok*/TRUE);
 + req = mpt_get_request(mpt, sleep_ok);
   if (req == NULL) {
   mpt_vol_prt(mpt, mpt_vol,
   mpt_verify_mwce: Get request failed!\n);
 @@ -965,7 +966,7 @@ static void
   rv = mpt_issue_raid_req(mpt, mpt_vol, /*disk*/NULL, req,
   MPI_RAID_ACTION_CHANGE_VOLUME_SETTINGS,
   data, /*addr*/0, /*len*/0,
 - /*write*/FALSE, /*wait*/TRUE);
 + /*write*/FALSE, /*wait*/TRUE, sleep_ok);
   if (rv == ETIMEDOUT) {
   mpt_vol_prt(mpt, mpt_vol, mpt_verify_mwce: 
   Write Cache Enable Timed-out\n);
 @@ -1018,7 +1019,8 @@ mpt_verify_resync_rate(struct mpt_softc *mpt, stru
   rv = mpt_issue_raid_req(mpt, mpt_vol, /*disk*/NULL, req,
   MPI_RAID_ACTION_SET_RESYNC_RATE,
   mpt-raid_resync_rate, /*addr*/0,
 - /*len*/0, 

Re: mpt irq timeout problem after reboot - only if non-verbose booting !?!

[John Baldwin]

On Wednesday, October 17, 2012 3:14:52 pm Harald Schmalzbauer (mobil) wrote:
 -Ursprüngliche Nachricht-
  Von: John Baldwin j...@freebsd.org
  An: freebsd-stable@freebsd.org
  Cc: h.schmalzba...@omnilan.de
  Gesendet: 17.10.'12,  20:46
  
  On Tuesday, October 16, 2012 5:24:44 am Harald Schmalzbauer wrote:
   Hello,
  
  I have 9.1-RC2 running in an ESXi 5.1 guest.
  I use 'lsisas' as virtual SCSI-Controller and mpt attaches and finds 1068E.
  
  Everything is working fine until the first 'shutdown -r now':
  The second boot pauses for ~2 minutes after probing disks and continues
  with this error:
  mpt0: Timedout requests already complete. Interrupts may not be 
  functioning.
  
  To be clear, you only see this at the end of reboot, and the hardware is 
  fine
  once the machine is back up?
 .
 
 Thanks for your attention!
 The timeout occurs after the first 'shutdown -r' while device probing during
 second boot process.  Perhaps this is amd64 specific. Today I had a new i386
 setup which doesn't exhibit this timeout. But it's on different hardware and
 hv-host was 5.0 inestead 5.1. So not really representative...

Hmmm, ok.  In that case my patch is not relevant.  It would only fix that
message occuring during the shutdown.

-- 
John Baldwin
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to freebsd-stable-unsubscr...@freebsd.org

Re: mpt irq timeout problem after reboot - only if non-verbose booting !?!

[John Baldwin]

On Tuesday, October 16, 2012 5:24:44 am Harald Schmalzbauer wrote:
  Hello,
 
 I have 9.1-RC2 running in an ESXi 5.1 guest.
 I use 'lsisas' as virtual SCSI-Controller and mpt attaches and finds 1068E.
 
 Everything is working fine until the first 'shutdown -r now':
 The second boot pauses for ~2 minutes after probing disks and continues
 with this error:
 mpt0: Timedout requests already complete. Interrupts may not be functioning.

To be clear, you only see this at the end of reboot, and the hardware is fine
once the machine is back up?

 This problem was also obeserved with real 1068 hardware:
 http://lists.freebsd.org/pipermail/freebsd-stable/2011-September/063937.html
 
 When I power off the virtual machine instead of rebooting, the problem
 doesn't occur.
 
 Accidentally I found a workarround ;-) :
 If I set 'verbose_boot' in loader.conf, the problem vanisehs!?!?!?
 
 Any idea how „verbose_boot“ affects the operation of the mpt driver?

Extra printfs affect the timing most likely.

Are you using any RAID volumes?  The only shutdown handler in mpt that looks
like it might want interrupts to work is mpt_raid_shutdown().  It needs to use
polled I/O instead of disabling interrupts I think.  Try this:

Index: mpt_raid.c
===
--- mpt_raid.c  (revision 241641)
+++ mpt_raid.c  (working copy)
@@ -115,7 +115,7 @@ static timeout_t mpt_raid_timer;
 static void mpt_enable_vol(struct mpt_softc *mpt,
   struct mpt_raid_volume *mpt_vol, int enable);
 #endif
-static void mpt_verify_mwce(struct mpt_softc *, struct mpt_raid_volume *);
+static void mpt_verify_mwce(struct mpt_softc *, struct mpt_raid_volume *, int);
 static void mpt_adjust_queue_depth(struct mpt_softc *, struct mpt_raid_volume 
*,
 struct cam_path *);
 #if __FreeBSD_version  50
@@ -135,7 +135,7 @@ static void mpt_disk_prt(struct mpt_softc *mpt, st
 static int mpt_issue_raid_req(struct mpt_softc *mpt,
 struct mpt_raid_volume *vol, struct mpt_raid_disk *disk, request_t *req,
 u_int Action, uint32_t ActionDataWord, bus_addr_t addr, bus_size_t len,
-int write, int wait);
+int write, int wait, int sleep_ok);
 
 static int mpt_refresh_raid_data(struct mpt_softc *mpt);
 static void mpt_schedule_raid_refresh(struct mpt_softc *mpt);
@@ -517,7 +517,7 @@ mpt_raid_shutdown(struct mpt_softc *mpt)
 
mpt-raid_mwce_setting = MPT_RAID_MWCE_OFF;
RAID_VOL_FOREACH(mpt, mpt_vol) {
-   mpt_verify_mwce(mpt, mpt_vol);
+   mpt_verify_mwce(mpt, mpt_vol, FALSE);
}
 }
 
@@ -592,7 +592,7 @@ static int
 mpt_issue_raid_req(struct mpt_softc *mpt, struct mpt_raid_volume *vol,
   struct mpt_raid_disk *disk, request_t *req, u_int Action,
   uint32_t ActionDataWord, bus_addr_t addr, bus_size_t len,
-  int write, int wait)
+  int write, int wait, int sleep_ok)
 {
MSG_RAID_ACTION_REQUEST *rap;
SGE_SIMPLE32 *se;
@@ -623,7 +623,7 @@ mpt_issue_raid_req(struct mpt_softc *mpt, struct m
 
if (wait) {
return (mpt_wait_req(mpt, req, REQ_STATE_DONE, REQ_STATE_DONE,
-/*sleep_ok*/FALSE, /*time_ms*/2000));
+sleep_ok, /*time_ms*/2000));
} else {
return (0);
}
@@ -763,7 +763,7 @@ mpt_raid_quiesce_disk(struct mpt_softc *mpt, struc
MPI_RAID_ACTION_QUIESCE_PHYS_IO,
/*ActionData*/0, /*addr*/0,
/*len*/0, /*write*/FALSE,
-   /*wait*/FALSE);
+   /*wait*/FALSE, /*sleep_ok*/FALSE);
if (rv != 0)
return (CAM_REQ_CMP_ERR);
 
@@ -882,7 +882,7 @@ mpt_enable_vol(struct mpt_softc *mpt, struct mpt_r
enable ? MPI_RAID_ACTION_ENABLE_VOLUME
   : MPI_RAID_ACTION_DISABLE_VOLUME,
/*data*/0, /*addr*/0, /*len*/0,
-   /*write*/FALSE, /*wait*/TRUE);
+   /*write*/FALSE, /*wait*/TRUE, /*sleep_ok*/TRUE);
if (rv == ETIMEDOUT) {
mpt_vol_prt(mpt, mpt_vol, mpt_enable_vol: 
%s Volume Timed-out\n,
@@ -903,7 +903,8 @@ mpt_enable_vol(struct mpt_softc *mpt, struct mpt_r
 #endif
 
 static void
-mpt_verify_mwce(struct mpt_softc *mpt, struct mpt_raid_volume *mpt_vol)
+mpt_verify_mwce(struct mpt_softc *mpt, struct mpt_raid_volume *mpt_vol,
+int sleep_ok)
 {
request_t *req;
struct mpt_raid_action_result *ar;
@@ -950,7 +951,7 @@ static void
return;
}
 
-   req = mpt_get_request(mpt, /*sleep_ok*/TRUE);
+   req = mpt_get_request(mpt, sleep_ok);
if (req == NULL) {
mpt_vol_prt(mpt, 

Re: mpt irq timeout problem after reboot - only if non-verbose booting !?!

[Harald Schmalzbauer (mobil)]

-Ursprüngliche Nachricht-

Von: John Baldwin j...@freebsd.org
An: freebsd-stable@freebsd.org
Cc: h.schmalzba...@omnilan.de
Gesendet: 17.10.'12,  20:46

On Tuesday, October 16, 2012 5:24:44 am Harald Schmalzbauer wrote:

 Hello,

I have 9.1-RC2 running in an ESXi 5.1 guest.
I use 'lsisas' as virtual SCSI-Controller and mpt attaches and finds 1068E.

Everything is working fine until the first 'shutdown -r now':
The second boot pauses for ~2 minutes after probing disks and continues
with this error:
mpt0: Timedout requests already complete. Interrupts may not be functioning.


To be clear, you only see this at the end of reboot, and the hardware is fine
once the machine is back up?

.

Thanks for your attention!
The timeout occurs after the first 'shutdown -r' while device probing during 
second boot process.
Perhaps this is amd64 specific. Today I had a new i386 setup which doesn't 
exhibit this timeout. But it's on different hardware and hv-host was 5.0 
inestead 5.1. So not really representative...


Extra printfs affect the timing most likely.

Are you using any RAID volumes?  The only shutdown handler in mpt that looks


The controller is 'virtual' SAS. But thers's been reports that also on real HW 
(sasuc8i) the same problem occurs.

I'll try your patch tomorrow morning; timezon shift...

Thanks,

-Harry
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to freebsd-stable-unsubscr...@freebsd.org