Re: [Bug-xorriso] building libisofs-1.4.4, libburn-1.4.4, libisoburn-1.4.4 on OpenBSD

[Thomas Schmitt]

Hi,

> Here is the log of OpenBSD-5.9/BH14NS48/PCI-IDE connection with IMMED bit.

The xorriso run looks like flawless Sense replies were delivered and like
the burn and blank runs were successful:

  SYNCHRONIZE CACHE
  35 02 00 00 00 00 00 00 00 00 
        240 us     [ 420112038 ]
  xorriso : UPDATE :  223 of  223 MB written (fifo  0%) [buf 100%]   2.4x.

  TEST UNIT READY
  00 00 00 00 00 00 
  +++ sense data = 71 00 02 00 00 00 00 0A 00 00 00 00 04 08 00 00 00 00
  +++ key=2  asc=04h  ascq=08h
        471 us     [ 420612904 ]
  ...
  
  TEST UNIT READY
  00 00 00 00 00 00 
        276 us     [ 449171695 ]
  ...

  Media status : is written , is closed
  Media summary: 1 session, 114243 data blocks,  223m data,     0 free

The drive needed 29 seconds for SYNCHRONIZE CACHE to complete.
Quite normal. (It depends on actual write speed and drive firmware.)
Subsequent inspection shows the expected media state.

The same with the xorriso -blank run:

  BLANK
  a1 11 00 00 00 00 00 00 00 00 00 00 
       1837 us     [ 5299934 ]

  TEST UNIT READY
  00 00 00 00 00 00 
  +++ sense data = 71 00 02 00 00 00 00 0A 00 00 00 00 04 07 00 80 00 00
  +++ key=2  asc=04h  ascq=07h
       1315 us     [ 5301504 ]
  ...

  TEST UNIT READY
  xorriso : UPDATE : Blanking  ( 99.0% done in 51 seconds )
  00 00 00 00 00 00 
       3145 us     [ 56805563 ]
  ...

  Media status : is blank
  Media summary: 0 sessions, 0 data blocks, 0 data,  703m free


> I will try on Linux later.

Not needed. Unless you experienced a failure with the OpenBSD test which
i missed when reading  pciide-xorriso-immed.log .

The cdrecord runs look ok, too. They report "retryable error" at the occasions
where the NOT READY Sense replies with correct ASC,ASCQ are shown in the
xorriso log. I.e. both programs experienced the same correct behavior.

-----------------------------------------------------------------------

This test confirms the theory that xorriso and the drives work with PCI-IDE.

We still have that aborted iHAS524 burn run log, which looks ok until it
suddenly ends: pciide-cdrw-2.log.
One should test again whether xorriso and iHAS524 work with PCI-IDE.
With or without -use_immed_bit "on" it will produce Sense replies, because
the drive has the habit to issue them when its write buffer is full.

-----------------------------------------------------------------------

When we are certain enough then it would be time to ask the AHCI experts
of OpenBSD what mangles the ASC,ASCQ of our NOT READY Sense replies to 0,0.

Are you involved in OpenBSD kernel programming ?
Do you have a proposal of a mailing list where to discuss this with
developers ?


Have a nice day :)

Thomas