#1362 closed defect (fixed)

bsp_reset() on older shared powerpc's

Reported by: Matt Rippa Owned by: Joel Sherrill
Priority: normal Milestone: 4.9
Component: bsps Version: 4.9
Severity: normal Keywords:
Cc: mrippa@… Blocked By:


Talked with Till-

This affects mvme2300...mvme2700:

There's a bug in libbsp/powerpc/shared/console/reboot.c which attempted a board reset via the keyboard port. This resulted in a NOOP (i.e., nothing happened and the board does not reset). I've proposed a patch which sets bit 1, port 0x92. This successfully resets *only* the board and I've also verified it does not reset the VME bus as described below.

Tests performed:

  1. I set bit 1 port 0x92 and removed the keyboard call. I verified that exit did (at least) a board reset. That works.
  1. I also performed an exit with the board SYSCON disabled (J20). I again verified the exit performed a board reset, but it did not affect any peripheral boards.
  1. SYSCON reenabled now, I included a peripheral board (vmic5588) and initialized it to make the fail light turn off. I again issued an exit and the 5588 fail light remained off, while the 2700 reset.
  1. I verified pushing the RST on the 2700 that my 5588 fail light transition from OFF->fail, implying this is what I should see during a VME bus reset.

So I think this is a sufficient test indicating the board reset works independently of resetting the VME bus.

Attachments (1)

reboot.patch (620 bytes) - added by Matt Rippa on Jan 26, 2009 at 9:32:59 PM.
patch for reboot.c

Download all attachments as: .zip

Change History (2)

Changed on Jan 26, 2009 at 9:32:59 PM by Matt Rippa

Attachment: reboot.patch added

patch for reboot.c

comment:1 Changed on Feb 11, 2009 at 3:14:50 PM by Joel Sherrill

Resolution: fixed
Status: newclosed

Patch applied to 4.9 and CVS head. Thanks.

Note: See TracTickets for help on using tickets.