Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 18 Apr 2018 22:13:11 +0300
From:      Vitalij Satanivskij <satan@ukr.net>
To:        John Baldwin <jhb@freebsd.org>
Cc:        Vitalij Satanivskij <satan@ukr.net>, freebsd-current@freebsd.org, cem@freebsd.org, Stephen Hurd <shurd@llnw.com>, Matthew Macy <mmacy@nextbsd.org>, "freebsd-hackers@freebsd.org" <hackers@freebsd.org>, Stephen Hurd <shurd@freebsd.org>
Subject:   Re: Current panic on boot on H11DSI motherboard with epyc cpu (nexus_add_irq: failed)
Message-ID:  <20180418191311.GA74540@hell.ukr.net>
In-Reply-To: <1616582.sIejGazfcv@ralph.baldwin.cx>
References:  <20180416102710.GA90028@hell.ukr.net> <3628282.XVdngBdGlp@ralph.baldwin.cx> <20180418105649.GA9989@hell.ukr.net> <1616582.sIejGazfcv@ralph.baldwin.cx>

next in thread | previous in thread | raw e-mail | index | archive | help
JB> Ohhhh, this is a different issue.  Sorry.  As a hack, try changing
JB> 'FIRST_MSI_INT' to 512 in sys/amd64/include/intr_machdep.h.  The issue
JB> is that some systems now include more than 256 interrupt pins on I/O
JB> APICs, so IRQ 256 is already reserved for use by one of those
JB> interrupt pins.  The real fix is that I need to make FIRST_MSI_INT
JB> dynamic instead of a constant and just define it as the first free IRQ
JB> after the I/O APICs have probed.
JB> 

Yep. That it.

But just one note 

irq585: ccp14:721 @cpu0(domain0): 0
irq586: ccp14:723 @cpu0(domain0): 0
irq587: ccp15:725 @cpu0(domain0): 0

If I understand correctly number of irq's even more then 512, so better to change to real number in system ?

Or this is another case ?

Any way thank you for help. Now I can use system with msix and msi enabled.




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20180418191311.GA74540>