netdev - RE: [E1000-devel] e1000 82546EB MSI support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <36D9DB17C6DE9E40B059440DB8D95F5205ABB3DA@orsmsx418.amr.corp.intel.com>
Date:	Mon, 21 Jul 2008 15:07:10 -0700
From:	"Brandeburg, Jesse" <jesse.brandeburg@...el.com>
To:	"Simon Arlott" <simon@...e.lp0.eu>, <netdev@...r.kernel.org>,
	<e1000-devel@...ts.sourceforge.net>
Subject: RE: [E1000-devel] e1000 82546EB MSI support

Simon Arlott wrote:
> MSI support was added to e1000 a long time ago:
> 
> Commit fa4f7ef3aaa6cee6b04ebe90266ee893e0b2ce07 (Wed Nov 1 08:48:10
> 2006 -0800) [e1000: MSI support for PCI-e adapters]
> +       if(adapter->hw.mac_type > e1000_82547_rev_2) {
> +               adapter->have_msi = TRUE;
> +               if((err = pci_enable_msi(adapter->pdev))) {
> +                       DPRINTK(PROBE, ERR,
> +                        "Unable to allocate MSI interrupt Error:
> %d\n", err); +                       adapter->have_msi = FALSE;
> +               }
> +       }
> 
> and commit 9ac98284428961bd5be285a6cc1f5e6f5b6644aa (Thu Apr 28
> 19:39:13 2005 -0700) [e1000: add dynamic generic MSI interrupt
> routine] -       if (adapter->hw.mac_type > e1000_82547_rev_2) {
> +       if (adapter->hw.mac_type >= e1000_82571) {
> 
> Is there are reason why it's only enabled for >= e1000_82571? This
> isn't explained in 
> the commit descriptions or in comments (aside from mentioning that
> the support is for 
> PCI-e cards).
> 
> 82546EB appears to support it (lspci below), but I get multiple TX
> hangs if I enable it: 

That hardware is PCI/PCI-X only and our first design to use MSI-X
interrupt (single vector message based interrupt).  Unfortunately it
didn't work[1].  The solution from the driver side was to make sure that
the driver never enabled it on those parts.  The only parts for gigabit
that correctly support MSI-X are the 82574, 82575 and 82576 based parts.
MSI (single vector) is supported by >= 82571, i.e. 82571/2/3, ESB2,
ICH8/9/10.

Jesse

[1] - http://download.intel.com/design/network/specupdt/82546eb.pdf
see item 26.  The same problem exists on 82546GB as well.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html