Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-19 Thread Bjorn Helgaas
On Tue, Mar 13, 2018 at 04:45:49PM -0500, Bjorn Helgaas wrote: > From: Bjorn Helgaas > > If a device already has MSI or MSI-X enabled, there's no need to set up its > legacy INTx interrupt. > > bba6f6fc68e7 ("[PATCH] MSI-X: fix resume crash") changed the cris, frv, > x86,

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-19 Thread Bjorn Helgaas
On Tue, Mar 13, 2018 at 04:45:49PM -0500, Bjorn Helgaas wrote: > From: Bjorn Helgaas > > If a device already has MSI or MSI-X enabled, there's no need to set up its > legacy INTx interrupt. > > bba6f6fc68e7 ("[PATCH] MSI-X: fix resume crash") changed the cris, frv, > x86, and ia64 arches to

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Andy Shevchenko
On Wed, Mar 14, 2018 at 7:49 PM, Bjorn Helgaas wrote: > On Wed, Mar 14, 2018 at 06:31:38PM +0200, Andy Shevchenko wrote: >> On Tue, Mar 13, 2018 at 11:45 PM, Bjorn Helgaas wrote: >> > If a device already has MSI or MSI-X enabled, there's no need to set up

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Andy Shevchenko
On Wed, Mar 14, 2018 at 7:49 PM, Bjorn Helgaas wrote: > On Wed, Mar 14, 2018 at 06:31:38PM +0200, Andy Shevchenko wrote: >> On Tue, Mar 13, 2018 at 11:45 PM, Bjorn Helgaas wrote: >> > If a device already has MSI or MSI-X enabled, there's no need to set up its >> > legacy INTx interrupt. >> >>

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Bjorn Helgaas
On Wed, Mar 14, 2018 at 06:31:38PM +0200, Andy Shevchenko wrote: > On Tue, Mar 13, 2018 at 11:45 PM, Bjorn Helgaas wrote: > > From: Bjorn Helgaas > > Agree with Christoph's comment. > > > If a device already has MSI or MSI-X enabled, there's no need to

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Bjorn Helgaas
On Wed, Mar 14, 2018 at 06:31:38PM +0200, Andy Shevchenko wrote: > On Tue, Mar 13, 2018 at 11:45 PM, Bjorn Helgaas wrote: > > From: Bjorn Helgaas > > Agree with Christoph's comment. > > > If a device already has MSI or MSI-X enabled, there's no need to set up its > > legacy INTx interrupt. >

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Bjorn Helgaas
On Wed, Mar 14, 2018 at 01:35:19AM -0700, Christoph Hellwig wrote: > Should this logic go into a little helper so that everyone is kept > in sync? Great point, thanks! What I'd really like is to completely get rid of most of the pcibios_enable_device() implementations. Most of them contain

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Bjorn Helgaas
On Wed, Mar 14, 2018 at 01:35:19AM -0700, Christoph Hellwig wrote: > Should this logic go into a little helper so that everyone is kept > in sync? Great point, thanks! What I'd really like is to completely get rid of most of the pcibios_enable_device() implementations. Most of them contain

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Andy Shevchenko
On Tue, Mar 13, 2018 at 11:45 PM, Bjorn Helgaas wrote: > From: Bjorn Helgaas > Agree with Christoph's comment. > If a device already has MSI or MSI-X enabled, there's no need to set up its > legacy INTx interrupt. Just point to the actual behaviour of

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Andy Shevchenko
On Tue, Mar 13, 2018 at 11:45 PM, Bjorn Helgaas wrote: > From: Bjorn Helgaas > Agree with Christoph's comment. > If a device already has MSI or MSI-X enabled, there's no need to set up its > legacy INTx interrupt. Just point to the actual behaviour of this. In some cases code in question has

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Rafael J. Wysocki
On 3/13/2018 10:45 PM, Bjorn Helgaas wrote: From: Bjorn Helgaas If a device already has MSI or MSI-X enabled, there's no need to set up its legacy INTx interrupt. bba6f6fc68e7 ("[PATCH] MSI-X: fix resume crash") changed the cris, frv, x86, and ia64 arches to skip INTx

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Rafael J. Wysocki
On 3/13/2018 10:45 PM, Bjorn Helgaas wrote: From: Bjorn Helgaas If a device already has MSI or MSI-X enabled, there's no need to set up its legacy INTx interrupt. bba6f6fc68e7 ("[PATCH] MSI-X: fix resume crash") changed the cris, frv, x86, and ia64 arches to skip INTx setup when MSI is

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Christoph Hellwig
Should this logic go into a little helper so that everyone is kept in sync?

Re: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-14 Thread Christoph Hellwig
Should this logic go into a little helper so that everyone is kept in sync?

RE: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-13 Thread Luck, Tony
> Change the remaining arches (cris, frv, and ia64) to skip INTx setup when > either MSI or MSI-X is enabled. My ia64 still boots with this patch applied (and doesn't have any new/different messages in the console log). So: Tested-by: Tony Luck -Tony

RE: [PATCH] PCI/MSI: Don't set up INTx if MSI or MSI-X is enabled

2018-03-13 Thread Luck, Tony
> Change the remaining arches (cris, frv, and ia64) to skip INTx setup when > either MSI or MSI-X is enabled. My ia64 still boots with this patch applied (and doesn't have any new/different messages in the console log). So: Tested-by: Tony Luck -Tony