Re: [PATCH 8/9] PCI: print out suggestion about using pci=realloc

[Jesse Barnes <jbarnes@virtuousgeek.org>]

[-- Attachment #1: Type: text/plain, Size: 1292 bytes --]

On Sat,  4 Feb 2012 22:55:07 -0800
Yinghai Lu <yinghai@kernel.org> wrote:

> let user know they could try if pci=realloc could help.
> 
> Suggested-by: Jesse Barnes <jbarnes@virtuousgeek.org>
> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
> ---
>  drivers/pci/setup-bus.c |    3 +++
>  1 files changed, 3 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
> index 22454c5..9526038 100644
> --- a/drivers/pci/setup-bus.c
> +++ b/drivers/pci/setup-bus.c
> @@ -1346,6 +1346,9 @@ again:
>  		goto enable_and_dump;
>  
>  	if (tried_times >= pci_try_num) {
> +		if (pci_realloc_enable == enable_not_set)
> +			printk(KERN_INFO "Some pci devices resources are not assigned, please try to boot with pci=realloc\n");
> +
>  		free_list(&fail_head);
>  		goto enable_and_dump;
>  	}

"Some PCI device resources are unassigned, try booting with pci=realloc"

That said, I wonder if this will cause more problems than it solves.
In some cases the BIOS may not allocate all resources, but this will be
harmless since the user won't actually use the device, or the driver
for the device won't even use the BARs in question anyway (as happens a
lot for IO regions).

-- 
Jesse Barnes, Intel Open Source Technology Center

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 836 bytes --]