[lkml]   [2013]   [Jan]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH v7u1 26/31] x86: Don't enable swiotlb if there is not enough ram for it
Yinghai Lu <> writes:

> On Mon, Jan 7, 2013 at 7:50 PM, Yinghai Lu <> wrote:
>> On Mon, Jan 7, 2013 at 7:13 PM, Eric W. Biederman <> wrote:
>>> I meant we should detect failure to allocate bounce buffers in in
>>> swiotlb_init() instead of panicing.
>>> I meant swiotlb_map_single() should either panic or simply fail.
>>> If I have read lib/swiotlb.c correctly the only place we allocate a
>>> bounce buffer is in swiotlb_map_single. If there are more places we can
>>> allocate bounce buffers those need to be handled as well.
>> ok, will give it a try.
> please check if you are ok with attached.

It looks like the right direction. Certainly enough to test and see if
the code will work.

I don't see the point of adding a nopanic case to the swiotlb
initialization. That just looks like unnecessary complications.
Certainly a nopanic case implemented by passing a nopanic parameter
looks like the wrong way to go. At most you want to return an error
code and do:

if (swiotlb_init_with_default_size() == -ENOMEM)
panic("Cannot allocate SWIOTLB buffer");

The page freeing in swiotlb_init_with_tbl appears to be in the wrong
function. I suggest looking at swiotlb_late_init which apparently is
allowed to fail for some ideas.

> looks like it need more change of lines.

The size of the change matters less than how clean and maintainable the
result is. If done carefully I expect you can have net fewer lines but
not needing to handle the case when the swiotlb apis are unavailable.


 \ /
  Last update: 2013-01-09 01:21    [W:0.097 / U:0.056 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site