Re: [PATCH v7 3/7] cxl/region: Don't leak tolerated RAM -EFAULT from unmapped poison scan

[Jonathan Cameron]

On Thu, 3 Sep 2026 22:25:30 -0700
Alison Schofield <alison.schofield@xxxxxxxxx> wrote:

> On Wed, Sep 02, 2026 at 01:38:35PM +0800, Richard Cheng wrote:
> > cxl_get_poison_unmapped() tolerates the -EFAULT a RAM partition returns
> > for Get Poison List by skipping that partition, but left rc holding the
> > error. If the tolerated RAM fault was the last poison query before the
> > loop ended, the function returned a spurious -EFAULT and the poison-list

Good to call out why it is doing this just to remind the reader.
+ what the path to actually get -EFAULT is.  That might be relevant
to anyone considering if the bug effects them.

Unhelpfully (maybe fix this) cxl_internal_send_cmd() is documented as returning
-EFAULT for "Hardware Error Occured" whereas I think it comes from an
Invalid PA being passed in the query but I'm not 100% sure there isn't
another path.

> > read failed even though enumeration succeeded. Reset rc to 0 when
> > tolerating the fault, matching poison_by_decoder().  
> 
> Reviewed-by: Alison Schofield <alison.schofield@xxxxxxxxx>