[v4,0/9] Add alternative API for Xen PTEs
mbox series

Message ID cover.1575477921.git.hongyxia@amazon.com
Headers show
Series
  • Add alternative API for Xen PTEs
Related show

Message

Xia, Hongyan Dec. 4, 2019, 5:10 p.m. UTC
NOTE: My email address has changed due to some HR management. I have
lost all my previous emails and I could only salvage some of the
comments to v3 from the mailing list archive. I will reply to the
comments from v3 in this v4 series.

This batch adds an alternative alloc-map-unmap-free Xen PTE API to the
normal alloc-free on the xenheap, in preparation of switching to domheap
for Xen page tables. Since map and unmap are basically no-ops now, and
other changes are cosmetic to ease future patches, this batch does not
introduce any functional changes.

tree:
https://xenbits.xen.org/git-http/people/hx242/xen.git xen_pte_map-v4

v3: 
https://lists.xenproject.org/archives/html/xen-devel/2019-10/msg00304.html

---
Changed since v3:
- change my email address in all patches
- address many style issues in v3
- rebase

Changed since v2:
- split into a smaller series
- drop the clear_page optimisation as Wei suggests
- rebase

Changed since v1:
- squash some commits
- merge bug fixes into this first batch
- rebase against latest master

Wei Liu (9):
  x86: move some xen mm function declarations
  x86: introduce a new set of APIs to manage Xen page tables
  x86/mm: introduce l{1,2}t local variables to map_pages_to_xen
  x86/mm: introduce l{1,2}t local variables to modify_xen_mappings
  x86/mm: map_pages_to_xen would better have one exit path
  x86/mm: add an end_of_loop label in map_pages_to_xen
  x86/mm: make sure there is one exit path for modify_xen_mappings
  x86/mm: add an end_of_loop label in modify_xen_mappings
  x86/mm: change pl*e to l*t in virt_to_xen_l*e

 xen/arch/x86/mm.c          | 280 ++++++++++++++++++++++---------------
 xen/include/asm-x86/mm.h   |  16 +++
 xen/include/asm-x86/page.h |   5 -
 3 files changed, 186 insertions(+), 115 deletions(-)

Comments

Jan Beulich Dec. 5, 2019, 9:14 a.m. UTC | #1
On 04.12.2019 18:10, Hongyan Xia wrote:
> NOTE: My email address has changed due to some HR management. I have
> lost all my previous emails and I could only salvage some of the
> comments to v3 from the mailing list archive. I will reply to the
> comments from v3 in this v4 series.

I'm afraid this isn't very helpful. In particular, does this mean
v4 is effectively v3, i.e. no review comments taken care of? Or
just some of them, and others left out? I'm not fancying re-
reviewing a version that doesn't have prior comments taken care
of. Please clarify.

Jan
Xia, Hongyan Dec. 5, 2019, 9:41 a.m. UTC | #2
I have addressed the comments that I can find in the archive. Some big
debates (like _new or not _new, whether to modularise map_pages_to_xen)
have not been touched. Various acked-by and reviewed-by added.

Hongyan

On Thu, 2019-12-05 at 10:14 +0100, Jan Beulich wrote:
> On 04.12.2019 18:10, Hongyan Xia wrote:
> > NOTE: My email address has changed due to some HR management. I
> > have
> > lost all my previous emails and I could only salvage some of the
> > comments to v3 from the mailing list archive. I will reply to the
> > comments from v3 in this v4 series.
> 
> I'm afraid this isn't very helpful. In particular, does this mean
> v4 is effectively v3, i.e. no review comments taken care of? Or
> just some of them, and others left out? I'm not fancying re-
> reviewing a version that doesn't have prior comments taken care
> of. Please clarify.
> 
> Jan
Jan Beulich Dec. 5, 2019, 9:51 a.m. UTC | #3
On 05.12.2019 10:41, Xia, Hongyan wrote:
> I have addressed the comments that I can find in the archive.

That's still pretty vague - is there reason to assume you were
not able to find some comments there?

> Some big
> debates (like _new or not _new, whether to modularise map_pages_to_xen)
> have not been touched.

The _new suffix discussion you've meanwhile responded you,
which is therefore fine. The modularization question, otoh,
I don't recall seeing any reply for, and hence for now I'd
skip re-reviewing the respective patches. Furthermore, is
your use of "like" implying there were more than the two
examples you gave? As much as I can understand difficulties
on your part resulting from your email issues, please also
understand my reservations regarding having to re-do things
where quite a bit of time had already been invested into.

Jan
Xia, Hongyan Dec. 5, 2019, 10:45 a.m. UTC | #4
Okay let me be explicit this time. Cross checked with mails from
lore.kernel.org, issues not touched from v3 to v4:

- _new or not _new
- splitting map_pages_to_xen; introduction of labels. Although I just
responded to these two issues.
- const not added to suggested variables since a lot of them are stuck
with the old API for now. I can review relevant functions and const
qualify other applicable ones.

Hongyan

On Thu, 2019-12-05 at 10:51 +0100, Jan Beulich wrote:
> On 05.12.2019 10:41, Xia, Hongyan wrote:
> > I have addressed the comments that I can find in the archive.
> 
> That's still pretty vague - is there reason to assume you were
> not able to find some comments there?
> 
> > Some big
> > debates (like _new or not _new, whether to modularise
> > map_pages_to_xen)
> > have not been touched.
> 
> The _new suffix discussion you've meanwhile responded you,
> which is therefore fine. The modularization question, otoh,
> I don't recall seeing any reply for, and hence for now I'd
> skip re-reviewing the respective patches. Furthermore, is
> your use of "like" implying there were more than the two
> examples you gave? As much as I can understand difficulties
> on your part resulting from your email issues, please also
> understand my reservations regarding having to re-do things
> where quite a bit of time had already been invested into.
> 
> Jan