Re: [PATCH drm-next v2 04/16] maple_tree: add flag MT_FLAGS_LOCK_NONE

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 2/21/23 19:31, Matthew Wilcox wrote:
On Tue, Feb 21, 2023 at 03:37:49PM +0100, Danilo Krummrich wrote:
On Mon, Feb 20, 2023 at 08:33:35PM +0000, Matthew Wilcox wrote:
On Mon, Feb 20, 2023 at 06:06:03PM +0100, Danilo Krummrich wrote:
On 2/20/23 16:10, Matthew Wilcox wrote:
This is why we like people to use the spinlock embedded in the tree.
There's nothing for the user to care about.  If the access really is
serialised, acquiring/releasing the uncontended spinlock is a minimal
cost compared to all the other things that will happen while modifying
the tree.

I think as for the users of the GPUVA manager we'd have two cases:

1) Accesses to the manager (and hence the tree) are serialized, no lock
needed.

2) Multiple operations on the tree must be locked in order to make them
appear atomic.

Could you give an example here of what you'd like to do?  Ideally
something complicated so I don't say "Oh, you can just do this" when
there's a more complex example for which "this" won't work.  I'm sure
that's embedded somewhere in the next 20-odd patches, but it's probably
quicker for you to describe in terms of tree operations that have to
appear atomic than for me to try to figure it out.


Absolutely, not gonna ask you to read all of that. :-)

One thing the GPUVA manager does is to provide drivers the (sub-)operations
that need to be processed in order to fulfill a map or unmap request from
userspace. For instance, when userspace asks the driver to map some memory
the GPUVA manager calculates which existing mappings must be removed, split up
or can be merged with the newly requested mapping.

A driver has two ways to fetch those operations from the GPUVA manager. It can
either obtain a list of operations or receive a callback for each operation
generated by the GPUVA manager.

In both cases the GPUVA manager walks the maple tree, which keeps track of
existing mappings, for the given range in __drm_gpuva_sm_map() (only considering
the map case, since the unmap case is a subset basically). For each mapping
found in the given range the driver, as mentioned, either receives a callback or
a list entry is added to the list of operations.

Typically, for each operation / callback one entry within the maple tree is
removed and, optionally at the beginning and end of a new mapping's range, a
new entry is inserted. An of course, as the last operation, there is the new
mapping itself to insert.

The GPUVA manager delegates locking responsibility to the drivers. Typically,
a driver either serializes access to the VA space managed by the GPUVA manager
(no lock needed) or need to lock the processing of a full set of operations
generated by the GPUVA manager.

OK, that all makes sense.  It does make sense to have the driver use its
own mutex and then take the spinlock inside the maple tree code.  It
shouldn't ever be contended.

In either case the embedded spinlock wouldn't be useful, we'd either need an
external lock or no lock at all.

If there are any internal reasons why specific tree operations must be
mutually excluded (such as those you explain below), wouldn't it make more
sense to always have the internal lock and, optionally, allow users to
specify an external lock additionally?

So the way this works for the XArray, which is a little older than the
Maple tree, is that we always use the internal spinlock for
modifications (possibly BH or IRQ safe), and if someone wants to
use an external mutex to make some callers atomic with respect to each
other, they're free to do so.  In that case, the XArray doesn't check
the user's external locking at all, because it really can't know.

I'd advise taking that approach; if there's really no way to use the
internal spinlock to make your complicated updates appear atomic
then just let the maple tree use its internal spinlock, and you can
also use your external mutex however you like.


That sounds like the right thing to do.

However, I'm using the advanced API of the maple tree (and that's the reason
why the above example appears a little more detailed than needed) because I
think with the normal API I can't insert / remove tree entries while walking
the tree, right?

Right.  The normal API is for simple operations while the advanced API
is for doing compound operations.

As by the documentation the advanced API, however, doesn't take care of locking
itself, hence just letting the maple tree use its internal spinlock doesn't
really work - I need to take care of that myself, right?

Yes; once you're using the advanced API, you get to compose the entire
operation yourself.

It feels a bit weird that I, as a user of the API, would need to lock certain
(or all?) mas_*() functions with the internal spinlock in order to protect
(future) internal features of the tree, such as the slab cache defragmentation
you mentioned. Because from my perspective, as the generic component that tells
it's users (the drivers) to take care of locking VA space operations (and hence
tree operations) I don't have an own purpose of this internal spinlock, right?

You don't ... but we can't know that.

Thanks for the clarification. I think I should now know what to for the GPUVA manager in terms of locking the maple tree in general.

Though I still have very limited insights on the maple tree I want to share some further thoughts.

From what I got so far it really seems to me that it would be better to just take the internal spinlock for both APIs (normal and advanced) whenever you need to internally.

This way users would not need to take care of locking maple tree internals, which I still think is a little odd.

Another plus would probably be maintainability. Once you got quite a few maple tree users using external locks (either in the sense of calling mt_set_external_lock() or in the way I'll potentially do it by using the internal lock with the advanced API and an additional external lock) it might be hard to apply any changes to the locking requirements, because you would either need to check every users implementation by hand or be able to run it in order to check it with lockdep.

If I got this correctly (please tell me if I don't) the only reason the internal lock is not managed by the advanced API internally is to let users do more complex transactions, without the need of having a separate external lock, as long as they fulfill the locking requirements of the maple tree, which are enforced by lockdep.

However, you already mentioned that "acquiring/releasing the uncontended spinlock is a minimal cost compared to all the other things that will happen while modifying the tree".

Do I miss something?


Also I'm a little confused how I'd know where to take the spinlock? E.g. for
inserting entries in the tree I use mas_store_gfp() with GFP_KERNEL.

Lockdep will shout at you if you get it wrong ;-)  But you can safely
take the spinlock before calling mas_store_gfp(GFP_KERNEL) because
mas_nomem() knows to drop the lock before doing a sleeping allocation.
Essentially you're open-coding mtree_store_range() but doing your own
thing in addition to the store.

Just asking lockdep was my plan already, however I thought I still better ask. :D

If you will keep the current approach of handling the internal lock I think its necessary to somewhere document where users need to take the lock and, even more important, where the maple tree implementation will drop the lock.

For instance, if I would rely on using the internal spinlock for locking sets of transactions to the maple tree this would cause nasty bugs if I use functions like mas_store_gfp() dropping the lock. Even though I must admit that this is not a great example, since it should raise some red flags if a user would expect a spinlock is held for a sleeping allocation without questioning it. However, you get the point I guess.




[Index of Archives]     [Linux DRI Users]     [Linux Intel Graphics]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux