Messages in this thread | | | Date | Fri, 20 Dec 2019 21:23:46 +0100 | From | Peter Zijlstra <> | Subject | Re: [RFC PATCH v1] devres: align devres.data strictly only for devm_kmalloc() |
| |
On Fri, Dec 20, 2019 at 07:32:16PM +0000, Alexey Brodkin wrote:
> Well it somehow used to work for quite some time now with the data-buffer > being allocated with 4 words offset (which is 16 bytes for 32-bit platform
3 words, devres_node is 3 words.
Which is exactly why we had to change it, the odd alignment caused ARC to explode.
> and 32 for 64-bit which is still much less than mentioned 128 bytes). > Or we just never managed to identify those rare cases when data corruption > really happened?
The races are rather rare methinks, you'd have to get a list-op concurrently with a DMA.
If you get the list corrupted, I'm thinking the crash is fairly likely, albeit really difficuly to debug.
> > No matter which way round you allocate devres and data, by necessity > > they're always going to consume the same total amount of memory. > > So then the next option I guess is to separate meta-data from data buffers > completely. Are there any reasons to not do that
Dunno, should work just fine I think.
> other than the hack we're > discussing here (meta-data in the beginning of the buffer) used to work OK-ish?
If meta-data at the beginngin used to work, I don't see why meta-data at the end wouldn't work equally well. They'd be equally broken.
But I'm still flabbergasted at the fact that they're doing non-coherent DMA to kmalloc memory, I thought we had a DMA api for that, with a special allocator and everything (but what do I know, I've never used that).
| |