Re: [PATCH] null_blk: mem garbage on NUMA systems during init

2013-12-15 Thread Jens Axboe
On Sun, Dec 15 2013, Linus Torvalds wrote: > Jens, ping? > > Patch looks fine, and it looks completely bogus to set > > null_mq_reg.nr_hw_queues > > to 'nr_online_nodes', when we've potentially done "setup_queues()" > with just a single queue. > > So I'm going to apply this as-is, I'm

Re: [PATCH] null_blk: mem garbage on NUMA systems during init

2013-12-15 Thread Linus Torvalds
Jens, ping? Patch looks fine, and it looks completely bogus to set null_mq_reg.nr_hw_queues to 'nr_online_nodes', when we've potentially done "setup_queues()" with just a single queue. So I'm going to apply this as-is, I'm just wondering why I haven't heard anything from you..

Re: [PATCH] null_blk: mem garbage on NUMA systems during init

2013-12-15 Thread Linus Torvalds
Jens, ping? Patch looks fine, and it looks completely bogus to set null_mq_reg.nr_hw_queues to 'nr_online_nodes', when we've potentially done setup_queues() with just a single queue. So I'm going to apply this as-is, I'm just wondering why I haven't heard anything from you..

Re: [PATCH] null_blk: mem garbage on NUMA systems during init

2013-12-15 Thread Jens Axboe
On Sun, Dec 15 2013, Linus Torvalds wrote: Jens, ping? Patch looks fine, and it looks completely bogus to set null_mq_reg.nr_hw_queues to 'nr_online_nodes', when we've potentially done setup_queues() with just a single queue. So I'm going to apply this as-is, I'm just wondering

[PATCH] null_blk: mem garbage on NUMA systems during init

2013-12-10 Thread Matias Bjorling
For NUMA systems, initializing the blk-mq layer and using per node hctx. We initialize submit queues to 1, while blk-mq nr_hw_queues is initialized to the number of NUMA nodes. This makes the null_init_hctx function overwrite memory outside of what it allocated. In my case it lead to writing

[PATCH] null_blk: mem garbage on NUMA systems during init

2013-12-10 Thread Matias Bjorling
For NUMA systems, initializing the blk-mq layer and using per node hctx. We initialize submit queues to 1, while blk-mq nr_hw_queues is initialized to the number of NUMA nodes. This makes the null_init_hctx function overwrite memory outside of what it allocated. In my case it lead to writing