lkml.org 
[lkml]   [2020]   [Nov]   [30]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: scheduling while atomic in z3fold
On Mon, Nov 30, 2020 at 02:20:14PM +0100, Sebastian Andrzej Siewior wrote:
> On 2020-11-29 12:41:14 [+0100], Mike Galbraith wrote:
> > On Sun, 2020-11-29 at 12:29 +0100, Oleksandr Natalenko wrote:
> > >
> > > Ummm so do compressors explode under non-rt kernel in your tests as
> > > well, or it is just -rt that triggers this?
> >
> > I only tested a non-rt kernel with z3fold, which worked just fine.
>
> I tried this and it did not not explode yet. Mike, can you please
> confirm?
>
> diff --git a/mm/z3fold.c b/mm/z3fold.c
> index 18feaa0bc5377..0bf70f624a4bd 100644
> --- a/mm/z3fold.c
> +++ b/mm/z3fold.c
> @@ -642,14 +642,17 @@ static inline void add_to_unbuddied(struct z3fold_pool *pool,
> {
> if (zhdr->first_chunks == 0 || zhdr->last_chunks == 0 ||
> zhdr->middle_chunks == 0) {
> - struct list_head *unbuddied = get_cpu_ptr(pool->unbuddied);
> -
> + struct list_head *unbuddied;
> int freechunks = num_free_chunks(zhdr);
> +
> + migrate_disable();
> + unbuddied = this_cpu_ptr(pool->unbuddied);
> +
> spin_lock(&pool->lock);
> list_add(&zhdr->buddy, &unbuddied[freechunks]);
> spin_unlock(&pool->lock);
> zhdr->cpu = smp_processor_id();
> - put_cpu_ptr(pool->unbuddied);
> + migrate_enable();
> }
> }
>
> @@ -887,7 +890,8 @@ static inline struct z3fold_header *__z3fold_alloc(struct z3fold_pool *pool,
>
> lookup:
> /* First, try to find an unbuddied z3fold page. */
> - unbuddied = get_cpu_ptr(pool->unbuddied);
> + migrate_disable();
> + unbuddied = this_cpu_ptr(pool->unbuddied);
> for_each_unbuddied_list(i, chunks) {
> struct list_head *l = &unbuddied[i];
>
> @@ -905,7 +909,7 @@ static inline struct z3fold_header *__z3fold_alloc(struct z3fold_pool *pool,
> !z3fold_page_trylock(zhdr)) {
> spin_unlock(&pool->lock);
> zhdr = NULL;
> - put_cpu_ptr(pool->unbuddied);
> + migrate_enable();
> if (can_sleep)
> cond_resched();
> goto lookup;
> @@ -919,7 +923,7 @@ static inline struct z3fold_header *__z3fold_alloc(struct z3fold_pool *pool,
> test_bit(PAGE_CLAIMED, &page->private)) {
> z3fold_page_unlock(zhdr);
> zhdr = NULL;
> - put_cpu_ptr(pool->unbuddied);
> + migrate_enable();
> if (can_sleep)
> cond_resched();
> goto lookup;
> @@ -934,7 +938,7 @@ static inline struct z3fold_header *__z3fold_alloc(struct z3fold_pool *pool,
> kref_get(&zhdr->refcount);
> break;
> }
> - put_cpu_ptr(pool->unbuddied);
> + migrate_enable();
>
> if (!zhdr) {
> int cpu;
> diff --git a/mm/zswap.c b/mm/zswap.c
> index 78a20f7b00f2c..b24f761b9241c 100644
> --- a/mm/zswap.c
> +++ b/mm/zswap.c
> @@ -394,7 +394,9 @@ struct zswap_comp {
> u8 *dstmem;
> };
>
> -static DEFINE_PER_CPU(struct zswap_comp, zswap_comp);
> +static DEFINE_PER_CPU(struct zswap_comp, zswap_comp) = {
> + .lock = INIT_LOCAL_LOCK(lock),

Is it a residue?

> +};
>
> static int zswap_dstmem_prepare(unsigned int cpu)
> {
>
> > -Mike
>
> Sebastian

--
Oleksandr Natalenko (post-factum)

\
 
 \ /
  Last update: 2020-11-30 14:55    [W:0.091 / U:0.288 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site