On Mon 13-01-25 18:19:17, Alexei Starovoitov wrote: > From: Alexei Starovoitov <ast@xxxxxxxxxx> > > Tracing BPF programs execute from tracepoints and kprobes where > running context is unknown, but they need to request additional > memory. The prior workarounds were using pre-allocated memory and > BPF specific freelists to satisfy such allocation requests. > Instead, introduce gfpflags_allow_spinning() condition that signals > to the allocator that running context is unknown. > Then rely on percpu free list of pages to allocate a page. > The rmqueue_pcplist() should be able to pop the page from. > If it fails (due to IRQ re-entrancy or list being empty) then > try_alloc_pages() attempts to spin_trylock zone->lock > and refill percpu freelist as normal. > BPF program may execute with IRQs disabled and zone->lock is > sleeping in RT, so trylock is the only option. In theory we can > introduce percpu reentrance counter and increment it every time > spin_lock_irqsave(&zone->lock, flags) is used, but we cannot rely > on it. Even if this cpu is not in page_alloc path the > spin_lock_irqsave() is not safe, since BPF prog might be called > from tracepoint where preemption is disabled. So trylock only. > > Note, free_page and memcg are not taught about gfpflags_allow_spinning() > condition. The support comes in the next patches. > > This is a first step towards supporting BPF requirements in SLUB > and getting rid of bpf_mem_alloc. > That goal was discussed at LSFMM: https://lwn.net/Articles/974138/ > > Signed-off-by: Alexei Starovoitov <ast@xxxxxxxxxx> LGTM, I am not entirely clear on kmsan_alloc_page part though. As long as that part is correct you can add Acked-by: Michal Hocko <mhocko@xxxxxxxx> Other than that try_alloc_pages_noprof begs some user documentation. /** * try_alloc_pages_noprof - opportunistic reentrant allocation from any context * @nid - node to allocate from * @order - allocation order size * * Allocates pages of a given order from the given node. This is safe to * call from any context (from atomic, NMI but also reentrant * allocator -> tracepoint -> try_alloc_pages_noprof). * Allocation is best effort and to be expected to fail easily so nobody should * rely on the succeess. Failures are not reported via warn_alloc(). * * Return: allocated page or NULL on failure. */ > +struct page *try_alloc_pages_noprof(int nid, unsigned int order) > +{ > + /* > + * Do not specify __GFP_DIRECT_RECLAIM, since direct claim is not allowed. > + * Do not specify __GFP_KSWAPD_RECLAIM either, since wake up of kswapd > + * is not safe in arbitrary context. > + * > + * These two are the conditions for gfpflags_allow_spinning() being true. > + * > + * Specify __GFP_NOWARN since failing try_alloc_pages() is not a reason > + * to warn. Also warn would trigger printk() which is unsafe from > + * various contexts. We cannot use printk_deferred_enter() to mitigate, > + * since the running context is unknown. > + * > + * Specify __GFP_ZERO to make sure that call to kmsan_alloc_page() below > + * is safe in any context. Also zeroing the page is mandatory for > + * BPF use cases. > + * > + * Though __GFP_NOMEMALLOC is not checked in the code path below, > + * specify it here to highlight that try_alloc_pages() > + * doesn't want to deplete reserves. > + */ > + gfp_t alloc_gfp = __GFP_NOWARN | __GFP_ZERO | __GFP_NOMEMALLOC; > + unsigned int alloc_flags = ALLOC_TRYLOCK; > + struct alloc_context ac = { }; > + struct page *page; > + > + /* > + * In RT spin_trylock() may call raw_spin_lock() which is unsafe in NMI. > + * If spin_trylock() is called from hard IRQ the current task may be > + * waiting for one rt_spin_lock, but rt_spin_trylock() will mark the > + * task as the owner of another rt_spin_lock which will confuse PI > + * logic, so return immediately if called form hard IRQ or NMI. > + * > + * Note, irqs_disabled() case is ok. This function can be called > + * from raw_spin_lock_irqsave region. > + */ > + if (IS_ENABLED(CONFIG_PREEMPT_RT) && (in_nmi() || in_hardirq())) > + return NULL; > + if (!pcp_allowed_order(order)) > + return NULL; > + > +#ifdef CONFIG_UNACCEPTED_MEMORY > + /* Bailout, since try_to_accept_memory_one() needs to take a lock */ > + if (has_unaccepted_memory()) > + return NULL; > +#endif > + /* Bailout, since _deferred_grow_zone() needs to take a lock */ > + if (deferred_pages_enabled()) > + return NULL; > + > + if (nid == NUMA_NO_NODE) > + nid = numa_node_id(); > + > + prepare_alloc_pages(alloc_gfp, order, nid, NULL, &ac, > + &alloc_gfp, &alloc_flags); > + > + /* > + * Best effort allocation from percpu free list. > + * If it's empty attempt to spin_trylock zone->lock. > + */ > + page = get_page_from_freelist(alloc_gfp, order, alloc_flags, &ac); > + > + /* Unlike regular alloc_pages() there is no __alloc_pages_slowpath(). */ > + > + trace_mm_page_alloc(page, order, alloc_gfp, ac.migratetype); > + kmsan_alloc_page(page, order, alloc_gfp); > + return page; > +} > -- > 2.43.5 -- Michal Hocko SUSE Labs
