search for: balloon_defl

...p the "cat file > > /dev/null" process. Total inflation time was 542 seconds. The longest > period that made no net forward progress was 315 seconds (see attached > graph). > Result of "grep balloon /proc/vmstat" after the test: > balloon_inflate 154828377 > balloon_deflate 154828377 > > With patch (kernel 5.6.0-rc4+): > Total inflation duration was 63 seconds. No deflate-queue activity > occurs when pressuring the page-cache. > Result of "grep balloon /proc/vmstat" after the test: > balloon_inflate 12968539 > balloon_deflate 1296853...

...p the "cat file > > /dev/null" process. Total inflation time was 542 seconds. The longest > period that made no net forward progress was 315 seconds (see attached > graph). > Result of "grep balloon /proc/vmstat" after the test: > balloon_inflate 154828377 > balloon_deflate 154828377 > > With patch (kernel 5.6.0-rc4+): > Total inflation duration was 63 seconds. No deflate-queue activity > occurs when pressuring the page-cache. > Result of "grep balloon /proc/vmstat" after the test: > balloon_inflate 12968539 > balloon_deflate 1296853...

.../dev/null" process. Total inflation time was 542 seconds. The longest >> period that made no net forward progress was 315 seconds (see attached >> graph). >> Result of "grep balloon /proc/vmstat" after the test: >> balloon_inflate 154828377 >> balloon_deflate 154828377 >> >> With patch (kernel 5.6.0-rc4+): >> Total inflation duration was 63 seconds. No deflate-queue activity >> occurs when pressuring the page-cache. >> Result of "grep balloon /proc/vmstat" after the test: >> balloon_inflate 1296...

.../dev/null" process. Total inflation time was 542 seconds. The longest >> period that made no net forward progress was 315 seconds (see attached >> graph). >> Result of "grep balloon /proc/vmstat" after the test: >> balloon_inflate 154828377 >> balloon_deflate 154828377 >> >> With patch (kernel 5.6.0-rc4+): >> Total inflation duration was 63 seconds. No deflate-queue activity >> occurs when pressuring the page-cache. >> Result of "grep balloon /proc/vmstat" after the test: >> balloon_inflate 1296...

...;page' while we get around > @@ -75,15 +76,14 @@ struct page *balloon_page_dequeue(struct balloon_dev_info *b_dev_info) > continue; > } > #endif > - spin_lock_irqsave(&b_dev_info->pages_lock, flags); > balloon_page_delete(page); > __count_vm_event(BALLOON_DEFLATE); > - spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); > unlock_page(page); > dequeued_page = true; > break; > } > } > + spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); > > if (!dequeued_page) { > /* I think th...

...;page' while we get around > @@ -75,15 +76,14 @@ struct page *balloon_page_dequeue(struct balloon_dev_info *b_dev_info) > continue; > } > #endif > - spin_lock_irqsave(&b_dev_info->pages_lock, flags); > balloon_page_delete(page); > __count_vm_event(BALLOON_DEFLATE); > - spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); > unlock_page(page); > dequeued_page = true; > break; > } > } > + spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); > > if (!dequeued_page) { > /* I think th...

On 05.02.20 17:34, David Hildenbrand wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be > shrunk when under memory pressure. The

On 05.02.20 17:34, David Hildenbrand wrote: > Commit 71994620bb25 ("virtio_balloon: replace oom notifier with shrinker") > changed the behavior when deflation happens automatically. Instead of > deflating when called by the OOM handler, the shrinker is used. > > However, the balloon is not simply some slab cache that should be > shrunk when under memory pressure. The

...arget until we stop the "cat file > /dev/null" process. Total inflation time was 542 seconds. The longest period that made no net forward progress was 315 seconds (see attached graph). Result of "grep balloon /proc/vmstat" after the test: balloon_inflate 154828377 balloon_deflate 154828377 With patch (kernel 5.6.0-rc4+): Total inflation duration was 63 seconds. No deflate-queue activity occurs when pressuring the page-cache. Result of "grep balloon /proc/vmstat" after the test: balloon_inflate 12968539 balloon_deflate 12968539 Conclusion: This p...

...arget until we stop the "cat file > /dev/null" process. Total inflation time was 542 seconds. The longest period that made no net forward progress was 315 seconds (see attached graph). Result of "grep balloon /proc/vmstat" after the test: balloon_inflate 154828377 balloon_deflate 154828377 With patch (kernel 5.6.0-rc4+): Total inflation duration was 63 seconds. No deflate-queue activity occurs when pressuring the page-cache. Result of "grep balloon /proc/vmstat" after the test: balloon_inflate 12968539 balloon_deflate 12968539 Conclusion: This p...

...> /dev/null" process. Total inflation time was 542 seconds. The longest > > period that made no net forward progress was 315 seconds (see attached > > graph). > > Result of "grep balloon /proc/vmstat" after the test: > > balloon_inflate 154828377 > > balloon_deflate 154828377 > > > > With patch (kernel 5.6.0-rc4+): > > Total inflation duration was 63 seconds. No deflate-queue activity > > occurs when pressuring the page-cache. > > Result of "grep balloon /proc/vmstat" after the test: > > balloon_inflate 1296853...

...if (!trylock_page(page)) > + continue; > + > + if (IS_ENABLED(CONFIG_BALLOON_COMPACTION) && > + PageIsolated(page)) { > + /* raced with isolation */ > + unlock_page(page); > + continue; > + } > + balloon_page_delete(page); > + __count_vm_event(BALLOON_DEFLATE); > + unlock_page(page); > + list_add(&page->lru, pages); I'm not sure whether this list_add must be under the page lock, but enqueue does list_del under page lock, so I think it's a good idea to keep dequeue consistent, operating in the reverse order of enqueue. > +...

...ct page *balloon_page_dequeue(struct balloon_dev_info *b_dev_info) > > > continue; > > > } > > > #endif > > > - spin_lock_irqsave(&b_dev_info->pages_lock, flags); > > > balloon_page_delete(page); > > > __count_vm_event(BALLOON_DEFLATE); > > > - spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); > > > unlock_page(page); > > > dequeued_page = true; > > > break; > > > } > > > } > > > + spin_unlock_irqrestore(&b_dev_info->pages_lo...

...ct page *balloon_page_dequeue(struct balloon_dev_info *b_dev_info) > > > continue; > > > } > > > #endif > > > - spin_lock_irqsave(&b_dev_info->pages_lock, flags); > > > balloon_page_delete(page); > > > __count_vm_event(BALLOON_DEFLATE); > > > - spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); > > > unlock_page(page); > > > dequeued_page = true; > > > break; > > > } > > > } > > > + spin_unlock_irqrestore(&b_dev_info->pages_lo...

During my compaction-related stuff, I encountered a bug with ballooning. With repeated inflating and deflating cycle, guest memory( ie, cat /proc/meminfo | grep MemTotal) is decreased and couldn't be recovered. The reason is balloon_lock doesn't cover release_pages_balloon so struct virtio_balloon fields could be overwritten by race of fill_balloon(e,g, vb->*pfns could be critical).

During my compaction-related stuff, I encountered a bug with ballooning. With repeated inflating and deflating cycle, guest memory( ie, cat /proc/meminfo | grep MemTotal) is decreased and couldn't be recovered. The reason is balloon_lock doesn't cover release_pages_balloon so struct virtio_balloon fields could be overwritten by race of fill_balloon(e,g, vb->*pfns could be critical).

...cat file > > /dev/null" process. Total inflation time was 542 seconds. The longest > period that made no net forward progress was 315 seconds (see attached > graph). > Result of "grep balloon /proc/vmstat" after the test: > balloon_inflate 154828377 > balloon_deflate 154828377 > > With patch (kernel 5.6.0-rc4+): > Total inflation duration was 63 seconds. No deflate-queue activity > occurs when pressuring the page-cache. > Result of "grep balloon /proc/vmstat" after the test: > balloon_inflate 12968539 > balloon_def...

...Total inflation time was 542 seconds. The longest > >> period that made no net forward progress was 315 seconds (see attached > >> graph). > >> Result of "grep balloon /proc/vmstat" after the test: > >> balloon_inflate 154828377 > >> balloon_deflate 154828377 > >> > >> With patch (kernel 5.6.0-rc4+): > >> Total inflation duration was 63 seconds. No deflate-queue activity > >> occurs when pressuring the page-cache. > >> Result of "grep balloon /proc/vmstat" after the test: > &...

...* to be released by the balloon driver. + */ + if (!trylock_page(page)) + continue; + + if (IS_ENABLED(CONFIG_BALLOON_COMPACTION) && + PageIsolated(page)) { + /* raced with isolation */ + unlock_page(page); + continue; + } + balloon_page_delete(page); + __count_vm_event(BALLOON_DEFLATE); + unlock_page(page); + list_add(&page->lru, pages); + ++n_pages; + } + spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); + + return n_pages; +} +EXPORT_SYMBOL_GPL(balloon_page_list_dequeue); + /* * balloon_page_alloc - allocates a new page for insertion into the ballo...

...* to be released by the balloon driver. + */ + if (!trylock_page(page)) + continue; + + if (IS_ENABLED(CONFIG_BALLOON_COMPACTION) && + PageIsolated(page)) { + /* raced with isolation */ + unlock_page(page); + continue; + } + balloon_page_delete(page); + __count_vm_event(BALLOON_DEFLATE); + list_add(&page->lru, pages); + unlock_page(page); + n_pages++; + } + spin_unlock_irqrestore(&b_dev_info->pages_lock, flags); + + return n_pages; +} +EXPORT_SYMBOL_GPL(balloon_page_list_dequeue); + /* * balloon_page_alloc - allocates a new page for insertion into the ballo...