Re: [PATCH v6 3/6] Uprobes: Support SDT markers having reference count (semaphore)

From: Oleg Nesterov
Date: Mon Jul 23 2018 - 12:26:36 EST


I have a mixed feeling about this series... I'll try to summarise my thinking
tomorrow, but I do not see any obvious problem so far. Although I have some
concerns about 5/6, I need to re-read it after sleep.


On 07/16, Ravi Bangoria wrote:
>
> +static int delayed_uprobe_install(struct vm_area_struct *vma)
> +{
> + struct list_head *pos, *q;
> + struct delayed_uprobe *du;
> + unsigned long vaddr;
> + int ret = 0, err = 0;
> +
> + mutex_lock(&delayed_uprobe_lock);
> + list_for_each_safe(pos, q, &delayed_uprobe_list) {
> + du = list_entry(pos, struct delayed_uprobe, list);
> +
> + if (!du->uprobe->ref_ctr_offset ||

Is it possible to see ->ref_ctr_offset == 0 in delayed_uprobe_list?

> @@ -1072,7 +1282,13 @@ int uprobe_mmap(struct vm_area_struct *vma)
> struct uprobe *uprobe, *u;
> struct inode *inode;
>
> - if (no_uprobe_events() || !valid_vma(vma, true))
> + if (no_uprobe_events())
> + return 0;
> +
> + if (vma->vm_flags & VM_WRITE)
> + delayed_uprobe_install(vma);

Obviously not nice performance-wise... OK, I do not know if it will actually
hurt in practice and probably we can use the better data structures if necessary.
But can't we check MMF_HAS_UPROBES at least? I mean,

if (vma->vm_flags & VM_WRITE && test_bit(MMF_HAS_UPROBES, &vma->vm_mm->flags))
delayed_uprobe_install(vma);

?


Perhaps we can even add another flag later, MMF_HAS_DELAYED_UPROBES set by
delayed_uprobe_add().

Oleg.