Messages in this thread | | | Date | Thu, 6 Oct 2011 12:23:26 +0530 | From | Srikar Dronamraju <> | Subject | Re: [PATCH v5 3.1.0-rc4-tip 9/26] Uprobes: Background page replacement. |
| |
* Oleg Nesterov <oleg@redhat.com> [2011-10-05 18:19:14]:
> On 09/20, Srikar Dronamraju wrote: > > > > +int __weak read_opcode(struct task_struct *tsk, unsigned long vaddr, > > + uprobe_opcode_t *opcode) > > +{ > > + struct vm_area_struct *vma; > > + struct page *page; > > + void *vaddr_new; > > + int ret; > > + > > + ret = get_user_pages(tsk, tsk->mm, vaddr, 1, 0, 0, &page, &vma); > > + if (ret <= 0) > > + return ret; > > + ret = -EINVAL; > > + > > + /* > > + * We are interested in text pages only. Our pages of interest > > + * should be mapped for read and execute only. We desist from > > + * adding probes in write mapped pages since the breakpoints > > + * might end up in the file copy. > > + */ > > + if (!valid_vma(vma)) > > + goto put_out; > > Another case when valid_vma() looks suspicious. We are going to restore > the original instruction. We shouldn't fail (at least we shouldn't "leak" > ->mm_uprobes_count) if ->vm_flags was changed between register_uprobe() > and unregister_uprobe(). >
Agree.
-- Thanks and Regards Srikar
| |