This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [PATCH v2 2.6.38-rc8-tip 11/20] 11: uprobes: slot allocation for uprobes
- From: Srikar Dronamraju <srikar at linux dot vnet dot ibm dot com>
- To: Jonathan Corbet <corbet at lwn dot net>
- Cc: Peter Zijlstra <peterz at infradead dot org>, Ingo Molnar <mingo at elte dot hu>, Steven Rostedt <rostedt at goodmis dot org>, Linux-mm <linux-mm at kvack dot org>, Arnaldo Carvalho de Melo <acme at infradead dot org>, Linus Torvalds <torvalds at linux-foundation dot org>, Ananth N Mavinakayanahalli <ananth at in dot ibm dot com>, Christoph Hellwig <hch at infradead dot org>, Andi Kleen <andi at firstfloor dot org>, Masami Hiramatsu <masami dot hiramatsu dot pt at hitachi dot com>, Oleg Nesterov <oleg at redhat dot com>, LKML <linux-kernel at vger dot kernel dot org>, Jim Keniston <jkenisto at linux dot vnet dot ibm dot com>, Roland McGrath <roland at hack dot frob dot com>, SystemTap <systemtap at sources dot redhat dot com>, Andrew Morton <akpm at linux-foundation dot org>, "Paul E. McKenney" <paulmck at linux dot vnet dot ibm dot com>
- Date: Wed, 16 Mar 2011 10:28:20 +0530
- Subject: Re: [PATCH v2 2.6.38-rc8-tip 11/20] 11: uprobes: slot allocation for uprobes
- References: <20110314133403.27435.7901.sendpatchset@localhost6.localdomain6> <20110314133610.27435.93666.sendpatchset@localhost6.localdomain6> <20110315131020.36477a1c@bike.lwn.net>
- Reply-to: Srikar Dronamraju <srikar at linux dot vnet dot ibm dot com>
* Jonathan Corbet <corbet@lwn.net> [2011-03-15 13:10:20]:
> Just a couple of minor notes while I was looking at this code...
>
> > +static struct uprobes_xol_area *xol_alloc_area(void)
> > +{
> > + struct uprobes_xol_area *area = NULL;
> > +
> > + area = kzalloc(sizeof(*area), GFP_USER);
> > + if (unlikely(!area))
> > + return NULL;
> > +
> > + area->bitmap = kzalloc(BITS_TO_LONGS(UINSNS_PER_PAGE) * sizeof(long),
> > + GFP_USER);
>
> Why GFP_USER? That causes extra allocation limits to be enforced. Given
> that in part 14 you have:
Okay, Will use GFP_KERNEL.
We used GFP_USER because we thought its going to represent part of
process address space;
>
> +/* Prepare to single-step probed instruction out of line. */
> +static int pre_ssout(struct uprobe *uprobe, struct pt_regs *regs,
> + unsigned long vaddr)
> +{
> + xol_get_insn_slot(uprobe, vaddr);
> + BUG_ON(!current->utask->xol_vaddr);
>
> It seems to me that you really don't want those allocations to fail.
>
> back to xol_alloc_area():
>
> > + if (!area->bitmap)
> > + goto fail;
> > +
> > + spin_lock_init(&area->slot_lock);
> > + if (!xol_add_vma(area) && !current->mm->uprobes_xol_area) {
> > + task_lock(current);
> > + if (!current->mm->uprobes_xol_area) {
> > + current->mm->uprobes_xol_area = area;
> > + task_unlock(current);
> > + return area;
> > + }
> > + task_unlock(current);
> > + }
> > +
> > +fail:
> > + if (area) {
> > + if (area->bitmap)
> > + kfree(area->bitmap);
> > + kfree(area);
> > + }
>
> You've already checked area against NULL, and kfree() can handle null
> pointers, so both of those tests are unneeded.
Okay,
>
> > + return current->mm->uprobes_xol_area;
> > +}
>
> jon
--
Thanks and Regards
Srikar