Messages in this thread | | | Date | Thu, 25 Sep 2008 09:59:30 -0500 | From | "Serge E. Hallyn" <> | Subject | Re: [PATCH 3/4] TPM: rcu locking |
| |
Quoting Paul E. McKenney (paulmck@linux.vnet.ibm.com): > On Thu, Sep 25, 2008 at 08:36:45AM -0500, Serge E. Hallyn wrote: > > Quoting Paul E. McKenney (paulmck@linux.vnet.ibm.com): > > > On Tue, Sep 23, 2008 at 05:18:17PM -0300, Rajiv Andrade wrote: > > > > Paul, > > > > > > > > On Tue, 2008-09-23 at 11:19 -0700, Paul E. McKenney wrote: > > > > > > > > > > But here we are deleting from what appears to be some other list. > > > > > And I don't see any insertiong into either list. > > > > > > > > > > What am I missing here? > > > > > > > > > > Thanx, Paul > > > > > > > > Sorry, forgot to change list_add() to list_add_rcu() in the code section > > > > below: > > > > > > > > > > + /* Make chip available */ > > > > > > + spin_lock(&driver_lock); > > > > > > + list_add(&chip->list, &tpm_chip_list); > > > > > > + spin_unlock(&driver_lock); > > > > > > > > I'll resubmit. > > > > > > Cool! > > > > > > So tpm_chip_list and the not-obviously-identical list manipulated > > > in tpm_remove_hardware() really are the same list? > > > > > > Thanx, Paul > > > > Hey Paul, > > > > curious, why do they not look like the same list? > > Because one of them is named &tpm_chip_list, a global variable, and the > other seemed to be returned from a function taking a struct device as an > argument. This is indeed consistent with an element in this list being > hung off of the struct device, so perhaps I was just being insufficiently > persistent in tracking things down. > > Thanx, Paul
Right, tpm_register_hardware both does dev_set_drvdata(dev, chip) to set dev->driver_data = chip, and list_add(chip->list, &tpm_chip_list). The tpm_remove_hardware(), then, gets the device, gets the chip from dev->driver_data, and removes it from the tpm_chip_list.
It does look kosher.
Though once this is all applied to some public git tree, I want to take some time to look at the full result.
-serge
| |