RE: [PATCH v2 3/5] pstore: support multiple pmsg instances

From: åæäæ / IWAMATSUïNOBUHIRO
Date: Wed Oct 05 2016 - 00:37:23 EST


Hi,

> -----Original Message-----
> From: keescook@xxxxxxxxxx [mailto:keescook@xxxxxxxxxx] On Behalf Of Kees
> Cook
> Sent: Friday, September 09, 2016 6:34 AM
> To: åæäæ / IWAMATSUïNOBUHIRO
> Cc: Anton Vorontsov; Colin Cross; Tony Luck; LKML; Hiraku Toyooka; Mark
> Salyzyn; éåèå / AGUCHIïSEIJI
> Subject: Re: [PATCH v2 3/5] pstore: support multiple pmsg instances
>
> On Sun, Jul 24, 2016 at 8:56 PM, Nobuhiro Iwamatsu
> <nobuhiro.iwamatsu.kw@xxxxxxxxxxx> wrote:
> > From: Hiraku Toyooka <hiraku.toyooka.gu@xxxxxxxxxxx>
> >
> > This enables pmsg to deal with multiple instances. One possible use is
> > content priority control on limited persistent store space. By using
> > different buffers, we can prevent important messages from being
> > overwritten by less important messages.
> >
> > When pstore backend module specifies the number of instances
> > (num_pmsg) in pstore_info, multiple /dev/pmsg[ID] appear. (ID is an
> > integer starting at 0. It corresponds to minor number of the each char
> > device.) Writes to each /dev/pmsg[ID] are isolated each other. After
> > reboot, the contents are available in
> /sys/fs/pstore/pmsg-[backend]-[ID].
> >
> > Signed-off-by: Hiraku Toyooka <hiraku.toyooka.gu@xxxxxxxxxxx>
> > Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.kw@xxxxxxxxxxx>
> > Cc: Mark Salyzyn <salyzyn@xxxxxxxxxxx>
> > Cc: Seiji Aguchi <seiji.aguchi.tr@xxxxxxxxxxx>
> > ---
> > fs/pstore/pmsg.c | 20 ++++++++++++++++++--
> > include/linux/pstore.h | 1 +
> > 2 files changed, 19 insertions(+), 2 deletions(-)
> >
> > diff --git a/fs/pstore/pmsg.c b/fs/pstore/pmsg.c index
> > 7de20cd..2e281ed 100644
> > --- a/fs/pstore/pmsg.c
> > +++ b/fs/pstore/pmsg.c
> > @@ -52,8 +52,8 @@ static ssize_t write_pmsg(struct file *file, const char
> __user *buf,
> > vfree(buffer);
> > return -EFAULT;
> > }
> > - psinfo->write_buf(PSTORE_TYPE_PMSG, 0, &id, 0, buffer,
> 0, c,
> > - psinfo);
> > + psinfo->write_buf(PSTORE_TYPE_PMSG, 0, &id,
> > + iminor(file->f_inode), buffer, 0, c,
> > + psinfo);
> >
> > i += c;
> > }
> > @@ -85,6 +85,7 @@ static char *pmsg_devnode(struct device *dev,
> > umode_t *mode) void pstore_register_pmsg(void) {
> > struct device *pmsg_device;
> > + int i = 0;
> >
> > pmsg_major = register_chrdev(0, PMSG_NAME, &pmsg_fops);
> > if (pmsg_major < 0) {
> > @@ -105,9 +106,24 @@ void pstore_register_pmsg(void)
> > pr_err("failed to create device\n");
> > goto err_device;
> > }
> > +
> > + /* allocate additional /dev/pmsg[ID] */
> > + for (i = 1; i < psinfo->num_pmsg; i++) {
> > + struct device *pmsg_device;
> > +
> > + pmsg_device = device_create(pmsg_class, NULL,
> > + MKDEV(pmsg_major, i), NULL,
> "%s%d",
> > + PMSG_NAME, i);
> > + if (IS_ERR(pmsg_device)) {
> > + pr_err("failed to create device\n");
> > + goto err_device;
> > + }
> > + }
>
> This should just be merged into the first device_create call (i.e.
> drop the first one, keep your loop, but have "i" start at 0). There's no
> reason that I can see to do it separately.

Indeed, I will fix.
>
> > return;
> >
> > err_device:
> > + for (i--; i >= 0; i--)
> > + device_destroy(pmsg_class, MKDEV(pmsg_major, i));
>
> Was the device_destroy() for /dev/pmsg0 missing before? (i.e. does
> class_destroy() or unregister_chrdev() already clean up the devices?)

Your are right.
I will fix as /dev/pmsg0 also treated with device_destroy().

>
> > class_destroy(pmsg_class);
> > err_class:
> > unregister_chrdev(pmsg_major, PMSG_NAME); diff --git
> > a/include/linux/pstore.h b/include/linux/pstore.h index
> > 831479f..b0c24cc 100644
> > --- a/include/linux/pstore.h
> > +++ b/include/linux/pstore.h
> > @@ -54,6 +54,7 @@ struct pstore_info {
> > size_t bufsize;
> > struct mutex read_mutex; /* serialize open/read/close
> */
> > int flags;
> > + unsigned int num_pmsg;
> > int (*open)(struct pstore_info *psi);
> > int (*close)(struct pstore_info *psi);
> > ssize_t (*read)(u64 *id, enum pstore_type_id *type,
> > --
> > 2.8.1
> >
> >
>
> -Kees
>
>
> --
> Kees Cook
> Nexus Security

Best regards,
Nobuhiro