All of lore.kernel.org
 help / color / mirror / Atom feed
* [bug report] dm verity: add support for forward error correction
@ 2018-09-25 12:07 Dan Carpenter
  2018-09-25 16:03 ` Sami Tolvanen
  0 siblings, 1 reply; 5+ messages in thread
From: Dan Carpenter @ 2018-09-25 12:07 UTC (permalink / raw)
  To: dm-devel; +Cc: samitolvanen

Hi DM devs,

The patch a739ff3f543a: "dm verity: add support for forward error
correction" from Dec 3, 2015, leads to the following static checker
warning:

	drivers/md/dm-verity-fec.c:147 fec_decode_bufs()
	warn: 'par' can also be NULL

drivers/md/dm-verity-fec.c
   124  /*
   125   * Decode all RS blocks from buffers and copy corrected bytes into fio->output
   126   * starting from block_offset.
   127   */
   128  static int fec_decode_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio,
   129                             u64 rsb, int byte_index, unsigned block_offset,
   130                             int neras)
   131  {
   132          int r, corrected = 0, res;
   133          struct dm_buffer *buf;
   134          unsigned n, i, offset;
   135          u8 *par, *block;
   136  
   137          par = fec_read_parity(v, rsb, block_offset, &offset, &buf);
   138          if (IS_ERR(par))
                           ^^^

The problem is that fec_read_parity() returns NULL from new_read().
It's possible that this is intentional...  There isn't any documentation
but in my experience there never is so that doesn't necessarily mean
anything...

   139                  return PTR_ERR(par);
   140  
   141          /*
   142           * Decode the RS blocks we have in bufs. Each RS block results in
   143           * one corrected target byte and consumes fec->roots parity bytes.
   144           */
   145          fec_for_each_buffer_rs_block(fio, n, i) {
   146                  block = fec_buffer_rs_block(v, fio, n, i);
   147                  res = fec_decode_rs8(v, fio, block, &par[offset], neras);
   148                  if (res < 0) {
   149                          r = res;
   150                          goto error;
   151                  }
   152  
   153                  corrected += res;
   154                  fio->output[block_offset] = block[byte_index];
   155  
   156                  block_offset++;
   157                  if (block_offset >= 1 << v->data_dev_block_bits)
   158                          goto done;
   159  
   160                  /* read the next block when we run out of parity bytes */
   161                  offset += v->fec->roots;
   162                  if (offset >= 1 << v->data_dev_block_bits) {
   163                          dm_bufio_release(buf);
   164  

regards,
dan carpenter

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [bug report] dm verity: add support for forward error correction
  2018-09-25 12:07 [bug report] dm verity: add support for forward error correction Dan Carpenter
@ 2018-09-25 16:03 ` Sami Tolvanen
  2018-09-26 10:14   ` Dan Carpenter
  0 siblings, 1 reply; 5+ messages in thread
From: Sami Tolvanen @ 2018-09-25 16:03 UTC (permalink / raw)
  To: dan.carpenter; +Cc: device-mapper development

On Tue, Sep 25, 2018 at 5:07 AM Dan Carpenter <dan.carpenter@oracle.com> wrote:
> The patch a739ff3f543a: "dm verity: add support for forward error
> correction" from Dec 3, 2015, leads to the following static checker
> warning:

Has something changed since the last time you reported this warning?

https://www.redhat.com/archives/dm-devel/2018-June/msg00164.html

Sami

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [bug report] dm verity: add support for forward error correction
  2018-09-25 16:03 ` Sami Tolvanen
@ 2018-09-26 10:14   ` Dan Carpenter
  0 siblings, 0 replies; 5+ messages in thread
From: Dan Carpenter @ 2018-09-26 10:14 UTC (permalink / raw)
  To: Sami Tolvanen; +Cc: device-mapper development

On Tue, Sep 25, 2018 at 09:03:40AM -0700, Sami Tolvanen wrote:
> On Tue, Sep 25, 2018 at 5:07 AM Dan Carpenter <dan.carpenter@oracle.com> wrote:
> > The patch a739ff3f543a: "dm verity: add support for forward error
> > correction" from Dec 3, 2015, leads to the following static checker
> > warning:
> 
> Has something changed since the last time you reported this warning?
> 
> https://www.redhat.com/archives/dm-devel/2018-June/msg00164.html
> 

Gar...  Sorry.  The only thing that changed is my computer.  I try to
keep a record of old warnings I've sent but it must be on my other
system.

regards,
dan carpenter

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [bug report] dm verity: add support for forward error correction
  2018-06-14 14:33 Dan Carpenter
@ 2018-06-14 15:54 ` Sami Tolvanen
  0 siblings, 0 replies; 5+ messages in thread
From: Sami Tolvanen @ 2018-06-14 15:54 UTC (permalink / raw)
  To: dan.carpenter; +Cc: device-mapper development

On Thu, Jun 14, 2018 at 7:33 AM Dan Carpenter <dan.carpenter@oracle.com> wrote:
> Apparently, if the read is a blocking read then new_read() returns NULL.

It looks like new_read can return NULL only if it's called with NF_GET
or NF_PREFETCH. dm_bufio_read calls it with NF_READ, which means it
shouldn't return NULL. Someone more familiar with dm-bufio, please
correct me if I'm wrong.

Sami

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [bug report] dm verity: add support for forward error correction
@ 2018-06-14 14:33 Dan Carpenter
  2018-06-14 15:54 ` Sami Tolvanen
  0 siblings, 1 reply; 5+ messages in thread
From: Dan Carpenter @ 2018-06-14 14:33 UTC (permalink / raw)
  To: samitolvanen; +Cc: dm-devel

Hello Sami Tolvanen,

The patch a739ff3f543a: "dm verity: add support for forward error
correction" from Dec 3, 2015, leads to the following static checker
warning:

	drivers/md/dm-verity-fec.c:147 fec_decode_bufs()
	warn: 'par' can also be NULL

drivers/md/dm-verity-fec.c
   128  static int fec_decode_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio,
   129                             u64 rsb, int byte_index, unsigned block_offset,
   130                             int neras)
   131  {
   132          int r, corrected = 0, res;
   133          struct dm_buffer *buf;
   134          unsigned n, i, offset;
   135          u8 *par, *block;
   136  
   137          par = fec_read_parity(v, rsb, block_offset, &offset, &buf);
   138          if (IS_ERR(par))
   139                  return PTR_ERR(par);

Apparently, if the read is a blocking read then new_read() returns NULL.
The documentation is unclear (just joking, there is no documentation).

   140  
   141          /*
   142           * Decode the RS blocks we have in bufs. Each RS block results in
   143           * one corrected target byte and consumes fec->roots parity bytes.
   144           */
   145          fec_for_each_buffer_rs_block(fio, n, i) {
   146                  block = fec_buffer_rs_block(v, fio, n, i);
   147                  res = fec_decode_rs8(v, fio, block, &par[offset], neras);
   148                  if (res < 0) {
   149                          r = res;
   150                          goto error;
   151                  }
   152  
   153                  corrected += res;
   154                  fio->output[block_offset] = block[byte_index];
   155  
   156                  block_offset++;
   157                  if (block_offset >= 1 << v->data_dev_block_bits)
   158                          goto done;
   159  
   160                  /* read the next block when we run out of parity bytes */
   161                  offset += v->fec->roots;
   162                  if (offset >= 1 << v->data_dev_block_bits) {
   163                          dm_bufio_release(buf);
   164  
   165                          par = fec_read_parity(v, rsb, block_offset, &offset, &buf);
   166                          if (unlikely(IS_ERR(par)))
   167                                  return PTR_ERR(par);
   168                  }
   169          }
   170  done:
   171          r = corrected;

regards,
dan carpenter

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2018-09-26 10:14 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-09-25 12:07 [bug report] dm verity: add support for forward error correction Dan Carpenter
2018-09-25 16:03 ` Sami Tolvanen
2018-09-26 10:14   ` Dan Carpenter
  -- strict thread matches above, loose matches on Subject: below --
2018-06-14 14:33 Dan Carpenter
2018-06-14 15:54 ` Sami Tolvanen

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.