Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>The two approaches are not quite the same thing, depending on the underlying data types. The standard (C11, 6.5.7 Bitwise shift operators) states that:</p> <blockquote> <p>If the value of the right operand is negative or is greater than or equal to the width of the promoted left operand, the behavior is undefined.</p> </blockquote> <p>Hence, if you were to shift a 32-bit wide integer type 32 bits to the right, it would result in undefined behaviour. However, shifting it 31 bits and then another bit <em>is</em> well defined.</p> <p>That particular point, while a reason why someone may do this, is probably moot in this particular case. Given your type is <code>unsigned long long</code> (guaranteed 64 bits width), there should be no difference between <code>(x &gt;&gt; 31) &gt;&gt; 1</code> and <code>x &gt;&gt; 32</code>.</p> <p>However, if some platform actually defines <code>ext4_fsblk_t</code> as a 32-bit type (or if it's carried forward from an earlier implementation which allowed a 32-bit type <sup>(a)</sup>), you would find yourself having to resort to the two-stage shift to guarantee defined behaviour.</p> <hr> <p><sup>(a)</sup>: There was an interim stage between <code>ext3</code> and <code>ext4</code> which allowed <code>ext3</code> to move to 48 bits with its extents, a stepping stone on the way to <code>ext4</code> with its 64-bit data types. Prior to that, 8TB was the practical limit to file systems.</p> <p>It's <em>possible,</em> though I haven't confirmed, that this code snippet is a hangover from that transition, before the underlying data types went up to 64 bits wide.</p> <p>The type that came to be used, <code>sector_t</code>, was conditionally defined as either <code>u64</code> or <code>unsigned long</code>, which may explain why the code is there in the first place. In situations where it was defined as the latter, a two-stage shift may have been needed.</p> <p>However, given that <code>ext4</code> has left this interim stage behind, I'm not sure the two-stage shift is needed any more. Best bet is to raise a change request and see if it gets shot down by the developers :-)</p>
    singulars
    1. This table or related slice is empty.
    1. POC, bit shift . 2 consecutive shifts
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USpaxdiablo
    1. USpaxdiablo
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. POC, bit shift . 2 consecutive shifts
      singulars
      1. PTQuestion
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COWell, shifting by 31 bits (for a signed 32-bit integer) already has the effect of copying the sign bit to all the bits. So I fail to see the purpose of shifting by 32.
      singulars
      1. PO
      1. USMysticial
    2. CO@Mystical: It's unsigned so should be a logical shift. I guess long long may be 32 bits though so this is to prevent undefined behavior in that case.
      singulars
      1. PO
      1. USGuy Sirton
    3. CO@Mystical, yes, that may apply for signed integers, but that's not necessarily the case here.
      singulars
      1. PO
      1. USpaxdiablo
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload