[HN Gopher] Linux's Slab Allocator Is Officially Deprecated
       ___________________________________________________________________
        
       Linux's Slab Allocator Is Officially Deprecated
        
       Author : mfiguiere
       Score  : 82 points
       Date   : 2023-06-30 17:24 UTC (5 hours ago)
        
 (HTM) web link (www.phoronix.com)
 (TXT) w3m dump (www.phoronix.com)
        
       | dathinab wrote:
       | How does something ends up on the front page which is so
       | uninteresting(1) that both on the site it comes from and here,
       | there isn't a single comment.
       | 
       | (1): Uninteresting but not irrelevant. Like a step in a stair,
       | each step is uninteresting but relevant as missing a step is a
       | huge issue.
        
         | ape4 wrote:
         | How an OS allocates memory is pretty important
        
           | astrange wrote:
           | This detail of an in kernel allocator isn't very important or
           | novel, but Linux changes are easy to report on since they're
           | done in public, so they get articles.
        
           | dathinab wrote:
           | Like I sayed important but uninteresting.
           | 
           | Because this is not about changing how Linux allocates
           | memory.
           | 
           | It's about deprecating a method Linux used in the past but
           | has by default switched away from already in the past.
        
             | arp242 wrote:
             | Clearly people disagree that it's uninteresting.
             | 
             | You don't need to personally like every story posted on HN.
             | It's fine to skip the some of them. You can even hide them.
        
               | dathinab wrote:
               | it's not weather its interesting for me or you but
               | weather its interesting enough for HN readers to interact
               | with it in the comment section ...
               | 
               | and I was wondering how something which seemed to be not
               | leading to much engagement at all ended up on the front
               | page, I wasn't complaining or saying it's bad, just
               | wondering
        
             | JohnFen wrote:
             | > Like I sayed important but uninteresting
             | 
             | Uninteresting to you. Interesting to me.
        
         | mat_epice wrote:
         | Most links here have dozens of points, this (and a couple of
         | others) only has seven. Perhaps a "hot" metric saw a few votes
         | come in quickly and overestimated its general interest. It
         | should fall off the front page very quickly.
        
       | mmastrac wrote:
       | SLAB vs SLOB vs SLUB:
       | https://events.static.linuxfound.org/sites/events/files/slid...
        
         | sedatk wrote:
         | Huh, thanks to the article, I learned that SLOB's maintainer
         | was Mercurial's creator, Olivia Mackall. I'm in awe of these
         | people who are involved in more than one gigantic project in
         | their lifetime.
        
         | gumby wrote:
         | Fortunately SLIB and SLEB are still in our future. I hope they
         | don't need to resort to SLYB, much less SLOEB and the like.
        
           | codetrotter wrote:
           | > SLOEB
           | 
           | Excuse me, I think you mean "SLOB"!
        
             | Iwan-Zotow wrote:
             | OSLOEB
        
             | gumby wrote:
             | Isn't the kernel USASCII-only (POSIX charset)?
        
       | corbet wrote:
       | LWN coverage of the LSFMM session that led to this decision:
       | https://lwn.net/Articles/932201/
        
       | trufas wrote:
       | SLUB is a nice story of a simpler implementation winning out,
       | even when SLAB may have been more technically sophisticated.
        
       ___________________________________________________________________
       (page generated 2023-06-30 23:02 UTC)