| 288,7 → 288,7 |
|---|
| <section id="frame_allocator"> |
| <title>Frame allocator</title> |
| |
| <para><mediaobject id="frame_alloc"> |
| <figure><mediaobject id="frame_alloc"> |
| <imageobject role="html"> |
| <imagedata fileref="images/frame_alloc.png" format="PNG" /> |
| </imageobject> |
| 296,7 → 296,9 |
|---|
| <imageobject role="fop"> |
| <imagedata fileref="images.vector/frame_alloc.svg" format="SVG" /> |
| </imageobject> |
| </mediaobject></para> |
| </mediaobject> |
| <title>Frame allocator scheme.</title> |
| </figure> |
| |
| <formalpara> |
| <title>Overview</title> |
| 339,7 → 341,7 |
|---|
| <section> |
| <title>Overview</title> |
| |
| <para><mediaobject id="buddy_alloc"> |
| <figure><mediaobject id="buddy_alloc"> |
| <imageobject role="html"> |
| <imagedata fileref="images/buddy_alloc.png" format="PNG" /> |
| </imageobject> |
| 347,7 → 349,9 |
|---|
| <imageobject role="fop"> |
| <imagedata fileref="images.vector/buddy_alloc.svg" format="SVG" /> |
| </imageobject> |
| </mediaobject></para> |
| </mediaobject> |
| <title>Buddy system scheme.</title> |
| </figure> |
| |
| <para>In the buddy allocator, the memory is broken down into |
| power-of-two sized naturally aligned blocks. These blocks are |
| 483,7 → 487,7 |
|---|
| <section> |
| <title>Implementation</title> |
| |
| <para><mediaobject id="slab_alloc"> |
| <figure><mediaobject id="slab_alloc"> |
| <imageobject role="html"> |
| <imagedata fileref="images/slab_alloc.png" format="PNG" /> |
| </imageobject> |
| 491,14 → 495,16 |
|---|
| <imageobject role="fop"> |
| <imagedata fileref="images.vector/slab_alloc.svg" format="SVG" /> |
| </imageobject> |
| </mediaobject></para> |
| </mediaobject> |
| <title>Slab allocator scheme.</title> |
| </figure> |
| |
| <para>The SLAB allocator is closely modelled after <ulink |
| <para>The slab allocator is closely modelled after <ulink |
| url="http://www.usenix.org/events/usenix01/full_papers/bonwick/bonwick_html/"> |
| OpenSolaris SLAB allocator by Jeff Bonwick and Jonathan Adams </ulink> |
| OpenSolaris slab allocator by Jeff Bonwick and Jonathan Adams </ulink> |
| with the following exceptions: <itemizedlist> |
| <listitem> |
| empty SLABS are deallocated immediately (in Linux they are kept in linked list, in Solaris ???) |
| empty slabs are deallocated immediately (in Linux they are kept in linked list, in Solaris ???) |
| </listitem> |
| |
| <listitem> |
| 519,7 → 525,7 |
|---|
| <title>Magazine layer</title> |
| |
| <para>Due to the extensive bottleneck on SMP architures, caused by |
| global SLAB locking mechanism, making processing of all slab |
| global slab locking mechanism, making processing of all slab |
| allocation requests serialized, a new layer was introduced to the |
| classic slab allocator design. Slab allocator was extended to |
| support per-CPU caches 'magazines' to achieve good SMP scaling. |