Re: User-space controlled raw ethernet - Is this the way to go?

From: David Schwartz <davids@xxxxxxxxxxxxx>
Date: Mon, 25 Feb 2008 04:24:57 -0800 (PST)

On Feb 25, 2:54 am, Andrew Greensted <ajg...@xxxxxxxxxxxxxx> wrote:

Hi David,

Actually this was my first approach. But I really wasn't that happy with
the overhead of the IP and UDP headers.

Then you will need to provide a driver to figure out what goes where.
Otherwise, there is simply no way for the kernel to arbitrate who gets
the data.

DS
.

Follow-Ups:
- Re: User-space controlled raw ethernet - Is this the way to go?
  - From: phil-news-nospam

References:
- User-space controlled raw ethernet - Is this the way to go?
  - From: Andrew Greensted
- Re: User-space controlled raw ethernet - Is this the way to go?
  - From: David Schwartz
- Re: User-space controlled raw ethernet - Is this the way to go?
  - From: Andrew Greensted

Prev by Date: Re: User-space controlled raw ethernet - Is this the way to go?
Next by Date: Re: Journalling in VM?
Previous by thread: Re: User-space controlled raw ethernet - Is this the way to go?
Next by thread: Re: User-space controlled raw ethernet - Is this the way to go?
Index(es):
- Date
- Thread

Relevant Pages

[benchmark] 1% performance overhead of paravirt_ops on native kernels
... Are there plans to analyze and fix this overhead too, ... I cannot cite a single other kernel feature that has so much ... # Core Netfilter Configuration ... # Device Drivers ...
(Linux-Kernel)
Re: [tip:perfcounters/core] perf_counter: x86: Fix call-chain support to use NMI-safe method
... speed up the kernel entry and exit, the few tens of cycles we ... # Overhead Symbol ... If you sort by a specific user-space symbol, ... The page fault overhead is down the bottom. ...
(Linux-Kernel)
Re: [patch 00/13] Syslets, "Threadlets", generic AIO support, v3
... threadlet that has user space code (and the stack overhead), ... I agree that threadlets are much more flexible - and they might in fact ... processing need to drop out of kernel mode to handle signals. ... I think syslets might also act as a prototyping platform for new system ...
(Linux-Kernel)
Re: lmbench lat_mmap slowdown with CONFIG_PARAVIRT
... promise to have no measurable runtime overhead. ... any real $cache overhead. ... The PV kernel has over 100K larger text size, nearly 40K alone in mm/ and ... User time also ...
(Linux-Kernel)
Re: [benchmark] 1% performance overhead of paravirt_ops on native kernels
... Thanks for taking the time to make these measurements. ... I also saw about 1% boot to boot variation with the non-pvops kernel. ... While I think pvops does add *some* overhead, ...
(Linux-Kernel)