All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Michel Dänzer" <michel@daenzer.net>
To: "Timur Kristóf" <timur.kristof@gmail.com>,
	"Mika Westerberg" <mika.westerberg@linux.intel.com>,
	"Alex Deucher" <alexdeucher@gmail.com>
Cc: "michael.jamet@intel.com" <michael.jamet@intel.com>,
	"dri-devel@lists.freedesktop.org"
	<dri-devel@lists.freedesktop.org>
Subject: Re: Why is Thunderbolt 3 limited to 2.5 GT/s on Linux?
Date: Thu, 4 Jul 2019 10:26:16 +0200	[thread overview]
Message-ID: <eee4eea0-ad27-9d05-192d-c1207da5d0bc@daenzer.net> (raw)
In-Reply-To: <4ee5439cb01e355436d618f5d6f2e109410dc35e.camel@gmail.com>

On 2019-07-03 1:04 p.m., Timur Kristóf wrote:
> 
>>> There may be other factors, yes. I can't offer a good explanation
>>> on
>>> what exactly is happening, but it's pretty clear that amdgpu can't
>>> take
>>> full advantage of the TB3 link, so it seemed like a good idea to
>>> start
>>> investigating this first.
>>
>> Yeah, actually it would be consistent with ~16-32 KB granularity
>> transfers based on your measurements above, which is plausible. So
>> making sure that the driver doesn't artificially limit the PCIe
>> bandwidth might indeed help.
> 
> Can you point me to the place where amdgpu decides the PCIe link speed?
> I'd like to try to tweak it a little bit to see if that helps at all.

I'm not sure offhand, Alex or anyone?


>> OTOH this also indicates a similar potential for improvement by using
>> larger transfers in Mesa and/or the kernel.
> 
> Yes, that sounds like it would be worth looking into.
> 
> Out of curiosity, is there a performace decrease with small transfers
> on a "normal" PCIe port too, or is this specific to TB3?

It's not TB3 specific. With a "normal" 8 GT/s x16 port, I get between
~256 MB/s for 4 KB transfers and ~12 GB/s for 4 MB transfers (even
larger transfers seem slightly slower again). This also looks consistent
with your measurements in that the practical limit seems to be around
75% of the theoretical bandwidth.


-- 
Earthling Michel Dänzer               |              https://www.amd.com
Libre software enthusiast             |             Mesa and X developer
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  reply	other threads:[~2019-07-04  8:26 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-28 10:23 Why is Thunderbolt 3 limited to 2.5 GT/s on Linux? Timur Kristóf
2019-06-28 10:32 ` Mika Westerberg
2019-06-28 11:08   ` Timur Kristóf
2019-06-28 11:34     ` Mika Westerberg
2019-06-28 12:21       ` Timur Kristóf
2019-06-28 12:53         ` Mika Westerberg
2019-06-28 13:33           ` Timur Kristóf
2019-06-28 14:14             ` Mika Westerberg
2019-06-28 14:53               ` Timur Kristóf
2019-07-01 11:44                 ` Mika Westerberg
2019-07-01 14:25                   ` Timur Kristóf
2019-07-01 14:28         ` Alex Deucher
2019-07-01 14:38           ` Timur Kristóf
2019-07-01 14:46             ` Alex Deucher
2019-07-01 15:10               ` Mika Westerberg
2019-07-01 14:54         ` Michel Dänzer
2019-07-01 16:01           ` Timur Kristóf
2019-07-02  8:09             ` Michel Dänzer
2019-07-02  9:49               ` Timur Kristóf
2019-07-03  8:07                 ` Michel Dänzer
2019-07-03 11:04                   ` Timur Kristóf
2019-07-04  8:26                     ` Michel Dänzer [this message]
2019-07-05  9:17                       ` Timur Kristóf
2019-07-05 13:36                       ` Alex Deucher
2019-07-18  9:11                         ` Timur Kristóf
2019-07-18 13:50                           ` Alex Deucher
     [not found]                             ` <172a41d97d383a8989ebd213bb4230a2df4d636d.camel@gmail.com>
2019-07-19 14:29                               ` Alex Deucher
2019-07-03 18:44                   ` Marek Olšák
2019-07-05  9:27                     ` Timur Kristóf
2019-07-05 15:35                       ` Marek Olšák
2019-07-05 16:01                         ` Timur Kristóf
     [not found]                         ` <8f0c2d7780430d40dd1e17a82484d236eae3f981.camel@gmail.com>
2019-07-18 10:29                           ` Michel Dänzer
2019-07-22  9:39                             ` Timur Kristóf
2019-07-23  8:11                               ` Michel Dänzer

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=eee4eea0-ad27-9d05-192d-c1207da5d0bc@daenzer.net \
    --to=michel@daenzer.net \
    --cc=alexdeucher@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=michael.jamet@intel.com \
    --cc=mika.westerberg@linux.intel.com \
    --cc=timur.kristof@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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.