From: Mika Westerberg <mika.westerberg@linux.intel.com>
To: Lukas Wunner <lukas@wunner.de>
Cc: linux-kernel@vger.kernel.org,
Michael Jamet <michael.jamet@intel.com>,
Yehezkel Bernat <YehezkelShB@gmail.com>,
Andreas Noever <andreas.noever@gmail.com>,
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Subject: Re: [PATCH v2 16/28] thunderbolt: Discover preboot PCIe paths the boot firmware established
Date: Tue, 12 Feb 2019 20:34:15 +0200 [thread overview]
Message-ID: <20190212183415.GI7875@lahna.fi.intel.com> (raw)
In-Reply-To: <20190212174942.if2ene44f5m5mern@wunner.de>
On Tue, Feb 12, 2019 at 06:49:42PM +0100, Lukas Wunner wrote:
> On Wed, Feb 06, 2019 at 04:17:26PM +0300, Mika Westerberg wrote:
> > /* dword 0 */
> > hop.next_hop = path->hops[i].next_hop_index;
> > hop.out_port = path->hops[i].out_port->port;
> > - /* TODO: figure out why these are good values */
> > - hop.initial_credits = (i == path->path_length - 1) ? 16 : 7;
> > + hop.initial_credits = path->hops[i].initial_credits;
> > hop.unknown1 = 0;
> > hop.enable = 1;
> [...]
> > @@ -78,6 +78,74 @@ static void tb_pci_init_path(struct tb_path *path)
> > path->weight = 1;
> > path->drop_packages = 0;
> > path->nfc_credits = 0;
> > + path->hops[0].initial_credits = 7;
> > + path->hops[1].initial_credits = 16;
>
> I guess Andreas' algorithm (the last hop in the path is assigned
> 16 and all the ones before are assigned 7) was reverse-engineered
> from Apple's driver. The fact that this algorithm works for paths
> of arbitrary length could indicate that Apple indeed does establish
> tunnels between non-adjacent switches.
We do it as well for DP and DMA paths in subsequent patches. For those
there are NULL ports in the middle which get assigned different amount
of credits. PCIe paths on the other hand only need two hops when we
daisy-chain them in this patch series.
> Also, why are these good values? (You've deleted the comment.)
To be honest, I don't know all the details. Credits are used for flow
control to make sure receiving port always has enough buffers before
sending port can send more packets (assuming the path is using flow
control). I don't know where 7 and 16 came but they seem to work pretty
well for PCIe path so I kept using them.
next prev parent reply other threads:[~2019-02-12 18:34 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-06 13:17 [PATCH v2 00/28] thunderbolt: Software connection manager improvements Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 01/28] net: thunderbolt: Unregister ThunderboltIP protocol handler when suspending Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 02/28] thunderbolt: Do not allocate switch if depth is greater than 6 Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 03/28] thunderbolt: Enable TMU access when accessing port space on legacy devices Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 04/28] thunderbolt: Add dummy read after port capability list walk on Light Ridge Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 05/28] thunderbolt: Move LC specific functionality into a separate file Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 06/28] thunderbolt: Configure lanes when switch is initialized Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 07/28] thunderbolt: Set sleep bit when suspending switch Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 08/28] thunderbolt: Properly disable path Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 09/28] thunderbolt: Cache adapter specific capability offset into struct port Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 10/28] thunderbolt: Rename tunnel_pci to tunnel Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 11/28] thunderbolt: Generalize tunnel creation functionality Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 12/28] thunderbolt: Add functions for allocating and releasing hop IDs Mika Westerberg
2019-02-10 12:13 ` Lukas Wunner
2019-02-11 8:30 ` Mika Westerberg
2019-02-12 12:43 ` Lukas Wunner
2019-02-12 12:51 ` Mika Westerberg
2019-02-12 12:59 ` Lukas Wunner
2019-02-12 13:02 ` Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 13/28] thunderbolt: Add helper function to iterate from one port to another Mika Westerberg
2019-02-07 14:33 ` Andy Shevchenko
2019-02-11 6:16 ` Lukas Wunner
2019-02-11 9:54 ` Mika Westerberg
2019-02-12 13:55 ` Lukas Wunner
2019-02-12 19:03 ` Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 14/28] thunderbolt: Extend tunnel creation to more than 2 adjacent switches Mika Westerberg
2019-02-10 15:33 ` Lukas Wunner
2019-02-11 8:45 ` Mika Westerberg
2019-02-12 12:54 ` Lukas Wunner
2019-02-06 13:17 ` [PATCH v2 15/28] thunderbolt: Deactivate all paths before restarting them Mika Westerberg
2019-02-11 6:35 ` Lukas Wunner
2019-02-11 8:52 ` Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 16/28] thunderbolt: Discover preboot PCIe paths the boot firmware established Mika Westerberg
2019-02-12 17:49 ` Lukas Wunner
2019-02-12 18:34 ` Mika Westerberg [this message]
2019-02-12 19:42 ` Lukas Wunner
2019-02-13 8:30 ` Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 17/28] thunderbolt: Add support for full PCIe daisy chains Mika Westerberg
2019-03-24 11:31 ` Lukas Wunner
2019-03-25 9:57 ` Mika Westerberg
2019-03-25 11:16 ` Lukas Wunner
2019-03-25 11:22 ` Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 18/28] thunderbolt: Scan only valid NULL adapter ports in hotplug Mika Westerberg
2019-02-12 14:04 ` Lukas Wunner
2019-02-12 18:46 ` Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 19/28] thunderbolt: Generalize port finding routines to support all port types Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 20/28] thunderbolt: Rework NFC credits handling Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 21/28] thunderbolt: Add support for Display Port tunnels Mika Westerberg
2019-02-07 14:28 ` Andy Shevchenko
2019-02-06 13:17 ` [PATCH v2 22/28] thunderbolt: Run tb_xdp_handle_request() in system workqueue Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 23/28] thunderbolt: Add XDomain UUID exchange support Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 24/28] thunderbolt: Add support for DMA tunnels Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 25/28] thunderbolt: Make tb_switch_alloc() return ERR_PTR() Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 26/28] thunderbolt: Add support for XDomain connections Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 27/28] thunderbolt: Make rest of the logging to happen at debug level Mika Westerberg
2019-02-06 13:17 ` [PATCH v2 28/28] thunderbolt: Start firmware on Titan Ridge Apple systems Mika Westerberg
2019-02-07 14:22 ` [PATCH v2 00/28] thunderbolt: Software connection manager improvements Andy Shevchenko
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=20190212183415.GI7875@lahna.fi.intel.com \
--to=mika.westerberg@linux.intel.com \
--cc=YehezkelShB@gmail.com \
--cc=andreas.noever@gmail.com \
--cc=andriy.shevchenko@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lukas@wunner.de \
--cc=michael.jamet@intel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).