linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Kirill kapranov <kirill.kapranov@compulab.co.il>
To: Mark Brown <broonie@kernel.org>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>,
	linux-spi@vger.kernel.org, linux-renesas-soc@vger.kernel.org,
	linux-kernel@vger.kernel.org, festevam@gmail.com,
	linux@roeck-us.net
Subject: Re: [PATCH -next] spi: Fix double IDR allocation with DT aliases
Date: Tue, 28 Aug 2018 22:47:05 +0300 (IDT)	[thread overview]
Message-ID: <2024176715.614.1535485625914.JavaMail.root@compulab.co.il> (raw)
In-Reply-To: <20180826132415.GB2414@sirena.org.uk>

> Right, that clearly wasn't an intended effect, though - should be using
> the max of the big constant and the maximum static ID.

Due to difficulties of enumeration of all device in a system, I propose
to assign safe sub-range for dynamic IDs in the upper-half of ID range the following way.

NOTE-0: This patch has to be applied after Geert's patch,
which fixes double call of idr_alloc. (above in the thread). Many thanks, Geert!

NOTE-1: The best solution, IMHO, would be migration to property_get_*
functions, declared in linux/property.h 

[PATCH] Eliminate possibility of conflict between static and dynamic IDs

For systems without DT support allocate dynamical bus ID in a safe sub-range
(if given), otherwise in upper half of the range so as to avoid possible
conflict between statically and dynamically allocated IDs.

Signed-off-by: Kirill Kapranov <kirill.kapranov@compulab.co.il>
---
 drivers/spi/Kconfig |  8 ++++++++
 drivers/spi/spi.c   | 22 +++++++++++++++++++++-
 2 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 671d078349cc..7498fae0113b 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -47,6 +47,14 @@ config SPI_MASTER
 
 if SPI_MASTER
 
+config SPI_MASTER_DYN_BUS_NUM_FIRST
+	int "First dynamically assigned SPI bus ID" if EXPERT
+	default 0
+	help
+	  This value can be used as the beginning of sub-range for dynamic
+	  allocation of SPI bus ID in case of absence of DT. If -1 chosen, the
+	  sub-range will be allocated in upper half of the SPI bus ID range.
+
 config SPI_MEM
 	bool "SPI memory extension"
 	help
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 9da0bc5a036c..3ac0cf0ab49c 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -49,6 +49,18 @@
 
 #include "internals.h"
 
+#ifndef CHAR_BIT
+#define CHAR_BIT 8	/* Normally in <limits.h> */
+#endif
+#if !defined(CONFIG_SPI_MASTER_DYN_BUS_NUM_FIRST) || \
+	CONFIG_SPI_MASTER_DYN_BUS_NUM_FIRST < 0
+//Half of the biggest signed int of this size
+#define SPI_DYN_FIRST_NUM (((1 << \
+	(sizeof(((struct spi_controller*)0)->bus_num) * CHAR_BIT - 1)) - 1) / 2)
+#else
+#define SPI_DYN_FIRST_NUM CONFIG_SPI_MASTER_DYN_BUS_NUM_FIRST
+#endif
+
 static DEFINE_IDR(spi_master_idr);
 
 static void spidev_release(struct device *dev)
@@ -2167,7 +2179,15 @@ int spi_register_controller(struct spi_controller *ctlr)
 	}
 	if (ctlr->bus_num < 0) {
 		first_dynamic = of_alias_get_highest_id("spi");
-		if (first_dynamic < 0)
+
+		if (first_dynamic == -ENOSYS)
+			/*
+			 * In case of system without DT support, allocate
+			 * dynamic bus ID in safe range, higher than the bound,
+			 * to avoid conflict between static and dynamic ID
+			 */
+			first_dynamic = SPI_DYN_FIRST_NUM;
+		else if (first_dynamic < 0)
 			first_dynamic = 0;
 		else
 			first_dynamic++;
-- 
2.11.0

  reply	other threads:[~2018-08-28 19:47 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-21  9:53 [PATCH -next] spi: Fix double IDR allocation with DT aliases Geert Uytterhoeven
2018-08-21 13:40 ` Greg KH
2018-08-21 17:42   ` Geert Uytterhoeven
2018-08-22 17:51 ` Kirill Kapranov
2018-08-23 10:21   ` Mark Brown
2018-08-25 17:54     ` Kirill Kapranov
2018-08-26 13:24       ` Mark Brown
2018-08-28 19:47         ` Kirill kapranov [this message]
2018-08-27 14:13 ` Fabio Estevam
2018-08-28 20:58 ` Applied "spi: Fix double IDR allocation with DT aliases" to the spi tree Mark Brown

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=2024176715.614.1535485625914.JavaMail.root@compulab.co.il \
    --to=kirill.kapranov@compulab.co.il \
    --cc=broonie@kernel.org \
    --cc=festevam@gmail.com \
    --cc=geert+renesas@glider.be \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-renesas-soc@vger.kernel.org \
    --cc=linux-spi@vger.kernel.org \
    --cc=linux@roeck-us.net \
    /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).