Skip to contentSkip to conversation

drivers: make early_platform code SuperH-specific

PATCH In review Bartosz Golaszewski 6 participants 15 messages Oct 3, 2019 – Dec 3, 2019 1 message not in the archive

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

v4 cover drivers: make early_platform code SuperH-specific

From: Bartosz Golaszewski <[email protected]>

Some time ago I started a discussion about the need for a proper early device probing mechanism[1]. One that would be based on real platform drivers and support both platform data and device tree.

While we're far from reaching any consensus on the implementation, Arnd suggested that I start off by moving the SuperH-specific early platform drivers implementation to arch/sh[2].

This series is the first attempt at making way for a new, less hacky implementation.

The first patch moves all the early_platform code to arch/sh.

The second patch prefixes all early_platform symbols with 'sh_'.

[1] https://lkml.org/lkml/2018/4/26/657
[2] https://lkml.org/lkml/2018/4/27/239

v1 -> v2:
- certain drivers are compiled for arm/mach-shmobile too - we need to add ifdefs for CONFIG_SUPERH around early_platform calls

v2 -> v3:
- added a stub for is_early_platform_device() which always returns false on non-SuperH architectures

v3 -> v4:
- rebased on top of v5.4-rc1
- removed patches that are already upstream from the series

Bartosz Golaszewski (2):
drivers: move the early platform device support to arch/sh
sh: add the sh_ prefix to early platform symbols

 arch/sh/drivers/Makefile               |   2 +-
 arch/sh/drivers/platform_early.c       | 347 +++++++++++++++++++++++++
 arch/sh/include/asm/platform_early.h   |  61 +++++
 arch/sh/kernel/cpu/sh2/setup-sh7619.c  |   3 +-
 arch/sh/kernel/cpu/sh2a/setup-mxg.c    |   3 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7201.c |   3 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7203.c |   3 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7206.c |   3 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7264.c |   3 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7269.c |   3 +-
 arch/sh/kernel/cpu/sh3/setup-sh3.c     |   1 +
 arch/sh/kernel/cpu/sh3/setup-sh7705.c  |   3 +-
 arch/sh/kernel/cpu/sh3/setup-sh770x.c  |   3 +-
 arch/sh/kernel/cpu/sh3/setup-sh7710.c  |   3 +-
 arch/sh/kernel/cpu/sh3/setup-sh7720.c  |   3 +-
 arch/sh/kernel/cpu/sh4/setup-sh4-202.c |   3 +-
 arch/sh/kernel/cpu/sh4/setup-sh7750.c  |   9 +-
 arch/sh/kernel/cpu/sh4/setup-sh7760.c  |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7343.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7366.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7722.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7723.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7724.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7734.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7757.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7763.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7770.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7780.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7785.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7786.c |   3 +-
 arch/sh/kernel/cpu/sh4a/setup-shx3.c   |   3 +-
 arch/sh/kernel/cpu/sh5/setup-sh5.c     |   3 +-
 arch/sh/kernel/setup.c                 |   3 +-
 arch/sh/kernel/time.c                  |   5 +-
 drivers/base/platform.c                | 288 --------------------
 drivers/clocksource/sh_cmt.c           |  13 +-
 drivers/clocksource/sh_mtu2.c          |  13 +-
 drivers/clocksource/sh_tmu.c           |  14 +-
 drivers/tty/serial/sh-sci.c            |  11 +-
 include/linux/platform_device.h        |  64 +----
 40 files changed, 525 insertions(+), 387 deletions(-)
 create mode 100644 arch/sh/drivers/platform_early.c
 create mode 100644 arch/sh/include/asm/platform_early.h
Signature · 1 line
2.23.0
#1 lore raw

PATCH v4 1/2 drivers: move the early platform device support to arch/sh

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

---
 arch/sh/drivers/Makefile               |   2 +-
 arch/sh/drivers/platform_early.c       | 347 +++++++++++++++++++++++++
 arch/sh/include/asm/platform_early.h   |  61 +++++
 arch/sh/kernel/cpu/sh2/setup-sh7619.c  |   1 +
 arch/sh/kernel/cpu/sh2a/setup-mxg.c    |   1 +
 arch/sh/kernel/cpu/sh2a/setup-sh7201.c |   1 +
 arch/sh/kernel/cpu/sh2a/setup-sh7203.c |   1 +
 arch/sh/kernel/cpu/sh2a/setup-sh7206.c |   1 +
 arch/sh/kernel/cpu/sh2a/setup-sh7264.c |   1 +
 arch/sh/kernel/cpu/sh2a/setup-sh7269.c |   1 +
 arch/sh/kernel/cpu/sh3/setup-sh3.c     |   1 +
 arch/sh/kernel/cpu/sh3/setup-sh7705.c  |   1 +
 arch/sh/kernel/cpu/sh3/setup-sh770x.c  |   1 +
 arch/sh/kernel/cpu/sh3/setup-sh7710.c  |   1 +
 arch/sh/kernel/cpu/sh3/setup-sh7720.c  |   1 +
 arch/sh/kernel/cpu/sh4/setup-sh4-202.c |   1 +
 arch/sh/kernel/cpu/sh4/setup-sh7750.c  |   1 +
 arch/sh/kernel/cpu/sh4/setup-sh7760.c  |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7343.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7366.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7722.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7723.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7724.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7734.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7757.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7763.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7770.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7780.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7785.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-sh7786.c |   1 +
 arch/sh/kernel/cpu/sh4a/setup-shx3.c   |   1 +
 arch/sh/kernel/cpu/sh5/setup-sh5.c     |   1 +
 arch/sh/kernel/setup.c                 |   1 +
 arch/sh/kernel/time.c                  |   1 +
 drivers/base/platform.c                | 288 --------------------
 drivers/clocksource/sh_cmt.c           |   7 +
 drivers/clocksource/sh_mtu2.c          |   7 +
 drivers/clocksource/sh_tmu.c           |   8 +
 drivers/tty/serial/sh-sci.c            |   7 +-
 include/linux/platform_device.h        |  64 +----
 40 files changed, 480 insertions(+), 342 deletions(-)
 create mode 100644 arch/sh/drivers/platform_early.c
 create mode 100644 arch/sh/include/asm/platform_early.h
arch/sh/drivers/Makefile +1 -1
index 3e93b434e604..56b0acace6e7 100644
@@ -3,7 +3,7 @@
 # Makefile for the Linux SuperH-specific device drivers.
 #
 
-obj-y		+= dma/
+obj-y		+= dma/ platform_early.o
 
 obj-$(CONFIG_PCI)		+= pci/
 obj-$(CONFIG_SUPERHYWAY)	+= superhyway/
arch/sh/drivers/platform_early.c new file +347 -0

347 lines not on this page · raw message

arch/sh/include/asm/platform_early.h new file +61 -0

61 lines not on this page · raw message

arch/sh/kernel/cpu/sh2/setup-sh7619.c +1 -0
index f5b6841ef7e1..27d459e8a598 100644
@@ -12,6 +12,7 @@
 #include <linux/sh_eth.h>
 #include <linux/sh_timer.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh2a/setup-mxg.c +1 -0
index 52350ad0b0a2..8a19c5754978 100644
@@ -9,6 +9,7 @@
 #include <linux/serial.h>
 #include <linux/serial_sci.h>
 #include <linux/sh_timer.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh2a/setup-sh7201.c +1 -0
index b51ed761ae08..d975e4d404bd 100644
@@ -11,6 +11,7 @@
 #include <linux/serial_sci.h>
 #include <linux/sh_timer.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh2a/setup-sh7203.c +1 -0
index 89b3e49fc250..318b8cd8d5b7 100644
@@ -10,6 +10,7 @@
 #include <linux/serial_sci.h>
 #include <linux/sh_timer.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh2a/setup-sh7206.c +1 -0
index 36ff3a3139da..8e6fae60074a 100644
@@ -11,6 +11,7 @@
 #include <linux/serial_sci.h>
 #include <linux/sh_timer.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh2a/setup-sh7264.c +1 -0
index d199618d877c..2d95c9b38330 100644
@@ -11,6 +11,7 @@
 #include <linux/usb/r8a66597.h>
 #include <linux/sh_timer.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh2a/setup-sh7269.c +1 -0
index 9095c960b455..ef4ff76cbe9e 100644
@@ -12,6 +12,7 @@
 #include <linux/usb/r8a66597.h>
 #include <linux/sh_timer.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh3/setup-sh3.c +1 -0
index 8058c01cf09d..cf2a3f09fee4 100644
@@ -8,6 +8,7 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 /* All SH3 devices are equipped with IRQ0->5 (except sh7708) */
 
arch/sh/kernel/cpu/sh3/setup-sh7705.c +1 -0
index e19d1ce7b6ad..9b34d1f4e797 100644
@@ -14,6 +14,7 @@
 #include <linux/sh_intc.h>
 #include <asm/rtc.h>
 #include <cpu/serial.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh3/setup-sh770x.c +1 -0
index 5c5144bee6bc..39f09504bd96 100644
@@ -18,6 +18,7 @@
 #include <linux/sh_timer.h>
 #include <linux/sh_intc.h>
 #include <cpu/serial.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh3/setup-sh7710.c +1 -0
index 4776e2495738..c006c8fcfa0f 100644
@@ -13,6 +13,7 @@
 #include <linux/sh_timer.h>
 #include <linux/sh_intc.h>
 #include <asm/rtc.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh3/setup-sh7720.c +1 -0
index 1d4c34e7b7db..cf2eec9f6abd 100644
@@ -19,6 +19,7 @@
 #include <linux/sh_intc.h>
 #include <linux/usb/ohci_pdriver.h>
 #include <asm/rtc.h>
+#include <asm/platform_early.h>
 #include <cpu/serial.h>
 
 static struct resource rtc_resources[] = {
arch/sh/kernel/cpu/sh4/setup-sh4-202.c +1 -0
index a40ef35d101a..2480ce4a3b1e 100644
@@ -12,6 +12,7 @@
 #include <linux/sh_timer.h>
 #include <linux/sh_intc.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif0_platform_data = {
 	.scscr		= SCSCR_REIE,
arch/sh/kernel/cpu/sh4/setup-sh7750.c +1 -0
index b37bda66a532..4592d660b942 100644
@@ -13,6 +13,7 @@
 #include <linux/sh_intc.h>
 #include <linux/serial_sci.h>
 #include <generated/machtypes.h>
+#include <asm/platform_early.h>
 
 static struct resource rtc_resources[] = {
 	[0] = {
arch/sh/kernel/cpu/sh4/setup-sh7760.c +1 -0
index 86845da85997..e64eb96e3589 100644
@@ -11,6 +11,7 @@
 #include <linux/sh_intc.h>
 #include <linux/serial_sci.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 enum {
 	UNUSED = 0,
arch/sh/kernel/cpu/sh4a/setup-sh7343.c +1 -0
index a15e25690b5f..0fa56128bdca 100644
@@ -12,6 +12,7 @@
 #include <linux/sh_timer.h>
 #include <linux/sh_intc.h>
 #include <asm/clock.h>
+#include <asm/platform_early.h>
 
 /* Serial */
 static struct plat_sci_port scif0_platform_data = {
arch/sh/kernel/cpu/sh4a/setup-sh7366.c +1 -0
index 7bd2776441ba..9f2f75285ec3 100644
@@ -15,6 +15,7 @@
 #include <linux/sh_intc.h>
 #include <linux/usb/r8a66597.h>
 #include <asm/clock.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif0_platform_data = {
 	.scscr		= SCSCR_REIE,
arch/sh/kernel/cpu/sh4a/setup-sh7722.c +1 -0
index 1ce65f88f060..8e5eaac789a7 100644
@@ -18,6 +18,7 @@
 #include <asm/clock.h>
 #include <asm/mmzone.h>
 #include <asm/siu.h>
+#include <asm/platform_early.h>
 
 #include <cpu/dma-register.h>
 #include <cpu/sh7722.h>
arch/sh/kernel/cpu/sh4a/setup-sh7723.c +1 -0
index edb649950662..b4b54aa0c267 100644
@@ -16,6 +16,7 @@
 #include <linux/io.h>
 #include <asm/clock.h>
 #include <asm/mmzone.h>
+#include <asm/platform_early.h>
 #include <cpu/sh7723.h>
 
 /* Serial */
arch/sh/kernel/cpu/sh4a/setup-sh7724.c +1 -0
index 3e9825031d3d..ec237ab112cf 100644
@@ -24,6 +24,7 @@
 #include <asm/suspend.h>
 #include <asm/clock.h>
 #include <asm/mmzone.h>
+#include <asm/platform_early.h>
 
 #include <cpu/dma-register.h>
 #include <cpu/sh7724.h>
arch/sh/kernel/cpu/sh4a/setup-sh7734.c +1 -0
index 06a91569697a..7ea9a8b43955 100644
@@ -18,6 +18,7 @@
 #include <linux/io.h>
 #include <asm/clock.h>
 #include <asm/irq.h>
+#include <asm/platform_early.h>
 #include <cpu/sh7734.h>
 
 /* SCIF */
arch/sh/kernel/cpu/sh4a/setup-sh7757.c +1 -0
index 2501ce656511..922d9886be18 100644
@@ -19,6 +19,7 @@
 #include <linux/usb/ohci_pdriver.h>
 #include <cpu/dma-register.h>
 #include <cpu/sh7757.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif2_platform_data = {
 	.scscr		= SCSCR_REIE,
arch/sh/kernel/cpu/sh4a/setup-sh7763.c +1 -0
index 419c5efe4a17..8725d4751baf 100644
@@ -14,6 +14,7 @@
 #include <linux/io.h>
 #include <linux/serial_sci.h>
 #include <linux/usb/ohci_pdriver.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif0_platform_data = {
 	.scscr		= SCSCR_REIE,
arch/sh/kernel/cpu/sh4a/setup-sh7770.c +1 -0
index 5fb4cf9b58c6..035a0d4f2a2e 100644
@@ -11,6 +11,7 @@
 #include <linux/sh_timer.h>
 #include <linux/sh_intc.h>
 #include <linux/io.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif0_platform_data = {
 	.scscr		= SCSCR_REIE | SCSCR_TOIE,
arch/sh/kernel/cpu/sh4a/setup-sh7780.c +1 -0
index ab7d6b715865..983be87cb7be 100644
@@ -13,6 +13,7 @@
 #include <linux/sh_timer.h>
 #include <linux/sh_intc.h>
 #include <cpu/dma-register.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif0_platform_data = {
 	.scscr		= SCSCR_REIE | SCSCR_CKE1,
arch/sh/kernel/cpu/sh4a/setup-sh7785.c +1 -0
index a438da47285d..bb0cb710c16e 100644
@@ -14,6 +14,7 @@
 #include <linux/sh_timer.h>
 #include <linux/sh_intc.h>
 #include <asm/mmzone.h>
+#include <asm/platform_early.h>
 #include <cpu/dma-register.h>
 
 static struct plat_sci_port scif0_platform_data = {
arch/sh/kernel/cpu/sh4a/setup-sh7786.c +1 -0
index d894165a0ef6..b5d94366eed3 100644
@@ -23,6 +23,7 @@
 #include <linux/usb/ohci_pdriver.h>
 #include <cpu/dma-register.h>
 #include <asm/mmzone.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif0_platform_data = {
 	.scscr		= SCSCR_REIE | SCSCR_CKE1,
arch/sh/kernel/cpu/sh4a/setup-shx3.c +1 -0
index 14aa4552bc45..e82398d80b74 100644
@@ -14,6 +14,7 @@
 #include <linux/sh_intc.h>
 #include <cpu/shx3.h>
 #include <asm/mmzone.h>
+#include <asm/platform_early.h>
 
 /*
  * This intentionally only registers SCIF ports 0, 1, and 3. SCIF 2
arch/sh/kernel/cpu/sh5/setup-sh5.c +1 -0
index 41c1673afc0b..584a6aeeca26 100644
@@ -12,6 +12,7 @@
 #include <linux/mm.h>
 #include <linux/sh_timer.h>
 #include <asm/addrspace.h>
+#include <asm/platform_early.h>
 
 static struct plat_sci_port scif0_platform_data = {
 	.flags		= UPF_IOREMAP,
arch/sh/kernel/setup.c +1 -0
index 2c0e0f37a318..e535507851e1 100644
@@ -44,6 +44,7 @@
 #include <asm/mmu_context.h>
 #include <asm/mmzone.h>
 #include <asm/sparsemem.h>
+#include <asm/platform_early.h>
 
 /*
  * Initialize loops_per_jiffy as 10000000 (1000MIPS).
arch/sh/kernel/time.c +1 -0
index e16b2cd269a3..8ad9fd5c4ff6 100644
@@ -18,6 +18,7 @@
 #include <linux/rtc.h>
 #include <asm/clock.h>
 #include <asm/rtc.h>
+#include <asm/platform_early.h>
 
 static void __init sh_late_time_init(void)
 {
drivers/base/platform.c +0 -288

297 lines not on this page · raw message

drivers/clocksource/sh_cmt.c +7 -0
index ef773db080e9..3a185485300e 100644
@@ -25,6 +25,10 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
+#ifdef CONFIG_SUPERH
+#include <asm/platform_early.h>
+#endif
+
 struct sh_cmt_device;
 
 /*
@@ -1109,7 +1113,10 @@ static void __exit sh_cmt_exit(void)
 	platform_driver_unregister(&sh_cmt_device_driver);
 }
 
+#ifdef CONFIG_SUPERH
 early_platform_init("earlytimer", &sh_cmt_device_driver);
+#endif
+
 subsys_initcall(sh_cmt_init);
 module_exit(sh_cmt_exit);
 
drivers/clocksource/sh_mtu2.c +7 -0
index 354b27d14a19..d581060d33a2 100644
@@ -23,6 +23,10 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
+#ifdef CONFIG_SUPERH
+#include <asm/platform_early.h>
+#endif
+
 struct sh_mtu2_device;
 
 struct sh_mtu2_channel {
@@ -511,7 +515,10 @@ static void __exit sh_mtu2_exit(void)
 	platform_driver_unregister(&sh_mtu2_device_driver);
 }
 
+#ifdef CONFIG_SUPERH
 early_platform_init("earlytimer", &sh_mtu2_device_driver);
+#endif
+
 subsys_initcall(sh_mtu2_init);
 module_exit(sh_mtu2_exit);
 
drivers/clocksource/sh_tmu.c +8 -0
index 8c4f3753b36e..50e30ead050e 100644
@@ -24,6 +24,10 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 
+#ifdef CONFIG_SUPERH
+#include <asm/platform_early.h>
+#endif
+
 enum sh_tmu_model {
 	SH_TMU,
 	SH_TMU_SH3,
@@ -615,6 +619,7 @@ static int sh_tmu_probe(struct platform_device *pdev)
 		pm_runtime_idle(&pdev->dev);
 		return ret;
 	}
+
 	if (is_early_platform_device(pdev))
 		return 0;
 
@@ -665,7 +670,10 @@ static void __exit sh_tmu_exit(void)
 	platform_driver_unregister(&sh_tmu_device_driver);
 }
 
+#ifdef CONFIG_SUPERH
 early_platform_init("earlytimer", &sh_tmu_device_driver);
+#endif
+
 subsys_initcall(sh_tmu_init);
 module_exit(sh_tmu_exit);
 
drivers/tty/serial/sh-sci.c +6 -1
index 4e754a4850e6..2a645fd7f07b 100644
@@ -54,6 +54,7 @@
 
 #ifdef CONFIG_SUPERH
 #include <asm/sh_bios.h>
+#include <asm/platform_early.h>
 #endif
 
 #include "serial_mctrl_gpio.h"
@@ -3086,6 +3087,7 @@ static struct console serial_console = {
 	.data		= &sci_uart_driver,
 };
 
+#ifdef CONFIG_SUPERH
 static struct console early_serial_console = {
 	.name           = "early_ttySC",
 	.write          = serial_console_write,
@@ -3114,6 +3116,7 @@ static int sci_probe_earlyprintk(struct platform_device *pdev)
 	register_console(&early_serial_console);
 	return 0;
 }
+#endif
 
 #define SCI_CONSOLE	(&serial_console)
 
@@ -3314,8 +3317,10 @@ static int sci_probe(struct platform_device *dev)
 	 * the special early probe. We don't have sufficient device state
 	 * to make it beyond this yet.
 	 */
+#ifdef CONFIG_SUPERH
 	if (is_early_platform_device(dev))
 		return sci_probe_earlyprintk(dev);
+#endif
 
 	if (dev->dev.of_node) {
 		p = sci_parse_dt(dev, &dev_id);
@@ -3410,7 +3415,7 @@ static void __exit sci_exit(void)
 		uart_unregister_driver(&sci_uart_driver);
 }
 
-#ifdef CONFIG_SERIAL_SH_SCI_CONSOLE
+#if defined(CONFIG_SUPERH) && defined(CONFIG_SERIAL_SH_SCI_CONSOLE)
 early_platform_init_buffer("earlyprintk", &sci_driver,
 			   early_serial_buf, ARRAY_SIZE(early_serial_buf));
 #endif
include/linux/platform_device.h +12 -52

74 lines not on this page · raw message

Signature · 1 line
2.23.0
#2 lore raw

PATCH v4 2/2 sh: add the sh_ prefix to early platform symbols

In reply to Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Old early platform device support is now sh-specific. Before moving on to implementing new early platform framework based on real platform devices, prefix all early platform symbols with 'sh_'.

---
 arch/sh/drivers/platform_early.c       | 66 +++++++++++++-------------
 arch/sh/include/asm/platform_early.h   | 30 ++++++------
 arch/sh/kernel/cpu/sh2/setup-sh7619.c  |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-mxg.c    |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7201.c |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7203.c |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7206.c |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7264.c |  2 +-
 arch/sh/kernel/cpu/sh2a/setup-sh7269.c |  2 +-
 arch/sh/kernel/cpu/sh3/setup-sh7705.c  |  2 +-
 arch/sh/kernel/cpu/sh3/setup-sh770x.c  |  2 +-
 arch/sh/kernel/cpu/sh3/setup-sh7710.c  |  2 +-
 arch/sh/kernel/cpu/sh3/setup-sh7720.c  |  2 +-
 arch/sh/kernel/cpu/sh4/setup-sh4-202.c |  2 +-
 arch/sh/kernel/cpu/sh4/setup-sh7750.c  |  8 ++--
 arch/sh/kernel/cpu/sh4/setup-sh7760.c  |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7343.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7366.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7722.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7723.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7724.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7734.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7757.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7763.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7770.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7780.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7785.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-sh7786.c |  2 +-
 arch/sh/kernel/cpu/sh4a/setup-shx3.c   |  2 +-
 arch/sh/kernel/cpu/sh5/setup-sh5.c     |  2 +-
 arch/sh/kernel/setup.c                 |  2 +-
 arch/sh/kernel/time.c                  |  4 +-
 drivers/clocksource/sh_cmt.c           |  6 +--
 drivers/clocksource/sh_mtu2.c          |  6 +--
 drivers/clocksource/sh_tmu.c           |  6 +--
 drivers/tty/serial/sh-sci.c            |  4 +-
 include/linux/platform_device.h        |  2 +-
 37 files changed, 94 insertions(+), 94 deletions(-)
arch/sh/drivers/platform_early.c +33 -33
index 3dfe9f3541bb..f6d148451dfc 100644
@@ -4,8 +4,8 @@
 #include <linux/mod_devicetable.h>
 #include <linux/pm.h>
 
-static __initdata LIST_HEAD(early_platform_driver_list);
-static __initdata LIST_HEAD(early_platform_device_list);
+static __initdata LIST_HEAD(sh_early_platform_driver_list);
+static __initdata LIST_HEAD(sh_early_platform_device_list);
 
 static const struct platform_device_id *
 platform_match_id(const struct platform_device_id *id,
@@ -58,13 +58,13 @@ static void pm_runtime_early_init(struct device *dev) {}
 #endif
 
 /**
- * early_platform_driver_register - register early platform driver
- * @epdrv: early_platform driver structure
+ * sh_early_platform_driver_register - register early platform driver
+ * @epdrv: sh_early_platform driver structure
  * @buf: string passed from early_param()
  *
- * Helper function for early_platform_init() / early_platform_init_buffer()
+ * Helper function for sh_early_platform_init() / sh_early_platform_init_buffer()
  */
-int __init early_platform_driver_register(struct early_platform_driver *epdrv,
+int __init sh_early_platform_driver_register(struct sh_early_platform_driver *epdrv,
 					  char *buf)
 {
 	char *tmp;
@@ -75,7 +75,7 @@ int __init early_platform_driver_register(struct early_platform_driver *epdrv,
 	 */
 	if (!epdrv->list.next) {
 		INIT_LIST_HEAD(&epdrv->list);
-		list_add_tail(&epdrv->list, &early_platform_driver_list);
+		list_add_tail(&epdrv->list, &sh_early_platform_driver_list);
 	}
 
 	/* If the user has specified device then make sure the driver
@@ -84,7 +84,7 @@ int __init early_platform_driver_register(struct early_platform_driver *epdrv,
 	 */
 	n = strlen(epdrv->pdrv->driver.name);
 	if (buf && !strncmp(buf, epdrv->pdrv->driver.name, n)) {
-		list_move(&epdrv->list, &early_platform_driver_list);
+		list_move(&epdrv->list, &sh_early_platform_driver_list);
 
 		/* Allow passing parameters after device name */
 		if (buf[n] == '\0' || buf[n] == ',')
@@ -114,14 +114,14 @@ int __init early_platform_driver_register(struct early_platform_driver *epdrv,
 }
 
 /**
- * early_platform_add_devices - adds a number of early platform devices
+ * sh_early_platform_add_devices - adds a number of early platform devices
  * @devs: array of early platform devices to add
  * @num: number of early platform devices in array
  *
  * Used by early architecture code to register early platform devices and
  * their platform data.
  */
-void __init early_platform_add_devices(struct platform_device **devs, int num)
+void __init sh_early_platform_add_devices(struct platform_device **devs, int num)
 {
 	struct device *dev;
 	int i;
@@ -134,20 +134,20 @@ void __init early_platform_add_devices(struct platform_device **devs, int num)
 			pm_runtime_early_init(dev);
 			INIT_LIST_HEAD(&dev->devres_head);
 			list_add_tail(&dev->devres_head,
-				      &early_platform_device_list);
+				      &sh_early_platform_device_list);
 		}
 	}
 }
 
 /**
- * early_platform_driver_register_all - register early platform drivers
+ * sh_early_platform_driver_register_all - register early platform drivers
  * @class_str: string to identify early platform driver class
  *
  * Used by architecture code to register all early platform drivers
  * for a certain class. If omitted then only early platform drivers
  * with matching kernel command line class parameters will be registered.
  */
-void __init early_platform_driver_register_all(char *class_str)
+void __init sh_early_platform_driver_register_all(char *class_str)
 {
 	/* The "class_str" parameter may or may not be present on the kernel
 	 * command line. If it is present then there may be more than one
@@ -160,23 +160,23 @@ void __init early_platform_driver_register_all(char *class_str)
 	 * We use parse_early_options() to make sure the early_param() gets
 	 * called at least once. The early_param() may be called more than
 	 * once since the name of the preferred device may be specified on
-	 * the kernel command line. early_platform_driver_register() handles
+	 * the kernel command line. sh_early_platform_driver_register() handles
 	 * this case for us.
 	 */
 	parse_early_options(class_str);
 }
 
 /**
- * early_platform_match - find early platform device matching driver
+ * sh_early_platform_match - find early platform device matching driver
  * @epdrv: early platform driver structure
  * @id: id to match against
  */
 static struct platform_device * __init
-early_platform_match(struct early_platform_driver *epdrv, int id)
+sh_early_platform_match(struct sh_early_platform_driver *epdrv, int id)
 {
 	struct platform_device *pd;
 
-	list_for_each_entry(pd, &early_platform_device_list, dev.devres_head)
+	list_for_each_entry(pd, &sh_early_platform_device_list, dev.devres_head)
 		if (platform_match(&pd->dev, &epdrv->pdrv->driver))
 			if (pd->id == id)
 				return pd;
@@ -185,16 +185,16 @@ early_platform_match(struct early_platform_driver *epdrv, int id)
 }
 
 /**
- * early_platform_left - check if early platform driver has matching devices
+ * sh_early_platform_left - check if early platform driver has matching devices
  * @epdrv: early platform driver structure
  * @id: return true if id or above exists
  */
-static int __init early_platform_left(struct early_platform_driver *epdrv,
+static int __init sh_early_platform_left(struct sh_early_platform_driver *epdrv,
 				       int id)
 {
 	struct platform_device *pd;
 
-	list_for_each_entry(pd, &early_platform_device_list, dev.devres_head)
+	list_for_each_entry(pd, &sh_early_platform_device_list, dev.devres_head)
 		if (platform_match(&pd->dev, &epdrv->pdrv->driver))
 			if (pd->id >= id)
 				return 1;
@@ -203,22 +203,22 @@ static int __init early_platform_left(struct early_platform_driver *epdrv,
 }
 
 /**
- * early_platform_driver_probe_id - probe drivers matching class_str and id
+ * sh_early_platform_driver_probe_id - probe drivers matching class_str and id
  * @class_str: string to identify early platform driver class
  * @id: id to match against
  * @nr_probe: number of platform devices to successfully probe before exiting
  */
-static int __init early_platform_driver_probe_id(char *class_str,
+static int __init sh_early_platform_driver_probe_id(char *class_str,
 						 int id,
 						 int nr_probe)
 {
-	struct early_platform_driver *epdrv;
+	struct sh_early_platform_driver *epdrv;
 	struct platform_device *match;
 	int match_id;
 	int n = 0;
 	int left = 0;
 
-	list_for_each_entry(epdrv, &early_platform_driver_list, list) {
+	list_for_each_entry(epdrv, &sh_early_platform_driver_list, list) {
 		/* only use drivers matching our class_str */
 		if (strcmp(class_str, epdrv->class_str))
 			continue;
@@ -229,7 +229,7 @@ static int __init early_platform_driver_probe_id(char *class_str,
 
 		} else {
 			match_id = id;
-			left += early_platform_left(epdrv, id);
+			left += sh_early_platform_left(epdrv, id);
 
 			/* skip requested id */
 			switch (epdrv->requested_id) {
@@ -251,7 +251,7 @@ static int __init early_platform_driver_probe_id(char *class_str,
 			match = NULL;
 			break;
 		default:
-			match = early_platform_match(epdrv, match_id);
+			match = sh_early_platform_match(epdrv, match_id);
 		}
 
 		if (match) {
@@ -293,7 +293,7 @@ static int __init early_platform_driver_probe_id(char *class_str,
 }
 
 /**
- * early_platform_driver_probe - probe a class of registered drivers
+ * sh_early_platform_driver_probe - probe a class of registered drivers
  * @class_str: string to identify early platform driver class
  * @nr_probe: number of platform devices to successfully probe before exiting
  * @user_only: only probe user specified early platform devices
@@ -302,7 +302,7 @@ static int __init early_platform_driver_probe_id(char *class_str,
  * within a certain class. For probe to happen a registered early platform
  * device matching a registered early platform driver is needed.
  */
-int __init early_platform_driver_probe(char *class_str,
+int __init sh_early_platform_driver_probe(char *class_str,
 				       int nr_probe,
 				       int user_only)
 {
@@ -310,7 +310,7 @@ int __init early_platform_driver_probe(char *class_str,
 
 	n = 0;
 	for (i = -2; n < nr_probe; i++) {
-		k = early_platform_driver_probe_id(class_str, i, nr_probe - n);
+		k = sh_early_platform_driver_probe_id(class_str, i, nr_probe - n);
 
 		if (k < 0)
 			break;
@@ -325,14 +325,14 @@ int __init early_platform_driver_probe(char *class_str,
 }
 
 /**
- * early_platform_cleanup - clean up early platform code
+ * sh_early_platform_cleanup - clean up early platform code
  */
-static int __init early_platform_cleanup(void)
+static int __init sh_early_platform_cleanup(void)
 {
 	struct platform_device *pd, *pd2;
 
 	/* clean up the devres list used to chain devices */
-	list_for_each_entry_safe(pd, pd2, &early_platform_device_list,
+	list_for_each_entry_safe(pd, pd2, &sh_early_platform_device_list,
 				 dev.devres_head) {
 		list_del(&pd->dev.devres_head);
 		memset(&pd->dev.devres_head, 0, sizeof(pd->dev.devres_head));
@@ -344,4 +344,4 @@ static int __init early_platform_cleanup(void)
  * This must happen once after all early devices are probed but before probing
  * real platform devices.
  */
-subsys_initcall(early_platform_cleanup);
+subsys_initcall(sh_early_platform_cleanup);
arch/sh/include/asm/platform_early.h +15 -15
index 4590ab757d5f..fc802137c37d 100644
@@ -8,7 +8,7 @@
 #include <linux/pm_runtime.h>
 #include <linux/slab.h>
 
-struct early_platform_driver {
+struct sh_early_platform_driver {
 	const char *class_str;
 	struct platform_driver *pdrv;
 	struct list_head list;
@@ -20,39 +20,39 @@ struct early_platform_driver {
 #define EARLY_PLATFORM_ID_UNSET -2
 #define EARLY_PLATFORM_ID_ERROR -3
 
-extern int early_platform_driver_register(struct early_platform_driver *epdrv,
+extern int sh_early_platform_driver_register(struct sh_early_platform_driver *epdrv,
 					  char *buf);
-extern void early_platform_add_devices(struct platform_device **devs, int num);
+extern void sh_early_platform_add_devices(struct platform_device **devs, int num);
 
-static inline int is_early_platform_device(struct platform_device *pdev)
+static inline int is_sh_early_platform_device(struct platform_device *pdev)
 {
 	return !pdev->dev.driver;
 }
 
-extern void early_platform_driver_register_all(char *class_str);
-extern int early_platform_driver_probe(char *class_str,
+extern void sh_early_platform_driver_register_all(char *class_str);
+extern int sh_early_platform_driver_probe(char *class_str,
 				       int nr_probe, int user_only);
 
-#define early_platform_init(class_string, platdrv)		\
-	early_platform_init_buffer(class_string, platdrv, NULL, 0)
+#define sh_early_platform_init(class_string, platdrv)		\
+	sh_early_platform_init_buffer(class_string, platdrv, NULL, 0)
 
 #ifndef MODULE
-#define early_platform_init_buffer(class_string, platdrv, buf, bufsiz)	\
-static __initdata struct early_platform_driver early_driver = {		\
+#define sh_early_platform_init_buffer(class_string, platdrv, buf, bufsiz)	\
+static __initdata struct sh_early_platform_driver early_driver = {		\
 	.class_str = class_string,					\
 	.buffer = buf,							\
 	.bufsize = bufsiz,						\
 	.pdrv = platdrv,						\
 	.requested_id = EARLY_PLATFORM_ID_UNSET,			\
 };									\
-static int __init early_platform_driver_setup_func(char *buffer)	\
+static int __init sh_early_platform_driver_setup_func(char *buffer)	\
 {									\
-	return early_platform_driver_register(&early_driver, buffer);	\
+	return sh_early_platform_driver_register(&early_driver, buffer);	\
 }									\
-early_param(class_string, early_platform_driver_setup_func)
+early_param(class_string, sh_early_platform_driver_setup_func)
 #else /* MODULE */
-#define early_platform_init_buffer(class_string, platdrv, buf, bufsiz)	\
-static inline char *early_platform_driver_setup_func(void)		\
+#define sh_early_platform_init_buffer(class_string, platdrv, buf, bufsiz)	\
+static inline char *sh_early_platform_driver_setup_func(void)		\
 {									\
 	return bufsiz ? buf : NULL;					\
 }
arch/sh/kernel/cpu/sh2/setup-sh7619.c +1 -1
index 27d459e8a598..b1c877b6a420 100644
@@ -200,6 +200,6 @@ void __init plat_early_device_setup(void)
 	/* enable CMT clock */
 	__raw_writeb(__raw_readb(STBCR3) & ~0x10, STBCR3);
 
-	early_platform_add_devices(sh7619_early_devices,
+	sh_early_platform_add_devices(sh7619_early_devices,
 				   ARRAY_SIZE(sh7619_early_devices));
 }
arch/sh/kernel/cpu/sh2a/setup-mxg.c +1 -1
index 8a19c5754978..cefa07924c16 100644
@@ -170,6 +170,6 @@ static struct platform_device *mxg_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(mxg_early_devices,
+	sh_early_platform_add_devices(mxg_early_devices,
 				   ARRAY_SIZE(mxg_early_devices));
 }
arch/sh/kernel/cpu/sh2a/setup-sh7201.c +1 -1
index d975e4d404bd..28f1bebf3405 100644
@@ -413,6 +413,6 @@ void __init plat_early_device_setup(void)
 	/* enable MTU2 clock */
 	__raw_writeb(__raw_readb(STBCR3) & ~0x20, STBCR3);
 
-	early_platform_add_devices(sh7201_early_devices,
+	sh_early_platform_add_devices(sh7201_early_devices,
 				   ARRAY_SIZE(sh7201_early_devices));
 }
arch/sh/kernel/cpu/sh2a/setup-sh7203.c +1 -1
index 318b8cd8d5b7..4839f3aaeb4c 100644
@@ -350,6 +350,6 @@ void __init plat_early_device_setup(void)
 	/* enable MTU2 clock */
 	__raw_writeb(__raw_readb(STBCR3) & ~0x20, STBCR3);
 
-	early_platform_add_devices(sh7203_early_devices,
+	sh_early_platform_add_devices(sh7203_early_devices,
 				   ARRAY_SIZE(sh7203_early_devices));
 }
arch/sh/kernel/cpu/sh2a/setup-sh7206.c +1 -1
index 8e6fae60074a..68add5af4cc5 100644
@@ -286,6 +286,6 @@ void __init plat_early_device_setup(void)
 	/* enable MTU2 clock */
 	__raw_writeb(__raw_readb(STBCR3) & ~0x20, STBCR3);
 
-	early_platform_add_devices(sh7206_early_devices,
+	sh_early_platform_add_devices(sh7206_early_devices,
 				   ARRAY_SIZE(sh7206_early_devices));
 }
arch/sh/kernel/cpu/sh2a/setup-sh7264.c +1 -1
index 2d95c9b38330..8a1cb613dd2e 100644
@@ -547,6 +547,6 @@ static struct platform_device *sh7264_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7264_early_devices,
+	sh_early_platform_add_devices(sh7264_early_devices,
 				   ARRAY_SIZE(sh7264_early_devices));
 }
arch/sh/kernel/cpu/sh2a/setup-sh7269.c +1 -1
index ef4ff76cbe9e..8b1ef3028320 100644
@@ -563,6 +563,6 @@ static struct platform_device *sh7269_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7269_early_devices,
+	sh_early_platform_add_devices(sh7269_early_devices,
 				   ARRAY_SIZE(sh7269_early_devices));
 }
arch/sh/kernel/cpu/sh3/setup-sh7705.c +1 -1
index 9b34d1f4e797..0544134b3f20 100644
@@ -179,7 +179,7 @@ static struct platform_device *sh7705_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7705_early_devices,
+	sh_early_platform_add_devices(sh7705_early_devices,
 				   ARRAY_SIZE(sh7705_early_devices));
 }
 
arch/sh/kernel/cpu/sh3/setup-sh770x.c +1 -1
index 39f09504bd96..4947f57748bc 100644
@@ -231,7 +231,7 @@ static struct platform_device *sh770x_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh770x_early_devices,
+	sh_early_platform_add_devices(sh770x_early_devices,
 				   ARRAY_SIZE(sh770x_early_devices));
 }
 
arch/sh/kernel/cpu/sh3/setup-sh7710.c +1 -1
index c006c8fcfa0f..381910761579 100644
@@ -178,7 +178,7 @@ static struct platform_device *sh7710_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7710_early_devices,
+	sh_early_platform_add_devices(sh7710_early_devices,
 				   ARRAY_SIZE(sh7710_early_devices));
 }
 
arch/sh/kernel/cpu/sh3/setup-sh7720.c +1 -1
index cf2eec9f6abd..425d067dae9b 100644
@@ -212,7 +212,7 @@ static struct platform_device *sh7720_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7720_early_devices,
+	sh_early_platform_add_devices(sh7720_early_devices,
 				   ARRAY_SIZE(sh7720_early_devices));
 }
 
arch/sh/kernel/cpu/sh4/setup-sh4-202.c +1 -1
index 2480ce4a3b1e..e6737f3d0df2 100644
@@ -77,7 +77,7 @@ static struct platform_device *sh4202_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh4202_early_devices,
+	sh_early_platform_add_devices(sh4202_early_devices,
 				   ARRAY_SIZE(sh4202_early_devices));
 }
 
arch/sh/kernel/cpu/sh4/setup-sh7750.c +4 -4
index 4592d660b942..19c8f1d69071 100644
@@ -162,15 +162,15 @@ void __init plat_early_device_setup(void)
 	if (mach_is_rts7751r2d()) {
 		scif_platform_data.scscr |= SCSCR_CKE1;
 		dev[0] = &scif_device;
-		early_platform_add_devices(dev, 1);
+		sh_early_platform_add_devices(dev, 1);
 	} else {
 		dev[0] = &sci_device;
-		early_platform_add_devices(dev, 1);
+		sh_early_platform_add_devices(dev, 1);
 		dev[0] = &scif_device;
-		early_platform_add_devices(dev, 1);
+		sh_early_platform_add_devices(dev, 1);
 	}
 
-	early_platform_add_devices(sh7750_early_devices,
+	sh_early_platform_add_devices(sh7750_early_devices,
 				   ARRAY_SIZE(sh7750_early_devices));
 }
 
arch/sh/kernel/cpu/sh4/setup-sh7760.c +1 -1
index e64eb96e3589..14212f5d803c 100644
@@ -272,7 +272,7 @@ static struct platform_device *sh7760_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7760_early_devices,
+	sh_early_platform_add_devices(sh7760_early_devices,
 				   ARRAY_SIZE(sh7760_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7343.c +1 -1
index 0fa56128bdca..b6015188fab1 100644
@@ -297,7 +297,7 @@ static struct platform_device *sh7343_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7343_early_devices,
+	sh_early_platform_add_devices(sh7343_early_devices,
 				   ARRAY_SIZE(sh7343_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7366.c +1 -1
index 9f2f75285ec3..6676beef053e 100644
@@ -241,7 +241,7 @@ static struct platform_device *sh7366_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7366_early_devices,
+	sh_early_platform_add_devices(sh7366_early_devices,
 				   ARRAY_SIZE(sh7366_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7722.c +1 -1
index 8e5eaac789a7..0c6757ef63f4 100644
@@ -513,7 +513,7 @@ static struct platform_device *sh7722_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7722_early_devices,
+	sh_early_platform_add_devices(sh7722_early_devices,
 				   ARRAY_SIZE(sh7722_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7723.c +1 -1
index b4b54aa0c267..83ae1ad4a86e 100644
@@ -411,7 +411,7 @@ static struct platform_device *sh7723_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7723_early_devices,
+	sh_early_platform_add_devices(sh7723_early_devices,
 				   ARRAY_SIZE(sh7723_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7724.c +1 -1
index ec237ab112cf..0d990ab1ba2a 100644
@@ -831,7 +831,7 @@ static struct platform_device *sh7724_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7724_early_devices,
+	sh_early_platform_add_devices(sh7724_early_devices,
 				   ARRAY_SIZE(sh7724_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7734.c +1 -1
index 7ea9a8b43955..9911da794358 100644
@@ -281,7 +281,7 @@ static struct platform_device *sh7734_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7734_early_devices,
+	sh_early_platform_add_devices(sh7734_early_devices,
 		ARRAY_SIZE(sh7734_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7757.c +1 -1
index 922d9886be18..67e330b7ea46 100644
@@ -768,7 +768,7 @@ static struct platform_device *sh7757_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7757_early_devices,
+	sh_early_platform_add_devices(sh7757_early_devices,
 				   ARRAY_SIZE(sh7757_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7763.c +1 -1
index 8725d4751baf..b0608664785f 100644
@@ -222,7 +222,7 @@ static struct platform_device *sh7763_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7763_early_devices,
+	sh_early_platform_add_devices(sh7763_early_devices,
 				   ARRAY_SIZE(sh7763_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7770.c +1 -1
index 035a0d4f2a2e..5efec6ceb04d 100644
@@ -317,7 +317,7 @@ static struct platform_device *sh7770_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7770_early_devices,
+	sh_early_platform_add_devices(sh7770_early_devices,
 				   ARRAY_SIZE(sh7770_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7780.c +1 -1
index 983be87cb7be..c818b788ecb0 100644
@@ -286,7 +286,7 @@ void __init plat_early_device_setup(void)
 		scif1_platform_data.scscr &= ~SCSCR_CKE1;
 	}
 
-	early_platform_add_devices(sh7780_early_devices,
+	sh_early_platform_add_devices(sh7780_early_devices,
 				   ARRAY_SIZE(sh7780_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7785.c +1 -1
index bb0cb710c16e..3b4a414d60a9 100644
@@ -354,7 +354,7 @@ static struct platform_device *sh7785_early_devices[] __initdata = {
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7785_early_devices,
+	sh_early_platform_add_devices(sh7785_early_devices,
 				   ARRAY_SIZE(sh7785_early_devices));
 }
 
arch/sh/kernel/cpu/sh4a/setup-sh7786.c +1 -1
index b5d94366eed3..4b0db8259e3d 100644
@@ -835,6 +835,6 @@ arch_initcall(sh7786_devices_setup);
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh7786_early_devices,
+	sh_early_platform_add_devices(sh7786_early_devices,
 				   ARRAY_SIZE(sh7786_early_devices));
 }
arch/sh/kernel/cpu/sh4a/setup-shx3.c +1 -1
index e82398d80b74..7014d6d199b3 100644
@@ -153,7 +153,7 @@ arch_initcall(shx3_devices_setup);
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(shx3_early_devices,
+	sh_early_platform_add_devices(shx3_early_devices,
 				   ARRAY_SIZE(shx3_early_devices));
 }
 
arch/sh/kernel/cpu/sh5/setup-sh5.c +1 -1
index 584a6aeeca26..dc8476d67244 100644
@@ -116,6 +116,6 @@ arch_initcall(sh5_devices_setup);
 
 void __init plat_early_device_setup(void)
 {
-	early_platform_add_devices(sh5_early_devices,
+	sh_early_platform_add_devices(sh5_early_devices,
 				   ARRAY_SIZE(sh5_early_devices));
 }
arch/sh/kernel/setup.c +1 -1
index e535507851e1..914174a125a4 100644
@@ -329,7 +329,7 @@ void __init setup_arch(char **cmdline_p)
 	sh_mv_setup();
 
 	/* Let earlyprintk output early console messages */
-	early_platform_driver_probe("earlyprintk", 1, 1);
+	sh_early_platform_driver_probe("earlyprintk", 1, 1);
 
 #ifdef CONFIG_OF_FLATTREE
 #ifdef CONFIG_USE_BUILTIN_DTB
arch/sh/kernel/time.c +2 -2
index 8ad9fd5c4ff6..821a09cbd605 100644
@@ -31,8 +31,8 @@ static void __init sh_late_time_init(void)
 	 * clocksource and the jiffies clocksource is used transparently
 	 * instead. No error handling is necessary here.
 	 */
-	early_platform_driver_register_all("earlytimer");
-	early_platform_driver_probe("earlytimer", 2, 0);
+	sh_early_platform_driver_register_all("earlytimer");
+	sh_early_platform_driver_probe("earlytimer", 2, 0);
 }
 
 void __init time_init(void)
drivers/clocksource/sh_cmt.c +3 -3
index 3a185485300e..9cde50cb3220 100644
@@ -1056,7 +1056,7 @@ static int sh_cmt_probe(struct platform_device *pdev)
 	struct sh_cmt_device *cmt = platform_get_drvdata(pdev);
 	int ret;
 
-	if (!is_early_platform_device(pdev)) {
+	if (!is_sh_early_platform_device(pdev)) {
 		pm_runtime_set_active(&pdev->dev);
 		pm_runtime_enable(&pdev->dev);
 	}
@@ -1076,7 +1076,7 @@ static int sh_cmt_probe(struct platform_device *pdev)
 		pm_runtime_idle(&pdev->dev);
 		return ret;
 	}
-	if (is_early_platform_device(pdev))
+	if (is_sh_early_platform_device(pdev))
 		return 0;
 
  out:
@@ -1114,7 +1114,7 @@ static void __exit sh_cmt_exit(void)
 }
 
 #ifdef CONFIG_SUPERH
-early_platform_init("earlytimer", &sh_cmt_device_driver);
+sh_early_platform_init("earlytimer", &sh_cmt_device_driver);
 #endif
 
 subsys_initcall(sh_cmt_init);
drivers/clocksource/sh_mtu2.c +3 -3
index d581060d33a2..c40eef091a04 100644
@@ -446,7 +446,7 @@ static int sh_mtu2_probe(struct platform_device *pdev)
 	struct sh_mtu2_device *mtu = platform_get_drvdata(pdev);
 	int ret;
 
-	if (!is_early_platform_device(pdev)) {
+	if (!is_sh_early_platform_device(pdev)) {
 		pm_runtime_set_active(&pdev->dev);
 		pm_runtime_enable(&pdev->dev);
 	}
@@ -466,7 +466,7 @@ static int sh_mtu2_probe(struct platform_device *pdev)
 		pm_runtime_idle(&pdev->dev);
 		return ret;
 	}
-	if (is_early_platform_device(pdev))
+	if (is_sh_early_platform_device(pdev))
 		return 0;
 
  out:
@@ -516,7 +516,7 @@ static void __exit sh_mtu2_exit(void)
 }
 
 #ifdef CONFIG_SUPERH
-early_platform_init("earlytimer", &sh_mtu2_device_driver);
+sh_early_platform_init("earlytimer", &sh_mtu2_device_driver);
 #endif
 
 subsys_initcall(sh_mtu2_init);
drivers/clocksource/sh_tmu.c +3 -3
index 50e30ead050e..d49690d15536 100644
@@ -599,7 +599,7 @@ static int sh_tmu_probe(struct platform_device *pdev)
 	struct sh_tmu_device *tmu = platform_get_drvdata(pdev);
 	int ret;
 
-	if (!is_early_platform_device(pdev)) {
+	if (!is_sh_early_platform_device(pdev)) {
 		pm_runtime_set_active(&pdev->dev);
 		pm_runtime_enable(&pdev->dev);
 	}
@@ -620,7 +620,7 @@ static int sh_tmu_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	if (is_early_platform_device(pdev))
+	if (is_sh_early_platform_device(pdev))
 		return 0;
 
  out:
@@ -671,7 +671,7 @@ static void __exit sh_tmu_exit(void)
 }
 
 #ifdef CONFIG_SUPERH
-early_platform_init("earlytimer", &sh_tmu_device_driver);
+sh_early_platform_init("earlytimer", &sh_tmu_device_driver);
 #endif
 
 subsys_initcall(sh_tmu_init);
drivers/tty/serial/sh-sci.c +2 -2
index 2a645fd7f07b..970f1f86b945 100644
@@ -3318,7 +3318,7 @@ static int sci_probe(struct platform_device *dev)
 	 * to make it beyond this yet.
 	 */
 #ifdef CONFIG_SUPERH
-	if (is_early_platform_device(dev))
+	if (is_sh_early_platform_device(dev))
 		return sci_probe_earlyprintk(dev);
 #endif
 
@@ -3416,7 +3416,7 @@ static void __exit sci_exit(void)
 }
 
 #if defined(CONFIG_SUPERH) && defined(CONFIG_SERIAL_SH_SCI_CONSOLE)
-early_platform_init_buffer("earlyprintk", &sci_driver,
+sh_early_platform_init_buffer("earlyprintk", &sci_driver,
 			   early_serial_buf, ARRAY_SIZE(early_serial_buf));
 #endif
 #ifdef CONFIG_SERIAL_SH_SCI_EARLYCON
include/linux/platform_device.h +1 -1
index 85aa28a041a9..f086b6a1bc6b 100644
@@ -332,7 +332,7 @@ extern int platform_dma_configure(struct device *dev);
  * drivers. It should go away once we introduce the new platform_device-based
  * early driver framework.
  */
-static inline int is_early_platform_device(struct platform_device *pdev)
+static inline int is_sh_early_platform_device(struct platform_device *pdev)
 {
 	return 0;
 }
Signature · 1 line
2.23.0
#3 lore raw
In reply to Bartosz Golaszewski
Quoted 28 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Some time ago I started a discussion about the need for a proper early device probing mechanism[1]. One that would be based on real platform drivers and support both platform data and device tree.

While we're far from reaching any consensus on the implementation, Arnd suggested that I start off by moving the SuperH-specific early platform drivers implementation to arch/sh[2].

This series is the first attempt at making way for a new, less hacky implementation.

The first patch moves all the early_platform code to arch/sh.

The second patch prefixes all early_platform symbols with 'sh_'.

[1] https://lkml.org/lkml/2018/4/26/657
[2] https://lkml.org/lkml/2018/4/27/239

v1 -> v2:
- certain drivers are compiled for arm/mach-shmobile too - we need to add ifdefs for CONFIG_SUPERH around early_platform calls

v2 -> v3:
- added a stub for is_early_platform_device() which always returns false on non-SuperH architectures

Quoted 7 lines

v3 -> v4:
- rebased on top of v5.4-rc1
- removed patches that are already upstream from the series

Bartosz Golaszewski (2):
drivers: move the early platform device support to arch/sh
sh: add the sh_ prefix to early platform symbols

I like this, any objection from anyone if I take this in my driver-core tree for 5.5-rc1?

thanks,

greg k-h

maintainer
#4 lore raw
Quoted 33 lines · Greg Kroah-Hartman
Quoted 32 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Some time ago I started a discussion about the need for a proper early device probing mechanism[1]. One that would be based on real platform drivers and support both platform data and device tree.

While we're far from reaching any consensus on the implementation, Arnd suggested that I start off by moving the SuperH-specific early platform drivers implementation to arch/sh[2].

This series is the first attempt at making way for a new, less hacky implementation.

The first patch moves all the early_platform code to arch/sh.

The second patch prefixes all early_platform symbols with 'sh_'.

[1] https://lkml.org/lkml/2018/4/26/657
[2] https://lkml.org/lkml/2018/4/27/239

v1 -> v2:
- certain drivers are compiled for arm/mach-shmobile too - we need to add ifdefs for CONFIG_SUPERH around early_platform calls

v2 -> v3:
- added a stub for is_early_platform_device() which always returns false on non-SuperH architectures

v3 -> v4:
- rebased on top of v5.4-rc1
- removed patches that are already upstream from the series

Quoted 6 lines
Quoted 3 lines

Bartosz Golaszewski (2):
drivers: move the early platform device support to arch/sh
sh: add the sh_ prefix to early platform symbols

I like this, any objection from anyone if I take this in my driver-core tree for 5.5-rc1?

I don't think I have any objection. It will probably make gratuitous merge conflicts with Sato-san's old device tree sh4 work when we get back to finishing that, but that's not really a big deal.

Rich

#5 lore raw
Quoted 38 lines · Rich Felker
Quoted 37 lines · Greg Kroah-Hartman
Quoted 35 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Some time ago I started a discussion about the need for a proper early device probing mechanism[1]. One that would be based on real platform drivers and support both platform data and device tree.

While we're far from reaching any consensus on the implementation, Arnd suggested that I start off by moving the SuperH-specific early platform drivers implementation to arch/sh[2].

This series is the first attempt at making way for a new, less hacky implementation.

The first patch moves all the early_platform code to arch/sh.

The second patch prefixes all early_platform symbols with 'sh_'.

[1] https://lkml.org/lkml/2018/4/26/657
[2] https://lkml.org/lkml/2018/4/27/239

v1 -> v2:
- certain drivers are compiled for arm/mach-shmobile too - we need to add ifdefs for CONFIG_SUPERH around early_platform calls

v2 -> v3:
- added a stub for is_early_platform_device() which always returns false on non-SuperH architectures

v3 -> v4:
- rebased on top of v5.4-rc1
- removed patches that are already upstream from the series

Bartosz Golaszewski (2):
drivers: move the early platform device support to arch/sh
sh: add the sh_ prefix to early platform symbols

Quoted 6 lines
Quoted 2 lines

I like this, any objection from anyone if I take this in my driver-core tree for 5.5-rc1?

I don't think I have any objection. It will probably make gratuitous merge conflicts with Sato-san's old device tree sh4 work when we get back to finishing that, but that's not really a big deal.

Ok, I've queued it up in my tree now, thanks,

greg k-h

A message that is not in the archive (sent off-list or not yet received): 20191003092912.G3pupKBmyct1r9ScP5Skuw9D-_ALcMVSnfMfHAlwe0Y@z

#7 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

Quoted 7 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
#8 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisaÅ‚(a):

Quoted 54 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

Quoted 1 line
# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

Bart

#9 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

Quoted 57 lines · Bartosz Golaszewski

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisaÅ‚(a):

Quoted 54 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

Quoted 1 line
# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
Quoted 4 lines

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

I don't have any useful logs. The boot fails too early for that.

Sorry, I don't understand "everywhere". It fails in mainline and -next, if that is what you mean. And, yes, it has failed in -next for a while, looking back at the log history.

Guenter

#10 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

pon., 2 gru 2019 o 14:03 Guenter Roeck <[email protected]> napisał(a):

Quoted 59 lines · Bartosz Golaszewski
Quoted 57 lines

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisał(a):

Quoted 55 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
Quoted 7 lines
Quoted 4 lines

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

I don't have any useful logs. The boot fails too early for that.

Yeah, it touched "early" devices after all...

Quoted 1 line

Sorry, I don't understand "everywhere". It fails in mainline and -next, if that is what you mean.

I referred to "All my qemu boot tests" when saying everywhere.

I don't have any relevant HW - the idea for this patch is to stop compiling for everyone a bunch of code that's only used by one largely irrelevant architecture and free the namespace for a generic early platform drivers implementation.

How are you creating your SH qemu images? I'm seeing there's something in buildroot for superh - maybe I'll be able to build something useful.

Bartosz

Quoted 3 lines

And, yes, it has failed in -next for a while, looking back at the log history.

Guenter

#11 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

Quoted 65 lines · Bartosz Golaszewski

pon., 2 gru 2019 o 14:03 Guenter Roeck <[email protected]> napisał(a):

Quoted 64 lines · Bartosz Golaszewski
Quoted 61 lines

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisał(a):

Quoted 54 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

Quoted 1 line
# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

Quoted 9 lines
Quoted 2 lines

I don't have any useful logs. The boot fails too early for that.

Yeah, it touched "early" devices after all...

Quoted 1 line

Sorry, I don't understand "everywhere". It fails in mainline and -next, if that is what you mean.

I referred to "All my qemu boot tests" when saying everywhere.

Yes, it affects all my sh and sheb qemu boot tests. See https://kerneltests.org/builders/.

Quoted 9 lines

I don't have any relevant HW - the idea for this patch is to stop compiling for everyone a bunch of code that's only used by one largely irrelevant architecture and free the namespace for a generic early platform drivers implementation.

How are you creating your SH qemu images? I'm seeing there's something in buildroot for superh - maybe I'll be able to build something useful.

I use buildroot. Various root file systems and scripts are at
https://github.com/groeck/linux-build-test/tree/master/rootfs/{sh,sheb}

Guenter

#12 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

In reply to Bartosz Golaszewski
Quoted 74 lines · Bartosz Golaszewski

pon., 2 gru 2019 o 14:03 Guenter Roeck <[email protected]> napisał(a):

Quoted 59 lines · Bartosz Golaszewski
Quoted 57 lines

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisał(a):

Quoted 55 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
Quoted 7 lines
Quoted 4 lines

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

I don't have any useful logs. The boot fails too early for that.

Yeah, it touched "early" devices after all...

Quoted 1 line

Sorry, I don't understand "everywhere". It fails in mainline and -next, if that is what you mean.

I referred to "All my qemu boot tests" when saying everywhere.

Quoted 9 lines

I don't have any relevant HW - the idea for this patch is to stop compiling for everyone a bunch of code that's only used by one largely irrelevant architecture and free the namespace for a generic early platform drivers implementation.

How are you creating your SH qemu images? I'm seeing there's something in buildroot for superh - maybe I'll be able to bearly_platform_cleanupuild something useful.

Below is a possible fix. As I had suspected, the call to early_platform_cleanup() is in the wrong place.

Guenter

---
  arch/sh/drivers/platform_early.c | 11 ++---------
  drivers/base/platform.c          |  4 ++++
  2 files changed, 6 insertions(+), 9 deletions(-)
arch/sh/drivers/platform_early.c +2 -9
index f6d148451dfc..16f33bffd8fc 100644
@@ -325,9 +325,9 @@ int __init sh_early_platform_driver_probe(char *class_str,
  }
 
  /**
- * sh_early_platform_cleanup - clean up early platform code
+ * early_platform_cleanup - clean up early platform code
   */
-static int __init sh_early_platform_cleanup(void)
+void early_platform_cleanup(void)
  {
  	struct platform_device *pd, *pd2;
 
@@ -337,11 +337,4 @@ static int __init sh_early_platform_cleanup(void)
  		list_del(&pd->dev.devres_head);
  		memset(&pd->dev.devres_head, 0, sizeof(pd->dev.devres_head));
  	}
-
-	return 0;
  }
-/*
- * This must happen once after all early devices are probed but before probing
- * real platform devices.
- */
-subsys_initcall(sh_early_platform_cleanup);
drivers/base/platform.c +4 -0
index 7c532548b0a6..3ba153e356ee 100644
@@ -1325,10 +1325,14 @@ struct device *platform_find_device_by_driver(struct device *start,
  }
  EXPORT_SYMBOL_GPL(platform_find_device_by_driver);
 
+void __weak early_platform_cleanup(void) { }
+
  int __init platform_bus_init(void)
  {
  	int error;
 
+	early_platform_cleanup();
+
  	error = device_register(&platform_bus);
  	if (error) {
  		put_device(&platform_bus);
#13 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

wt., 3 gru 2019 o 06:24 Guenter Roeck <[email protected]> napisał(a):

Quoted 81 lines · Bartosz Golaszewski
Quoted 79 lines

pon., 2 gru 2019 o 14:03 Guenter Roeck <[email protected]> napisał(a):

Quoted 59 lines · Bartosz Golaszewski
Quoted 57 lines

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisał(a):

Quoted 55 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
Quoted 7 lines
Quoted 4 lines

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

I don't have any useful logs. The boot fails too early for that.

Yeah, it touched "early" devices after all...

Quoted 1 line

Sorry, I don't understand "everywhere". It fails in mainline and -next, if that is what you mean.

I referred to "All my qemu boot tests" when saying everywhere.

I don't have any relevant HW - the idea for this patch is to stop compiling for everyone a bunch of code that's only used by one largely irrelevant architecture and free the namespace for a generic early platform drivers implementation.

Quoted 8 lines
Quoted 4 lines

How are you creating your SH qemu images? I'm seeing there's something in buildroot for superh - maybe I'll be able to bearly_platform_cleanupuild something useful.

Below is a possible fix. As I had suspected, the call to early_platform_cleanup() is in the wrong place.

This makes sense. If it fixes the issue - do you want to send it to Greg?

Bart

Quoted 55 lines

Guenter

---
  arch/sh/drivers/platform_early.c | 11 ++---------
  drivers/base/platform.c          |  4 ++++
  2 files changed, 6 insertions(+), 9 deletions(-)
arch/sh/drivers/platform_early.c +2 -9
index f6d148451dfc..16f33bffd8fc 100644
@@ -325,9 +325,9 @@ int __init sh_early_platform_driver_probe(char *class_str,
  }
 
  /**
- * sh_early_platform_cleanup - clean up early platform code
+ * early_platform_cleanup - clean up early platform code
   */
-static int __init sh_early_platform_cleanup(void)
+void early_platform_cleanup(void)
  {
        struct platform_device *pd, *pd2;
 
@@ -337,11 +337,4 @@ static int __init sh_early_platform_cleanup(void)
                list_del(&pd->dev.devres_head);
                memset(&pd->dev.devres_head, 0, sizeof(pd->dev.devres_head));
        }
-
-       return 0;
  }
-/*
- * This must happen once after all early devices are probed but before probing
- * real platform devices.
- */
-subsys_initcall(sh_early_platform_cleanup);
drivers/base/platform.c +4 -0
index 7c532548b0a6..3ba153e356ee 100644
@@ -1325,10 +1325,14 @@ struct device *platform_find_device_by_driver(struct device *start,
  }
  EXPORT_SYMBOL_GPL(platform_find_device_by_driver);
 
+void __weak early_platform_cleanup(void) { }
+
  int __init platform_bus_init(void)
  {
        int error;
 
+       early_platform_cleanup();
+
        error = device_register(&platform_bus);
        if (error) {
                put_device(&platform_bus);
#14 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

Hi Bartosz, Günter,

Quoted 86 lines · Bartosz Golaszewski

wt., 3 gru 2019 o 06:24 Guenter Roeck <[email protected]> napisał(a):

Quoted 85 lines · Bartosz Golaszewski
Quoted 83 lines

pon., 2 gru 2019 o 14:03 Guenter Roeck <[email protected]> napisał(a):

Quoted 59 lines · Bartosz Golaszewski
Quoted 57 lines

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisał(a):

Quoted 55 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
Quoted 7 lines
Quoted 4 lines

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

I don't have any useful logs. The boot fails too early for that.

Yeah, it touched "early" devices after all...

Quoted 1 line

Sorry, I don't understand "everywhere". It fails in mainline and -next, if that is what you mean.

I referred to "All my qemu boot tests" when saying everywhere.

I don't have any relevant HW - the idea for this patch is to stop compiling for everyone a bunch of code that's only used by one largely irrelevant architecture and free the namespace for a generic early platform drivers implementation.

How are you creating your SH qemu images? I'm seeing there's something in buildroot for superh - maybe I'll be able to bearly_platform_cleanupuild something useful.

Quoted 5 lines
Quoted 3 lines

Below is a possible fix. As I had suspected, the call to early_platform_cleanup() is in the wrong place.

This makes sense. If it fixes the issue - do you want to send it to Greg?

Which tree does this patch apply to?
Oh, it's whitespace-damaged.

After applying it manually on top of current Linus' tree, (remote) SH7722-based Migo-R boots again, so

There's still an issue with crashing userspace, but that happens on v5.4 too. Bisecting...

Gr{oetje,eeting}s,

Geert

Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
#15 lore raw

Re: [PATCH v4 1/2] drivers: move the early platform device support to arch/sh

Quoted 89 lines · Geert Uytterhoeven

Hi Bartosz, Günter,

Quoted 86 lines · Bartosz Golaszewski

wt., 3 gru 2019 o 06:24 Guenter Roeck <[email protected]> napisał(a):

Quoted 85 lines · Bartosz Golaszewski
Quoted 83 lines

pon., 2 gru 2019 o 14:03 Guenter Roeck <[email protected]> napisał(a):

Quoted 59 lines · Bartosz Golaszewski
Quoted 57 lines

pon., 2 gru 2019 o 05:14 Guenter Roeck <[email protected]> napisał(a):

Quoted 55 lines · Bartosz Golaszewski
Quoted 7 lines

From: Bartosz Golaszewski <[email protected]>

SuperH is the only user of the current implementation of early platform device support. We want to introduce a more robust approach to early probing. As the first step - move all the current early platform code to arch/sh.

Quoted 8 lines

In order not to export internal drivers/base functions to arch code for this temporary solution - copy the two needed routines for driver matching from drivers/base/platform.c to arch/sh/drivers/platform_early.c.

Also: call early_platform_cleanup() from subsys_initcall() so that it's called after all early devices are probed.

Wondering ... has anyone tested this patch on affected hardware ? All my qemu boot tests (both sh and sheb) fail because of it. Bisect log below.

Guenter

---
# bad: [72c0870e3a05d9cd5466d08c3d2a3069ed0a2f9f] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
# good: [89d57dddd7d319ded00415790a0bb3c954b7e386] Merge tag 'media/v5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect start 'HEAD' '89d57dddd7d3'
# good: [0a6cad5df541108cfd3fbd79eef48eb824c89bdc] Merge branch 'vmwgfx-coherent' of git://people.freedesktop.org/~thomash/linux into drm-next git bisect good 0a6cad5df541108cfd3fbd79eef48eb824c89bdc
# bad: [9a3d7fd275be4559277667228902824165153c80] Merge tag 'driver-core-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
git bisect bad 9a3d7fd275be4559277667228902824165153c80
# good: [59274c7164807d27b24e6c068dfe734f7bea4623] Merge tag 'usb-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
git bisect good 59274c7164807d27b24e6c068dfe734f7bea4623
# good: [e71903106721dc53923e90aa484d78bc86c039a9] staging: mt7621-dma: align to match open parenthesis
git bisect good e71903106721dc53923e90aa484d78bc86c039a9
# good: [8f56e4ebe05c26c30e167519273843476e39e244] Merge tag 'char-misc-5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
git bisect good 8f56e4ebe05c26c30e167519273843476e39e244
# good: [8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50] Merge tag 'iio-for-5.5c' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next git bisect good 8bde9f3d2a217d1635a7c7bdf8ad4c25c9a34b50
# bad: [0628cda318df6baec439ca6e6e274007492f1ccd] mac80211: Use debugfs_create_xul() helper
git bisect bad 0628cda318df6baec439ca6e6e274007492f1ccd
# bad: [03324507e66c7664c754b1ef92c5c3be24c78aa2] driver core: Allow fwnode_operations.add_links to differentiate errors
git bisect bad 03324507e66c7664c754b1ef92c5c3be24c78aa2
# bad: [313f5dbba41d905d59c820bb2d91ee6c661aff99] debugfs: remove return value of debugfs_create_u16()
git bisect bad 313f5dbba41d905d59c820bb2d91ee6c661aff99
# good: [d4387cd117414ba80230f27a514be5ca4a09cfcc] of: property: Create device links for all child-supplier depencencies
git bisect good d4387cd117414ba80230f27a514be5ca4a09cfcc
# bad: [c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1] base: soc: Handle custom soc information sysfs entries
git bisect bad c31e73121f4c1ec45a3e523ac6ce3ce6dafdcec1
# bad: [201e91091b1d47047f55580b5474e1239f4d17aa] sh: add the sh_ prefix to early platform symbols
git bisect bad 201e91091b1d47047f55580b5474e1239f4d17aa
# bad: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
git bisect bad 507fd01d5333338753a1cc26322dfc9f856c109f

# first bad commit: [507fd01d5333338753a1cc26322dfc9f856c109f] drivers: move the early platform device support to arch/sh
Quoted 7 lines
Quoted 4 lines

Hi Guenter,

can you post some bootlogs? Is it the same problem everywhere?

I don't have any useful logs. The boot fails too early for that.

Yeah, it touched "early" devices after all...

Quoted 1 line

Sorry, I don't understand "everywhere". It fails in mainline and -next, if that is what you mean.

I referred to "All my qemu boot tests" when saying everywhere.

I don't have any relevant HW - the idea for this patch is to stop compiling for everyone a bunch of code that's only used by one largely irrelevant architecture and free the namespace for a generic early platform drivers implementation.

How are you creating your SH qemu images? I'm seeing there's something in buildroot for superh - maybe I'll be able to bearly_platform_cleanupuild something useful.

Quoted 9 lines
Quoted 5 lines
Quoted 3 lines

Below is a possible fix. As I had suspected, the call to early_platform_cleanup() is in the wrong place.

This makes sense. If it fixes the issue - do you want to send it to Greg?

Which tree does this patch apply to?
Oh, it's whitespace-damaged.

Sorry, that wasn't meant to be applicable, just a hack to confirm my suspicion. I'll clean it up and submit it.

Thanks,
Guenter

Quoted 11 lines

After applying it manually on top of current Linus' tree, (remote) SH7722-based Migo-R boots again, so

There's still an issue with crashing userspace, but that happens on v5.4 too. Bisecting...

Gr{oetje,eeting}s,

Geert