Re: [PATCH v7 0/5] clocksource/timer-econet-en751221: Support irq number per timer
From: Caleb James DeLisle
Date: Wed Aug 12 2026 - 15:32:23 EST
On 12/08/2026 10:26, Daniel Lezcano wrote:
On 7/31/26 14:02, Caleb James DeLisle wrote:
This timer appears on both the EcoNet EN751221 (34Kc) and on the
EN751627 / EN7528 (1004Kc) SoCs. On the 34Kc the interrupt controller
uses a percpu interrupt while on the others it uses an individual
IRQ number for each CPU timer. Add support for EN751627 and EN7528.
Add support for IRQ number for each CPU timer under the
econet,en751627-timer compatible.
Changes since v6:
* Rebase to tip/timers/core (base-commit below)
Got this:
In file included from /home/dlezcano/Work/src/linux/timers/include/asm-generic/bug.h:31,
from /home/dlezcano/Work/src/linux/timers/arch/arm64/include/asm/bug.h:26,
from /home/dlezcano/Work/src/linux/timers/arch/arm64/include/asm/pgtable.h:8,
from /home/dlezcano/Work/src/linux/timers/include/linux/pgtable.h:6,
from /home/dlezcano/Work/src/linux/timers/arch/arm64/include/asm/io.h:12,
from /home/dlezcano/Work/src/linux/timers/include/linux/io.h:12,
from /home/dlezcano/Work/src/linux/timers/drivers/clocksource/timer-econet-en751221.c:8:
/home/dlezcano/Work/src/linux/timers/drivers/clocksource/timer-econet-en751221.c: In function ‘timer_init’:
/home/dlezcano/Work/src/linux/timers/include/linux/kern_levels.h:5:25: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘long unsigned int’ [-Wformat=]
5 | #define KERN_SOH "\001" /* ASCII Start Of Header */
| ^~~~~~
/home/dlezcano/Work/src/linux/timers/include/linux/printk.h:483:25: note: in definition of macro ‘printk_index_wrap’
483 | _p_func(_fmt, ##__VA_ARGS__); \
| ^~~~
/home/dlezcano/Work/src/linux/timers/include/linux/printk.h:554:9: note: in expansion of macro ‘printk’
554 | printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
| ^~~~~~
/home/dlezcano/Work/src/linux/timers/include/linux/kern_levels.h:11:25: note: in expansion of macro ‘KERN_SOH’
11 | #define KERN_ERR KERN_SOH "3" /* error conditions */
| ^~~~~~~~
/home/dlezcano/Work/src/linux/timers/include/linux/printk.h:554:16: note: in expansion of macro ‘KERN_ERR’
554 | printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
| ^~~~~~~~
/home/dlezcano/Work/src/linux/timers/drivers/clocksource/timer-econet-en751221.c:201:17: note: in expansion of macro ‘pr_err’
201 | pr_err("%pOFn: Too many IRQs max %d got %d\n", np,
| ^~~~~~
/home/dlezcano/Work/src/linux/timers/include/linux/kern_levels.h:5:25: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘long unsigned int’ [-Wformat=]
5 | #define KERN_SOH "\001" /* ASCII Start Of Header */
| ^~~~~~
/home/dlezcano/Work/src/linux/timers/include/linux/printk.h:483:25: note: in definition of macro ‘printk_index_wrap’
483 | _p_func(_fmt, ##__VA_ARGS__); \
| ^~~~
/home/dlezcano/Work/src/linux/timers/include/linux/printk.h:554:9: note: in expansion of macro ‘printk’
554 | printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
| ^~~~~~
/home/dlezcano/Work/src/linux/timers/include/linux/kern_levels.h:11:25: note: in expansion of macro ‘KERN_SOH’
11 | #define KERN_ERR KERN_SOH "3" /* error conditions */
| ^~~~~~~~
/home/dlezcano/Work/src/linux/timers/include/linux/printk.h:554:16: note: in expansion of macro ‘KERN_ERR’
554 | printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
| ^~~~~~~~
/home/dlezcano/Work/src/linux/timers/drivers/clocksource/timer-econet-en751221.c:206:17: note: in expansion of macro ‘pr_err’
206 | pr_err("%pOFn: Too many regs: max %d got %d\n", np,
| ^~~~~~
I gather you must have built this on a 64 bit machine, so the size_t of ARRAY_SIZE() is a uint64_t (?)
I just double-checked building it for MIPS 24Kc GCC and it all passes. This is a built-in CPU timer for EcoNet MIPS SoCs. Those were all 32 bit and they're not making any more of them so we don't need to worry about a future 64 bit port.
But I'm not that knowledgeable in best practices, perhaps its expected to compile on 64 bit? Or perhaps I should forbid 64 bit in the Kconfig? There is no way to get CONFIG_ECONET and 64 bit, but through TEST_COMPILE I suppose it is.
Any suggestions would be appreciated.
Thanks,
Caleb
Changes since v5:
* Integrate patch fix refcount leak from:
Link: https://lore.kernel.org/linux-mips/20260526184105.18962-3-grandmaster@xxxxxxxxxxxx/
* NULL econet_timer.membase after unmapping
* NULL check dev->event_handler in case clockevents_config_and_register
failed to fully register.
* v5: https://lore.kernel.org/linux-mips/20260516233943.49502-1-cjd@xxxxxxxx/
Changes from v4:
* Fixed incorrect compatible in DT schema
* Bracing in out_irq_free if statement
* I did not move cpuhp_setup_state past the point of no return
* v4: https://lore.kernel.org/linux-mips/20260516215720.4160831-1-cjd@xxxxxxxx
Changes from v3:
* off-by-one in out_irq_free
* Gracefully fail if too many address resources are provided in the DT
* Allow fewer actual timers than num_possible_cpus, as long as they
never start
* v3: https://lore.kernel.org/linux-mips/20260516182648.3987792-1-cjd@xxxxxxxx
v3 note:
I decided to take this patchset in a slightly different direction from
v2. There is a certain amount of initialization that once it has run,
it cannot be reverted in case of error. The original driver init was a
mixture of revertible and non-revertible code which made it too easy
for code which is designed with best practices in mind to in fact
introduce a bug. Therefore I start by making the init process more
legible but without any business logic changes, then fix a race
condition by keeping IRQ disabled until after cevt is setup, then
finally add non-percpu IRQ support.
* v2: https://lore.kernel.org/linux-mips/20260514000601.3430262-1-cjd@xxxxxxxx
Changes from v1:
* Split changes over 3 refactoring patches + main patch
* Remove driver discussion from dt commit message
* v1: https://lore.kernel.org/linux-mips/20260416175101.958073-1-cjd@xxxxxxxx/
Alexander A. Klimov (1):
clocksource/timer-econet-en751221: fix refcount leak
Caleb James DeLisle (4):
dt-bindings: timer: econet: Update EN751627 for multi-IRQ
clocksource/timer-econet-en751221: Init teardown on error if possible
clocksource/timer-econet-en751221: Disable IRQ until cevt registered
clocksource/timer-econet-en751221: Support EN751627 without percpu IRQ
.../bindings/timer/econet,en751221-timer.yaml | 31 +--
drivers/clocksource/timer-econet-en751221.c | 181 ++++++++++++++----
2 files changed, 159 insertions(+), 53 deletions(-)
base-commit: ecc330e3096173f433659aa64ab3674d0d48440e