drivers: uart: fix IRQ Config

[Shreyas-Shankar155]

Fix UART_NS16550_IRQ_FLAGS macro to recognize TI VIM interrupt-controller.

[natto1784]

Do not use CONFIG_DT_HAS* in driver code. It maybe so that the UART instance does not have VIM as its parent, and yet, just the presence of VIM in the DT will set this to 1.

[Shreyas-Shankar155]

i've addressed this comment, but my solution is rather brute-force.
i'm searching if 'flags' exists, then if 'sense' exists.

however, if in the future, what to do if some new name has to be supported?
it wouldn't look too good if we have a chain of macros in this driver.

[natto1784]

This PR intended to make it consistent: #80790. Maybe you can pick it up and rename all "sense" cells to "flags"

[Shreyas-Shankar155]

i don't think its a good idea to rename all "sense" cells to "flags".

what if there are some drivers make assumptions that their compatible interrupt-controller use "sense"?

[glneo]

The core issue is making the driver care about the details of the interrupt-controller in the first place. What about interrupt-controllers that don't have a "type/sense/flags" cell at all? The interrupt-controller should provide a method to parse out the information it needs in a way that can be passed back to that interrupt-controller as it expects it. Having every end consumer driver do the DT parsing for the IRQ controller is never going to be flexible enough for all IRQ controllers.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[natto1784]

Good idea, can definitely use a macro in the header file for interrupt controller subsystem

…

On Mon, 14 Jul, 2025, 22:09 Andrew Davis, ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In drivers/serial/uart_ns16550.c <#92940 (comment)> : > @@ -1808,9 +1808,15 @@ static DEVICE_API(uart, uart_ns16550_driver_api) = { #endif }; -#define UART_NS16550_IRQ_FLAGS(n) \ - COND_CODE_1(DT_INST_IRQ_HAS_CELL(n, sense), \ - (DT_INST_IRQ(n, sense)), \ +#ifdef CONFIG_DT_HAS_TI_VIM_ENABLED The core issue is making the driver care about the details of the interrupt-controller in the first place. What about interrupt-controllers that don't have a "type/sense/flags" cell at all? The interrupt-controller should provide a method to parse out the information it needs in a way that can be passed back to that interrupt-controller as it expects it. Having every end consumer driver do the DT parsing for the IRQ controller is never going to be flexible enough for all IRQ controllers. — Reply to this email directly, view it on GitHub <#92940 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANNVFPXY4VBQUH7CTBRCDQT3IPMMVAVCNFSM6AAAAACBF4BCYGVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZTAMJXGAZDKMJWGU> . You are receiving this because you commented.Message ID: ***@***.***>