Messages in this thread | | | Date | Sat, 26 Dec 2009 02:20:17 -0200 | From | Henrique de Moraes Holschuh <> | Subject | Re: [PATCH] thinkpad_acpi.c: Remove #define TPACPI_<level>s for printks |
| |
On Fri, 25 Dec 2009, Joe Perches wrote: > The use of printk(TPACPI_<level> is unusual. > Make it a bit more kernel normal style.
NAK.
This is likely purely a taste thing, but I completely fail to see how this:
> - printk(TPACPI_DEBUG "%s: access by process with PID %d\n", > - what, task_tgid_vnr(current)); > + printk(KERN_DEBUG pr_fmt("%s: access by process with PID %d\n"), > + what, task_tgid_vnr(current));
is a good thing at all. Maybe you could convince me to change my mind about it but I dislike the resuling uglyness, so I will need a good technical reason to even consider it.
Why would I want a parameter macro instead of a straightforward bunch of #defines ? It is not like the #defines are harder to understand than this pr_fmt thing.
> if (!*fmt) { > - printk(TPACPI_ERR "acpi_evalf() called with empty format\n"); > + pr_err("acpi_evalf() called with empty format\n"); > return 0;
I am with Peter Zijlstra on this one. I don't want the pr_foo() siliness in code I maintain. You could perhaps convince me to change what is inside the printk(), but the fact that my printk helpers end with "_printk" is not a coincidence.
-- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh
| |