Messages in this thread | | | Date | Mon, 3 Aug 1998 20:12:27 -0400 (EDT) | From | "Richard B. Johnson" <> | Subject | Re: Linux version 2.1.113 |
| |
On 3 Aug 1998, James H. Cloos Jr. wrote:
> Looking further in my archives, it was in 2.1.109 that WCHAN died. > > The change in fs/proc/array.c in 109 is irrelevant to wchan on i386. > The changes to arch/i386/kernel/process.c and include/asm-i386/ > processor.h, however, seem more significant. > > -JimC > -- > James H. Cloos, Jr. > <cloos@jhcloos.com> >
Yes. There are comments in the sources about adding a new wchan member to the tss structure. This was done for the power-pc. At the time a context-switch occurs, one could get the return address off the stack and write it to the wchan member. The problem is that this would occur at every context-switch which would add some overhead. The time to add overhead, in my opinion, is the time at which one actually needs this information.
Adding a new member would make this function much cleaner, but I don't think others would appreciate the increased overhead. So, I will wait and see. Presently, a simple hack results in the address of the scheduler, which is not what we want.
Cheers, Dick Johnson ***** FILE SYSTEM MODIFIED ***** Penguin : Linux version 2.1.113 on an i586 machine (66.15 BogoMips). Warning : It's hard to remain at the trailing edge of technology.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
| |