On Sat, 8 Dec 2001, Norm Dresner wrote:
> Okay. If you want to be _that_ literal, yes, you're right. But I was > thinking along the functional lines of "anything that enters the Linux > kernel is a function call", and I think that most people who read the Well, I disagree. I think "system calls" has a very specific definition. You aregue that most people would understand what you are talking about when you take "system call" to mean "any linux kernel function that is not a system call" (the reason I say not a system call is because you can't really call true system calls from within the kernel safely anyway). > message understood what I was talking about. After all, I was taught, > long ago, that "the primary purpose of language is to communicate", > not to provide a living for lawyers. > > Norm > > > ----- Original Message ----- > From: Calin A. Culianu <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Cc: <[EMAIL PROTECTED]> > Sent: Saturday, December 08, 2001 1:33 PM > Subject: RE: [rtl] malloc, timers, and crashes. > > > > On Fri, 7 Dec 2001, Dresner, Norman A. wrote: > > > > > You CAN NOT use any normal Linux system calls from RT software. > > > > I don't mean to be picky--(erm.. ok I mean to be picky) I just > want to > > clarify some teminology: Something like kmalloc, which is a function > > internal to the linux kernel is something quite different from a > system > > call. A system call has a very specific meaning, and kernel-space > > functions do not fall under that category at all.. :) > > > > -Calin > > > > > > > > If you can't pre-allocate it, you'll have to create a pool. > > > > > > > > > Norm > > > > > > > > > -----Original Message----- > > > From: Ken Emmons, Jr. [mailto:[EMAIL PROTECTED]] > > > Sent: Thursday, December 06, 2001 3:54 PM > > > To: [EMAIL PROTECTED] > > > Subject: [rtl] malloc, timers, and crashes. > > > > > > > > > Hello, > > > > > > I am trying to create a doubly linked list in order to keep track > of timers > > > on my system. It is similar to the way they do it in the Linux > kernel, but > > > obviously much faster polling for RT purposes. Certain function > calls in my > > > API will allocate memory for a structure and then call a function > to isert > > > this structure intot he list. I tried using: > > > > > > my_struct_t my_struct; > > > my_struct = kmalloc( sizeof(my_struct_t), GFP_ATOMIC); > > > > > > within a periodic RT task and it appears to crash my machine. > > > > > > Is this supposed to work in RTLinux?? If not how do I get free > memory?? Do I > > > have to set aside a free block of ram and create my own memory > allocation > > > algorithms?? Does RTLinux have a preferred mechanism? > > > > > > Thanks, > > > > > > ~Ken > > > > > > -- [rtl] --- > > > To unsubscribe: > > > echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR > > > echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED] > > > -- > > > For more information on Real-Time Linux see: > > > http://www.rtlinux.org/ > > > -- [rtl] --- > > > To unsubscribe: > > > echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR > > > echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED] > > > -- > > > For more information on Real-Time Linux see: > > > http://www.rtlinux.org/ > > > > > > > -- [rtl] --- > > To unsubscribe: > > echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR > > echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED] > > -- > > For more information on Real-Time Linux see: > > http://www.rtlinux.org/ > > > > -- [rtl] --- > To unsubscribe: > echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR > echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED] > -- > For more information on Real-Time Linux see: > http://www.rtlinux.org/ > -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED] -- For more information on Real-Time Linux see: http://www.rtlinux.org/
