[patch] exec() SMP-scalability bugfix, 2.3.49/pre-2.3.50

• Next message: Ingo Molnar: "Re: isn't __brelse() wrong?"
• Previous message: Alexander Viro: "Re: bug: mount on an open directory succeeds"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

i missed a case while SMP-threading exec(). The window is very small but
present, under very high exec() load this bug results in negative d_count
asserts in dput(). Attached patch is against pre-2.3.50-2.

(the patch also includes a fix to binfmt_elf.c, although i cannot see how
that #ifdef-ed code could possibly get compiled in. Plus the patch updates
em86 binformat to the new binary loader SMP locking semantics. Dynamic
loading of binary formats into the kernel is still SMP-unsafe.)

Tested ELF and script binformats under load, and they are ok.

-- mingo

• TEXT/PLAIN attachment: exec-scale-2.3.50-A1

-
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.tux.org/lkml/

• Next message: Ingo Molnar: "Re: isn't __brelse() wrong?"
• Previous message: Alexander Viro: "Re: bug: mount on an open directory succeeds"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:19 EST