[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20090126141646.GA30919@elte.hu>
Date: Mon, 26 Jan 2009 15:16:46 +0100
From: Ingo Molnar <mingo@...e.hu>
To: Yinghai Lu <yinghai@...nel.org>
Cc: Thomas Gleixner <tglx@...utronix.de>,
"H. Peter Anvin" <hpa@...or.com>,
Andrew Morton <akpm@...ux-foundation.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] x86: make irqinit_32.c more like irqoinit_64.c
* Ingo Molnar <mingo@...e.hu> wrote:
> > Impact: cleanup
> >
> > 1. add smp_intr_init and apic_intr_init for 32bit the same as 64bit
> > 2. move apic_intr_init() calling before set gate with interrupt[i]
> > 3. for 64bit, if ia32_emulation is not used, will make per_cpu to use 0x80 vector.
this change caused crashes/hangs on a 32-bit testsystem:
Checking if this processor honours the WP bit even in supervisor mode...Ok.
SLUB: Genslabs=12, HWalign=64, Order=0-3, MinObjects=0, CPUs=2, Nodes=1
Unknown interrupt or fault at EIP 00000286 00000060 c0117e1c
Calibrating delay loop (skipped), value calculated using timer frequency.. 4020.82 BogoMIPS (lpj=8041640)
Security Framework initialized
Mount-cache hash table entries: 512
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 512K (64 bytes/line)
CPU: Physical Processor ID: 0
CPU: Processor Core ID: 0
Checking 'hlt' instruction... [hard hang]
See the 'Unknown interrupt' entry (it is new) - which happens right after
irqs are enabled. Reverting your patch makes it boot fine. Config
attached.
Ingo
View attachment "config" of type "text/plain" (69220 bytes)
Powered by blists - more mailing lists