Date: Thu, 15 Nov 2001 16:54:27 -0800 (PST) From: John Baldwin <jhb@FreeBSD.org> To: Daniel Eischen <eischen@pcnet1.pcnet.com> Cc: Maxim Sobolev <sobomax@FreeBSD.org>, marcus@marcuscom.com, freebsd-ports@FreeBSD.org, hackers@FreeBSD.org Subject: Re: Using bit 21 of EFLAGS in user-mode [was: Re: sigreturn: efl Message-ID: <XFMail.011115165427.jhb@FreeBSD.org> In-Reply-To: <Pine.SUN.3.91.1011115185612.23017A-100000@pcnet1.pcnet.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 16-Nov-01 Daniel Eischen wrote:
> On Thu, 15 Nov 2001, John Baldwin wrote:
>> On 15-Nov-01 Daniel Eischen wrote:
>> > On Thu, 15 Nov 2001, Maxim Sobolev wrote:
>> >> On Thu, 15 Nov 2001 14:56:31 -0500 (EST), Joe Clarke wrote:
>> >> >
>> >> > I learned about this by reading through some of the -hackers archives.
>> >> > One person complained of similar errors trying to get xine to work on
>> >> > FreeBSD. Removing the MMX detection code fixed it. I remembered
>> >> > libpng
>> >> > also used MMX, so I removed the pnggccrd.c source, and voila!
>> >> >
>> >> > Based on core dumps, strace output, and a lot of code surfing, this
>> >> > makes
>> >> > sense to me. Basically, any png-dependent app's thread that runs
>> >> > longer
>> >> > than what ITIMER_PROF is set to gets hit with a SIGPROF. When that
>> >> > happens, things context switch. eflags must have been corrupted by the
>> >> > MMX code, thus sigreturn() bombs out, and causes uthread_kern to die as
>> >> > well. Here's what strace looks like when balsa tries to read a 33 MB
>> >> > mailbox:
>> >> >
>> >> > 74202 sigreturn(0x81f2c64
>> >> >
>> >> > When this happens, strace politely dies with a bus error.
>> >> >
>> >> > Thanks for testing this, Maxim. Hopefully someone can find the problem
>> >> > and fix it for good.
>> >>
>> >> That explains... After a quick glance at png code I found that
>> >> the only place where EFLAGS is altered is CPUID code, where
>> >> the library flips bit 21 of EFLAGS in order to ensure that the
>> >> CPUID instruction is supported (otherwise it will get SIGILL
>> >> on older processors). Unfortunately, for some reason FreeBSB
>> >
>> > Does it need to keep bit 21 of EFLAGS flipped, or can libpng
>> > set it back and keep knowledge that CPUID is supported? Or
>> > does that bit need to remain set for CPUID to work?
>>
>> It needs to be able to change it. If you can change the value of the bit
>> (done
>> by pushf ; pop %eax ; mov %eax,%ebx ; xor $PSL_ID,%eax ; push %eax ; popf ;
>> pushf ; pop %eax ; compare bit PSL_ID of eax ebx to see if they match).
>> The problem is if a signal comes in during the middle of that bit toggling
>> due
>> to a profiling timer. I think the problem may be that it uses a sequence
>> that
>> leaves the bit set, thus the kernel freaks out thinking that the user has
>> changed a kernel only flag. The solution is Maxim's patch to make the
>> kernel
>> not care about the flag (which it shouldn't since cpuid is not a privileged
>> instruction).
>
> I just thought perhaps libpng could do something like:
>
> static int init_done = 0;
> static int cpuid_supported = 0;
>
> ...
>
> if (init_done == 0) {
> block_all_sigs();
> cpuid_supported = check_cpuid();
> init_done = 1;
> unblock_sigs();
> }
>
> But if it always needs to change the bit, I guess the above doesn't help.
No, that's correct. The problem is the check_cpuid() function left the state
of the PSL_ID flag bit changed. :) So you need to change the check_cpuid() to
include a wrapper to save/restore the flags. Maxim just committed a change to
do this. I'm not sure if he included the signal blocking though.
--
John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve!" - http://www.FreeBSD.org/
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.011115165427.jhb>