Date: Sat, 4 Oct 2008 17:49:23 -0500 (CDT) From: "Sean C. Farley" <scf@FreeBSD.org> To: Chagin Dmitry <dchagin@FreeBSD.org> Cc: freebsd-emulation@FreeBSD.org, Alexander Logvinov <ports@logvinov.com> Subject: Re: SEGV with games/linux-ut Message-ID: <alpine.BSF.2.00.0810041741590.4282@thor.farley.org> In-Reply-To: <20081004211850.GA44165@dchagin.dialup.corbina.ru> References: <alpine.BSF.2.00.0810041148140.4282@thor.farley.org> <20081004211850.GA44165@dchagin.dialup.corbina.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 5 Oct 2008, Chagin Dmitry wrote:
> On Sat, Oct 04, 2008 at 11:55:45AM -0500, Sean C. Farley wrote:
>> Alexander and I are getting SEGV's when trying to run games/linux-ut.
>> He also sees problems with Serious Sam. It does not occur for him
>> with an FC4 base.
>>
>> Criteria for SEGV:
>> 1. linux_base-f8-8_{4,6}
>> 2. compat.linux.osrelease: 2.6.16
>> 3. RELENG_7 (r183556) with or without patches for Flash 9 support
>> 4. nvidia-driver-173.14.12
>>
>> ktrace:
>>
>> 7055 ut-bin CALL geteuid
>> 7055 ut-bin RET geteuid 1001/0x3e9
>> 7055 ut-bin CALL geteuid
>> 7055 ut-bin RET geteuid 1001/0x3e9
>> 7055 ut-bin CALL linux_open(0xbfbfc628,0x8002,0)
>> 7055 ut-bin NAMI "/compat/linux/dev/nvidiactl"
>> 7055 ut-bin NAMI "/dev/nvidiactl"
>> 7055 ut-bin RET linux_open 50/0x32
>> 7055 ut-bin CALL linux_ioctl(0x32,0xc04846d2 ,0xbfbfc588)
>> 7055 ut-bin RET linux_ioctl 0
>> 7055 ut-bin CALL linux_ioctl(0x32,0xc0e046c8 ,0x2bab5700)
>> 7055 ut-bin RET linux_ioctl 0
>> 7055 ut-bin CALL linux_ioctl(0x32,0xc00c4622 ,0xbfbfc6d8)
>> 7055 ut-bin RET linux_ioctl 0
>> 7055 ut-bin CALL geteuid
>> 7055 ut-bin RET geteuid 1001/0x3e9
>> 7055 ut-bin CALL linux_open(0xbfbfc528,0x8002,0)
>> 7055 ut-bin NAMI "/compat/linux/dev/nvidia0"
>> 7055 ut-bin NAMI "/dev/nvidia0"
>> 7055 ut-bin RET linux_open 51/0x33
>> 7055 ut-bin CALL linux_ioctl(0x32,0xc040464d ,0xbfbfc338)
>> 7055 ut-bin RET linux_ioctl 0
>
> 0xc040464d = _IOWR('F', 77, xxxxx), aka FrameBuffer which never
> supported in linuxulator (as far as I understand)
Maybe new functionality in linuxulator encouraged the code to try
farther?
Here is the entire trace minus the I/O data:
http://people.freebsd.org/~scf/ut-ktrace.log.bz2
> please, show /compat/linux/bin/sh /usr/bin/ldd /path_to/ut-bin
Is this what you want?
# ll /compat/linux/bin/{ba,}sh /usr/bin/ldd /compat/linux/usr/bin/ldd /usr/local/bin/ut /usr/local/share/linux-ut/System/ut-bin
-rwxr-xr-x 1 root wheel 729352 Oct 3 17:09 /compat/linux/bin/bash*
lrwxrwxrwx 1 root wheel 4 Oct 3 17:10 /compat/linux/bin/sh@ -> bash
-rwxr-xr-x 1 root wheel 5770 Oct 3 17:09 /compat/linux/usr/bin/ldd*
-r-xr-xr-x 1 root wheel 16596 Oct 2 18:56 /usr/bin/ldd*
-r-xr-xr-x 1 root wheel 1727 Sep 15 19:23 /usr/local/bin/ut*
-r-xr-xr-x 1 root wheel 91336 Sep 15 19:23 /usr/local/share/linux-ut/System/ut-bin*
>> 7055 ut-bin PSIG SIGSEGV caught handler=0x283a4d38 mask=0x0 code=0x0
>> 7055 ut-bin PSIG SIGSEGV SIG_DFL
>> 7055 ut-bin NAMI "ut-bin.core"
Sean
--
scf@FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.0810041741590.4282>