Date: Sat, 04 Feb 2006 23:06:41 +0900 From: NAKAJI Hiroyuki <nakaji@heimat.jp> To: freebsd-x11@freebsd.org Subject: Re: "Running X twice" problem: system hangup Message-ID: <864q3fkyku.fsf@ra333.heimat.gr.jp> In-Reply-To: <864q4rt0dt.fsf_-_@ra333.heimat.gr.jp> (NAKAJI Hiroyuki's message of "Fri, 30 Dec 2005 18:11:42 %2B0900") References: <86psnjavwy.fsf@wagner.bjencks.net> <200512300423.25326.dejan.lesjak@ijs.si> <86fyob42p0.fsf@ra333.heimat.gr.jp> <1135919112.933.27.camel@leguin> <864q4rt0dt.fsf_-_@ra333.heimat.gr.jp>
next in thread | previous in thread | raw e-mail | index | archive | help
I managed to get some logs of Xorg, but no kernel crash dump. With "-logverbose 10" option, I checked 4 cases. 1. http://heimat.jp/~nakaji/FreeBSD/Millenium/Xorg.0.log.ok Run X for the first time. "startx -- -logverbose 10 -logfile ~/public_html/FreeBSD/Millenium/Xorg.0.log.ok" 2. http://heimat.jp/~nakaji/FreeBSD/Millenium/Xorg.0.log.ng Run X after first X session end. "startx -- -logverbose 10 -logfile ~/public_html/FreeBSD/Millenium/Xorg.0.log.ng" 3. http://heimat.jp/~nakaji/FreeBSD/Millenium/PCIDEBUG=2/Xorg.0.log.ok Same as case 1, compiled with PCIDEBUG=2 in Pci.h "startx -- -logverbose 10 -logfile ~/public_html/FreeBSD/Millenium/PCIDEBUG=2/Xorg.0.log.ok" 4. http://heimat.jp/~nakaji/FreeBSD/Millenium/PCIDEBUG=2/Xorg.0.log.ng Same as case 2, compiled with PCIDEBUG=2 in Pci.h "startx -- -logverbose 10 -logfile ~/public_html/FreeBSD/Millenium/PCIDEBUG=2/Xorg.0.log.ng" In all cases, Xorg is installed using ports/x11-servers/xorg-server. In the case 3 and 4, Pci.h is modified manually after "make patch". === Pci.h change === Index: Pci.h =================================================================== RCS file: /cvs/xorg/xc/programs/Xserver/hw/xfree86/os-support/bus/Pci.h,v retrieving revision 1.10 diff -u -u -r1.10 Pci.h --- Pci.h 8 Nov 2005 19:04:56 -0000 1.10 +++ Pci.h 4 Feb 2006 13:53:09 -0000 @@ -204,7 +204,7 @@ /* * Debug Macros/Definitions */ -/* #define DEBUGPCI 2 */ /* Disable/enable trace in PCI code */ +#define DEBUGPCI 2 /* Disable/enable trace in PCI code */ #if defined(DEBUGPCI) === Pci.h change === The diff between case 1 and 2 shows some pci register values are changed after the first X execution. @@ -70,7 +70,7 @@ (II) PCI: Probing config type using method 1 (II) PCI: Config type is 1 -(II) PCI: stages = 0x03, oldVal1 = 0x8000005c, mode1Res1 = 0x80000000 +(II) PCI: stages = 0x03, oldVal1 = 0x00000000, mode1Res1 = 0x80000000 (II) PCI: PCI scan (all values are in hex) (II) PCI: 00:00:0: chip 8086,1237 card 0000,0000 rev 02 class 06,00,00 hdr 00 (II) PCI: 00:06:0: chip 1033,002c card 0000,0000 rev 01 class 06,80,00 hdr 00 @@ -336,9 +336,8 @@ (--) MGA(0): Linear framebuffer at 0x24000000 (--) MGA(0): MMIO registers at 0x20414000 (II) xf86ReadPciBios: modifying membase[1] for device 0:13:0 -(II) xf86ReadPciBios: modifying membase[1] for device 0:13:0 -(II) xf86ReadPciBios: modifying membase[1] for device 0:13:0 -(WW) MGA(0): Video BIOS info block not detected! +(II) Truncating PCI BIOS Length to 32768 +(--) MGA(0): Video BIOS info block at offset 0x07D60 (==) MGA(0): Write-combining range (0x20414000,0x4000) was already clear (==) MGA(0): Write-combining range (0x24000000,0x800000) (**) MGA(0): VideoRAM: 4096 kByte I saved the output of pcitweak of case 3 and 4 http://heimat.jp/~nakaji/FreeBSD/Millenium/PCIDEBUG=2/pcitweak.out.ok and http://heimat.jp/~nakaji/FreeBSD/Millenium/PCIDEBUG=2/pcitweak.out.after respectively. -- NAKAJI Hiroyuki
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?864q3fkyku.fsf>