Testing 2.6.15-rc4
Hi,
Recently I have tested the new kernel on my Ultra 10. Here is a short
summary of my observations:
1) Parallel port (parport_pc) works. I noticed that FIFO/DMA was
disabled by default. With FIFO/DMA enabled I experienced some problems
with previous kernel versions (oopses or complete lockup).
# CONFIG_PARPORT_PC_FIFO is not set
2) snd-sun-cs4231 works
3) snd-ens1371 works
The machine is very stable. I'm running apache, exim4 and Arno's
iptables firewall.
I must mention that I recompiled iptables from source because of some
Invalid argument messages.
The only remaining problem is the unstability of the bttv driver under
SPARC (complete lockup after approx. 15-20 min). I tried just about
every possible settings but until now without success. The strange
thing is that the same card (Jetway bt878) works fine under Solaris
with the bt848x driver, so hardware problems are out of question. Here
are some error messages
Dec 9 18:14:55 solaris kernel: videobuf_dma_pci_map: pci_map_sg failed
Dec 9 18:14:55 solaris kernel: Badness in pci_map_sg at
arch/sparc64/kernel/pci_iommu.c:633
Dec 9 18:14:55 solaris kernel: Call Trace:
Dec 9 18:14:55 solaris kernel: [1020393c]
videobuf_iolock+0x9c/0x160 [video_buf]
Dec 9 18:14:55 solaris kernel: [1020c4b4]
bttv_prepare_buffer+0xf4/0x1c0 [bttv]
Dec 9 18:14:55 solaris kernel: [102113e0]
bttv_do_ioctl+0x1080/0x1e80 [bttv]
Dec 9 18:14:55 solaris kernel: [101da168]
video_usercopy+0xc8/0x2a0 [videodev]
Dec 9 18:14:55 solaris kernel: [0049645c] do_ioctl+0x5c/0x80
Dec 9 18:14:55 solaris kernel: [004964ec]
vfs_ioctl+0x6c/0x340
Dec 9 18:14:55 solaris kernel: [00496834] sys_ioctl+0x74/0xa0
Dec 9 18:14:55 solaris kernel: [004af3b8]
compat_sys_ioctl+0x158/0x3a0
Dec 9 18:14:55 solaris kernel: [00407194]
linux_sparc_syscall32+0x34/0x40
Dec 9 18:14:55 solaris kernel: [00047db0] 0x47db0
Dec 9 18:21:38 solaris kernel: bttv: Bt8xx card found (0).
Dec 9 18:21:38 solaris kernel: bttv0: Bt878 (rev 17) at :02:01.0,
irq: 7474944, latency: 128, mmio: 0x1ff2000
Dec 9 18:21:38 solaris kernel: bttv0: using: Jetway TV/Capture
JW-TV878-FBK, Kworld KW-TV878RF [card=78,insmod option]
Dec 9 18:21:38 solaris kernel: bttv0: using tuner=5
Dec 9 18:21:38 solaris kernel: bttv0: i2c: checking for TDA9875 @
0xb0... not found
Dec 9 18:21:38 solaris kernel: bttv0: i2c: checking for TDA7432 @
0x8a... not found
Dec 9 18:21:38 solaris kernel: bttv0: i2c: checking for TDA9887 @
0x86... not found
Dec 9 18:21:38 solaris kernel: tuner 0-0060: All bytes are equal. It
is not a TEA5767
Dec 9 18:21:38 solaris kernel: tuner 0-0060: chip found @ 0xc0 (bt878
#0 [sw])
Dec 9 18:21:38 solaris kernel: tuner 0-0060: type set to 5 (Philips
PAL_BG (FI1216 and compatibles))
Dec 9 18:21:38 solaris kernel: bttv: Overlay support disabled.
Dec 9 18:21:38 solaris kernel: bttv0: registered device video0
Dec 9 18:21:38 solaris kernel: bttv0: registered device vbi0
Dec 9 18:21:38 solaris kernel: bttv0: registered device radio0
Dec 9 18:21:38 solaris kernel: bttv0: PLL: 28636363 = 35468950 . ok
Dec 9 18:21:49 solaris kernel: ioctl32(xdtv:2122): Unknown cmd fd(4)
cmd(40685600){00} arg(effad4f4) on /dev/video0
Dec 9 18:21:49 solaris kernel: ioctl32(xdtv:2122): Unknown cmd fd(4)
cmd(40685600){00} arg(effad4f4) on /dev/video0
Dec 9 18:21:57 solaris kernel: ioctl32(xdtv:2122): Unknown cmd fd(4)
cmd(40685600){00} arg(effacf4c) on /dev/video0
Dec 9 18:28:11 solaris kernel: bttv0: timeout: drop=39
irq=29062/29062, risc=daa92024, bits: HSYNC OFLOW FDSR
Dec 9 18:28:11 solaris kernel: bttv0: reset, reinitialize
Dec 9 18:28:11 solaris kernel: bttv0: PLL: 28636363 = 35468950 . ok
Dec 9 18:28:12 solaris kernel: bttv0: timeout: drop=52
irq=29076/29076, risc=db6d0024, bits: VSYNC HSYNC OFLOW FDSR
Dec 9 18:28:12 solaris kernel: bttv0: reset, reinitialize
Dec 9 18:28:12 solaris kernel: bttv0: PLL: 28636363 = 35468950 . ok
Dec 9 18:28:56 solaris kernel: bttv0: timeout: drop=77
irq=30472/30472, risc=c3b74024, bits: VSYNC HSYNC OFLOW FDSR
Best regards,
Attila
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: problem with compile
Andrey Chernomyrdin wrote: Have a nice day, I have a problem with compile nginx web server on SunBlade-100 $ gcc -c -Wall -g -O2 -I src/core -I src/event -I src/event/modules -I src/os/unix -I objs \ -o objs/src/core/ngx_times.o \ src/core/ngx_times.c /home/andrey/tmp/ccjb1kos.s: Assembler messages: /home/andrey/tmp/ccjb1kos.s:377: Error: Architecture mismatch on casa. /home/andrey/tmp/ccjb1kos.s:377: (Requires v9|v9a|v9b; requested architecture is sparclite.) /home/andrey/tmp/ccjb1kos.s:585: Error: Architecture mismatch on membar. /home/andrey/tmp/ccjb1kos.s:585: (Requires v9|v9a|v9b; requested architecture is sparclite.) Hi, I'm not a programmer, but I think you can try this: gcc -c -Wall -g -O2 -I src/core -I src/event -I src/event/modules -I src/os/unix -I objs -S -o objs/src/core/ngx_times.s src/core/ngx_times.c as -Av9 -64 -o objs/src/core/ngx_times.o objs/src/core/ngx_times.s Regards, Attila -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Please test pre-release 2.6.14-3 packages
These are the last messsages from strace:
ioctl(3, 0x4004667f, 0xef81751c)= -1 EINTR (Interrupted system
call)
gettimeofday({1131785220, 929401}, NULL) = 4
gettimeofday({0, 0}, NULL) = 8
ioctl(3, 0x4004667f, 0xef81751c)= 3
ioctl(3, 0x4004667f, 0xef81751c)= -1 EINTR (Interrupted system
call)
gettimeofday({1131785220, 968939}, NULL) = 4
gettimeofday({0, 0}, NULL) = 8
ioctl(3, 0x4004667f, 0xef81751c)= 3
ioctl(3, 0x4004667f, 0xef81751c)= -1 EINTR (Interrupted system
call)
gettimeofday({1131785221, 9991}, NULL) = 4
gettimeofday({0, 0}, NULL) = 8
ioctl(3, 0x4004667f, 0xef81751c)= 3
ioctl(3, 0x4004667f, 0xef81751c)= -1 EINTR (Interrupted system
call)
gettimeofday({1131785221, 47414}, NULL) = 4
gettimeofday({0, 0}, NULL) = 8
ioctl(3, 0x4004667f, 0xef81751c)= 3
ioctl(3, 0x4004667f, 0xef81751c)= -1 EINTR (Interrupted system
call)
This time i found nothing interesting in /var/log/messages, only these:
Nov 12 10:28:48 solaris kernel: bttv0: timeout: drop=37
irq=72636/72636, risc=df7c801c, bits: HSYNC OFLOW
Nov 12 10:28:48 solaris kernel: bttv0: reset, reinitialize
Nov 12 10:28:48 solaris kernel: bttv0: PLL: 28636363 = 35468950 . ok
Nov 12 10:28:49 solaris kernel: bttv0: timeout: drop=50
irq=72652/72652, risc=c03b0024, bits: VSYNC HSYNC OFLOW FDSR
Nov 12 10:28:49 solaris kernel: bttv0: reset, reinitialize
Nov 12 10:28:49 solaris kernel: bttv0: PLL: 28636363 = 35468950 . ok
Best regards,
Attila
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Please test pre-release 2.6.14-3 packages
I tried various options for bttv (triton1, vsfx, gbuffers, no_overlay etc.) but without success. This is the error message before bad things happen: bttv0: OCERR @ c3a6e01c,bits: HSYNC OFLOW OCERR* spitfire_data_access_exception: SFSR[00801009] SFAR[f80010918f64], g oing. \|/ \|/ @'/ .. \`@ /_| \__/ |_\ \__U_/ sh(2359): Dax [#1] TSTATE: 009911009607 TPC: 004babd0 TNPC: 004babd4 Y: 000 0Not tainted TPC: proc_lookup+0x30/0xc0 g0: f80011522fe1 g1: 0006 g2: 0004 g3: f ff8 g4: f8001329e000 g5: 6f6e6600 g6: f8001152 g7: 0 06b o0: fff8 o1: f800104dcd61 o2: 0004 o3: fbf51 83a o4: f80011b41844 o5: f80011b417e8 sp: f80011523051 ret_pc: 0 04babf0 RPC: proc_lookup+0x50/0xc0 l0: fffbf80010918f60 l1: f80011b41868 l2: 00702c00 l3: 0 006 l4: f80015bf000a l5: l6: l7: 701d1 000 i0: f842ef48 i1: f80011b41790 i2: f80011523d68 i3: 0 000 i4: i5: i6: f80011523111 i7: 00491 6a0 I7: do_lookup+0x140/0x180 Caller[004916a0]: do_lookup+0x140/0x180 Caller[00491820]: __link_path_walk+0x140/0xe60 Caller[00492598]: link_path_walk+0x58/0x140 Caller[00493c90]: open_namei+0x70/0x660 Caller[00481e60]: filp_open+0x40/0xa0 Caller[00481f58]: do_sys_open+0x38/0xe0 Caller[00407194]: linux_sparc_syscall32+0x34/0x40 Caller[70136bcc]: 0x70136bcc Instruction DUMP: e05c2048 22c4001e b0103ffe c4142004 c206602c 80a08001 12 4a d4142004 952ab030 Best regards, Attila -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bttv and SPARC
[EMAIL PROTECTED] wrote: Hi all. I would like to know if bttv is supposed to work under sparc (Ultra 10). pci_iommu: alloc_streaming_cluster of npages(260) failed! I confirm that my TV tuner works with the 2.6.14 kernel. It seems that the problem was that the fast allocator in the sparc64 PCI IOMMU code had a hard limit of 256 pages. (according to the kernel changelog). I experienced a hard lockup and an oops; still needs some investigation. Nov 1 14:09:18 solaris kernel: spitfire_data_access_exception: SFSR[00801009] SFAR[0004], going. Nov 1 14:09:18 solaris kernel: \|/ \|/ Nov 1 14:09:18 solaris kernel: @'/ .. \`@ Nov 1 14:09:18 solaris kernel: /_| \__/ |_\ Nov 1 14:09:18 solaris kernel: \__U_/ Nov 1 14:09:18 solaris kernel: sh(2613): Dax [#1] Nov 1 14:09:18 solaris kernel: TSTATE: 009911009607 TPC: 004babd0 TNPC: 004babd4 Y: Not tainted Nov 1 14:09:18 solaris kernel: TPC: proc_lookup+0x30/0xc0 Nov 1 14:09:18 solaris kernel: g0: f80007a86fe1 g1: 0006 g2: g3: fff8 Nov 1 14:09:18 solaris kernel: g4: f8001319c800 g5: 6f6e6600 g6: f80007a84000 g7: 006b Nov 1 14:09:18 solaris kernel: o0: fff8 o1: f80016ff6d61 o2: o3: 0a433ef2 Nov 1 14:09:18 solaris kernel: o4: f8001226cefc o5: f8001226cea0 sp: f80007a87051 ret_pc: 004babf0 Nov 1 14:09:18 solaris kernel: RPC: proc_lookup+0x50/0xc0 Nov 1 14:09:18 solaris kernel: l0: c000 l1: f8001226d5e0 l2: 00702c00 l3: 0006 Nov 1 14:09:18 solaris kernel: l4: f80007e3900a l5: l6: l7: 701d1000 Nov 1 14:09:18 solaris kernel: i0: f80007e54328 i1: f8001226ce48 i2: f80007a87d68 i3: Nov 1 14:09:18 solaris kernel: i4: i5: i6: f80007a87111 i7: 004916a0 Nov 1 14:09:18 solaris kernel: I7: do_lookup+0x140/0x180 Nov 1 14:09:18 solaris kernel: Caller[004916a0]: do_lookup+0x140/0x180 Nov 1 14:09:18 solaris kernel: Caller[00491820]: __link_path_walk+0x140/0xe60 Nov 1 14:09:18 solaris kernel: Caller[00492598]: link_path_walk+0x58/0x140 Nov 1 14:09:18 solaris kernel: Caller[00493c90]: open_namei+0x70/0x660 Nov 1 14:09:18 solaris kernel: Caller[00481e60]: filp_open+0x40/0xa0 Nov 1 14:09:18 solaris kernel: Caller[00481f58]: do_sys_open+0x38/0xe0 Nov 1 14:09:18 solaris kernel: Caller[00407194]: linux_sparc_syscall32+0x34/0x40 Nov 1 14:09:18 solaris kernel: Caller[70136bcc]: 0x70136bcc Nov 1 14:09:18 solaris kernel: Instruction DUMP: e05c2048 22c4001e b0103ffe c4142004 c206602c 80a08001 124a d4142004 952ab030 ... Nov 1 19:02:24 solaris kernel: Badness in pci_map_sg at arch/sparc64/kernel/pci_iommu.c:633 Nov 1 19:02:24 solaris kernel: Call Trace: Nov 1 19:02:24 solaris kernel: [1021d99c] videobuf_iolock+0x9c/0x160 [video_buf] Nov 1 19:02:24 solaris kernel: [102264b4] bttv_prepare_buffer+0xf4/0x1c0 [bttv] Nov 1 19:02:24 solaris kernel: [1022b4a4] bttv_do_ioctl+0x1084/0x1ea0 [bttv] Nov 1 19:02:24 solaris kernel: [101ee168] video_usercopy+0xc8/0x2a0 [videodev] Nov 1 19:02:24 solaris kernel: [0049585c] do_ioctl+0x5c/0x80 Nov 1 19:02:24 solaris kernel: [004958ec] vfs_ioctl+0x6c/0x340 Nov 1 19:02:24 solaris kernel: [00495be8] sys_ioctl+0x28/0x60 Nov 1 19:02:24 solaris kernel: [004ad904] compat_sys_ioctl+0xe4/0x3c0 Nov 1 19:02:24 solaris kernel: [00407194] linux_sparc_syscall32+0x34/0x40 Nov 1 19:02:24 solaris kernel: [00047db0] 0x47db0 Nov 1 19:02:24 solaris kernel: videobuf_dma_pci_map: pci_map_sg failed -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bttv and SPARC
Hi all. I would like to know if bttv is supposed to work under sparc (Ultra 10). I have a Jetway TV tuner (card=78). I can insert the bttv module succesfully Here is the dmesg output Linux video capture interface: v1.00 bttv: driver version 0.9.16 loaded bttv: using 8 buffers with 2080k (260 pages) each for capture bttv: Bt8xx card found (0). bttv0: Bt878 (rev 17) at :02:01.0, irq: 7499520, latency: 128, mmio: 0x1ff2000 bttv0: using: Jetway TV/Capture JW-TV878-FBK, Kworld KW-TV878RF [card=78,insmod option] bttv0: gpio: en=, out= in=003f [init] bttv0: using tuner=5 bttv0: i2c: checking for TDA9875 @ 0xb0... not found bttv0: i2c: checking for TDA7432 @ 0x8a... not found bttv0: i2c: checking for TDA9887 @ 0x86... not found : chip found @ 0xc0 (bt878 #0 [sw]) : All bytes are equal. It is not a TEA5767 tuner 0-0060: type set to 5 (Philips PAL_BG (FI1216 and compatibles)) bttv0: registered device video0 bttv0: registered device vbi0 bttv0: registered device radio0 bttv0: PLL: 28636363 = 35468950 .. ok Everything seems to be fine but with xawtv I get only a black window. I can tune the stations, the radio and the sound works but no picture at all. Here is the output of v4l-conf v4l-conf: using X11 display :0.0 dga: version 2.0 WARNING: No DGA support available for this display. mode: 1152x900, depth=24, bpp=32, bpl=4608, base=unknown /dev/video0 [v4l2]: ioctl VIDIOC_QUERYCAP: Invalid argument /dev/video0 [v4l]: configuration done If I try xawtv -noxv I get these error messages: pci_iommu: alloc_streaming_cluster of npages(260) failed! bttv0: SCERR @ c3fea014,bits: VSYNC HSYNC OFLOW FBUS PABORT SCERR* bttv0: SCERR @ c3fea01c,bits: VSYNC HSYNC OFLOW FBUS PABORT SCERR* bttv0: SCERR @ c3fea01c,bits: VSYNC HSYNC OFLOW FBUS PABORT SCERR* bttv0: SCERR @ c3fea01c,bits: VSYNC HSYNC OFLOW FBUS PABORT SCERR* bttv0: timeout: drop=0 irq=3687/3687, risc=c3fea01c, bits: HSYNC OFLOW bttv0: reset, reinitialize bttv0: PLL: 28636363 = 35468950 . ok Btw. the card works fine under Solaris 10 with the bt848x driver. Somehow this is ridiculous because this driver is actually a port of the linux bttv driver. Any suggestion would be appreciated. Best regards, Attila -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Printing problems on Ultra 2
Hi to everyone, Recently I have tried Debian Stable on my new Ultra 2. Everything seems to be working fine excepting the parallel port. I have a HP Laserjet 4L. The 2.4.27 kernel contains 2 drivers for the Sun parallel port: bpp (obsolete) and parport_sunbpp. With the parport_sunbpp module one cannot print anything useful; the output contains only garbage. With the bpp module the situation is somehow better, but the printout contains additional strange characters; some characters are missing too. With dvilj4l the printout is almost perfect but some characters are missing or they are in the wrong position. I know there are problems with the sun parallel port under Linux, but I would like to know if someone can print correctly using the sbus parallel port. I mention that the same printer works just fine with Debian SPARC on a Sun Ultra 10, which uses the parport_pc module. So we can be sure that the problem is connected with the parport_sunbpp/bpp driver. Any advice would be appreciated. Best regards, Attila -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Printing problems on Ultra 2
Hi to everyone, Recently I have tried Debian Stable on my new Ultra 2. Everything seems to be working fine excepting the parallel port. I have an HP Laserjet 4L. The 2.4.27 kernel contains 2 drivers for the Sun parallel port: bpp (obsolete) and parport_sunbpp. With the parport_sunbpp module one cannot print anything useful; the output contains only garbage. With the bpp module the situation is somehow better, but the printout contains additional strange characters; some characters are missing also. With dvilj4l the printout is almost perfect, some characters are missing though. I have tried messing with some udelay-s but without success. I would like to know if anyone managed to print using the bpp/parport_sunbpp driver under SPARC-linux. I mention that the printer works fine with my Ultra 10 under Debian-SPARC, but with the parport_pc driver (pc-style parallel port). So we can be absolutely sure that the problem is connected with the bpp/parport_sunbpp drivers. Any suggestion would be appreciated. Best regards, Attila -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
