ACPI suspend lockup 2.6.10, works with 2.6.11
on 15 Feb I complained about 2.6.10, now it works now, with 2.6.11. However, there are two warnings still, below. PM: Preparing system for suspend Stopping tasks: | PM: Entering state. Back to C! Debug: sleeping function called from invalid context at mm/slab.c:2082 in_atomic():0, irqs_disabled():1 [] __might_sleep+0xac/0xc0 [] kmem_cache_alloc+0x5b/0x60 [] acpi_pci_link_set+0x43/0x193 [] irqrouter_resume+0x1c/0x30 [] sysdev_resume+0xf7/0xfc [] device_power_up+0x5/0xa [] suspend_enter+0x35/0x50 [] enter_state+0x55/0x90 [] acpi_suspend+0x25/0x33 [] copy_from_user+0x58/0x90 [] acpi_system_write_sleep+0x69/0x7a [] vfs_write+0xc3/0x130 [] sys_write+0x47/0x80 [] sysenter_past_esp+0x52/0x75 PM: Finishing up. > point of previous lockup here PCI: cache line size of 32 is not supported by device :00:1d.7 ehci_hcd :00:1d.7: USB 2.0 restarted, EHCI 1.00, driver 10 Dec 2004 ACPI: PCI interrupt :00:1f.1[A] -> GSI 11 (level, low) -> IRQ 11 ACPI: PCI interrupt :00:1f.5[B] -> GSI 11 (level, low) -> IRQ 11 PCI: Setting latency timer of device :00:1f.5 to 64 ACPI: PCI interrupt :02:00.0[A] -> GSI 11 (level, low) -> IRQ 11 PCI: Enabling device :02:02.0 ( -> 0002) ACPI: PCI interrupt :02:02.0[A] -> GSI 11 (level, low) -> IRQ 11 Restarting tasks... done ALPS Touchpad (Glidepoint) detected Disabling hardware tapping input: AlpsPS/2 ALPS TouchPad on isa0060/serio4 lspci: 00:00.0 Host bridge: Intel Corp. 82855PM Processor to I/O Controller (rev 03) 00:01.0 PCI bridge: Intel Corp. 82855PM Processor to AGP Controller (rev 03) 00:1d.0 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #1 (rev 03) 00:1d.1 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #2 (rev 03) 00:1d.2 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #3 (rev 03) 00:1d.7 USB Controller: Intel Corp. 82801DB (ICH4) USB2 EHCI Controller (rev 03) 00:1e.0 PCI bridge: Intel Corp. 82801BAM/CAM PCI Bridge (rev 83) 00:1f.0 ISA bridge: Intel Corp. 82801DBM LPC Interface Controller (rev 03) 00:1f.1 IDE interface: Intel Corp. 82801DBM (ICH4) Ultra ATA Storage Controller (rev 03) 00:1f.3 SMBus: Intel Corp. 82801DB/DBM (ICH4) SMBus Controller (rev 03) 00:1f.5 Multimedia audio controller: Intel Corp. 82801DB (ICH4) AC'97 Audio Controller (rev 03) 00:1f.6 Modem: Intel Corp. 82801DB (ICH4) AC'97 Modem Controller (rev 03) 01:00.0 VGA compatible controller: ATI Technologies Inc Radeon R250 Lf [Radeon Mobility 9000 M9] (rev 01) 02:00.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host Controller (rev 80) 02:01.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) 02:02.0 Network controller: Intel Corp. PRO/Wireless LAN 2100 3B Mini PCI Adapter (rev 04) 02:03.0 CardBus bridge: ENE Technology Inc CB1410 Cardbus Controller - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
ACPI suspend lockup 2.6.10, works with 2.6.11
on 15 Feb I complained about 2.6.10, now it works now, with 2.6.11. However, there are two warnings still, below. PM: Preparing system for suspend Stopping tasks: | PM: Entering state. Back to C! Debug: sleeping function called from invalid context at mm/slab.c:2082 in_atomic():0, irqs_disabled():1 [c012102c] __might_sleep+0xac/0xc0 [c016047b] kmem_cache_alloc+0x5b/0x60 [c026c5aa] acpi_pci_link_set+0x43/0x193 [c026ca09] irqrouter_resume+0x1c/0x30 [c02a8e47] sysdev_resume+0xf7/0xfc [c02ad8e5] device_power_up+0x5/0xa [c014b995] suspend_enter+0x35/0x50 [c014ba35] enter_state+0x55/0x90 [c02697b6] acpi_suspend+0x25/0x33 [c0238088] copy_from_user+0x58/0x90 [c0269895] acpi_system_write_sleep+0x69/0x7a [c0181b83] vfs_write+0xc3/0x130 [c0181cb7] sys_write+0x47/0x80 [c0103b49] sysenter_past_esp+0x52/0x75 PM: Finishing up. point of previous lockup here PCI: cache line size of 32 is not supported by device :00:1d.7 ehci_hcd :00:1d.7: USB 2.0 restarted, EHCI 1.00, driver 10 Dec 2004 ACPI: PCI interrupt :00:1f.1[A] - GSI 11 (level, low) - IRQ 11 ACPI: PCI interrupt :00:1f.5[B] - GSI 11 (level, low) - IRQ 11 PCI: Setting latency timer of device :00:1f.5 to 64 ACPI: PCI interrupt :02:00.0[A] - GSI 11 (level, low) - IRQ 11 PCI: Enabling device :02:02.0 ( - 0002) ACPI: PCI interrupt :02:02.0[A] - GSI 11 (level, low) - IRQ 11 Restarting tasks... done ALPS Touchpad (Glidepoint) detected Disabling hardware tapping input: AlpsPS/2 ALPS TouchPad on isa0060/serio4 lspci: 00:00.0 Host bridge: Intel Corp. 82855PM Processor to I/O Controller (rev 03) 00:01.0 PCI bridge: Intel Corp. 82855PM Processor to AGP Controller (rev 03) 00:1d.0 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #1 (rev 03) 00:1d.1 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #2 (rev 03) 00:1d.2 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #3 (rev 03) 00:1d.7 USB Controller: Intel Corp. 82801DB (ICH4) USB2 EHCI Controller (rev 03) 00:1e.0 PCI bridge: Intel Corp. 82801BAM/CAM PCI Bridge (rev 83) 00:1f.0 ISA bridge: Intel Corp. 82801DBM LPC Interface Controller (rev 03) 00:1f.1 IDE interface: Intel Corp. 82801DBM (ICH4) Ultra ATA Storage Controller (rev 03) 00:1f.3 SMBus: Intel Corp. 82801DB/DBM (ICH4) SMBus Controller (rev 03) 00:1f.5 Multimedia audio controller: Intel Corp. 82801DB (ICH4) AC'97 Audio Controller (rev 03) 00:1f.6 Modem: Intel Corp. 82801DB (ICH4) AC'97 Modem Controller (rev 03) 01:00.0 VGA compatible controller: ATI Technologies Inc Radeon R250 Lf [Radeon Mobility 9000 M9] (rev 01) 02:00.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host Controller (rev 80) 02:01.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) 02:02.0 Network controller: Intel Corp. PRO/Wireless LAN 2100 3B Mini PCI Adapter (rev 04) 02:03.0 CardBus bridge: ENE Technology Inc CB1410 Cardbus Controller - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
ACPI lockup 2.6.10
My laptop, intel Centrino M based, all intel chips except graphics. After opening laptop, I have to push power button, then it goes: Back to C! Debug: sleeping function called from invalid context at mm/slab.c:2055 in_atomic():0,irqs_disabled():1 __might_sleep __kmalloc acpi_os_allocate acpi_ut_callocate acpi_ut_initialize_buffer acpi_rc_create_byte_stream acpi_rs_set_srs_method_data acpi_pci_link_set irqrouter_resume sysdev_resume device_power_up suspend_enter enter_state acpi_suspend copy_from_user acpi_system_write_sleep vfs_write strncpy_from_user sys_write sysenter_past_esp PM: Finishing up. and there it is , every day :) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
ACPI lockup 2.6.10
My laptop, intel Centrino M based, all intel chips except graphics. After opening laptop, I have to push power button, then it goes: Back to C! Debug: sleeping function called from invalid context at mm/slab.c:2055 in_atomic():0,irqs_disabled():1 __might_sleep __kmalloc acpi_os_allocate acpi_ut_callocate acpi_ut_initialize_buffer acpi_rc_create_byte_stream acpi_rs_set_srs_method_data acpi_pci_link_set irqrouter_resume sysdev_resume device_power_up suspend_enter enter_state acpi_suspend copy_from_user acpi_system_write_sleep vfs_write strncpy_from_user sys_write sysenter_past_esp PM: Finishing up. and there it is , every day :) - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
HANG 2.4.5-ac5: Netfinity 3000 IDE cdrom DMA enable
when dma enabled by default, hangs No time for exact investigation as the computer is here only for installation. There is SMP MB, UP kernel. SCSI HDD IDE CDROM: CRD-8400B,ATAPI CD/DVD-ROM drive hdparm utility says on /dev/hdc: getmultcount getnovers getgeo all failed without DMA, it just works, Elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
HANG 2.4.5-ac5: Netfinity 3000 IDE cdrom DMA enable
when dma enabled by default, hangs No time for exact investigation as the computer is here only for installation. There is SMP MB, UP kernel. SCSI HDD IDE CDROM: CRD-8400B,ATAPI CD/DVD-ROM drive hdparm utility says on /dev/hdc: getmultcount getnovers getgeo all failed without DMA, it just works, Elmer. - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Aironet doesn't work
On Mon, 30 Apr 2001, Jeff Garzik wrote: > Not correct -- you do not need I82365 if you have CardBus. However, if > you are running 2.4.4 you should be ok. So it is nice I dont have to prove it. Never seen cardbus laptop with linux yet. ( But Mandrake can send me one :) ) Elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Aironet doesn't work
On Mon, 30 Apr 2001, Jeff Garzik wrote: Not correct -- you do not need I82365 if you have CardBus. However, if you are running 2.4.4 you should be ok. So it is nice I dont have to prove it. Never seen cardbus laptop with linux yet. ( But Mandrake can send me one :) ) Elmer. - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Aironet doesn't work
the whole pcmcia does not work in 2.4. Not with latest cardmgr. What makes airo_cs to work is that pcmcia package and kernel modules are replaced. That is what most of distros do. Which overwrites kernel standard ones and really fucks things up for pcmcia drivers being in kernel. Elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: Aironet doesn't work
the whole pcmcia does not work in 2.4. Not with latest cardmgr. What makes airo_cs to work is that pcmcia package and kernel modules are replaced. That is what most of distros do. Which overwrites kernel standard ones and really fucks things up for pcmcia drivers being in kernel. Elmer. - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
2.4.3 SMP: nfs stale handle, fb dualhead hardlock, G400/450 misnaming
1. stale NFS file handle 2.4.2-ac28 serving nfs3 from reiserfs 2.4.3 being nfs3 client, nfs damn slow on 100Mbps p2p link. mounted nfs with rsize=8192,wsize=8192 nfs fast. soon got : bash-2.04$ls ls .: stale NFS file handle just about home directory it says so. 2. Hard lockup: G450, I set con2fb, switch consoles some times and there it comes. swithc between X and single console is OK. 3. seems that I have G450 and linux shows it as G400. bash-2.04$ /sbin/lspci: 01:00.0 VGA compatible controller: Matrox Graphics, Inc. MGA G400 AGP (rev 82) /proc/pci: VGA compatible controller: Matrox Graphics, Inc. MGA G400 AGP (rev 130). /sbin/lspci -n 01:00.0 Class 0300: 102b:0525 (rev 82) at least when I finally, after a longlong braindamage, started using G450 stuff everywhere, it become more and more usable. for dual fb X it is stable and usable now. G400 drivers also work, but matroxset aint switching second head to monitor output, neither does anything else. It remains blank. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
2.4.3 SMP: nfs stale handle, fb dualhead hardlock, G400/450 misnaming
1. stale NFS file handle 2.4.2-ac28 serving nfs3 from reiserfs 2.4.3 being nfs3 client, nfs damn slow on 100Mbps p2p link. mounted nfs with rsize=8192,wsize=8192 nfs fast. soon got : bash-2.04$ls ls .: stale NFS file handle just about home directory it says so. 2. Hard lockup: G450, I set con2fb, switch consoles some times and there it comes. swithc between X and single console is OK. 3. seems that I have G450 and linux shows it as G400. bash-2.04$ /sbin/lspci: 01:00.0 VGA compatible controller: Matrox Graphics, Inc. MGA G400 AGP (rev 82) /proc/pci: VGA compatible controller: Matrox Graphics, Inc. MGA G400 AGP (rev 130). /sbin/lspci -n 01:00.0 Class 0300: 102b:0525 (rev 82) at least when I finally, after a longlong braindamage, started using G450 stuff everywhere, it become more and more usable. for dual fb X it is stable and usable now. G400 drivers also work, but matroxset aint switching second head to monitor output, neither does anything else. It remains blank. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
ipc/shm.c ifdef CONFIG_PROC_FS
missing for line 73 at 2.4.0 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
ipc/shm.c ifdef CONFIG_PROC_FS
missing for line 73 at 2.4.0 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: OOPS: reiserfs, 2.4.2-ac26 SMP
Chris Mason wrote: > > Most likely compiled with redhat gcc 2.96. Please upgrade to their latest, > or use kgcc. Ok, after 1. upgrading redhat gcc 2. applying that BKL in vmtruncate minipatch now it copies about 50MB before cp gets stuck on do_journal 2.4.0 with reiserfs patch and reiserfs quota patch just works (not the quota part yet). - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: OOPS: reiserfs, 2.4.2-ac26 SMP
On Wed, 28 Mar 2001, Chris Mason wrote: > Most likely compiled with redhat gcc 2.96. Please upgrade to their latest, > or use kgcc. umm, upgraded to their latest, the only difference is that it wont happen now right away, but after some time. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: OOPS: reiserfs, 2.4.2-ac26 SMP
On Wed, 28 Mar 2001, Chris Mason wrote: Most likely compiled with redhat gcc 2.96. Please upgrade to their latest, or use kgcc. umm, upgraded to their latest, the only difference is that it wont happen now right away, but after some time. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: OOPS: reiserfs, 2.4.2-ac26 SMP
Chris Mason wrote: Most likely compiled with redhat gcc 2.96. Please upgrade to their latest, or use kgcc. Ok, after 1. upgrading redhat gcc 2. applying that BKL in vmtruncate minipatch now it copies about 50MB before cp gets stuck on do_journal 2.4.0 with reiserfs patch and reiserfs quota patch just works (not the quota part yet). - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: OOPS: reiserfs, 2.4.2-ac26 SMP
On Wed, 28 Mar 2001, Elmer Joandi wrote: > 2.4.2-ac26, > mkreiserfs /dev/hda11 > mount /dev/hda11 /mnt/space > cp -dpR /usr/* /mnt/space/ followup: if using older 3.5 reiserfs format cp just gets stuck in WCHAN=do_jour... problem is reproducable, also 2.4.3-pre8, not tested others. first file is where it happens probably, it is /usr/X11R6/bin something on redhat70. At least X11R6/bin is made as only directory and attemt to look into that ends up with waiting on semaphore. the same time, peaceful "touch somefile" just works. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
OOPS: reiserfs, 2.4.2-ac26 SMP
Tyan 260 Dual PIII, 512M RAM, 2.4.2-ac26, mkreiserfs /dev/hda11 mount /dev/hda11 /mnt/space cp -dpR /usr/* /mnt/space/ immediately: Mar 28 04:23:17 server kernel: Unable to handle kernel NULL pointer dereference at virtual address 0014 Mar 28 04:23:17 server kernel: printing eip: Mar 28 04:23:17 server kernel: c0166488 Mar 28 04:23:17 server kernel: *pde = Mar 28 04:23:17 server kernel: Oops: Mar 28 04:23:17 server kernel: CPU:1 Mar 28 04:23:17 server kernel: EIP:0010:[create_virtual_node+664/1168] Mar 28 04:23:17 server kernel: EFLAGS: 00010213 Mar 28 04:23:17 server kernel: eax: ebx: d0607c0c ecx: 0044 edx: Mar 28 04:23:17 server kernel: esi: dc538000 edi: dc5381b4 ebp: 0011 esp: d0607aac Mar 28 04:23:17 server kernel: ds: 0018 es: 0018 ss: 0018 Mar 28 04:23:17 server kernel: Process cp (pid: 5494, stackpage=d0607000) Mar 28 04:23:17 server kernel: Stack: dc538000 dc5381b4 0044 0198 d0007bc0 cfff9018 Mar 28 04:23:17 server kernel:0020 d0607c0c d0007bc0 c0167a5f d0607c0c dc538000 Mar 28 04:23:17 server kernel:0084 d0607ba4 c01694c0 102c 1000 0044 0069 Mar 28 04:23:17 server kernel: Call Trace: [ip_check_balance+847/2784] [unfix_nodes+352/368] [fix_nodes+277/1104] [reiserfs_insert_item+136/272] [reiserfs_new_inode+986/1280] [reiserfs_create+101/320] [dput+28/336] Mar 28 04:23:17 server kernel:[vfs_create+271/352] [lookup_hash+143/208] [open_namei+342/1648] [dentry_open+192/320] [filp_open+54/96] [sys_open+52/192] [system_call+51/56] Mar 28 04:23:17 server kernel: Mar 28 04:23:17 server kernel: Code: 8b 40 14 ff d0 89 c2 8b 06 83 c4 10 01 c2 89 16 8b 83 8c 01 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
2.2.16, 2.4.*, harddeadlocks with XFree86 and framebuffer
So, the problem is on different hardware and kernel versions. 1. looks like VT switch with multiple X copies running hangs on certain conditions 2. More critical: Matrox G400 dualhead AGP 16M hangs immediately with fb. Tested computers: 1. Tyan 230 SMP Dual PIII 667Mhz , 512MB 2. MSI SMP Dual PIII 667Mhz 512MB 3. Tyan 1834 SMP Dual Celeron 256MB Software: 1 compiled: 2.4.0,2.4.2,.2.4.2-ac8 2 RH7 stock: 2.2.16 1. True multihead hang on computers 2, 3, both software : First, to get it fly: I must log in via network, because executing X behind X/console causes vt switch. First I run USB keyboard patched X for ATI cards, non-fb. That one I can run forever, normally, until: if to execute matrox (two cards : Millenium + Mystique) server, it hangs or wont work if I modprobe matrofb_base, then now I have single possibility to run Xfree86-4 _once_ for matrox fb. If either of X's(for two matroxes or ATIs) quits, there is hardlock. ping stops, capslock doesnt work(for console keyboard, USB one naturally doesnt work at all), 2. Single dualhed G400 hang, on computers 1 2 3, software 1,2 . Well, it is simpler, it hangs with framebuffer and with normal X mode, with 2.2.16 and with 2.4.*, quite soon. inserting framebuffer module hangs the beast. framebuffer matrox compiled into kernel does not hang until I try to start X in some ways or do other console change/swtiching operations... Actually, RedHat 7.0 doesnt install on SMP Matrox dualhead G400 as XWindows setup probe lock the box up. It looks like I should tear it down and use old nfsroot based Xtermian again... And sell G400, with which there is nothing to do. Pity, everything works, but once a day ATI XFree86 sucks some dust in and gets killed and the whole box goes with it., cant stand it... Elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
2.2.16, 2.4.*, harddeadlocks with XFree86 and framebuffer
So, the problem is on different hardware and kernel versions. 1. looks like VT switch with multiple X copies running hangs on certain conditions 2. More critical: Matrox G400 dualhead AGP 16M hangs immediately with fb. Tested computers: 1. Tyan 230 SMP Dual PIII 667Mhz , 512MB 2. MSI SMP Dual PIII 667Mhz 512MB 3. Tyan 1834 SMP Dual Celeron 256MB Software: 1 compiled: 2.4.0,2.4.2,.2.4.2-ac8 2 RH7 stock: 2.2.16 1. True multihead hang on computers 2, 3, both software : First, to get it fly: I must log in via network, because executing X behind X/console causes vt switch. First I run USB keyboard patched X for ATI cards, non-fb. That one I can run forever, normally, until: if to execute matrox (two cards : Millenium + Mystique) server, it hangs or wont work if I modprobe matrofb_base, then now I have single possibility to run Xfree86-4 _once_ for matrox fb. If either of X's(for two matroxes or ATIs) quits, there is hardlock. ping stops, capslock doesnt work(for console keyboard, USB one naturally doesnt work at all), 2. Single dualhed G400 hang, on computers 1 2 3, software 1,2 . Well, it is simpler, it hangs with framebuffer and with normal X mode, with 2.2.16 and with 2.4.*, quite soon. inserting framebuffer module hangs the beast. framebuffer matrox compiled into kernel does not hang until I try to start X in some ways or do other console change/swtiching operations... Actually, RedHat 7.0 doesnt install on SMP Matrox dualhead G400 as XWindows setup probe lock the box up. It looks like I should tear it down and use old nfsroot based Xtermian again... And sell G400, with which there is nothing to do. Pity, everything works, but once a day ATI XFree86 sucks some dust in and gets killed and the whole box goes with it., cant stand it... Elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
OOPS: reiserfs, 2.4.2-ac26 SMP
Tyan 260 Dual PIII, 512M RAM, 2.4.2-ac26, mkreiserfs /dev/hda11 mount /dev/hda11 /mnt/space cp -dpR /usr/* /mnt/space/ immediately: Mar 28 04:23:17 server kernel: Unable to handle kernel NULL pointer dereference at virtual address 0014 Mar 28 04:23:17 server kernel: printing eip: Mar 28 04:23:17 server kernel: c0166488 Mar 28 04:23:17 server kernel: *pde = Mar 28 04:23:17 server kernel: Oops: Mar 28 04:23:17 server kernel: CPU:1 Mar 28 04:23:17 server kernel: EIP:0010:[create_virtual_node+664/1168] Mar 28 04:23:17 server kernel: EFLAGS: 00010213 Mar 28 04:23:17 server kernel: eax: ebx: d0607c0c ecx: 0044 edx: Mar 28 04:23:17 server kernel: esi: dc538000 edi: dc5381b4 ebp: 0011 esp: d0607aac Mar 28 04:23:17 server kernel: ds: 0018 es: 0018 ss: 0018 Mar 28 04:23:17 server kernel: Process cp (pid: 5494, stackpage=d0607000) Mar 28 04:23:17 server kernel: Stack: dc538000 dc5381b4 0044 0198 d0007bc0 cfff9018 Mar 28 04:23:17 server kernel:0020 d0607c0c d0007bc0 c0167a5f d0607c0c dc538000 Mar 28 04:23:17 server kernel:0084 d0607ba4 c01694c0 102c 1000 0044 0069 Mar 28 04:23:17 server kernel: Call Trace: [ip_check_balance+847/2784] [unfix_nodes+352/368] [fix_nodes+277/1104] [reiserfs_insert_item+136/272] [reiserfs_new_inode+986/1280] [reiserfs_create+101/320] [dput+28/336] Mar 28 04:23:17 server kernel:[vfs_create+271/352] [lookup_hash+143/208] [open_namei+342/1648] [dentry_open+192/320] [filp_open+54/96] [sys_open+52/192] [system_call+51/56] Mar 28 04:23:17 server kernel: Mar 28 04:23:17 server kernel: Code: 8b 40 14 ff d0 89 c2 8b 06 83 c4 10 01 c2 89 16 8b 83 8c 01 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: OOPS: reiserfs, 2.4.2-ac26 SMP
On Wed, 28 Mar 2001, Elmer Joandi wrote: 2.4.2-ac26, mkreiserfs /dev/hda11 mount /dev/hda11 /mnt/space cp -dpR /usr/* /mnt/space/ followup: if using older 3.5 reiserfs format cp just gets stuck in WCHAN=do_jour... problem is reproducable, also 2.4.3-pre8, not tested others. first file is where it happens probably, it is /usr/X11R6/bin something on redhat70. At least X11R6/bin is made as only directory and attemt to look into that ends up with waiting on semaphore. the same time, peaceful "touch somefile" just works. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: atyfb,matrox hardlocks, multihead, USB broken, 2.4.2-ac8
Got it to work somewhat... that was real long f*** the only sequence - no dga, kernel boots and BIOS uses Matrox PCI as first graphics via matrox start up two ATI cards(one AGP, one PCI)(xinerama mode, screwed output ) with Xserver hacked to read /dev/input/event* ant talking direct to ATI. now insert matroxfb_base now start two normal XFree86 on framebuffer /dev/fb0 and /dev/fb1, i.e. matroxes, xinerama keyboard from system console... log into matroxes ATI dualscreen picture clears itself by magic spell and becomes usable. everything works... until ATI exits, then there is kernel hardlock. Weird that all locks are hardlocks - no ping no sysreq... SMP here. ah - message from matrox framebuffer - complaining no irq A assigned to slot, and suggesting that BIOS is buggy. Will I be more happy when using a dualhead matrox AGP instead of AGP+PCI ATI pair ? 2.4.0 kernel, 2.4.2-ac8 USB looks like very very broken. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
atyfb,matrox hardlocks, multihead, USB broken, 2.4.2-ac8
2.4.2-ac8, with 4 graphics cards, Dual Celeron now with 2.4.2-ac8 it is even more clear any attempt to insert module ends with straight lockup video mode swithc occurs and then ping to the box stops immediately. more, starting X locks kernel the same way. meantime I changed from BIOS the AGP to be primary video. this was probably what made the most of difference as AGP is with largest PCI ID. USB is also nonworking. 2.4.0 was most part OK. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
True multihead, lots of crashes
Hi, DUAL-Celeron, 2.4.0, ATI64 Rage pro AGP, Matrox Millenium, Matrox Mystique, ATI Rage II+ PCI, normal keyboard USB keyboard, 2x USB mouse. USB hub and USB hub in keyboard. 1. Just plain Xfree 4.0.2 with xinerama works stable, just gets killed by memory-hogging netscape. 2. framebuffer leads to hardlock sometimes quite soon. whatever, ati or matrox. Looks like at VT switch. No way to debug, 3. I have here second patched Xfree, got someones patch and rebuilt rpm, to support USB keyboard. the box crashes also without it. It looks like all places where VT switch may occur, are commented out from xfree. Without framebuffer, everything works, but vt switch still occurs and only one X is usable... first X stalls until second gets killed and then works again. sometimes I can even switch between them from console. Does the console input trigger console switch... dont understand. strace doesnt show anything ... looks like two copies of X talk each other via shared memory or something... elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
True multihead, lots of crashes
Hi, DUAL-Celeron, 2.4.0, ATI64 Rage pro AGP, Matrox Millenium, Matrox Mystique, ATI Rage II+ PCI, normal keyboard USB keyboard, 2x USB mouse. USB hub and USB hub in keyboard. 1. Just plain Xfree 4.0.2 with xinerama works stable, just gets killed by memory-hogging netscape. 2. framebuffer leads to hardlock sometimes quite soon. whatever, ati or matrox. Looks like at VT switch. No way to debug, 3. I have here second patched Xfree, got someones patch and rebuilt rpm, to support USB keyboard. the box crashes also without it. It looks like all places where VT switch may occur, are commented out from xfree. Without framebuffer, everything works, but vt switch still occurs and only one X is usable... first X stalls until second gets killed and then works again. sometimes I can even switch between them from console. Does the console input trigger console switch... dont understand. strace doesnt show anything ... looks like two copies of X talk each other via shared memory or something... elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
atyfb,matrox hardlocks, multihead, USB broken, 2.4.2-ac8
2.4.2-ac8, with 4 graphics cards, Dual Celeron now with 2.4.2-ac8 it is even more clear any attempt to insert module ends with straight lockup video mode swithc occurs and then ping to the box stops immediately. more, starting X locks kernel the same way. meantime I changed from BIOS the AGP to be primary video. this was probably what made the most of difference as AGP is with largest PCI ID. USB is also nonworking. 2.4.0 was most part OK. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: atyfb,matrox hardlocks, multihead, USB broken, 2.4.2-ac8
Got it to work somewhat... that was real long f*** the only sequence - no dga, kernel boots and BIOS uses Matrox PCI as first graphics via matrox start up two ATI cards(one AGP, one PCI)(xinerama mode, screwed output ) with Xserver hacked to read /dev/input/event* ant talking direct to ATI. now insert matroxfb_base now start two normal XFree86 on framebuffer /dev/fb0 and /dev/fb1, i.e. matroxes, xinerama keyboard from system console... log into matroxes ATI dualscreen picture clears itself by magic spell and becomes usable. everything works... until ATI exits, then there is kernel hardlock. Weird that all locks are hardlocks - no ping no sysreq... SMP here. ah - message from matrox framebuffer - complaining no irq A assigned to slot, and suggesting that BIOS is buggy. Will I be more happy when using a dualhead matrox AGP instead of AGP+PCI ATI pair ? 2.4.0 kernel, 2.4.2-ac8 USB looks like very very broken. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
RE: USB mouse jumping
On Tue, 13 Feb 2001, Dunlap, Randy wrote: > > If USB mice had serial numbers (like some USB storage devices > do), then we could tell that it's the same mouse on the > same connector and not change from mouse0 to mouse1. > Currently it looks like a new device attachment. > > One possible solution is for you to use /dev/usb/mice, > which is all USB mice merged into one input stream. Please, if it is possible, make it simple and sensible. if to have true multihead, with 5 keyboards and mice, I would really wish that the device numbers are connected to physical holes for plugs, otherwise anyone (of schoolchildren for example) can do simple nasty stupid things. And also it is the dream for true dumbuser with one mouse, because it then just works out of box. Or you tell that with USB internal design you can not know physical connector unique number ? elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
USB mouse jumping
I dont know, if it is bug or feature, but, USB mouse jumps around (between) /dev/input/mouse0 and mouse1 when taken out and put back in(to same connector), 2.4.0 kernel. Annoys, should not be the default behaviour, IMHO. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
RE: USB mouse jumping
On Tue, 13 Feb 2001, Dunlap, Randy wrote: If USB mice had serial numbers (like some USB storage devices do), then we could tell that it's the same mouse on the same connector and not change from mouse0 to mouse1. Currently it looks like a new device attachment. One possible solution is for you to use /dev/usb/mice, which is all USB mice merged into one input stream. Please, if it is possible, make it simple and sensible. if to have true multihead, with 5 keyboards and mice, I would really wish that the device numbers are connected to physical holes for plugs, otherwise anyone (of schoolchildren for example) can do simple nasty stupid things. And also it is the dream for true dumbuser with one mouse, because it then just works out of box. Or you tell that with USB internal design you can not know physical connector unique number ? elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: aironet4500_card (2.4.1-ac8), The PCI BIOS has not enabled thisdevice!
Sorry, no time to test, neither I have cisco cards. However, general notes: 1. Aironet did (cisco may do) weird tricks on bus. 2. insmod driver -> leds go out, that may be normal. ifconfig up should bring leds on. 3. People who fail with both drivers (Bens and mine), have had weird BIOS or BIOS settings in most of cases. IO conflict with bios configuration port (ICL 486 ), old PCI BIOS (Intel Pentium 200Mhz board) , etc. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: aironet4500_card (2.4.1-ac8), The PCI BIOS has not enabled thisdevice!
Sorry, no time to test, neither I have cisco cards. However, general notes: 1. Aironet did (cisco may do) weird tricks on bus. 2. insmod driver - leds go out, that may be normal. ifconfig up should bring leds on. 3. People who fail with both drivers (Bens and mine), have had weird BIOS or BIOS settings in most of cases. IO conflict with bios configuration port (ICL 486 ), old PCI BIOS (Intel Pentium 200Mhz board) , etc. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: prerelease total nonmodular compile, compiler warnings, linkingerrors
On Tue, 2 Jan 2001, Andreas Bombe wrote: > You're then using the ieee1394.o module object which doesn't include the > hardware and highlevel drivers. I've sent a patch to Linus already and > cc'd the mailing list also. !!! there were heaps of other (not related to ieee1394 stuff ) linking errors. hack was for to see other errors I hope I did not confuse anybody, and those ohter linking errors are attached again with this letter. I'd like to see if by grub can load a nonmodular kernel with everything compiled in (should be about 20MB ? ). Not that it would of any value other than being nice to see. elmer. drivers/sound/sounddrivers.o: In function `cleanup_module': drivers/sound/sounddrivers.o(.text.exit+0xf30): multiple definition of `cleanup_module' drivers/isdn/isdn.a(.text+0x176d4): first defined here ld: Warning: size of symbol `cleanup_module' changed from 105 to 45 in drivers/sound/sounddrivers.o drivers/sound/sounddrivers.o: In function `init_module': drivers/sound/sounddrivers.o(.text.init+0xade0): multiple definition of `init_module' drivers/isdn/isdn.a(.text+0x17670): first defined here ld: Warning: size of symbol `init_module' changed from 97 to 25 in drivers/sound/sounddrivers.o drivers/net/net.o: In function `network_ldisc_init': drivers/net/net.o(.text.init+0x532f): undefined reference to `mkiss_init_ctrl_dev' drivers/net/net.o(.data.init+0x52890): undefined reference to `yam_init' drivers/net/tokenring/tr.a(smctr.o): In function `smctr_reset_adapter': smctr.o(.text+0x392a): undefined reference to `__bad_udelay' smctr.o(.text+0x3938): undefined reference to `__bad_udelay' drivers/ieee1394/ieee1394.o: In function `register_builtin_highlevels': drivers/ieee1394/ieee1394.o(.text+0x27b9): undefined reference to `init_raw1394' drivers/ieee1394/ieee1394.o: In function `register_builtin_lowlevels': drivers/ieee1394/ieee1394.o(.text.init+0x24): undefined reference to `get_lynx_template' drivers/ieee1394/ieee1394.o(.text.init+0x51): undefined reference to `get_ohci_template' drivers/video/video.o: In function `aty_set_pll18818': drivers/video/video.o(.text+0xc121): undefined reference to `__bad_udelay' drivers/video/video.o: In function `init_vgachip': drivers/video/video.o(.text.init+0x41ad): undefined reference to `__bad_udelay' drivers/net/irda/irda.o: In function `toshoboe_gotosleep': drivers/net/irda/irda.o(.text+0x6b81): undefined reference to `__bad_udelay' make: *** [vmlinux] Error 1
prerelease total nonmodular compile, compiler warnings, linkingerrors
compiling everything builtin, (exept RCPCI, which does not compile)
linking errors:
drivers/ieee1394/ieee1394.a is not made, quick hack to use .o to see other
errors.
compiling warnings and linking errors are included
drivers/sound/sounddrivers.o: In function `cleanup_module':
drivers/sound/sounddrivers.o(.text.exit+0xf30): multiple definition of
`cleanup_module'
drivers/isdn/isdn.a(.text+0x176d4): first defined here
ld: Warning: size of symbol `cleanup_module' changed from 105 to 45 in
drivers/sound/sounddrivers.o
drivers/sound/sounddrivers.o: In function `init_module':
drivers/sound/sounddrivers.o(.text.init+0xade0): multiple definition of `init_module'
drivers/isdn/isdn.a(.text+0x17670): first defined here
ld: Warning: size of symbol `init_module' changed from 97 to 25 in
drivers/sound/sounddrivers.o
drivers/net/net.o: In function `network_ldisc_init':
drivers/net/net.o(.text.init+0x532f): undefined reference to `mkiss_init_ctrl_dev'
drivers/net/net.o(.data.init+0x52890): undefined reference to `yam_init'
drivers/net/tokenring/tr.a(smctr.o): In function `smctr_reset_adapter':
smctr.o(.text+0x392a): undefined reference to `__bad_udelay'
smctr.o(.text+0x3938): undefined reference to `__bad_udelay'
drivers/ieee1394/ieee1394.o: In function `register_builtin_highlevels':
drivers/ieee1394/ieee1394.o(.text+0x27b9): undefined reference to `init_raw1394'
drivers/ieee1394/ieee1394.o: In function `register_builtin_lowlevels':
drivers/ieee1394/ieee1394.o(.text.init+0x24): undefined reference to
`get_lynx_template'
drivers/ieee1394/ieee1394.o(.text.init+0x51): undefined reference to
`get_ohci_template'
drivers/video/video.o: In function `aty_set_pll18818':
drivers/video/video.o(.text+0xc121): undefined reference to `__bad_udelay'
drivers/video/video.o: In function `init_vgachip':
drivers/video/video.o(.text.init+0x41ad): undefined reference to `__bad_udelay'
drivers/net/irda/irda.o: In function `toshoboe_gotosleep':
drivers/net/irda/irda.o(.text+0x6b81): undefined reference to `__bad_udelay'
make: *** [vmlinux] Error 1
md5sum: WARNING: 11 of 12 computed checksums did NOT match
ec.c:279: warning: `ec_space_setup' defined but not used
{standard input}: Assembler messages:
{standard input}:765: Warning: Indirect lcall without `*'
{standard input}:849: Warning: Indirect lcall without `*'
{standard input}:936: Warning: Indirect lcall without `*'
{standard input}:976: Warning: Indirect lcall without `*'
{standard input}:1008: Warning: Indirect lcall without `*'
{standard input}:1040: Warning: Indirect lcall without `*'
{standard input}:1071: Warning: Indirect lcall without `*'
{standard input}:1100: Warning: Indirect lcall without `*'
{standard input}:1129: Warning: Indirect lcall without `*'
{standard input}:1411: Warning: Indirect lcall without `*'
{standard input}:1504: Warning: Indirect lcall without `*'
net/network.o: In function `atm_ioctl':
net/network.o(.text+0x3c742): undefined reference to `atm_lane_init'
net/network.o(.text+0x3c7f2): undefined reference to `atm_mpoa_init'
make: *** [vmlinux] Error 1
objcopy: Warning: Output file cannot represent architecture UNKNOWN!
ip2/i2cmd.c:142: warning: `ct89' defined but not used
sx.c:1623: warning: `do_memtest_w' defined but not used
i2o_block.c:595: warning: #warning "RACE"
md5sum: WARNING: 11 of 12 computed checksums did NOT match
bttv-cards.c: In function `bttv_check_chipset':
bttv-cards.c:1389: warning: unused variable `i'
bttv-cards.c: At top level:
bttv-cards.c:1379: warning: `needs_etbf' defined but not used
mtdchar.c: In function `init_mtdchar':
mtdchar.c:452: warning: unused variable `mtd'
mtdchar.c:451: warning: unused variable `name'
mtdchar.c:450: warning: unused variable `i'
ftl.c:139: warning: `debug' defined but not used
nftlmount.c: In function `check_and_mark_free_block':
nftlmount.c:363: warning: unused variable `buf'
nftlmount.c:362: warning: unused variable `i'
sunhme.c:2791: warning: #warning This needs to be corrected... -DaveM
sdla_chdlc.c: In function `if_send':
sdla_chdlc.c:936: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_chdlc.c: In function `wpc_isr':
sdla_chdlc.c:1501: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_ppp.c: In function `if_send':
sdla_ppp.c:901: warning: unsigned int format, long unsigned int arg (arg 3)
qla1280.c:1609: warning: `qla1280_do_dpc' defined but not used
NCR5380.c:795: warning: `NCR5380_print_options' defined but not used
sr.c: In function `sr_init_command':
sr.c:347: warning: `block' might be used uninitialized in this function
cs46xx.c:2867: warning: `amp_voyetra_4294' defined but not used
cs4281.c: In function `cs4281_write_ac97':
cs4281.c:607: warning: `status' might be used uninitialized in this function
plusb.c:985: warning: initialization from incompatible pointer type
mdacon.c:133: warning: `test_mda_b' defined but not used
matroxfb_g450.c:7: warning: `matroxfb_g450_get_reg'
Compile errors: RCPCI, LANE, and others
Did full compile, just for fun:
CONFIG_for Red Creek whatever RCPCI has a syntax error
other warnings and errors, compiled on 2.4.0-prerelease, nonSMP, PIII
md5sum: WARNING: 11 of 12 computed checksums did NOT match
ec.c:279: warning: `ec_space_setup' defined but not used
{standard input}: Assembler messages:
{standard input}:765: Warning: Indirect lcall without `*'
{standard input}:849: Warning: Indirect lcall without `*'
{standard input}:936: Warning: Indirect lcall without `*'
{standard input}:976: Warning: Indirect lcall without `*'
{standard input}:1008: Warning: Indirect lcall without `*'
{standard input}:1040: Warning: Indirect lcall without `*'
{standard input}:1071: Warning: Indirect lcall without `*'
{standard input}:1100: Warning: Indirect lcall without `*'
{standard input}:1129: Warning: Indirect lcall without `*'
{standard input}:1411: Warning: Indirect lcall without `*'
{standard input}:1504: Warning: Indirect lcall without `*'
net/network.o: In function `atm_ioctl':
net/network.o(.text+0x3c742): undefined reference to `atm_lane_init'
net/network.o(.text+0x3c7f2): undefined reference to `atm_mpoa_init'
make: *** [vmlinux] Error 1
objcopy: Warning: Output file cannot represent architecture UNKNOWN!
ip2/i2cmd.c:142: warning: `ct89' defined but not used
sx.c:1623: warning: `do_memtest_w' defined but not used
i2o_block.c:595: warning: #warning "RACE"
md5sum: WARNING: 11 of 12 computed checksums did NOT match
bttv-cards.c: In function `bttv_check_chipset':
bttv-cards.c:1389: warning: unused variable `i'
bttv-cards.c: At top level:
bttv-cards.c:1379: warning: `needs_etbf' defined but not used
mtdchar.c: In function `init_mtdchar':
mtdchar.c:452: warning: unused variable `mtd'
mtdchar.c:451: warning: unused variable `name'
mtdchar.c:450: warning: unused variable `i'
ftl.c:139: warning: `debug' defined but not used
nftlmount.c: In function `check_and_mark_free_block':
nftlmount.c:363: warning: unused variable `buf'
nftlmount.c:362: warning: unused variable `i'
sunhme.c:2791: warning: #warning This needs to be corrected... -DaveM
sdla_chdlc.c: In function `if_send':
sdla_chdlc.c:936: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_chdlc.c: In function `wpc_isr':
sdla_chdlc.c:1501: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_ppp.c: In function `if_send':
sdla_ppp.c:901: warning: unsigned int format, long unsigned int arg (arg 3)
qla1280.c:1609: warning: `qla1280_do_dpc' defined but not used
NCR5380.c:795: warning: `NCR5380_print_options' defined but not used
sr.c: In function `sr_init_command':
sr.c:347: warning: `block' might be used uninitialized in this function
cs46xx.c:2867: warning: `amp_voyetra_4294' defined but not used
cs4281.c: In function `cs4281_write_ac97':
cs4281.c:607: warning: `status' might be used uninitialized in this function
plusb.c:985: warning: initialization from incompatible pointer type
mdacon.c:133: warning: `test_mda_b' defined but not used
matroxfb_g450.c:7: warning: `matroxfb_g450_get_reg' defined but not used
intrep.c:96: warning: `jffs_hexdump' defined but not used
dn_table.c:872: warning: `dn_fib_del_tree' defined but not used
irlap.c: In function `irlap_change_speed':
irlap.c:892: warning: implicit declaration of function `irlap_queue_xmit'
qos.c:609: warning: `byte_value' defined but not used
irias_object.c:37: warning: braces around scalar initializer
irias_object.c:37: warning: (near initialization for `missing.len')
irsyms.c:222: warning: `irda_cleanup' defined but not used
{standard input}: Assembler messages:
{standard input}:193: Warning: Indirect lcall without `*'
{standard input}:270: Warning: Indirect lcall without `*'
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
Compile errors: RCPCI, LANE, and others
Did full compile, just for fun:
CONFIG_for Red Creek whatever RCPCI has a syntax error
other warnings and errors, compiled on 2.4.0-prerelease, nonSMP, PIII
md5sum: WARNING: 11 of 12 computed checksums did NOT match
ec.c:279: warning: `ec_space_setup' defined but not used
{standard input}: Assembler messages:
{standard input}:765: Warning: Indirect lcall without `*'
{standard input}:849: Warning: Indirect lcall without `*'
{standard input}:936: Warning: Indirect lcall without `*'
{standard input}:976: Warning: Indirect lcall without `*'
{standard input}:1008: Warning: Indirect lcall without `*'
{standard input}:1040: Warning: Indirect lcall without `*'
{standard input}:1071: Warning: Indirect lcall without `*'
{standard input}:1100: Warning: Indirect lcall without `*'
{standard input}:1129: Warning: Indirect lcall without `*'
{standard input}:1411: Warning: Indirect lcall without `*'
{standard input}:1504: Warning: Indirect lcall without `*'
net/network.o: In function `atm_ioctl':
net/network.o(.text+0x3c742): undefined reference to `atm_lane_init'
net/network.o(.text+0x3c7f2): undefined reference to `atm_mpoa_init'
make: *** [vmlinux] Error 1
objcopy: Warning: Output file cannot represent architecture UNKNOWN!
ip2/i2cmd.c:142: warning: `ct89' defined but not used
sx.c:1623: warning: `do_memtest_w' defined but not used
i2o_block.c:595: warning: #warning "RACE"
md5sum: WARNING: 11 of 12 computed checksums did NOT match
bttv-cards.c: In function `bttv_check_chipset':
bttv-cards.c:1389: warning: unused variable `i'
bttv-cards.c: At top level:
bttv-cards.c:1379: warning: `needs_etbf' defined but not used
mtdchar.c: In function `init_mtdchar':
mtdchar.c:452: warning: unused variable `mtd'
mtdchar.c:451: warning: unused variable `name'
mtdchar.c:450: warning: unused variable `i'
ftl.c:139: warning: `debug' defined but not used
nftlmount.c: In function `check_and_mark_free_block':
nftlmount.c:363: warning: unused variable `buf'
nftlmount.c:362: warning: unused variable `i'
sunhme.c:2791: warning: #warning This needs to be corrected... -DaveM
sdla_chdlc.c: In function `if_send':
sdla_chdlc.c:936: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_chdlc.c: In function `wpc_isr':
sdla_chdlc.c:1501: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_ppp.c: In function `if_send':
sdla_ppp.c:901: warning: unsigned int format, long unsigned int arg (arg 3)
qla1280.c:1609: warning: `qla1280_do_dpc' defined but not used
NCR5380.c:795: warning: `NCR5380_print_options' defined but not used
sr.c: In function `sr_init_command':
sr.c:347: warning: `block' might be used uninitialized in this function
cs46xx.c:2867: warning: `amp_voyetra_4294' defined but not used
cs4281.c: In function `cs4281_write_ac97':
cs4281.c:607: warning: `status' might be used uninitialized in this function
plusb.c:985: warning: initialization from incompatible pointer type
mdacon.c:133: warning: `test_mda_b' defined but not used
matroxfb_g450.c:7: warning: `matroxfb_g450_get_reg' defined but not used
intrep.c:96: warning: `jffs_hexdump' defined but not used
dn_table.c:872: warning: `dn_fib_del_tree' defined but not used
irlap.c: In function `irlap_change_speed':
irlap.c:892: warning: implicit declaration of function `irlap_queue_xmit'
qos.c:609: warning: `byte_value' defined but not used
irias_object.c:37: warning: braces around scalar initializer
irias_object.c:37: warning: (near initialization for `missing.len')
irsyms.c:222: warning: `irda_cleanup' defined but not used
{standard input}: Assembler messages:
{standard input}:193: Warning: Indirect lcall without `*'
{standard input}:270: Warning: Indirect lcall without `*'
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
prerelease total nonmodular compile, compiler warnings, linkingerrors
compiling everything builtin, (exept RCPCI, which does not compile)
linking errors:
drivers/ieee1394/ieee1394.a is not made, quick hack to use .o to see other
errors.
compiling warnings and linking errors are included
drivers/sound/sounddrivers.o: In function `cleanup_module':
drivers/sound/sounddrivers.o(.text.exit+0xf30): multiple definition of
`cleanup_module'
drivers/isdn/isdn.a(.text+0x176d4): first defined here
ld: Warning: size of symbol `cleanup_module' changed from 105 to 45 in
drivers/sound/sounddrivers.o
drivers/sound/sounddrivers.o: In function `init_module':
drivers/sound/sounddrivers.o(.text.init+0xade0): multiple definition of `init_module'
drivers/isdn/isdn.a(.text+0x17670): first defined here
ld: Warning: size of symbol `init_module' changed from 97 to 25 in
drivers/sound/sounddrivers.o
drivers/net/net.o: In function `network_ldisc_init':
drivers/net/net.o(.text.init+0x532f): undefined reference to `mkiss_init_ctrl_dev'
drivers/net/net.o(.data.init+0x52890): undefined reference to `yam_init'
drivers/net/tokenring/tr.a(smctr.o): In function `smctr_reset_adapter':
smctr.o(.text+0x392a): undefined reference to `__bad_udelay'
smctr.o(.text+0x3938): undefined reference to `__bad_udelay'
drivers/ieee1394/ieee1394.o: In function `register_builtin_highlevels':
drivers/ieee1394/ieee1394.o(.text+0x27b9): undefined reference to `init_raw1394'
drivers/ieee1394/ieee1394.o: In function `register_builtin_lowlevels':
drivers/ieee1394/ieee1394.o(.text.init+0x24): undefined reference to
`get_lynx_template'
drivers/ieee1394/ieee1394.o(.text.init+0x51): undefined reference to
`get_ohci_template'
drivers/video/video.o: In function `aty_set_pll18818':
drivers/video/video.o(.text+0xc121): undefined reference to `__bad_udelay'
drivers/video/video.o: In function `init_vgachip':
drivers/video/video.o(.text.init+0x41ad): undefined reference to `__bad_udelay'
drivers/net/irda/irda.o: In function `toshoboe_gotosleep':
drivers/net/irda/irda.o(.text+0x6b81): undefined reference to `__bad_udelay'
make: *** [vmlinux] Error 1
md5sum: WARNING: 11 of 12 computed checksums did NOT match
ec.c:279: warning: `ec_space_setup' defined but not used
{standard input}: Assembler messages:
{standard input}:765: Warning: Indirect lcall without `*'
{standard input}:849: Warning: Indirect lcall without `*'
{standard input}:936: Warning: Indirect lcall without `*'
{standard input}:976: Warning: Indirect lcall without `*'
{standard input}:1008: Warning: Indirect lcall without `*'
{standard input}:1040: Warning: Indirect lcall without `*'
{standard input}:1071: Warning: Indirect lcall without `*'
{standard input}:1100: Warning: Indirect lcall without `*'
{standard input}:1129: Warning: Indirect lcall without `*'
{standard input}:1411: Warning: Indirect lcall without `*'
{standard input}:1504: Warning: Indirect lcall without `*'
net/network.o: In function `atm_ioctl':
net/network.o(.text+0x3c742): undefined reference to `atm_lane_init'
net/network.o(.text+0x3c7f2): undefined reference to `atm_mpoa_init'
make: *** [vmlinux] Error 1
objcopy: Warning: Output file cannot represent architecture UNKNOWN!
ip2/i2cmd.c:142: warning: `ct89' defined but not used
sx.c:1623: warning: `do_memtest_w' defined but not used
i2o_block.c:595: warning: #warning "RACE"
md5sum: WARNING: 11 of 12 computed checksums did NOT match
bttv-cards.c: In function `bttv_check_chipset':
bttv-cards.c:1389: warning: unused variable `i'
bttv-cards.c: At top level:
bttv-cards.c:1379: warning: `needs_etbf' defined but not used
mtdchar.c: In function `init_mtdchar':
mtdchar.c:452: warning: unused variable `mtd'
mtdchar.c:451: warning: unused variable `name'
mtdchar.c:450: warning: unused variable `i'
ftl.c:139: warning: `debug' defined but not used
nftlmount.c: In function `check_and_mark_free_block':
nftlmount.c:363: warning: unused variable `buf'
nftlmount.c:362: warning: unused variable `i'
sunhme.c:2791: warning: #warning This needs to be corrected... -DaveM
sdla_chdlc.c: In function `if_send':
sdla_chdlc.c:936: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_chdlc.c: In function `wpc_isr':
sdla_chdlc.c:1501: warning: unsigned int format, long unsigned int arg (arg 3)
sdla_ppp.c: In function `if_send':
sdla_ppp.c:901: warning: unsigned int format, long unsigned int arg (arg 3)
qla1280.c:1609: warning: `qla1280_do_dpc' defined but not used
NCR5380.c:795: warning: `NCR5380_print_options' defined but not used
sr.c: In function `sr_init_command':
sr.c:347: warning: `block' might be used uninitialized in this function
cs46xx.c:2867: warning: `amp_voyetra_4294' defined but not used
cs4281.c: In function `cs4281_write_ac97':
cs4281.c:607: warning: `status' might be used uninitialized in this function
plusb.c:985: warning: initialization from incompatible pointer type
mdacon.c:133: warning: `test_mda_b' defined but not used
matroxfb_g450.c:7: warning: `matroxfb_g450_get_reg'
Re: prerelease total nonmodular compile, compiler warnings, linkingerrors
On Tue, 2 Jan 2001, Andreas Bombe wrote: You're then using the ieee1394.o module object which doesn't include the hardware and highlevel drivers. I've sent a patch to Linus already and cc'd the mailing list also. !!! there were heaps of other (not related to ieee1394 stuff ) linking errors. hack was for to see other errors I hope I did not confuse anybody, and those ohter linking errors are attached again with this letter. I'd like to see if by grub can load a nonmodular kernel with everything compiled in (should be about 20MB ? ). Not that it would of any value other than being nice to see. elmer. drivers/sound/sounddrivers.o: In function `cleanup_module': drivers/sound/sounddrivers.o(.text.exit+0xf30): multiple definition of `cleanup_module' drivers/isdn/isdn.a(.text+0x176d4): first defined here ld: Warning: size of symbol `cleanup_module' changed from 105 to 45 in drivers/sound/sounddrivers.o drivers/sound/sounddrivers.o: In function `init_module': drivers/sound/sounddrivers.o(.text.init+0xade0): multiple definition of `init_module' drivers/isdn/isdn.a(.text+0x17670): first defined here ld: Warning: size of symbol `init_module' changed from 97 to 25 in drivers/sound/sounddrivers.o drivers/net/net.o: In function `network_ldisc_init': drivers/net/net.o(.text.init+0x532f): undefined reference to `mkiss_init_ctrl_dev' drivers/net/net.o(.data.init+0x52890): undefined reference to `yam_init' drivers/net/tokenring/tr.a(smctr.o): In function `smctr_reset_adapter': smctr.o(.text+0x392a): undefined reference to `__bad_udelay' smctr.o(.text+0x3938): undefined reference to `__bad_udelay' drivers/ieee1394/ieee1394.o: In function `register_builtin_highlevels': drivers/ieee1394/ieee1394.o(.text+0x27b9): undefined reference to `init_raw1394' drivers/ieee1394/ieee1394.o: In function `register_builtin_lowlevels': drivers/ieee1394/ieee1394.o(.text.init+0x24): undefined reference to `get_lynx_template' drivers/ieee1394/ieee1394.o(.text.init+0x51): undefined reference to `get_ohci_template' drivers/video/video.o: In function `aty_set_pll18818': drivers/video/video.o(.text+0xc121): undefined reference to `__bad_udelay' drivers/video/video.o: In function `init_vgachip': drivers/video/video.o(.text.init+0x41ad): undefined reference to `__bad_udelay' drivers/net/irda/irda.o: In function `toshoboe_gotosleep': drivers/net/irda/irda.o(.text+0x6b81): undefined reference to `__bad_udelay' make: *** [vmlinux] Error 1
Re: Universal debug macros.
well, really, look the other side: We dont make a way to take info away, we just put a lot more into it and give the option to take it away if it is not needed. With this you get your usual amount of debug info plus a way to have lots more. Oh, and one more point: if linux is going to have nonprofessional endusers space comparable to MSWin, then you probably do not want to have every bug report, because these will be stupid anyway, with or without debug info. But if ideological wars stop development in nonsense places, then that day wont come anyway. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Universal debug macros.
On Mon, 27 Nov 2000, Rogier Wolff wrote: > Turns out that people will > prefer to run the "performance" kernel, and they will send in useless > bugreports like "my just hangs" much more often than now. But look at positive side: 1. really few people run development kernels despite the "performance" so it probably will be with nondebug kernels. 2. production kernels get more solid 3. because there could be a lot more debug points in development kernels 4. Distributors are interested in shipping debug-kernels. You see the part that lots of asserts and debug prints may go. I see the advantage, that a lot of them can come, at no cost. Besides, if you want to have some assert anyway, then do not write it with system-wide macro but make your own or mark it as "included allways". Faulty logic. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Universal debug macros.
well, really, look the other side: We dont make a way to take info away, we just put a lot more into it and give the option to take it away if it is not needed. With this you get your usual amount of debug info plus a way to have lots more. Oh, and one more point: if linux is going to have nonprofessional endusers space comparable to MSWin, then you probably do not want to have every bug report, because these will be stupid anyway, with or without debug info. But if ideological wars stop development in nonsense places, then that day wont come anyway. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Universal debug macros.
On Mon, 27 Nov 2000, Rogier Wolff wrote: > Now, how is say "Red Hat" (*) going to ship kernels? Of course they are > going to turn off debugging. Then I'll be stuck with a non-recompiling > user-in-trouble with a non-debugging-enabled kernel. Red Hat will ship two kernels. Well, they actually ship now about 4 ones or something. So they will ship 8. Plus they will ship a script that recompiles kernel without user crawling in documentation. It should be an option in linuxconf: recompile local kernel: debug-nodebug-optimized-localized-nonmodular-server-router-workstation which compiles and installs 2 hardware/situation optimized/configured kernels: debug and production. I am sure company like redhat can ship a little new linuxconf module. Don't worry, people can make good use of multiple options. If you provide orthogonal tools they will provide orthogonal solutions. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Universal debug macros.
On Sun, 26 Nov 2000, Rogier Wolff wrote: > Sure it will slow the driver down a bit, because of all those bit-test > instructions in the driver. If it bothers you, you get to turn it > off. If you are capable of that, you are also capable enough to turn > it back on when neccesary. Now if there would be simple _unified_ system for switching debug code on/off, it would be a real win. That recompilation-capable enduser would not need much knowledge to go "General Setup" or newly created "Optimization" section and switch debugging off/on for _all_ network drivers or ide drivers for example. > The debug asserts that trigger during normal operation are what make > the Linux kernel stable. Problems get spotted at an early > stage. Problems get fixed. Yess Lets say LDBG_* namespace, macros being in general form: LDBG_OPERATION(OPTIMIZATION_LEVEL,SUBSYSTEM,MODULE,ACTION, operation params..) OPERATIONS would be first likely: ASSERT_PRINT, PRINT, ASSERT_PANIC OPTIMIZATION_LEVELs would be first: DEVELOP, ALPHA, BETA, TEST, RELEASE, PRODUCTION, FINETUNED, EMBEDDED SUBSYSTEMS: memory, fs, network, drivers(network, fs,...), divided to about 20 sections or so. MODULE: would be current module ACTION: division inside module : DATA_UP, DATA_DOWN, INIT, CLEANUP, CONFIGURE, ToUserpace,FromUserSpace ... etc. about 15-25 divisions. LDBG_DECLARE_DEBUG_VAR(OPTIMIZATION_LEVEL, SUBSYSTEM, MODULE) woudl declare a global unsigned int subsystem_module_debug = 0 and also sysctl interface to change it. the var would have ACTION bitfields, so user can control debug output runtime for the module. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] removal of "static foo = 0"
On Sun, 26 Nov 2000, Alexander Viro wrote: > I would suggest you to read through the following book and files: > * Kernighan & Pike, "The Practice of Programming" > * Documentation/CodingStyle > * drivers/net/aironet4500_proc.c > and consider, erm, discrepancies. On the second thought, reading K > might also be useful. IOW, no offense, but your C is bad beyond belief. Yep, very true. aironet4500_proc.c is ugly. And is because there is quickly handwirtten something that should have been generic for kernel for some long time, not for every driver-writer to reinvent a wheel. Note that there is something that virtually elliminates need for exact user<->kernel level interfaces and userlevel kerneldata manipulation programs and lots of other maintenance pains. And it does it in quite short sentences. Plus, half of that file is direct repeating of some non-exported kernel functions. But, if you think you can do better, then look into aironet4500_rid.c and handcode it (like real K people do), instead of using aironet4500_proc.c to operate on it. Also, pcmcia/aironet4500_cs.c has lots of ugly parts. Those which are related to stupid masohistic code repetitions due to pcmcia package interface being "cutting edge optimal stupid" The same true is that 2.4 kernel is, in commercial production sense, late for 6 months. And reason being that the codebase and testing becomes unmanageable. And it becomes unmanageable, because some people only read K and try to optimize last bit out of it with using and old book. I'd really propose again: 1. universal debug macros 2. universalt user-kernelspace configuration interface via proc/sys I'd really like C++, but it can be done with C and macros. Some years ago I even managed to write something like stl container withing C and with macros. That was really screwy thing. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] removal of "static foo = 0"
Nice to see again a two cutting-edge-killing opinions.
Every time I really wonder, how such brilliant hackers can be that stupid
that they can not have cake and eat it the same time, and have to scratch
each-others eyes every time.
Use macros.
Kernel has become so big that it really needs universal debugging macros
instead of comments. Comments are waste of brain, if the same
can be explained by long variable names and debug macros.
static Subsystem_module_LocalVariableForThisPurpose;
int Subsytem_module_function_this_and_that(){
DEBUG_ASSERT( Subsystem_module_LocalVariableForThisPurpose == 0 );
DEBUG_ASSERT(MOST_OF_TIME,FS_AREA,MYFS_MODULE, somethingaboutIndodes->node !=
NULL )
}
Those macros would be acceptable if they are unified and taken to
kernel configuration level, so it would be easy to switch them in/out
not only as boolean option but systematically for different levels,
subsystems and modules.
elmer.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] removal of static foo = 0
Nice to see again a two cutting-edge-killing opinions.
Every time I really wonder, how such brilliant hackers can be that stupid
that they can not have cake and eat it the same time, and have to scratch
each-others eyes every time.
Use macros.
Kernel has become so big that it really needs universal debugging macros
instead of comments. Comments are waste of brainfingerpower, if the same
can be explained by long variable names and debug macros.
static Subsystem_module_LocalVariableForThisPurpose;
int Subsytem_module_function_this_and_that(){
DEBUG_ASSERT( Subsystem_module_LocalVariableForThisPurpose == 0 );
DEBUG_ASSERT(MOST_OF_TIME,FS_AREA,MYFS_MODULE, somethingaboutIndodes-node !=
NULL )
}
Those macros would be acceptable if they are unified and taken to
kernel configuration level, so it would be easy to switch them in/out
not only as boolean option but systematically for different levels,
subsystems and modules.
elmer.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] removal of static foo = 0
On Sun, 26 Nov 2000, Alexander Viro wrote: I would suggest you to read through the following book and files: * Kernighan Pike, "The Practice of Programming" * Documentation/CodingStyle * drivers/net/aironet4500_proc.c and consider, erm, discrepancies. On the second thought, reading KR might also be useful. IOW, no offense, but your C is bad beyond belief. Yep, very true. aironet4500_proc.c is ugly. And is because there is quickly handwirtten something that should have been generic for kernel for some long time, not for every driver-writer to reinvent a wheel. Note that there is something that virtually elliminates need for exact user-kernel level interfaces and userlevel kerneldata manipulation programs and lots of other maintenance pains. And it does it in quite short sentences. Plus, half of that file is direct repeating of some non-exported kernel functions. But, if you think you can do better, then look into aironet4500_rid.c and handcode it (like real KR people do), instead of using aironet4500_proc.c to operate on it. Also, pcmcia/aironet4500_cs.c has lots of ugly parts. Those which are related to stupid masohistic code repetitions due to pcmcia package interface being "cutting edge optimal stupid" The same true is that 2.4 kernel is, in commercial production sense, late for 6 months. And reason being that the codebase and testing becomes unmanageable. And it becomes unmanageable, because some people only read KR and try to optimize last bit out of it with using and old book. I'd really propose again: 1. universal debug macros 2. universalt user-kernelspace configuration interface via proc/sys I'd really like C++, but it can be done with C and macros. Some years ago I even managed to write something like stl container withing C and with macros. That was really screwy thing. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Universal debug macros.
On Sun, 26 Nov 2000, Rogier Wolff wrote: Sure it will slow the driver down a bit, because of all those bit-test instructions in the driver. If it bothers you, you get to turn it off. If you are capable of that, you are also capable enough to turn it back on when neccesary. Now if there would be simple _unified_ system for switching debug code on/off, it would be a real win. That recompilation-capable enduser would not need much knowledge to go "General Setup" or newly created "Optimization" section and switch debugging off/on for _all_ network drivers or ide drivers for example. The debug asserts that trigger during normal operation are what make the Linux kernel stable. Problems get spotted at an early stage. Problems get fixed. Yess Lets say LDBG_* namespace, macros being in general form: LDBG_OPERATION(OPTIMIZATION_LEVEL,SUBSYSTEM,MODULE,ACTION, operation params..) OPERATIONS would be first likely: ASSERT_PRINT, PRINT, ASSERT_PANIC OPTIMIZATION_LEVELs would be first: DEVELOP, ALPHA, BETA, TEST, RELEASE, PRODUCTION, FINETUNED, EMBEDDED SUBSYSTEMS: memory, fs, network, drivers(network, fs,...), divided to about 20 sections or so. MODULE: would be current module ACTION: division inside module : DATA_UP, DATA_DOWN, INIT, CLEANUP, CONFIGURE, ToUserpace,FromUserSpace ... etc. about 15-25 divisions. LDBG_DECLARE_DEBUG_VAR(OPTIMIZATION_LEVEL, SUBSYSTEM, MODULE) woudl declare a global unsigned int subsystem_module_debug = 0 and also sysctl interface to change it. the var would have ACTION bitfields, so user can control debug output runtime for the module. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Universal debug macros.
On Mon, 27 Nov 2000, Rogier Wolff wrote: Now, how is say "Red Hat" (*) going to ship kernels? Of course they are going to turn off debugging. Then I'll be stuck with a non-recompiling user-in-trouble with a non-debugging-enabled kernel. Red Hat will ship two kernels. Well, they actually ship now about 4 ones or something. So they will ship 8. Plus they will ship a script that recompiles kernel without user crawling in documentation. It should be an option in linuxconf: recompile local kernel: debug-nodebug-optimized-localized-nonmodular-server-router-workstation which compiles and installs 2 hardware/situation optimized/configured kernels: debug and production. I am sure company like redhat can ship a little new linuxconf module. Don't worry, people can make good use of multiple options. If you provide orthogonal tools they will provide orthogonal solutions. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
reiserfs lockup 2.4.0-t11 SMP.
SMP Dual celeron, 128MB ram, 3.6GB part newly created, untar'ing 1GB newsspool, gave kreiserfsd priority -19, got not very easily reproducible lockup. Sysreq showd kreiserfsd running in state L-TLB or something. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
reiserfs lockup 2.4.0-t11 SMP.
SMP Dual celeron, 128MB ram, 3.6GB part newly created, untar'ing 1GB newsspool, gave kreiserfsd priority -19, got not very easily reproducible lockup. Sysreq showd kreiserfsd running in state L-TLB or something. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
OOPS: 2.4.0-test10 3c509,isapnp,SMP
At mandrake bootup, both isapnp and 3c509 as modules Nov 4 20:29:46 news kernel: isapnp: Scanning for Pnp cards... Nov 4 20:29:46 news kernel: isapnp: No Plug & Play device found Nov 4 20:29:46 news kernel: Unable to handle kernel NULL pointer dereference at virtual address 0070 Nov 4 20:29:46 news kernel: printing eip: Nov 4 20:29:46 news kernel: c89c25a5 Nov 4 20:29:46 news kernel: *pde = Nov 4 20:29:46 news kernel: Oops: 0002 Nov 4 20:29:46 news kernel: CPU:0 Nov 4 20:29:46 news kernel: EIP:0010:[3c509:el3_probe+1349/4832] Nov 4 20:29:46 news kernel: EFLAGS: 00010202 Nov 4 20:29:47 news kernel: eax: bfaf2000 ebx: 0004 ecx: 0070 edx: 0070 Nov 4 20:29:47 news kernel: esi: edi: ebp: 0200 esp: c6b19f08 Nov 4 20:29:47 news kernel: ds: 0018 es: 0018 ss: 0018 Nov 4 20:29:47 news kernel: Process modprobe (pid: 305, stackpage=c6b19000) Nov 4 20:29:47 news kernel: Stack: c89c204e ffea 0070 c6b19f3c c6b19f2c 0003 Nov 4 20:29:47 news PAM_pwdb[398]: (su) session opened for user xfs by (uid=0) Nov 4 20:29:47 news kernel:51ff0001 c89c204e ffea c1044010 c0238cdc bfaf2000 5c5c c89c3295 Nov 4 20:29:47 news kernel: c89c2000 0001 c0119c18 c6b18000 0804b21b 080806e8 bfffe100 Nov 4 20:29:47 news kernel: Call Trace: [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+78/96] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+78/96] [3c509:el3_probe+4661/4832] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+0/96] [sys_init_module+1016/1184] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+72/96] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+-32768/96] Nov 4 20:29:47 news kernel: [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+72/96] [system_call+51/56] [call_spurious_interrupt+28439/33124] Nov 4 20:29:47 news kernel: Code: 89 02 8b 44 24 38 66 89 42 04 8b 4c 24 40 89 69 20 8b 44 24 Nov 4 20:29:47 news kernel: usb.c: registered new driver usbdevfs - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
OOPS: 2.4.0-test10-pre6 around reiserfs 3.6.18
under serious memory shortage, memory hog running and doing random access over 133 MB(128MB ram) and disk output as fast as it could. swap(128M) free = 0M, stable high disk io for long time, then me killing X with -9 , got oops. /home is on reiserfs, which is on raid, which has 5 slices all on same disk (for fun). Nov 4 07:22:32 fw kernel: printing eip: Nov 4 07:22:32 fw kernel: c0133296 Nov 4 07:22:32 fw kernel: *pde = Nov 4 07:22:32 fw kernel: Oops: Nov 4 07:22:32 fw kernel: CPU:0 Nov 4 07:22:32 fw kernel: EIP:0010:[block_read_full_page+14/500] Nov 4 07:22:32 fw kernel: EFLAGS: 00010282 Nov 4 07:22:32 fw kernel: eax: ebx: ecx: c51c1f24 edx: c11d4f2c Nov 4 07:22:32 fw kernel: esi: c11d4f2c edi: c04c157c ebp: esp: c7e75ee4 Nov 4 07:22:32 fw kernel: ds: 0018 es: 0018 ss: 0018 Nov 4 07:22:32 fw kernel: Process bash (pid: 24057, stackpage=c7e75000) Nov 4 07:22:32 fw kernel: Stack: c11d4f2c c04c157c c11d4f2c c04c157c Nov 4 07:22:32 fw kernel: c021f0f0 c7e75f10 c51c1f24 c11d4f58 c11d4f2c c04c157c Nov 4 07:22:32 fw kernel: c0278e40 c015faf7 c11d4f2c c015d4b8 c0124f56 c68d8e40 c11d4f2c Nov 4 07:22:32 fw kernel: Call Trace: [call_apic_timer_interrupt+5/13] [reiserfs_readpage+15/20] [reiserfs_get_block+0/3548] [do_generic_file_read+698/1284] [generic_file_read+91/120] [file_read_actor+0/84] [sys_read+143/196] Nov 4 07:22:32 fw kernel:[system_call+51/56] Nov 4 07:22:32 fw kernel: Code: 8b 40 10 89 44 24 24 c7 44 24 18 00 00 00 00 8b 42 18 a8 01 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
OOPS: 2.4.0-test10-pre6 around reiserfs 3.6.18
under serious memory shortage, memory hog running and doing random access over 133 MB(128MB ram) and disk output as fast as it could. swap(128M) free = 0M, stable high disk io for long time, then me killing X with -9 , got oops. /home is on reiserfs, which is on raid, which has 5 slices all on same disk (for fun). Nov 4 07:22:32 fw kernel: printing eip: Nov 4 07:22:32 fw kernel: c0133296 Nov 4 07:22:32 fw kernel: *pde = Nov 4 07:22:32 fw kernel: Oops: Nov 4 07:22:32 fw kernel: CPU:0 Nov 4 07:22:32 fw kernel: EIP:0010:[block_read_full_page+14/500] Nov 4 07:22:32 fw kernel: EFLAGS: 00010282 Nov 4 07:22:32 fw kernel: eax: ebx: ecx: c51c1f24 edx: c11d4f2c Nov 4 07:22:32 fw kernel: esi: c11d4f2c edi: c04c157c ebp: esp: c7e75ee4 Nov 4 07:22:32 fw kernel: ds: 0018 es: 0018 ss: 0018 Nov 4 07:22:32 fw kernel: Process bash (pid: 24057, stackpage=c7e75000) Nov 4 07:22:32 fw kernel: Stack: c11d4f2c c04c157c c11d4f2c c04c157c Nov 4 07:22:32 fw kernel: c021f0f0 c7e75f10 c51c1f24 c11d4f58 c11d4f2c c04c157c Nov 4 07:22:32 fw kernel: c0278e40 c015faf7 c11d4f2c c015d4b8 c0124f56 c68d8e40 c11d4f2c Nov 4 07:22:32 fw kernel: Call Trace: [call_apic_timer_interrupt+5/13] [reiserfs_readpage+15/20] [reiserfs_get_block+0/3548] [do_generic_file_read+698/1284] [generic_file_read+91/120] [file_read_actor+0/84] [sys_read+143/196] Nov 4 07:22:32 fw kernel:[system_call+51/56] Nov 4 07:22:32 fw kernel: Code: 8b 40 10 89 44 24 24 c7 44 24 18 00 00 00 00 8b 42 18 a8 01 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
OOPS: 2.4.0-test10 3c509,isapnp,SMP
At mandrake bootup, both isapnp and 3c509 as modules Nov 4 20:29:46 news kernel: isapnp: Scanning for Pnp cards... Nov 4 20:29:46 news kernel: isapnp: No Plug Play device found Nov 4 20:29:46 news kernel: Unable to handle kernel NULL pointer dereference at virtual address 0070 Nov 4 20:29:46 news kernel: printing eip: Nov 4 20:29:46 news kernel: c89c25a5 Nov 4 20:29:46 news kernel: *pde = Nov 4 20:29:46 news kernel: Oops: 0002 Nov 4 20:29:46 news kernel: CPU:0 Nov 4 20:29:46 news kernel: EIP:0010:[3c509:el3_probe+1349/4832] Nov 4 20:29:46 news kernel: EFLAGS: 00010202 Nov 4 20:29:47 news kernel: eax: bfaf2000 ebx: 0004 ecx: 0070 edx: 0070 Nov 4 20:29:47 news kernel: esi: edi: ebp: 0200 esp: c6b19f08 Nov 4 20:29:47 news kernel: ds: 0018 es: 0018 ss: 0018 Nov 4 20:29:47 news kernel: Process modprobe (pid: 305, stackpage=c6b19000) Nov 4 20:29:47 news kernel: Stack: c89c204e ffea 0070 c6b19f3c c6b19f2c 0003 Nov 4 20:29:47 news PAM_pwdb[398]: (su) session opened for user xfs by (uid=0) Nov 4 20:29:47 news kernel:51ff0001 c89c204e ffea c1044010 c0238cdc bfaf2000 5c5c c89c3295 Nov 4 20:29:47 news kernel: c89c2000 0001 c0119c18 c6b18000 0804b21b 080806e8 bfffe100 Nov 4 20:29:47 news kernel: Call Trace: [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+78/96] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+78/96] [3c509:el3_probe+4661/4832] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+0/96] [sys_init_module+1016/1184] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+72/96] [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+-32768/96] Nov 4 20:29:47 news kernel: [3c509:__insmod_3c509_O/lib/modules/2.4.0-test10/kernel/drivers/ne+72/96] [system_call+51/56] [call_spurious_interrupt+28439/33124] Nov 4 20:29:47 news kernel: Code: 89 02 8b 44 24 38 66 89 42 04 8b 4c 24 40 89 69 20 8b 44 24 Nov 4 20:29:47 news kernel: usb.c: registered new driver usbdevfs - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
240-test[589] SMP lockups, confirming.
Running here SMP dual celeron, 440BX, 2HD( Maxtor 1G IBM 9G) ISA 3c509, 128MB ram, nfsroot server for 2 clients, X, workstation, Once a day of serious usage computer locks up. Most part sysrq works. Now I had to work on text-console(test9 clean) for some days and got 2 lockups: 1. nfs was in D state if I remember correctly (or R)(sysrq works) but userland is stuck. 2. httpd(not livin on nfs) stop caused then cpu1 to detect lockup (and sysrq didnt work) httpd stop is a weird operation, on some computers with e2compr this caused a logmessage about async io. logs show nothing. It has been so for whole 2.4.0-test[5-9] time(I did not try earlier much, 2.3.99-pre6 was latest, nfs activity seems to be most guilty, however, I can not see any way to debug it when it happens under X. Anyway, the complaining about SMP instablity seems very justified, because, I had to pump 1GB of logs into mysql and 2.2.18-pre9 did the job, so it is very probably not a hardware problem. 2.2.18-pre9 is up and running for 35 hours now, incuding 10 hours mysql with loadavg 3. nfsroot clients are 2.2 and 2.4. If anyone wants me to do some setup-specific stresstests, u are welcome. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
240-test[589] SMP lockups, confirming.
Running here SMP dual celeron, 440BX, 2HD( Maxtor 1G IBM 9G) ISA 3c509, 128MB ram, nfsroot server for 2 clients, X, workstation, Once a day of serious usage computer locks up. Most part sysrq works. Now I had to work on text-console(test9 clean) for some days and got 2 lockups: 1. nfs was in D state if I remember correctly (or R)(sysrq works) but userland is stuck. 2. httpd(not livin on nfs) stop caused then cpu1 to detect lockup (and sysrq didnt work) httpd stop is a weird operation, on some computers with e2compr this caused a logmessage about async io. logs show nothing. It has been so for whole 2.4.0-test[5-9] time(I did not try earlier much, 2.3.99-pre6 was latest, nfs activity seems to be most guilty, however, I can not see any way to debug it when it happens under X. Anyway, the complaining about SMP instablity seems very justified, because, I had to pump 1GB of logs into mysql and 2.2.18-pre9 did the job, so it is very probably not a hardware problem. 2.2.18-pre9 is up and running for 35 hours now, incuding 10 hours mysql with loadavg 3. nfsroot clients are 2.2 and 2.4. If anyone wants me to do some setup-specific stresstests, u are welcome. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: experimental, non-production bits.
Alexander Viro wrote: > ??? You had explicitly enabled the code that was marked "experimental". If > the warning from make config was not enough, how the hell would runtime > warning be more useful? Yeah, you see, If you have about 100 Linux servers to maintain, part your own, part installed by someone else, part standard installation, then this question does make sense too, but is still irrelevant to outcome which is needed "now, after few counted hours and for sure". In such a situation you just want that feature, however principially wrong the whole setup of the question and path of your own actions might be. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
experimental, non-production bits.
Alexander Viro wrote: > How about syslog? Well, I read syslog a lot at home and servers, but customer on-site production computer deep reconfiguration , there is another paradigm - it either works 100% or I dont care. Looks from general talk here, that some kernel people should try servicing some customers on-site with house full of people waiting their stable results within 24 hours and their ass being kicked if they do not succeed. > > and non-production modules. > > So after a year someone would not try to use that stuff on production system. > > Well, the better way is to fix the bloody thing... It is a bit more generic. Just to get an automatic warning for each module which is marked experimental, so that I could rely on that, so I dont need to dig back 100 versions of source code. Another idea: experimental bit. Kernel would have two bits: non-production and experimental. Whenever such a subsystem is activated, those bits remain on until reboot. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: ufs fs at 2.2.x and 2.4.x
Alexander Viro wrote: > Looks like I'm taking care of the UFS for a while. Yes, 2.4 is currently > broken. 2.2.16, migrated FreeBSD to Linux on production system this weekend. Dreamed that I just leave ufs there without copying the stuff. 1. First - it mounted a ufs but showed nothing long time to find the 44bsd senseless option. 2. ok, linux up, samba and MS Windows Nx100MB profiles copying by few users. After a while - MS tells me : media is read-only what a hell. going to command line - the partition IS now readonly 3. -o remount,rw : disk full (not really) It would be nice, if for 2.4 there would be mandatory printk for expermiental and non-production modules. So after a year someone would not try to use that stuff on production system. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
several lockups and performance
Hi, 2.4.0-test8, 2.4.0-test7, 2.4.0-test5 1. bridge. If to set bridge addr on C level after bridge is created, but before any interfaces are ever attached to it, it oopses. The box I am doing it is quite embedded and I am too lazy to set up any serial line debugging or nfs root, so I could say about it at the moment, that it occurs somewhere in bridge code around kmallocing. 2. 3c509 isapnp 3c509 goes up straight, but, with cards configured pnp-mode, with 4 cards, there will be a lockup after the second one. It is probably the generic problem about 3c509 no tolerating foreign interrupts. weird is that sysreq works, but with about 30 sec latency. No oops onscreen. Looks like being in cycle somewhere. P75, 16MB memory, Compaq, PCnet32 onboard, 4x3c509B cards 3. SMP ACPI ? this is on server box. Dont know where it is, but once a day or so, after being away from computer, it wont be happy when I touch it again. Never happened while sitting there, allways when coming back. Sometimes X locks up, but most usually locks whole box. SysReq works, but it allways happnes on graphics console after display coming on and I have typed some letters into X screensaver lock password dialog. 4. performance problem: 10Mbit bridge with very minimalistic kernel is 20% slower with 8MB ram as compared to 16MB ram. Other interesting feature is, that it is 100% slower at start and after some minutes of run, comes to 20%... ultimately weird. kernel is 450kB compressed, has vfat as root fs, and whatever else possibly disabled. TCP is supported but real minimalistic. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
several lockups and performance
Hi, 2.4.0-test8, 2.4.0-test7, 2.4.0-test5 1. bridge. If to set bridge addr on C level after bridge is created, but before any interfaces are ever attached to it, it oopses. The box I am doing it is quite embedded and I am too lazy to set up any serial line debugging or nfs root, so I could say about it at the moment, that it occurs somewhere in bridge code around kmallocing. 2. 3c509 isapnp 3c509 goes up straight, but, with cards configured pnp-mode, with 4 cards, there will be a lockup after the second one. It is probably the generic problem about 3c509 no tolerating foreign interrupts. weird is that sysreq works, but with about 30 sec latency. No oops onscreen. Looks like being in cycle somewhere. P75, 16MB memory, Compaq, PCnet32 onboard, 4x3c509B cards 3. SMP ACPI ? this is on server box. Dont know where it is, but once a day or so, after being away from computer, it wont be happy when I touch it again. Never happened while sitting there, allways when coming back. Sometimes X locks up, but most usually locks whole box. SysReq works, but it allways happnes on graphics console after display coming on and I have typed some letters into X screensaver lock password dialog. 4. performance problem: 10Mbit bridge with very minimalistic kernel is 20% slower with 8MB ram as compared to 16MB ram. Other interesting feature is, that it is 100% slower at start and after some minutes of run, comes to 20%... ultimately weird. kernel is 450kB compressed, has vfat as root fs, and whatever else possibly disabled. TCP is supported but real minimalistic. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
from binary towards source level scalability
Martin Dalecki wrote: > Elmer Joandi wrote: > > strict standard template for linux kernel functions: > > INLINE(context,level,for_speed, fixed) returntype functionname > > Please have a tought look at the floppy tape streamer driver to see why > this is a BAD IDEA. Couldnt see much else than half of it being implemented. 1. My point is more in source level scalability, whatever techincal way it is done. There is an natural potential for open source software which is not quite completely used. RedHat still ships i386 binaries which run 30% slower( and are still with debug info default on) than Mandrakes i586. Neither of them offers on-install automatic recompilation. Before imitation of commercial binary vendors they could think about using their native potential. 2. About macroplay, if you dislike it: if few macros would be used all over the code, it would be very clear, cleaner than without them. Ftape driver trace macros have some strings in them :), if strings are forced , then someone gets to comment his code a lot more. It is just a matter of getting used to style. 3. Lets assume for a while, that for every container(array, hash, btree) for which there is currently runtime dynamicly changeable default size or other parameter there would be a compile-time option to turn it static and compiled in with both intentions: a) to keep it small an operational on 386sx25(your cellular phone) and b) fast on NxGB memory top-tuned SMP superbox. i.e. instead of #define MY_HASH_SIZE 123 or #define MY_HASH_SIZE ((whatever)->size) there would be CONTAINER_SIZE(MY, 10, user_min, 40, 123, 1200, user_max, 1) and it would by default compile to dynamically changeable 123, but could also be user-specified size or developer-specified minimum statically compiled in. And, would not go out of developer specified reasonable values. 4. symbol, printk and other text-based information, inline regulation... all of those disputes wheter to have or not to have something new like that in kernel could just end up being configuration options. PRINTK(subsystem, module, level, "my networking whatever") could be elliminated by configuration option like: not verbose for subsystem=networking. Top dream would be to have enduser to specify his intentions(file,webserving, desktop,development) and then a program gathering memory and cpu speed-size info, generating proper kernel and libraries config and compiling it static then and doing some stress test just after that.That could be standard procedure for linux installation :) Instead of compiling it by hand and then getting 30% faster. On some places source code must not be an art, but just systematic structure or a bunch of data in table. Currently, for example, inlining is an artistic act. And debugging. And /proc interface. The other way would be to make it a configure-time option and/or flexibly configurable/dynamicly changeable. Then it would be compile-time art for distributors. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
kernel debugging
> understanding the
> > underlying principles and the code.
Speaking about that, I have been long time dreaming
about following strict standard template for linux kernel functions:
(macroplay intended)
--
INLINE(context,level,for_speed, fixed) returntype functionname
(PARM(parameters)){
DEBUG(ENTRY,SUBSYSTEM, module, functionname);
/* function body starts */
do something
do something
on success goto ok;
on failure goto failure_reason_name;
/* function body ends */
ok:
on ok do something
DEBUG(EXIT,SUBSYSTEM, module functionname, OK);
exit or return whatever
failure_reason_name;
on this reason do something
DEBUG(EXIT,SUBSYSTEM, module, functionname, reason_name);
exit or return whatever
failure_reason_name1;
on this reason do something
DEBUG(EXIT,SUBSYSTEM, module, functionname, reason_name1);
exit or return whatever
};
--
You see, you try to catch somebody by following their steps.
Instead, try to enchanche your own advantages.
Those two macros: DEBUG and INLINE would let the code be
generated as effective as possible and as debuggable as possible.
strict rule that all returns must be explicitly goto'd to the end would
make code also much easier to read and would force people to
formalize all failure reasons.
Closed source people have only one way to go, it is to make
kernel binary fully debuggable.
Imagine kernel configuration options: which subsystems
to optimize for speeed and which for size, for which to have
debug options turned on, etc. Different debug macros
for different situations - debugging, logging, timing.
All printk's made macros and configure-time option to dump
them all... that way the kernel may run in 2MB 386sx16 again
for embedded devices and have lots of inlines for 2MB cache
Xeon processors.
For to stay competitive, linux should less try to repeat others
tricks and more try to advance those, whitch are naturally its own.
Essentially the trick that everyone may recompile the kernel.
elmer.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
kernel debugging
understanding the
underlying principles and the code.
Speaking about that, I have been long time dreaming
about following strict standard template for linux kernel functions:
(macroplay intended)
--
INLINE(context,level,for_speed, fixed) returntype functionname
(PARM(parameters)){
DEBUG(ENTRY,SUBSYSTEM, module, functionname);
/* function body starts */
do something
do something
on success goto ok;
on failure goto failure_reason_name;
/* function body ends */
ok:
on ok do something
DEBUG(EXIT,SUBSYSTEM, module functionname, OK);
exit or return whatever
failure_reason_name;
on this reason do something
DEBUG(EXIT,SUBSYSTEM, module, functionname, reason_name);
exit or return whatever
failure_reason_name1;
on this reason do something
DEBUG(EXIT,SUBSYSTEM, module, functionname, reason_name1);
exit or return whatever
};
--
You see, you try to catch somebody by following their steps.
Instead, try to enchanche your own advantages.
Those two macros: DEBUG and INLINE would let the code be
generated as effective as possible and as debuggable as possible.
strict rule that all returns must be explicitly goto'd to the end would
make code also much easier to read and would force people to
formalize all failure reasons.
Closed source people have only one way to go, it is to make
kernel binary fully debuggable.
Imagine kernel configuration options: which subsystems
to optimize for speeed and which for size, for which to have
debug options turned on, etc. Different debug macros
for different situations - debugging, logging, timing.
All printk's made macros and configure-time option to dump
them all... that way the kernel may run in 2MB 386sx16 again
for embedded devices and have lots of inlines for 2MB cache
Xeon processors.
For to stay competitive, linux should less try to repeat others
tricks and more try to advance those, whitch are naturally its own.
Essentially the trick that everyone may recompile the kernel.
elmer.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
Re: GPL violations: make it harder
Andre Hedrick wrote: > > There is overproduction of generic-purpose software in world and > > of course lots of companies are going to bancrupt soon, but if you > > continue this way, GPL is going the same way... > > Do not follow the thought, sorry. The Novell stuff. Sorry to say, but who needs it in real world ? 99% of consumers are happy with NT and are happy with samba. See the generated revenue per marketshare as divided between small-average-large bussinesses and you see, that there is very little place for cutting edge software for file sharing. People just solve software problems in hardware. I have constantly problems getting Pentium 100Mhz to work full-loaded and if I get, I replace it with Pentium 266. For last two years there is more hardware and software power that people can make use of. Interfaces and protocols incompatibility and version instability is the only thing that keeps the money coming from generic-purpose software(operating systems), otherwise everybody just would use linux and samba on top of it, however bad and inefective it may sound from non-linux benchmarkers mouth. If fileserver becomes slow, they replace it and take old to home for kids to play. There is more general conclusion from that: 1. times of revolutionary rapid changes in generic purpose(OS) software is over and there will be one or two (binary and open source) generic purpose software distributions in the world as fast as interoperability will be achieved. 2. It also means that besides MS, there will be no one who manages to charge money for generic-prupose software and if full interoprerability is achieved, neither do they. 3. A lots of programmers get fired and go back to roots - farming work for example :) 4. Innovation currently generates new values, but public can not make use of it in reasonable ways and therefore it will not be rewarded. GPL(linux) may also go bancrupt in sense that if MS drops NT server price down to 70$ and copies all new ideas, linux will be for freaks and paranoid and MS can earn revenue from applications. However, there is another possible way: 100% interoperability is achieved and however ineffective the linux solution is, people will use that. But the worst nightmare for linux would be if MS goes open source with its OS. Because then US foreign affairs department will enforce strict copyrigth laws and forging and stealing software will be made impossible and all new ideas would be patented and bought up. You see, if binary software is banned, that makes possible way more deep monopoly schemes. And then it would be either GPL or MS (patent-or-legal-rights-buy-up-office) who owns the software of the world. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: GPL violations: make it harder
Andre Hedrick wrote: There is overproduction of generic-purpose software in world and of course lots of companies are going to bancrupt soon, but if you continue this way, GPL is going the same way... Do not follow the thought, sorry. The Novell stuff. Sorry to say, but who needs it in real world ? 99% of consumers are happy with NT and are happy with samba. See the generated revenue per marketshare as divided between small-average-large bussinesses and you see, that there is very little place for cutting edge software for file sharing. People just solve software problems in hardware. I have constantly problems getting Pentium 100Mhz to work full-loaded and if I get, I replace it with Pentium 266. For last two years there is more hardware and software power that people can make use of. Interfaces and protocols incompatibility and version instability is the only thing that keeps the money coming from generic-purpose software(operating systems), otherwise everybody just would use linux and samba on top of it, however bad and inefective it may sound from non-linux benchmarkers mouth. If fileserver becomes slow, they replace it and take old to home for kids to play. There is more general conclusion from that: 1. times of revolutionary rapid changes in generic purpose(OS) software is over and there will be one or two (binary and open source) generic purpose software distributions in the world as fast as interoperability will be achieved. 2. It also means that besides MS, there will be no one who manages to charge money for generic-prupose software and if full interoprerability is achieved, neither do they. 3. A lots of programmers get fired and go back to roots - farming work for example :) 4. Innovation currently generates new values, but public can not make use of it in reasonable ways and therefore it will not be rewarded. GPL(linux) may also go bancrupt in sense that if MS drops NT server price down to 70$ and copies all new ideas, linux will be for freaks and paranoid and MS can earn revenue from applications. However, there is another possible way: 100% interoperability is achieved and however ineffective the linux solution is, people will use that. But the worst nightmare for linux would be if MS goes open source with its OS. Because then US foreign affairs department will enforce strict copyrigth laws and forging and stealing software will be made impossible and all new ideas would be patented and bought up. You see, if binary software is banned, that makes possible way more deep monopoly schemes. And then it would be either GPL or MS (patent-or-legal-rights-buy-up-office) who owns the software of the world. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
GPL violations: make it harder
Andre Hedrick wrote: > My and co-worker's code for doing full taskfile access under linux was > rejected here but is being used in MicroSoft Whistler 2001. They are > quick to grab the very best of Linux and adopt it for their own. ? You mean that they did it illegally and you can show a way to prove it ? If it is true and cant be justified in court, then it can be justified trough good reverse-engineering programs, by lynching. I have been following it in this list for longer - you cut the thing you are sitting on when telling people who say "I saw GNU licence violation" to not to bother or turn to other organization... There is overproduction of generic-purpose software in world and of course lots of companies are going to bancrupt soon, but if you continue this way, GPL is going the same way... Nobody really needs cutting edge software anymore, people need working software. It is very generic change - after every revolution there come merchants who take over. I'd prefer it to be open source. Then it is fair. Of course it wont be any giveaway there then. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: www.crucial.com won't talk to 2.4.0-test7 system
Mark Hahn wrote: > I'm curious to know what you mean. That is your websurfing session. But try on some 100+ size network to set some hops go trough a tunnel... the variety of behaviour and reasons for it are way larger there. I mean a network that is not completely under your control. neither at client neither at servers end... elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
GPL violations: make it harder
Andre Hedrick wrote: My and co-worker's code for doing full taskfile access under linux was rejected here but is being used in MicroSoft Whistler 2001. They are quick to grab the very best of Linux and adopt it for their own. ? You mean that they did it illegally and you can show a way to prove it ? If it is true and cant be justified in court, then it can be justified trough good reverse-engineering programs, by lynching. I have been following it in this list for longer - you cut the thing you are sitting on when telling people who say "I saw GNU licence violation" to not to bother or turn to other organization... There is overproduction of generic-purpose software in world and of course lots of companies are going to bancrupt soon, but if you continue this way, GPL is going the same way... Nobody really needs cutting edge software anymore, people need working software. It is very generic change - after every revolution there come merchants who take over. I'd prefer it to be open source. Then it is fair. Of course it wont be any giveaway there then. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: www.crucial.com won't talk to 2.4.0-test7 system
Matti Aarnio wrote: > From personal experience I can say that even cisco router training > includes example of: "block ALL of ICMP", which of course makes > TCP PMTU discovery non-functional. Yeah, there is the general problem with such a stuff: if something becomes de facto standard it messes up de jure standard. And, if that hurts more people who use de jure than it hurts the people who misconfigure, then it is just plain impossible in practice to use de jure standard and/or change the "de facto" situation. The real solution is to change "the hurting ratio". Changing MTU from 1500 to 1498 on several US and US<-> Europe backbones would create major mess which would be painful enough to change the situation with small MTU for example. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: www.crucial.com won't talk to 2.4.0-test7 system
Alan Cox wrote: > > There are a -lot- of large sites that give us issues like this. > > So mail lots of people. Cisco are I think now aware that their firewall > products dont handle ECN correctly but others might not be. > > Or wait until more vendors roll out ECN There is another big problem like that... tunnels actually do not work on todays real internet... MTU 1500 is so much a standard that it starts killing tunnels. MTU < 1500 is not a working solution today thanks to (mostly linux based ? ) broken firewalls elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: www.crucial.com won't talk to 2.4.0-test7 system
Matti Aarnio wrote: From personal experience I can say that even cisco router training includes example of: "block ALL of ICMP", which of course makes TCP PMTU discovery non-functional. Yeah, there is the general problem with such a stuff: if something becomes de facto standard it messes up de jure standard. And, if that hurts more people who use de jure than it hurts the people who misconfigure, then it is just plain impossible in practice to use de jure standard and/or change the "de facto" situation. The real solution is to change "the hurting ratio". Changing MTU from 1500 to 1498 on several US and US- Europe backbones would create major mess which would be painful enough to change the situation with small MTU for example. elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: www.crucial.com won't talk to 2.4.0-test7 system
Alan Cox wrote: There are a -lot- of large sites that give us issues like this. So mail lots of people. Cisco are I think now aware that their firewall products dont handle ECN correctly but others might not be. Or wait until more vendors roll out ECN There is another big problem like that... tunnels actually do not work on todays real internet... MTU 1500 is so much a standard that it starts killing tunnels. MTU 1500 is not a working solution today thanks to (mostly linux based ? ) broken firewalls elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: If loadable modules are covered by Linux GPL?
Mike Coleman wrote: > Microsoft is about taking control > away from users and giving it to vendors, and the GPL is, to a degree, the > reverse. True, but for very specific stuff there is no good to go to edges with that. Very much better solution would be if company defines a good generic interface for devices of such-and-such type, commits it to GPL and then uses binary modules. If drivers could run outside ring 0, it would be enev nicer. Hiding driver source back is most part useless. I have here hacked "lotsa of code", legally, "for compatiblity purposes", as the law says. Of course, one side of it was that I hurt companies bussiness. Any move has its positive and negative consequences. The other side was, that they remained in profits, as I 1. discovered a lots of bugs in it(hw, firmware), including several DoS. 2. put it to run 50% faster 3. made solution more cost-effective. Most importantly: 4. created trust - GPL plays against vendors and for public, but vendors who go with it, gain trust by public that vendor probably will not play against public in future. Lots of people have used to choose hardware if it has linux drivers. Not because they would need it right away, but because openess generates trust - about that in next driver version some feature will not be turned off for M$ platform, just for to enforce public to buy additional 15$ pseudo-module and have lots of PiA(pain in *ss) with it. Finally, hiding software interfaces because there is money invested in it is usually result of incompetent bussiness people who understand only movement of dollars. For hardware piece there should be most of difference in hardware... And if software does not make much difference then is no point in hiding it as it can be used only with that hardware. They question to ask yourself: is the loss for gain for competitors from your work higher than gain from being open source, first, fast and trusted by public ? elmer. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
