ChangeSet@1.1303, 2003-12-16 15:43:16-02:00, daniela@cyclades.com [PATCH] Add Cyclades PC300 driver ChangeSet@1.1302, 2003-12-16 13:59:16-02:00, t-kochi@bq.jp.nec.com [PATCH] duplicate PID fix Hello Marcelo, This fix was sent to lkml in April, and was merged to -ac tree, but is not merged in the main tree yet. Please consider taking this in. Without this, duplicate pids can be allocated, which will make one of them unkillable (signals are deliverd to only one of them), and this can be exploitable (I don't know for sure, but maybe, like brk() ;) This situation happens only when all pid space is full. Usually, users cannot fork processes more than 32768 (PID_MAX), but default user limit of max processes can be more than PID_MAX on large memory machines such as 64bit platforms (although it's adjustable by threads-max sysctl). This patch modifies common code and affects all architectures, but modifies code only executed when no pid is available, so it doesn't hurt any normal path anyway. (BTW, once I sent this patch to Rusty's Trivial patch monkey, but his reply was non-trivial, and he also said this is scary ;) The details are described below: In get_pid(), an available pid is searched through all task_structs even when there is no available pid. If a new pid is not available, the kernel exits the loop with static variable 'next_safe' untouched, which usually is no problem. spin_lock(&lastpid_lock); beginpid = last_pid; if((++last_pid) & 0xffff8000) { last_pid = 300; /* Skip daemons etc. */ goto inside; } if(last_pid >= next_safe) { inside: next_safe = PID_MAX; read_lock(&tasklist_lock); repeat: for_each_task(p) { if(p->pid == last_pid || p->pgrp == last_pid || p->tgid == last_pid || p->session == last_pid) { <= (A) if(++last_pid >= next_safe) { <= (B) if(last_pid & 0xffff8000) last_pid = 300; next_safe = PID_MAX; } if(unlikely(last_pid == beginpid)) <= (C) goto nomorepids; goto repeat; } if(p->pid > last_pid && next_safe > p->pid) next_safe = p->pid; if(p->pgrp > last_pid && next_safe > p->pgrp) next_safe = p->pgrp; if(p->tgid > last_pid && next_safe > p->tgid) next_safe = p->tgid; if(p->session > last_pid && next_safe > p->session) next_safe = p->session; } In a rare case, both (B) and (C) can be true and then, next_safe will remain PID_MAX (32768). If that happens, following get_pid() will always succeed until last_pid reaches 32768 and there may be many duplicate pids. For example, this happens when * PID space are full (300-32767 are all occupied) * the last pid allocated is 10000 * task list chain is like: ...(pids < 9999), 9999, ...(pids 300~9998, 10001~32767)... , 10000 The loop starts searching an available pid with beginpid=10000 and last_pid=10001. last_pid is incremented until it gets PID_MAX and then wraps around to 300, then is incremented again. At the point that p->pid=9999 is found in tasklist (condition (A)), last_pid = 9999 next_safe <= 9998 therefore condition (B) is true, and then last_pid = 10000 next_safe = PID_MAX and then, condition (C) is also true, and exits the loop. To protect this case is simple; when the condition (C) is true, set next_safe to 0 or any safe value to guarantee that a free pid will be searched through next time. Thanks, ChangeSet@1.1301, 2003-12-16 11:30:53-02:00, khali@linux-fr.org [PATCH] Remove sa1100 unused header This small patch fixes the pcmcia sa1100 driver including the i2c.h header, while it doesn't make any use of it. I found that while working on my various i2c patches. Please apply. ChangeSet@1.1300, 2003-12-16 11:30:20-02:00, khali@linux-fr.org [PATCH] i2c printk fixes This patch fixes many (122) printk calls missing their KERN_* constant. There also are some other trivial fixes that were passing by (indentation changes and string changes). Please apply. ChangeSet@1.1299, 2003-12-16 11:29:46-02:00, khali@linux-fr.org [PATCH] i2c documentation This patch updates the i2c documentation files to what we have in our repository. There are various fixes, and a more complete protocol summary. Please apply. ChangeSet@1.1296.1.2, 2003-12-16 10:44:34-02:00, khali@linux-fr.org [PATCH] i2c core cleanups This patch does some trivial changes to a few i2c drivers. The changes are only white space and comment changes, and line reordering. There are also two simple changes to i2c-id.h to keep it in line with the ones in our repository and Linux 2.6.0-test11. These changes should make subsequent patches more readable. Please apply. ChangeSet@1.1296.1.1, 2003-12-15 22:55:45-02:00, marcelo@logos.cnet Ernie Petrides: Fix __alloc_pages PF_MEMDIE handling ChangeSet@1.1297, 2003-12-15 10:21:37-08:00, maxk@qualcomm.com Merge bk://linux-mh.bkbits.net/bluetooth-2.4 into qualcomm.com:/home/kernel/bt-2.4 ChangeSet@1.1270.3.13, 2003-12-15 17:29:07+11:00, linas@us.ibm.com [PPC64] Use the correct functions to access user memory in proc file routines. ChangeSet@1.1270.4.10, 2003-12-15 07:25:08+01:00, marcel@holtmann.org [Bluetooth] Cosmetic cleanup of the HCI USB driver This patch removes a left over entry in the Configure.help file that is no longer used and corrects the config name of the SCO audio support. ChangeSet@1.1270.3.12, 2003-12-15 17:08:30+11:00, engebret@us.ibm.com [PPC64] Fix setting/clearing of the RI (recoverable interrupt) bit. RI needs to be clear whenever taking an exception or interrupt would lead to loss of state (i.e. when the SRR0/1 registers are live) and set at other times. ChangeSet@1.1270.4.9, 2003-12-15 07:07:25+01:00, marcel@holtmann.org [Bluetooth] Remove copy of sockfd_lookup() The function sockfd_lookup() is now exported and so there is no need to keep our own copy. ChangeSet@1.1270.3.11, 2003-12-15 16:55:24+11:00, paulus@quango.ozlabs.ibm.com [PPC64] Fix and extend 32-bit syscall emulation code. This adds a 32-bit ustat, and fixes the quotactl emulation as well as various ioctls (including ethtool and LVM ioctls). ChangeSet@1.1270.3.10, 2003-12-15 16:42:19+11:00, paulus@quango.ozlabs.ibm.com [PPC64] Fix compile error in arch/ppc64/kernel/pmc.c. ChangeSet@1.1270.4.8, 2003-12-15 06:35:41+01:00, marcel@holtmann.org [Bluetooth] Support inquiry results with RSSI The Bluetooth 1.2 specification defines a new event for returning inquiry results. This patch converts this event into the standard event and add its information to the inquiry cache. ChangeSet@1.1270.3.9, 2003-12-15 15:44:43+11:00, engebret@us.ibm.com [PPC64] Fix save_flags/restore_flags on iSeries. ChangeSet@1.1270.3.8, 2003-12-15 15:41:23+11:00, paulus@quango.ozlabs.ibm.com [PPC64] Fix stack expansion bug. Previously, any access to an address in the unmapped area below the stack would cause the stack to be expanded. Now we check whether the address is either at or above the stack pointer, or would be if the instruction has succeeded (for the store-with-update instructions). If not we don't expand the stack. ChangeSet@1.1270.3.7, 2003-12-15 15:29:43+11:00, engebret@us.ibm.com [PPC64] Boot wrapper improvements. ChangeSet@1.1270.4.7, 2003-12-15 05:28:59+01:00, marcel@holtmann.org [Bluetooth] Update HCI security filter This patch is a very big update to the HCI security filter. It consists of every forgotten commands/events from Bluetooth 1.1 and also the new parts from the Bluetooth 1.2 specification. ChangeSet@1.1270.3.6, 2003-12-15 15:18:05+11:00, engebret@us.ibm.com [PPC64] Makefile fixes: use $(CC), -fno-zero-initialized-in-bss ChangeSet@1.1270.3.5, 2003-12-15 15:14:57+11:00, olof@austin.ibm.com [PPC64] Better default port, irq and flag settings for ttyS2/3 on pSeries machines. ChangeSet@1.1270.3.4, 2003-12-15 15:00:12+11:00, paulus@quango.ozlabs.ibm.com [PPC64] Make sure the user stack pointer is 16-byte aligned on signal delivery. This is required by the ABI. ChangeSet@1.1270.3.3, 2003-12-15 14:36:48+11:00, olof@austin.ibm.com [PPC64] Make sure we don't take a segment miss in a critical region. Backport of 2.5 fix: Don't throw out the current stack segment in make_slbe. Since the entry is still in the ERAT, and the ERAT entry might be expired at any time, we might hit an SLB fault at critical sections where we can't handle it. One subtle difference between 2.4 and 2.5 in this aspect: We still need to do the slbie in 2.4, since this code might be running on the interrupt stack and we might cast out the kernel stack segment. Without the slbie the kernel stack segment might still be in the ERAT and we'd be back at the same problem. ChangeSet@1.1270.3.2, 2003-12-15 14:20:20+11:00, olof@austin.ibm.com [PPC64] Fix smp_call_function so we don't crash if an IPI is very late. ChangeSet@1.1294, 2003-12-14 21:54:36-02:00, dhinds@sonic.net [PATCH] PATCH: fix logging levels for yenta socket driver Hi Marcelo, This patch fixes printk() calls in the yenta driver to specify logging levels. -- Dave ChangeSet@1.1293, 2003-12-14 10:05:16-02:00, marcelo@logos.cnet Merge bk://linuxusb.bkbits.net/usb-devel-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1270.1.7, 2003-12-13 21:52:29-08:00, davem@nuts.ninka.net [SPARC32]: Fix build after asm/system.h include was added to linux/spinlock.h - Do not include asm/oplib.h in system.h - Remove system.h's halt() macro, prom_halt() is equivalent so replace the two users of halt() with that. - Put 'romvec' extern decl where it belongs, in openprom.h Based upon a patch from Rob Radez. ChangeSet@1.1270.4.6, 2003-12-13 13:54:15-02:00, dhinds@sonic.net [PATCH] PATCH: fix PCMCIA memory resource management bug Hi Marcelo, o Fix PCMCIA memory resource management list traversal bug. -- Dave ChangeSet@1.1270.4.5, 2003-12-13 13:53:41-02:00, dhinds@sonic.net [PATCH] PATCH: fix PCMCIA interrupt allocation Hi Marcelo, Please apply: o Fix ISA/PCI PCMCIA interrupt allocation priorities. -- Dave ChangeSet@1.1270.4.4, 2003-12-13 13:53:30-02:00, dhinds@sonic.net [PATCH] Re: UPD: "do_IRQ: near stack overflow" when inserting CF disk On Fri, Dec 12, 2003 at 04:52:22PM +0100, Roger Larsson wrote: > > Summary: > Some routines allocate to much stack: validate_cis? > several subtypes in the union cisparse_t are quite > big (>250 bytes) > This eats stack (to much?) or only close enough to > trigger 'stackdefer' which might stress the drivers SMPness? Marcelo, Please apply this patch. It reduces kernel stack usage for the PCMCIA CIS processing code. -- Dave ChangeSet@1.1270.4.3, 2003-12-13 11:20:08-02:00, jt@bougret.hpl.hp.com [PATCH] IrDA kernel log buster Hi Marcelo, I just ran 2.4.23, and after a few min the disk reached 100% capacity. A quick check lead to to oversized kernel log, and to the following changeset : http://linux.bkbits.net:8080/linux-2.4/cset@1.1136.23.2?nav=index.html|ChangeSet@-12w Patch to fix this problem is attached below, I've just backported the proper fixes from 2.5.X into 2.4.X. Probably this person did too much Python, but in C you need braces around multiple statements part of the same branch, so the second printf was always executed even when logging was disabled. I also don't understand why this person didn't decided to backport the 2.5.X fix. I'm also bit surprised that this kind of patch went into the kernel behind my back, because I though that freeze meant not accepting untested patch from random hacker. Have fun... Jean ChangeSet@1.1270.4.2, 2003-12-13 10:56:38-02:00, marcelo@logos.cnet Merge http://linux-mh.bkbits.net/uinput-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1270.4.1, 2003-12-13 10:50:56-02:00, ralf@linux-mips.org [PATCH] Undo accidental deletion of MWave config bits ===== drivers/char/Config.in 1.61 vs edited ===== ChangeSet@1.1291, 2003-12-12 14:04:18-08:00, luca.risolia@studio.unibo.it [PATCH] USB: W996[87]CF driver update ChangeSet@1.1290, 2003-12-12 13:13:46-08:00, greg@kroah.com [PATCH] USB: add support for Sony UX50 device to visor driver Thanks to Ralf Dietrich for the information. ChangeSet@1.1289, 2003-12-12 12:18:04-08:00, marr@flex.com [PATCH] USB: MCT-U232 Patch for cts Brief Patch Description: Fix a problem in the 'mct_u232' driver whereby output data gets held up in the USB/RS-232 adapter for RS-232 devices which don't assert the 'CTS' signal. Background: The Belkin F5U109 is a 9-pin USB/RS-232 adapter that is supported by the existing 'mct_u232' kernel module.  Recently, I've been testing it under the 2.4.22 (Slackware 9.1) kernel and the 2.6.0-test9 kernel. I've connected a Garmin 'GPS35 TracPak' GPS receiver (RS-232 interface) and an ordinary RS-232 external modem to my PC's USB port via the Belkin F5U109 adapter. Problem: Although _reads_ from either of the RS-232 devices mentioned above work fine via the Belkin adapter, _writes_ to the GPS receiver are not being seen by the GPS.  Writes to the modem, however, work perfectly. Aside: The 'Linux USB Users' archives show that at least one other person (circa May 2002) had the exact same problem I'm having, but it sounds like no solution was ever determined because the person in question just bought a different USB/RS-232 adapter. Investigation: Using the 'seyon' terminal emulator in Linux and a crude hardware RS-232 "breakout box" that I hacked together, I've determined that the problem is related to the RTS/CTS RS-232 hardware handshaking. After further investigation, I've concluded that RS-232 devices which do not assert the 'Clear To Send' ('CTS') signal prevent the Belkin F5U109 adapter from transmitting data to the RS-232 device when the current (version 1.1) 'mct_u232' module is used. The data gets "queued up" (up to a point -- 16 bytes, I think) in the adapter but never transmitted. Since this GPS receiver works perfectly (reads and writes) when connected to a PC running W98se using the same Belkin adapter and the Belkin-supplied Windows driver, the Linux driver became suspect. After some testing with SniffUSB, I found that the Windows driver sends a couple of unique undocumented USB 'device requests' that the Linux driver does not. As it turns out, the second of those 2 requests is critical in making the adapter transmit data to a device which doesn't assert 'CTS'. For completeness, the Windows driver in use was determined from the 'Device Manager', 'Driver File Details' page:    U2SPORT.VXD    Provider: Magic Control Technology    File version: 1.21P.0104 for Win98/Me Solution: My patch adds the 2 missing USB 'device request' commands right after a baud-change command. This mimics the operation of the W98 driver. Unfortunately, after much testing, I found no other operation (besides a baud-change request) under Windows that triggers either of these 2 'device request' commands. This makes it impossible to fully document the behavior of these requests, but I've made entries for them alongside the others in the 'mct_u232.h' file. Purely for clarity, the patch also modifies various comments in 'mct_u232.h', mostly to reflect proper sizes of the various 'USB Device Request' fields per the USB 1.1 specification. The patch also updates the version number of the driver, corrects a minor typographical error, and documents a difference in the length of the data in a 'baud rate change' command for certain adapters which use a coded baud-rate rather than the conventional RS-232 baud rate divisor. ChangeSet@1.1288, 2003-12-12 11:49:05-08:00, zaitcev@redhat.com [PATCH] USB: Backport of printer 2.6=>2.4 > Pete, I tested your backport + this change above on 2.4.23 and it works > well on my HP psc 2110. > I did have to apply part of the patch by hand, so I rediffed it against > 2.4.23. Patch is below. OK, either way is good. However, I don't understand why it didn't apply for you. I tried this way and that, it all matches. The patch you posted applies cleanly, too. Hmmm.... I think the attached should be perfect. I do not much like the syntax of "while ( 1==1 )", but if 2.6 has it... ChangeSet@1.1287, 2003-12-12 10:18:22-08:00, tchen@on-go.com [PATCH] USB: fix io_edgeport driver alignment issues. ChangeSet@1.1286, 2003-12-12 10:18:02-08:00, tchen@on-go.com [PATCH] USB: fix bug when errors happen in ioedgeport driver ChangeSet@1.1285, 2003-12-12 10:02:22-08:00, greg@kroah.com [PATCH] USB: add support for another pl2303 device Info came from John Zhuge ChangeSet@1.1270.3.1, 2003-12-12 16:48:12+11:00, engebret@us.ibm.com PPC64: Use r13 to point to a per-processor data area rather than current. ChangeSet@1.1284, 2003-12-11 17:35:40-08:00, stern@rowland.harvard.edu [PATCH] USB storage: Unusual_devs.h addition This patch adds to unusual_devs.h an entry reported by Andries Brouwer and it moves another entry to the correct position in the numerical ordering. ChangeSet@1.1283, 2003-12-11 17:33:25-08:00, _nessuno_@katamail.com [PATCH] USB storage: Medion 6047 Digital Camera ...a patch for the "Medion 6047 Digital Camera" *** a/drivers/usb/storage/unusual_devs.h Sun Nov 23 22:31:51 2003 ChangeSet@1.1282, 2003-12-11 17:04:41-08:00, berentsen@sent5.uni-duisburg.de [PATCH] USB storage: Minolta Dimage S414 usb patch here I submitt you the vendor/id patch for the Minolta Dimage S414 Camera, which runs fine with the usb under linux. cat /proc/bus/usb/device -> ChangeSet@1.1281, 2003-12-11 17:02:03-08:00, stern@rowland.harvard.edu [PATCH] USB storage: Another unusual_devs.h update On Thu, 20 Nov 2003, Stefan J. Betz wrote: > Hello People, > > i have some Mitsumi USB Floppy Drive with the following Data: > Manufactur: Mitsumi > Typ : D353FUE > > When i plug this Device into my Linux Box (Kernel 2.6.0-test9), i get > the following messages in my Syslog: > > Nov 20 22:17:57 mobileone kernel: hub 1-0:1.0: new USB device on port 1, assigned address 2 > Nov 20 22:17:57 mobileone kernel: usb-storage: This device (03ee,6901,0100 S 04 P 00) has unneeded SubClass and Protocol entries in unusual_devs.h > Nov 20 22:17:57 mobileone kernel: Please send a copy of this message to > Nov 20 22:17:57 mobileone kernel: scsi2 : SCSI emulation for USB Mass Storage devices > Nov 20 22:17:57 mobileone kernel: Vendor: MITSUMI Model: USB FDD Rev: 1039 > Nov 20 22:17:57 mobileone kernel: Type: Direct-Access ANSI SCSI revision: 02 > Nov 20 22:17:57 mobileone kernel: Attached scsi generic sg2 at scsi2, channel 0, id 0, lun 0, type 0 > I that is enough information to Support that drive (or how can i use ist > today?) > > Greeting Betz Stefan Thank you for sending this in. The usb-storage driver will be updated sometime after 2.6.0-final is released. Alan Stern ChangeSet@1.1270.1.5, 2003-12-11 16:34:23-08:00, ebrower@usa.net [SPARC64]: SUNW,lombus device has nonstandard ebus child regs too. ChangeSet@1.1280, 2003-12-11 16:32:42-08:00, alexander@all-2.com [PATCH] USB storage: patch for unusual_devs.h I send a patch and copy of /proc/bus/usb/devices for my 5`25 external USB enclosure. I don't know exactly manufacturer of this device, but model is CD-509. It will be nice if it helps somebody else. T: Bus=01 Lev=00 Prnt=00 Port=00 Cnt=00 Dev#= 1 Spd=12 MxCh= 2 B: Alloc= 93/900 us (10%), #Int= 1, #Iso= 0 D: Ver= 1.00 Cls=09(hub ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1 P: Vendor=0000 ProdID=0000 Rev= 0.00 S: Product=USB UHCI Root Hub S: SerialNumber=14a0 C:* #Ifs= 1 Cfg#= 1 Atr=40 MxPwr= 0mA I: If#= 0 Alt= 0 #EPs= 1 Cls=09(hub ) Sub=00 Prot=00 Driver=hub E: Ad=81(I) Atr=03(Int.) MxPS= 8 Ivl=255ms T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=1.5 MxCh= 0 D: Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1 P: Vendor=045e ProdID=0040 Rev= 3.00 S: Manufacturer=Microsoft S: Product=Microsoft 3-Button Mouse with IntelliEye(TM) C:* #Ifs= 1 Cfg#= 1 Atr=a0 MxPwr=100mA I: If#= 0 Alt= 0 #EPs= 1 Cls=03(HID ) Sub=01 Prot=02 Driver=usbmouse E: Ad=81(I) Atr=03(Int.) MxPS= 4 Ivl=10ms T: Bus=01 Lev=01 Prnt=01 Port=01 Cnt=02 Dev#= 15 Spd=12 MxCh= 0 D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1 P: Vendor=05e3 ProdID=0701 Rev= 0.02 S: Product=USB TO IDE C:* #Ifs= 1 Cfg#= 1 Atr=c0 MxPwr= 96mA I: If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage E: Ad=81(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms ChangeSet@1.1279, 2003-12-11 16:23:45-08:00, herbert@gondor.apana.org.au [PATCH] USB Storage: freecom dvd-rw fx-50 usb-ide patch ChangeSet@1.1278, 2003-12-11 16:22:58-08:00, stern@rowland.harvard.edu [PATCH] USB storage: unusual_devs.h entry revision Here is another update for unusual_devs.h in both 2.6 and 2.4. No urgency. On Wed, 12 Nov 2003, Aris Basic wrote: > Device Sony Memory Stick Reader MSAC-US1 > usb-storage: This device (054c,002d,0100 S 04 P 01) has unneeded SubClass and Protocol entries in unusual_devs.h > Please send a copy of this message to Thanks for sending this in. ChangeSet@1.1277, 2003-12-11 11:55:29-08:00, mbp@samba.org [PATCH] USB storage: add unusual storage device entry for Minolta DiMAGE Yes, it seems to work OK on the 7i with this updated patch. I don't have a 7 or 7Hi to try, but everything on the web seems to say the USB firmware works the same way. ChangeSet@1.1276, 2003-12-11 11:51:17-08:00, stephane.galles@free.fr [PATCH] USB storage: patch for Kyocera S5 camera I've seen some entries in 2.4.22 and 2.6.0 unusual_devs.h for Kyocera Finecam S3 et S4 cameras and I own a Finecam S5 that does not work out of the box either (here is the beast : http://www.yashica.com/digital/finecams5/finecams5.html) so I found the unusual_devs.h entry and submitted it some month ago at http://www.qbik.ch/usb/devices/showdev.php?id=1626 for the 2.4 kernels I thought It would be nice to have the whole Finecam family in Unusual_devs.h for 2.6.0 The patch for the 2.6.0-test9 is attached with this mail It differs from the entry I submitted at www.qbik.ch as I used the new SC/PR_DEVICE flags and got rid of the IGNORE_SER flag from 2.4 Do you want a patch for 2.4 too ? If so, I should test my old 2.4 entry with the lastest 2.4 Kernels, coz on a daily basis I use a 2.4.20, which is rather old. Moreover, I could used the SC/PR_DEVICE flags too for 2.4.22 (keeping the IGNORE_SER flag though) By the way, several entries with the running patch : /proc/bus/usb/devices : T: Bus=03 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=12 MxCh= 0 D: Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1 P: Vendor=0482 ProdID=0103 Rev= 1.00 C:* #Ifs= 1 Cfg#= 1 Atr=c0 MxPwr= 2mA I: If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage E: Ad=81(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms I: If#= 0 Alt= 1 #EPs= 3 Cls=ff(vend.) Sub=06 Prot=50 Driver=usb-storage E: Ad=81(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=83(I) Atr=03(Int.) MxPS= 8 Ivl=32ms ChangeSet@1.1275, 2003-12-11 11:48:09-08:00, fello@libero.it [PATCH] USB storage: patch for Fujifilm EX-20 ChangeSet@1.1274, 2003-12-11 11:42:50-08:00, per.winkvist@uk.com [PATCH] USB storage: Make Pentax Optio S4 work The change below is needed to get the S4 camera working. Tested with both Optio S/S4 ChangeSet@1.1273, 2003-12-11 11:37:20-08:00, petkan@nucleusys.com [PATCH] USB: pegasus driver update This is a patch for pegasus driver. Nothing exciting, just 2 more devices and a small fix. correct handling of external PHYs; better output to ethtool and suchlike; two new vendors. one of them Microsoft :) ChangeSet@1.1272, 2003-12-11 11:23:51-08:00, henning@meier-geinitz.de [PATCH] USB: scanner driver: new device ids Added vendor/product ids for Epson, Genius, Microtek, Plustek, Reflecta, and Visioneer scanners. Removed ids for HP PSC devices as these are supported by the hpoj userspace driver. ChangeSet@1.1270.1.4, 2003-12-10 18:50:18-02:00, bunk@fs.tum.de [PATCH] Ask for CONFIG_INDYDOG only on mips On Wed, Dec 10, 2003 at 02:23:14PM -0200, Marcelo Tosatti wrote: >... > Summary of changes from v2.4.23 to v2.4.24-pre1 > ============================================ >... > Ralf Bächle: >... > o MIPS char driver update >... This change contains the following bogus change for drivers/char/Config.in: ... @@ -237,9 +251,6 @@ tristate ' Eurotech CPU-1220/1410 Watchdog Timer' CONFIG_EUROTECH_WDT tristate ' IB700 SBC Watchdog Timer' CONFIG_IB700_WDT tristate ' ICP ELectronics Wafer 5823 Watchdog' CONFIG_WAFER_WDT - if [ "$CONFIG_SGI_IP22" = "y" ]; then - dep_tristate ' Indy/I2 Hardware Watchdog' CONFIG_INDYDOG $CONFIG_SGI_IP22 - fi tristate ' Intel i810 TCO timer / Watchdog' CONFIG_I810_TCO tristate ' Mixcom Watchdog' CONFIG_MIXCOMWD tristate ' SBC-60XX Watchdog Timer' CONFIG_60XX_WDT @@ -256,6 +267,7 @@ fi fi tristate ' ZF MachZ Watchdog' CONFIG_MACHZ_WDT + dep_tristate ' Indy/I2 Hardware Watchdog' CONFIG_INDYDOG $CONFIG_SGI_IP22 dep_tristate ' AMD 766/768 TCO Timer/Watchdog' CONFIG_AMD7XX_TCO $CONFIG_EXPERIMENTAL fi endmenu ... A dependency on a possibly undefined variable doesn't work with the 2.4 config system, and "make oldconfig" asks me on i386 for CONFIG_INDYDOG. The following patch fixes it: ChangeSet@1.1270.1.3, 2003-12-10 18:30:50-02:00, trini@mvista.com [PATCH] Fix rtc leak ===== arch/cris/drivers/ds1302.c 1.6 vs edited ===== ChangeSet@1.1270.1.2, 2003-12-09 21:39:08-02:00, marcelo@logos.cnet Merge master.kernel.org:/home/davem/BK/sparc-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 TAG: v2.4.24-pre1 ChangeSet@1.1270.2.1, 2003-12-09 12:35:10-08:00, davem@nuts.ninka.net [SPARC]: Add CONFIG_OOM_KILLER entries. ChangeSet@1.1270.1.1, 2003-12-09 16:57:53-02:00, wli@holomorphy.com [PATCH] fix 2.4 BLK_BOUNCE_ANY The bitshift defining BLK_BOUNCE_ANY can overflow. This patch casts to u64 before shifting there as well as in BLK_BOUNCE_HIGH to ensure integer overflow does not occur. Originally discovered by Zwane Mwaikambo during pgcl development, submitted by me to mainline to 2.6, and already included in 2.6. -- wli ===== include/linux/blkdev.h 1.25 vs edited ===== ChangeSet@1.1270, 2003-12-09 09:52:38-02:00, marcelo@logos.cnet Readd the OOM killer as configurable option, defaulted to off. ChangeSet@1.1136.74.6, 2003-12-08 21:56:39-08:00, xose@wanadoo.es [TG3]: Add new device IDs. ChangeSet@1.1268, 2003-12-08 17:43:11-02:00, torvalds@osdl.org [PATCH] Fix x86 kernel page fault error codes ChangeSet@1.1267, 2003-12-08 14:12:52-02:00, trini@kernel.crashing.org [PATCH] Fix PPC compilation On Mon, Dec 08, 2003 at 12:02:44PM +0200, Meelis Roos wrote: > gcc -D__ASSEMBLY__ -D__KERNEL__ -I/home/mroos/compile/linux-2.4/include -I/home/mroos/compile/linux-2.4/arch/ppc -c -o cpu_setup_6xx.o cpu_setup_6xx.S > cpu_setup_6xx.S: Assembler messages: > cpu_setup_6xx.S:164: Error: Unrecognized opcode: `andi' [ donning brown paper bag ] Marcelo, please apply the following: PPC32: Fix a thinko in arch/ppc/kernel/cpu_setup_6xx.S ChangeSet@1.1266, 2003-12-08 11:31:31-02:00, geert@linux-m68k.org [PATCH] MAINTAINERS vger.rutgers.edu Mailing lists at vger.rutgers.edu are obsolete, use vger.kernel.org instead. Applies to 2.4.23 and 2.6.0-test11. ChangeSet@1.1265, 2003-12-08 11:31:11-02:00, geert@linux-m68k.org [PATCH] Amiga Gayle IDE cleanup Amiga Gayle IDE: Kill old test code for the IDE doubler ChangeSet@1.1264, 2003-12-08 11:30:42-02:00, geert@linux-m68k.org [PATCH] Mac extern Mac: Use common header files instead of explicit extern declaration ChangeSet@1.1263, 2003-12-08 11:30:27-02:00, geert@linux-m68k.org [PATCH] Mac SCSI Mac SCSI fixes (from Matthias Urlichs in 2.6.0): - Inline functions need to be defined before being used. - out_8() takes an address and a value, not the other way round. ChangeSet@1.1262, 2003-12-08 11:28:46-02:00, geert@linux-m68k.org [PATCH] Genrtc warning Genrtc: Move code to kill warning if CONFIG_PROC_FS is disabled ChangeSet@1.1261, 2003-12-08 11:28:23-02:00, geert@linux-m68k.org [PATCH] NCR53C9x SCSI inline NCR53C9x SCSI: Kill bogus inline ChangeSet@1.1260, 2003-12-08 11:24:49-02:00, geert@linux-m68k.org [PATCH] Amiga Buddha/CatWeasel IDE Buddha/CatWeasel IDE: Make sure the core IDE driver doesn't try to request the MMIO ports a second time, since this will fail. ChangeSet@1.1259, 2003-12-08 11:24:40-02:00, geert@linux-m68k.org [PATCH] M68k Documentation M68k Documentation: framebuffer.txt no longer exists in the m68k directory (from Nikita Melnikov) ChangeSet@1.1258, 2003-12-08 09:28:55-02:00, geert@linux-m68k.org [PATCH] M68k extern inline M68k core: Replace (variants of) `extern inline' by `static inline' ChangeSet@1.1257, 2003-12-08 09:28:31-02:00, geert@linux-m68k.org [PATCH] M68k asm/system.h M68k: Add missing #ifdef __KERNEL / #endif (from Christian T. Steigies) ChangeSet@1.1256, 2003-12-08 09:27:30-02:00, geert@linux-m68k.org [PATCH] Mac II VIA Mac II VIA: Don't include directly ChangeSet@1.1255, 2003-12-08 09:26:52-02:00, geert@linux-m68k.org [PATCH] Amiga NCR53c710 SCSI Amiga NCR53c710: Coalesce all Amiga NCR53c710 SCSI host adapters configuration options into one config option, as suggested by Matthew Wilcox. ChangeSet@1.1254, 2003-12-08 09:26:05-02:00, geert@linux-m68k.org [PATCH] Amiga Gayle E-Matrix 530 IDE Amiga Gayle IDE: Add support for the IDE interface on the M-Tech E-Matrix 530 expansion card ChangeSet@1.1253, 2003-12-08 09:25:31-02:00, geert@linux-m68k.org [PATCH] Amiga debug fix Amiga: Fix `debug=mem' (record all kernel messages in ChipRAM): virt_to_phys() no longer works for Zorro II memory space, we must use ZTWO_PADDR() ChangeSet@1.1252, 2003-12-08 09:24:28-02:00, geert@linux-m68k.org [PATCH] ncr53c7xx SCSI ncr53c7xx: Cleanup prototypes for ncr53c7xx_init() ChangeSet@1.1136.1.84, 2003-12-08 08:37:45-02:00, geert@linux-m68k.org [PATCH] Mac ADB ADB: Disable the ADB clock code when CONFIG_ADB is not selected (from Matthias Urlichs in 2.6.0). ChangeSet@1.1136.1.83, 2003-12-08 08:37:20-02:00, geert@linux-m68k.org [PATCH] Macfb setup Macfb: Update setup routine (from Matthias Urlichs in 2.6.0) ChangeSet@1.1136.1.82, 2003-12-08 08:37:05-02:00, geert@linux-m68k.org [PATCH] Mac ESP SCSI setup Mac ESP SCSI: Update argument parsing (from Matthias Urlichs in 2.6.0) ChangeSet@1.1136.1.81, 2003-12-08 08:36:39-02:00, geert@linux-m68k.org [PATCH] Mac89x0 Ethernet Macintosh CS89x0 Ethernet: Netif updates (from Matthias Urlichs in 2.6.0) ChangeSet@1.1136.1.80, 2003-12-08 08:36:19-02:00, geert@linux-m68k.org [PATCH] M68k symbol exports M68k: Export missing symbols (from Matthias Urlichs in 2.6.0) ChangeSet@1.1136.1.79, 2003-12-08 08:35:58-02:00, geert@linux-m68k.org [PATCH] Mac ADB IOP fix Mac ADB IOP: Fix improperly initialized request struct in the reset code, causing a bogus pointer (from Matthias Urlichs in 2.6.0) ChangeSet@1.1136.1.78, 2003-12-08 08:35:35-02:00, geert@linux-m68k.org [PATCH] M68k cache mode M68k: Use a constant m68k_supervisor_cachemode only if we know it's safe, otherwise use the value from head.S (from Roman Zippel) ChangeSet@1.1136.1.77, 2003-12-08 08:34:39-02:00, geert@linux-m68k.org [PATCH] Zorro include guard Zorro: Make closing include guards consistent with opening include guards ChangeSet@1.1136.1.76, 2003-12-08 08:34:22-02:00, geert@linux-m68k.org [PATCH] M68k RMW accesses M68k: Avoid bus fault for certain RMW accesses (from Roman Zippel) ChangeSet@1.1136.1.75, 2003-12-07 21:20:48-02:00, willy@debian.org [PATCH] Remove broken file lock accounting On Mon, Jul 01, 2002 at 11:13:55PM +0100, Matthew Wilcox wrote: > The file lock accounting code is horribly broken (and I wrote it, I > should know). I think the best solution to 2.4 is simply to delete it, > at least for BSD-style flocks. > > Patch to follow. Note that 2.5 has the same issue, but I'll fix it > differently there. Here's the patch for 2.4: ChangeSet@1.1136.1.74, 2003-12-07 15:14:54-02:00, mason@suse.com [PATCH] From -aa tree: Fix end_buffer_io_kiobuf() locking Must wakeup with end_buffer_io_kiobuf as last thing to be sure the wakeup will happen on a still allocated kiobuf. ChangeSet@1.1136.1.73, 2003-12-07 15:10:38-02:00, mikulas@artax.karlin.mff.cuni.cz [PATCH] from -aa tree: Fix potential fsync() race condition > 00_ll_rw_block-sync-race-1 first appeared in 2.4.21pre4aa3 - 470 bytes > > Add lock_page in ll_rw_block to fix a fs race > condition. Fix suggested by Mikulas Patocka. Yes. You have two inodes placed in the same buffer. Process 1 modifies inode 1 and calls fsync on it. fsync initiates write of the block. ll_rw_block returns, write is in progress. Process 2 modifies inode 2 and calls fsync on it. Filesystem calls ll_rw_block write on the same buffer. ll_rw_block immediatelly returns, because it sees there is already IO on the buffer (there used to be something like if (buffer_locked(bh)) return;). Process 2 waits on buffer. The write finished. Both processes are waken up. Both processes return out of fsync function. Process 2 returns from fsync while it did not write its inode modification to disk --- it waited on process 1's write. ChangeSet@1.1136.82.2, 2003-12-07 12:43:34-02:00, wli@holomorphy.com [PATCH] out_of_memory() locking On Sun, Nov 30, 2003 at 08:18:02AM -0800, William Lee Irwin III wrote: > (1) the timestamps/etc. weren't locked, and when cpus raced, it caused > false OOM kills > (2) the mm could go away while scanning the tasklist, causing the thing > to try to kill kernel threads > Here's a preliminary backport (please do _NOT_ apply until I or someone > tests it) for you to comment on. Basically, do you want (1) and (2) > split out, is the basic thing okay, etc.? out_of_memory()'s operational variables are not locked, and can be reset by multiple cpus simultaneously, causing false OOM kills. This patch adds an oom_lock to out_of_memory() to protect its operational variables. -- wli ===== mm/oom_kill.c 1.14 vs edited ===== ChangeSet@1.1136.82.1, 2003-12-06 18:43:00-02:00, marcelo@logos.cnet Ernie Petrides: Readd exec_mmap() fastpath with correct locking ChangeSet@1.1136.1.69, 2003-12-06 16:35:42-02:00, johnstul@us.ibm.com [PATCH] Always make TSC available via get_cycles() when TSC is present On Tue, 2003-11-04 at 15:22, Joel Becker wrote: > Folks, > Certain distributions are building all of their SMP kernels > NUMA-aware. This is great, as the kernels support boxes like the x440 > with no trouble. However, this implicitly disables CONFIG_X86_TSC. > While that is good for NUMA systems, and fine from a kernel timing > standpoint, it also eliminates any generic access to the TSC via > get_cycles(). With CONFIG_X86_TSC not defined, get_cycles() always > returns 0. > Given that >95% of machines will not be x440s, this means that a > user of that kernel cannot access a high resolution timer via > get_cycles(). I don't want to have to litter my code with rdtscll() > when I managed to remove it! > The proposed patch is trivial. If the system has a TSC, it is > available get_cycles(). This makes no change to the other parts of the > kernel protected by CONFIG_X86_TSC. CONFIG_X86_TSC be the devil. Personally, I'd much prefer dropping the compile time option and using dynamic detection. Something like (not recently tested and i believe against 2.5.something, but you get the idea): ChangeSet@1.1136.1.68, 2003-12-06 16:25:16-02:00, wli@holomorphy.com [PATCH] Fixup smb_boot_cpus(): Fix HT detection bug On Wed, Dec 03, 2003 at 06:41:36PM -0500, Ethan Weinstein wrote: > Ok, setting CONFIG_NR_CPUS=8 does indeed solve the HT issue, looks like > it was the numbering scheme: Something like this might do the trick. NR_CPUS is already checked indirectly via max_cpus. -- wli ===== arch/i386/kernel/smpboot.c 1.17 vs edited ===== ChangeSet@1.1136.81.1, 2003-12-06 14:49:55+11:00, nathans@bruce.melbourne.sgi.com Merge nathans@xfs.org://export/hose/bkroot/linux-2.4+coreXFS into bruce.melbourne.sgi.com:/source1/linux-2.4+coreXFS ChangeSet@1.1136.71.12, 2003-12-06 14:30:56+11:00, nathans@bruce.melbourne.sgi.com Add XFS documentation and incorporate XFS into the kernel build. ChangeSet@1.1136.71.11, 2003-12-06 14:25:48+11:00, nathans@bruce.melbourne.sgi.com Export several kernel symbols used by the XFS filesystem. ChangeSet@1.1136.80.1, 2003-12-05 12:46:52-06:00, Matt_Domsch@dell.com EDD: s/DISKSIG_BUFFER/DISK80_SIG_BUFFER so it compiles bump EDD version number. ChangeSet@1.1136.79.2, 2003-12-05 10:42:16-08:00, dancy@dancysoft.com [PATCH] USB: add TIOCMIWAIT support to pl2303 driver ChangeSet@1.1136.1.66, 2003-12-05 16:14:13-02:00, marcelo@logos.cnet Merge bk://stop.crashing.org/linux-2.4.24-pre into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1136.1.65, 2003-12-05 15:53:34-02:00, mikpe@csd.uu.se [PATCH] fix reboot/no_idt bug When compiling 2.4.23 with gcc-3.3.2, gcc generates the following warning for arch/i386/kernel/process.c: process.c: In function `machine_restart': process.c:427: warning: use of memory input without lvalue in asm operand 0 is deprecated The warning identifies a real bug. no_idt is passed to lidt with an "m" constraint, which requires an l-value. Since no_idt is faked as an array, gcc creates an anonymous variable pointing to no_idt and passes that to lidt(*), so at runtime lidt sees the wrong address. Not good. (The bug, while real, is unlikely to trigger since it sits in an infrequently used path in the reboot code.) The fix is to make no_idt a struct (and thus an l-lvalue) like the other gdt/idt descriptors. This patch is a backport of the fix Linus made for the same bug in 2.6.0-test4. [Andi: x86-64 appears to have the same bug] (*) Verified by inspection of the assembly code. /Mikael ChangeSet@1.1136.70.7, 2003-12-05 11:39:32-02:00, cel@citi.umich.edu [PATCH] Make readahead last page of file hi marcelo- i posted this a while back on fsdevel for comments, but never heard anything. i'd like this patch to be included in 2.4.24, as it improves NFS client read performance by a significant margin, and doesn't appear to have any negative impact (see fsdevel archives for benchmarks). this is against 2.4.23. generic_file_readahead never reads the last page of a file. this means the last page is always read synchronously by do_generic_file_read. normally this is not an issue, as most local disk reads are fast. however, this means that the NFS client is never given the opportunity to coalesce the last page of a file, which must be read in a separate synchronous read operation. this is especially honerous if the network round trip time is long, and/or the workload consists of reading many uncached small files. i also explored changing the way that generic_file_readahead computes the index of the last page of the file. the fix below appears to be the best solution. ChangeSet@1.1136.77.5, 2003-12-05 11:17:04-02:00, dhinds@sonic.net [PATCH] PATCH: update/bugfix for pcnet_cs driver Hi Marcelo, o fixed duplex selection for DL10019 cards in pcnet_cs driver. -- Dave ChangeSet@1.1136.77.4, 2003-12-05 10:59:20-02:00, elenstev@mesatop.com [PATCH] 2.4.23 update Documentation/Changes for quota-tools This patch updates Documentation/Changes for quota-tools. This information may be needed by those wanting to use the new quota code (CONFIG_QFMT_V2) which went in about 6 months ago. The patch to update ver_linux for quota-tools was applied 5 months ago, but this part slipped through the cracks somehow. Steven ChangeSet@1.1136.77.3, 2003-12-05 10:33:35-02:00, aspicht@arkeia.com [PATCH] ac97_plugin_ad1980 fixes Hi, I have a MSI875P Neo motherboard which includes an AD1980 codec. I tried to use ac97_plugin_ad1980 as a module, but it failed to load because of a GPL issue. I had to add a MODULE_LICENSE("GPL") to make it work. Also, I had to force the ADI compatibility mode to have the switch input/ouput working. Now, it's working perfectly. Here is the diff -u with kernel version 2.4.23 ChangeSet@1.1136.73.53, 2003-12-04 10:34:54-08:00, laforge@netfilter.org [IPV6]: Do not bypass netfilter for MLD/IGMP messages. ChangeSet@1.1136.73.52, 2003-12-04 10:32:56-08:00, laforge@netfilter.org [NETFILTER]: Fix ipchains MASQUERADE oops. ChangeSet@1.1136.77.1, 2003-12-04 12:13:04-02:00, mikael.starvik@axis.com [PATCH] CRIS architecture update Hi Marcelo! The attached patch contains the latest changes to the CRIS architechture that makes it up to date with 2.4.23. Would be nice if you could apply it for inclusion in 2.4.24. Hope it applies without fuzz... ChangeSet@1.1242, 2003-12-03 22:31:56-06:00, cattelan@lupo.thebarn.com Gone dmapi ChangeSet@1.1136.71.10, 2003-12-04 11:23:52+11:00, nathans@bruce.melbourne.sgi.com Provide an iget variant without unlocking the inode and without the read_inode call (iget_locked). Used by XFS and backported from 2.6. ChangeSet@1.1136.71.9, 2003-12-04 11:18:56+11:00, nathans@bruce.melbourne.sgi.com Provide a simple try-lock based dirty page flushing routine. ChangeSet@1.1136.71.8, 2003-12-04 11:13:59+11:00, nathans@bruce.melbourne.sgi.com Support for delayed allocation. Used by XFS and backported from 2.6. ChangeSet@1.1136.71.7, 2003-12-04 11:09:52+11:00, nathans@bruce.melbourne.sgi.com Add a process flag to identify a process performing a transaction. Used by XFS and backported from 2.6. ChangeSet@1.1136.71.6, 2003-12-04 11:07:43+11:00, nathans@bruce.melbourne.sgi.com Remove some unused macros and related comment from the XFS quota header. ChangeSet@1.1136.71.5, 2003-12-04 11:05:44+11:00, nathans@bruce.melbourne.sgi.com Fix utimes(2) and immutable/append-only files. Originally by Ethan Benson. ChangeSet@1.1136.45.12, 2003-12-03 08:48:57-07:00, trini@kernel.crashing.org PPC32: Update an errata on the MPC745x line. Motorola has decided that 7450 and 7451 do not have the BTIC errata. ChangeSet@1.1136.76.10, 2003-12-03 10:05:02-02:00, geert@linux-m68k.org [PATCH] 2.4.23 ext3 warning Kill warning if CONFIG_QUOTA is disabled. ChangeSet@1.1136.76.9, 2003-12-03 10:04:17-02:00, geert@linux-m68k.org [PATCH] Atyfb on Mach64 GX or Atari Here's the latest incarnation of the atyfb fix for Mach64 GX (compiles with warnings and cannot work) and Atari. There's only one chance since the last version I sent in September: the probing for the crystal frequency is now done for all chips with integrated clock generator. ChangeSet@1.1136.76.8, 2003-12-03 07:51:25-02:00, adam@nmt.edu [PATCH] PATCH: 3ware driver update for 2.4.23-bk2 ChangeSet@1.1231.1.3, 2003-12-03 17:01:12+11:00, nathans@sgi.com [XFS] Fix comment in xfs_rename.c SGI Modid: xfs-linux:slinx:162760a ChangeSet@1.1231.1.2, 2003-12-03 16:56:07+11:00, nathans@sgi.com [XFS] Prevent log ktrace code from sleeping in an invalid context. SGI Modid: xfs-linux:slinx:162768a ChangeSet@1.1231.1.1, 2003-12-03 16:51:38+11:00, nathans@sgi.com [XFS] Use vnode timespec modifiers for atime/mtime/ctime, keeps last code hunk in sync. SGI Modid: xfs-linux:slinx:162782a ChangeSet@1.1136.74.5, 2003-12-02 19:45:58-08:00, davem@nuts.ninka.net [TG3]: Update to latest non-5705 TSO firmware. ChangeSet@1.1136.73.51, 2003-12-02 16:56:55-08:00, geert@linux-m68k.org [NET]: Fix atm/br2684 build with procfs disabled. ChangeSet@1.1136.73.50, 2003-12-02 16:29:34-08:00, pavlin@icir.org [RTNETLINK]: Add RTPROT_XORP. ChangeSet@1.1136.1.63, 2003-12-02 16:20:42-07:00, bjorn.helgaas@hp.com ia64: update default configs ChangeSet@1.1136.76.7, 2003-12-02 15:21:27-02:00, ralf@linux-mips.org [PATCH] MIPS network driver updates Update the MIPS network drivers. These are all changes to MIPS specific drivers. This patch removes the meth driver for the SGI IP32; support for this machine is only developped for 2.6. The gt64240eth driver supports the integrated ethernet interface of the Galileo GT-64240 system controller. ChangeSet@1.1136.76.6, 2003-12-02 15:08:01-02:00, mikpe@csd.uu.se [PATCH] fix some DRM43 warnings [Resend. Was ignored for 2.4.23-pre :-(] This patch fixes three sources of warnings in the new 4.3 DRM code: - drm_os_linux.h has a #warning on broken list_entry usage. The patch backports a fix from the 2.6 code. - radeon_drv.h has a #warning PCI posting bug. The #warning refers to an obsolete and unused macro. Simply remove it. It's also gone from the 2.6 code. - drm_agpsupport.h generates a gcc warning on an assignment used as a truth value. The patch backports a fix from the 2.6 code. These changes are well-tested: they have been in 2.6 for ages, and I've been running them in 2.4.23-pre/rc for months. Please apply. /Mikael ChangeSet@1.1136.76.5, 2003-12-02 15:06:32-02:00, pp@ee.oulu.fi [PATCH] 2.4 lacks dummy SET_NETDEV_DEV I agree, someone should merge in the dummy SET_NETDEV_DEV once Marcelo starts up 2.4.24-preX Ping :-) ChangeSet@1.1136.76.4, 2003-12-02 12:02:00-02:00, neilb@cse.unsw.edu.au [PATCH] Drop module count if lockd reclaimer thread failed to start. ----------- Diffstat output ------------ ./fs/lockd/clntlock.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletion(-) ChangeSet@1.1136.76.3, 2003-12-02 12:01:27-02:00, neilb@cse.unsw.edu.au [PATCH] Honour SUN NFSv2 hack for "set times to server time. >From Greg Banks , see comment in code. ----------- Diffstat output ------------ ./fs/nfsd/nfsxdr.c | 11 +++++++++++ 1 files changed, 11 insertions(+) ChangeSet@1.1136.76.2, 2003-12-02 11:58:06-02:00, neilb@cse.unsw.edu.au [PATCH] Make root a special case for per-user process limits. This is needed because when a setuid-root program calls setuid(0) to become really-root, p->user becomes root_user, but ->rlim stays as the original user's limit, and now the process cannot fork - becuase root has more processes than the original user had. The real problem is that NPROC is not really a per-process limit, but its a per-user limit, and including it with the rlim structure was not a good idea :-( This fix is already in 2.6 ----------- Diffstat output ------------ ./kernel/fork.c | 1 + 1 files changed, 1 insertion(+) ChangeSet@1.1136.75.9, 2003-12-02 11:29:52-02:00, ralf@linux-mips.org [PATCH] Misc MIPS video bits ChangeSet@1.1136.75.8, 2003-12-02 11:27:56-02:00, ralf@linux-mips.org [PATCH] Supply default values to rtc.c This supplies default values (from PC-ish architectures) to the rtc.c driver. ChangeSet@1.1136.75.7, 2003-12-02 11:26:51-02:00, ralf@linux-mips.org [PATCH] No stone-age compat stuff for MIPS Yeah ugly, let's add one more to that expression ... Somehow the logic of this expression is the wrong way but heck, this is 2.4 ... ChangeSet@1.1136.75.6, 2003-12-02 11:26:08-02:00, ralf@linux-mips.org [PATCH] Cleanup lk201-map.c drivers/tc/lk201-map.c is a generated file so needs to be removed on make distclean. ChangeSet@1.1136.75.5, 2003-12-02 11:25:55-02:00, ralf@linux-mips.org [PATCH] Remove remaining drivers/sgi bits ChangeSet@1.1136.75.4, 2003-12-02 11:25:48-02:00, ralf@linux-mips.org [PATCH] MIPS ioaddr_t is 32-bit ChangeSet@1.1136.75.3, 2003-12-02 11:25:34-02:00, ralf@linux-mips.org [PATCH] Add more MIPS bits in ChangeSet@1.1136.75.2, 2003-12-02 11:25:25-02:00, ralf@linux-mips.org [PATCH] Include into spinlock.h uses local_irq_save() etc. from but relies on this header file having been dragged in on some other way. So if things are just right the build may blow up ... ChangeSet@1.1136.75.1, 2003-12-02 10:37:25-02:00, ralf@linux-mips.org [PATCH] New PCMCIA drivers Add PCMCIA bridge drivers for AMD Alchemy Au1x00, NEC VRc4173 and Sibyte. ChangeSet@1.1136.74.4, 2003-12-02 02:26:36-08:00, davem@nuts.ninka.net [TG3]: Update version and release date. ChangeSet@1.1136.74.3, 2003-12-02 02:16:23-08:00, davem@nuts.ninka.net [TG3]: Clear on-chip stats/status block after resetting flow-through queues. On systems where the config cycles might take a long time, we can end up with the ASF firmware using the FTQs before we get to resetting them. ChangeSet@1.1136.74.2, 2003-12-02 02:09:52-08:00, davem@nuts.ninka.net [TG3]: Do not set RX_MODE_KEEP_VLAN_TAG when ASF is enabled. ChangeSet@1.1136.74.1, 2003-12-02 02:01:49-08:00, davem@nuts.ninka.net [TG3]: Do not drop existing GRC_MODE_HOST_STACKUP when writing to GRC_MODE. ChangeSet@1.1136.73.48, 2003-12-02 01:43:17-08:00, kaber@trash.net [NETFILTER]: Export conntrack bucket count via read-only sysctl. With help from Harald Welte. ChangeSet@1.1136.73.47, 2003-12-02 01:41:36-08:00, kadlec@blackhole.kfki.hu [NETFILTER]: Avoid order n^2 lookup of whole conntrack hash in ip_ct_selective_cleanup(). ChangeSet@1.1136.73.46, 2003-12-02 01:38:04-08:00, kaber@trash.net [NETFILTER]: Fix expectation evict order This patch fixes the order in which expectations are evicted when a helper specifies IP_CT_HELPER_F_REUSE_EXPECT. Prior to that patch, each new expectation is added at the beginning of the masters sibling list. When max_expected is reached, the first non-established one on the list is reused (which gave LIFO behaviour). The patch - adds new expectations at the end of the masters sibling list - uses list_for_each_entry for finding the oldes one - changes nat core to walk backwards throuth the sibling list when calling nat helpers ChangeSet@1.1136.73.45, 2003-12-02 01:35:05-08:00, kadlec@blackhole.kfki.hu [NETFILTER]: Use list macros instead of reimplementation. ChangeSet@1.1136.73.44, 2003-12-02 01:30:36-08:00, shemminger@osdl.org [NETFILTER]: Trivial -- Get rid of warnings in netfilter if /proc is not configured on. ChangeSet@1.1136.73.43, 2003-12-02 01:29:27-08:00, kaber@trash.net [NETFILTER]: Fix various issues with the amanda conntrack+NAT helpers ip_conntrack_amanda: - udp checksum checked even if not set - memory not validated to be inside packet at all or validated like this: if (*data and data < data_limit) - would create expectation with port 0 which makes nat-helper throw away packet (which makes exploiting the race-condition trivial even on UP) - expectation had no mask.src.* set so connections were accepted from anywhere - helper_unregister called after failed helper_register ip_nat_amanda: - drops packet if amanda_data_fixup fails without clearing the offset field in the expectation which is used as a flag to indicate that the current packet needs mangling. Problem arises from the fact that the conntrack helper creates multiple expectations at once and even if it clears the offset field for the current expectation the following expectations will not have it cleared and will mangle the next packet they see. - same unregister after failed register problem ChangeSet@1.1136.73.42, 2003-12-02 01:27:14-08:00, laforge@netfilter.org [NETFILTER]: Kill extraneous memset()s in nat/conntrack TFTP code. ChangeSet@1.1136.73.41, 2003-12-02 01:25:27-08:00, rusty@rustcorp.com.au [NETFILTER]: Do not flush MASQ if IP did not change. Herve Eychenne noted that many ADSL connections get the same address when the interface comes back up, so some users use SNAT instead of masquerade. The answer is that MASQUERADE should only drop connections when the interface comes up, and then only if the interface address has actually changed. ChangeSet@1.1136.73.40, 2003-12-01 22:35:24-08:00, davem@nuts.ninka.net [VLAN]: Kill build warning due to missing declarations. ChangeSet@1.1136.73.39, 2003-12-01 21:51:38-08:00, davem@nuts.ninka.net Merge nuts.ninka.net:/disk1/davem/BK/net-work-2.4 into nuts.ninka.net:/disk1/davem/BK/net-2.4 ChangeSet@1.1136.1.62, 2003-12-01 12:54:06-07:00, kaos@sgi.com ia64: sync salinfo.c with 2.6 (suser -> capable, use standard macros). Make salinfo.c more compatible with 2.6. This is the 2.4 bit of http://marc.theaimsgroup.com/?l=linux-ia64&m=106974968032730&w=2 ChangeSet@1.1136.73.38, 2003-12-01 17:51:56-02:00, ralf@linux-mips.org [PATCH] Add IP22 parallel port driver ChangeSet@1.1136.73.37, 2003-12-01 17:51:01-02:00, ralf@linux-mips.org [PATCH] Add Pete Popov to CREDITS ChangeSet@1.1136.73.36, 2003-12-01 17:50:46-02:00, ralf@linux-mips.org [PATCH] Probe SGI partitions earlier Probe the disk volume header earlier so in case of doubt partitions will be recogniced as Disk Volume Header partitions instead of MSDOS style partitions. The is a problem at times on SGI system where both types of partitions are activated. ChangeSet@1.1136.73.35, 2003-12-01 17:50:01-02:00, ralf@linux-mips.org [PATCH] Add two new NEC PCI IDs ChangeSet@1.1136.73.34, 2003-12-01 17:49:47-02:00, ralf@linux-mips.org [PATCH] Fix VRc5477 remove method declaration ChangeSet@1.1136.45.11, 2003-12-01 12:49:10-07:00, trini@kernel.crashing.org PPC32: Print the correct ammount of memory not covered by BATs. From Mich Lanners . ChangeSet@1.1136.45.10, 2003-12-01 12:47:16-07:00, trini@kernel.crashing.org PPC32: gcc-3.4 build fixes from Olaf Hering . ChangeSet@1.1136.1.61, 2003-12-01 12:43:59-07:00, davidm@tiger.hpl.hp.com ia64: Fix a bug in sigtramp() which corrupted ar.rnat when unwinding across a signal trampoline (in user space). Reported by Laurent Morichetti. ChangeSet@1.1136.73.33, 2003-12-01 17:43:30-02:00, ralf@linux-mips.org [PATCH] Add GT-064011/GT-64111 PCI ID ChangeSet@1.1136.73.32, 2003-12-01 17:43:20-02:00, ralf@linux-mips.org [PATCH] Update NEC Osprey defconfig ChangeSet@1.1136.73.31, 2003-12-01 17:42:51-02:00, ralf@linux-mips.org [PATCH] ARC library updates New flag PROM_FLAG_DONT_FREE_TEMP to tell the ARC library to not free firmware memory. This is needed for the buggy SNI RM200 firmware which passed wrong values. Remove IP32 ARC bits. ChangeSet@1.1136.73.30, 2003-12-01 17:42:36-02:00, ralf@linux-mips.org [PATCH] Update Philips Nino defconfig ChangeSet@1.1136.73.29, 2003-12-01 17:42:18-02:00, ralf@linux-mips.org [PATCH] Update Cobalt defconfig file ChangeSet@1.1136.73.28, 2003-12-01 17:42:06-02:00, ralf@linux-mips.org [PATCH] Update Ocelot defconfig ChangeSet@1.1136.73.27, 2003-12-01 17:39:42-02:00, ralf@linux-mips.org [PATCH] Update HP Laserjet Update HP Laserjet defconfig file. ChangeSet@1.1136.73.26, 2003-12-01 17:39:30-02:00, ralf@linux-mips.org [PATCH] Update Jazz Change extern inline to static inline. Make a little attempt to implement proper locking for the Jazz DMA allocator. Reformat to Linux style. ChangeSet@1.1136.73.25, 2003-12-01 17:39:09-02:00, ralf@linux-mips.org [PATCH] DEC updates Update the DECstation support code to use the new HPT code; update defconfig file. ChangeSet@1.1136.73.24, 2003-12-01 17:38:27-02:00, ralf@linux-mips.org [PATCH] Galileo boards Update the defconfigs for the Galileo EV-64120 and EV-96100 eval boards. ChangeSet@1.1136.73.23, 2003-12-01 17:37:14-02:00, ralf@linux-mips.org [PATCH] ITE updates Update ITE8172 and IVR code to use the new high precision time code; update defconfigs. ChangeSet@1.1136.73.22, 2003-12-01 17:36:33-02:00, ralf@linux-mips.org [PATCH] NEC DDB updates This updates the defconfig for two of the NEC DDB eval boards; is also changes the time code to use the new MIPS high precision timer code. ChangeSet@1.1136.73.21, 2003-12-01 17:35:55-02:00, ralf@linux-mips.org [PATCH] SGI I2C driver This adds the I2C driver for the Indy and yet another SGI system. This driver is required in order to use the VINO and Indycam drivers. ChangeSet@1.1136.73.20, 2003-12-01 17:34:54-02:00, ralf@linux-mips.org [PATCH] Update JMR3927 support Update the default config and RTC code for the JMR3927. ChangeSet@1.1136.73.19, 2003-12-01 17:34:24-02:00, ralf@linux-mips.org [PATCH] Sibyte updates Various minor fixes for the SB1125 and SB1250 support code. ChangeSet@1.1136.73.18, 2003-12-01 17:33:41-02:00, ralf@linux-mips.org [PATCH] Update sgiwd93 driver Cleanup the mess that the sgiwd93 SCSI driver was. Primary intend was to get rid of MIPS-specific APIs in favor for equivalent generic APIs. ChangeSet@1.1136.73.17, 2003-12-01 17:33:01-02:00, ralf@linux-mips.org [PATCH] Update Atlas, Malta and SEAD boards Update the board support code for the MIPS Atlas, Malta and SEAD boards. Minor changes to the time and interrupt code and a formatting change plus update of the kernel config files. ChangeSet@1.1136.73.16, 2003-12-01 17:28:53-02:00, ralf@linux-mips.org [PATCH] MIPS video driver updates This updates some of the video drivers for Digital's DECstation workstations and the AMD Alchemy Au1100 framebuffer driver. ChangeSet@1.1136.73.15, 2003-12-01 17:26:01-02:00, ralf@linux-mips.org [PATCH] Update Origin code Header micro-cleanup and updated default configuration files. ChangeSet@1.1136.73.14, 2003-12-01 17:25:38-02:00, ralf@linux-mips.org [PATCH] MIPS char driver update Minor fix for reading the hour from the DS1286 / DS1386 driver used in Indys. New RTC driver for the Dallas DS1742 which is used on the RBTX4927. Misc fixes and updates to the dz.c driver; disabled debugging code in dz.c. Change rtc.c which really should get a different name because surprise not everybody in the world uses the same RTC chip to support memory mapped RTC chips properly. This is needed for Digital's DECstations. ChangeSet@1.1136.73.13, 2003-12-01 17:23:12-02:00, ralf@linux-mips.org [PATCH] Update TX4927 code This updates code to support for Toshiba's RXTX4927 evaluation board. ChangeSet@1.1136.73.12, 2003-12-01 17:20:06-02:00, ralf@linux-mips.org [PATCH] Update for NEC VR41xx systems Update support for NEC VR41xx systems. ChangeSet@1.1136.73.11, 2003-12-01 17:19:42-02:00, ralf@linux-mips.org [PATCH] Update support for Lasat boards Update the support for Lasat 100 and Lasat 200 systems. ChangeSet@1.1136.73.10, 2003-12-01 17:19:21-02:00, ralf@linux-mips.org [PATCH] Update RM200 C support Update the default config file for the SNI RM200 C. ChangeSet@1.1136.73.9, 2003-12-01 17:19:07-02:00, ralf@linux-mips.org [PATCH] Update SGI IP22 support Fix 64-bit kernel for IP22. Plenty of cleanups. Update the Indy's HAL2 sound driver. ChangeSet@1.1136.73.8, 2003-12-01 17:08:51-02:00, ralf@linux-mips.org [PATCH] Add VINO and Indycam drivers Add drivers for the Indycam and VINO hardware of the SGI Indy. ChangeSet@1.1136.73.7, 2003-12-01 17:07:22-02:00, ralf@linux-mips.org [PATCH] Add support for Momentum Ocelot C and Jaguar ATX Add support for Momentum's CompactPCI Ocelot C system and the ATX form factor Jaguar ATX board. ChangeSet@1.1136.73.6, 2003-12-01 17:06:42-02:00, ralf@linux-mips.org [PATCH] Remove support for compressed EV-64120 kernels Remove support for compressed kernels on EV-64120. This code was carrying half a firmware in it's baggage, introduced yet another copy of zlib and in general was just unmaintained and ugly beyond believe ... ChangeSet@1.1136.73.5, 2003-12-01 17:00:36-02:00, ralf@linux-mips.org [PATCH] AMD Alchemy updates Update of MIPS Alchemy code including PCMCIA and sound driver and default configuration files for various evaluation boards and systems. ChangeSet@1.1136.73.4, 2003-12-01 16:59:08-02:00, ralf@linux-mips.org [PATCH] mips64 updates Update the generic mips64 code. Support for the RM9000 and VR4133 processors. Fixes to the o32 / N32 binary compatibility code, spelling fixes. Change the linker script such that ECOFF conversion using elf2ecoff works again. Partial support for compile time configurable support of 4k, 16k and 64k page size. Fixes to RM7000 second level cache handling. ChangeSet@1.1136.73.3, 2003-12-01 16:57:42-02:00, ralf@linux-mips.org [PATCH] Update generic MIPS code Update the generic MIPS code. Highlights is support for PMC-Sierra's new RM9000 processor; also plenty of small bug fixes. ChangeSet@1.1136.73.2, 2003-12-01 16:56:43-02:00, ralf@linux-mips.org [PATCH] Update MIPS MAINTAINERS entries Update a few MIPS entries in MAINTAINERS. ChangeSet@1.1136.73.1, 2003-12-01 16:56:36-02:00, ralf@linux-mips.org [PATCH] MIPS Configure.help updates Updates the MIPS Documentation/Configure.help bits. ChangeSet@1.1136.1.60, 2003-12-01 11:51:55-07:00, davidm@tiger.hpl.hp.com ia64: Fix a alternate-signal-stack bug which could corrupt RNaT bits when bspstore happened to point to an RNaT-slot. Bug reported by Matt Chapman. ChangeSet@1.1136.71.2, 2003-12-01 09:23:57-06:00, Matt_Domsch@dell.com Merge dell.com:/home/mdomsch/bk/linux-2.4 into dell.com:/home/mdomsch/bk/edd/linux-2.4-edd-disksig ChangeSet@1.1231, 2003-12-01 10:36:38+11:00, nathans@bruce.melbourne.sgi.com [XFS] Remove some unused pagebuf source and header files. ChangeSet@1.1230, 2003-12-01 10:21:09+11:00, nathans@sgi.com [XFS] Add the noikeep mount option, make ikeep the default for now. SGI Modid: xfs-linux:slinx:162621a ChangeSet@1.1229, 2003-12-01 10:05:21+11:00, nathans@sgi.com [XFS] Update the way we hook into the generic direct IO code so we share more code. This means we no longer need to dup remove_suid within xfs_write_clear_setuid. SGI Modid: xfs-linux:slinx:162446a ChangeSet@1.1228, 2003-12-01 09:59:33+11:00, nathans@sgi.com [XFS] Convert to revised kmem shake interface. SGI Modid: xfs-linux:slinx:162426a ChangeSet@1.1227, 2003-12-01 09:56:21+11:00, nathans@sgi.com [XFS] Use a kmem shaking interface for 2.4 which is much more like the 2.6 one. SGI Modid: xfs-linux:slinx:162377a ChangeSet@1.1136.72.1, 2003-11-30 20:13:06+01:00, marcel@holtmann.org [PATCH] User level driver support for input subsystem This driver adds support for user level drivers for input subsystem accessible under char device /dev/input/uinput. ChangeSet@1.1136.71.1, 2003-11-30 14:55:41-02:00, marcelo@logos.cnet Changed Makefile to 2.4.23 TAG: v2.4.23