aboutsummaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2002-02-11Merge mainline to 2.5.4 final. This is veryHEADmasterdavem347-1150/+1584
possibly the last vger CVS merge I'll be doing as Linus moves over to Bitkeeper which I am already using to push 2.5.x changes to him. The 2.4.x CVS will remain active assuming Marcelo decides to not use BK for his work.
2002-02-11Sorry about forgetting to commitdavem1-3/+3
this.
2002-02-10implement writev() support (patch also sent to maintainer)ecd1-8/+55
2002-02-10make it compileecd1-3/+4
2002-02-10add TUNSET* ioctlsecd1-1/+8
2002-02-10use pci_memspace_mask for mmap calculationsecd1-4/+4
2002-02-10Merge error.davem2-4/+5
2002-02-10Make thread_saved_pc take a task_struct ptr.davem3-7/+7
2002-02-10Merge mainline to 2.5.4-pre5davem113-1110/+1263
2002-02-10Merge errors.davem2-2/+22
2002-02-10Forgotten commit.davem1-0/+210
2002-02-10No longer needed.davem1-20/+0
2002-02-09Merge mainline to 2.5.4-pre3davem159-2086/+4049
2002-02-09Merge mainline to 2.5.4-pre3davem4-72/+481
2002-02-08Fix build.davem1-5/+4
2002-02-08Merge mainline to 2.5.4-pre2davem251-2135/+2293
2002-02-07Merge mainline to 2.5.4-pre1davem110-1262/+1432
2002-02-02Forward port cpu_relax fix from 2.4: Implement a somewhat randomizedkanoj3-4/+12
version of cpu_relax() to alleviate livelock problems in the brlock code, seen when writer and reader cpus get instep with each other.
2002-02-01Update netfilter entry.davem1-3/+4
2002-02-01Revert bug-fix-undoings from acme patches.davem2-3/+3
2002-02-01Cure typos.davem1-2/+4
2002-02-01Move protocol specific state out of struct sock.davem118-2800/+3279
This is phase 1. Work by Arnaldo Carvalho de Melo.
2002-02-01When resetting due to a hard error condition, forcedavem1-3/+4
the PHY to be reset and auto-negotiation restarted.
2002-02-01Now remove central_bus pci_first_slot hackdavem1-4/+1
because the real bug is fixed.
2002-02-01Do not try to calculate IRQ of empty EBUSdavem1-1/+15
devices. Needed to work around a bug in OBP in Ex000 systems with PCI I/O boards. In that case it does not compute the fully specified INO for the interrupts property even though it is an onboard device and has forth firmware.
2002-01-31Mask out syscall tracer count correctly wrt.davem1-2/+2
2.5.3-final task_work changes.
2002-01-31Remove debugging printks.davem1-12/+0
2002-01-31Merge mainline to 2.5.3 final.davem220-16585/+11451
2002-01-30Do not offer sunlance twice.davem1-1/+0
2002-01-30Merge mainline to 2.5.3-pre6davem122-1959/+3997
2002-01-29Forward port 2.4 fix: Eliminate error printk when booting singlekanoj1-2/+4
cpu via maxcpus kernel command line. Also, make maxcpus work with the wait_init_idle changes.
2002-01-25Kill empty statement warning.davem1-0/+1
Posting by Ingo to linux-kernel.
2002-01-25Set p->cpu in wake_up_forked_process.davem1-0/+1
If the parent reschedules before we hit this code, the cpu will be inaccurate. Posted by Ingo on linux-kernel.
2002-01-25Cure missing initialize of i_frag_size anddavem2-0/+2
i_dir_acl.
2002-01-24Merge mainline to 2.5.3-pre5davem158-25361/+29482
2002-01-24Update.davem1-0/+4
2002-01-24Fix MOD_{INC,DEC}_USE_COUNT usage, it is notdavem1-5/+5
a function. :-)
2002-01-24Fix prototype of afinet6_get_info, from Jeff Garzik.davem1-2/+2
2002-01-24Fix sbus_init declaration.davem2-5/+9
Include linux/init.h in sysv/inode.c
2002-01-24Quick untested merge to 2.5.3-pre4davem157-1218/+4425
2002-01-24Fix typo.davem1-1/+1
2002-01-23Cure warnings.davem2-1/+2
2002-01-23Update.davem1-0/+2
2002-01-23Move route_me_harder into a header,davem4-96/+79
fixes all cases where this duplicated function was buggy. From Harald Welte
2002-01-23Mac PHY updates from Ben.davem2-9/+34
2002-01-23Kill get_fast_time.davem20-56/+15
2002-01-23Forgotten ntohs on ipv6 header payload_len.davem1-5/+7
2002-01-23Netfilter fixes to the mainline.davem29-53/+1646
2002-01-23Merge mainline to 2.5.3-pre3davem431-8007/+20557
2002-01-19Remove __FUNCTION__ cpp pasting, gcc-3.1 warnsdavem2-5/+3
about it.
2002-01-17Correct netdev_fastroute_obstacles packet socketdavem2-1/+5
check. Export netdev_fastroute for drivers.
2002-01-17Same MAC on multiple interfaces is legal.davem1-2/+3
From Lennert Buytenhek.
2002-01-16Merge mainline to 2.5.3-pre1davem97-2262/+6767
2002-01-16Fix the build.davem1-4/+8
2002-01-15Linus wants lowercase names for the PCI_UNMAP_foo interfaces.davem13-79/+79
2002-01-15Revert 2.4.x changes.davem4-449/+161
2002-01-15Mark cleanup function with __exit.davem1-1/+1
2002-01-15Missed during previous merge.davem1-7/+4
2002-01-15Eliminate use of strtok. From Rene Scharfe.davem1-17/+12
2002-01-15Fix destroyme timer handling, for real.davem1-18/+34
From Paul Russell.
2002-01-15Kill spurious assignment. From acme.davem1-2/+2
2002-01-15Do not use v9 branches out of fixupdavem1-6/+6
section.
2002-01-15Revert local change, will go to 2.5.x whendavem1-10/+0
Trond merges his latest nfs stuff in.
2002-01-15Revert bogus local changes.davem1-9/+77
2002-01-15Missed commit during 2.5.2 merge.davem1-1/+3
2002-01-15Fix filp_open error return check, by Al Viro.davem1-1/+1
2002-01-15Fixup the build, kill warnings.davem11-56/+56
2002-01-15Make warning msg more politically correct :)davem20-21/+20
2002-01-15Fixes from Benjamin. Make sure we stop thedavem1-24/+12
chip and (re-)init the PHY equally on all variants.
2002-01-15Merge mainline to 2.5.2-final.davem152-1649/+3391
2002-01-15Fix build...davem1-1/+1
2002-01-15Fix build.davem1-1/+1
2002-01-15Fix build...davem1-2/+7
2002-01-15Back out a lot of local changes which willdavem142-3772/+2829
get propagated from 2.4.x anyways and are only getting in the way of my work currently.
2002-01-14Stop the lance before requesting the IRQ.davem1-3/+3
2002-01-14sg_ioctl_trans needs to copy in and sizedavem1-4/+9
the sbp data area properly. Fix from Luis F. Ortiz
2002-01-14When skb_recv_data can return NULL skb withdavem1-2/+6
err == 0 when receive side is shutdown. unix_accept needs to take care about this and really return on newsocket resulting in an OOPS.
2002-01-14When output interface changes, so doesdavem1-11/+23
hh_len, realloc if necessary. From Paul Russell.
2002-01-14Fetch assigned-address property, even for EBUS.davem1-11/+12
2002-01-12Include rtnetlink.h to get BUG_TRAPdavem1-1/+2
2002-01-12Fix compile failures when CONFIG_INET is notdavem3-4/+3
set.
2002-01-12extern+inline --> static+inlinedavem3-10/+10
From Keith Owens.
2002-01-12Lower memory clock used on Rage/XL.davem1-1/+1
2002-01-12Do not GC embrions about to be attacheddavem2-5/+19
to listening socket. Fix from Alexey.
2002-01-12Fix potential local buffer overflowdavem1-2/+2
in rt_cache_get_info.
2002-01-12Allow IP packets with payload shorter thandavem2-4/+7
8 bytes.
2002-01-12SunGEM updates from Benjamin Herrenschmidt.davem2-182/+154
2002-01-12Add entry for CONFIG_VLAN_8021Qdavem1-0/+10
2002-01-12Add /proc/tty/driver/serial support todavem1-4/+85
Zilog driver, from Alex Buell.
2002-01-12Revert meaningless local change.davem1-1/+1
2002-01-12Remove my debugging code :-)davem1-10/+10
2002-01-12Merge mainline to 2.5.2-pre11 + Ingo sched H6.davem31-220/+560
2002-01-11Do not send reschedules to self, fixes sparc64 SMPdavem1-1/+1
problems.
2002-01-11Known broken commit of 2.5.2-pre10 + sched-01-2.5.2-pre10-E1davem144-1632/+4162
I want to move my debugging over to a machine other than my workstation :-)
2002-01-09Back out local 2.5.x idsn differences.davem14-438/+146
2002-01-08Merge mainline to 2.5.2-pre9davem1049-14386/+42463
2002-01-08Fix swift_flush_cache_range first argdavem1-1/+2
handling even tho it is commented out.
2002-01-07Revert bogus cleanup.davem1-2/+0
2002-01-05On PSYCHO, only set pci_first_slot for PBM-B to 1davem1-10/+6
if we are on an Enterprise-xxxx system. This needs to be done better at some point...
2002-01-05The rs_init changes are causing occaisional problemsdavem4-8/+12
on my SBUS dual-Ultra so revert this for now.
2002-01-04Use R3 for interrupt status reading insteaddavem1-124/+79
of R2. Change receive_chars to read the status byte with each incomming character.
2002-01-04Add missing dev_put, from Lennert Buytenhek.davem1-1/+2
2002-01-03Fix sparc32 build.davem1-1/+1
2001-12-26Forward port 2.4 fix: Fix SMP race during bootup. The slaves shouldkanoj1-1/+1
not be allowed to do cpu_idle() and clear their masks by calling init_idle() before the master has a chance to set up wait_init_idle properly.
2001-12-24Port state handling fixes to br input processing.davem1-58/+67
From Lennert Buytenhek.
2001-12-24Remove erroneous branch in __handle_user_windows.davem1-3/+2
Noticed by Kanoj.
2001-12-24Correct printk message in conntrack handling.davem3-32/+33
Move some conntrack enums to header visible to userspace. From netfilter maintainers.
2001-12-24Bonding updates from bonding driver sourceforgedavem2-66/+131
project.
2001-12-24Add multiple inclusion protection.davem10-3/+46
2001-12-24Do not allow enslavement of a bridge device todavem3-4/+8
another bridge device, causes recursion and does not make any sense anyways. From Lennert Buytenhek.
2001-12-24Make bridge dep_tristate on CONFIG_INET.davem2-6/+2
Remove CONFIG_INET ifdefs from bridging code. From Lennert Buytenhek.
2001-12-24Remove rpciod_tcp_dispatcher extern.davem1-1/+0
2001-12-24Add ARPHRD_IEEE80211_PRISM.davem1-0/+1
2001-12-22Include sunrpc/xprt.hdavem1-0/+1
2001-12-21Include linux/init.hdavem1-0/+1
2001-12-21Quick unchecked merge to 2.5.2-pre1davem40-912/+809
2001-12-21Recognize timestamp negotiation properlydavem1-1/+2
for ipv6, whoops. Fix from Alexey.
2001-12-21Make flush_{tlb,cache}_range() take adavem131-429/+600
vm_area_struct instead of an mm_struct. Update all callers and documentation. On sparc64, when we are unmapping a SHARED+WRITABLE mapping call flush_dcache_page on all cpus for dirty pages mapped into that area.
2001-12-21Define PCI_DMA_BUS_IS_PHYS to 0.davem1-0/+1
2001-12-21Kill virt_to_bus/bus_to_virt.davem1-3/+2
Define page_to_phys.
2001-12-21Define a dma64_addr_tdavem1-1/+2
2001-12-21Include tty.hdavem1-1/+2
2001-12-21Max check_asm more robust, from Jakub.davem2-4/+4
2001-12-21Sparc32 build fixes for mainline.davem8-11/+18
2001-12-20Kill memclear_highpage, unused.davem1-11/+1
Do not forget to call flush_dcache_page in memclear_highpage_flush, this is used for truncate operations.
2001-12-19Bring range/reg fixes to the mainline.davem5-38/+14
2001-12-18Kill explicit rs_init calls and make it module_init.davem4-12/+8
2001-12-18Sparc64 thread fix for 2.5.x mainlinedavem1-1/+2
2001-12-17Bring back invariant that arch-specificdavem1-0/+1
portions of scatterlist entries will be zero'd out.
2001-12-17Revert kill(-1) behavioral change in 2.5.1davem1-5/+3
2001-12-17Fix some remaining references to dead 'address'davem14-102/+110
member of scatterlist. INQUIRY scsi commands never use scatterlists so consider such a case a BUG in aic7xxx_old.c
2001-12-17bus_to_virt/virt_to_bus are now illegal.davem96-488/+376
Kill 'address' member of scatterlist. The 'bounce_buffers' member of the scsi command struct is now never set, kill it. Convert some ISA devices over to isa_virt_to_bus et al. Start marking PCI drivers as broken if they reference virt_to_bus or bus_to_virt.
2001-12-17Free old *PSKB before assigning NSKBdavem1-1/+1
to it. From linux-kernel.
2001-12-17Merge mainline to 2.5.1-final.davem149-1692/+1187
2001-12-16Trident DMA mask fix for the mainline.davem1-2/+10
2001-12-13Turn back on RAID1 as a module.davem1-1/+1
2001-12-13Remove if 0 around asm/io.h include.davem1-2/+0
2001-12-13Synchronize with Axboe's latest sources.davem15-104/+112
2001-12-13Quick merge to 2.5.1-pre11davem52-2214/+2452
2001-12-13Purge ISA DMA bouncing garbage from scsi_merge.cdavem4-252/+13
Handle unchecked_isa_dma in scsi layer by setting the bounce_limit for the queue not the segment boundary mask :-) Patch sent to Jens.
2001-12-13Forward port netfilter fixes from 2.4.x branch.davem8-16/+98
2001-12-13Reimplement IOMMU bio layer support so thatdavem23-398/+302
it works. Patch sent to Jens.
2001-12-13Fix icmp docs in mainline too.davem2-10/+17
2001-12-13Kill DMA_CHUNK_SIZE and related code.davem14-142/+37
Rename request nr_segments to nr_phys_segments.
2001-12-12Fix MERGEABLE_BUFFERS check in backdavem1-1/+1
merge function, args should be req->biotail, bio
2001-12-12Put back asm/irq.h include.davem1-2/+3
Bump driver version/reldate.
2001-12-12Reference iounmap arg to avoid unuseddavem1-2/+2
variable warnings.
2001-12-12Revert previous change.davem1-1/+1
2001-12-11Fix to use proper size_t printfdavem3-4/+1
format string. Kill unused local car in old-tulip driver. Forward port parport 2.4.17-preX merge error fix.
2001-12-11Merge mainline to 2.5.1-pre10davem24-525/+550
2001-12-11Remove forced compile failuredavem1-2/+4
and replace with hack to make no writes actually go to disk. You can run with this thing until page/block memory starts getting reclaimed. Basically every write request just gets marks UPTODATE and no I/O is actually initiated.
2001-12-11Disable DMA_CHUNK_SIZE, less variablesdavem1-1/+3
to debug.
2001-12-11Merge error.davem4-1542/+0
2001-12-11Dont code very late at night kiddies or you toodavem1-2/+5
will make mistakes like this...
2001-12-11Resolve merge errors.davem1-2/+14
2001-12-11Include asm/irq.hdavem1-0/+1
2001-12-11Forgotten commit.davem2-4/+5
2001-12-11Allow to compile with DMA_CHUNK_SIZE undefined.davem1-1/+6
2001-12-11Quick unchecked merge to 2.5.1-pre9davem49-277/+2629
2001-12-11Merge mainline to 2.5.1-pre8davem439-9259/+12533
This is being checked in knowing it is BROKEN on sparc64. My SBUS/ESP workstation gets full on filesystem corruption with this tree. I would advise staying away from this tree until I figure things out.
2001-12-11Use PCI_IRQ_NONE, only run powerd if IRQ isdavem1-8/+11
present.
2001-12-08Bump to 2.5.0davem1-3/+3
2001-12-08Fixing ,v file in repository.davem1-0/+60
2001-12-08Merge mainline to 2.4.17-pre6davem154-1743/+2601
2001-12-07Add in missing __raw_readq definition similar to other 64bit alpha/ia64kanoj1-1/+2
platforms.
2001-12-07Merge mainline to 2.4.17-pre5davem71-3127/+3253
2001-12-06Perform resched+signal checks atomicallydavem1-29/+101
with PSTATE_IE disabled, and keep it disabled all the way back to userspace. Also make register l6 handling more consistent.
2001-12-06Merge errordavem2-281/+0
2001-12-05Delete obsolete comment about MODULES_VADDRdavem1-4/+1
being defined and described in modutil.c
2001-12-05Merge errordavem1-1/+1
2001-12-05Merge mainline to 2.4.17-pre4davem40-228/+2296
2001-12-05Fix OOPS on module unload, fromdavem2-4/+1
Eran Man. Set NETIF_F_DYNALLOC in netdev features. Do not kfree dev in vlan_dev_destruct.
2001-12-05Make ipt_unclean deal with ECE/CWR flagdavem1-1/+4
bits correctly. Fix from Harald Welte.
2001-12-05Fix tcp_v4_search_req for the ICMP case,davem1-9/+12
we were using reversed src/dest. Discovered by Julian Anastasov. Note: IPv6 has the same bug but I need to ask Alexey about something before I change it.
2001-12-05SunGEM fixes from Banjamin.davem1-18/+36
Only depend upon BMSR_LSTATUS in gem_link_timer. Call gem_init_pause_thresholds after gem_init_mac not before or else we will lose the settings. Set MIF_CFG bits correctly in gem_apple_powerup. Make gem_ioctl take the pm_sem.
2001-12-05Move kernel/obp/modules/vmalloc area defines intodavem3-31/+20
one place, asm/pgtable.h
2001-12-05Merge mainline to 2.4.17-pre2davem106-1143/+1689
2001-12-05Add Cheetah fully assosciative TLB loaddavem4-7/+27
workaround to blockops, forgot this. Move MODULES_VADDR up a notch so it does not conflict with TLBTEMP_BASE. In all such critical spots, point the reader at some new commentary in mm/modutil.c that explains the location of all these things.
2001-12-05Do not assume KERNBASE == 4MB.davem1-7/+6
2001-11-30I really should not be let out of thedavem1-2/+2
cage sometimes...
2001-11-30Don't set bigkernel only on spitfire :-)davem1-2/+2
2001-11-30Make it build.davem1-2/+2
2001-11-30Add support for >4MB kernels, only reallydavem4-15/+88
useful right now for net booting as SILO would have big problems with such huge kernels. From Kanoj.
2001-11-30_copy_page has no more references, it may nowdavem3-64/+3
be killed.
2001-11-30copy_highpage can now die, no more references existdavem1-11/+0
2001-11-30Do not use plain copy_highpage on a normaldavem1-1/+1
COW user mapping, that is what copy_user_highpage is for.
2001-11-30Typo.davem1-1/+1
2001-11-30Fix cheetah nloops claculation.davem1-4/+6
2001-11-30Add CONFIG_HOTPLUG option, remove spuriousdavem2-3/+4
CONFIG_FB option as drivers/video has it.
2001-11-29With the revert of the bogus kernel/exec_domain.cdavem1-19/+7
changes in 2.4.17-pre1, we can go back to the older and simpler sparc64_personality implementation from revision 1.52 of arch/sparc64/kernel/sys_sparc.c
2001-11-29Include asm/irq.hdavem1-1/+2
2001-11-29Fix MII register name error.davem1-2/+2
2001-11-29Make the spitfire #tlb entries dependency clear in the tlbflush code.kanoj2-5/+6
2001-11-29DecNET updates from Steven Whitehouse:davem3-55/+126
1) Fix port allocation bug. 2) Doc updates.
2001-11-29Use INIT_BLOCK_TX_RING_SIZE _not_davem1-2/+2
TX_RING_SIZE to calculate rx ring DMA base address.
2001-11-29Only reports gigabit as supported ifdavem1-2/+11
gp->gigabit_capable is true. Also, recognize Enable Semiconductor phys.
2001-11-29Recognize 0x18074c0 as Lucent PHY.davem1-31/+42
2001-11-29Fix locking error.davem1-2/+6
2001-11-29Do not include linux/irq.hdavem1-2/+1
2001-11-29PPC support updates from Banjamin Herrenscmidtdavem2-242/+893
generated by cgit 1.2.3-korg (git 2.39.0) at 2024-05-19 02:43:45 +0000