Marcelo Tosatti: Linux 2.4.27-pre5

Kernel.org Mirrors

FullChangelog

Hi,

Here goes -pre5.

This time we have merges from Jeff's -netdrivers tree, David's -nettree, including a fix for compilation error without CONFIG_SCTP set,SPARC64 update,i810_audio fixes, amongst others.

Summary of changes from v2.4.27-pre4 to v2.4.27-pre5

:

• x86-64 only: ia32entry.S reg changes dropped during debugging

:

• [netdrvr wan] farsync update

:

• tulip PCI ID for D-Link DFE-680TXD

:

• fealnx #0: replace dev->base_addr with ioaddr; sync with 2.6
• fealnx #1: replace magic constants with enums
• fealnx #2: add 'static'; fix wrapped comment
• fealnx #3: fix pointer substraction bug
• fealnx #4: stop doing stop_nic_rx/writel(np->crvalue) inrest_rx_descriptors()
• fealnx #5: introduce stop_nic_rxtx(), use it where makes sense
• fealnx #6: Francois' fixes for low memory handling; removefree_one_rx_descriptor (not used anymore)
• fealnx #7: Garzik fix (IIRC): add locking to tx_timeout
• fealnx #8
• fealnx #9
• fealnx #10
• fealnx #11

Andrew Morton:

• 8139too not running s3 suspend/resume pci fix

Arjan van de Ven:

• small change for scsi 2.6 header compatibility
• [libata] Use standard headers from include/scsi, not drivers/scsi

Daniele Venzano:

• [netdrvr sis900] fix ISA bridge detection
• [netdrvr sis900] cosmetic header cleanups
• [netdrvr sis900] fix missing netif_device_detach() in suspend

David S. Miller:

• [TCP]: Fix build in 2.4.x with SCTP disabled
• [PKT_SCHED]: Missing rta_len init in sch_delay
• [SPARC64]: gcc-3.4.x build fixes

Don Fry:

• 2.4.27-pre3 pcnet32 add static to two routines
• 2.4.27-pre3 pcnet32 avoid hard hang with some chip variants
• 2.4.27-pre3 pcnet32 correct 79C976 variant string
• 2.4.27-pre3 pcnet32 fix boundary comparison bug
• 2.4.27-pre3 pcnet32 remove timer and complexity
• pcnet32: limit frames received during interrupt
• pcnet32: fix bogus carrier errors with 79c973
• pcnet32: correct printk for big-endian arch
• pcnet32: avoid timeout with tcpdump
• pcnet32: fix for patch 8 le16_to_cpu

François Romieu:

• [netdrvr r8169] update to 2.6.x version; many fixes and changes

Ganesh Venkatesan:

• e1000 1/7: Clear auto-mdix mode when forcing link to
• e1000 2/7: Workaround for link LED staying ON even when
• e1000 3/7: Determine Link Status correctly while using
• e1000 4/7: Rewrite logic to estimate # of tx descriptors
• e1000 6/7: ethtool_ops support
• e1000 7/7: Error Logging support (enabled/disabled via
• e100 1/1: Update to sync up version numbers

Jeff Garzik:

• [sound i810] pci id cleanups
• [libata] add new ->bmdma_setup hook
• [libata] use new ->bmdma_{start,setup} method to properlysupport ATAPI
• [libata] more ATAPI work - translate SCSI CDB to ATA PACKET
• [libata] random minor bug fixes
• [libata] kill ATA_QCFLAG_POLL flag
• [libata] internal cleanups
• [libata] minor stuff
• [libata] handle non-data ATAPI commands via interrupt
• [libata] DMADIR support
• [libata] remove redundant use of ATA_QCFLAG_SG in ATAPI packettranslation
• [libata] SCSI->ATA simulator hacking: INQUIRY command
• [libata] comments and constants
• [libata] scsi simulator improvements: MODE SENSE, SEEK(6,10),REZERO_UNIT
• [libata] replace ATA_QCFLAG_ATAPI with inline helper
• [libata] polish DocBook docs a bit
• [netdrvr sis900] sync with 2.6.x
• [netdrvr e1000] use generic ethtool_ops provided innet/core/ethtool.c
• [libata promise] revert broken taskfile delivery change
• [libata scsi] ack SYNCHRONIZE CACHE command

Marcelo Tosatti:

• Changed EXTRAVERSION to -pre5

Patrick McHardy:

• [IPV4,6]: Fix off-by-one in max protocol-type check

Stefan Rompf:

• [netdrvr b44] always restore PCI config on resume