Andrew Gallatin 0617522889 Fix an alpha-only race which causes the transmit side of the chip to ... lock up under moderate to heavy load. The status & command fields share a 32-bit longword. The programming API of the eepro apparently requires that you update the command field of a transmit slot that you've already given to the card. This means the card could be updating the status field of the same longword at the same time. Since alphas can only operate on 32-bit chunks of memory, both the status & command fields are loaded from memory & operated on in registers when the following line of C is executed: sc->cbl_last->cb_command &= ~FXP_CB_COMMAND_S; The race is caused by the card DMA'ing up the status at just the wrong time -- after it has been loaded into a register & before it has been written back. The old value of the status is written back, clobbering the status the card just DMA'ed up. The fact that the card has sent this frame is missed & the transmit engine appears to hang. Luckily, as numerous people on the freebsd-alpha list pointed out, the load-locked/store-conditional instructions used by the atomic functions work with respect changes in memory due to I/O devices. We now use them to safely update the command field. Tested by: Bernd Walter <ticso@mail.cicely.de>		2000-07-19 14:33:52 +00:00
..
agp_ali.c	A driver for programming the AGP hardware. This is only very lightly	2000-06-09 16:04:30 +00:00
agp_amd.c	Release resources properly in detach.	2000-06-10 17:53:20 +00:00
agp_i810.c	Add support for Intel's i810 chipset with integrated graphics. An	2000-07-12 10:13:07 +00:00
agp_if.m	A driver for programming the AGP hardware. This is only very lightly	2000-06-09 16:04:30 +00:00
agp_intel.c	A driver for programming the AGP hardware. This is only very lightly	2000-06-09 16:04:30 +00:00
agp_sis.c	A driver for programming the AGP hardware. This is only very lightly	2000-06-09 16:04:30 +00:00
agp_via.c	A driver for programming the AGP hardware. This is only very lightly	2000-06-09 16:04:30 +00:00
agp.c	Add support for Intel's i810 chipset with integrated graphics. An	2000-07-12 10:13:07 +00:00
agppriv.h	Add support for Intel's i810 chipset with integrated graphics. An	2000-07-12 10:13:07 +00:00
agpreg.h	Add support for Intel's i810 chipset with integrated graphics. An	2000-07-12 10:13:07 +00:00
agpvar.h	A driver for programming the AGP hardware. This is only very lightly	2000-06-09 16:04:30 +00:00
alpm.c	Remove ~25 unneeded #include <sys/conf.h>	2000-04-19 14:58:28 +00:00
amd.c	Remove ~25 unneeded #include <sys/conf.h>	2000-04-19 14:58:28 +00:00
amd.h	Back out the previous change to the queue(3) interface.	2000-05-26 02:09:24 +00:00
cy_pci.c	Ahhrggg. Put the test for the compat shims AFTER the file that includes	2000-03-27 20:24:02 +00:00
cy_pcireg.h
dc21040reg.h
ide_pcireg.h
if_ar_p.c	Ahhrggg. Put the test for the compat shims AFTER the file that includes	2000-03-27 20:24:02 +00:00
if_dc.c	Add the PCI IDs for the Macronix 98727 and 98732 parts. These are	2000-07-17 19:27:41 +00:00
if_dcreg.h	Add the PCI IDs for the Macronix 98727 and 98732 parts. These are	2000-07-17 19:27:41 +00:00
if_de.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_devar.h	Use bus_space for all register accesses.	2000-05-13 18:47:24 +00:00
if_en_pci.c	Ahhrggg. Put the test for the compat shims AFTER the file that includes	2000-03-27 20:24:02 +00:00
if_fxp.c	Fix an alpha-only race which causes the transmit side of the chip to	2000-07-19 14:33:52 +00:00
if_fxpreg.h	Added support for the i82559ER (10/100Mbps NIC for embedded applications).	2000-06-18 10:26:09 +00:00
if_fxpvar.h	Added support for cards and on-motherboard NICs that use an SEEPROM	2000-03-28 04:41:42 +00:00
if_mn.c	Newbusify this driver at an expense of 40 extra lines of code.	2000-07-10 19:33:41 +00:00
if_rl.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_rlreg.h
if_sf.c	Close PR# 19904: remember to call sf_setmulti() when initializing the	2000-07-14 19:11:02 +00:00
if_sfreg.h
if_sis.c	sis_miibus_writereg() was checking for SIS_TYPE_900 and phy != 0 twice	2000-07-14 05:30:48 +00:00
if_sisreg.h	Add support for the National Semiconductor DP83815 fast ethernet	2000-07-06 06:02:04 +00:00
if_sk.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_skreg.h	Back out the previous change to the queue(3) interface.	2000-05-26 02:09:24 +00:00
if_sr_p.c	Ahhrggg. Put the test for the compat shims AFTER the file that includes	2000-03-27 20:24:02 +00:00
if_ste.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_stereg.h
if_ti.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_tireg.h	Back out the previous change to the queue(3) interface.	2000-05-26 02:09:24 +00:00
if_tl.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_tlreg.h
if_tx.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_txvar.h	Added support for SMC9432BTX cards.	2000-06-21 19:19:49 +00:00
if_vr.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_vrreg.h	Close PR# 19617: add support for VIA VT6102 NICs to VIA Rhine driver.	2000-07-05 21:37:21 +00:00
if_wb.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_wbreg.h
if_wx.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_wxreg.h	Fix this driver to (finally) work with switches. Some more black	2000-06-16 06:28:31 +00:00
if_wxvar.h	Fix this driver to (finally) work with switches. Some more black	2000-06-16 06:28:31 +00:00
if_xl.c	Make all Ethernet drivers attach using ether_ifattach() and detach using	2000-07-13 22:54:34 +00:00
if_xlreg.h
intpm.c	Fix some leakage from inside #ifdef PCI_COMPAT (the FreeBSD 2.2 compat	2000-05-28 16:17:18 +00:00
intpmreg.h
isp_pci.c	Keep interrupts blocked for all of isp_pci_attach. Redo DMA routines	2000-07-18 06:40:22 +00:00
locate.pl	use warnings	2000-06-27 22:41:12 +00:00
meteor_reg.h	Add $FreeBSD$	2000-05-01 20:32:07 +00:00
meteor.c	Ahhrggg. Put the test for the compat shims AFTER the file that includes	2000-03-27 20:24:02 +00:00
ncr.c	ncr.c is no longer a COMPAT_OLDPCI device - remove the warning.	2000-05-28 10:45:06 +00:00
ncrreg.h
ohci_pci.c	Only display the 'Disable PNP-OS BIOS option' message on the i386 arch.	2000-05-11 20:17:28 +00:00
pccbb.c
pccbbreg.h
pccbbvar.h
pci_compat.c	Encapsulate the old PCI compatability support and APIs completely under	2000-05-28 16:35:57 +00:00
pci_if.m	* Factor out the object system from new-bus so that it can be used by	2000-04-08 14:17:18 +00:00
pci.c	Nuke the useless chip driver. It gets in the way when you want to load	2000-06-09 16:00:29 +00:00
pcic_p.c	Make this compile without depending on the FreeBSD 2.2 compatability	2000-05-28 16:31:35 +00:00
pcic_p.h	Make this compile without depending on the FreeBSD 2.2 compatability	2000-05-28 16:31:35 +00:00
pcireg.h	Encapsulate the old PCI compatability support and APIs completely under	2000-05-28 16:35:57 +00:00
pcisupport.c	Nuke the useless chip driver. It gets in the way when you want to load	2000-06-09 16:00:29 +00:00
pcivar.h	Nuke the useless chip driver. It gets in the way when you want to load	2000-06-09 16:00:29 +00:00
scsiiom.c	Add $FreeBSD$	2000-05-01 20:32:07 +00:00
simos.c	Separate the struct bio related stuff out of <sys/buf.h> into	2000-05-05 09:59:14 +00:00
simos.h
ti_fw2.h	Update the Tigon firmware to 12.3.21. This fixes a few bugs and adds support	2000-04-24 17:44:45 +00:00
ti_fw.h	Update the Tigon firmware to 12.3.21. This fixes a few bugs and adds support	2000-04-24 17:44:45 +00:00
uhci_pci.c	Add 440MX chipset.	2000-05-24 02:24:38 +00:00
xmaciireg.h	Reoganize/update the SysKonnect driver:	2000-04-22 02:16:41 +00:00
xrpu.c	Do not depend on FreeBSD 2.2 compatability defines.	2000-05-28 16:18:58 +00:00