Age | Commit message (Collapse) | Author | |
---|---|---|---|
2006-10-03 | make sure to call SLIST_INIT on sc->rx_freelist. | Damien Bergamini | |
sanity check length field in RX descriptors. | |||
2006-10-03 | don't use IF_PREPEND() on altq's. | Damien Bergamini | |
use IFQ_POLL()/IFQ_DEQUEUE() logic instead as described in altq(4). tested by jolan@ on macppc "diffs look ok" brad@ | |||
2006-10-03 | move the advice about smart, readattr, and sec* to before the command | Jason McIntyre | |
list, since it doesn;t make sense to put these in any one single command description (nor indeed to repeat them for every command); ok grange | |||
2006-10-03 | Enable altq in acx(4) the tx path is already handling all cases correctly. | Claudio Jeker | |
OK mglocker@ | |||
2006-10-03 | Do not use IF_PREPEND. Instead use IFQ_POLL and IFQ_DEQUEUE. This should | Claudio Jeker | |
"fix" altq. OK mglocker@ | |||
2006-10-03 | correct check for proper return value of dup2 | Kurt Miller | |
2006-10-03 | provide a summary of the available commands, since there's a ton of | Jason McIntyre | |
them; | |||
2006-10-03 | put the command list into some semblance of order; | Jason McIntyre | |
2006-10-03 | when using SHA1Data(), no CTX is needed, so zap that line in the example | Alexander von Gernler | |
ok jmc@ | |||
2006-10-03 | sync | Theo de Raadt | |
2006-10-03 | Last Part of file descriptor race and deadlock corrections. | Kurt Miller | |
When a fd enters the closing state prevent any threads from polling the fd and reschedule the thread with the closing_fd flag set. This fixes a class of deadlocks where a thread is blocked waiting for data (that may never arrive) and a later thread calls close() or dup2() on the fd. okay brad@ | |||
2006-10-03 | upon success dup2() returns newfd so don't overwrite it with | Kurt Miller | |
_thread_fd_table_init()'s return value and blow it away. fixes firefox plugin problem noticed by jolan@ | |||
2006-10-03 | Don't load bad144 bad sector information into disklabels for archs that | Kenneth R Westerback | |
don't have wd* in GENERIC, since wd* is the only place such info is being used. First step of eliminating bad144 bad sector info from all disklabels. 'go go go' deraadt@ | |||
2006-10-03 | strtonum() with INT_MAX intead of LONG_MAX, problem pointed out by ↵ | Theo de Raadt | |
pierre-yves@spootnik.org | |||
2006-10-02 | Fix error messages, okay ray@ deraadt@ | Pedro Martelletto | |
2006-10-02 | Zap unused MLME routines. Probably they were just used in WPA mode. | Marcus Glocker | |
ok claudio@ | |||
2006-10-02 | Fill October | Jasper Lievisse Adriaanse | |
ok mickey | |||
2006-10-02 | Fully sort MLINKS. | Michael Knudsen | |
ok jmc | |||
2006-10-02 | various tweaks to smarten this page up a bit; | Jason McIntyre | |
2006-10-02 | Remove old ACPI device page builds (reorder build order alphabetically | Michael Knudsen | |
while there). MLINKS+= the new acpi(4) page to the ACPI device page names. ok jmc | |||
2006-10-02 | Remove old ACPI pages. | Michael Knudsen | |
2006-10-02 | Merge the ACPI device man pages into acpi(4) after discussion with and | Michael Knudsen | |
input from jmc and marco. ok jmc marco | |||
2006-10-02 | standard EXAMPLES; | Jason McIntyre | |
2006-10-02 | simplify SYNOPSIS and sync usage(); | Jason McIntyre | |
2006-10-02 | Simplify device attachment. From brad@ with a tiny correction. | Marcus Glocker | |
2006-10-02 | sync the ServerWorks ids. | Brad Smith | |
2006-10-02 | regen | Brad Smith | |
2006-10-02 | revert this change. after a little more digging I found some conflicting | Brad Smith | |
information. I'm not really sure if this is a compat mode id or not. so just revert back to SATA and label it with the chipset it comes from. | |||
2006-10-02 | One <arm/exec.h> to bind them all. | Miod Vallat | |
2006-10-02 | regen | Brad Smith | |
2006-10-02 | gah. ensure the entries are unique. | Brad Smith | |
2006-10-02 | add the ServerWorks PCIe bridge ids and correct an unknown ServerWorks | Brad Smith | |
SATA entry; it seems this id is the HT-1000 SATA controller but in an IDE compat mode depending on a BIOS setting. | |||
2006-10-02 | Fix off by one comparison. FSMAXTYPES is the last valid index into | Kenneth R Westerback | |
fstypesnames[], but that entry is NULL. The last valid file system name is at fstypesnames[FSMAXTYPES-1]. Encountered and diagnosed by pedro@. ok pedro@ deraadt@ | |||
2006-10-02 | get rid of a boolean typedef. this is c, we have ints, deal with it. | David Gwynne | |
ok marco@ krw@ | |||
2006-10-02 | remove some unneccessary escaping; | Jason McIntyre | |
2006-10-02 | - simplify a list item | Jason McIntyre | |
- use .Ex | |||
2006-10-02 | reword a clunky sentence; | Jason McIntyre | |
2006-10-02 | remove variable we do not need | Theo de Raadt | |
2006-10-02 | Since monitor mode works now, mention it. | Marcus Glocker | |
2006-10-02 | do not bother to read the esr and isr registers in the status routine just | Brad Smith | |
to end up not using the information read. | |||
2006-10-02 | return EINVAL for the Gig manual media types when attached to a | Brad Smith | |
Fast Ethernet PHY. | |||
2006-10-02 | sync | Theo de Raadt | |
2006-10-02 | enable bnx firmware builds | Theo de Raadt | |
2006-10-02 | actually put all the require firmware bits into the file | Theo de Raadt | |
2006-10-02 | new structure for the firmware blob; ok brad | Theo de Raadt | |
2006-10-02 | move the checksum stuff under IXGB_CSUM_OFFLOAD. | Brad Smith | |
2006-10-01 | Make monitor mode work by prepending the right kind headers and fixing the | Claudio Jeker | |
channel selection code. Joint work with mglocker@, OK mglocker@ | |||
2006-10-01 | Get rid of this strange pgt_data_frame which is unused. Nice side-effect is | Claudio Jeker | |
an easier to read pgt_datarx_completion. OK mglocker@ | |||
2006-10-01 | bye bye maestro_reg* | Marc Espie | |
2006-10-01 | zap maestro_reg.h, since it's only used in the kernel. | Marc Espie | |
okay deraadt@ |