- To get 16 bit one's complement value from uint32_t variable,
higher 16 bits should be ignored.
- RFC 1624 describes methods to recalculate checksum field in headers,
i.e. one's complement of one's complement sum that could be 0x0000,
but we don't have to use the strategy to deduct one's complement sum
itself which won't be zero but should be 0xffff.
Found on debugging mec(4) on sgimips O2.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)
- bus_dmamap fixes, with checks against NULL
- move the rings' allocation code later in iwi_attach() to avoid NULL pointer
dereference if allocation fails
- avoid double free() in case of failure during attach. If an allocation fails,
do not free the ring directly, as it is handled by iwi_detach()
- only set ring->count for RX/TX rings when ring allocation is successful, or
else the for loop during detach will fail with a NULL dereference.
- call pci_intr_disestablish() if iwi_reset() fails during attach; driver is
in an incoherent state, interrupt handler should not be used.
Proposed by me on tech-net@, approved by christos@. Compiled and tested
with GENERIC and XEN3_DOM0 kernels.
in IP headers, so we have to deduct not only IP option headers but all
IP headers. But in TCP/UDP layer we can assume the IP header is valid
and a sum of the IP header part should be 0xffff, so we don't have to
bother to deduct it from the computed checksum.
which don't have EXT_RFA and IPCB support. From hme(4) driver and
FreeBSD's fxp(4). Tested on i82559.
XXX: Probably we should have a common function to parse RX packet headers
XXX: to handle a raw checksum value and share it among hme(4) and gem(4) etc.
already powered down. This change prevents PCI bus exceptions
(indicated by NMI) when, for example, I detach the whole device
tree on a Soekris net4521:
com1: detached
NMI ... going to debugger
Stopped in pid 0.1 (system) at netbsd:sipcom_stop+0x4d: pushl $0
db{0}> continue
NMI ... going to debugger
Stopped in pid 0.1 (system) at netbsd:sipcom_stop+0x149: addl $0x10,%e
sp
db{0}> continue
sip0: detached
(On the AMD Elan SC520, the NMI occurs at the instruction after
the write or, if the write was posted, at either that instruction
or some later one.)
enabled:
- don't enable interrupts before calling uhci_run()
- check if the controller's interrupt is enabled in uhci_intr()
The issue is that uhci_run() may tsleep(), uhci_intr1() may be called before
uhci_run() is complete and disable it because it found it halted.
Now a uhci controller can be successfully exported to a NetBSD Xen domU :)
not 82550, so don't set FXPF_EXT_RFA (available only on 82550/82551)
for those variants.
Tested on Intel PRO/100 VM with 82562EM/EX PHY by ISHII Kimio
on current-users, and should also fix PR kern/40431.
because EXT_RFA for RX cksum is always available with IPCB for TX cksum
but i82558 and i82559 have only EXT_TXCB without IPCB.
Tested on the following fxp cards:
fxp0 at pci0 dev 14 function 0: Intel i82557 Ethernet, rev 2
fxp0 at pci0 dev 14 function 0: i82559 Ethernet, rev 8
fxp0 at pci0 dev 14 function 0: i82550 Ethernet, rev 12
when bus_dmamap_create fails, then the map pointer is undefined.
age_dma_free() assumes, the map pointer is NULL. Make the assumption
always true by setting the map pointer to NULL in bus_dmamap_create()
error path in order to prevent age_dma_free() to call bus_dmamap_destroy()
on an invalid map.
