created | 2024-01-07T16:13:53Z |
---|---|
begin | 2024-01-04T00:00:00Z |
end | 2024-01-05T00:00:00Z |
path | src/sys |
commits | 8 |
date | 2024-01-04T01:32:06Z | |||
---|---|---|---|---|
author | mvs | |||
files | src/sys/dev/pv/hypervic.c | log | diff | annotate |
message |
Revert previous. splx(9) can call kvp_get_ip_info() from any place with netlock held and cause recursive lock acquisition issue. |
date | 2024-01-04T07:08:47Z | |||
---|---|---|---|---|
author | jmatthew | |||
files | src/sys/dev/pci/if_bnxt.c | log | diff | annotate |
message |
fix up barriers in bnxt_down() - use barriers for all interrupts and for the rx refill timeouts. tested by hrvoje ok bluhm@ |
date | 2024-01-04T08:41:59Z | |||
---|---|---|---|---|
author | kevlo | |||
files | src/sys/dev/usb/if_axen.c | log | diff | annotate |
src/sys/dev/usb/if_axenreg.h | log | diff | annotate | |
src/sys/dev/usb/if_cdce.c | log | diff | annotate | |
src/sys/dev/usb/if_cdcereg.h | log | diff | annotate | |
message |
Add support for AX88179A. AX88179A interweave dummies alongside valid packet headers in axen_rxeof(). However current driver records these dummy headers as dropped frames, leading to stats misreporting one Ifail per Ipkt. This skips those dummy headers silently, thereby not generating Ifail for them. From FreeBSD commit 70fbcd451b68b7f6038d8a602cd8d5e1bb890f1d Tested by landry@ and myself. ok claudio@, landry@ |
date | 2024-01-04T12:22:35Z | |||
---|---|---|---|---|
author | mglocker | |||
files | src/sys/dev/ic/ufshci.c | log | diff | annotate |
message |
Pass SCSI command directly to the UFS command descriptor instead of decoding/encoding it. Suggested and OK kettenis@ |
date | 2024-01-04T13:30:20Z | |||
---|---|---|---|---|
author | mglocker | |||
files | src/sys/dev/ic/ufshci.c | log | diff | annotate |
message | Fix timeout value for write command (typo). |
date | 2024-01-04T20:50:43Z | |||
---|---|---|---|---|
author | kettenis | |||
files | src/sys/arch/amd64/include/intrdefs.h | log | diff | annotate |
message |
Adjust IPLs to give us more interrupt vectors for IPL_NET. That is where we need them most since multi-queue NICs seem to be a thing now. ok patrick@, mlarkin@, jan@ |
date | 2024-01-04T21:02:30Z | |||
---|---|---|---|---|
author | mglocker | |||
files | src/sys/dev/ic/ufshcireg.h | log | diff | annotate |
message |
Set the interrupt aggregation counter down to 1, which fixes the read performance from ~20MB/s to ~220MB/s. |
date | 2024-01-04T21:35:56Z | |||
---|---|---|---|---|
author | mglocker | |||
files | src/sys/dev/ic/ufshci.c | log | diff | annotate |
message | Zap some '#if 0' code which was initially required for debugging. |