| created | 2022-12-25T09:08:33Z |
|---|---|
| begin | 2022-12-22T00:00:00Z |
| end | 2022-12-23T00:00:00Z |
| path | src/sys |
| commits | 5 |
| date | 2022-12-22T04:43:52Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/toeplitz.h | log | diff | annotate |
| message | provide stoeplitz_n32() | |||
| date | 2022-12-22T05:59:26Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/conf/files | log | diff | annotate |
| message |
use stoeplitz to generate a hash/flowid for state keys. the hash will be used to partition work in pf and pfsync in the future, and right now it is used as the first comparison in the rb tree state lookup. using stoeplitz means that pf will hash traffic the same way that hardware using a stoeplitz key will hash incoming traffic on rings. stoeplitz is also used by the tcp stack to generate a flow id, which is used to pick which transmit ring is used on nics with multiple queues too. using the same algorithm throughout the stack encourages affinity of packets to rings and softnet threads the whole way through. using the hash as the first comparison in the state rb tree comparison should encourage faster traversal of the state tree by having all the address/port bits summarised into the single hash value. however, tests by hrvoje popovski don't show performance changing. on the plus side, if this change is free from a performance point of view then it makes the future steps more straightforward. discussed at length at h2k22 tested by sashan@ and hrvoje popovski ok tb@ sashan@ claudio@ jmatthew@ |
|||
| date | 2022-12-22T05:59:27Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/pf.c | log | diff | annotate |
| src/sys/net/pfvar.h | log | diff | annotate | |
| src/sys/net/pfvar_priv.h | log | diff | annotate | |
| message |
use stoeplitz to generate a hash/flowid for state keys. the hash will be used to partition work in pf and pfsync in the future, and right now it is used as the first comparison in the rb tree state lookup. using stoeplitz means that pf will hash traffic the same way that hardware using a stoeplitz key will hash incoming traffic on rings. stoeplitz is also used by the tcp stack to generate a flow id, which is used to pick which transmit ring is used on nics with multiple queues too. using the same algorithm throughout the stack encourages affinity of packets to rings and softnet threads the whole way through. using the hash as the first comparison in the state rb tree comparison should encourage faster traversal of the state tree by having all the address/port bits summarised into the single hash value. however, tests by hrvoje popovski don't show performance changing. on the plus side, if this change is free from a performance point of view then it makes the future steps more straightforward. discussed at length at h2k22 tested by sashan@ and hrvoje popovski ok tb@ sashan@ claudio@ jmatthew@ |
|||
| date | 2022-12-22T15:44:02Z | |||
|---|---|---|---|---|
| author | kettenis | |||
| files | src/sys/arch/armv7/stand/efiboot/conf.c | log | diff | annotate |
| src/sys/arch/armv7/stand/efiboot/efiboot.c | log | diff | annotate | |
| src/sys/arch/armv7/stand/efiboot/efiboot.h | log | diff | annotate | |
| message |
Bring over various changes from the arm64 version of this code. ok patrick@ |
|||
| date | 2022-12-22T19:51:11Z | |||
|---|---|---|---|---|
| author | cheloha | |||
| files | src/sys/arch/sparc64/sparc64/clock.c | log | diff | annotate |
| src/sys/arch/sparc64/sparc64/locore.s | log | diff | annotate | |
| message |
sparc64: move retry logic out of stickcmpr_set() In some cases, stickcmpr_set() fails to ensure that %STICK_CMPR leads %STICK before returning to the caller. Rewriting the retry logic in C trivially fixes the issue. So move the retry logic out into a new function, stick_rearm(). Issue discovered by miod@, fix tested by miod@. Link: https://marc.info/?l=openbsd-tech&m=167122933414238&w=2 ok kettenis@ |
|||