| created | 2021-07-01T19:15:01Z |
|---|---|
| begin | 2021-06-02T00:00:00Z |
| end | 2021-06-03T00:00:00Z |
| path | src/sys |
| commits | 23 |
| date | 2021-06-02T00:09:57Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/netinet/in.h | log | diff | annotate |
| src/sys/netinet/ip_input.c | log | diff | annotate | |
| message |
factor out the code that does basic sanity checks on ipv4 headers. this will allow these checks to be reused by bridge (where they're currently duplicated), veb, and tpmr. ok bluhm@ sashan@ |
|||
| date | 2021-06-02T00:20:50Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/netinet6/in6.h | log | diff | annotate |
| src/sys/netinet6/ip6_input.c | log | diff | annotate | |
| message |
factor out the code that does sanity checks on ipv6 headers and addresses. this will allow these checks to be reused for ip packet inspection in bridge, veb, and tpmr. ok bluhm@ sashan@ |
|||
| date | 2021-06-02T00:27:03Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/netinet6/ip6_input.c | log | diff | annotate |
| message |
don't init a pointer just to immediately set it again. this is in ip6_input_if just before ipv6_check returns the pointer we end up using. pointed out by bluhm@ |
|||
| date | 2021-06-02T00:39:25Z | |||
|---|---|---|---|---|
| author | cheloha | |||
| files | src/sys/ddb/db_command.c | log | diff | annotate |
| src/sys/kern/subr_prf.c | log | diff | annotate | |
| src/sys/sys/systm.h | log | diff | annotate | |
| message |
kernel: introduce per-CPU panic(9) message buffers Add a 512-byte buffer (ci_panicbuf) to each cpu_info struct on each platform for use by panic(9). The first panic on a given CPU writes its message to this buffer. Subsequent panics on a given CPU print the panic message to the console but do not modify the buffer. This aids debugging in two cases: - If 2+ CPUs panic simultaneously there is no risk of garbled messages in the panic buffer. - If a CPU panics and then the operator causes a second panic while using ddb(4), the operator can still recall the first failure on a particular CPU. Misc. changes to support this bigger change: - Set panicstr atomically to identify the first CPU to reach panic(). - Tweak db_show_panic_cmd() to print all panic messages across all CPUs. Prefix the first panic with an asterisk ('*'). - Prefer db_printf() to printf() during a panic if we have it. Apparently it disturbs less global state. - On amd64, tweak fault() to write the local panic buffer. This needs more work. Prompted by bluhm@ and deraadt@. Mostly written by deraadt@. Discussed with bluhm@, deraadt@ and kettenis@. Borne from a discussion on tech@ about making panic(9) more MP-safe: https://marc.info/?l=openbsd-tech&m=162086462316143&w=2 ok kettenis@, visa@, bluhm@, deraadt@ |
|||
| date | 2021-06-02T00:39:26Z | |||
|---|---|---|---|---|
| author | cheloha | |||
| files | src/sys/arch/alpha/include/cpu.h | log | diff | annotate |
| src/sys/arch/amd64/amd64/trap.c | log | diff | annotate | |
| src/sys/arch/amd64/include/cpu.h | log | diff | annotate | |
| src/sys/arch/arm/include/cpu.h | log | diff | annotate | |
| src/sys/arch/arm64/include/cpu.h | log | diff | annotate | |
| src/sys/arch/hppa/include/cpu.h | log | diff | annotate | |
| src/sys/arch/i386/include/cpu.h | log | diff | annotate | |
| src/sys/arch/m88k/include/cpu.h | log | diff | annotate | |
| src/sys/arch/mips64/include/cpu.h | log | diff | annotate | |
| src/sys/arch/powerpc/include/cpu.h | log | diff | annotate | |
| message |
kernel: introduce per-CPU panic(9) message buffers Add a 512-byte buffer (ci_panicbuf) to each cpu_info struct on each platform for use by panic(9). The first panic on a given CPU writes its message to this buffer. Subsequent panics on a given CPU print the panic message to the console but do not modify the buffer. This aids debugging in two cases: - If 2+ CPUs panic simultaneously there is no risk of garbled messages in the panic buffer. - If a CPU panics and then the operator causes a second panic while using ddb(4), the operator can still recall the first failure on a particular CPU. Misc. changes to support this bigger change: - Set panicstr atomically to identify the first CPU to reach panic(). - Tweak db_show_panic_cmd() to print all panic messages across all CPUs. Prefix the first panic with an asterisk ('*'). - Prefer db_printf() to printf() during a panic if we have it. Apparently it disturbs less global state. - On amd64, tweak fault() to write the local panic buffer. This needs more work. Prompted by bluhm@ and deraadt@. Mostly written by deraadt@. Discussed with bluhm@, deraadt@ and kettenis@. Borne from a discussion on tech@ about making panic(9) more MP-safe: https://marc.info/?l=openbsd-tech&m=162086462316143&w=2 ok kettenis@, visa@, bluhm@, deraadt@ |
|||
| date | 2021-06-02T00:39:27Z | |||
|---|---|---|---|---|
| author | cheloha | |||
| files | src/sys/arch/powerpc64/include/cpu.h | log | diff | annotate |
| src/sys/arch/riscv64/include/cpu.h | log | diff | annotate | |
| src/sys/arch/sh/include/cpu.h | log | diff | annotate | |
| src/sys/arch/sparc64/include/cpu.h | log | diff | annotate | |
| message |
kernel: introduce per-CPU panic(9) message buffers Add a 512-byte buffer (ci_panicbuf) to each cpu_info struct on each platform for use by panic(9). The first panic on a given CPU writes its message to this buffer. Subsequent panics on a given CPU print the panic message to the console but do not modify the buffer. This aids debugging in two cases: - If 2+ CPUs panic simultaneously there is no risk of garbled messages in the panic buffer. - If a CPU panics and then the operator causes a second panic while using ddb(4), the operator can still recall the first failure on a particular CPU. Misc. changes to support this bigger change: - Set panicstr atomically to identify the first CPU to reach panic(). - Tweak db_show_panic_cmd() to print all panic messages across all CPUs. Prefix the first panic with an asterisk ('*'). - Prefer db_printf() to printf() during a panic if we have it. Apparently it disturbs less global state. - On amd64, tweak fault() to write the local panic buffer. This needs more work. Prompted by bluhm@ and deraadt@. Mostly written by deraadt@. Discussed with bluhm@, deraadt@ and kettenis@. Borne from a discussion on tech@ about making panic(9) more MP-safe: https://marc.info/?l=openbsd-tech&m=162086462316143&w=2 ok kettenis@, visa@, bluhm@, deraadt@ |
|||
| date | 2021-06-02T00:40:51Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/if_bridge.c | log | diff | annotate |
| message |
use ipv4_check and ipv6_check provided by the network stacks. this removes the duplication of the check code, and lets the v6 code in particular pick up a lot more sanity checks around valid addresses on the wire. ok bluhm@ sashan@ |
|||
| date | 2021-06-02T00:44:18Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/if_tpmr.c | log | diff | annotate |
| src/sys/net/if_veb.c | log | diff | annotate | |
| message |
use ipv4_check and ipv6_check to well, check ip headers before running pf. unlike bridge(4), these checks are only run when the packet is entering the veb/tpmr topology. the assumption is that only valid IP packets end up inside the topology so we don't have to check them when they're leaving. ok bluhm@ sashan@ |
|||
| date | 2021-06-02T01:30:30Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/if_tpmr.c | log | diff | annotate |
| message |
read the tpmr if_flags once in tpmr_input so link flags apply consistently. this avoids IFF_LINK1 getting set by another cpu halfway through tpmr_input. if LINK1 is not set when a packet enters a tpmr pair it skips ip/pf checks, but if it is then set then only pf is run against it. this way you either get the ip checks and pf when the packet enters and leaves tpmr, or you dont get the ip and pf checks at all. |
|||
| date | 2021-06-02T01:36:10Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/if_tpmr.c | log | diff | annotate |
| message |
tpmr_input is called in an smr crit section, so it doesnt need its own. this simplifies the code a little bit. |
|||
| date | 2021-06-02T01:37:10Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/if_tpmr.c | log | diff | annotate |
| message | only read the if_bpf pointer once. | |||
| date | 2021-06-02T07:46:22Z | |||
|---|---|---|---|---|
| author | dlg | |||
| files | src/sys/net/pf_ioctl.c | log | diff | annotate |
| message | whitespace tweaks, no functional change. | |||
| date | 2021-06-02T11:30:23Z | |||
|---|---|---|---|---|
| author | mvs | |||
| files | src/sys/kern/syscalls.master | log | diff | annotate |
| src/sys/kern/uipc_syscalls.c | log | diff | annotate | |
| message |
Unlock setrtable(2). Local copy of `ps_rtableid' used to make checks consistent. ok mpi@ |
|||
| date | 2021-06-02T11:31:10Z | |||
|---|---|---|---|---|
| author | mvs | |||
| files | src/sys/kern/init_sysent.c | log | diff | annotate |
| src/sys/kern/syscalls.c | log | diff | annotate | |
| src/sys/sys/syscall.h | log | diff | annotate | |
| src/sys/sys/syscallargs.h | log | diff | annotate | |
| message | regen | |||
| date | 2021-06-02T13:56:28Z | |||
|---|---|---|---|---|
| author | visa | |||
| files | src/sys/kern/init_main.c | log | diff | annotate |
| src/sys/kern/kern_event.c | log | diff | annotate | |
| src/sys/sys/event.h | log | diff | annotate | |
| message |
Enable pool cache on knote pool Use the pool cache to reduce the overhead of memory management in function kqueue_register(). When EV_ADD is given, kqueue_register() pre-allocates a knote to avoid potential sleeping in the middle of the critical section that spans from knote lookup to insertion. However, the pre-allocation is useless if the lookup finds a matching knote. The cost of knote allocation will become significant with kqueue-based poll(2) and select(2) because the frequency of allocation will increase. Most of the cost appears to come from the locking inside the pool. The pool cache amortizes it by using CPU-local caches of free knotes as buffers. OK dlg@ mpi@ |
|||
| date | 2021-06-02T15:31:15Z | |||
|---|---|---|---|---|
| author | kettenis | |||
| files | src/sys/arch/arm64/stand/efiboot/conf.c | log | diff | annotate |
| src/sys/arch/arm64/stand/efiboot/efidev.c | log | diff | annotate | |
| message |
Add support for booting from disks with 4k sectors. Inspired by the code we already have for amd64, but changed to use EFI memory allocation interfaces as some implementations seem to insist on page-aligned memory. ok krw@ |
|||
| date | 2021-06-02T19:11:02Z | |||
|---|---|---|---|---|
| author | patrick | |||
| files | src/sys/dev/ic/nvme.c | log | diff | annotate |
| message |
Correctly calculate number of PRPL entries we have to sync by adding brackets to manage operator precedence. Otherwise we'd attempt to sync more than needed, which doesn't cause issues, but it's still wrong. ok dlg@ jmatthew@ |
|||
| date | 2021-06-02T19:16:11Z | |||
|---|---|---|---|---|
| author | patrick | |||
| files | src/sys/dev/pci/if_mcx.c | log | diff | annotate |
| message |
When processing a received packet, only sync the amount of bytes mcx(4) told us has arrived. The DMA map's mapsize on RX packets is the length of the allocated buffer. For mcx(4), this can be more than around 9000 bytes, as each buffer will be at least as big as the maximum supported MTU. There's no need to sync the whole buffer, if it's only a small packet. ok dlg@ jmatthew@ |
|||
| date | 2021-06-02T19:38:14Z | |||
|---|---|---|---|---|
| author | kettenis | |||
| files | src/sys/arch/powerpc64/dev/xive.c | log | diff | annotate |
| message |
Avoid spinning on the kernel lock with interrupts disabled. ok gkoehler@ |
|||
| date | 2021-06-02T21:41:38Z | |||
|---|---|---|---|---|
| author | kettenis | |||
| files | src/sys/dev/fdt/if_mvpp.c | log | diff | annotate |
| message |
Linux folks introduced "10gbase-r" since it is "more correct" and we're starting to see firmware that uses this instead of the "10gbase-kr" that was used before to describe the PHYs. So recognize both and treat that the same. ok deraadt@, patrick@ |
|||
| date | 2021-06-02T21:49:31Z | |||
|---|---|---|---|---|
| author | sashan | |||
| files | src/sys/net/if_pfsync.c | log | diff | annotate |
| message |
With parallel execution of pf_test() two packets may try to update the same state in pfsync(4) queue. pfsync_q_ins() takes that race into account with one exception: the KASSERT() at line 2352. That KASSERT() needs to be removed. 2346 void 2347 pfsync_q_ins(struct pf_state *st, int q) 2348 { 2349 struct pfsync_softc *sc = pfsyncif; 2350 size_t nlen, sc_len; 2351 2352 KASSERT(st->sync_state == PFSYNC_S_NONE); 2353 2354 #if defined(PFSYNC_DEBUG) 2355 if (sc->sc_len < PFSYNC_MINPKT) 2356 panic("pfsync pkt len is too low %zd", sc->sc_len); 2357 #endif 2358 do { 2359 mtx_enter(&sc->sc_mtx[q]); 2360 2361 /* 2362 * If two threads are competing to insert the same state, then 2363 * there must be just single winner. 2364 */ 2365 if (st->sync_state != PFSYNC_S_NONE) { 2366 mtx_leave(&sc->sc_mtx[q]); 2367 break; 2368 } OK bluhm@ |
|||
| date | 2021-06-02T22:44:26Z | |||
|---|---|---|---|---|
| author | krw | |||
| files | src/sys/arch/amd64/stand/efiboot/efidev.c | log | diff | annotate |
| src/sys/arch/amd64/stand/libsa/softraid_amd64.c | log | diff | annotate | |
| message |
Use the same logic in all copies of gpt_chk_mbr(), relaxing the media length check to allow EFI GPT partitions to be smaller that the entire disk. Consistently use GPTSECTOR instead of randomly tossing in some literal '1's. ok kettenis@ |
|||
| date | 2021-06-02T22:44:27Z | |||
|---|---|---|---|---|
| author | krw | |||
| files | src/sys/arch/arm64/stand/efiboot/efidev.c | log | diff | annotate |
| src/sys/arch/arm64/stand/efiboot/softraid_arm64.c | log | diff | annotate | |
| src/sys/arch/armv7/stand/efiboot/efidev.c | log | diff | annotate | |
| src/sys/arch/riscv64/stand/efiboot/efidev.c | log | diff | annotate | |
| src/sys/arch/riscv64/stand/efiboot/softraid_riscv64.c | log | diff | annotate | |
| src/sys/kern/subr_disk.c | log | diff | annotate | |
| message |
Use the same logic in all copies of gpt_chk_mbr(), relaxing the media length check to allow EFI GPT partitions to be smaller that the entire disk. Consistently use GPTSECTOR instead of randomly tossing in some literal '1's. ok kettenis@ |
|||