Measured on firmware 13.60: the first poll or wait that reports a
finished request returns its plain state and later ones add bit
0x10000, while cancel returns the plain state and does not count as a
read; a request that fails completes with state 3 and the error in its
result, and the other requests of its batch still run.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Measured on firmware 13.60: an id that is not a live request, including
one already deleted, makes poll, wait, cancel and delete return 0 and
write 0x80020003 for it; delete makes an id unknown, and cancel leaves a
finished request's state unchanged.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Measured on firmware 13.60: submits take 1 to 128 requests at priority
1 to 3, the id lists of poll, wait, cancel and delete take 1 to 128 ids,
wait takes mode 1 or 2, and sceKernelAioSetParam takes a window and a
delayed count of 0 to 128, a split flag of 0 or 1 and, with split on,
split and chunk sizes of 1 to 0x1000000; everything else returns EINVAL
without writing. sceKernelAioInitializeParam writes the console's three
default scheduling structs.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Denormal mode 2 flushes only the sources and mode 1 only the result, as
measured on RDNA2 (gfx1030), also with an inline or literal constant
factor. Both need the device to keep the f32 denormals the shader keeps,
so they translate only when the module declares DenormPreserve 32 and
throw otherwise; mode 1 with another f32 rounding throws as before.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
v_min_f32, v_max_f32, v_min3_f32, v_max3_f32 and v_med3_f32 follow only
bit 0 of the FLOAT_MODE f32 denormal field, as measured on RDNA2
(gfx1030): mode 2 flushes denormal inputs as mode 0 does, and mode 1
keeps them and returns a selected denormal unflushed as mode 3 does.
Mode 1 needs the device to keep those denormals, so it translates only
when the module declares DenormPreserve 32 and throws otherwise. Both
modes threw before.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
v_add_f32, v_sub_f32, v_subrev_f32, v_mul_f32 and v_mul_legacy_f32
follow the FLOAT_MODE f32 denormal field with f32 round to nearest
even, as measured on RDNA2 (gfx1030): mode 1 keeps denormal inputs and
flushes a denormal result, and mode 2 flushes denormal inputs and keeps
a denormal result. Both need the device to keep the f32 denormals the
shader keeps, so they translate only when the module declares
DenormPreserve 32 and throw otherwise. The result flush now applies
when only the result flush bit is set; it was skipped unless the input
flush bit was set too. A constant factor no longer skips the tiny
product check while denormal inputs are kept, as a kept denormal times
such a constant can still be tiny. Directed rounding in these modes
still throws.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
sceNpUnregisterStateCallbackA returns 0x80550015 for a callback id
outside 1..8, compared unsigned so 0, negative ids and INT_MIN are
included, as libSceNpManager.sprx does at RVA 0x15b10 in prosper's
reading. An empty slot still returns 0x80550009.
FreeBSD 11.4's sys_fchdir resolves the descriptor with getvnode, which fails with EINVAL for a descriptor that has no vnode; only a vnode that is not a directory gives ENOTDIR from change_dir. fchdir returned ENOTDIR for a pipe or socket. A descriptor that is neither a vnode nor a random device now gets EINVAL; the debt entry records the order of the three errors.
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
A kqueue descriptor is ready for POLLIN and POLLRDNORM while one of its
events is triggered and for no other event, as FreeBSD's kqueue_poll
answers; poll threw for it on both hosts. A kqueue cannot be waited on,
so while one is polled poll checks it every millisecond and waits on the
other descriptors in slices of 1 ms. This is the loop Windows already
used for pipes, which Linux now shares.
A kqueue closed during the wait reports POLLNVAL, and one whose number
was reused is polled again with the rest of the timeout, as FreeBSD's
pollrescan looks the descriptor up again.
Depends on #7248.
Co-Authored-By: Claude Opus 5.5 <[email protected]>
PPSA12361 binds color format 7 (COLOR_11_11_10, float) with COMP_SWAP 0
and stops on "unsupported color format 7 number type 7 component swap 0".
Only the reversed order (COMP_SWAP 2) was accepted.
The format keeps its 10-bit component in the low bits. The standard
swap puts red in the first slot, as Mesa pairs SWAP_STD with
COLOR_10_11_11 for R11G11B10_FLOAT, so COMP_SWAP 0 puts red there and
COMP_SWAP 2 blue. Swap 0 now decodes with the identity export mapping
0xe4 and the same packing; the other swaps still throw.
The decode test accepts swap 0 and keeps refusing swaps 1 and 3. The
execution test renders both orders into tiled and linear targets and
compares each stored word with the packing of the order. Its program
pool grows to fit the second pass. The debt entry names the accepted
orders and says the swap 0 layout is derived, not measured.
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
sosend_generic fails a send on a socket that can send no more with EPIPE
before it looks at so_error. Winsock reports the same send as WSAECONNABORTED
or WSAECONNRESET, which reached the guest as ECONNABORTED or ECONNRESET.
A stream send now maps both to EPIPE. The send-after-close checks of
guest_tcp, guest_socket_read_write and guest_socket_pair run on Windows too.
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
FreeBSD 11.4 serves kern.boottime ({CTL_KERN, KERN_BOOTTIME}) from
sysctl_kern_boottime in sys/kern/kern_tc.c as a read-only 16-byte struct
timeval, so a size query reports 16, a short buffer gets a partial copy
and ENOMEM, and a write gets EPERM from sysctl_root. sysctl, sysctlbyname
and sysctlnametomib now answer it with the guest CLOCK_REALTIME minus
CLOCK_UPTIME, taken at the first read and kept. HotSpot reads it while
writing its error report, which aborted with "unsupported MIB 1.21".
FreeBSD 11.4's soconnectat() (sys/kern/uipc_socket.c) disconnects a
connected socket of a protocol without PR_CONNREQUIRED before it calls
pru_connect. udp_disconnect() and udp6_disconnect() drop the peer and
reset the local address, then in_pcbconnect_setup() and in6_pcbladdr()
fail port 0 with EADDRNOTAVAIL, so the socket is left unconnected.
connect to port 0 on a connected datagram socket now dissolves the
association through the AF_UNSPEC disconnect path before failing with 49,
where it kept the old peer. Stream sockets keep EISCONN.
Braxton Atkinson