Use macros to avoid duplication. Arguably, this is somewhat less
legible, but future additions would grow this part of the file to
completely unmanageable dimensions.
The EMU*_COMMON_TEXTS macros will save duplication in a future commit;
I pulled them ahead to reduce churn.
While rewriting the tables anyway, rearrange them such that each card's
strings and registers are adjacent.
Also, add some static asserts to verify that the array sizes match.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230516093612.3536508-3-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Many registers are meaningless for stereo slaves and the extra voices.
This patch cleans up these unnecessary register writes.
snd_emu10k1_playback_{trigger,stop}_voice() is not called for stereo
slaves any more.
snd_emu10k1_playback_prepare_voice() is renamed to
snd_emu10k1_playback_unmute_voice(), as this better reflects its
remaining function. It's not called for the extra voices any more.
Accordingly, snd_emu10k1_playback_mute_voice() is factored out from
snd_emu10k1_playback_stop_voice(), and is called selectively as well.
This doesn't add conditionals which would avoid initializing
sub-registers, as that wouldn't pull its weight.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230516093612.3536451-6-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
We now enable ints even before triggering, and disable them only after
stopping - otherwise there is a race condition we may plausibly run into
when we pause/resume near the end of the buffer.
Updating the epcm->running flag is moved the same way, as it affects the
*_pointer() functions, which are called by the interrupt handler.
Also, factor these out to own functions, for clarity.
For multi-channel, the extra voice is now triggered after all regular
voices - we wouldn't want to receive an int before all channels have
passed the period boundary.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230516093612.3536451-5-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
We initialize them at card init and don't touch them later, so there is
no need to reset them again at voice start.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230516093612.3536451-4-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
We (rightfully) don't enable the envelope engine for PCM voices, so any
related setup is entirely pointless - the EMU8K documentation makes that
very clear, and the fact that the various open drivers all use different
values to no observable detriment pretty much confirms it.
The remaining initializations are regrouped for clarity.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230516093612.3536451-3-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The mixer structures were filled in two places: on driver init, and when
the devices are opened. The latter made the former pointless, so we
remove the former. This implies that mixer dumps may now return all
zeroes, which is OK, as restoring them is meaningless as well.
Things were even weirder for the (generally unused) secondary sends:
Some of the initialization loops were forgotten when support for Audigy
was added, thus creating the technically illegal state of multiple sends
being routed to the same FX accumulator (though it apparently doesn't
matter when the amount is zero).
The global multi-channel init used some rather bizarre values for the
secondary sends, and the init on open actually forgot to re-initialize
them. We now use a not really more useful, but simpler formula.
The direct register init was also bogus. This doesn't really matter, as
the value is overwritten when a voice comes into use, but still.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230516093612.3536451-2-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
SNDRV_CTL_ELEM_TYPE_* are type of snd_ctl_elem_type_t, we
have to __force cast them to int when comparing them with int
to fix the following sparse warnings.
sound/core/control_compat.c:203:14: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
sound/core/control_compat.c:205:14: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
sound/core/control_compat.c:207:14: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
sound/core/control_compat.c:209:14: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
sound/core/control_compat.c:237:21: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
sound/core/control_compat.c:238:21: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
sound/core/control_compat.c:270:21: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
sound/core/control_compat.c:271:21: sparse: warning: restricted snd_ctl_elem_type_t degrades to integer
Signed-off-by: Min-Hua Chen <minhuadotchen@gmail.com>
Link: https://lore.kernel.org/r/20230516223806.185683-1-minhuadotchen@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The voice volume is a raw fractional multiplier that can't actually
represent 1.0. To still enable real pass-through, we now set the volume
to 0.5 (which results in no loss of precision, as the FX bus provides
fractional values) and scale up the samples in DSP code.
To maintain backwards compatibility with existing configuration files,
we rescale the values in the mixer controls. The range is extended
upwards from 0xffff to 0x1fffd, which actually introduces the
possibility of specifying an amplification.
There is still a minor incompatibility with user space, namely if
someone loaded custom DSP code. They'll just get half the volume, so
this doesn't seem like a big deal.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408834-8-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Fractional multiplication with the maximal value 2^31-1 causes some tiny
distortion. Instead, we want to multiply with the full 2^31. The catch
is of course that this cannot be represented in the DSP's signed 32 bit
registers.
One way to deal with this is to encode 1.0 as a negative number and
special-case it. As a matter of fact, the SbLive! code path already
contained such code, though the controls never actually exercised it.
A more efficient approach is to use negative values, which actually
extend to -2^31. Accordingly, for all the volume adjustments we now use
the MAC1 instruction which negates the X operand.
The range of the controls in highres mode is extended downwards, so -1
is the new zero/mute. At maximal excursion, real zero is not mute any
more, but I don't think anyone will notice this behavior change. ;-)
That also required making the min/max/values in the control structs
signed. This technically changes the user space interface, but it seems
implausible that someone would notice - the numbers were actually
treated as if they were signed anyway (and in the actual mixer iface
they _are_). And without this change, the min value didn't even make
sense in the first place (and no-one noticed, because it was always 0).
Tested-by: Jonathan Dowland <jon@dow.land>
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408834-7-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The microphone capture device is a feature of the AC97 codec, so its
availability should be coupled to the presence of that codec.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408834-6-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The E-MU cards don't try very hard to be Sound Blasters. All sound I/O
goes through the Hana FPGA, thus making the regular extin/out controls
useless. Still showing them just serves to clutter up the interface and
confuse the user.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408834-5-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
User space could pass arbitrary ranges, which were uncritically
accepted. This could lead to table lookups out of range.
I don't think that this is a security issue, as it only allowed someone
with CAP_SYS_ADMIN to crash the kernel, but still.
Setting an invalid translation mode will also be rejected now. That did
no harm, but it's still better to detect errors.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408834-4-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
- Pull ahead all fixed allocations, so we don't rely on the semi-
dynamic ones not crossing the arbitrarily determined limit
- Use an enum for the fixed allocations
- Stop arbitrarily wasting registers on unexplained "reservations"
- Don't reserve two regs for the master volume control - it's mono
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408834-2-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Rather than applying masks to the provided values, make assertions
about them being valid - otherwise we'd just try to paper over bugs.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408798-2-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The idea to encode the bitfield manipulation in the register address is
quite clever, but doing that by hand is ugly and error-prone. So derive
it automatically from the mask instead.
Macros cannot #define other macros, so we now declare enums instead.
This also adds macros for decoding the register definitions. These will
be used by later commits.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230514170323.3408798-1-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Instead of using lots of instructions to mix wet and dry signals,
simply skip over the whole code block if tone control is disabled.
This also allows us doing away with the "shadow" playback channels.
Tested-by: Jonathan Dowland <jon@dow.land>
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230510173917.3073107-7-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Evidently, the channel delay bug exists in all E-MU cards; it's in the
Hana FPGA program, and was never fixed.
Note that the implementation is somewhat lazy - to localize the code
paths, we actually waste a GPR and a DSP instruction by keeping two
delay registers for the same physical source.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230510173917.3073107-6-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Instead of spending lots of instructions on masking and transplanting
the sign bit, sidestep the issue by replacing the last bit shift with
a wrapping addition to self.
Solution stolen from kX-project, after I pondered other ideas first.
Also, the function really doesn't need to return a constant int value.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230510173917.3073107-5-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Presumably, JDC added the seemingly superfluous indirection over the
temporary register because without it he'd get only zero readings.
However, switching the X and Y operands (or using EMU32 as the A
operand in the temporary load) works just fine. Presumably a DSP bug?
The original code was also actually buggy, though: both channels used
the left volume control.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230510173917.3073107-4-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
It controls the whole surround set, so stereo can't work. As a
consequence, only the left channel was paid attention to.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230510173917.3073107-2-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Introduce SNDRV_PCM_INFO_PERFECT_DRAIN and SNDRV_PCM_HW_PARAMS_NO_DRAIN_SILENCE
flags to fully control the filling of the silence samples in the drain ioctl.
Actually, the configurable silencing is going to be implemented in the user
space [1], but drivers (hardware) may not require this operation. Those flags
do the bidirectional setup for this operation:
1) driver may notify the presence of the perfect drain
2) user space may not require the filling of the silence samples to inhibit clicks
If we decide to move this operation to the kernel space in future, the
SNDRV_PCM_INFO_PERFECT_DRAIN flag may handle this situation without
double "silence" processing (user + kernel space).
The ALSA API should be universal, so forcing the behaviour (modifying of
the ring buffer with any samples) for the drain operation is not ideal.
[1] https://lore.kernel.org/alsa-devel/20230502115010.986325-1-perex@perex.cz/
[ fixed a typo in comment by tiwai ]
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Link: https://lore.kernel.org/r/20230502115536.986900-1-perex@perex.cz
Signed-off-by: Takashi Iwai <tiwai@suse.de>
- Fix mixer source port names. These will require some users to
re-adjust their mixer settings, which seems acceptable:
- The S/PDIF port is on the main 1010 card, not the 0202 daughter card
- The 1616m CardBus card has all inputs on the dock, so there is
no point in specifying it
- Conversely, the 1010 card has "dispersed" inputs, so say where the
ADAT port is, consistently with the S/PDIF port
- The 1616m CardBus card is actually named E-MU 02 (due to the headphone
output jack it has)
- Fix capitalization of "E-MU"
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230428095941.1706335-1-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Since commit 5bbb1ab5bd ("control: use counting semaphore as write lock
for ELEM_WRITE operation"), mixer values have been fully read-write
locked. This means that it is now unnecessary to apply any additional
locks to values that are accessed solely by mixer callbacks. Values that
are read outside mixer callbacks still need write locking. There are no
cases of mixer values being written outside mixer callbacks, so no read
locks remain in mixer callbacks.
Note that the removed locks refer only to the emu data structure, not
the card's registers as the lock's name suggests.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230428095941.1706278-7-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Since commit 5bbb1ab5bd ("control: use counting semaphore as write lock
for ELEM_WRITE operation"), this has been locking the controls including
their values, not just the list of controls.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230428095941.1706278-6-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The function does read-modify-write cycles on the card's registers, and
doesn't access mutable members of the emu data structure.
I suppose this might have been a mixup due to the lock names being
logically swapped.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230428095941.1706278-4-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Contrary to its name, reg_lock locks the emu data structure, not the
registers. As the functions access only data which is set once at card
initialization, there is no point in locking it.
Actually locking the registers would be pointless as well, as
snd_emu10k1_intr_{en,dis}able() does its own locking, and TIMER is
accessed only in this one place.
Locking snd_emu10k1_timer_{start,stop}() against each other also
wouldn't buy us anything; the functions interleaving their I/O accesses
wouldn't introduce new problems.
Signed-off-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Link: https://lore.kernel.org/r/20230428095941.1706278-2-oswald.buddenhagen@gmx.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
We need to better polish building with BPF skels, so revert back to
making it an experimental feature that has to be explicitely enabled
using BUILD_BPF_SKEL=1.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCZFbCXwAKCRCyPKLppCJ+
J7cHAP97erKY4hBXArjpfzcvpFmboh/oqhbTLntyIpS6TEnOyQEAyervAPGIjQYC
DCo4foyXmOWn3dhNtK9M+YiRl3o2SgQ=
=7G78
-----END PGP SIGNATURE-----
Merge tag 'perf-tools-for-v6.4-3-2023-05-06' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux
Pull perf tool updates from Arnaldo Carvalho de Melo:
"Third version of perf tool updates, with the build problems with with
using a 'vmlinux.h' generated from the main build fixed, and the bpf
skeleton build disabled by default.
Build:
- Require libtraceevent to build, one can disable it using
NO_LIBTRACEEVENT=1.
It is required for tools like 'perf sched', 'perf kvm', 'perf
trace', etc.
libtraceevent is available in most distros so installing
'libtraceevent-devel' should be a one-time event to continue
building perf as usual.
Using NO_LIBTRACEEVENT=1 produces tooling that is functional and
sufficient for lots of users not interested in those libtraceevent
dependent features.
- Allow Python support in 'perf script' when libtraceevent isn't
linked, as not all features requires it, for instance Intel PT does
not use tracepoints.
- Error if the python interpreter needed for jevents to work isn't
available and NO_JEVENTS=1 isn't set, preventing a build without
support for JSON vendor events, which is a rare but possible
condition. The two check error messages:
$(error ERROR: No python interpreter needed for jevents generation. Install python or build with NO_JEVENTS=1.)
$(error ERROR: Python interpreter needed for jevents generation too old (older than 3.6). Install a newer python or build with NO_JEVENTS=1.)
- Make libbpf 1.0 the minimum required when building with out of
tree, distro provided libbpf.
- Use libsdtc++'s and LLVM's libcxx's __cxa_demangle, a portable C++
demangler, add 'perf test' entry for it.
- Make binutils libraries opt in, as distros disable building with it
due to licensing, they were used for C++ demangling, for instance.
- Switch libpfm4 to opt-out rather than opt-in, if libpfm-devel (or
equivalent) isn't installed, we'll just have a build warning:
Makefile.config:1144: libpfm4 not found, disables libpfm4 support. Please install libpfm4-dev
- Add a feature test for scandirat(), that is not implemented so far
in musl and uclibc, disabling features that need it, such as
scanning for tracepoints in /sys/kernel/tracing/events.
perf BPF filters:
- New feature where BPF can be used to filter samples, for instance:
$ sudo ./perf record -e cycles --filter 'period > 1000' true
$ sudo ./perf script
perf-exec 2273949 546850.708501: 5029 cycles: ffffffff826f9e25 finish_wait+0x5 ([kernel.kallsyms])
perf-exec 2273949 546850.708508: 32409 cycles: ffffffff826f9e25 finish_wait+0x5 ([kernel.kallsyms])
perf-exec 2273949 546850.708526: 143369 cycles: ffffffff82b4cdbf xas_start+0x5f ([kernel.kallsyms])
perf-exec 2273949 546850.708600: 372650 cycles: ffffffff8286b8f7 __pagevec_lru_add+0x117 ([kernel.kallsyms])
perf-exec 2273949 546850.708791: 482953 cycles: ffffffff829190de __mod_memcg_lruvec_state+0x4e ([kernel.kallsyms])
true 2273949 546850.709036: 501985 cycles: ffffffff828add7c tlb_gather_mmu+0x4c ([kernel.kallsyms])
true 2273949 546850.709292: 503065 cycles: 7f2446d97c03 _dl_map_object_deps+0x973 (/usr/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2)
- In addition to 'period' (PERF_SAMPLE_PERIOD), the other
PERF_SAMPLE_ can be used for filtering, and also some other sample
accessible values, from tools/perf/Documentation/perf-record.txt:
Essentially the BPF filter expression is:
<term> <operator> <value> (("," | "||") <term> <operator> <value>)*
The <term> can be one of:
ip, id, tid, pid, cpu, time, addr, period, txn, weight, phys_addr,
code_pgsz, data_pgsz, weight1, weight2, weight3, ins_lat, retire_lat,
p_stage_cyc, mem_op, mem_lvl, mem_snoop, mem_remote, mem_lock,
mem_dtlb, mem_blk, mem_hops
The <operator> can be one of:
==, !=, >, >=, <, <=, &
The <value> can be one of:
<number> (for any term)
na, load, store, pfetch, exec (for mem_op)
l1, l2, l3, l4, cxl, io, any_cache, lfb, ram, pmem (for mem_lvl)
na, none, hit, miss, hitm, fwd, peer (for mem_snoop)
remote (for mem_remote)
na, locked (for mem_locked)
na, l1_hit, l1_miss, l2_hit, l2_miss, any_hit, any_miss, walk, fault (for mem_dtlb)
na, by_data, by_addr (for mem_blk)
hops0, hops1, hops2, hops3 (for mem_hops)
perf lock contention:
- Show lock type with address.
- Track and show mmap_lock, siglock and per-cpu rq_lock with address.
This is done for mmap_lock by following the current->mm pointer:
$ sudo ./perf lock con -abl -- sleep 10
contended total wait max wait avg wait address symbol
...
16344 312.30 ms 2.22 ms 19.11 us ffff8cc702595640
17686 310.08 ms 1.49 ms 17.53 us ffff8cc7025952c0
3 84.14 ms 45.79 ms 28.05 ms ffff8cc78114c478 mmap_lock
3557 76.80 ms 68.75 us 21.59 us ffff8cc77ca3af58
1 68.27 ms 68.27 ms 68.27 ms ffff8cda745dfd70
9 54.53 ms 7.96 ms 6.06 ms ffff8cc7642a48b8 mmap_lock
14629 44.01 ms 60.00 us 3.01 us ffff8cc7625f9ca0
3481 42.63 ms 140.71 us 12.24 us ffffffff937906ac vmap_area_lock
16194 38.73 ms 42.15 us 2.39 us ffff8cd397cbc560
11 38.44 ms 10.39 ms 3.49 ms ffff8ccd6d12fbb8 mmap_lock
1 5.43 ms 5.43 ms 5.43 ms ffff8cd70018f0d8
1674 5.38 ms 422.93 us 3.21 us ffffffff92e06080 tasklist_lock
581 4.51 ms 130.68 us 7.75 us ffff8cc9b1259058
5 3.52 ms 1.27 ms 703.23 us ffff8cc754510070
112 3.47 ms 56.47 us 31.02 us ffff8ccee38b3120
381 3.31 ms 73.44 us 8.69 us ffffffff93790690 purge_vmap_area_lock
255 3.19 ms 36.35 us 12.49 us ffff8d053ce30c80
- Update default map size to 16384.
- Allocate single letter option -M for --map-nr-entries, as it is
proving being frequently used.
- Fix struct rq lock access for older kernels with BPF's CO-RE
(Compile once, run everywhere).
- Fix problems found with MSAn.
perf report/top:
- Add inline information when using --call-graph=fp or lbr, as was
already done to the --call-graph=dwarf callchain mode.
- Improve the 'srcfile' sort key performance by really using an
optimization introduced in 6.2 for the 'srcline' sort key that
avoids calling addr2line for comparision with each sample.
perf sched:
- Make 'perf sched latency/map/replay' to use "sched:sched_waking"
instead of "sched:sched_waking", consistent with 'perf record'
since d566a9c2d4 ("perf sched: Prefer sched_waking event when it
exists").
perf ftrace:
- Make system wide the default target for latency subcommand, run the
following command then generate some network traffic and press
control+C:
# perf ftrace latency -T __kfree_skb
^C
DURATION | COUNT | GRAPH |
0 - 1 us | 27 | ############# |
1 - 2 us | 22 | ########### |
2 - 4 us | 8 | #### |
4 - 8 us | 5 | ## |
8 - 16 us | 24 | ############ |
16 - 32 us | 2 | # |
32 - 64 us | 1 | |
64 - 128 us | 0 | |
128 - 256 us | 0 | |
256 - 512 us | 0 | |
512 - 1024 us | 0 | |
1 - 2 ms | 0 | |
2 - 4 ms | 0 | |
4 - 8 ms | 0 | |
8 - 16 ms | 0 | |
16 - 32 ms | 0 | |
32 - 64 ms | 0 | |
64 - 128 ms | 0 | |
128 - 256 ms | 0 | |
256 - 512 ms | 0 | |
512 - 1024 ms | 0 | |
1 - ... s | 0 | |
#
perf top:
- Add --branch-history (LBR: Last Branch Record) option, just like
already available for 'perf record'.
- Fix segfault in thread__comm_len() where thread->comm was being
used outside thread->comm_lock.
perf annotate:
- Allow configuring objdump and addr2line in ~/.perfconfig., so that
you can use alternative binaries, such as llvm's.
perf kvm:
- Add TUI mode for 'perf kvm stat report'.
Reference counting:
- Add reference count checking infrastructure to check for use after
free, done to the 'cpumap', 'namespaces', 'maps' and 'map' structs,
more to come.
To build with it use -DREFCNT_CHECKING=1 in the make command line
to build tools/perf. Documented at:
https://perf.wiki.kernel.org/index.php/Reference_Count_Checking
- The above caught, for instance, fix, present in this series:
- Fix maps use after put in 'perf test "Share thread maps"':
'maps' is copied from leader, but the leader is put on line 79
and then 'maps' is used to read the reference count below - so
a use after put, with the put of maps happening within
thread__put.
Fixed by reversing the order of puts so that the leader is put
last.
- Also several fixes were made to places where reference counts were
not being held.
- Make this one of the tests in 'make -C tools/perf build-test' to
regularly build test it and to make sure no direct access to the
reference counted structs are made, doing that via accessors to
check the validity of the struct pointer.
ARM64:
- Fix 'perf report' segfault when filtering coresight traces by
sparse lists of CPUs.
- Add support for 'simd' as a sort field for 'perf report', to show
ARM's NEON SIMD's predicate flags: "partial" and "empty".
arm64 vendor events:
- Add N1 metrics.
Intel vendor events:
- Add graniterapids, grandridge and sierraforrest events.
- Refresh events for: alderlake, aldernaken, broadwell, broadwellde,
broadwellx, cascadelakx, haswell, haswellx, icelake, icelakex,
jaketown, meteorlake, knightslanding, sandybridge, sapphirerapids,
silvermont, skylake, tigerlake and westmereep-dp
- Refresh metrics for alderlake-n, broadwell, broadwellde,
broadwellx, haswell, haswellx, icelakex, ivybridge, ivytown and
skylakex.
perf stat:
- Implement --topdown using JSON metrics.
- Add TopdownL1 JSON metric as a default if present, but disable it
for now for some Intel hybrid architectures, a series of patches
addressing this is being reviewed and will be submitted for v6.5.
- Use metrics for --smi-cost.
- Update topdown documentation.
Vendor events (JSON) infrastructure:
- Add support for computing and printing metric threshold values. For
instance, here is one found in thesapphirerapids json file:
{
"BriefDescription": "Percentage of cycles spent in System Management Interrupts.",
"MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)",
"MetricGroup": "smi",
"MetricName": "smi_cycles",
"MetricThreshold": "smi_cycles > 0.1",
"ScaleUnit": "100%"
},
- Test parsing metric thresholds with the fake PMU in 'perf test
pmu-events'.
- Support for printing metric thresholds in 'perf list'.
- Add --metric-no-threshold option to 'perf stat'.
- Add rand (reverse and) and has_pmem (optane memory) support to
metrics.
- Sort list of input files to avoid depending on the order from
readdir() helping in obtaining reproducible builds.
S/390:
- Add common metrics: - CPI (cycles per instruction), prbstate (ratio
of instructions executed in problem state compared to total number
of instructions), l1mp (Level one instruction and data cache misses
per 100 instructions).
- Add cache metrics for z13, z14, z15 and z16.
- Add metric for TLB and cache.
ARM:
- Add raw decoding for SPE (Statistical Profiling Extension) v1.3 MTE
(Memory Tagging Extension) and MOPS (Memory Operations) load/store.
Intel PT hardware tracing:
- Add event type names UINTR (User interrupt delivered) and UIRET
(Exiting from user interrupt routine), documented in table 32-50
"CFE Packet Type and Vector Fields Details" in the Intel Processor
Trace chapter of The Intel SDM Volume 3 version 078.
- Add support for new branch instructions ERETS and ERETU.
- Fix CYC timestamps after standalone CBR
ARM CoreSight hardware tracing:
- Allow user to override timestamp and contextid settings.
- Fix segfault in dso lookup.
- Fix timeless decode mode detection.
- Add separate decode paths for timeless and per-thread modes.
auxtrace:
- Fix address filter entire kernel size.
Miscellaneous:
- Fix use-after-free and unaligned bugs in the PLT handling routines.
- Use zfree() to reduce chances of use after free.
- Add missing 0x prefix for addresses printed in hexadecimal in 'perf
probe'.
- Suppress massive unsupported target platform errors in the unwind
code.
- Fix return incorrect build_id size in elf_read_build_id().
- Fix 'perf scripts intel-pt-events.py' IPC output for Python 2 .
- Add missing new parameter in kfree_skb tracepoint to the python
scripts using it.
- Add 'perf bench syscall fork' benchmark.
- Add support for printing PERF_MEM_LVLNUM_UNC (Uncached access) in
'perf mem'.
- Fix wrong size expectation for perf test 'Setup struct
perf_event_attr' caused by the patch adding
perf_event_attr::config3.
- Fix some spelling mistakes"
* tag 'perf-tools-for-v6.4-3-2023-05-06' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (365 commits)
Revert "perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL"
Revert "perf build: Warn for BPF skeletons if endian mismatches"
perf metrics: Fix SEGV with --for-each-cgroup
perf bpf skels: Stop using vmlinux.h generated from BTF, use subset of used structs + CO-RE
perf stat: Separate bperf from bpf_profiler
perf test record+probe_libc_inet_pton: Fix call chain match on x86_64
perf test record+probe_libc_inet_pton: Fix call chain match on s390
perf tracepoint: Fix memory leak in is_valid_tracepoint()
perf cs-etm: Add fix for coresight trace for any range of CPUs
perf build: Fix unescaped # in perf build-test
perf unwind: Suppress massive unsupported target platform errors
perf script: Add new parameter in kfree_skb tracepoint to the python scripts using it
perf script: Print raw ip instead of binary offset for callchain
perf symbols: Fix return incorrect build_id size in elf_read_build_id()
perf list: Modify the warning message about scandirat(3)
perf list: Fix memory leaks in print_tracepoint_events()
perf lock contention: Rework offset calculation with BPF CO-RE
perf lock contention: Fix struct rq lock access
perf stat: Disable TopdownL1 on hybrid
perf stat: Avoid SEGV on counter->name
...
The recent fix to ensure atomicity of lookup and allocation inadvertently
broke the pool refill mechanism, so that debugobject OOMs now in certain
situations. The reason is that the functions which got updated no longer
invoke debug_objecs_init(), which is now the only place to care about
refilling the tracking object pool.
Restore the original behaviour by adding explicit refill opportunities to
those places.
-----BEGIN PGP SIGNATURE-----
iQJHBAABCgAxFiEEQp8+kY+LLUocC4bMphj1TA10mKEFAmRWoFATHHRnbHhAbGlu
dXRyb25peC5kZQAKCRCmGPVMDXSYocNID/9e1fU2Nf32woHokzBGgARKb69Kl/hb
6yVdMpOnZtxmluheJLnqCWI4WbAB6NjulEMFv+KkwRZ+QndBKVEo8NMZ9RjbXDBb
HEehI6DvsqRDjaytOLEZj+/8afcZ7bUBKk7JuUK+y5B1gZViazfp1eF3hpiKsIV9
aowpH6c9lL/9sPgFe2qpp21MUmNTUQbHpz0vbYC0QjqSEU2zTlu8p//P6VLA3xpl
qoh8Gu5qo/L8lPspN2v8TRVXdiqH67J+KpbGO9IuUQWYPQqFdc6WchhHwomAk8nr
Nyn9Q1Lred96pTdW3B0Cumnxuf0VPt4X/uQxPSP0kCo/h0Q0Mh6fq59Z66H/Mhjk
TAvM52w3VzfTmQB6WgaCD1HyRRqIK5Nd+XqXnenCkHN4kjmGXNLg9MUGxua5CVgF
iQTSRYtN18rF9OevDOFGzsEig2RN1JFi9MnJg9Q/L8SoDUn5ZUfhPaSA/HcOBnSe
m+9aeRxlb0hAP7+upFKsJkDYzJTtbP6LSx6qqZMyQWqYdsUVHpdiPtJpXb7mLIqQ
wo83i/Ohq8+dF6ykd89ZcKJ8vLBrnE1rPFKKmvS5ov1eRt/hZbtR3tmMviCNna0M
2nrJE2fKClbs8Dmc6NNboJdz51ASgZEi32XmdFkATiuZqiD1id7ne0f85ju7DHD9
sOjfo4ZtIKD/Fw==
=/0Kc
-----END PGP SIGNATURE-----
Merge tag 'core-debugobjects-2023-05-06' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull debugobjects fix from Thomas Gleixner:
"A single fix for debugobjects:
The recent fix to ensure atomicity of lookup and allocation
inadvertently broke the pool refill mechanism, so that debugobject
OOMs now in certain situations. The reason is that the functions which
got updated no longer invoke debug_objecs_init(), which is now the
only place to care about refilling the tracking object pool.
Restore the original behaviour by adding explicit refill opportunities
to those places"
* tag 'core-debugobjects-2023-05-06' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
debugobject: Ensure pool refill (again)
- A long-standing bug in crypto_engine.
- A buggy but harmless check in the sun8i-ss driver.
- A regression in the CRYPTO_USER interface.
-----BEGIN PGP SIGNATURE-----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=9Us7
-----END PGP SIGNATURE-----
Merge tag 'v6.4-p2' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Pull crypto fixes from Herbert Xu:
- A long-standing bug in crypto_engine
- A buggy but harmless check in the sun8i-ss driver
- A regression in the CRYPTO_USER interface
* tag 'v6.4-p2' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: api - Fix CRYPTO_USER checks for report function
crypto: engine - fix crypto_queue backlog handling
crypto: sun8i-ss - Fix a test in sun8i_ss_setup_ivs()
- Revert an i.MX patch that's causing video failures because division
math goes sideways
- Fix a clang + W=1 build isue where FIELD_PREP() is taking a 32-bit
variable instead of the usual u64 type
- Fix a Kconfig bug in the StarFive JH7110 clk config that selects a
reset controller when it can't be selected
-----BEGIN PGP SIGNATURE-----
iQJFBAABCAAvFiEE9L57QeeUxqYDyoaDrQKIl8bklSUFAmRW7qARHHNib3lkQGtl
cm5lbC5vcmcACgkQrQKIl8bklSU8gA/8Dee5F/zZR+Y7TJGld2ZF5gTjo+5vWrPP
SQ4FTCSZwnslrYcESXL+tfKxGv4JaP+bbWfufbbBYAlapNQZchSnRurg0rvYCIc2
s9NeUbLvPhxJTaBJIOoUzOfzezvPpS9+eEnT6goyvrWrYBz+wolTrsOzI6t2ocyb
9KpBEiaqAUSOmno6mNmqNg48UvBytgtXBnlC3YsqLlKuBZcRBrK/nHO+dUIqb2kO
s0fbbnNI2Mal0e3Se25ehW803fl+OhnxshHO4EO+aRvCLhXINgyapFNR2vk1woN2
UM38va+BPyzaI9goXcxMN5JROxGZFTmHMYtrqgimvFwxVDnHdz2jdokaY0fLhAMV
mN5Y+J1XBdNui4Pz/iWzldkPWULtbLTCZZQFl1lbAdBcvPSQC+hIcQEFKxg0O4rk
d0+B3a2SPfK1X5Ebl+BtOogbTGr616zVHgiiHeU0vxxFbqXboGh+iKJJdvtHA4WF
t/y4gH3cx67NneUFMihNs+UqVSVnE6N1+EghTzAubnEIz3ONIQDyvWuTNagDZBWH
QRMJz51691OGuSAW6lw9Nnfe0qNFiSzQpJptBT1bJ2GvTzOUI8izNLZ0ah/effuR
avHhYGkf7DYggyldEf5eL/iVqUTJ1JYKpRdcuyfLAg4q0UPmRP43n4KdTtZ98wC3
UdusEJ8EqDc=
=IxvN
-----END PGP SIGNATURE-----
Merge tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fixes from Stephen Boyd:
"A couple more patches that would be good to get into -rc1:
- Revert an i.MX patch that's causing video failures because division
math goes sideways
- Fix a clang + W=1 build isue where FIELD_PREP() is taking a 32-bit
variable instead of the usual u64 type
- Fix a Kconfig bug in the StarFive JH7110 clk config that selects a
reset controller when it can't be selected"
* tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: starfive: Fix RESET_STARFIVE_JH7110 can't be selected in a specified case
clk: sp7021: Adjust width of _m in HWM_FIELD_PREP()
Revert "clk: imx: composite-8m: Add support to determine_rate"
Convert omap and pcc to use mbox_bind_client
- omap and hi6220 : use of_property_read_bool
- test: fix double-free and use spinlock header
- rockchip and bcm-pdc: drop of_match_ptr
- mpfs: change config symbol
- mediatek gce: support MT6795
- qcom apcs: consolidate of_device_id
support IPQ9574
-----BEGIN PGP SIGNATURE-----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=NHj4
-----END PGP SIGNATURE-----
Merge tag 'mailbox-v6.4' of git://git.linaro.org/landing-teams/working/fujitsu/integration
Pull mailbox updates from Jassi Brar:
- mailbox api: allow direct registration to a channel and convert omap
and pcc to use mbox_bind_client
- omap and hi6220 : use of_property_read_bool
- test: fix double-free and use spinlock header
- rockchip and bcm-pdc: drop of_match_ptr
- mpfs: change config symbol
- mediatek gce: support MT6795
- qcom apcs: consolidate of_device_id and support IPQ9574
* tag 'mailbox-v6.4' of git://git.linaro.org/landing-teams/working/fujitsu/integration:
dt-bindings: mailbox: qcom: add compatible for IPQ9574 SoC
mailbox: qcom-apcs-ipc: do not grow the of_device_id
dt-bindings: mailbox: qcom,apcs-kpss-global: use fallbacks for few variants
dt-bindings: mailbox: mediatek,gce-mailbox: Add support for MT6795
mailbox: mpfs: convert SOC_MICROCHIP_POLARFIRE to ARCH_MICROCHIP_POLARFIRE
mailbox: bcm-pdc: drop of_match_ptr for ID table
mailbox: rockchip: drop of_match_ptr for ID table
mailbox: mailbox-test: Fix potential double-free in mbox_test_message_write()
mailbox: mailbox-test: Explicitly include header for spinlock support
mailbox: Use of_property_read_bool() for boolean properties
mailbox: pcc: Use mbox_bind_client
mailbox: omap: Use mbox_bind_client
mailbox: Allow direct registration to a channel
-----BEGIN PGP SIGNATURE-----
iQJEBAABCAAuFiEEwPw5LcreJtl1+l5K99NY+ylx4KYFAmRXkp8QHGF4Ym9lQGtl
cm5lbC5kawAKCRD301j7KXHgpsH3EADZ4ex4vvc2/giYjRYTdyda00GIdYGiSZaZ
OBW59OBYJfDjsS99Z2U4638shyqM3rA6BwzvWD6hPUTShD7PUddIKDaanpCYBs+x
GhcDhzeJxARjsp/5qZPQ4r3Hdp+kWRnOFu1BSniCEdWt43J6AOm0l5duuNxl1db5
Jz+BkSgF1pICAVB5YfSPDShyFiDfc0qcv3rHZo/rVRjwmX3x4x+OVr97zbZInnH4
fm2vcv2e3Dc8uQEB0XFEyBWNeC8hdhTNe60t1y3MlILaPjHrQlzl5lkQFZ9dtLG6
/FDK+bD6Ex8t6SQnnQGrLa9q4vQYA6zlT5u/wdvzAkakmrSxt6phf23U1TbyZSQ8
clSg6MSuyJAW7YhiT3OqAsNMvBZ5g/nKHjLxfkAe86N9YIrMebZW9GDF+fWf0q0K
em7O4XPehjkCXKclCInMv2bkv+/pncrd84D7Jt3OmCPljBL6oXfWF3vJJQ6FaY+t
G3tNxl1hMHqwX8uaaSqyeuMYAEYizZa+ebMmHJwm/151d0EKylAOh3i6O1e4bpk3
Kn5irF6pfTV1g1TcmY3fOwNgMffcHc9Y4E8bXxMJAN9WiHDuopgn2C9+MrJ+KRrx
pLN7VD9UK/5yFzHSMBhyM8lm0/oTIhfTVgb+WyuHmJLcRhAnyNy9RrIBhSD0AoH+
UMyOexZR8w==
=l7Mo
-----END PGP SIGNATURE-----
Merge tag 'for-6.4/io_uring-2023-05-07' of git://git.kernel.dk/linux
Pull more io_uring updates from Jens Axboe:
"Nothing major in here, just two different parts:
- A small series from Breno that enables passing the full SQE down
for ->uring_cmd().
This is a prerequisite for enabling full network socket operations.
Queued up a bit late because of some stylistic concerns that got
resolved, would be nice to have this in 6.4-rc1 so the dependent
work will be easier to handle for 6.5.
- Fix for the huge page coalescing, which was a regression introduced
in the 6.3 kernel release (Tobias)"
* tag 'for-6.4/io_uring-2023-05-07' of git://git.kernel.dk/linux:
io_uring: Remove unnecessary BUILD_BUG_ON
io_uring: Pass whole sqe to commands
io_uring: Create a helper to return the SQE size
io_uring/rsrc: check for nonconsecutive pages
This reverts commit a980755beb.
We need to better polish building with BPF skels, so revert back to
making it an experimental feature that has to be explicitely enabled
using BUILD_BPF_SKEL=1.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This reverts commit 51924ae69e.
We need to better polish building with BPF skels, so revert back to
making it an experimental feature that has to be explicitely enabled
using BUILD_BPF_SKEL=1.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2d55c16c0c ("dmapool: create/destroy cleanup").
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQTTMBEPP41GrTpTJgfdBJ7gKXxAjgUCZFaTPAAKCRDdBJ7gKXxA
ji1zAP4s/mWPQOJkoTR8cn9wRdGUB4FKgDiJwwYeOdcHRuylvwEAlgOD8qlizXYu
z4tGx6J8p3X+IdEApqH6Up56wjW+Wgk=
=X19M
-----END PGP SIGNATURE-----
Merge tag 'mm-stable-2023-05-06-10-49' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull dmapool updates - again - from Andrew Morton:
"Reinstate the dmapool changes which were accidentally removed by a
mishap on the last commit in the previous attempt at the series"
Fixes: 2d55c16c0c ("dmapool: create/destroy cleanup").
[ The whole old series: def8574308ed..2d55c16c0c54 results in an empty
diff because that last commit ended up being just a revert of all that
came everything before it. - Linus ]
* tag 'mm-stable-2023-05-06-10-49' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
dmapool: link blocks across pages
dmapool: don't memset on free twice
dmapool: simplify freeing
dmapool: consolidate page initialization
dmapool: rearrange page alloc failure handling
dmapool: move debug code to own functions
dmapool: speedup DMAPOOL_DEBUG with init_on_alloc
dmapool: cleanup integer types
dmapool: use sysfs_emit() instead of scnprintf()
dmapool: remove checks for dev == NULL
changes.
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQTTMBEPP41GrTpTJgfdBJ7gKXxAjgUCZFaSnAAKCRDdBJ7gKXxA
jskTAPwIkL6xVMIEdtWfQNIy751o0onzfOQAIWgM3dSL83cMJQEA5OXxXt4LjrJw
4q0zf4GLYXjMjhnfYr8zXF39y05iyQA=
=nIhV
-----END PGP SIGNATURE-----
Merge tag 'mm-hotfixes-stable-2023-05-06-10-45' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull hotfixes from Andrew Morton:
"Five hotfixes.
Three are cc:stable, two pertain to merge window changes"
* tag 'mm-hotfixes-stable-2023-05-06-10-45' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
afs: fix the afs_dir_get_folio return value
nilfs2: do not write dirty data after degenerating to read-only
mm: do not reclaim private data from pinned page
nilfs2: fix infinite loop in nilfs_mdt_get_block()
mm/mmap/vma_merge: always check invariants
The allocated dmapool pages are never freed for the lifetime of the pool.
There is no need for the two level list+stack lookup for finding a free
block since nothing is ever removed from the list. Just use a simple
stack, reducing time complexity to constant.
The implementation inserts the stack linking elements and the dma handle
of the block within itself when freed. This means the smallest possible
dmapool block is increased to at most 16 bytes to accommodate these
fields, but there are no exisiting users requesting a dma pool smaller
than that anyway.
Removing the list has a significant change in performance. Using the
kernel's micro-benchmarking self test:
Before:
# modprobe dmapool_test
dmapool test: size:16 blocks:8192 time:57282
dmapool test: size:64 blocks:8192 time:172562
dmapool test: size:256 blocks:8192 time:789247
dmapool test: size:1024 blocks:2048 time:371823
dmapool test: size:4096 blocks:1024 time:362237
After:
# modprobe dmapool_test
dmapool test: size:16 blocks:8192 time:24997
dmapool test: size:64 blocks:8192 time:26584
dmapool test: size:256 blocks:8192 time:33542
dmapool test: size:1024 blocks:2048 time:9022
dmapool test: size:4096 blocks:1024 time:6045
The module test allocates quite a few blocks that may not accurately
represent how these pools are used in real life. For a more marco level
benchmark, running fio high-depth + high-batched on nvme, this patch shows
submission and completion latency reduced by ~100usec each, 1% IOPs
improvement, and perf record's time spent in dma_pool_alloc/free were
reduced by half.
[kbusch@kernel.org: push new blocks in ascending order]
Link: https://lkml.kernel.org/r/20230221165400.1595247-1-kbusch@meta.com
Link: https://lkml.kernel.org/r/20230126215125.4069751-12-kbusch@meta.com
Fixes: 2d55c16c0c ("dmapool: create/destroy cleanup")
Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Tony Battersby <tonyb@cybernetics.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
If debug is enabled, dmapool will poison the range, so no need to clear it
to 0 immediately before writing over it.
Link: https://lkml.kernel.org/r/20230126215125.4069751-11-kbusch@meta.com
Fixes: 2d55c16c0c ("dmapool: create/destroy cleanup")
Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Tony Battersby <tonyb@cybernetics.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
The actions for busy and not busy are mostly the same, so combine these
and remove the unnecessary function. Also, the pool is about to be freed
so there's no need to poison the page data since we only check for poison
on alloc, which can't be done on a freed pool.
Link: https://lkml.kernel.org/r/20230126215125.4069751-10-kbusch@meta.com
Fixes: 2d55c16c0c ("dmapool: create/destroy cleanup")
Signed-off-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Tony Battersby <tonyb@cybernetics.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Oswald Buddenhagen
Min-Hua Chen
Jaroslav Kysela
Linus Torvalds
Arnaldo Carvalho de Melo
Keith Busch