Commit Graph

425 Commits

Author SHA1 Message Date
Xiang Xiao 53c6789bef mm: Add mm_uninitialize to release the resource
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-06-12 19:41:14 +09:00
Gustavo Henrique Nihei 183da24fc5 mm: Log name of the Heap to which the new region is being added
Signed-off-by: Gustavo Henrique Nihei <gustavo.nihei@espressif.com>
2022-06-08 01:10:10 +08:00
Abdelatif Guettouche 4896623be2 build: Fix dependencies of kernel targets.
Targets build during the kernel phase did not have their dependencies
specified and thus they were not rebuilt after their dependencies have
changed, for example by changing options in menuconfig.

Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com>
2022-05-24 10:45:39 +08:00
YAMAMOTO Takashi 5edf5e5585 mm: Add CONFIG_MM_PANIC_ON_FAILURE option
While allocation failure here is not necessarily abnormal enough to
warrant a panic in general, it can be useful in certain situations.
2022-05-21 14:28:41 +08:00
YAMAMOTO Takashi b05320cac2 mm: Move backtrace stuff into a separate option
The functionality has too much overhead for CONFIG_DEBUG_MM.
2022-05-21 14:28:41 +08:00
YAMAMOTO Takashi c546c0b647 mm: Do not abort on allocation failue with CONFIG_DEBUG_MM
When allocation failed, it isn't too uncommon for the caller
to fall back to other allocation method.
(eg. esp32 textheap code tries iram heap when an allocation from rtc heap
failed.)
DEBUGASSERT(false) is too much in that case.

This commit removes the DEBUGASSERT, and also makes the heap dump
a separate option.
2022-05-21 14:28:41 +08:00
Xiang Xiao 1fb8c13e5e Replace nxsem_timedwait_uninterruptible with nxsem_tickwait_uninterruptible
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-05-15 13:55:58 +03:00
Xiang Xiao 9072eecc30 sched/wqueue: Change the return type of work_notifier_teardown to void
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-05-14 00:35:29 +03:00
wangbowen6 d4978bfba4 mm_initialize: malloc() return aligend pointer.
malloc() should return aligned (with MM_MIN_CHUNK) pointer, but
pr #5906 destroy that, this pr find a better method to solve
these questions.

Signed-off-by: YAMAMOTO Takashi <yamamoto@midokura.com>
and
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
2022-04-11 12:46:17 +08:00
wangbowen6 1a4ccd2d70 Revert "mm_heap: heapsize align with MM_MIN_CHUNK."
This reverts commit 69e69740b5.
2022-04-11 12:46:17 +08:00
wangbowen6 69e69740b5 mm_heap: heapsize align with MM_MIN_CHUNK.
Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
2022-03-31 19:50:25 +08:00
wangbowen6 a0bde21f0f mm/mm_heap: place alloc-bit at the LSB of node->preceeding.
Heap always allocate the memory at least 2bytes alignment.

Signed-off-by: wangbowen6 <wangbowen6@xiaomi.com>
2022-03-31 14:43:56 +03:00
Jiuzhu Dong 2ffca6d16d mm/mm_heap: output mallinfo when malloc failed
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-03-27 13:21:48 +08:00
Jiuzhu Dong 1ab58aff55 mm/heap: add MM_BACKTRACE_DEFAULT to config backtrace record by default
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-03-27 13:21:48 +08:00
Jiuzhu Dong 709207b8d3 mm/memdump: dynamic turn on backtrace in heap when enable DEBUG_MM
default turn off.
turn on: echo on > /proc/memdump
turn off: echo off > proc/memdump

Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-03-27 13:21:48 +08:00
Petro Karashchenko 68902d8732 pid_t: unify usage of special task IDs
Signed-off-by: Petro Karashchenko <petro.karashchenko@gmail.com>
2022-03-22 21:22:32 +08:00
Xiang Xiao c2a1d0f5ae procfs: Remove mallinfo from struct procfs_meminfo_entry_s
let's call mm_mallinfo directly

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-03-17 13:59:03 -03:00
Xiang Xiao d9a442a859 mm/iob: Remove initialized static variable inside iob_initialize
since it's impossible to call iob_initialize twice

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-03-12 15:06:39 -03:00
Xiang Xiao cf2538c277 mm/shm: Initialize shm_info_s at the definition place
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-03-12 15:06:39 -03:00
Jiuzhu Dong 018d7d605c mm/mm_malloc: dump all allocated memory info when malloc failed
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-26 14:32:42 +08:00
Jiuzhu Dong 168074fe2e mm/memdump: fix print backtrace abnormal
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-26 14:32:42 +08:00
Jiuzhu Dong 7ae3c572dc procfs: add heap info for every task
cat /proc/2/heap
AllocSize:  512
AllocBlks:  10

Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-26 14:32:42 +08:00
Jiuzhu Dong 2dcc4a359d mm/dump: add pid and backtrace for every memory node
usage:
echo <pid/used/free> trace > /proc/memdump
echo used > /proc/memdump //output all used memory info with backtrace
echo free > /proc/memdump //output all free memory info
echo 22 > /proc/memdump //output used memory info for task pid is 22 with backtrace

Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-26 14:32:42 +08:00
Jiuzhu Dong c5ba9261bc procfs: add memdump interface to dump used/free memory info
usage:echo <used/free> > /proc/memdump
echo used > /proc/memdump // output all used memory node info(address + size)
echo free > /proc/memdump // output all free memory node info(address + size)

Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-26 14:32:42 +08:00
Jiuzhu Dong 39eaeefb78 mm/mm_heap: remove the unnecessary check for memory node size
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-26 14:32:42 +08:00
ligd 419bc2fd04 mm: handle take mm sem in IRQ
This is a fix of:
0169a51220
This is caused by wrong memory sem operation in IDLE.

Fix:
Obey the original design, don't check the IDLE in mm_takesemaphore()

Signed-off-by: ligd <liguiding1@xiaomi.com>
2022-02-21 23:41:06 +08:00
Xiang Xiao 1d1bdd85a3 Remove the double blank line from source files
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-02-20 20:10:14 +01:00
Oki Minabe 484eb1795b fix #if condition of g_mmheap for BUILD_KERNEL. 2022-02-19 19:33:15 +08:00
Oki Minabe 979124928c fix array index range checking of gat[] in gran_alloc(). 2022-02-18 11:44:32 +08:00
chao.an 0b18cfbd9a mm/iob: add support of alloc with timeout iob_timedalloc()
Signed-off-by: chao.an <anchao@xiaomi.com>
2022-02-17 11:55:29 -03:00
Masayuki Ishikawa 41d62033a0 mm: umm_heap: Fix umm_heap for BUILD_KERNEL & ADDRENV
Summary:
- I noticed that the user heap is corrupted
- This commit fixes this issue by reverting the change to
  the NuttX-9.0.0

Impact:
- None

Testing:
- sabre6-quad:netknsh (not merged yet)

Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Co-authored-by: Oki Minabe <minabe.oki@gmail.com>
2022-02-17 11:31:01 +01:00
ligd 0169a51220 mm: handle take mm sem in IRQ
Signed-off-by: ligd <liguiding1@xiaomi.com>
2022-02-16 09:53:42 -03:00
Jiuzhu Dong 12b256f860 mm/mm_mallinfo: using mm_foreach to do mm_mallinfo
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-13 03:20:31 +08:00
Jiuzhu Dong b1a51a5b30 mm/mm_checkcorruption: using mm_foreach to do mm_checkcorruption
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-13 03:20:31 +08:00
Jiuzhu Dong eba0323eae mm/mm_heap: add mm_foreach to run handler for every memory node
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2022-02-13 03:20:31 +08:00
ligd b316611ef0 Revert "sem: remove limitation of irq context when do sem_trywait"
This reverts commit 7c547b3ebd.
2022-01-29 00:53:47 +08:00
Xiang Xiao 77792a1598 sched: Define CONFIG_SMP_NCPUS to 1 in no SMP case
to simplify the SMP related code logic

Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2022-01-20 23:21:21 +08:00
Jiuzhu Dong 9fb130cdea mm/circbuf: add sanity check for resize
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
2021-12-17 23:36:07 -06:00
ligd 5b369c5cec libs/lbc: remove CHAR_BIT = 16 support
For CEVA platform CHAR_BIT is 16, and will do lots of extra work
when use IPC.
We will not support this platform anymore, so remove all the b2c operations.

Signed-off-by: ligd <liguiding1@xiaomi.com>
2021-12-13 11:08:28 -06:00
ligd 7c547b3ebd sem: remove limitation of irq context when do sem_trywait
Signed-off-by: ligd <liguiding1@xiaomi.com>
2021-11-17 02:20:08 -06:00
Chery Dan 8d799f2148 Fix IOB functions
The operations of struct iob_queue_s in qh_head & qh_tail are performed with interrupts disabled.

change iflags to flags

add header file ref
update for check
2021-11-11 08:47:03 -06:00
chenwei23 39cdd99d77 mm: Support the kernel address sanitizer
Signed-off-by: chenwei23 <chenwei23@xiaomi.com>
2021-11-02 13:32:47 -03:00
Xiang Xiao 5cb6b042aa mm/heap: Simplify the condition check in mm_realloc
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-11-02 13:32:47 -03:00
ligd 3615592e82 mm: fix realloc overwrite memcpy when use a new block
Signed-off-by: ligd <liguiding1@xiaomi.com>
2021-11-01 21:27:52 -05:00
Abdelatif Guettouche f86ceea099 mm/mm_heap/mm.h: Fix some typos.
Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com>
2021-08-03 10:11:26 -07:00
YAMAMOTO Takashi 5b1f2dff5f iob_tryalloc: Fix throttled allocation
If threads are blocking for throttled allocation,
g_throttle_sem.semcount doesn't represent the number of
usable IOBs well.

Note: For non-throttled allocations, the g_iob_committed mechanism
is a rescue. But there is no equivalent for throttled allocations.
2021-08-02 11:25:55 -07:00
YAMAMOTO Takashi e96a870485 iob_tryalloc: Remove a wrong assertion 2021-08-02 11:25:55 -07:00
Masanari Iida b64e9050c4 mm: Fix a typo in Kconfig
This patch fixes a spelling typo in mm/Kconfig.

Signed-off-by: Masanari Iida <standby24x7@gmail.com>
2021-07-30 01:31:01 -07:00
Xiang Xiao 007adc7736 Replace all __attribute__((section(x)) with locate_data(x)
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-07-29 21:55:21 -03:00
YAMAMOTO Takashi 7a580ae3bc iob_trimhead: fix an integer truncation
I guess this fixes https://github.com/apache/incubator-nuttx/issues/4181
2021-07-20 14:20:29 +08:00