Commit Graph

1287 Commits

Author SHA1 Message Date
cae6b9f154 move McKernel out of Linux kernel virtual 2018-09-04 19:51:09 +09:00
5fcbfa2eb5 page_fault_process_memory_range: Remove ihk_mc_map_virtual for CoW of device map
Device map with MAP_PRIVATE is copied when forking using copy_user_pte.
So the map isn't copied by those statements.

Futjitsu: POSTK_TEMP_FIX_14
Refs: #1039
Change-Id: I1a697ed2e003055d66a8eebd3e8d5e9e49d094ad
2018-08-30 02:21:42 +00:00
9a20cfaefb mem: Check if phys-mem is within the range of McKernel memory
Fujitsu: POSTK_DEBUG_TEMP_FIX_52
Refs: #1164
Change-Id: Idb9a6eac1d2e1df4c663c3171925c774421177fd
2018-08-30 02:18:37 +00:00
f57b0c5d4f wait: Delay wake-up parent within switch context
Fujitsu: POSTK_DEBUG_TEMP_FIX_41
Refs: #1006
Change-Id: Ia98e896505ad0f6549766604ade84550eee8bd2d
2018-08-30 02:13:51 +00:00
0fdeb254b3 switch context: Move to arch-dependent (arch_switch_context())
Fujitsu: POSTK_DEBUG_ARCH_DEP_22
Change-Id: I6faf8d9daa1e639350c2cd83db9bb27b9d37ba01
2018-08-30 02:13:34 +00:00
895a8c4099 procfs: Support multiple reads of e.g. /proc/*/maps
Refs: #1021
Change-Id: If36e1a0f3f41f0215868daf578e96775d96a59a3
2018-08-30 01:48:06 +00:00
e531ee626e mcctrl pager: handle pagers more properly
the pagers are all destroyed when linux thinks there is no process left,
but there is no synchronisation with mcexec on that and some new process
might have spawned and started using these pagers in the meantime,
leading to weird crashes because an invalid pager was used.

The reason we're cleaning up pagers when no process is left is that
mcctrl does not handle pager_req_release is the linux-side process got
killed or died before the mckernel one for some reason, so:
 - move pager_req_release to a new __do_in_kernel_irq_syscall() helper
 - have free_all_process_memory_range not set MF_HOST_RELEASED on the
memobj
 - just in case, clean up everything like before on mcctrl shutdown
instead of when no process is left.

Change-Id: I53b8b9b81b1e5b807593850af17b5ea5e8471174
Refs: #1154
2018-08-24 09:18:20 +09:00
94d093f058 fileobj_create: Suppress message on getting -ESRCH
-ESRCH from mcctrl doesn't mean an error but the file is not a regular
file and mcctrl wants McKernel to treat it as a device file.

Change-Id: Ie121f0e6a8b1f0a29c2f2cf193a51f4f52337809
2018-08-23 04:01:20 +00:00
ebc702624b devobj: fix object size (POSTK_DEBUG_TEMP_FIX_36)
Fujitsu: POSTK_DEBUG_TEMP_FIX_36
Change-Id: I5f020708f97b7468f19496b44c98e164d856598d
2018-08-22 07:26:50 +00:00
affe3e9010 do_fork: Increase tid table size when allowing oversubscription
The size of tid table needs to be more than #CPUs when CPU oversubscription
is needed.

Note that the max number of simultaneous threads are the min of the
following two:
(1) Number of mcexec worker threads
(2) NR_TID defined in kernel/syscall.c

Change-Id: I425189da415e1d3a763ad62567950d001850cf0d
2018-08-22 06:42:13 +00:00
0b2169964a futex_wait_queue_me: Spin-sleep when timeout and idle_halt is specified
schedule_timeout() with idle_halt should use spin sleep because sleep
with timeout is not implemented.

Change-Id: Ia0bebcc10ddfb872bffeece7f13fb35a4791db18
2018-08-22 06:36:43 +00:00
f18d1f5383 __sched_wakeup_thread: Notify interrupt_exit() of re-schedule
Change-Id: I438eb168f818eb5649857e22bdc7e68a145872f7
2018-08-22 06:33:23 +00:00
a6a9bac5b7 Protect more code by #ifdef PERF_ENABLE
Change-Id: I20a67c56c4d7817fdb87cc6a2aa47d68fe3eae8d
2018-08-22 06:03:12 +00:00
20368dd317 syscall: move sync_child_event up a bit
The function was between two perf functions when perf functions don't
use it...
It seemed simpler to move the function than to add an extra ifdef

Use that occasion to fix style warnings, no actual code changes were
made.

Change-Id: Ie8b5fa7968a3d5e54a690d079874db54f5e6c8c9
2018-08-22 05:55:26 +00:00
a7c9988aeb schedule: Don't reschedule immediately when wake up on migrate
Refs: #1027
Change-Id: Ibe563c45c42611170273f1e437566c20fbef68d3
2018-08-02 02:28:25 +00:00
786649d2a3 perf_event: Move changing monitoring-status into perf_stop
Change-Id: I84a13c2a825de24bfdada533c7049e8770a07061
2018-08-02 02:23:38 +00:00
d7416c6f79 perf_event: Specify counter by bit_mask on start/stop
Fujitsu: POSTK_DEBUG_TEMP_FIX_30
Refs: #1002
Change-Id: Iea51e9aef78927a5033e3a226d5efc6298da056a
2018-08-02 11:22:28 +09:00
cb1522ca92 perf_event: Handle fixed-pmc in arch-dep part
Fujitsu: POSTK_DEBUG_TEMP_FIX_31
Refs: #1003
Change-Id: I66c7d18b9137894cf5764464482e2ebd5ecb9d52
2018-08-02 02:14:04 +00:00
14660a10c3 Fix to procfs read returns EIO
Refs: #1152
Change-Id: I48b330953fd7674ba1a3ac35744f9f50a5712730
2018-08-02 01:48:51 +00:00
c716e87c53 execve: Clear sigaltstack and fp_regs
Fujitsu: POSTK_DEBUG_TEMP_FIX_19
Refs: #976
Change-Id: I16895eab13eecbb47b7e6da961fae82ee5e570ee
2018-08-01 15:11:05 +09:00
f01a883971 devobj: fix out of bounds shift
Similarily, pgoff << PAGE_SHIFT would need pgoff to be unsigned to fit,
but off_t is signed.
The reason for this shift was to truncate the offset argument to be
aligned to page boundaries, do that instead

Change-Id: I36c3de34b1834fdb0503942a6f3212e94986effd
2018-07-26 05:20:19 +00:00
3185334c1c debug messages: implement dynamic debug
Heavily inspired off linux kernel's dynamic debug:
 * add a /sys/kernel/debug/dynamic_debug/control file
 (accessible from linux side in /sys/class/mcos/mcos0/sys/kernel/debug/dynamic_debug/control)
 * read from file to list debug statements (currently limited to 4k in size)
 * write to file with '[file foo ][func bar ][line [x][-[y]]] [+-]p' to change values

Side effects:
 * reindented all linker scripts, there is a new __verbose section
 * added string function strpbrk

Change-Id: I36d7707274dcc3ecaf200075a31a2f0f76021059
2018-07-26 14:16:31 +09:00
bc887aab44 x86 futex: fix out of bounds shift
8 << 28 needs unsigned to fit, other shifts were done to truncate
the input, use a mask instead

Change-Id: I81ba41595f4629f1df554e34392116440ff3b641
2018-07-26 05:10:36 +00:00
6f7c428a34 terminate: fix oversubscribe hang when waiting for other threads on same CPU to die
Change-Id: I8c4fbdd3aab9d0567ce5457a4a6405490608925d
2018-07-26 05:02:13 +00:00
68c702d024 process_procfs_request: Add Pid to /proc/<PID>/status
The standard UNIX tool to get processes information, need to have the
process id inside /proc/<PID>/status.

Using ps without PID in /proc/<PID>/status gives :

  PID TTY          TIME CMD
 2551 pts/0    00:00:00 bash
    0 pts/0    00:00:00 exe
    0 pts/0    00:00:00 exe

With this patch:
  PID TTY          TIME CMD
 2551 pts/0    00:00:00 bash
11966 pts/0    00:00:00 exe
12619 pts/0    00:00:00 exe

Change-Id: Ic9d255cbef4d49e49bdaedcfc8e3545d9c144325
2018-07-26 05:00:21 +00:00
ad2cb6375a kprintf: only call eventfd() if it is safe to interrupt
Missing ARM64 implementation, cannot test right now

Change-Id: Ia05e8b7952b19bcd8fdac1f920d9bfe341be8b97
2018-07-26 04:57:30 +00:00
6df4bd8f8c Fix a few more warnings
Some are important, e.g. the seemingly harmless braces around if with dprintf,
since that dprintf is defined as empty, will screw things up and grab the next
line

Change-Id: Ie5e1cf813178ad708ff42ae5e477fbc96034471c
2018-07-26 04:52:17 +00:00
0994c3300e search_free_space: remove POSTK_DEBUG_ARCH_DEP_27 side
search_free_space changed since this was implemented and the code is
no longer compatible
Looking at it again, the function is not used anywhere other than syscall.c
and the second function does not seem to fix anything specific so this
just removes the untested side.

Change-Id: If28d35ec4da083a40dc6936fcb21f05fb64e378a
Fujitsu: POSTK_DEBUG_ARCH_DEP_27
2018-07-26 04:43:05 +00:00
a5c3e48843 search_free_space(): manage region->map_end internally
Change-Id: If9176773868c44fa1eb801c0815c35cea9f4b54b
2018-07-26 04:43:05 +00:00
df2c993721 fileobj_create: only allocate new object if one wasn't found
Change-Id: I5e12439333bf0c9cc7dad6e3cf410bfee616f77e
2018-07-26 04:41:03 +00:00
9d587dcbe8 fileobj_release: do not notify linux of surplus refs
Surplus refs on the linux side will not change anything, so spare
ourselves a message.
The final message will free all refs at once when the object is
destroyed.

Change-Id: Ie086b9dda663729962037c67e8233370509234a5
2018-07-26 04:08:43 +00:00
fd429ecc5b rusage_private: fix null pointer dereference
Change-Id: Id1f066699a41c249203073c5937e34012f5fe6c3
2018-07-26 03:50:44 +00:00
ed7f5abc28 schedule: fix null pointer dereferences
Change-Id: I1d4b0a2fabb5810a89cca4c6a0a837db3a9813ee
2018-07-26 03:50:44 +00:00
9a79920ef9 Static analysis fixes
Change-Id: I7bc42545a1c497f704d7bfa6ea1b7e3893acc697
2018-07-26 03:36:50 +00:00
bc3e6ded65 disable sse for everyone
GCC optimizes big switches with sse so we could clobber users floating
point registers when they would do a syscall

Reproducer:
```
 #include <stdio.h>
 #include <stdlib.h>

 union num {
 	float f;
 	unsigned long long i;
 };

 #define WORKSIZE (1024 * 1024 * 32)

 int main(int argc, char **argv) {
 	char *work = malloc(WORKSIZE);
 	char *fromaddr;
 	char sink;
 	union num r;
 	unsigned long long int offset;

 	r.f = drand48();
 	printf("r: %llx\n", (long long)r.i);
 	offset = (long long int)(r.f * (double)WORKSIZE);
 	fromaddr = work + offset;
 	printf("%e %llx %llx\n", r.f, offset, fromaddr);
 	sink = *fromaddr;

 	return 0;
 }
```

Change-Id: I7bb0883ec8ef2f245ab98064e308025422afc115
2018-07-26 03:26:25 +00:00
eae5c40f60 init_process_stack: Support "ulimit -s unlimited"
Refs: #1109
Change-Id: I395f012fd747cb6a2f93be71e34c7f6f3666ed67
2018-07-26 02:40:27 +00:00
67ebcca74d Fix to VMAP virtual address leak
Fujitsu: POSTK_DEBUG_TEMP_FIX_51
Refs: #1024
Change-Id: I1692ee4f004cb4d1f725baf47a8ed31fce1bf42a
2018-07-26 02:17:55 +00:00
94e96927a6 mremap: Do nothing when no size change and !MREMAP_FIXED
Behave in the same way as Linux which returns old_address when
old_size == new_size && !MREMAP_FIXED.

Refs: #1112
Change-Id: Ice1421a8a77f962d087de8475aa2cd40c59be5f7
2018-07-26 01:49:01 +00:00
3636c8e7e4 setrlimit: Check arguments in the same order as in Linux
(1) Check if rlim's address is valid
(2) Check if soft-limit does not exceed hard-limit

Fujitsu: POSTK_DEBUG_TEMP_FIX_3
Refs: #1050
Change-Id: I5bf1008ce172f9dff64ec89b1f97614926abaf13
2018-07-26 01:48:05 +00:00
4ce4c9f264 init_process: Inherit parent cpu_set
Fujitsu: POSTK_DEBUG_TEMP_FIX_69
Refs: #1028
Change-Id: I1628bb5bf35fa670bb0019e1f3ae295277b1566e
2018-07-26 00:44:41 +00:00
9bb8076dc0 shmget: Make shmobj underwent IPC_RMID invisible to shmget
Refs: #926
Change-Id: I16120623b581da5d5d484fd05d5111788c8ad5e2
2018-07-10 02:13:00 +00:00
992705d465 pager_get_path: Append \0 to path
Change-Id: Iaabd89a649bb20b37b35cd345da0f468fd5dd0b5
2018-07-10 02:10:19 +00:00
1cbe389879 do_fork: Propagate error code returned by mcexec
Refs: #731
Change-Id: I7eb52c1c76103d65d108b18b7beaf8041b51cd03
2018-07-03 09:19:54 +00:00
0758f6254e headers: declare void arguments for functions
Not giving any argument means that any argument is OK,
this is not what is meant here.

Change-Id: Ide651c1dec973d4b8709cf00646988f4c4f3acdd
2018-07-03 09:18:25 +00:00
db732a245c execve: Reinitialize vm_regions's map area on execve
Reinitialize vm->region.map_end in sys_execve()
in the same way as when creating a new process.

Change-Id: I7fc048a187e619ba4b5a578976e2a6774d13a6a7
2018-07-03 08:58:50 +00:00
08f2840f7d procfs: Show file names in /proc/<PID>/maps
Refs: #1065
Change-Id: I2f1603b02d12e60972c8f2e5f059d0025f4ceaea
2018-07-03 08:56:44 +00:00
521bdc6181 mremap: Fix type of size arguments (from ssize_t to size_t)
Refs: #1112
Change-Id: I3987d3a20a1e7c4b60f3880e91a670bc0bdc240f
2018-07-03 08:54:14 +00:00
e7b6a3472b sched_getaffinity: Check arguments in the same order as in Linux
(1) Check if size is large enough
(2) Check if size is positive

Fujitsu: POSTK_DEBUG_TEMP_FIX_5
Refs: #1121
Change-Id: I3e41720c89ef89294820f7f4fa8df1a69a7011b0
2018-07-03 08:53:30 +00:00
11756d96ef mmap, mremap: Check arguments in the same order as in Linux
Refs: #1137
Change-Id: I4fd2ac83b013a2741a3facce4dd7e0c37b14fd25
2018-07-03 08:41:30 +00:00
854bc85602 mcctrl: convert send_signal to mcctrl_ihk_send_wait
Change-Id: Ibd2fc834444d83341a96579f0c9c22080a53e8fa
2018-07-02 16:11:01 +09:00