rpms/crash
7
0
Fork 0
Code Issues Pull Requests Releases Activity

Rebase to upstream crash-8.0.2

Browse Source 
Release: crash-8.0.2-1

Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
This commit is contained in:
Lianbo Jiang 2022-11-17 12:04:08 +08:00
parent a76f3f3679
commit 8edbb22a5c
50 changed files with 19 additions and 4464 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@ -46,5 +46,6 @@ crash-5.0.6.tar.gz
/crash-7.3.0.tar.gz /crash-7.3.0.tar.gz
/crash-8.0.0.tar.gz /crash-8.0.0.tar.gz
/crash-8.0.1.tar.gz /crash-8.0.1.tar.gz
/crash-8.0.2.tar.gz
/gdb-7.6.tar.gz /gdb-7.6.tar.gz
/gdb-10.2.tar.gz /gdb-10.2.tar.gz

31
0001-ppc64-update-the-NR_CPUS-to-8192.patch
View File

@ -1,31 +0,0 @@
From ae52398a13fa9a238279114ed671c7c514c154ee Mon Sep 17 00:00:00 2001
From: Sourabh Jain <sourabhjain@linux.ibm.com>
Date: Mon, 9 May 2022 12:49:56 +0530
Subject: [PATCH 01/16] ppc64: update the NR_CPUS to 8192
Since the kernel commit 2d8ae638bb86 ("powerpc: Make the NR_CPUS max 8192")
the NR_CPUS on Linux kernel ranges from 1-8192. So let's match NR_CPUS with
the max NR_CPUS count on the Linux kernel.
Signed-off-by: Sourabh Jain <sourabhjain@linux.ibm.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/defs.h b/defs.h
index 1e8360d65a3b..a6735d07b32f 100644
--- a/defs.h
+++ b/defs.h
@@ -136,7 +136,7 @@
#define NR_CPUS (4096)
#endif
#ifdef PPC64
-#define NR_CPUS (2048)
+#define NR_CPUS (8192)
#endif
#ifdef S390
#define NR_CPUS (512)
--
2.30.2

44
0001-sbitmapq-Fix-for-kernels-without-struct-wait_queue_h.patch
View File

@ -1,44 +0,0 @@
From 6bc3b74c6e2b0aaebe1bc164594e53b010efef56 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 10 Jun 2022 15:52:34 +0900
Subject: [PATCH 01/29] sbitmapq: Fix for kernels without struct
wait_queue_head
The current struct wait_queue_head was renamed by kernel commit
9d9d676f595b ("sched/wait: Standardize internal naming of wait-queue heads")
at Linux 4.13. Without the patch, on earlier kernels the "sbitmapq"
command fails with the following error:
crash> sbitmapq ffff8801790b3b50
depth = 128
busy = 0
bits_per_word = 32
...
sbitmapq: invalid structure member offset: wait_queue_head_head
FILE: sbitmap.c LINE: 344 FUNCTION: sbitmap_queue_show()
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
sbitmap.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/sbitmap.c b/sbitmap.c
index bb2f19e6207b..be5d30a8ea88 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -341,7 +341,10 @@ static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
sbq_wait_state_size = SIZE(sbq_wait_state);
wait_cnt_off = OFFSET(sbq_wait_state_wait_cnt);
wait_off = OFFSET(sbq_wait_state_wait);
- list_head_off = OFFSET(wait_queue_head_head);
+ if (VALID_MEMBER(wait_queue_head_head)) /* 4.13 and later */
+ list_head_off = OFFSET(wait_queue_head_head);
+ else
+ list_head_off = OFFSET(__wait_queue_head_task_list);
sbq_wait_state_buf = GETBUF(sbq_wait_state_size);
--
2.37.1

43
0002-sbitmapq-Limit-kernels-without-sbitmap-again.patch
View File

@ -1,43 +0,0 @@
From b8f2ae6b494d706b1e4855b439c4930a6a6a2f5c Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 10 Jun 2022 16:00:14 +0900
Subject: [PATCH 02/29] sbitmapq: Limit kernels without sbitmap again
commit 364b2e413c69 ("sbitmapq: remove struct and member validation
in sbitmapq_init()") allowed the use of the "sbitmapq" command
unconditionally. Without the patch, the command fails with the
following error on kernels without sbitmap:
crash> sbitmapq ffff88015796e550
sbitmapq: invalid structure member offset: sbitmap_queue_sb
FILE: sbitmap.c LINE: 385 FUNCTION: sbitmap_queue_context_load()
Now the command supports Linux 4.9 and later kernels since it was
abstracted out, so it can be limited by the non-existence of the
sbitmap structure.
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
sbitmap.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/sbitmap.c b/sbitmap.c
index be5d30a8ea88..12d6512a1e4d 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -540,6 +540,10 @@ void sbitmapq_init(void)
STRUCT_SIZE_INIT(sbitmap_queue, "sbitmap_queue");
STRUCT_SIZE_INIT(sbq_wait_state, "sbq_wait_state");
+ /* sbitmap was abstracted out by commit 88459642cba4 on Linux 4.9. */
+ if (INVALID_SIZE(sbitmap))
+ command_not_supported();
+
MEMBER_OFFSET_INIT(sbitmap_word_depth, "sbitmap_word", "depth");
MEMBER_OFFSET_INIT(sbitmap_word_word, "sbitmap_word", "word");
MEMBER_OFFSET_INIT(sbitmap_word_cleared, "sbitmap_word", "cleared");
--
2.37.1

62
0002-sbitmapq-remove-struct-and-member-validation-in-sbit.patch
View File

@ -1,62 +0,0 @@
From 364b2e413c69daf189d2bc0238e3ba9b0dcbd937 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Mon, 23 May 2022 18:04:13 +0800
Subject: [PATCH 02/16] sbitmapq: remove struct and member validation in
sbitmapq_init()
Let's remove the struct and member validation from sbitmapq_init(), which
will help the crash to display the actual error when the sbitmapq fails.
Without the patch:
crash> sbitmapq ffff8e99d0dc8010
sbitmapq: command not supported or applicable on this architecture or kernel
With the patch:
crash> sbitmapq ffff8e99d0dc8010
sbitmapq: invalid structure member offset: sbitmap_queue_alloc_hint
FILE: sbitmap.c LINE: 365 FUNCTION: sbitmap_queue_context_load()
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
sbitmap.c | 24 ------------------------
1 file changed, 24 deletions(-)
diff --git a/sbitmap.c b/sbitmap.c
index 96a61e6c2c71..7693eef6cebd 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -525,30 +525,6 @@ void sbitmapq_init(void)
MEMBER_OFFSET_INIT(sbq_wait_state_wait_cnt, "sbq_wait_state", "wait_cnt");
MEMBER_OFFSET_INIT(sbq_wait_state_wait, "sbq_wait_state", "wait");
- if (!VALID_SIZE(sbitmap_word) ||
- !VALID_SIZE(sbitmap) ||
- !VALID_SIZE(sbitmap_queue) ||
- !VALID_SIZE(sbq_wait_state) ||
- INVALID_MEMBER(sbitmap_word_depth) ||
- INVALID_MEMBER(sbitmap_word_word) ||
- INVALID_MEMBER(sbitmap_word_cleared) ||
- INVALID_MEMBER(sbitmap_depth) ||
- INVALID_MEMBER(sbitmap_shift) ||
- INVALID_MEMBER(sbitmap_map_nr) ||
- INVALID_MEMBER(sbitmap_map) ||
- INVALID_MEMBER(sbitmap_queue_sb) ||
- INVALID_MEMBER(sbitmap_queue_alloc_hint) ||
- INVALID_MEMBER(sbitmap_queue_wake_batch) ||
- INVALID_MEMBER(sbitmap_queue_wake_index) ||
- INVALID_MEMBER(sbitmap_queue_ws) ||
- INVALID_MEMBER(sbitmap_queue_ws_active) ||
- INVALID_MEMBER(sbitmap_queue_round_robin) ||
- INVALID_MEMBER(sbitmap_queue_min_shallow_depth) ||
- INVALID_MEMBER(sbq_wait_state_wait_cnt) ||
- INVALID_MEMBER(sbq_wait_state_wait)) {
- command_not_supported();
- }
-
sb_flags |= SB_FLAG_INIT;
}
--
2.30.2

146
0003-Fix-for-dev-command-on-Linux-5.11-and-later.patch
View File

@ -1,146 +0,0 @@
From 85f39061390f095e73d9037f015cec077441eb13 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Wed, 15 Jun 2022 10:50:13 +0900
Subject: [PATCH 03/29] Fix for "dev" command on Linux 5.11 and later
The following kernel commits eventually removed the bdev_map array in
Linux v5.11 kernel:
e418de3abcda ("block: switch gendisk lookup to a simple xarray")
22ae8ce8b892 ("block: simplify bdev/disk lookup in blkdev_get")
Without the patch, the "dev" command fails to dump block device data
with the following error:
crash> dev
...
dev: blkdevs or all_bdevs: symbols do not exist
To get block device's gendisk, search blockdev_superblock.s_inodes
instead of bdev_map.
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
dev.c | 77 +++++++++++++++++++++++++++++++++++++++++++++++++++++++----
1 file changed, 72 insertions(+), 5 deletions(-)
diff --git a/dev.c b/dev.c
index db97f8aebdc2..75d30bd022a1 100644
--- a/dev.c
+++ b/dev.c
@@ -24,6 +24,7 @@ static void dump_blkdevs_v2(ulong);
static void dump_blkdevs_v3(ulong);
static ulong search_cdev_map_probes(char *, int, int, ulong *);
static ulong search_bdev_map_probes(char *, int, int, ulong *);
+static ulong search_blockdev_inodes(int, ulong *);
static void do_pci(void);
static void do_pci2(void);
static void do_io(void);
@@ -493,9 +494,10 @@ dump_blkdevs(ulong flags)
ulong ops;
} blkdevs[MAX_DEV], *bp;
- if (kernel_symbol_exists("major_names") &&
- kernel_symbol_exists("bdev_map")) {
- dump_blkdevs_v3(flags);
+ if (kernel_symbol_exists("major_names") &&
+ (kernel_symbol_exists("bdev_map") ||
+ kernel_symbol_exists("blockdev_superblock"))) {
+ dump_blkdevs_v3(flags);
return;
}
@@ -717,6 +719,7 @@ dump_blkdevs_v3(ulong flags)
char buf[BUFSIZE];
uint major;
ulong gendisk, addr, fops;
+ int use_bdev_map = kernel_symbol_exists("bdev_map");
if (!(len = get_array_length("major_names", NULL, 0)))
len = MAX_DEV;
@@ -745,8 +748,11 @@ dump_blkdevs_v3(ulong flags)
strncpy(buf, blk_major_name_buf +
OFFSET(blk_major_name_name), 16);
- fops = search_bdev_map_probes(buf, major == i ? major : i,
- UNUSED, &gendisk);
+ if (use_bdev_map)
+ fops = search_bdev_map_probes(buf, major == i ? major : i,
+ UNUSED, &gendisk);
+ else /* v5.11 and later */
+ fops = search_blockdev_inodes(major, &gendisk);
if (CRASHDEBUG(1))
fprintf(fp, "blk_major_name: %lx block major: %d name: %s gendisk: %lx fops: %lx\n",
@@ -829,6 +835,67 @@ search_bdev_map_probes(char *name, int major, int minor, ulong *gendisk)
return fops;
}
+/* For bdev_inode. See block/bdev.c */
+#define I_BDEV(inode) (inode - SIZE(block_device))
+
+static ulong
+search_blockdev_inodes(int major, ulong *gendisk)
+{
+ struct list_data list_data, *ld;
+ ulong addr, bd_sb, disk, fops = 0;
+ int i, inode_count, gendisk_major;
+ char *gendisk_buf;
+
+ ld = &list_data;
+ BZERO(ld, sizeof(struct list_data));
+
+ get_symbol_data("blockdev_superblock", sizeof(void *), &bd_sb);
+
+ addr = bd_sb + OFFSET(super_block_s_inodes);
+ if (!readmem(addr, KVADDR, &ld->start, sizeof(ulong),
+ "blockdev_superblock.s_inodes", QUIET|RETURN_ON_ERROR))
+ return 0;
+
+ if (empty_list(ld->start))
+ return 0;
+
+ ld->flags |= LIST_ALLOCATE;
+ ld->end = bd_sb + OFFSET(super_block_s_inodes);
+ ld->list_head_offset = OFFSET(inode_i_sb_list);
+
+ inode_count = do_list(ld);
+
+ gendisk_buf = GETBUF(SIZE(gendisk));
+
+ for (i = 0; i < inode_count; i++) {
+ addr = I_BDEV(ld->list_ptr[i]) + OFFSET(block_device_bd_disk);
+ if (!readmem(addr, KVADDR, &disk, sizeof(ulong),
+ "block_device.bd_disk", QUIET|RETURN_ON_ERROR))
+ continue;
+
+ if (!disk)
+ continue;
+
+ if (!readmem(disk, KVADDR, gendisk_buf, SIZE(gendisk),
+ "gendisk buffer", QUIET|RETURN_ON_ERROR))
+ continue;
+
+ gendisk_major = INT(gendisk_buf + OFFSET(gendisk_major));
+ if (gendisk_major != major)
+ continue;
+
+ fops = ULONG(gendisk_buf + OFFSET(gendisk_fops));
+ if (fops) {
+ *gendisk = disk;
+ break;
+ }
+ }
+
+ FREEBUF(ld->list_ptr);
+ FREEBUF(gendisk_buf);
+ return fops;
+}
+
void
dump_dev_table(void)
{
--
2.37.1

118
0003-sbitmapq-fix-invalid-offset-for-sbitmap_queue_alloc_.patch
View File

@ -1,118 +0,0 @@
From a295cb40cd5d24fb5995cc78d29c5def3843d285 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Mon, 23 May 2022 18:04:14 +0800
Subject: [PATCH 03/16] sbitmapq: fix invalid offset for
"sbitmap_queue_alloc_hint" on Linux v5.13-rc1
Kernel commit c548e62bcf6a ("scsi: sbitmap: Move allocation hint
into sbitmap") moved the alloc_hint member from struct sbitmap_queue
to struct sbitmap. Without the patch, the sbitmapq will fail:
crash> sbitmapq 0xffff8e99d0dc8010
sbitmapq: invalid structure member offset: sbitmap_queue_alloc_hint
FILE: sbitmap.c LINE: 365 FUNCTION: sbitmap_queue_context_load()
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 2 ++
sbitmap.c | 14 ++++++++++++--
symbols.c | 2 ++
3 files changed, 16 insertions(+), 2 deletions(-)
diff --git a/defs.h b/defs.h
index a6735d07b32f..0aeb98c4f654 100644
--- a/defs.h
+++ b/defs.h
@@ -2168,6 +2168,7 @@ struct offset_table { /* stash of commonly-used offsets */
long sbitmap_queue_min_shallow_depth;
long sbq_wait_state_wait_cnt;
long sbq_wait_state_wait;
+ long sbitmap_alloc_hint;
};
struct size_table { /* stash of commonly-used sizes */
@@ -5907,6 +5908,7 @@ struct sbitmap_context {
unsigned shift;
unsigned map_nr;
ulong map_addr;
+ ulong alloc_hint;
};
typedef bool (*sbitmap_for_each_fn)(unsigned int idx, void *p);
diff --git a/sbitmap.c b/sbitmap.c
index 7693eef6cebd..2921d5447c65 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -285,6 +285,7 @@ void sbitmap_for_each_set(const struct sbitmap_context *sc,
static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
const struct sbitmap_context *sc)
{
+ ulong alloc_hint_addr = 0;
int cpus = get_cpus_possible();
int sbq_wait_state_size, wait_cnt_off, wait_off, list_head_off;
char *sbq_wait_state_buf;
@@ -297,6 +298,11 @@ static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
fprintf(fp, "bits_per_word = %u\n", 1U << sc->shift);
fprintf(fp, "map_nr = %u\n", sc->map_nr);
+ if (VALID_MEMBER(sbitmap_queue_alloc_hint))
+ alloc_hint_addr = sqc->alloc_hint;
+ else if (VALID_MEMBER(sbitmap_alloc_hint)) /* 5.13 and later */
+ alloc_hint_addr = sc->alloc_hint;
+
fputs("alloc_hint = {", fp);
first = true;
for (i = 0; i < cpus; i++) {
@@ -307,7 +313,7 @@ static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
fprintf(fp, ", ");
first = false;
- ptr = kt->__per_cpu_offset[i] + sqc->alloc_hint;
+ ptr = kt->__per_cpu_offset[i] + alloc_hint_addr;
readmem(ptr, KVADDR, &val, sizeof(val), "alloc_hint", FAULT_ON_ERROR);
fprintf(fp, "%u", val);
@@ -362,7 +368,8 @@ static void sbitmap_queue_context_load(ulong addr, struct sbitmap_queue_context
error(FATAL, "cannot read sbitmap_queue\n");
}
- sqc->alloc_hint = ULONG(sbitmap_queue_buf + OFFSET(sbitmap_queue_alloc_hint));
+ if (VALID_MEMBER(sbitmap_queue_alloc_hint))
+ sqc->alloc_hint = ULONG(sbitmap_queue_buf + OFFSET(sbitmap_queue_alloc_hint));
sqc->wake_batch = UINT(sbitmap_queue_buf + OFFSET(sbitmap_queue_wake_batch));
sqc->wake_index = INT(sbitmap_queue_buf + OFFSET(sbitmap_queue_wake_index));
sqc->ws_addr = ULONG(sbitmap_queue_buf + OFFSET(sbitmap_queue_ws));
@@ -387,6 +394,8 @@ void sbitmap_context_load(ulong addr, struct sbitmap_context *sc)
sc->shift = UINT(sbitmap_buf + OFFSET(sbitmap_shift));
sc->map_nr = UINT(sbitmap_buf + OFFSET(sbitmap_map_nr));
sc->map_addr = ULONG(sbitmap_buf + OFFSET(sbitmap_map));
+ if (VALID_MEMBER(sbitmap_alloc_hint))
+ sc->alloc_hint = ULONG(sbitmap_buf + OFFSET(sbitmap_alloc_hint));
FREEBUF(sbitmap_buf);
}
@@ -512,6 +521,7 @@ void sbitmapq_init(void)
MEMBER_OFFSET_INIT(sbitmap_shift, "sbitmap", "shift");
MEMBER_OFFSET_INIT(sbitmap_map_nr, "sbitmap", "map_nr");
MEMBER_OFFSET_INIT(sbitmap_map, "sbitmap", "map");
+ MEMBER_OFFSET_INIT(sbitmap_alloc_hint, "sbitmap", "alloc_hint");
MEMBER_OFFSET_INIT(sbitmap_queue_sb, "sbitmap_queue", "sb");
MEMBER_OFFSET_INIT(sbitmap_queue_alloc_hint, "sbitmap_queue", "alloc_hint");
diff --git a/symbols.c b/symbols.c
index ba5e2741347d..fd0eb06899f0 100644
--- a/symbols.c
+++ b/symbols.c
@@ -10708,6 +10708,8 @@ dump_offset_table(char *spec, ulong makestruct)
OFFSET(sbitmap_map_nr));
fprintf(fp, " sbitmap_map: %ld\n",
OFFSET(sbitmap_map));
+ fprintf(fp, " sbitmap_alloc_hint: %ld\n",
+ OFFSET(sbitmap_alloc_hint));
fprintf(fp, " sbitmap_queue_sb: %ld\n",
OFFSET(sbitmap_queue_sb));
fprintf(fp, " sbitmap_queue_alloc_hint: %ld\n",
--
2.30.2

84
0004-Extend-field-length-of-task-attributes.patch
View File

@ -1,84 +0,0 @@
From d8869b08548362345fc34e4cf17a1eac9bddec6b Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Wed, 22 Jun 2022 08:32:59 +0900
Subject: [PATCH 04/29] Extend field length of task attributes
Nowadays, some machines have many CPU cores and memory, and some
distributions have a larger kernel.pid_max parameter, e.g. 7 digits.
This impairs the readability of a few commands, especially "ps" and
"ps -l|-m" options.
Let's extend the field length of the task attributes, PID, CPU, VSZ,
and RSS to improve the readability.
Without the patch:
crash> ps
PID PPID CPU TASK ST %MEM VSZ RSS COMM
...
2802197 2699997 2 ffff916f63c40000 IN 0.0 307212 10688 timer
2802277 1 0 ffff9161a25bb080 IN 0.0 169040 2744 gpg-agent
2806711 3167854 10 ffff9167fc498000 IN 0.0 127208 6508 su
2806719 2806711 1 ffff91633c3a48c0 IN 0.0 29452 6416 bash
2988346 1 5 ffff916f7c629840 IN 2.8 9342476 1917384 qemu-kvm
With the patch:
crash> ps
PID PPID CPU TASK ST %MEM VSZ RSS COMM
...
2802197 2699997 2 ffff916f63c40000 IN 0.0 307212 10688 timer
2802277 1 0 ffff9161a25bb080 IN 0.0 169040 2744 gpg-agent
2806711 3167854 10 ffff9167fc498000 IN 0.0 127208 6508 su
2806719 2806711 1 ffff91633c3a48c0 IN 0.0 29452 6416 bash
2988346 1 5 ffff916f7c629840 IN 2.8 9342476 1917384 qemu-kvm
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
task.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/task.c b/task.c
index 864c838637ee..071c787fbfa5 100644
--- a/task.c
+++ b/task.c
@@ -3828,7 +3828,7 @@ show_ps_data(ulong flag, struct task_context *tc, struct psinfo *psi)
} else
fprintf(fp, " ");
- fprintf(fp, "%5ld %5ld %2s %s %3s",
+ fprintf(fp, "%7ld %7ld %3s %s %3s",
tc->pid, task_to_pid(tc->ptask),
task_cpu(tc->processor, buf2, !VERBOSE),
task_pointer_string(tc, flag & PS_KSTACKP, buf3),
@@ -3838,8 +3838,8 @@ show_ps_data(ulong flag, struct task_context *tc, struct psinfo *psi)
if (strlen(buf1) == 3)
mkstring(buf1, 4, CENTER|RJUST, NULL);
fprintf(fp, "%s ", buf1);
- fprintf(fp, "%7ld ", (tm->total_vm * PAGESIZE())/1024);
- fprintf(fp, "%6ld ", (tm->rss * PAGESIZE())/1024);
+ fprintf(fp, "%8ld ", (tm->total_vm * PAGESIZE())/1024);
+ fprintf(fp, "%8ld ", (tm->rss * PAGESIZE())/1024);
if (is_kernel_thread(tc->task))
fprintf(fp, "[%s]\n", tc->comm);
else
@@ -3856,7 +3856,7 @@ show_ps(ulong flag, struct psinfo *psi)
if (!(flag & ((PS_EXCLUSIVE & ~PS_ACTIVE)|PS_NO_HEADER)))
fprintf(fp,
- " PID PPID CPU %s ST %%MEM VSZ RSS COMM\n",
+ " PID PPID CPU %s ST %%MEM VSZ RSS COMM\n",
flag & PS_KSTACKP ?
mkstring(buf, VADDR_PRLEN, CENTER|RJUST, "KSTACKP") :
mkstring(buf, VADDR_PRLEN, CENTER, "TASK"));
@@ -7713,7 +7713,7 @@ print_task_header(FILE *out, struct task_context *tc, int newline)
char buf[BUFSIZE];
char buf1[BUFSIZE];
- fprintf(out, "%sPID: %-5ld TASK: %s CPU: %-2s COMMAND: \"%s\"\n",
+ fprintf(out, "%sPID: %-7ld TASK: %s CPU: %-3s COMMAND: \"%s\"\n",
newline ? "\n" : "", tc->pid,
mkstring(buf1, VADDR_PRLEN, LJUST|LONG_HEX, MKSTR(tc->task)),
task_cpu(tc->processor, buf, !VERBOSE), tc->comm);
--
2.37.1

103
0004-sbitmapq-fix-invalid-offset-for-sbitmap_queue_round_.patch
View File

@ -1,103 +0,0 @@
From 530fe6ad7e4d7ff6254596c1219d25ed929e3867 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Mon, 23 May 2022 18:04:15 +0800
Subject: [PATCH 04/16] sbitmapq: fix invalid offset for
"sbitmap_queue_round_robin" on Linux v5.13-rc1
Kernel commit efe1f3a1d583 ("scsi: sbitmap: Maintain allocation
round_robin in sbitmap") moved the round_robin member from struct
sbitmap_queue to struct sbitmap. Without the patch, the sbitmapq
will fail:
crash> sbitmapq 0xffff8e99d0dc8010
sbitmapq: invalid structure member offset: sbitmap_queue_round_robin
FILE: sbitmap.c LINE: 378 FUNCTION: sbitmap_queue_context_load()
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 2 ++
sbitmap.c | 12 ++++++++++--
symbols.c | 2 ++
3 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/defs.h b/defs.h
index 0aeb98c4f654..ecbced24d2e3 100644
--- a/defs.h
+++ b/defs.h
@@ -2169,6 +2169,7 @@ struct offset_table { /* stash of commonly-used offsets */
long sbq_wait_state_wait_cnt;
long sbq_wait_state_wait;
long sbitmap_alloc_hint;
+ long sbitmap_round_robin;
};
struct size_table { /* stash of commonly-used sizes */
@@ -5909,6 +5910,7 @@ struct sbitmap_context {
unsigned map_nr;
ulong map_addr;
ulong alloc_hint;
+ bool round_robin;
};
typedef bool (*sbitmap_for_each_fn)(unsigned int idx, void *p);
diff --git a/sbitmap.c b/sbitmap.c
index 2921d5447c65..7b318b533702 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -352,7 +352,11 @@ static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
FREEBUF(sbq_wait_state_buf);
- fprintf(fp, "round_robin = %d\n", sqc->round_robin);
+ if (VALID_MEMBER(sbitmap_queue_round_robin))
+ fprintf(fp, "round_robin = %d\n", sqc->round_robin);
+ else if (VALID_MEMBER(sbitmap_round_robin)) /* 5.13 and later */
+ fprintf(fp, "round_robin = %d\n", sc->round_robin);
+
fprintf(fp, "min_shallow_depth = %u\n", sqc->min_shallow_depth);
}
@@ -374,7 +378,8 @@ static void sbitmap_queue_context_load(ulong addr, struct sbitmap_queue_context
sqc->wake_index = INT(sbitmap_queue_buf + OFFSET(sbitmap_queue_wake_index));
sqc->ws_addr = ULONG(sbitmap_queue_buf + OFFSET(sbitmap_queue_ws));
sqc->ws_active = INT(sbitmap_queue_buf + OFFSET(sbitmap_queue_ws_active));
- sqc->round_robin = BOOL(sbitmap_queue_buf + OFFSET(sbitmap_queue_round_robin));
+ if (VALID_MEMBER(sbitmap_queue_round_robin))
+ sqc->round_robin = BOOL(sbitmap_queue_buf + OFFSET(sbitmap_queue_round_robin));
sqc->min_shallow_depth = UINT(sbitmap_queue_buf + OFFSET(sbitmap_queue_min_shallow_depth));
FREEBUF(sbitmap_queue_buf);
@@ -396,6 +401,8 @@ void sbitmap_context_load(ulong addr, struct sbitmap_context *sc)
sc->map_addr = ULONG(sbitmap_buf + OFFSET(sbitmap_map));
if (VALID_MEMBER(sbitmap_alloc_hint))
sc->alloc_hint = ULONG(sbitmap_buf + OFFSET(sbitmap_alloc_hint));
+ if (VALID_MEMBER(sbitmap_round_robin))
+ sc->round_robin = BOOL(sbitmap_buf + OFFSET(sbitmap_round_robin));
FREEBUF(sbitmap_buf);
}
@@ -522,6 +529,7 @@ void sbitmapq_init(void)
MEMBER_OFFSET_INIT(sbitmap_map_nr, "sbitmap", "map_nr");
MEMBER_OFFSET_INIT(sbitmap_map, "sbitmap", "map");
MEMBER_OFFSET_INIT(sbitmap_alloc_hint, "sbitmap", "alloc_hint");
+ MEMBER_OFFSET_INIT(sbitmap_round_robin, "sbitmap", "round_robin");
MEMBER_OFFSET_INIT(sbitmap_queue_sb, "sbitmap_queue", "sb");
MEMBER_OFFSET_INIT(sbitmap_queue_alloc_hint, "sbitmap_queue", "alloc_hint");
diff --git a/symbols.c b/symbols.c
index fd0eb06899f0..5d12a021c769 100644
--- a/symbols.c
+++ b/symbols.c
@@ -10710,6 +10710,8 @@ dump_offset_table(char *spec, ulong makestruct)
OFFSET(sbitmap_map));
fprintf(fp, " sbitmap_alloc_hint: %ld\n",
OFFSET(sbitmap_alloc_hint));
+ fprintf(fp, " sbitmap_round_robin: %ld\n",
+ OFFSET(sbitmap_round_robin));
fprintf(fp, " sbitmap_queue_sb: %ld\n",
OFFSET(sbitmap_queue_sb));
fprintf(fp, " sbitmap_queue_alloc_hint: %ld\n",
--
2.30.2

45
0005-ppc64-fix-bt-for-S-case.patch
View File

@ -1,45 +0,0 @@
From c67ce5bbb8e37d28f1c26b239b203a6561f574c1 Mon Sep 17 00:00:00 2001
From: Hari Bathini <hbathini@linux.ibm.com>
Date: Mon, 4 Jul 2022 10:55:41 +0530
Subject: [PATCH 05/29] ppc64: fix bt for '-S' case
Passing '-S' option to 'bt' command was intended to specify the stack
pointer manually. But get_stack_frame() handling on ppc64 is ignoring
this option altogether. Fix it.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
ppc64.c | 16 ++++++++++++++++
1 file changed, 16 insertions(+)
diff --git a/ppc64.c b/ppc64.c
index 975caa53b812..0e1d8678eef5 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -2330,6 +2330,22 @@ ppc64_vmcore_stack_frame(struct bt_info *bt_in, ulong *nip, ulong *ksp)
pt_regs = (struct ppc64_pt_regs *)bt_in->machdep;
if (!pt_regs || !pt_regs->gpr[1]) {
+ if (bt_in->hp) {
+ if (bt_in->hp->esp) {
+ *ksp = bt_in->hp->esp;
+ if (!bt_in->hp->eip) {
+ if (IS_KVADDR(*ksp)) {
+ readmem(*ksp+16, KVADDR, &unip, sizeof(ulong),
+ "Regs NIP value", FAULT_ON_ERROR);
+ *nip = unip;
+ }
+ } else
+ *nip = bt_in->hp->eip;
+
+ }
+ return TRUE;
+ }
+
/*
* Not collected regs. May be the corresponding CPU not
* responded to an IPI in case of KDump OR f/w has not
--
2.37.1

101
0005-sbitmapq-fix-invalid-offset-for-sbitmap_word_depth-o.patch
View File

@ -1,101 +0,0 @@
From 3750803f6ae5f5ad071f86ca916dbbb17b7a83a5 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Mon, 23 May 2022 18:04:16 +0800
Subject: [PATCH 05/16] sbitmapq: fix invalid offset for "sbitmap_word_depth"
on Linux v5.18-rc1
Kernel commit 3301bc53358a ("lib/sbitmap: kill 'depth' from sbitmap_word")
removed the depth member from struct sbitmap_word. Without the patch, the
sbitmapq will fail:
crash> sbitmapq 0xffff8e99d0dc8010
sbitmapq: invalid structure member offset: sbitmap_word_depth
FILE: sbitmap.c LINE: 84 FUNCTION: __sbitmap_weight()
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
sbitmap.c | 19 +++++++++++--------
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/sbitmap.c b/sbitmap.c
index 7b318b533702..e8ebd62fe01c 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -78,10 +78,16 @@ static unsigned long bitmap_weight(unsigned long bitmap, unsigned int bits)
return w;
}
+static inline unsigned int __map_depth(const struct sbitmap_context *sc, int index)
+{
+ if (index == sc->map_nr - 1)
+ return sc->depth - (index << sc->shift);
+ return 1U << sc->shift;
+}
+
static unsigned int __sbitmap_weight(const struct sbitmap_context *sc, bool set)
{
const ulong sbitmap_word_size = SIZE(sbitmap_word);
- const ulong w_depth_off = OFFSET(sbitmap_word_depth);
const ulong w_word_off = OFFSET(sbitmap_word_word);
const ulong w_cleared_off = OFFSET(sbitmap_word_cleared);
@@ -99,7 +105,7 @@ static unsigned int __sbitmap_weight(const struct sbitmap_context *sc, bool set)
error(FATAL, "cannot read sbitmap_word\n");
}
- depth = ULONG(sbitmap_word_buf + w_depth_off);
+ depth = __map_depth(sc, i);
if (set) {
word = ULONG(sbitmap_word_buf + w_word_off);
@@ -142,7 +148,6 @@ static void sbitmap_emit_byte(unsigned int offset, uint8_t byte)
static void sbitmap_bitmap_show(const struct sbitmap_context *sc)
{
const ulong sbitmap_word_size = SIZE(sbitmap_word);
- const ulong w_depth_off = OFFSET(sbitmap_word_depth);
const ulong w_word_off = OFFSET(sbitmap_word_word);
const ulong w_cleared_off = OFFSET(sbitmap_word_cleared);
@@ -165,7 +170,7 @@ static void sbitmap_bitmap_show(const struct sbitmap_context *sc)
word = ULONG(sbitmap_word_buf + w_word_off);
cleared = ULONG(sbitmap_word_buf + w_cleared_off);
- word_bits = ULONG(sbitmap_word_buf + w_depth_off);
+ word_bits = __map_depth(sc, i);
word &= ~cleared;
@@ -213,7 +218,6 @@ static void __sbitmap_for_each_set(const struct sbitmap_context *sc,
unsigned int start, sbitmap_for_each_fn fn, void *data)
{
const ulong sbitmap_word_size = SIZE(sbitmap_word);
- const ulong w_depth_off = OFFSET(sbitmap_word_depth);
const ulong w_word_off = OFFSET(sbitmap_word_word);
const ulong w_cleared_off = OFFSET(sbitmap_word_cleared);
@@ -232,7 +236,7 @@ static void __sbitmap_for_each_set(const struct sbitmap_context *sc,
while (scanned < sc->depth) {
unsigned long w_addr = sc->map_addr + (sbitmap_word_size * index);
- unsigned long w_depth, w_word, w_cleared;
+ unsigned long w_word, w_cleared;
unsigned long word, depth;
if (!readmem(w_addr, KVADDR, sbitmap_word_buf, sbitmap_word_size, "sbitmap_word", RETURN_ON_ERROR)) {
@@ -240,11 +244,10 @@ static void __sbitmap_for_each_set(const struct sbitmap_context *sc,
error(FATAL, "cannot read sbitmap_word\n");
}
- w_depth = ULONG(sbitmap_word_buf + w_depth_off);
w_word = ULONG(sbitmap_word_buf + w_word_off);
w_cleared = ULONG(sbitmap_word_buf + w_cleared_off);
- depth = min(w_depth - nr, sc->depth - scanned);
+ depth = min(__map_depth(sc, index) - nr, sc->depth - scanned);
scanned += depth;
word = w_word & ~w_cleared;
--
2.30.2

27
0006-Makefile-add-missing-crash_target.o-to-be-cleaned.patch
View File

@ -1,27 +0,0 @@
From 9705669a49c341402efd8528e8fe809379dd798d Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Mon, 23 May 2022 14:48:50 +0900
Subject: [PATCH 06/16] Makefile: add missing crash_target.o to be cleaned
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/Makefile b/Makefile
index e520b1217a9d..162c2baa5164 100644
--- a/Makefile
+++ b/Makefile
@@ -184,7 +184,7 @@ GDB_7.6_FILES=
GDB_7.6_OFILES=${GDB}/gdb/symtab.o
GDB_10.2_FILES=
-GDB_10.2_OFILES=${GDB}/gdb/symtab.o
+GDB_10.2_OFILES=${GDB}/gdb/symtab.o crash_target.o
#
# GDB_FLAGS is passed up from the gdb Makefile.
--
2.30.2

147
0006-ppc64-dynamically-allocate-h-w-interrupt-stack.patch
View File

@ -1,147 +0,0 @@
From 3ee5956721d9a67fe8d4c6d5022aa022c5f9a11c Mon Sep 17 00:00:00 2001
From: Hari Bathini <hbathini@linux.ibm.com>
Date: Mon, 4 Jul 2022 10:55:42 +0530
Subject: [PATCH 06/29] ppc64: dynamically allocate h/w interrupt stack
Only older kernel (v2.4) used h/w interrupt stack to store frames when
CPU received IPI. Memory used for this in 'struct machine_specific' is
useless for later kernels. For the sake of backward compatibility keep
h/w interrupt stack but dynamically allocate memory for it and save
some bytes from being wasted.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 2 +-
ppc64.c | 51 +++++++++++++++++++++------------------------------
2 files changed, 22 insertions(+), 31 deletions(-)
diff --git a/defs.h b/defs.h
index 7d3b73422f48..d1d3ea919e70 100644
--- a/defs.h
+++ b/defs.h
@@ -6303,7 +6303,7 @@ struct ppc64_vmemmap {
* Used to store the HW interrupt stack. It is only for 2.4.
*/
struct machine_specific {
- ulong hwintrstack[NR_CPUS];
+ ulong *hwintrstack;
char *hwstackbuf;
uint hwstacksize;
diff --git a/ppc64.c b/ppc64.c
index 0e1d8678eef5..272eb207074a 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -256,7 +256,7 @@ static int set_ppc64_max_physmem_bits(void)
}
struct machine_specific ppc64_machine_specific = {
- .hwintrstack = { 0 },
+ .hwintrstack = NULL,
.hwstackbuf = 0,
.hwstacksize = 0,
.pte_rpn_shift = PTE_RPN_SHIFT_DEFAULT,
@@ -275,7 +275,7 @@ struct machine_specific ppc64_machine_specific = {
};
struct machine_specific book3e_machine_specific = {
- .hwintrstack = { 0 },
+ .hwintrstack = NULL,
.hwstackbuf = 0,
.hwstacksize = 0,
.pte_rpn_shift = PTE_RPN_SHIFT_L4_BOOK3E_64K,
@@ -676,6 +676,9 @@ ppc64_init(int when)
*/
offset = MEMBER_OFFSET("paca_struct", "xHrdIntStack");
paca_sym = symbol_value("paca");
+ if (!(machdep->machspec->hwintrstack =
+ (ulong *)calloc(NR_CPUS, sizeof(ulong))))
+ error(FATAL, "cannot malloc hwintrstack space.");
for (cpu = 0; cpu < kt->cpus; cpu++) {
readmem(paca_sym + (paca_size * cpu) + offset,
KVADDR,
@@ -686,14 +689,9 @@ ppc64_init(int when)
machdep->machspec->hwstacksize = 8 * machdep->pagesize;
if ((machdep->machspec->hwstackbuf = (char *)
malloc(machdep->machspec->hwstacksize)) == NULL)
- error(FATAL, "cannot malloc hwirqstack space.");
- } else
- /*
- * 'xHrdIntStack' member in "paca_struct" is not
- * available for 2.6 kernel.
- */
- BZERO(&machdep->machspec->hwintrstack,
- NR_CPUS*sizeof(ulong));
+ error(FATAL, "cannot malloc hwirqstack buffer space.");
+ }
+
if (!machdep->hz) {
machdep->hz = HZ;
if (THIS_KERNEL_VERSION >= LINUX(2,6,0))
@@ -846,23 +844,15 @@ ppc64_dump_machdep_table(ulong arg)
fprintf(fp, " is_vmaddr: %s\n",
machdep->machspec->is_vmaddr == book3e_is_vmaddr ?
"book3e_is_vmaddr()" : "ppc64_is_vmaddr()");
- fprintf(fp, " hwintrstack[%d]: ", NR_CPUS);
- for (c = 0; c < NR_CPUS; c++) {
- for (others = 0, i = c; i < NR_CPUS; i++) {
- if (machdep->machspec->hwintrstack[i])
- others++;
+ if (machdep->machspec->hwintrstack) {
+ fprintf(fp, " hwintrstack[%d]: ", NR_CPUS);
+ for (c = 0; c < NR_CPUS; c++) {
+ fprintf(fp, "%s%016lx ",
+ ((c % 4) == 0) ? "\n " : "",
+ machdep->machspec->hwintrstack[c]);
}
- if (!others) {
- fprintf(fp, "%s%s",
- c && ((c % 4) == 0) ? "\n " : "",
- c ? "(remainder unused)" : "(unused)");
- break;
- }
-
- fprintf(fp, "%s%016lx ",
- ((c % 4) == 0) ? "\n " : "",
- machdep->machspec->hwintrstack[c]);
- }
+ } else
+ fprintf(fp, " hwintrstack: (unused)");
fprintf(fp, "\n");
fprintf(fp, " hwstackbuf: %lx\n", (ulong)machdep->machspec->hwstackbuf);
fprintf(fp, " hwstacksize: %d\n", machdep->machspec->hwstacksize);
@@ -1683,9 +1673,10 @@ ppc64_check_sp_in_HWintrstack(ulong sp, struct bt_info *bt)
*
* Note: HW Interrupt stack is used only in 2.4 kernel.
*/
- if (is_task_active(bt->task) && (tt->panic_task != bt->task) &&
- machdep->machspec->hwintrstack[bt->tc->processor]) {
+ if (machdep->machspec->hwintrstack && is_task_active(bt->task) &&
+ (bt->task != tt->panic_task)) {
ulong newsp;
+
readmem(machdep->machspec->hwintrstack[bt->tc->processor],
KVADDR, &newsp, sizeof(ulong),
"stack pointer", FAULT_ON_ERROR);
@@ -1958,7 +1949,7 @@ ppc64_back_trace(struct gnu_request *req, struct bt_info *bt)
bt->stackbase = irqstack;
bt->stacktop = bt->stackbase + STACKSIZE();
alter_stackbuf(bt);
- } else if (ms->hwintrstack[bt->tc->processor]) {
+ } else if (ms->hwintrstack) {
bt->stacktop = ms->hwintrstack[bt->tc->processor] +
sizeof(ulong);
bt->stackbase = ms->hwintrstack[bt->tc->processor] -
@@ -2555,7 +2546,7 @@ retry:
goto retry;
}
- if (check_intrstack && ms->hwintrstack[bt->tc->processor]) {
+ if (check_intrstack && ms->hwintrstack) {
bt->stacktop = ms->hwintrstack[bt->tc->processor] +
sizeof(ulong);
bt->stackbase = ms->hwintrstack[bt->tc->processor] -
--
2.37.1

205
0007-bt-x86_64-filter-out-idle-task-stack.patch
View File

@ -1,205 +0,0 @@
From 6833262bf87177d8affe4f91b2e7d2c76ecdf636 Mon Sep 17 00:00:00 2001
From: Qi Zheng <zhengqi.arch@bytedance.com>
Date: Tue, 24 May 2022 20:25:53 +0800
Subject: [PATCH 07/16] bt: x86_64: filter out idle task stack
When we use crash to troubleshoot softlockup and other problems,
we often use the 'bt -a' command to print the stacks of running
processes on all CPUs. But now some servers have hundreds of CPUs
(such as AMD machines), which causes the 'bt -a' command to output
a lot of process stacks. And many of these stacks are the stacks
of the idle process, which are not needed by us.
Therefore, in order to reduce this part of the interference information,
this patch adds the -n option to the bt command. When we specify
'-n idle' (meaning no idle), the stack of the idle process will be
filtered out, thus speeding up our troubleshooting.
And the option works only for crash dumps captured by kdump.
The command output is as follows:
crash> bt -a -n idle
[...]
PID: 0 TASK: ffff889ff8c34380 CPU: 8 COMMAND: "swapper/8"
PID: 0 TASK: ffff889ff8c32d00 CPU: 9 COMMAND: "swapper/9"
PID: 0 TASK: ffff889ff8c31680 CPU: 10 COMMAND: "swapper/10"
PID: 0 TASK: ffff889ff8c35a00 CPU: 11 COMMAND: "swapper/11"
PID: 0 TASK: ffff889ff8c3c380 CPU: 12 COMMAND: "swapper/12"
PID: 150773 TASK: ffff889fe85a1680 CPU: 13 COMMAND: "bash"
#0 [ffffc9000d35bcd0] machine_kexec at ffffffff8105a407
#1 [ffffc9000d35bd28] __crash_kexec at ffffffff8113033d
#2 [ffffc9000d35bdf0] panic at ffffffff81081930
#3 [ffffc9000d35be70] sysrq_handle_crash at ffffffff814e38d1
#4 [ffffc9000d35be78] __handle_sysrq.cold.12 at ffffffff814e4175
#5 [ffffc9000d35bea8] write_sysrq_trigger at ffffffff814e404b
#6 [ffffc9000d35beb8] proc_reg_write at ffffffff81330d86
#7 [ffffc9000d35bed0] vfs_write at ffffffff812a72d5
#8 [ffffc9000d35bf00] ksys_write at ffffffff812a7579
#9 [ffffc9000d35bf38] do_syscall_64 at ffffffff81004259
RIP: 00007fa7abcdc274 RSP: 00007fffa731f678 RFLAGS: 00000246
RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007fa7abcdc274
RDX: 0000000000000002 RSI: 0000563ca51ee6d0 RDI: 0000000000000001
RBP: 0000563ca51ee6d0 R8: 000000000000000a R9: 00007fa7abd6be80
R10: 000000000000000a R11: 0000000000000246 R12: 00007fa7abdad760
R13: 0000000000000002 R14: 00007fa7abda8760 R15: 0000000000000002
ORIG_RAX: 0000000000000001 CS: 0033 SS: 002b
[...]
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Acked-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Acked-by: Lianbo Jiang <lijiang@redhat.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 1 +
help.c | 33 ++++++++++++++++++++++++++++++++-
kernel.c | 13 ++++++++++++-
x86_64.c | 8 ++++++++
4 files changed, 53 insertions(+), 2 deletions(-)
diff --git a/defs.h b/defs.h
index ecbced24d2e3..c8444b4e54eb 100644
--- a/defs.h
+++ b/defs.h
@@ -5832,6 +5832,7 @@ ulong cpu_map_addr(const char *type);
#define BT_SHOW_ALL_REGS (0x2000000000000ULL)
#define BT_REGS_NOT_FOUND (0x4000000000000ULL)
#define BT_OVERFLOW_STACK (0x8000000000000ULL)
+#define BT_SKIP_IDLE (0x10000000000000ULL)
#define BT_SYMBOL_OFFSET (BT_SYMBOLIC_ARGS)
#define BT_REF_HEXVAL (0x1)
diff --git a/help.c b/help.c
index 51a0fe3d687c..e1bbc5abe029 100644
--- a/help.c
+++ b/help.c
@@ -1909,12 +1909,14 @@ char *help_bt[] = {
"bt",
"backtrace",
"[-a|-c cpu(s)|-g|-r|-t|-T|-l|-e|-E|-f|-F|-o|-O|-v|-p] [-R ref] [-s [-x|d]]"
-"\n [-I ip] [-S sp] [pid | task]",
+"\n [-I ip] [-S sp] [-n idle] [pid | task]",
" Display a kernel stack backtrace. If no arguments are given, the stack",
" trace of the current context will be displayed.\n",
" -a displays the stack traces of the active task on each CPU.",
" (only applicable to crash dumps)",
" -A same as -a, but also displays vector registers (S390X only).",
+" -n idle filter the stack of idle tasks (x86_64).",
+" (only applicable to crash dumps)",
" -p display the stack trace of the panic task only.",
" (only applicable to crash dumps)",
" -c cpu display the stack trace of the active task on one or more CPUs,",
@@ -2004,6 +2006,35 @@ char *help_bt[] = {
" DS: 002b ESI: bfffc8a0 ES: 002b EDI: 00000000 ",
" SS: 002b ESP: bfffc82c EBP: bfffd224 ",
" CS: 0023 EIP: 400d032e ERR: 0000008e EFLAGS: 00000246 ",
+" ",
+" Display the stack trace of the active task(s) when the kernel panicked,",
+" and filter out the stack of the idle tasks:",
+" ",
+" %s> bt -a -n idle",
+" ...",
+" PID: 0 TASK: ffff889ff8c35a00 CPU: 11 COMMAND: \"swapper/11\"",
+" ",
+" PID: 0 TASK: ffff889ff8c3c380 CPU: 12 COMMAND: \"swapper/12\"",
+" ",
+" PID: 150773 TASK: ffff889fe85a1680 CPU: 13 COMMAND: \"bash\"",
+" #0 [ffffc9000d35bcd0] machine_kexec at ffffffff8105a407",
+" #1 [ffffc9000d35bd28] __crash_kexec at ffffffff8113033d",
+" #2 [ffffc9000d35bdf0] panic at ffffffff81081930",
+" #3 [ffffc9000d35be70] sysrq_handle_crash at ffffffff814e38d1",
+" #4 [ffffc9000d35be78] __handle_sysrq.cold.12 at ffffffff814e4175",
+" #5 [ffffc9000d35bea8] write_sysrq_trigger at ffffffff814e404b",
+" #6 [ffffc9000d35beb8] proc_reg_write at ffffffff81330d86",
+" #7 [ffffc9000d35bed0] vfs_write at ffffffff812a72d5",
+" #8 [ffffc9000d35bf00] ksys_write at ffffffff812a7579",
+" #9 [ffffc9000d35bf38] do_syscall_64 at ffffffff81004259",
+" RIP: 00007fa7abcdc274 RSP: 00007fffa731f678 RFLAGS: 00000246",
+" RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007fa7abcdc274",
+" RDX: 0000000000000002 RSI: 0000563ca51ee6d0 RDI: 0000000000000001",
+" RBP: 0000563ca51ee6d0 R8: 000000000000000a R9: 00007fa7abd6be80",
+" R10: 000000000000000a R11: 0000000000000246 R12: 00007fa7abdad760",
+" R13: 0000000000000002 R14: 00007fa7abda8760 R15: 0000000000000002",
+" ORIG_RAX: 0000000000000001 CS: 0033 SS: 002b",
+" ...",
"\n Display the stack trace of the active task on CPU 0 and 1:\n",
" %s> bt -c 0,1",
" PID: 0 TASK: ffffffff81a8d020 CPU: 0 COMMAND: \"swapper\"",
diff --git a/kernel.c b/kernel.c
index d0921cf567d9..411e9da1e54f 100644
--- a/kernel.c
+++ b/kernel.c
@@ -2503,7 +2503,7 @@ cmd_bt(void)
if (kt->flags & USE_OPT_BT)
bt->flags |= BT_OPT_BACK_TRACE;
- while ((c = getopt(argcnt, args, "D:fFI:S:c:aAloreEgstTdxR:Ovp")) != EOF) {
+ while ((c = getopt(argcnt, args, "D:fFI:S:c:n:aAloreEgstTdxR:Ovp")) != EOF) {
switch (c)
{
case 'f':
@@ -2672,6 +2672,13 @@ cmd_bt(void)
active++;
break;
+ case 'n':
+ if (machine_type("X86_64") && STREQ(optarg, "idle"))
+ bt->flags |= BT_SKIP_IDLE;
+ else
+ option_not_supported(c);
+ break;
+
case 'r':
bt->flags |= BT_RAW;
break;
@@ -3092,6 +3099,10 @@ back_trace(struct bt_info *bt)
} else
machdep->get_stack_frame(bt, &eip, &esp);
+ /* skip idle task stack */
+ if (bt->flags & BT_SKIP_IDLE)
+ return;
+
if (bt->flags & BT_KSTACKP) {
bt->stkptr = esp;
return;
diff --git a/x86_64.c b/x86_64.c
index ecaefd2f46a8..cfafbcc4dabe 100644
--- a/x86_64.c
+++ b/x86_64.c
@@ -4918,6 +4918,9 @@ x86_64_get_stack_frame(struct bt_info *bt, ulong *pcp, ulong *spp)
if (bt->flags & BT_DUMPFILE_SEARCH)
return x86_64_get_dumpfile_stack_frame(bt, pcp, spp);
+ if (bt->flags & BT_SKIP_IDLE)
+ bt->flags &= ~BT_SKIP_IDLE;
+
if (pcp)
*pcp = x86_64_get_pc(bt);
if (spp)
@@ -4960,6 +4963,9 @@ x86_64_get_dumpfile_stack_frame(struct bt_info *bt_in, ulong *rip, ulong *rsp)
estack = -1;
panic = FALSE;
+ if (bt_in->flags & BT_SKIP_IDLE)
+ bt_in->flags &= ~BT_SKIP_IDLE;
+
panic_task = tt->panic_task == bt->task ? TRUE : FALSE;
if (panic_task && bt->machdep) {
@@ -5098,6 +5104,8 @@ next_sysrq:
if (!panic_task && STREQ(sym, "crash_nmi_callback")) {
*rip = *up;
*rsp = bt->stackbase + ((char *)(up) - bt->stackbuf);
+ if ((bt->flags & BT_SKIP_IDLE) && is_idle_thread(bt->task))
+ bt_in->flags |= BT_SKIP_IDLE;
return;
}
--
2.30.2

56
0007-ppc64-rename-ppc64_paca_init-to-ppc64_paca_percpu_of.patch
View File

@ -1,56 +0,0 @@
From 4d1b968abb286ea39ea080ae073b0e2b5bfe6c4e Mon Sep 17 00:00:00 2001
From: Hari Bathini <hbathini@linux.ibm.com>
Date: Mon, 4 Jul 2022 10:55:43 +0530
Subject: [PATCH 07/29] ppc64: rename ppc64_paca_init to
ppc64_paca_percpu_offset_init
ppc64_paca_init() function is specifically used to initialize percpu
data_offset for kernels older than v2.6.36. So, the name is slightly
misleading. Rename it to ppc64_paca_percpu_offset_init to reflect its
purpose.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
ppc64.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/ppc64.c b/ppc64.c
index 272eb207074a..0a3aa5f7af91 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -52,7 +52,7 @@ static char * ppc64_check_eframe(struct ppc64_pt_regs *);
static void ppc64_print_eframe(char *, struct ppc64_pt_regs *,
struct bt_info *);
static void parse_cmdline_args(void);
-static int ppc64_paca_init(int);
+static int ppc64_paca_percpu_offset_init(int);
static void ppc64_init_cpu_info(void);
static int ppc64_get_cpu_map(void);
static void ppc64_clear_machdep_cache(void);
@@ -3285,7 +3285,7 @@ parse_cmdline_args(void)
* Initialize the per cpu data_offset values from paca structure.
*/
static int
-ppc64_paca_init(int map)
+ppc64_paca_percpu_offset_init(int map)
{
int i, cpus, nr_paca;
char *cpu_paca_buf;
@@ -3387,10 +3387,11 @@ ppc64_init_cpu_info(void)
* which was removed post v2.6.15 ppc64 and now we get the per cpu
* data_offset from __per_cpu_offset symbol during kernel_init()
* call. Hence for backward (pre-2.6.36) compatibility, call
- * ppc64_paca_init() only if symbol __per_cpu_offset does not exist.
+ * ppc64_paca_percpu_offset_init() only if symbol __per_cpu_offset
+ * does not exist.
*/
if (!symbol_exists("__per_cpu_offset"))
- cpus = ppc64_paca_init(map);
+ cpus = ppc64_paca_percpu_offset_init(map);
else {
if (!(nr_cpus = get_array_length("__per_cpu_offset", NULL, 0)))
nr_cpus = (kt->kernel_NR_CPUS ? kt->kernel_NR_CPUS :
--
2.37.1

96
0008-bt-arm64-add-support-for-bt-n-idle.patch
View File

@ -1,96 +0,0 @@
From 0f162febebc4d11a165dd40cee00f3b0ba691a52 Mon Sep 17 00:00:00 2001
From: Qi Zheng <zhengqi.arch@bytedance.com>
Date: Tue, 24 May 2022 20:25:54 +0800
Subject: [PATCH 08/16] bt: arm64: add support for 'bt -n idle'
The '-n idle' option of bt command can help us filter the
stack of the idle process when debugging the dumpfiles
captured by kdump.
This patch supports this feature on ARM64.
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
arm64.c | 19 ++++++++++++++++---
help.c | 2 +-
kernel.c | 3 ++-
3 files changed, 19 insertions(+), 5 deletions(-)
diff --git a/arm64.c b/arm64.c
index 65f6cdf69fa6..0f615cf52bef 100644
--- a/arm64.c
+++ b/arm64.c
@@ -3681,6 +3681,12 @@ arm64_get_dumpfile_stackframe(struct bt_info *bt, struct arm64_stackframe *frame
{
struct machine_specific *ms = machdep->machspec;
struct arm64_pt_regs *ptregs;
+ bool skip = false;
+
+ if (bt->flags & BT_SKIP_IDLE) {
+ skip = true;
+ bt->flags &= ~BT_SKIP_IDLE;
+ }
if (!ms->panic_task_regs ||
(!ms->panic_task_regs[bt->tc->processor].sp &&
@@ -3713,8 +3719,11 @@ try_kernel:
}
if (arm64_in_kdump_text(bt, frame) ||
- arm64_in_kdump_text_on_irq_stack(bt))
+ arm64_in_kdump_text_on_irq_stack(bt)) {
bt->flags |= BT_KDUMP_ADJUST;
+ if (skip && is_idle_thread(bt->task))
+ bt->flags |= BT_SKIP_IDLE;
+ }
return TRUE;
}
@@ -3738,10 +3747,14 @@ arm64_get_stack_frame(struct bt_info *bt, ulong *pcp, ulong *spp)
int ret;
struct arm64_stackframe stackframe = { 0 };
- if (DUMPFILE() && is_task_active(bt->task))
+ if (DUMPFILE() && is_task_active(bt->task)) {
ret = arm64_get_dumpfile_stackframe(bt, &stackframe);
- else
+ } else {
+ if (bt->flags & BT_SKIP_IDLE)
+ bt->flags &= ~BT_SKIP_IDLE;
+
ret = arm64_get_stackframe(bt, &stackframe);
+ }
if (!ret)
error(WARNING,
diff --git a/help.c b/help.c
index e1bbc5abe029..99214c1590fa 100644
--- a/help.c
+++ b/help.c
@@ -1915,7 +1915,7 @@ char *help_bt[] = {
" -a displays the stack traces of the active task on each CPU.",
" (only applicable to crash dumps)",
" -A same as -a, but also displays vector registers (S390X only).",
-" -n idle filter the stack of idle tasks (x86_64).",
+" -n idle filter the stack of idle tasks (x86_64, arm64).",
" (only applicable to crash dumps)",
" -p display the stack trace of the panic task only.",
" (only applicable to crash dumps)",
diff --git a/kernel.c b/kernel.c
index 411e9da1e54f..a521ef30cdb0 100644
--- a/kernel.c
+++ b/kernel.c
@@ -2673,7 +2673,8 @@ cmd_bt(void)
break;
case 'n':
- if (machine_type("X86_64") && STREQ(optarg, "idle"))
+ if ((machine_type("X86_64") || machine_type("ARM64")) &&
+ STREQ(optarg, "idle"))
bt->flags |= BT_SKIP_IDLE;
else
option_not_supported(c);
--
2.30.2

352
0008-ppc64-handle-backtrace-when-CPU-is-in-an-emergency-s.patch
View File

@ -1,352 +0,0 @@
From cdd57e8b16aba2f5714673368d6dbc7565d59841 Mon Sep 17 00:00:00 2001
From: Hari Bathini <hbathini@linux.ibm.com>
Date: Mon, 4 Jul 2022 10:55:44 +0530
Subject: [PATCH 08/29] ppc64: handle backtrace when CPU is in an emergency
stack
A CPU could be in an emergency stack when it is running in real mode
or any special scenario like TM bad thing. Also, there are dedicated
emergency stacks for machine check and system reset interrupt. Right
now, no backtrace is provided if a CPU is in any of these stacks.
This change ensures backtrace is processed appropriately even when
a CPU is in any one of these emergency stacks. Also, if stack info
cannot be found, print that message always instead of only when
verbose logs are enabled.
Related kernel commits:
729b0f715371 ("powerpc/book3s: Introduce exclusive emergency stack for machine check exception.")
b1ee8a3de579 ("powerpc/64s: Dedicated system reset interrupt stack")
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 12 ++++
ppc64.c | 203 ++++++++++++++++++++++++++++++++++++++++++++++++++++----
2 files changed, 203 insertions(+), 12 deletions(-)
diff --git a/defs.h b/defs.h
index d1d3ea919e70..9b1b69a4f421 100644
--- a/defs.h
+++ b/defs.h
@@ -6288,6 +6288,13 @@ struct ppc64_elf_prstatus {
#ifdef PPC64
+enum emergency_stack_type {
+ NONE_STACK = 0,
+ EMERGENCY_STACK,
+ NMI_EMERGENCY_STACK,
+ MC_EMERGENCY_STACK
+};
+
struct ppc64_opal {
uint64_t base;
uint64_t entry;
@@ -6307,6 +6314,11 @@ struct machine_specific {
char *hwstackbuf;
uint hwstacksize;
+ /* Emergency stacks */
+ ulong *emergency_sp;
+ ulong *nmi_emergency_sp;
+ ulong *mc_emergency_sp;
+
uint l4_index_size;
uint l3_index_size;
uint l2_index_size;
diff --git a/ppc64.c b/ppc64.c
index 0a3aa5f7af91..03047a85955d 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -48,6 +48,10 @@ static ulong ppc64_get_stackbase(ulong);
static ulong ppc64_get_stacktop(ulong);
void ppc64_compiler_warning_stub(void);
static ulong ppc64_in_irqstack(ulong);
+static enum emergency_stack_type ppc64_in_emergency_stack(int cpu, ulong addr,
+ bool verbose);
+static void ppc64_set_bt_emergency_stack(enum emergency_stack_type type,
+ struct bt_info *bt);
static char * ppc64_check_eframe(struct ppc64_pt_regs *);
static void ppc64_print_eframe(char *, struct ppc64_pt_regs *,
struct bt_info *);
@@ -56,6 +60,7 @@ static int ppc64_paca_percpu_offset_init(int);
static void ppc64_init_cpu_info(void);
static int ppc64_get_cpu_map(void);
static void ppc64_clear_machdep_cache(void);
+static void ppc64_init_paca_info(void);
static void ppc64_vmemmap_init(void);
static int ppc64_get_kvaddr_ranges(struct vaddr_range *);
static uint get_ptetype(ulong pte);
@@ -692,6 +697,8 @@ ppc64_init(int when)
error(FATAL, "cannot malloc hwirqstack buffer space.");
}
+ ppc64_init_paca_info();
+
if (!machdep->hz) {
machdep->hz = HZ;
if (THIS_KERNEL_VERSION >= LINUX(2,6,0))
@@ -1204,6 +1211,70 @@ ppc64_kvtop(struct task_context *tc, ulong kvaddr,
return ppc64_vtop(kvaddr, (ulong *)vt->kernel_pgd[0], paddr, verbose);
}
+static void
+ppc64_init_paca_info(void)
+{
+ struct machine_specific *ms = machdep->machspec;
+ ulong *paca_ptr;
+ int i;
+
+ if (!(paca_ptr = (ulong *)calloc(kt->cpus, sizeof(ulong))))
+ error(FATAL, "cannot malloc paca pointers space.\n");
+
+ /* Get paca pointers for all CPUs. */
+ if (symbol_exists("paca_ptrs")) {
+ ulong paca_loc;
+
+ readmem(symbol_value("paca_ptrs"), KVADDR, &paca_loc, sizeof(void *),
+ "paca double pointer", FAULT_ON_ERROR);
+ readmem(paca_loc, KVADDR, paca_ptr, sizeof(void *) * kt->cpus,
+ "paca pointers", FAULT_ON_ERROR);
+ } else if (symbol_exists("paca") &&
+ (get_symbol_type("paca", NULL, NULL) == TYPE_CODE_PTR)) {
+ readmem(symbol_value("paca"), KVADDR, paca_ptr, sizeof(void *) * kt->cpus,
+ "paca pointers", FAULT_ON_ERROR);
+ } else {
+ free(paca_ptr);
+ return;
+ }
+
+ /* Initialize emergency stacks info. */
+ if (MEMBER_EXISTS("paca_struct", "emergency_sp")) {
+ ulong offset = MEMBER_OFFSET("paca_struct", "emergency_sp");
+
+ if (!(ms->emergency_sp = (ulong *)calloc(kt->cpus, sizeof(ulong))))
+ error(FATAL, "cannot malloc emergency stack space.\n");
+ for (i = 0; i < kt->cpus; i++)
+ readmem(paca_ptr[i] + offset, KVADDR, &ms->emergency_sp[i],
+ sizeof(void *), "paca->emergency_sp",
+ FAULT_ON_ERROR);
+ }
+
+ if (MEMBER_EXISTS("paca_struct", "nmi_emergency_sp")) {
+ ulong offset = MEMBER_OFFSET("paca_struct", "nmi_emergency_sp");
+
+ if (!(ms->nmi_emergency_sp = (ulong *)calloc(kt->cpus, sizeof(ulong))))
+ error(FATAL, "cannot malloc NMI emergency stack space.\n");
+ for (i = 0; i < kt->cpus; i++)
+ readmem(paca_ptr[i] + offset, KVADDR, &ms->nmi_emergency_sp[i],
+ sizeof(void *), "paca->nmi_emergency_sp",
+ FAULT_ON_ERROR);
+ }
+
+ if (MEMBER_EXISTS("paca_struct", "mc_emergency_sp")) {
+ ulong offset = MEMBER_OFFSET("paca_struct", "mc_emergency_sp");
+
+ if (!(ms->mc_emergency_sp = (ulong *)calloc(kt->cpus, sizeof(ulong))))
+ error(FATAL, "cannot malloc machine check emergency stack space.\n");
+ for (i = 0; i < kt->cpus; i++)
+ readmem(paca_ptr[i] + offset, KVADDR, &ms->mc_emergency_sp[i],
+ sizeof(void *), "paca->mc_emergency_sp",
+ FAULT_ON_ERROR);
+ }
+
+ free(paca_ptr);
+}
+
/*
* Verify that the kernel has made the vmemmap list available,
* and if so, stash the relevant data required to make vtop
@@ -1755,6 +1826,11 @@ ppc64_eframe_search(struct bt_info *bt_in)
addr = bt->stackbase +
roundup(SIZE(thread_info), sizeof(ulong));
} else if (!INSTACK(addr, bt)) {
+ enum emergency_stack_type estype;
+
+ if ((estype = ppc64_in_emergency_stack(bt->tc->processor, addr, false)))
+ ppc64_set_bt_emergency_stack(estype, bt);
+
/*
* If the user specified SP is in HW interrupt stack
* (only for tasks running on other CPUs and in 2.4
@@ -1856,6 +1932,84 @@ ppc64_in_irqstack(ulong addr)
return 0;
}
+/*
+ * Check if the CPU is running in any of its emergency stacks.
+ * Returns
+ * NONE_STACK : if input is invalid or addr is not within any emergency stack.
+ * EMERGENCY_STACK : if the addr is within emergency stack.
+ * NMI_EMERGENCY_STACK : if the addr is within NMI emergency stack.
+ * MC_EMERGENCY_STACK : if the addr is within machine check emergency stack.
+ */
+static enum emergency_stack_type
+ppc64_in_emergency_stack(int cpu, ulong addr, bool verbose)
+{
+ struct machine_specific *ms = machdep->machspec;
+ ulong base, top;
+
+ if (cpu < 0 || cpu >= kt->cpus)
+ return NONE_STACK;
+
+ if (ms->emergency_sp) {
+ top = ms->emergency_sp[cpu];
+ base = top - STACKSIZE();
+ if (addr >= base && addr < top) {
+ if (verbose)
+ fprintf(fp, "---<Emergency Stack>---\n");
+ return EMERGENCY_STACK;
+ }
+ }
+
+ if (ms->nmi_emergency_sp) {
+ top = ms->nmi_emergency_sp[cpu];
+ base = top - STACKSIZE();
+ if (addr >= base && addr < top) {
+ if (verbose)
+ fprintf(fp, "---<NMI Emergency Stack>---\n");
+ return NMI_EMERGENCY_STACK;
+ }
+ }
+
+ if (ms->mc_emergency_sp) {
+ top = ms->mc_emergency_sp[cpu];
+ base = top - STACKSIZE();
+ if (addr >= base && addr < top) {
+ if (verbose)
+ fprintf(fp, "---<Machine Check Emergency Stack>---\n");
+ return MC_EMERGENCY_STACK;
+ }
+ }
+
+ return NONE_STACK;
+}
+
+static void
+ppc64_set_bt_emergency_stack(enum emergency_stack_type type, struct bt_info *bt)
+{
+ struct machine_specific *ms = machdep->machspec;
+ ulong top;
+
+ switch (type) {
+ case EMERGENCY_STACK:
+ top = ms->emergency_sp[bt->tc->processor];
+ break;
+ case NMI_EMERGENCY_STACK:
+ top = ms->nmi_emergency_sp[bt->tc->processor];
+ break;
+ case MC_EMERGENCY_STACK:
+ top = ms->mc_emergency_sp[bt->tc->processor];
+ break;
+ default:
+ top = 0;
+ break;
+ }
+
+ if (top) {
+ bt->stackbase = top - STACKSIZE();
+ bt->stacktop = top;
+ alter_stackbuf(bt);
+ }
+}
+
/*
* Unroll a kernel stack.
*/
@@ -1936,10 +2090,13 @@ ppc64_back_trace_cmd(struct bt_info *bt)
static void
ppc64_back_trace(struct gnu_request *req, struct bt_info *bt)
{
- int frame = 0;
- ulong lr = 0; /* hack...need to pass in initial lr reg */
+ enum emergency_stack_type estype;
ulong newpc = 0, newsp, marker;
+ int c = bt->tc->processor;
+ ulong nmi_sp = 0;
int eframe_found;
+ int frame = 0;
+ ulong lr = 0; /* hack...need to pass in initial lr reg */
if (!INSTACK(req->sp, bt)) {
ulong irqstack;
@@ -1949,6 +2106,10 @@ ppc64_back_trace(struct gnu_request *req, struct bt_info *bt)
bt->stackbase = irqstack;
bt->stacktop = bt->stackbase + STACKSIZE();
alter_stackbuf(bt);
+ } else if ((estype = ppc64_in_emergency_stack(c, req->sp, true))) {
+ if (estype == NMI_EMERGENCY_STACK)
+ nmi_sp = req->sp;
+ ppc64_set_bt_emergency_stack(estype, bt);
} else if (ms->hwintrstack) {
bt->stacktop = ms->hwintrstack[bt->tc->processor] +
sizeof(ulong);
@@ -1957,9 +2118,7 @@ ppc64_back_trace(struct gnu_request *req, struct bt_info *bt)
bt->stackbuf = ms->hwstackbuf;
alter_stackbuf(bt);
} else {
- if (CRASHDEBUG(1)) {
- fprintf(fp, "cannot find the stack info.\n");
- }
+ fprintf(fp, "cannot find the stack info.\n");
return;
}
}
@@ -1989,13 +2148,20 @@ ppc64_back_trace(struct gnu_request *req, struct bt_info *bt)
newsp =
*(ulong *)&bt->stackbuf[newsp - bt->stackbase];
if (!INSTACK(newsp, bt)) {
- /*
- * Switch HW interrupt stack to process's stack.
- */
- bt->stackbase = GET_STACKBASE(bt->task);
- bt->stacktop = GET_STACKTOP(bt->task);
- alter_stackbuf(bt);
- }
+ if ((estype = ppc64_in_emergency_stack(c, newsp, true))) {
+ if (!nmi_sp && estype == NMI_EMERGENCY_STACK)
+ nmi_sp = newsp;
+ ppc64_set_bt_emergency_stack(estype, bt);
+ } else {
+ /*
+ * Switch HW interrupt stack or emergency stack
+ * to process's stack.
+ */
+ bt->stackbase = GET_STACKBASE(bt->task);
+ bt->stacktop = GET_STACKTOP(bt->task);
+ alter_stackbuf(bt);
+ }
+ }
if (IS_KVADDR(newsp) && INSTACK(newsp, bt))
newpc = *(ulong *)&bt->stackbuf[newsp + 16 -
bt->stackbase];
@@ -2039,6 +2205,16 @@ ppc64_back_trace(struct gnu_request *req, struct bt_info *bt)
}
}
+ /*
+ * NMI stack may not be re-entrant. In so, an SP in the NMI stack
+ * is likely to point back to an SP within the NMI stack, in case
+ * of a nested NMI.
+ */
+ if (nmi_sp && nmi_sp == newsp) {
+ fprintf(fp, "---<Nested NMI>---\n");
+ break;
+ }
+
/*
* Some Linux 3.7 kernel threads have been seen to have
* their end-of-trace stack linkage pointer pointing
@@ -2416,6 +2592,9 @@ ppc64_get_dumpfile_stack_frame(struct bt_info *bt_in, ulong *nip, ulong *ksp)
pt_regs = (struct ppc64_pt_regs *)bt->machdep;
ur_nip = pt_regs->nip;
ur_ksp = pt_regs->gpr[1];
+ /* Print the collected regs for panic task. */
+ ppc64_print_regs(pt_regs);
+ ppc64_print_nip_lr(pt_regs, 1);
} else if ((pc->flags & KDUMP) ||
((pc->flags & DISKDUMP) &&
(*diskdump_flags & KDUMP_CMPRS_LOCAL))) {
--
2.37.1

67
0009-gdb-print-details-of-unnamed-struct-and-union.patch
View File

@ -1,67 +0,0 @@
From dda5b2d02b8d8de1264f84b6267582aa7a1e5a57 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Tue, 31 May 2022 17:12:16 +0900
Subject: [PATCH 09/16] gdb: print details of unnamed struct and union
Currently gdb's "ptype" command does not print the details of unnamed
structure and union deeper than second level in a structure, it prints
only "{...}" instead. And crash's "struct" and similar commands also
inherit this behavior, so we cannot get the full information of them.
To print the details of them, change the show variable when it is an
unnamed one like crash-7.x.
Without the patch:
crash> struct -o page
struct page {
[0] unsigned long flags;
union {
struct {...};
struct {...};
...
With the patch:
crash> struct -o page
struct page {
[0] unsigned long flags;
union {
struct {
[8] struct list_head lru;
[24] struct address_space *mapping;
[32] unsigned long index;
[40] unsigned long private;
};
struct {
[8] dma_addr_t dma_addr;
};
...
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
gdb-10.2.patch | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/gdb-10.2.patch b/gdb-10.2.patch
index b67db4e1ded9..577f5e45fc5a 100644
--- a/gdb-10.2.patch
+++ b/gdb-10.2.patch
@@ -1638,3 +1638,15 @@ exit 0
$(ECHO_CXXLD) $(CC_LD) $(INTERNAL_LDFLAGS) $(WIN32LDAPP) \
-o $(shell /bin/cat mergeobj) $(LIBGDB_OBS) \
$(TDEPLIBS) $(TUI_LIBRARY) $(CLIBS) $(LOADLIBES) $(shell /bin/cat mergelibs)
+--- gdb-10.2/gdb/c-typeprint.c.orig
++++ gdb-10.2/gdb/c-typeprint.c
+@@ -1202,6 +1202,9 @@ c_type_print_base_struct_union (struct t
+ = podata->end_bitpos
+ - TYPE_LENGTH (type->field (i).type ()) * TARGET_CHAR_BIT;
+ }
++ else if (strlen(TYPE_FIELD_NAME (type, i)) == 0)
++ /* crash: Print details for unnamed struct and union. */
++ newshow = show;
+
+ c_print_type_1 (type->field (i).type (),
+ TYPE_FIELD_NAME (type, i),
--
2.30.2

74
0009-ppc64-print-emergency-stacks-info-with-mach-command.patch
View File

@ -1,74 +0,0 @@
From 4dc2f1c32d1c99586e67032c9cd62c5c4334049c Mon Sep 17 00:00:00 2001
From: Hari Bathini <hbathini@linux.ibm.com>
Date: Mon, 4 Jul 2022 10:55:45 +0530
Subject: [PATCH 09/29] ppc64: print emergency stacks info with 'mach' command
Print top address of emergency stacks with 'mach' command.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
ppc64.c | 40 ++++++++++++++++++++++++++++++++++++++++
1 file changed, 40 insertions(+)
diff --git a/ppc64.c b/ppc64.c
index 03047a85955d..ad1d6e121e81 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -3161,6 +3161,44 @@ opalmsg(void)
fprintf(fp, "\n");
}
+static void ppc64_print_emergency_stack_info(void)
+{
+ struct machine_specific *ms = machdep->machspec;
+ char buf[32];
+ int i;
+
+ fprintf(fp, " EMERGENCY STACK: ");
+ if (ms->emergency_sp) {
+ fprintf(fp, "\n");
+ for (i = 0; i < kt->cpus; i++) {
+ sprintf(buf, "CPU %d", i);
+ fprintf(fp, "%19s: %lx\n", buf, ms->emergency_sp[i]);
+ }
+ } else
+ fprintf(fp, "(unused)\n");
+
+ fprintf(fp, "NMI EMERGENCY STACK: ");
+ if (ms->nmi_emergency_sp) {
+ fprintf(fp, "\n");
+ for (i = 0; i < kt->cpus; i++) {
+ sprintf(buf, "CPU %d", i);
+ fprintf(fp, "%19s: %lx\n", buf, ms->nmi_emergency_sp[i]);
+ }
+ } else
+ fprintf(fp, "(unused)\n");
+
+ fprintf(fp, " MC EMERGENCY STACK: ");
+ if (ms->mc_emergency_sp) {
+ fprintf(fp, "\n");
+ for (i = 0; i < kt->cpus; i++) {
+ sprintf(buf, "CPU %d", i);
+ fprintf(fp, "%19s: %lx\n", buf, ms->mc_emergency_sp[i]);
+ }
+ } else
+ fprintf(fp, "(unused)\n");
+ fprintf(fp, "\n");
+}
+
/*
* Machine dependent command.
*/
@@ -3241,6 +3279,8 @@ ppc64_display_machine_stats(void)
fprintf(fp, "%19s: %lx\n", buf, tt->softirq_ctx[c]);
}
}
+
+ ppc64_print_emergency_stack_info();
}
static const char *hook_files[] = {
--
2.37.1

392
0010-Enhance-dev-d-D-options-to-support-blk-mq-sbitmap.patch
View File

@ -1,392 +0,0 @@
From 7095c8fd029e3a33117e3b67de73f504686ebfe2 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Thu, 2 Jun 2022 20:12:55 +0800
Subject: [PATCH 10/16] Enhance "dev -d|-D" options to support blk-mq sbitmap
Since Linux 5.16-rc1, which kernel commit 9a14d6ce4135 ("block: remove
debugfs blk_mq_ctx dispatched/merged/completed attributes") removed the
members from struct blk_mq_ctx, crash has not displayed disk I/O statistics
for multiqueue (blk-mq) devices.
Let's parse the sbitmap in blk-mq layer to support it.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
---
defs.h | 11 +++
dev.c | 244 +++++++++++++++++++++++++++++++++++++++++++++---------
symbols.c | 22 +++++
3 files changed, 238 insertions(+), 39 deletions(-)
diff --git a/defs.h b/defs.h
index c8444b4e54eb..2681586a33dc 100644
--- a/defs.h
+++ b/defs.h
@@ -2170,6 +2170,16 @@ struct offset_table { /* stash of commonly-used offsets */
long sbq_wait_state_wait;
long sbitmap_alloc_hint;
long sbitmap_round_robin;
+ long request_cmd_flags;
+ long request_q;
+ long request_state;
+ long request_queue_queue_hw_ctx;
+ long request_queue_nr_hw_queues;
+ long blk_mq_hw_ctx_tags;
+ long blk_mq_tags_bitmap_tags;
+ long blk_mq_tags_breserved_tags;
+ long blk_mq_tags_nr_reserved_tags;
+ long blk_mq_tags_rqs;
};
struct size_table { /* stash of commonly-used sizes */
@@ -2339,6 +2349,7 @@ struct size_table { /* stash of commonly-used sizes */
long sbitmap;
long sbitmap_queue;
long sbq_wait_state;
+ long blk_mq_tags;
};
struct array_table {
diff --git a/dev.c b/dev.c
index a493e51ac95c..4be4c96df8b0 100644
--- a/dev.c
+++ b/dev.c
@@ -4238,19 +4238,176 @@ get_one_mctx_diskio(unsigned long mctx, struct diskio *io)
io->write = (dispatch[1] - comp[1]);
}
+typedef bool (busy_tag_iter_fn)(ulong rq, void *data);
+
+struct mq_inflight {
+ ulong q;
+ struct diskio *dio;
+};
+
+struct bt_iter_data {
+ ulong tags;
+ uint reserved;
+ uint nr_reserved_tags;
+ busy_tag_iter_fn *fn;
+ void *data;
+};
+
+/*
+ * See the include/linux/blk_types.h and include/linux/blk-mq.h
+ */
+#define MQ_RQ_IN_FLIGHT 1
+#define REQ_OP_BITS 8
+#define REQ_OP_MASK ((1 << REQ_OP_BITS) - 1)
+
+static uint op_is_write(uint op)
+{
+ return (op & REQ_OP_MASK) & 1;
+}
+
+static bool mq_check_inflight(ulong rq, void *data)
+{
+ uint cmd_flags = 0, state = 0;
+ ulong addr = 0, queue = 0;
+ struct mq_inflight *mi = data;
+
+ if (!IS_KVADDR(rq))
+ return TRUE;
+
+ addr = rq + OFFSET(request_q);
+ if (!readmem(addr, KVADDR, &queue, sizeof(ulong), "request.q", RETURN_ON_ERROR))
+ return FALSE;
+
+ addr = rq + OFFSET(request_cmd_flags);
+ if (!readmem(addr, KVADDR, &cmd_flags, sizeof(uint), "request.cmd_flags", RETURN_ON_ERROR))
+ return FALSE;
+
+ addr = rq + OFFSET(request_state);
+ if (!readmem(addr, KVADDR, &state, sizeof(uint), "request.state", RETURN_ON_ERROR))
+ return FALSE;
+
+ if (queue == mi->q && state == MQ_RQ_IN_FLIGHT) {
+ if (op_is_write(cmd_flags))
+ mi->dio->write++;
+ else
+ mi->dio->read++;
+ }
+
+ return TRUE;
+}
+
+static bool bt_iter(uint bitnr, void *data)
+{
+ ulong addr = 0, rqs_addr = 0, rq = 0;
+ struct bt_iter_data *iter_data = data;
+ ulong tag = iter_data->tags;
+
+ if (!iter_data->reserved)
+ bitnr += iter_data->nr_reserved_tags;
+
+ /* rqs */
+ addr = tag + OFFSET(blk_mq_tags_rqs);
+ if (!readmem(addr, KVADDR, &rqs_addr, sizeof(void *), "blk_mq_tags.rqs", RETURN_ON_ERROR))
+ return FALSE;
+
+ addr = rqs_addr + bitnr * sizeof(ulong); /* rqs[bitnr] */
+ if (!readmem(addr, KVADDR, &rq, sizeof(ulong), "blk_mq_tags.rqs[]", RETURN_ON_ERROR))
+ return FALSE;
+
+ return iter_data->fn(rq, iter_data->data);
+}
+
+static void bt_for_each(ulong q, ulong tags, ulong sbq, uint reserved, uint nr_resvd_tags, struct diskio *dio)
+{
+ struct sbitmap_context sc = {0};
+ struct mq_inflight mi = {
+ .q = q,
+ .dio = dio,
+ };
+ struct bt_iter_data iter_data = {
+ .tags = tags,
+ .reserved = reserved,
+ .nr_reserved_tags = nr_resvd_tags,
+ .fn = mq_check_inflight,
+ .data = &mi,
+ };
+
+ sbitmap_context_load(sbq + OFFSET(sbitmap_queue_sb), &sc);
+ sbitmap_for_each_set(&sc, bt_iter, &iter_data);
+}
+
+static void queue_for_each_hw_ctx(ulong q, ulong *hctx, uint cnt, struct diskio *dio)
+{
+ uint i;
+
+ for (i = 0; i < cnt; i++) {
+ ulong addr = 0, tags = 0;
+ uint nr_reserved_tags = 0;
+
+ /* Tags owned by the block driver */
+ addr = hctx[i] + OFFSET(blk_mq_hw_ctx_tags);
+ if (!readmem(addr, KVADDR, &tags, sizeof(ulong),
+ "blk_mq_hw_ctx.tags", RETURN_ON_ERROR))
+ break;
+
+ addr = tags + OFFSET(blk_mq_tags_nr_reserved_tags);
+ if (!readmem(addr, KVADDR, &nr_reserved_tags, sizeof(uint),
+ "blk_mq_tags_nr_reserved_tags", RETURN_ON_ERROR))
+ break;
+
+ if (nr_reserved_tags) {
+ addr = tags + OFFSET(blk_mq_tags_breserved_tags);
+ bt_for_each(q, tags, addr, 1, nr_reserved_tags, dio);
+ }
+ addr = tags + OFFSET(blk_mq_tags_bitmap_tags);
+ bt_for_each(q, tags, addr, 0, nr_reserved_tags, dio);
+ }
+}
+
+static void get_mq_diskio_from_hw_queues(ulong q, struct diskio *dio)
+{
+ uint cnt = 0;
+ ulong addr = 0, hctx_addr = 0;
+ ulong *hctx_array = NULL;
+
+ addr = q + OFFSET(request_queue_nr_hw_queues);
+ readmem(addr, KVADDR, &cnt, sizeof(uint),
+ "request_queue.nr_hw_queues", FAULT_ON_ERROR);
+
+ addr = q + OFFSET(request_queue_queue_hw_ctx);
+ readmem(addr, KVADDR, &hctx_addr, sizeof(void *),
+ "request_queue.queue_hw_ctx", FAULT_ON_ERROR);
+
+ hctx_array = (ulong *)GETBUF(sizeof(void *) * cnt);
+ if (!hctx_array)
+ error(FATAL, "fail to get memory for the hctx_array\n");
+
+ if (!readmem(hctx_addr, KVADDR, hctx_array, sizeof(void *) * cnt,
+ "request_queue.queue_hw_ctx[]", RETURN_ON_ERROR)) {
+ FREEBUF(hctx_array);
+ return;
+ }
+
+ queue_for_each_hw_ctx(q, hctx_array, cnt, dio);
+
+ FREEBUF(hctx_array);
+}
+
static void
get_mq_diskio(unsigned long q, unsigned long *mq_count)
{
int cpu;
unsigned long queue_ctx;
unsigned long mctx_addr;
- struct diskio tmp;
+ struct diskio tmp = {0};
if (INVALID_MEMBER(blk_mq_ctx_rq_dispatched) ||
- INVALID_MEMBER(blk_mq_ctx_rq_completed))
+ INVALID_MEMBER(blk_mq_ctx_rq_completed)) {
+ get_mq_diskio_from_hw_queues(q, &tmp);
+ mq_count[0] = tmp.read;
+ mq_count[1] = tmp.write;
return;
-
- memset(&tmp, 0x00, sizeof(struct diskio));
+ }
readmem(q + OFFSET(request_queue_queue_ctx), KVADDR, &queue_ctx,
sizeof(ulong), "request_queue.queue_ctx",
@@ -4479,41 +4636,24 @@ display_one_diskio(struct iter *i, unsigned long gendisk, ulong flags)
&& (io.read + io.write == 0))
return;
- if (use_mq_interface(queue_addr) &&
- (INVALID_MEMBER(blk_mq_ctx_rq_dispatched) ||
- INVALID_MEMBER(blk_mq_ctx_rq_completed)))
- fprintf(fp, "%s%s%s %s%s%s%s %s%s%s",
- mkstring(buf0, 5, RJUST|INT_DEC, (char *)(unsigned long)major),
- space(MINSPACE),
- mkstring(buf1, VADDR_PRLEN, LJUST|LONG_HEX, (char *)gendisk),
- space(MINSPACE),
- mkstring(buf2, 10, LJUST, disk_name),
- space(MINSPACE),
- mkstring(buf3, VADDR_PRLEN <= 11 ? 11 : VADDR_PRLEN,
- LJUST|LONG_HEX, (char *)queue_addr),
- space(MINSPACE),
- mkstring(buf4, 17, RJUST, "(not supported)"),
- space(MINSPACE));
-
- else
- fprintf(fp, "%s%s%s %s%s%s%s %s%5d%s%s%s%s%s",
- mkstring(buf0, 5, RJUST|INT_DEC, (char *)(unsigned long)major),
- space(MINSPACE),
- mkstring(buf1, VADDR_PRLEN, LJUST|LONG_HEX, (char *)gendisk),
- space(MINSPACE),
- mkstring(buf2, 10, LJUST, disk_name),
- space(MINSPACE),
- mkstring(buf3, VADDR_PRLEN <= 11 ? 11 : VADDR_PRLEN,
- LJUST|LONG_HEX, (char *)queue_addr),
- space(MINSPACE),
- io.read + io.write,
- space(MINSPACE),
- mkstring(buf4, 5, RJUST|INT_DEC,
- (char *)(unsigned long)io.read),
- space(MINSPACE),
- mkstring(buf5, 5, RJUST|INT_DEC,
- (char *)(unsigned long)io.write),
- space(MINSPACE));
+ fprintf(fp, "%s%s%s %s%s%s%s %s%5d%s%s%s%s%s",
+ mkstring(buf0, 5, RJUST|INT_DEC, (char *)(unsigned long)major),
+ space(MINSPACE),
+ mkstring(buf1, VADDR_PRLEN, LJUST|LONG_HEX, (char *)gendisk),
+ space(MINSPACE),
+ mkstring(buf2, 10, LJUST, disk_name),
+ space(MINSPACE),
+ mkstring(buf3, VADDR_PRLEN <= 11 ? 11 : VADDR_PRLEN,
+ LJUST|LONG_HEX, (char *)queue_addr),
+ space(MINSPACE),
+ io.read + io.write,
+ space(MINSPACE),
+ mkstring(buf4, 5, RJUST|INT_DEC,
+ (char *)(unsigned long)io.read),
+ space(MINSPACE),
+ mkstring(buf5, 5, RJUST|INT_DEC,
+ (char *)(unsigned long)io.write),
+ space(MINSPACE));
if (VALID_MEMBER(request_queue_in_flight)) {
if (!use_mq_interface(queue_addr)) {
@@ -4597,6 +4737,9 @@ void diskio_init(void)
MEMBER_OFFSET_INIT(kobject_entry, "kobject", "entry");
MEMBER_OFFSET_INIT(kset_list, "kset", "list");
MEMBER_OFFSET_INIT(request_list_count, "request_list", "count");
+ MEMBER_OFFSET_INIT(request_cmd_flags, "request", "cmd_flags");
+ MEMBER_OFFSET_INIT(request_q, "request", "q");
+ MEMBER_OFFSET_INIT(request_state, "request", "state");
MEMBER_OFFSET_INIT(request_queue_in_flight, "request_queue",
"in_flight");
if (MEMBER_EXISTS("request_queue", "rq"))
@@ -4608,10 +4751,33 @@ void diskio_init(void)
"mq_ops");
ANON_MEMBER_OFFSET_INIT(request_queue_queue_ctx,
"request_queue", "queue_ctx");
+ MEMBER_OFFSET_INIT(request_queue_queue_hw_ctx,
+ "request_queue", "queue_hw_ctx");
+ MEMBER_OFFSET_INIT(request_queue_nr_hw_queues,
+ "request_queue", "nr_hw_queues");
MEMBER_OFFSET_INIT(blk_mq_ctx_rq_dispatched, "blk_mq_ctx",
"rq_dispatched");
MEMBER_OFFSET_INIT(blk_mq_ctx_rq_completed, "blk_mq_ctx",
"rq_completed");
+ MEMBER_OFFSET_INIT(blk_mq_hw_ctx_tags, "blk_mq_hw_ctx", "tags");
+ MEMBER_OFFSET_INIT(blk_mq_tags_bitmap_tags, "blk_mq_tags",
+ "bitmap_tags");
+ MEMBER_OFFSET_INIT(blk_mq_tags_breserved_tags, "blk_mq_tags",
+ "breserved_tags");
+ MEMBER_OFFSET_INIT(blk_mq_tags_nr_reserved_tags, "blk_mq_tags",
+ "nr_reserved_tags");
+ MEMBER_OFFSET_INIT(blk_mq_tags_rqs, "blk_mq_tags", "rqs");
+ STRUCT_SIZE_INIT(blk_mq_tags, "blk_mq_tags");
+ STRUCT_SIZE_INIT(sbitmap, "sbitmap");
+ STRUCT_SIZE_INIT(sbitmap_word, "sbitmap_word");
+ MEMBER_OFFSET_INIT(sbitmap_word_word, "sbitmap_word", "word");
+ MEMBER_OFFSET_INIT(sbitmap_word_cleared, "sbitmap_word", "cleared");
+ MEMBER_OFFSET_INIT(sbitmap_depth, "sbitmap", "depth");
+ MEMBER_OFFSET_INIT(sbitmap_shift, "sbitmap", "shift");
+ MEMBER_OFFSET_INIT(sbitmap_map_nr, "sbitmap", "map_nr");
+ MEMBER_OFFSET_INIT(sbitmap_map, "sbitmap", "map");
+ MEMBER_OFFSET_INIT(sbitmap_queue_sb, "sbitmap_queue", "sb");
+
}
MEMBER_OFFSET_INIT(subsys_private_klist_devices, "subsys_private",
"klist_devices");
diff --git a/symbols.c b/symbols.c
index 5d12a021c769..c1f09556d710 100644
--- a/symbols.c
+++ b/symbols.c
@@ -10385,6 +10385,12 @@ dump_offset_table(char *spec, ulong makestruct)
OFFSET(kset_list));
fprintf(fp, " request_list_count: %ld\n",
OFFSET(request_list_count));
+ fprintf(fp, " request_cmd_flags: %ld\n",
+ OFFSET(request_cmd_flags));
+ fprintf(fp, " request_q: %ld\n",
+ OFFSET(request_q));
+ fprintf(fp, " request_state: %ld\n",
+ OFFSET(request_state));
fprintf(fp, " request_queue_in_flight: %ld\n",
OFFSET(request_queue_in_flight));
fprintf(fp, " request_queue_rq: %ld\n",
@@ -10393,10 +10399,25 @@ dump_offset_table(char *spec, ulong makestruct)
OFFSET(request_queue_mq_ops));
fprintf(fp, " request_queue_queue_ctx: %ld\n",
OFFSET(request_queue_queue_ctx));
+ fprintf(fp, " request_queue_queue_hw_ctx: %ld\n",
+ OFFSET(request_queue_queue_hw_ctx));
+ fprintf(fp, " request_queue_nr_hw_queues: %ld\n",
+ OFFSET(request_queue_nr_hw_queues));
fprintf(fp, " blk_mq_ctx_rq_dispatched: %ld\n",
OFFSET(blk_mq_ctx_rq_dispatched));
fprintf(fp, " blk_mq_ctx_rq_completed: %ld\n",
OFFSET(blk_mq_ctx_rq_completed));
+ fprintf(fp, " blk_mq_hw_ctx_tags: %ld\n",
+ OFFSET(blk_mq_hw_ctx_tags));
+ fprintf(fp, " blk_mq_tags_bitmap_tags: %ld\n",
+ OFFSET(blk_mq_tags_bitmap_tags));
+ fprintf(fp, " blk_mq_tags_breserved_tags: %ld\n",
+ OFFSET(blk_mq_tags_breserved_tags));
+ fprintf(fp, " blk_mq_tags_nr_reserved_tags: %ld\n",
+ OFFSET(blk_mq_tags_nr_reserved_tags));
+ fprintf(fp, " blk_mq_tags_rqs: %ld\n",
+ OFFSET(blk_mq_tags_rqs));
+
fprintf(fp, " subsys_private_klist_devices: %ld\n",
OFFSET(subsys_private_klist_devices));
fprintf(fp, " subsystem_kset: %ld\n",
@@ -11003,6 +11024,7 @@ dump_offset_table(char *spec, ulong makestruct)
fprintf(fp, " sbitmap: %ld\n", SIZE(sbitmap));
fprintf(fp, " sbitmap_queue: %ld\n", SIZE(sbitmap_queue));
fprintf(fp, " sbq_wait_state: %ld\n", SIZE(sbq_wait_state));
+ fprintf(fp, " blk_mq_tags: %ld\n", SIZE(blk_mq_tags));
fprintf(fp, "\n array_table:\n");
/*
--
2.30.2

389
0010-ppc64-use-a-variable-for-machdep-machspec.patch
View File

@ -1,389 +0,0 @@
From 93b880217de239268315be942c10dfce5649db8b Mon Sep 17 00:00:00 2001
From: Hari Bathini <hbathini@linux.ibm.com>
Date: Mon, 4 Jul 2022 10:55:46 +0530
Subject: [PATCH 10/29] ppc64: use a variable for machdep->machspec
machdpep->machspec is referred to multiple times. The compiler would
likely optimize this but nonetheless, use a variable to optimize in
coding and also improve readability. No functional change.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
ppc64.c | 224 ++++++++++++++++++++++++++++----------------------------
1 file changed, 111 insertions(+), 113 deletions(-)
diff --git a/ppc64.c b/ppc64.c
index ad1d6e121e81..4ea1f7c0c6f8 100644
--- a/ppc64.c
+++ b/ppc64.c
@@ -307,6 +307,8 @@ struct machine_specific book3e_machine_specific = {
void
ppc64_init(int when)
{
+ struct machine_specific *ms;
+
#if defined(__x86_64__)
if (ACTIVE())
error(FATAL, "compiled for the PPC64 architecture\n");
@@ -416,16 +418,16 @@ ppc64_init(int when)
break;
case POST_GDB:
- if (!(machdep->flags & BOOK3E)) {
- struct machine_specific *m = machdep->machspec;
+ ms = machdep->machspec;
+ if (!(machdep->flags & BOOK3E)) {
/*
* To determine if the kernel was running on OPAL based platform,
* use struct opal, which is populated with relevant values.
*/
if (symbol_exists("opal")) {
- get_symbol_data("opal", sizeof(struct ppc64_opal), &(m->opal));
- if (m->opal.base == SKIBOOT_BASE)
+ get_symbol_data("opal", sizeof(struct ppc64_opal), &(ms->opal));
+ if (ms->opal.base == SKIBOOT_BASE)
machdep->flags |= OPAL_FW;
}
@@ -453,18 +455,18 @@ ppc64_init(int when)
* _PAGE_WRITETHRU can be used to infer it.
*/
if (THIS_KERNEL_VERSION >= LINUX(3,14,0))
- m->_page_coherent = 0x0UL;
+ ms->_page_coherent = 0x0UL;
/*
* In kernel v4.5, _PAGE_PTE bit is introduced to
* distinguish PTEs from pointers.
*/
if (THIS_KERNEL_VERSION >= LINUX(4,5,0)) {
- m->_page_pte = 0x1UL;
- m->_page_present = 0x2UL;
- m->_page_user = 0x4UL;
- m->_page_rw = 0x8UL;
- m->_page_guarded = 0x10UL;
+ ms->_page_pte = 0x1UL;
+ ms->_page_present = 0x2UL;
+ ms->_page_user = 0x4UL;
+ ms->_page_rw = 0x8UL;
+ ms->_page_guarded = 0x10UL;
}
/*
@@ -474,8 +476,8 @@ ppc64_init(int when)
* Also, page table entries store physical addresses.
*/
if (THIS_KERNEL_VERSION >= LINUX(4,6,0)) {
- m->_page_pte = 0x1UL << 62;
- m->_page_present = 0x1UL << 63;
+ ms->_page_pte = 0x1UL << 62;
+ ms->_page_present = 0x1UL << 63;
machdep->flags |= PHYS_ENTRY_L4;
}
@@ -504,118 +506,117 @@ ppc64_init(int when)
machdep->ptrs_per_pgd = PTRS_PER_PGD;
} else {
/* 2.6.14 layout */
- struct machine_specific *m = machdep->machspec;
if (machdep->pagesize == 65536) {
/* 64K pagesize */
if (machdep->flags & RADIX_MMU) {
- m->l1_index_size = PTE_INDEX_SIZE_RADIX_64K;
- m->l2_index_size = PMD_INDEX_SIZE_RADIX_64K;
- m->l3_index_size = PUD_INDEX_SIZE_RADIX_64K;
- m->l4_index_size = PGD_INDEX_SIZE_RADIX_64K;
+ ms->l1_index_size = PTE_INDEX_SIZE_RADIX_64K;
+ ms->l2_index_size = PMD_INDEX_SIZE_RADIX_64K;
+ ms->l3_index_size = PUD_INDEX_SIZE_RADIX_64K;
+ ms->l4_index_size = PGD_INDEX_SIZE_RADIX_64K;
} else if (!(machdep->flags & BOOK3E) &&
(THIS_KERNEL_VERSION >= LINUX(4,6,0))) {
- m->l1_index_size = PTE_INDEX_SIZE_L4_64K_3_10;
+ ms->l1_index_size = PTE_INDEX_SIZE_L4_64K_3_10;
if (THIS_KERNEL_VERSION >= LINUX(4,12,0)) {
- m->l2_index_size = PMD_INDEX_SIZE_L4_64K_4_12;
+ ms->l2_index_size = PMD_INDEX_SIZE_L4_64K_4_12;
if (THIS_KERNEL_VERSION >= LINUX(4,17,0))
- m->l3_index_size = PUD_INDEX_SIZE_L4_64K_4_17;
+ ms->l3_index_size = PUD_INDEX_SIZE_L4_64K_4_17;
else
- m->l3_index_size = PUD_INDEX_SIZE_L4_64K_4_12;
- m->l4_index_size = PGD_INDEX_SIZE_L4_64K_4_12;
+ ms->l3_index_size = PUD_INDEX_SIZE_L4_64K_4_12;
+ ms->l4_index_size = PGD_INDEX_SIZE_L4_64K_4_12;
} else {
- m->l2_index_size = PMD_INDEX_SIZE_L4_64K_4_6;
- m->l3_index_size = PUD_INDEX_SIZE_L4_64K_4_6;
- m->l4_index_size = PGD_INDEX_SIZE_L4_64K_3_10;
+ ms->l2_index_size = PMD_INDEX_SIZE_L4_64K_4_6;
+ ms->l3_index_size = PUD_INDEX_SIZE_L4_64K_4_6;
+ ms->l4_index_size = PGD_INDEX_SIZE_L4_64K_3_10;
}
} else if (THIS_KERNEL_VERSION >= LINUX(3,10,0)) {
- m->l1_index_size = PTE_INDEX_SIZE_L4_64K_3_10;
- m->l2_index_size = PMD_INDEX_SIZE_L4_64K_3_10;
- m->l3_index_size = PUD_INDEX_SIZE_L4_64K;
- m->l4_index_size = PGD_INDEX_SIZE_L4_64K_3_10;
+ ms->l1_index_size = PTE_INDEX_SIZE_L4_64K_3_10;
+ ms->l2_index_size = PMD_INDEX_SIZE_L4_64K_3_10;
+ ms->l3_index_size = PUD_INDEX_SIZE_L4_64K;
+ ms->l4_index_size = PGD_INDEX_SIZE_L4_64K_3_10;
} else {
- m->l1_index_size = PTE_INDEX_SIZE_L4_64K;
- m->l2_index_size = PMD_INDEX_SIZE_L4_64K;
- m->l3_index_size = PUD_INDEX_SIZE_L4_64K;
- m->l4_index_size = PGD_INDEX_SIZE_L4_64K;
+ ms->l1_index_size = PTE_INDEX_SIZE_L4_64K;
+ ms->l2_index_size = PMD_INDEX_SIZE_L4_64K;
+ ms->l3_index_size = PUD_INDEX_SIZE_L4_64K;
+ ms->l4_index_size = PGD_INDEX_SIZE_L4_64K;
}
if (!(machdep->flags & BOOK3E))
- m->pte_rpn_shift = symbol_exists("demote_segment_4k") ?
+ ms->pte_rpn_shift = symbol_exists("demote_segment_4k") ?
PTE_RPN_SHIFT_L4_64K_V2 : PTE_RPN_SHIFT_L4_64K_V1;
if (!(machdep->flags & BOOK3E) &&
(THIS_KERNEL_VERSION >= LINUX(4,6,0))) {
- m->pgd_masked_bits = PGD_MASKED_BITS_64K_4_6;
- m->pud_masked_bits = PUD_MASKED_BITS_64K_4_6;
- m->pmd_masked_bits = PMD_MASKED_BITS_64K_4_6;
+ ms->pgd_masked_bits = PGD_MASKED_BITS_64K_4_6;
+ ms->pud_masked_bits = PUD_MASKED_BITS_64K_4_6;
+ ms->pmd_masked_bits = PMD_MASKED_BITS_64K_4_6;
} else {
- m->pgd_masked_bits = PGD_MASKED_BITS_64K;
- m->pud_masked_bits = PUD_MASKED_BITS_64K;
+ ms->pgd_masked_bits = PGD_MASKED_BITS_64K;
+ ms->pud_masked_bits = PUD_MASKED_BITS_64K;
if ((machdep->flags & BOOK3E) &&
(THIS_KERNEL_VERSION >= LINUX(4,5,0)))
- m->pmd_masked_bits = PMD_MASKED_BITS_BOOK3E_64K_4_5;
+ ms->pmd_masked_bits = PMD_MASKED_BITS_BOOK3E_64K_4_5;
else if (THIS_KERNEL_VERSION >= LINUX(3,11,0))
- m->pmd_masked_bits = PMD_MASKED_BITS_64K_3_11;
+ ms->pmd_masked_bits = PMD_MASKED_BITS_64K_3_11;
else
- m->pmd_masked_bits = PMD_MASKED_BITS_64K;
+ ms->pmd_masked_bits = PMD_MASKED_BITS_64K;
}
} else {
/* 4K pagesize */
if (machdep->flags & RADIX_MMU) {
- m->l1_index_size = PTE_INDEX_SIZE_RADIX_4K;
- m->l2_index_size = PMD_INDEX_SIZE_RADIX_4K;
- m->l3_index_size = PUD_INDEX_SIZE_RADIX_4K;
- m->l4_index_size = PGD_INDEX_SIZE_RADIX_4K;
+ ms->l1_index_size = PTE_INDEX_SIZE_RADIX_4K;
+ ms->l2_index_size = PMD_INDEX_SIZE_RADIX_4K;
+ ms->l3_index_size = PUD_INDEX_SIZE_RADIX_4K;
+ ms->l4_index_size = PGD_INDEX_SIZE_RADIX_4K;
} else {
- m->l1_index_size = PTE_INDEX_SIZE_L4_4K;
- m->l2_index_size = PMD_INDEX_SIZE_L4_4K;
+ ms->l1_index_size = PTE_INDEX_SIZE_L4_4K;
+ ms->l2_index_size = PMD_INDEX_SIZE_L4_4K;
if (THIS_KERNEL_VERSION >= LINUX(3,7,0))
- m->l3_index_size = PUD_INDEX_SIZE_L4_4K_3_7;
+ ms->l3_index_size = PUD_INDEX_SIZE_L4_4K_3_7;
else
- m->l3_index_size = PUD_INDEX_SIZE_L4_4K;
- m->l4_index_size = PGD_INDEX_SIZE_L4_4K;
+ ms->l3_index_size = PUD_INDEX_SIZE_L4_4K;
+ ms->l4_index_size = PGD_INDEX_SIZE_L4_4K;
if (machdep->flags & BOOK3E)
- m->pte_rpn_shift = PTE_RPN_SHIFT_L4_BOOK3E_4K;
+ ms->pte_rpn_shift = PTE_RPN_SHIFT_L4_BOOK3E_4K;
else
- m->pte_rpn_shift = THIS_KERNEL_VERSION >= LINUX(4,5,0) ?
+ ms->pte_rpn_shift = THIS_KERNEL_VERSION >= LINUX(4,5,0) ?
PTE_RPN_SHIFT_L4_4K_4_5 : PTE_RPN_SHIFT_L4_4K;
}
- m->pgd_masked_bits = PGD_MASKED_BITS_4K;
- m->pud_masked_bits = PUD_MASKED_BITS_4K;
- m->pmd_masked_bits = PMD_MASKED_BITS_4K;
+ ms->pgd_masked_bits = PGD_MASKED_BITS_4K;
+ ms->pud_masked_bits = PUD_MASKED_BITS_4K;
+ ms->pmd_masked_bits = PMD_MASKED_BITS_4K;
}
- m->pte_rpn_mask = PTE_RPN_MASK_DEFAULT;
+ ms->pte_rpn_mask = PTE_RPN_MASK_DEFAULT;
if (!(machdep->flags & BOOK3E)) {
if (THIS_KERNEL_VERSION >= LINUX(4,6,0)) {
- m->pte_rpn_mask = PTE_RPN_MASK_L4_4_6;
- m->pte_rpn_shift = PTE_RPN_SHIFT_L4_4_6;
+ ms->pte_rpn_mask = PTE_RPN_MASK_L4_4_6;
+ ms->pte_rpn_shift = PTE_RPN_SHIFT_L4_4_6;
}
if (THIS_KERNEL_VERSION >= LINUX(4,7,0)) {
- m->pgd_masked_bits = PGD_MASKED_BITS_4_7;
- m->pud_masked_bits = PUD_MASKED_BITS_4_7;
- m->pmd_masked_bits = PMD_MASKED_BITS_4_7;
+ ms->pgd_masked_bits = PGD_MASKED_BITS_4_7;
+ ms->pud_masked_bits = PUD_MASKED_BITS_4_7;
+ ms->pmd_masked_bits = PMD_MASKED_BITS_4_7;
}
}
/* Compute ptrs per each level */
- m->l1_shift = machdep->pageshift;
- m->ptrs_per_l1 = (1 << m->l1_index_size);
- m->ptrs_per_l2 = (1 << m->l2_index_size);
- m->ptrs_per_l3 = (1 << m->l3_index_size);
- m->ptrs_per_l4 = (1 << m->l4_index_size);
- machdep->ptrs_per_pgd = m->ptrs_per_l4;
+ ms->l1_shift = machdep->pageshift;
+ ms->ptrs_per_l1 = (1 << ms->l1_index_size);
+ ms->ptrs_per_l2 = (1 << ms->l2_index_size);
+ ms->ptrs_per_l3 = (1 << ms->l3_index_size);
+ ms->ptrs_per_l4 = (1 << ms->l4_index_size);
+ machdep->ptrs_per_pgd = ms->ptrs_per_l4;
/* Compute shifts */
- m->l2_shift = m->l1_shift + m->l1_index_size;
- m->l3_shift = m->l2_shift + m->l2_index_size;
- m->l4_shift = m->l3_shift + m->l3_index_size;
+ ms->l2_shift = ms->l1_shift + ms->l1_index_size;
+ ms->l3_shift = ms->l2_shift + ms->l2_index_size;
+ ms->l4_shift = ms->l3_shift + ms->l3_index_size;
}
if (machdep->flags & VMEMMAP)
@@ -681,19 +682,15 @@ ppc64_init(int when)
*/
offset = MEMBER_OFFSET("paca_struct", "xHrdIntStack");
paca_sym = symbol_value("paca");
- if (!(machdep->machspec->hwintrstack =
- (ulong *)calloc(NR_CPUS, sizeof(ulong))))
+ if (!(ms->hwintrstack = (ulong *)calloc(NR_CPUS, sizeof(ulong))))
error(FATAL, "cannot malloc hwintrstack space.");
for (cpu = 0; cpu < kt->cpus; cpu++) {
- readmem(paca_sym + (paca_size * cpu) + offset,
- KVADDR,
- &machdep->machspec->hwintrstack[cpu],
- sizeof(ulong), "PPC64 HW_intr_stack",
- FAULT_ON_ERROR);
+ readmem(paca_sym + (paca_size * cpu) + offset, KVADDR,
+ &ms->hwintrstack[cpu], sizeof(ulong),
+ "PPC64 HW_intr_stack", FAULT_ON_ERROR);
}
- machdep->machspec->hwstacksize = 8 * machdep->pagesize;
- if ((machdep->machspec->hwstackbuf = (char *)
- malloc(machdep->machspec->hwstacksize)) == NULL)
+ ms->hwstacksize = 8 * machdep->pagesize;
+ if ((ms->hwstackbuf = (char *)malloc(ms->hwstacksize)) == NULL)
error(FATAL, "cannot malloc hwirqstack buffer space.");
}
@@ -756,6 +753,7 @@ ppc64_get_stacktop(ulong task)
void
ppc64_dump_machdep_table(ulong arg)
{
+ struct machine_specific *ms = machdep->machspec;
int i, c, others;
others = 0;
@@ -844,57 +842,57 @@ ppc64_dump_machdep_table(ulong arg)
i, machdep->cmdline_args[i] ?
machdep->cmdline_args[i] : "(unused)");
}
- fprintf(fp, " machspec: %lx\n", (ulong)machdep->machspec);
+ fprintf(fp, " machspec: %lx\n", (ulong)ms);
fprintf(fp, " is_kvaddr: %s\n",
- machdep->machspec->is_kvaddr == book3e_is_kvaddr ?
+ ms->is_kvaddr == book3e_is_kvaddr ?
"book3e_is_kvaddr()" : "generic_is_kvaddr()");
fprintf(fp, " is_vmaddr: %s\n",
- machdep->machspec->is_vmaddr == book3e_is_vmaddr ?
+ ms->is_vmaddr == book3e_is_vmaddr ?
"book3e_is_vmaddr()" : "ppc64_is_vmaddr()");
- if (machdep->machspec->hwintrstack) {
+ if (ms->hwintrstack) {
fprintf(fp, " hwintrstack[%d]: ", NR_CPUS);
for (c = 0; c < NR_CPUS; c++) {
fprintf(fp, "%s%016lx ",
((c % 4) == 0) ? "\n " : "",
- machdep->machspec->hwintrstack[c]);
+ ms->hwintrstack[c]);
}
} else
fprintf(fp, " hwintrstack: (unused)");
fprintf(fp, "\n");
- fprintf(fp, " hwstackbuf: %lx\n", (ulong)machdep->machspec->hwstackbuf);
- fprintf(fp, " hwstacksize: %d\n", machdep->machspec->hwstacksize);
- fprintf(fp, " l4_index_size: %d\n", machdep->machspec->l4_index_size);
- fprintf(fp, " l3_index_size: %d\n", machdep->machspec->l3_index_size);
- fprintf(fp, " l2_index_size: %d\n", machdep->machspec->l2_index_size);
- fprintf(fp, " l1_index_size: %d\n", machdep->machspec->l1_index_size);
- fprintf(fp, " ptrs_per_l4: %d\n", machdep->machspec->ptrs_per_l4);
- fprintf(fp, " ptrs_per_l3: %d\n", machdep->machspec->ptrs_per_l3);
- fprintf(fp, " ptrs_per_l2: %d\n", machdep->machspec->ptrs_per_l2);
- fprintf(fp, " ptrs_per_l1: %d\n", machdep->machspec->ptrs_per_l1);
- fprintf(fp, " l4_shift: %d\n", machdep->machspec->l4_shift);
- fprintf(fp, " l3_shift: %d\n", machdep->machspec->l3_shift);
- fprintf(fp, " l2_shift: %d\n", machdep->machspec->l2_shift);
- fprintf(fp, " l1_shift: %d\n", machdep->machspec->l1_shift);
- fprintf(fp, " pte_rpn_mask: %lx\n", machdep->machspec->pte_rpn_mask);
- fprintf(fp, " pte_rpn_shift: %d\n", machdep->machspec->pte_rpn_shift);
- fprintf(fp, " pgd_masked_bits: %lx\n", machdep->machspec->pgd_masked_bits);
- fprintf(fp, " pud_masked_bits: %lx\n", machdep->machspec->pud_masked_bits);
- fprintf(fp, " pmd_masked_bits: %lx\n", machdep->machspec->pmd_masked_bits);
+ fprintf(fp, " hwstackbuf: %lx\n", (ulong)ms->hwstackbuf);
+ fprintf(fp, " hwstacksize: %d\n", ms->hwstacksize);
+ fprintf(fp, " l4_index_size: %d\n", ms->l4_index_size);
+ fprintf(fp, " l3_index_size: %d\n", ms->l3_index_size);
+ fprintf(fp, " l2_index_size: %d\n", ms->l2_index_size);
+ fprintf(fp, " l1_index_size: %d\n", ms->l1_index_size);
+ fprintf(fp, " ptrs_per_l4: %d\n", ms->ptrs_per_l4);
+ fprintf(fp, " ptrs_per_l3: %d\n", ms->ptrs_per_l3);
+ fprintf(fp, " ptrs_per_l2: %d\n", ms->ptrs_per_l2);
+ fprintf(fp, " ptrs_per_l1: %d\n", ms->ptrs_per_l1);
+ fprintf(fp, " l4_shift: %d\n", ms->l4_shift);
+ fprintf(fp, " l3_shift: %d\n", ms->l3_shift);
+ fprintf(fp, " l2_shift: %d\n", ms->l2_shift);
+ fprintf(fp, " l1_shift: %d\n", ms->l1_shift);
+ fprintf(fp, " pte_rpn_mask: %lx\n", ms->pte_rpn_mask);
+ fprintf(fp, " pte_rpn_shift: %d\n", ms->pte_rpn_shift);
+ fprintf(fp, " pgd_masked_bits: %lx\n", ms->pgd_masked_bits);
+ fprintf(fp, " pud_masked_bits: %lx\n", ms->pud_masked_bits);
+ fprintf(fp, " pmd_masked_bits: %lx\n", ms->pmd_masked_bits);
fprintf(fp, " vmemmap_base: ");
- if (machdep->machspec->vmemmap_base)
- fprintf(fp, "%lx\n", machdep->machspec->vmemmap_base);
+ if (ms->vmemmap_base)
+ fprintf(fp, "%lx\n", ms->vmemmap_base);
else
fprintf(fp, "(unused)\n");
- if (machdep->machspec->vmemmap_cnt) {
+ if (ms->vmemmap_cnt) {
fprintf(fp, " vmemmap_cnt: %d\n",
- machdep->machspec->vmemmap_cnt);
+ ms->vmemmap_cnt);
fprintf(fp, " vmemmap_psize: %d\n",
- machdep->machspec->vmemmap_psize);
- for (i = 0; i < machdep->machspec->vmemmap_cnt; i++) {
+ ms->vmemmap_psize);
+ for (i = 0; i < ms->vmemmap_cnt; i++) {
fprintf(fp,
" vmemmap_list[%d]: virt: %lx phys: %lx\n", i,
- machdep->machspec->vmemmap_list[i].virt,
- machdep->machspec->vmemmap_list[i].phys);
+ ms->vmemmap_list[i].virt,
+ ms->vmemmap_list[i].phys);
}
} else {
fprintf(fp, " vmemmap_cnt: (unused)\n");
--
2.37.1

121
0011-Fix-for-dev-d-D-options-to-support-blk-mq-change-on-.patch
View File

@ -1,121 +0,0 @@
From 68ce0b9a35d77d767872dd1a729c50e4695a30a8 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Thu, 2 Jun 2022 20:12:56 +0800
Subject: [PATCH 11/16] Fix for "dev -d|-D" options to support blk-mq change on
Linux v5.18-rc1
Kernel commit 4e5cc99e1e48 ("blk-mq: manage hctx map via xarray") removed
the "queue_hw_ctx" member from struct request_queue at Linux v5.18-rc1,
and replaced it with a struct xarray "hctx_table". Without the patch, the
"dev -d|-D" options will print an error:
crash> dev -d
MAJOR GENDISK NAME REQUEST_QUEUE TOTAL READ WRITE
dev: invalid structure member offset: request_queue_queue_hw_ctx
With the patch:
crash> dev -d
MAJOR GENDISK NAME REQUEST_QUEUE TOTAL READ WRITE
8 ffff8e99d0a1ae00 sda ffff8e9c14c59980 10 6 4
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 1 +
dev.c | 42 +++++++++++++++++++++++++++++++++---------
symbols.c | 2 ++
3 files changed, 36 insertions(+), 9 deletions(-)
diff --git a/defs.h b/defs.h
index 2681586a33dc..7d3b73422f48 100644
--- a/defs.h
+++ b/defs.h
@@ -2180,6 +2180,7 @@ struct offset_table { /* stash of commonly-used offsets */
long blk_mq_tags_breserved_tags;
long blk_mq_tags_nr_reserved_tags;
long blk_mq_tags_rqs;
+ long request_queue_hctx_table;
};
struct size_table { /* stash of commonly-used sizes */
diff --git a/dev.c b/dev.c
index 4be4c96df8b0..0172c83ffaea 100644
--- a/dev.c
+++ b/dev.c
@@ -4369,20 +4369,42 @@ static void get_mq_diskio_from_hw_queues(ulong q, struct diskio *dio)
uint cnt = 0;
ulong addr = 0, hctx_addr = 0;
ulong *hctx_array = NULL;
+ struct list_pair *lp = NULL;
+
+ if (VALID_MEMBER(request_queue_hctx_table)) {
+ addr = q + OFFSET(request_queue_hctx_table);
+ cnt = do_xarray(addr, XARRAY_COUNT, NULL);
+ lp = (struct list_pair *)GETBUF(sizeof(struct list_pair) * (cnt + 1));
+ if (!lp)
+ error(FATAL, "fail to get memory for list_pair.\n");
+ lp[0].index = cnt;
+ cnt = do_xarray(addr, XARRAY_GATHER, lp);
+ } else {
+ addr = q + OFFSET(request_queue_nr_hw_queues);
+ readmem(addr, KVADDR, &cnt, sizeof(uint),
+ "request_queue.nr_hw_queues", FAULT_ON_ERROR);
- addr = q + OFFSET(request_queue_nr_hw_queues);
- readmem(addr, KVADDR, &cnt, sizeof(uint),
- "request_queue.nr_hw_queues", FAULT_ON_ERROR);
-
- addr = q + OFFSET(request_queue_queue_hw_ctx);
- readmem(addr, KVADDR, &hctx_addr, sizeof(void *),
- "request_queue.queue_hw_ctx", FAULT_ON_ERROR);
+ addr = q + OFFSET(request_queue_queue_hw_ctx);
+ readmem(addr, KVADDR, &hctx_addr, sizeof(void *),
+ "request_queue.queue_hw_ctx", FAULT_ON_ERROR);
+ }
hctx_array = (ulong *)GETBUF(sizeof(void *) * cnt);
- if (!hctx_array)
+ if (!hctx_array) {
+ if (lp)
+ FREEBUF(lp);
error(FATAL, "fail to get memory for the hctx_array\n");
+ }
+
+ if (lp && hctx_array) {
+ uint i;
+
+ /* copy it from list_pair to hctx_array */
+ for (i = 0; i < cnt; i++)
+ hctx_array[i] = (ulong)lp[i].value;
- if (!readmem(hctx_addr, KVADDR, hctx_array, sizeof(void *) * cnt,
+ FREEBUF(lp);
+ } else if (!readmem(hctx_addr, KVADDR, hctx_array, sizeof(void *) * cnt,
"request_queue.queue_hw_ctx[]", RETURN_ON_ERROR)) {
FREEBUF(hctx_array);
return;
@@ -4755,6 +4777,8 @@ void diskio_init(void)
"request_queue", "queue_hw_ctx");
MEMBER_OFFSET_INIT(request_queue_nr_hw_queues,
"request_queue", "nr_hw_queues");
+ MEMBER_OFFSET_INIT(request_queue_hctx_table,
+ "request_queue", "hctx_table");
MEMBER_OFFSET_INIT(blk_mq_ctx_rq_dispatched, "blk_mq_ctx",
"rq_dispatched");
MEMBER_OFFSET_INIT(blk_mq_ctx_rq_completed, "blk_mq_ctx",
diff --git a/symbols.c b/symbols.c
index c1f09556d710..bee1faf92c83 100644
--- a/symbols.c
+++ b/symbols.c
@@ -10403,6 +10403,8 @@ dump_offset_table(char *spec, ulong makestruct)
OFFSET(request_queue_queue_hw_ctx));
fprintf(fp, " request_queue_nr_hw_queues: %ld\n",
OFFSET(request_queue_nr_hw_queues));
+ fprintf(fp, " request_queue_hctx_table: %ld\n",
+ OFFSET(request_queue_hctx_table));
fprintf(fp, " blk_mq_ctx_rq_dispatched: %ld\n",
OFFSET(blk_mq_ctx_rq_dispatched));
fprintf(fp, " blk_mq_ctx_rq_completed: %ld\n",
--
2.30.2

39
0011-arm64-Fix-for-st-_stext_vmlinux-not-initialized-when.patch
View File

@ -1,39 +0,0 @@
From 6722ea102264b54529afc19d347a3a7473670fdd Mon Sep 17 00:00:00 2001
From: Qianli Zhao <qianli.zhao@horizon.ai>
Date: Mon, 4 Jul 2022 16:40:01 +0800
Subject: [PATCH 11/29] arm64: Fix for st->_stext_vmlinux not initialized when
set VA_BITS_ACTUAL
Setting st->_stext_vmlinux to UNINITIALIZED to search for "_stext"
from the vmlinux. In the scenario where kaslr is disabled and
without vmcoreinfo, crash will get the wrong MODULES/VMALLOC ranges
and cause a failure in parsing a raw RAM dumpfile.
Signed-off-by: Qianli Zhao <qianli.zhao@horizon.ai>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
arm64.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arm64.c b/arm64.c
index 0f615cf52bef..b6b7aa11f4fe 100644
--- a/arm64.c
+++ b/arm64.c
@@ -149,6 +149,14 @@ arm64_init(int when)
ms = machdep->machspec;
+ /*
+ * The st->_stext_vmlinux is needed in arm64_init(PRE_GDB) when a
+ * dumpfile does not have vmcoreinfo and we use -m vabits_actual
+ * option, e.g. a raw RAM dumpfile.
+ */
+ if (ms->VA_BITS_ACTUAL)
+ st->_stext_vmlinux = UNINITIALIZED;
+
if (!ms->kimage_voffset && STREQ(pc->live_memsrc, "/dev/crash"))
ioctl(pc->mfd, DEV_CRASH_ARCH_DATA, &ms->kimage_voffset);
--
2.37.1

43
0012-Doc-update-man-page-for-the-bpf-and-sbitmapq-command.patch
View File

@ -1,43 +0,0 @@
From c672d7a4c290712b32c54329cbdc1e74d122e813 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Mon, 6 Jun 2022 19:09:16 +0800
Subject: [PATCH 12/16] Doc: update man page for the "bpf" and "sbitmapq"
commands
The information of the "bpf" and "sbitmapq" commands is missing in the man
page of the crash utility. Let's add it to the man page.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
crash.8 | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/crash.8 b/crash.8
index 1f3657b11e4c..e553a0b4adb3 100644
--- a/crash.8
+++ b/crash.8
@@ -584,6 +584,9 @@ creates a single-word alias for a command.
.I ascii
displays an ascii chart or translates a numeric value into its ascii components.
.TP
+.I bpf
+provides information on currently-loaded eBPF programs and maps.
+.TP
.I bt
displays a task's kernel-stack backtrace. If it is given the
.I \-a
@@ -706,6 +709,11 @@ number of seconds between each command execution.
.I runq
displays the tasks on the run queue.
.TP
+.I sbitmapq
+dumps the contents of the sbitmap_queue structure and the used
+bits in the bitmap. Also, it shows the dump of a structure array
+associated with the sbitmap_queue.
+.TP
.I search
searches a range of user or kernel memory space for given value.
.TP
--
2.30.2

53
0012-Fix-gcc-11-compiler-warnings-on-filesys.c.patch
View File

@ -1,53 +0,0 @@
From f374aca364b7e8809f122678aefed1010e3c94bd Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 22 Jul 2022 13:44:50 +0900
Subject: [PATCH 12/29] Fix gcc-11 compiler warnings on filesys.c
Without the patch, the following gcc-11 compiler warnings are emitted
for filesys.c:
filesys.c: In function 'mount_point':
filesys.c:718:17: warning: 'pclose' called on pointer returned from a mismatched allocation function [-Wmismatched-dealloc]
718 | pclose(mp);
| ^~~~~~~~~~
filesys.c:709:27: note: returned from 'fopen'
709 | if ((mp = fopen(mntfile, "r")) == NULL)
| ^~~~~~~~~~~~~~~~~~~
filesys.c:738:17: warning: 'pclose' called on pointer returned from a mismatched allocation function [-Wmismatched-dealloc]
738 | pclose(mp);
| ^~~~~~~~~~
filesys.c:723:27: note: returned from 'fopen'
723 | if ((mp = fopen(mntfile, "r")) == NULL)
| ^~~~~~~~~~~~~~~~~~~
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
filesys.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/filesys.c b/filesys.c
index 43cbe826fc79..a863f04eb250 100644
--- a/filesys.c
+++ b/filesys.c
@@ -715,7 +715,7 @@ mount_point(char *name)
continue;
found++;
}
- pclose(mp);
+ fclose(mp);
if (!(mount_points = (char **)malloc(sizeof(char *) * found)))
return FALSE;
@@ -735,7 +735,7 @@ mount_point(char *name)
mount_points_gathered++, i++;
}
}
- pclose(mp);
+ fclose(mp);
if (CRASHDEBUG(2))
for (i = 0; i < mount_points_gathered; i++)
--
2.37.1

53
0013-Fix-gcc-11-compiler-warning-on-symbols.c.patch
View File

@ -1,53 +0,0 @@
From b9c0ed124e422b7e0b1526afa3a691ad0579607b Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 22 Jul 2022 13:44:50 +0900
Subject: [PATCH 13/29] Fix gcc-11 compiler warning on symbols.c
Without the patch, the following gcc-11 compiler warning is emitted for
symbols.c:
symbols.c: In function 'cmd_p':
symbols.c:7412:38: warning: writing 1 byte into a region of size 0 [-Wstringop-overflow=]
7412 | *(cpuspec-1) = ':';
| ~~~~~~~~~~~~~^~~~~
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
symbols.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/symbols.c b/symbols.c
index bee1faf92c83..42c4eb400044 100644
--- a/symbols.c
+++ b/symbols.c
@@ -7351,7 +7351,7 @@ cmd_p(void)
unsigned radix;
int do_load_module_filter;
char buf1[BUFSIZE];
- char *cpuspec;
+ char *cpuspec, *p;
do_load_module_filter = radix = 0;
@@ -7386,7 +7386,7 @@ cmd_p(void)
if (argerrs || !args[optind])
cmd_usage(pc->curcmd, SYNOPSIS);
- cpuspec = strrchr(args[optind], ':');
+ p = cpuspec = strrchr(args[optind], ':');
if (cpuspec)
*cpuspec++ = NULLCHAR;
@@ -7409,7 +7409,7 @@ cmd_p(void)
sp->name);
else
/* maybe a valid C expression (e.g. ':') */
- *(cpuspec-1) = ':';
+ *p = ':';
}
process_gdb_output(concat_args(buf1, 0, TRUE), radix,
--
2.37.1

48
0013-sbitmapq-Fix-for-sbitmap_queue-without-ws_active-mem.patch
View File

@ -1,48 +0,0 @@
From 9ce31a14d1083cbb2beb4a8e6eb7b88234b79a99 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 10 Jun 2022 11:49:47 +0900
Subject: [PATCH 13/16] sbitmapq: Fix for sbitmap_queue without ws_active
member
The sbitmap_queue.ws_active member was added by kernel commit 5d2ee7122c73
("sbitmap: optimize wakeup check") at Linux 5.0. Without the patch, on
earlier kernels the "sbitmapq" command fails with the following error:
crash> sbitmapq ffff8f1a3611cf10
sbitmapq: invalid structure member offset: sbitmap_queue_ws_active
FILE: sbitmap.c LINE: 393 FUNCTION: sbitmap_queue_context_load()
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
sbitmap.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/sbitmap.c b/sbitmap.c
index e8ebd62fe01c..152c28e6875f 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -325,7 +325,8 @@ static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
fprintf(fp, "wake_batch = %u\n", sqc->wake_batch);
fprintf(fp, "wake_index = %d\n", sqc->wake_index);
- fprintf(fp, "ws_active = %d\n", sqc->ws_active);
+ if (VALID_MEMBER(sbitmap_queue_ws_active)) /* 5.0 and later */
+ fprintf(fp, "ws_active = %d\n", sqc->ws_active);
sbq_wait_state_size = SIZE(sbq_wait_state);
wait_cnt_off = OFFSET(sbq_wait_state_wait_cnt);
@@ -380,7 +381,8 @@ static void sbitmap_queue_context_load(ulong addr, struct sbitmap_queue_context
sqc->wake_batch = UINT(sbitmap_queue_buf + OFFSET(sbitmap_queue_wake_batch));
sqc->wake_index = INT(sbitmap_queue_buf + OFFSET(sbitmap_queue_wake_index));
sqc->ws_addr = ULONG(sbitmap_queue_buf + OFFSET(sbitmap_queue_ws));
- sqc->ws_active = INT(sbitmap_queue_buf + OFFSET(sbitmap_queue_ws_active));
+ if (VALID_MEMBER(sbitmap_queue_ws_active))
+ sqc->ws_active = INT(sbitmap_queue_buf + OFFSET(sbitmap_queue_ws_active));
if (VALID_MEMBER(sbitmap_queue_round_robin))
sqc->round_robin = BOOL(sbitmap_queue_buf + OFFSET(sbitmap_queue_round_robin));
sqc->min_shallow_depth = UINT(sbitmap_queue_buf + OFFSET(sbitmap_queue_min_shallow_depth));
--
2.30.2

38
0014-Fix-gcc-11-compiler-warning-on-makedumpfile.c.patch
View File

@ -1,38 +0,0 @@
From 7591e3c07cef4900f6b0ca797270cb7527fb4e29 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 22 Jul 2022 13:44:50 +0900
Subject: [PATCH 14/29] Fix gcc-11 compiler warning on makedumpfile.c
Without the patch, the following gcc-11 compiler warning is emitted for
makedumpfile.c:
In function 'flattened_format_get_osrelease',
inlined from 'check_flattened_format' at makedumpfile.c:236:3:
makedumpfile.c:392:9: warning: 'fclose' called on pointer returned from a mismatched allocation function [-Wmismatched-dealloc]
392 | fclose(pipe);
| ^~~~~~~~~~~~
makedumpfile.c: In function 'check_flattened_format':
makedumpfile.c:380:21: note: returned from 'popen'
380 | if ((pipe = popen(buf, "r")) == NULL)
| ^~~~~~~~~~~~~~~
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
makedumpfile.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/makedumpfile.c b/makedumpfile.c
index ebf24f56da2c..26d12b638ecd 100644
--- a/makedumpfile.c
+++ b/makedumpfile.c
@@ -389,5 +389,5 @@ flattened_format_get_osrelease(char *file)
}
}
- fclose(pipe);
+ pclose(pipe);
}
--
2.37.1

110
0014-sbitmapq-Fix-for-sbitmap_word-without-cleared-member.patch
View File

@ -1,110 +0,0 @@
From 0d3e86fee5eead93b521a0e20a0e099ede4ab72b Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 10 Jun 2022 11:49:47 +0900
Subject: [PATCH 14/16] sbitmapq: Fix for sbitmap_word without cleared member
The sbitmap_word.cleared member was added by kernel commit ea86ea2cdced
("sbitmap: ammortize cost of clearing bits") at Linux 5.0. Without the
patch, on earlier kernels the "sbitmapq" command fails with the
following error:
crash> sbitmapq ffff8f1a3611cf10
sbitmapq: invalid structure member offset: sbitmap_word_cleared
FILE: sbitmap.c LINE: 92 FUNCTION: __sbitmap_weight()
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
sbitmap.c | 26 ++++++++++++++++++--------
1 file changed, 18 insertions(+), 8 deletions(-)
diff --git a/sbitmap.c b/sbitmap.c
index 152c28e6875f..c9f7209f9e3e 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -89,7 +89,6 @@ static unsigned int __sbitmap_weight(const struct sbitmap_context *sc, bool set)
{
const ulong sbitmap_word_size = SIZE(sbitmap_word);
const ulong w_word_off = OFFSET(sbitmap_word_word);
- const ulong w_cleared_off = OFFSET(sbitmap_word_cleared);
unsigned int weight = 0;
ulong addr = sc->map_addr;
@@ -111,7 +110,10 @@ static unsigned int __sbitmap_weight(const struct sbitmap_context *sc, bool set)
word = ULONG(sbitmap_word_buf + w_word_off);
weight += bitmap_weight(word, depth);
} else {
- cleared = ULONG(sbitmap_word_buf + w_cleared_off);
+ if (VALID_MEMBER(sbitmap_word_cleared))
+ cleared = ULONG(sbitmap_word_buf + OFFSET(sbitmap_word_cleared));
+ else
+ cleared = 0;
weight += bitmap_weight(cleared, depth);
}
@@ -130,7 +132,10 @@ static unsigned int sbitmap_weight(const struct sbitmap_context *sc)
static unsigned int sbitmap_cleared(const struct sbitmap_context *sc)
{
- return __sbitmap_weight(sc, false);
+ if (VALID_MEMBER(sbitmap_word_cleared)) /* 5.0 and later */
+ return __sbitmap_weight(sc, false);
+
+ return 0;
}
static void sbitmap_emit_byte(unsigned int offset, uint8_t byte)
@@ -149,7 +154,6 @@ static void sbitmap_bitmap_show(const struct sbitmap_context *sc)
{
const ulong sbitmap_word_size = SIZE(sbitmap_word);
const ulong w_word_off = OFFSET(sbitmap_word_word);
- const ulong w_cleared_off = OFFSET(sbitmap_word_cleared);
uint8_t byte = 0;
unsigned int byte_bits = 0;
@@ -169,7 +173,10 @@ static void sbitmap_bitmap_show(const struct sbitmap_context *sc)
}
word = ULONG(sbitmap_word_buf + w_word_off);
- cleared = ULONG(sbitmap_word_buf + w_cleared_off);
+ if (VALID_MEMBER(sbitmap_word_cleared))
+ cleared = ULONG(sbitmap_word_buf + OFFSET(sbitmap_word_cleared));
+ else
+ cleared = 0;
word_bits = __map_depth(sc, i);
word &= ~cleared;
@@ -219,7 +226,6 @@ static void __sbitmap_for_each_set(const struct sbitmap_context *sc,
{
const ulong sbitmap_word_size = SIZE(sbitmap_word);
const ulong w_word_off = OFFSET(sbitmap_word_word);
- const ulong w_cleared_off = OFFSET(sbitmap_word_cleared);
unsigned int index;
unsigned int nr;
@@ -245,7 +251,10 @@ static void __sbitmap_for_each_set(const struct sbitmap_context *sc,
}
w_word = ULONG(sbitmap_word_buf + w_word_off);
- w_cleared = ULONG(sbitmap_word_buf + w_cleared_off);
+ if (VALID_MEMBER(sbitmap_word_cleared))
+ w_cleared = ULONG(sbitmap_word_buf + OFFSET(sbitmap_word_cleared));
+ else
+ w_cleared = 0;
depth = min(__map_depth(sc, index) - nr, sc->depth - scanned);
@@ -297,7 +306,8 @@ static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
fprintf(fp, "depth = %u\n", sc->depth);
fprintf(fp, "busy = %u\n", sbitmap_weight(sc) - sbitmap_cleared(sc));
- fprintf(fp, "cleared = %u\n", sbitmap_cleared(sc));
+ if (VALID_MEMBER(sbitmap_word_cleared)) /* 5.0 and later */
+ fprintf(fp, "cleared = %u\n", sbitmap_cleared(sc));
fprintf(fp, "bits_per_word = %u\n", 1U << sc->shift);
fprintf(fp, "map_nr = %u\n", sc->map_nr);
--
2.30.2

61
0015-Fix-gcc-11-compiler-warning-on-kvmdump.c.patch
View File

@ -1,61 +0,0 @@
From f37df7df8a50519d80f04fb48499287892021575 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 22 Jul 2022 13:44:50 +0900
Subject: [PATCH 15/29] Fix gcc-11 compiler warning on kvmdump.c
Without the patch, the following gcc-11 compiler warning is emitted for
kvmdump.c:
In function 'write_mapfile_registers',
inlined from 'write_mapfile_trailer' at kvmdump.c:947:3,
inlined from 'kvmdump_init' at kvmdump.c:145:4:
kvmdump.c:972:13: warning: 'write' reading 8 bytes from a region of size 4 [-Wstringop-overread]
972 | if (write(kvm->mapfd, &kvm->cpu_devices, sizeof(uint64_t)) != sizeof(uint64_t))
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from kvmdump.c:19:
kvmdump.c: In function 'kvmdump_init':
kvmdump.h:67:18: note: source object 'cpu_devices' of size 4
67 | uint32_t cpu_devices;
| ^~~~~~~~~~~
In file included from defs.h:26,
from kvmdump.c:18:
/usr/include/unistd.h:378:16: note: in a call to function 'write' declared with attribute 'access (read_only, 2, 3)'
378 | extern ssize_t write (int __fd, const void *__buf, size_t __n) __wur
| ^~~~~
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
kvmdump.c | 2 +-
kvmdump.h | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/kvmdump.c b/kvmdump.c
index 4db96bd844e9..e515bf0ce9a1 100644
--- a/kvmdump.c
+++ b/kvmdump.c
@@ -297,7 +297,7 @@ kvmdump_memory_dump(FILE *ofp)
(ulonglong)kvm->page_cache[i].paddr);
}
- fprintf(ofp, " cpu_devices: %d\n", kvm->cpu_devices);
+ fprintf(ofp, " cpu_devices: %ld\n", kvm->cpu_devices);
fprintf(ofp, " iohole: %llx (%llx - %llx)\n",
(ulonglong)kvm->iohole, 0x100000000ULL - kvm->iohole,
0x100000000ULL);
diff --git a/kvmdump.h b/kvmdump.h
index 07e047bb171c..2e408aebef0b 100644
--- a/kvmdump.h
+++ b/kvmdump.h
@@ -64,7 +64,7 @@ struct kvmdump_data {
ulong compresses;
uint64_t kvbase;
ulong *debug;
- uint32_t cpu_devices;
+ uint64_t cpu_devices;
struct register_set *registers;
uint64_t iohole;
};
--
2.37.1

49
0015-sbitmapq-Fix-for-sbitmap_queue-without-min_shallow_d.patch
View File

@ -1,49 +0,0 @@
From 12fe6c7cdd768f87ce6e903a2bbfb0c0591585c5 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 10 Jun 2022 11:49:47 +0900
Subject: [PATCH 15/16] sbitmapq: Fix for sbitmap_queue without
min_shallow_depth member
The sbitmap_queue.min_shallow_depth member was added by kernel commit
a327553965de ("sbitmap: fix missed wakeups caused by sbitmap_queue_get_shallow()")
at Linux 4.18. Without the patch, on earlier kernels the "sbitmapq"
command fails with the following error:
crash> sbitmapq ffff89bb7638ee50
sbitmapq: invalid structure member offset: sbitmap_queue_min_shallow_depth
FILE: sbitmap.c LINE: 398 FUNCTION: sbitmap_queue_context_load()
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
sbitmap.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/sbitmap.c b/sbitmap.c
index c9f7209f9e3e..bb2f19e6207b 100644
--- a/sbitmap.c
+++ b/sbitmap.c
@@ -371,7 +371,8 @@ static void sbitmap_queue_show(const struct sbitmap_queue_context *sqc,
else if (VALID_MEMBER(sbitmap_round_robin)) /* 5.13 and later */
fprintf(fp, "round_robin = %d\n", sc->round_robin);
- fprintf(fp, "min_shallow_depth = %u\n", sqc->min_shallow_depth);
+ if (VALID_MEMBER(sbitmap_queue_min_shallow_depth)) /* 4.18 and later */
+ fprintf(fp, "min_shallow_depth = %u\n", sqc->min_shallow_depth);
}
static void sbitmap_queue_context_load(ulong addr, struct sbitmap_queue_context *sqc)
@@ -395,7 +396,8 @@ static void sbitmap_queue_context_load(ulong addr, struct sbitmap_queue_context
sqc->ws_active = INT(sbitmap_queue_buf + OFFSET(sbitmap_queue_ws_active));
if (VALID_MEMBER(sbitmap_queue_round_robin))
sqc->round_robin = BOOL(sbitmap_queue_buf + OFFSET(sbitmap_queue_round_robin));
- sqc->min_shallow_depth = UINT(sbitmap_queue_buf + OFFSET(sbitmap_queue_min_shallow_depth));
+ if (VALID_MEMBER(sbitmap_queue_min_shallow_depth))
+ sqc->min_shallow_depth = UINT(sbitmap_queue_buf + OFFSET(sbitmap_queue_min_shallow_depth));
FREEBUF(sbitmap_queue_buf);
}
--
2.30.2

84
0016-Make-dev-d-D-options-parse-sbitmap-on-Linux-4.18-and.patch
View File

@ -1,84 +0,0 @@
From c07068266b41450ca6821ee0a1a3adf34206015f Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Fri, 10 Jun 2022 15:21:53 +0900
Subject: [PATCH 16/16] Make "dev -d|-D" options parse sbitmap on Linux 4.18
and later
There have been a few reports that the "dev -d|-D" options displayed
incorrect I/O stats due to racy blk_mq_ctx.rq_* counters. To fix it,
make the options parse sbitmap to count I/O stats on Linux 4.18 and
later kernels, which include RHEL8 ones.
To do this, adjust to the blk_mq_tags structure of Linux 5.10 through
5.15 kernels, which contain kernel commit 222a5ae03cdd ("blk-mq: Use
pointers for blk_mq_tags bitmap tags") and do not contain ae0f1a732f4a
("blk-mq: Stop using pointers for blk_mq_tags bitmap tags").
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
dev.c | 25 +++++++++++++++++++++++--
1 file changed, 23 insertions(+), 2 deletions(-)
diff --git a/dev.c b/dev.c
index 0172c83ffaea..db97f8aebdc2 100644
--- a/dev.c
+++ b/dev.c
@@ -4339,6 +4339,10 @@ static void bt_for_each(ulong q, ulong tags, ulong sbq, uint reserved, uint nr_r
static void queue_for_each_hw_ctx(ulong q, ulong *hctx, uint cnt, struct diskio *dio)
{
uint i;
+ int bitmap_tags_is_ptr = 0;
+
+ if (MEMBER_TYPE("blk_mq_tags", "bitmap_tags") == TYPE_CODE_PTR)
+ bitmap_tags_is_ptr = 1;
for (i = 0; i < cnt; i++) {
ulong addr = 0, tags = 0;
@@ -4357,9 +4361,17 @@ static void queue_for_each_hw_ctx(ulong q, ulong *hctx, uint cnt, struct diskio
if (nr_reserved_tags) {
addr = tags + OFFSET(blk_mq_tags_breserved_tags);
+ if (bitmap_tags_is_ptr &&
+ !readmem(addr, KVADDR, &addr, sizeof(ulong),
+ "blk_mq_tags.bitmap_tags", RETURN_ON_ERROR))
+ break;
bt_for_each(q, tags, addr, 1, nr_reserved_tags, dio);
}
addr = tags + OFFSET(blk_mq_tags_bitmap_tags);
+ if (bitmap_tags_is_ptr &&
+ !readmem(addr, KVADDR, &addr, sizeof(ulong),
+ "blk_mq_tags.bitmap_tags", RETURN_ON_ERROR))
+ break;
bt_for_each(q, tags, addr, 0, nr_reserved_tags, dio);
}
}
@@ -4423,14 +4435,23 @@ get_mq_diskio(unsigned long q, unsigned long *mq_count)
unsigned long mctx_addr;
struct diskio tmp = {0};
- if (INVALID_MEMBER(blk_mq_ctx_rq_dispatched) ||
- INVALID_MEMBER(blk_mq_ctx_rq_completed)) {
+ /*
+ * Currently this function does not support old blk-mq implementation
+ * before 12f5b9314545 ("blk-mq: Remove generation seqeunce"), so
+ * filter them out.
+ */
+ if (VALID_MEMBER(request_state)) {
+ if (CRASHDEBUG(1))
+ fprintf(fp, "mq: using sbitmap\n");
get_mq_diskio_from_hw_queues(q, &tmp);
mq_count[0] = tmp.read;
mq_count[1] = tmp.write;
return;
}
+ if (CRASHDEBUG(1))
+ fprintf(fp, "mq: using blk_mq_ctx.rq_{completed,dispatched} counters\n");
+
readmem(q + OFFSET(request_queue_queue_ctx), KVADDR, &queue_ctx,
sizeof(ulong), "request_queue.queue_ctx",
FAULT_ON_ERROR);
--
2.30.2

156
0016-x86_64-Fix-for-AMD-SME-issue.patch
View File

@ -1,156 +0,0 @@
From 763e221388219b07bd949a9ba48768856908ec6d Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Thu, 28 Jul 2022 15:11:20 +0800
Subject: [PATCH 16/29] x86_64: Fix for AMD SME issue
Kernel commit changes(see [1]/[2]) may cause the failure of crash-utility
with the following error:
#./crash /home/vmlinux /home/vmcore
...
For help, type "help".
Type "apropos word" to search for commands related to "word"...
crash: seek error: physical address: 8000760a14000 type: "p4d page"
Let's get the "NUMBER(sme_mask)" from vmcoreinfo, and try to remove
the C-bit from the page table entries, the intention is to get the
true physical address.
Related kernel commits:
[1] aad983913d77 ("x86/mm/encrypt: Simplify sme_populate_pgd() and sme_populate_pgd_large()")
[2] e7d445ab26db ("x86/sme: Use #define USE_EARLY_PGTABLE_L5 in mem_encrypt_identity.c")
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 1 +
x86_64.c | 21 ++++++++++++++++++---
2 files changed, 19 insertions(+), 3 deletions(-)
diff --git a/defs.h b/defs.h
index 9b1b69a4f421..9d6d8916284b 100644
--- a/defs.h
+++ b/defs.h
@@ -6198,6 +6198,7 @@ struct machine_specific {
ulong cpu_entry_area_end;
ulong page_offset_force;
char **exception_functions;
+ ulong sme_mask;
};
#define KSYMS_START (0x1)
diff --git a/x86_64.c b/x86_64.c
index cfafbcc4dabe..dfada48d8b26 100644
--- a/x86_64.c
+++ b/x86_64.c
@@ -208,6 +208,10 @@ x86_64_init(int when)
machdep->machspec->kernel_image_size = dtol(string, QUIET, NULL);
free(string);
}
+ if ((string = pc->read_vmcoreinfo("NUMBER(sme_mask)"))) {
+ machdep->machspec->sme_mask = dtol(string, QUIET, NULL);
+ free(string);
+ }
if (SADUMP_DUMPFILE() || QEMU_MEM_DUMP_NO_VMCOREINFO() ||
VMSS_DUMPFILE())
/* Need for calculation of kaslr_offset and phys_base */
@@ -940,6 +944,7 @@ x86_64_dump_machdep_table(ulong arg)
ms->kernel_image_size/MEGABYTES(1));
else
fprintf(fp, "(uninitialized)\n");
+ fprintf(fp, " sme_mask: %lx\n", ms->sme_mask);
fprintf(fp, " physical_mask_shift: %ld\n", ms->physical_mask_shift);
fprintf(fp, " pgdir_shift: %ld\n", ms->pgdir_shift);
fprintf(fp, " GART_start: %lx\n", ms->GART_start);
@@ -1817,7 +1822,7 @@ x86_64_kpgd_offset(ulong kvaddr, int verbose, int IS_XEN)
if (IS_XEN)
fprintf(fp, "PAGE DIRECTORY: %lx [machine]\n", *pgd);
else
- fprintf(fp, "PAGE DIRECTORY: %lx\n", *pgd);
+ fprintf(fp, "PAGE DIRECTORY: %lx\n", *pgd & ~machdep->machspec->sme_mask);
}
return pgd;
@@ -1854,7 +1859,8 @@ x86_64_upgd_offset_legacy(struct task_context *tc, ulong uvaddr, int verbose, in
if (IS_XEN)
fprintf(fp, " PGD: %lx => %lx [machine]\n", (ulong)pud, pud_pte);
else
- fprintf(fp, " PGD: %lx => %lx\n", (ulong)pud, pud_pte);
+ fprintf(fp, " PGD: %lx => %lx\n",
+ (ulong)pud, pud_pte & ~machdep->machspec->sme_mask);
}
return pud_pte;
@@ -1885,7 +1891,8 @@ x86_64_upgd_offset(struct task_context *tc, ulong uvaddr, int verbose, int IS_XE
if (IS_XEN)
fprintf(fp, " PGD: %lx => %lx [machine]\n", (ulong)pgd, pgd_pte);
else
- fprintf(fp, " PGD: %lx => %lx\n", (ulong)pgd, pgd_pte);
+ fprintf(fp, " PGD: %lx => %lx\n",
+ (ulong)pgd, pgd_pte & ~machdep->machspec->sme_mask);
}
return pgd_pte;
@@ -1903,9 +1910,11 @@ x86_64_p4d_offset(ulong pgd_pte, ulong vaddr, int verbose, int IS_XEN)
ulong p4d_pte;
p4d_paddr = pgd_pte & PHYSICAL_PAGE_MASK;
+ p4d_paddr &= ~machdep->machspec->sme_mask;
FILL_P4D(p4d_paddr, PHYSADDR, PAGESIZE());
p4d = ((ulong *)p4d_paddr) + p4d_index(vaddr);
p4d_pte = ULONG(machdep->machspec->p4d + PAGEOFFSET(p4d));
+ p4d_pte &= ~machdep->machspec->sme_mask;
if (verbose) {
if (IS_XEN)
fprintf(fp, " P4D: %lx => %lx [machine]\n", (ulong)p4d, p4d_pte);
@@ -1928,6 +1937,7 @@ x86_64_pud_offset(ulong pgd_pte, ulong vaddr, int verbose, int IS_XEN)
ulong pud_pte;
pud_paddr = pgd_pte & PHYSICAL_PAGE_MASK;
+ pud_paddr &= ~machdep->machspec->sme_mask;
if (IS_XEN) {
pud_paddr = xen_m2p(pud_paddr);
@@ -1938,6 +1948,7 @@ x86_64_pud_offset(ulong pgd_pte, ulong vaddr, int verbose, int IS_XEN)
FILL_PUD(pud_paddr, PHYSADDR, PAGESIZE());
pud = ((ulong *)pud_paddr) + pud_index(vaddr);
pud_pte = ULONG(machdep->pud + PAGEOFFSET(pud));
+ pud_pte &= ~machdep->machspec->sme_mask;
if (verbose) {
if (IS_XEN)
fprintf(fp, " PUD: %lx => %lx [machine]\n", (ulong)pud, pud_pte);
@@ -1960,6 +1971,7 @@ x86_64_pmd_offset(ulong pud_pte, ulong vaddr, int verbose, int IS_XEN)
ulong pmd_pte;
pmd_paddr = pud_pte & PHYSICAL_PAGE_MASK;
+ pmd_paddr &= ~machdep->machspec->sme_mask;
if (IS_XEN) {
pmd_paddr = xen_m2p(pmd_paddr);
@@ -1970,6 +1982,7 @@ x86_64_pmd_offset(ulong pud_pte, ulong vaddr, int verbose, int IS_XEN)
FILL_PMD(pmd_paddr, PHYSADDR, PAGESIZE());
pmd = ((ulong *)pmd_paddr) + pmd_index(vaddr);
pmd_pte = ULONG(machdep->pmd + PAGEOFFSET(pmd));
+ pmd_pte &= ~machdep->machspec->sme_mask;
if (verbose) {
if (IS_XEN)
fprintf(fp, " PMD: %lx => %lx [machine]\n", (ulong)pmd, pmd_pte);
@@ -1991,6 +2004,7 @@ x86_64_pte_offset(ulong pmd_pte, ulong vaddr, int verbose, int IS_XEN)
ulong pte;
pte_paddr = pmd_pte & PHYSICAL_PAGE_MASK;
+ pte_paddr &= ~machdep->machspec->sme_mask;
if (IS_XEN) {
pte_paddr = xen_m2p(pte_paddr);
@@ -2001,6 +2015,7 @@ x86_64_pte_offset(ulong pmd_pte, ulong vaddr, int verbose, int IS_XEN)
FILL_PTBL(pte_paddr, PHYSADDR, PAGESIZE());
ptep = ((ulong *)pte_paddr) + pte_index(vaddr);
pte = ULONG(machdep->ptbl + PAGEOFFSET(ptep));
+ pte &= ~machdep->machspec->sme_mask;
if (verbose) {
if (IS_XEN)
fprintf(fp, " PTE: %lx => %lx [machine]\n", (ulong)ptep, pte);
--
2.37.1

34
0017-Makefile-Fix-unnecessary-re-patching-with-coreutils-.patch
View File

@ -1,34 +0,0 @@
From c2743ad474529951ace2b8ec712bf373f3a07d4c Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Mon, 22 Aug 2022 11:59:46 +0900
Subject: [PATCH 17/29] Makefile: Fix unnecessary re-patching with
coreutils-9.0
"sum" command in coreutils-9.0 (e.g. Fedora 36) started to output a file
name. As a result, "make" always detects a change of gdb-10.2.patch
wrongly and re-applies it unnecessarily.
Use standard input to fix it and "md5sum" to improve detection.
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/Makefile b/Makefile
index 162c2baa5164..79aef1769444 100644
--- a/Makefile
+++ b/Makefile
@@ -272,7 +272,7 @@ rebuild:
@if [ ! -f ${GDB}/${GDB}.patch ]; then \
touch ${GDB}/${GDB}.patch; fi
@if [ -f ${GDB}.patch ] && [ -s ${GDB}.patch ] && \
- [ "`sum ${GDB}.patch`" != "`sum ${GDB}/${GDB}.patch`" ]; then \
+ [ "`md5sum < ${GDB}.patch`" != "`md5sum < ${GDB}/${GDB}.patch`" ]; then \
(sh -x ${GDB}.patch ${TARGET}; patch -N -p0 -r- --fuzz=0 < ${GDB}.patch; cp ${GDB}.patch ${GDB}; cd ${GDB}; \
$(MAKE) CRASH_TARGET=${TARGET}) \
else (cd ${GDB}/gdb; $(MAKE) CRASH_TARGET=${TARGET}); fi
--
2.37.1

52
0018-gdb-fix-for-assigning-NULL-to-std-string.patch
View File

@ -1,52 +0,0 @@
From 4c85e982d25a259f81b5e8c230a67d40d4527ddf Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Wed, 24 Aug 2022 10:19:20 +0800
Subject: [PATCH 18/29] gdb: fix for assigning NULL to std::string
When trying to load a module with "mod -s" without its separated debug
info file installed, the crash utility will abort as below:
crash> mod -s kpatch_test kpatch_test.ko
...
terminate called after throwing an instance of 'std::logic_error'
what(): basic_string::_M_construct null not valid
Aborted (core dumped)
Let's return the std::string() instead of std::string(NULL) when a
string is null, because the check_specified_kernel_debug_file() may
return NULL.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
gdb-10.2.patch | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/gdb-10.2.patch b/gdb-10.2.patch
index 577f5e45fc5a..f0034ed626a0 100644
--- a/gdb-10.2.patch
+++ b/gdb-10.2.patch
@@ -9,6 +9,7 @@
# to all subsequent patch applications.
tar xvzmf gdb-10.2.tar.gz \
+ gdb-10.2/gdb/symfile.c \
gdb-10.2/gdb/Makefile.in
exit 0
@@ -1650,3 +1651,13 @@ exit 0
c_print_type_1 (type->field (i).type (),
TYPE_FIELD_NAME (type, i),
+--- gdb-10.2/gdb/symfile.c.orig
++++ gdb-10.2/gdb/symfile.c
+@@ -1610,7 +1610,7 @@ find_separate_debug_file_by_debuglink (struct objfile *objfile)
+ if (debugfile.empty ()) {
+ char *name_copy;
+ name_copy = check_specified_kernel_debug_file();
+- return std::string (name_copy);
++ return name_copy ? std::string (name_copy) : std::string ();
+ }
+ #endif
--
2.37.1

111
0019-arm64-use-TCR_EL1_T1SZ-to-get-the-correct-info-if-va.patch
View File

@ -1,111 +0,0 @@
From f02c8e87fccb1a92fbc025883bc69b6467a4e6c8 Mon Sep 17 00:00:00 2001
From: Huang Shijie <shijie@os.amperecomputing.com>
Date: Mon, 22 Aug 2022 09:29:32 +0000
Subject: [PATCH 19/29] arm64: use TCR_EL1_T1SZ to get the correct info if
vabits_actual is missing
After kernel commit 0d9b1ffefabe ("arm64: mm: make vabits_actual a build
time constant if possible"), the vabits_actual is not compiled to kernel
symbols when "VA_BITS > 48" is false.
So the crash will not find the vabits_actual symbol, and it will fail
in the end like this:
# ./crash
...
WARNING: VA_BITS: calculated: 46 vmcoreinfo: 48
crash: invalid kernel virtual address: ffff88177ffff000 type: "pud page"
This patch introduces the arm64_set_va_bits_by_tcr(), and if crash cannot
find vabits_actual symbol, it will use the TCR_EL1_T1SZ register to get
the correct VA_BITS_ACTUAL/VA_BITS/VA_START.
Tested this patch with:
1.) the live mode with /proc/kcore
2.) the kdump file with /proc/vmcore.
Signed-off-by: Huang Shijie <shijie@os.amperecomputing.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
arm64.c | 51 ++++++++++++++++++++++++++++++++++-----------------
1 file changed, 34 insertions(+), 17 deletions(-)
diff --git a/arm64.c b/arm64.c
index b6b7aa11f4fe..c3e26a371a61 100644
--- a/arm64.c
+++ b/arm64.c
@@ -4586,6 +4586,36 @@ arm64_IS_VMALLOC_ADDR(ulong vaddr)
(vaddr >= ms->modules_vaddr && vaddr <= ms->modules_end));
}
+/* Return TRUE if we succeed, return FALSE on failure. */
+static int
+arm64_set_va_bits_by_tcr(void)
+{
+ ulong value;
+ char *string;
+
+ if ((string = pc->read_vmcoreinfo("NUMBER(TCR_EL1_T1SZ)")) ||
+ (string = pc->read_vmcoreinfo("NUMBER(tcr_el1_t1sz)"))) {
+ /* See ARMv8 ARM for the description of
+ * TCR_EL1.T1SZ and how it can be used
+ * to calculate the vabits_actual
+ * supported by underlying kernel.
+ *
+ * Basically:
+ * vabits_actual = 64 - T1SZ;
+ */
+ value = 64 - strtoll(string, NULL, 0);
+ if (CRASHDEBUG(1))
+ fprintf(fp, "vmcoreinfo : vabits_actual: %ld\n", value);
+ free(string);
+ machdep->machspec->VA_BITS_ACTUAL = value;
+ machdep->machspec->VA_BITS = value;
+ machdep->machspec->VA_START = _VA_START(machdep->machspec->VA_BITS_ACTUAL);
+ return TRUE;
+ }
+
+ return FALSE;
+}
+
static void
arm64_calc_VA_BITS(void)
{
@@ -4616,23 +4646,8 @@ arm64_calc_VA_BITS(void)
} else if (ACTIVE())
error(FATAL, "cannot determine VA_BITS_ACTUAL: please use /proc/kcore\n");
else {
- if ((string = pc->read_vmcoreinfo("NUMBER(TCR_EL1_T1SZ)")) ||
- (string = pc->read_vmcoreinfo("NUMBER(tcr_el1_t1sz)"))) {
- /* See ARMv8 ARM for the description of
- * TCR_EL1.T1SZ and how it can be used
- * to calculate the vabits_actual
- * supported by underlying kernel.
- *
- * Basically:
- * vabits_actual = 64 - T1SZ;
- */
- value = 64 - strtoll(string, NULL, 0);
- if (CRASHDEBUG(1))
- fprintf(fp, "vmcoreinfo : vabits_actual: %ld\n", value);
- free(string);
- machdep->machspec->VA_BITS_ACTUAL = value;
- machdep->machspec->VA_BITS = value;
- machdep->machspec->VA_START = _VA_START(machdep->machspec->VA_BITS_ACTUAL);
+ if (arm64_set_va_bits_by_tcr()) {
+ /* nothing */
} else if (machdep->machspec->VA_BITS_ACTUAL) {
machdep->machspec->VA_BITS = machdep->machspec->VA_BITS_ACTUAL;
machdep->machspec->VA_START = _VA_START(machdep->machspec->VA_BITS_ACTUAL);
@@ -4654,6 +4669,8 @@ arm64_calc_VA_BITS(void)
*/
machdep->flags |= FLIPPED_VM;
return;
+ } else if (arm64_set_va_bits_by_tcr()) {
+ return;
}
if (!(sp = symbol_search("swapper_pg_dir")) &&
--
2.37.1

66
0020-Fix-task-R-by-adding-end-identifier-for-union-in-tas.patch
View File

@ -1,66 +0,0 @@
From 9cbfea67eb4f094d47cd841b73ddbbdbe6b58696 Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Thu, 25 Aug 2022 14:39:44 +0800
Subject: [PATCH 20/29] Fix "task -R" by adding end identifier for union in
task_struct
Previously, the start and end identifiers for union are " {\n" and
" }, \n". However the end identifier is not always as expected.
" },\n" can also be the end identifier with gdb-10.2. As a result,
variable "randomized" is in incorrect state after union, and fails to
identify the later struct members. For example, we can reproduce the
issue as follows:
crash> task
PID: 847 TASK: ffff94f8038f4000 CPU: 72 COMMAND: "khungtaskd"
struct task_struct {
thread_info = {
flags = 2148024320,
status = 0,
preempt_lazy_count = 0
},
{
<the union>
},
...
wake_entry = {
next = 0x0
},
...
Before patch:
crash> task -R wake_entry
PID: 847 TASK: ffff94f8038f4000 CPU: 72 COMMAND: "khungtaskd"
After patch:
crash> task -R wake_entry
PID: 847 TASK: ffff94f8038f4000 CPU: 72 COMMAND: "khungtaskd"
wake_entry = {
next = 0x0
},
Signed-off-by: Tao Liu <ltao@redhat.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
task.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/task.c b/task.c
index 071c787fbfa5..db2abc8106a2 100644
--- a/task.c
+++ b/task.c
@@ -3436,7 +3436,8 @@ parse_task_thread(int argcnt, char *arglist[], struct task_context *tc) {
while (fgets(buf, BUFSIZE, pc->tmpfile)) {
if (STREQ(buf, " {\n"))
randomized = TRUE;
- else if (randomized && STREQ(buf, " }, \n"))
+ else if (randomized &&
+ (STREQ(buf, " }, \n") || STREQ(buf, " },\n")))
randomized = FALSE;
if (strlen(lookfor2)) {
--
2.37.1

142
0021-Let-gdb-get-kernel-module-symbols-info-from-crash.patch
View File

@ -1,142 +0,0 @@
From 2145b2bb79c59aa25c5155a8f9851554d1813fb9 Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Wed, 31 Aug 2022 11:54:13 +0800
Subject: [PATCH 21/29] Let gdb get kernel module symbols info from crash
Gdb will try to resolve an address to its corresponding symbol name such as
when printing a structure. It works fine for kernel symbols, because gdb can
find them through vmlinux. However as for kernel modules symbols, crash
resolves them by dig into "struct module", which gdb don't know. As a result,
gdb fails to translate a kernel module address to its symbol name without
"mod -s|-S" options. For example we can reproduce the issue as follows.
crash> timer
....
4331308176 336 ffff94ea24240860 ffffffffc03762c0 <estimation_timer>
....
crash> sym 0xffffffffc03762c0
ffffffffc03762c0 (t) estimation_timer [ip_vs]
Before patch:
crash> timer_list ffff94ea24240860
struct timer_list {
....
function = 0xffffffffc03762c0,
....
}
After patch:
crash> timer_list ffff94ea24240860
struct timer_list {
....
function = 0xffffffffc03762c0 <estimation_timer>,
....
}
In this patch, we add an interface for gdb, when gdb trying to build kernel
module's address symbolic, the info can be get from crash.
Signed-off-by: Tao Liu <ltao@redhat.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
defs.h | 2 ++
gdb-10.2.patch | 35 +++++++++++++++++++++++++++++++++++
gdb_interface.c | 12 ++++++++++++
3 files changed, 49 insertions(+)
diff --git a/defs.h b/defs.h
index 9d6d8916284b..afdcf6c4ac20 100644
--- a/defs.h
+++ b/defs.h
@@ -4874,6 +4874,7 @@ extern "C" int patch_kernel_symbol(struct gnu_request *);
struct syment *symbol_search(char *);
int gdb_line_number_callback(ulong, ulong, ulong);
int gdb_print_callback(ulong);
+char *gdb_lookup_module_symbol(ulong, ulong *);
extern "C" int same_file(char *, char *);
#endif
@@ -7284,6 +7285,7 @@ int gdb_pass_through(char *, FILE *, ulong);
int gdb_readmem_callback(ulong, void *, int, int);
int gdb_line_number_callback(ulong, ulong, ulong);
int gdb_print_callback(ulong);
+char *gdb_lookup_module_symbol(ulong, ulong *);
void gdb_error_hook(void);
void restore_gdb_sanity(void);
int is_gdb_command(int, ulong);
diff --git a/gdb-10.2.patch b/gdb-10.2.patch
index f0034ed626a0..642cd6a78a1d 100644
--- a/gdb-10.2.patch
+++ b/gdb-10.2.patch
@@ -9,6 +9,7 @@
# to all subsequent patch applications.
tar xvzmf gdb-10.2.tar.gz \
+ gdb-10.2/gdb/printcmd.c \
gdb-10.2/gdb/symfile.c \
gdb-10.2/gdb/Makefile.in
@@ -1661,3 +1662,37 @@ exit 0
+ return name_copy ? std::string (name_copy) : std::string ();
}
#endif
+
+--- gdb-10.2/gdb/printcmd.c.orig
++++ gdb-10.2/gdb/printcmd.c
+@@ -576,6 +576,10 @@ print_address_symbolic (struct gdbarch *gdbarch, CORE_ADDR addr,
+
+ /* See valprint.h. */
+
++#ifdef CRASH_MERGE
++extern "C" char *gdb_lookup_module_symbol(unsigned long, unsigned long *);
++#endif
++
+ int
+ build_address_symbolic (struct gdbarch *gdbarch,
+ CORE_ADDR addr, /* IN */
+@@ -682,7 +686,19 @@ build_address_symbolic (struct gdbarch *gdbarch,
+ }
+ }
+ if (symbol == NULL && msymbol.minsym == NULL)
++#ifdef CRASH_MERGE
++ {
++ char *name_ptr = gdb_lookup_module_symbol(addr, (unsigned long *)offset);
++ if (name_ptr) {
++ *name = name_ptr;
++ return 0;
++ } else {
++ return 1;
++ }
++ }
++#else
+ return 1;
++#endif
+
+ /* If the nearest symbol is too far away, don't print anything symbolic. */
+
diff --git a/gdb_interface.c b/gdb_interface.c
index 3a7fcc9e3ade..b14319c66147 100644
--- a/gdb_interface.c
+++ b/gdb_interface.c
@@ -935,6 +935,18 @@ gdb_print_callback(ulong addr)
return IS_KVADDR(addr);
}
+char *
+gdb_lookup_module_symbol(ulong addr, ulong *offset)
+{
+ struct syment *sp;
+
+ if ((sp = value_search_module(addr, offset))) {
+ return sp->name;
+ } else {
+ return NULL;
+ }
+}
+
/*
* Used by gdb_interface() to catch gdb-related errors, if desired.
*/
--
2.37.1

65
0022-x86_64-Correct-the-identifier-when-locating-the-call.patch
View File

@ -1,65 +0,0 @@
From 3ed9ec5c8d09cffac9772abbf54214125ade9127 Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Wed, 31 Aug 2022 11:54:15 +0800
Subject: [PATCH 22/29] x86_64: Correct the identifier when locating the call
instruction
The previous implementation to locate the call instruction is
to strstr "call", then check whether the previous char is ' '
or '\t'. The implementation is problematic. For example it
cannot resolve the following disassembly string:
"0xffffffffc0995378 <nfs41_callback_svc+344>:\tcall 0xffffffff8ecfa4c0 <schedule>\n"
strstr will locate the "_call" and char check fails,
as a result, extract_hex fails to get the calling address.
NOTE: the issue is more likely to be reproduced when patch[1] applied.
Because without patch[1], the disassembly string will be as follows,
so the issue is no longer reproducible.
"0xffffffffc0995378:\tcall 0xffffffff8ecfa4c0 <schedule>\n"
Before the patch:
crash> bt 1472
PID: 1472 TASK: ffff8c121fa72f70 CPU: 18 COMMAND: "nfsv4.1-svc"
#0 [ffff8c16231a3db8] __schedule at ffffffff8ecf9ef3
#1 [ffff8c16231a3e40] schedule at ffffffff8ecfa4e9
After the patch:
crash> bt 1472
PID: 1472 TASK: ffff8c121fa72f70 CPU: 18 COMMAND: "nfsv4.1-svc"
#0 [ffff8c16231a3db8] __schedule at ffffffff8ecf9ef3
#1 [ffff8c16231a3e40] schedule at ffffffff8ecfa4e9
#2 [ffff8c16231a3e50] nfs41_callback_svc at ffffffffc099537d [nfsv4]
#3 [ffff8c16231a3ec8] kthread at ffffffff8e6b966f
#4 [ffff8c16231a3f50] ret_from_fork at ffffffff8ed07898
This patch fix the issue by strstr "\tcall" and " call", to
locate the correct call instruction.
[1]: https://listman.redhat.com/archives/crash-utility/2022-August/010085.html
Signed-off-by: Tao Liu <ltao@redhat.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
x86_64.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/x86_64.c b/x86_64.c
index dfada48d8b26..74bd1bbde41c 100644
--- a/x86_64.c
+++ b/x86_64.c
@@ -4432,8 +4432,7 @@ x86_64_function_called_by(ulong rip)
if (gdb_pass_through(buf, pc->tmpfile2, GNU_RETURN_ON_ERROR)) {
rewind(pc->tmpfile2);
while (fgets(buf, BUFSIZE, pc->tmpfile2)) {
- if ((p1 = strstr(buf, "call")) &&
- whitespace(*(p1-1))) {
+ if ((p1 = strstr(buf, " call")) || (p1 = strstr(buf, "\tcall"))) {
if (extract_hex(p1, &value, NULLCHAR, TRUE))
break;
}
--
2.37.1

40
0023-Add-debian-ubuntu-vmlinux-location-to-default-search.patch
View File

@ -1,40 +0,0 @@
From 5b9d3e98cda9d99f3277aabec30d076e62cc5e71 Mon Sep 17 00:00:00 2001
From: "Chunguang.Xu" <chunguang.xu@shopee.com>
Date: Thu, 25 Aug 2022 12:07:20 +0800
Subject: [PATCH 23/29] Add debian/ubuntu vmlinux location to default search
dirs
Now crash cannot find debian/ubuntu kernel vmlinux, we need to
explicitly specify the path to vmlinux. Try to add the debian
vmlinux location to default search directories.
Signed-off-by: Chunguang Xu <chunguang.xu@shopee.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
filesys.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/filesys.c b/filesys.c
index a863f04eb250..c2ea78de821d 100644
--- a/filesys.c
+++ b/filesys.c
@@ -319,7 +319,7 @@ match_proc_version(void)
#define CREATE 1
#define DESTROY 0
-#define DEFAULT_SEARCHDIRS 5
+#define DEFAULT_SEARCHDIRS 6
#define EXTRA_SEARCHDIRS 5
static char **
@@ -336,6 +336,7 @@ build_searchdirs(int create, int *preferred)
"/boot/",
"/boot/efi/redhat",
"/boot/efi/EFI/redhat",
+ "/usr/lib/debug/boot/",
"/",
NULL
};
--
2.37.1

102
0024-Fix-gcc-12-compiler-warnings-on-lkcd_-.c.patch
View File

@ -1,102 +0,0 @@
From 51acac75cdb20caab30a85ebfec5906efe034477 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Thu, 1 Sep 2022 14:03:09 +0900
Subject: [PATCH 24/29] Fix gcc-12 compiler warnings on lkcd_*.c
Without the patch, the following gcc-12 compiler warnings are emitted
for lkcd_*.c:
lkcd_v1.c: In function 'dump_lkcd_environment_v1':
lkcd_v1.c:252:20: warning: the comparison will always evaluate as 'true' for the address of 'dh_panic_string' will never be NULL [-Waddress]
252 | dh && dh->dh_panic_string &&
| ^~
In file included from lkcd_v1.c:21:
lkcd_vmdump_v1.h:108:30: note: 'dh_panic_string' declared here
108 | char dh_panic_string[DUMP_PANIC_LEN];
| ^~~~~~~~~~~~~~~
...
Reported-by: Lianbo Jiang <lijiang@redhat.com>
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
lkcd_v1.c | 3 +--
lkcd_v2_v3.c | 3 +--
lkcd_v5.c | 3 +--
lkcd_v7.c | 3 +--
lkcd_v8.c | 3 +--
5 files changed, 5 insertions(+), 10 deletions(-)
diff --git a/lkcd_v1.c b/lkcd_v1.c
index 5f891aeb1ce3..31bdc04c5fdd 100644
--- a/lkcd_v1.c
+++ b/lkcd_v1.c
@@ -249,8 +249,7 @@ dump_header_only:
lkcd_print(" dh_eip: %lx\n", dh->dh_eip);
lkcd_print(" dh_num_pages: %d\n", dh->dh_num_pages);
lkcd_print(" dh_panic_string: %s%s", dh->dh_panic_string,
- dh && dh->dh_panic_string &&
- strstr(dh->dh_panic_string, "\n") ? "" : "\n");
+ dh && strstr(dh->dh_panic_string, "\n") ? "" : "\n");
lkcd_print(" dh_time: %s\n",
strip_linefeeds(ctime(&(dh->dh_time.tv_sec))));
diff --git a/lkcd_v2_v3.c b/lkcd_v2_v3.c
index 8635a7b07c77..edcb6637a55e 100644
--- a/lkcd_v2_v3.c
+++ b/lkcd_v2_v3.c
@@ -307,8 +307,7 @@ dump_header_only:
lkcd_print(" dh_num_pages: ");
lkcd_print(BITS32() ? "%ld\n" : "%d\n", dh->dh_num_pages);
lkcd_print(" dh_panic_string: %s%s", dh->dh_panic_string,
- dh && dh->dh_panic_string &&
- strstr(dh->dh_panic_string, "\n") ? "" : "\n");
+ dh && strstr(dh->dh_panic_string, "\n") ? "" : "\n");
lkcd_print(" dh_time: %s\n",
strip_linefeeds(ctime(&(dh->dh_time.tv_sec))));
diff --git a/lkcd_v5.c b/lkcd_v5.c
index cb7634d094ec..e3bfa6f4b272 100644
--- a/lkcd_v5.c
+++ b/lkcd_v5.c
@@ -270,8 +270,7 @@ dump_header_only:
lkcd_print(" dh_num_pages: ");
lkcd_print(BITS32() ? "%ld\n" : "%d\n", dh->dh_num_pages);
lkcd_print(" dh_panic_string: %s%s", dh->dh_panic_string,
- dh && dh->dh_panic_string &&
- strstr(dh->dh_panic_string, "\n") ? "" : "\n");
+ dh && strstr(dh->dh_panic_string, "\n") ? "" : "\n");
lkcd_print(" dh_time: %s\n",
strip_linefeeds(ctime(&(dh->dh_time.tv_sec))));
diff --git a/lkcd_v7.c b/lkcd_v7.c
index 608e1481f01d..97d99008635a 100644
--- a/lkcd_v7.c
+++ b/lkcd_v7.c
@@ -347,8 +347,7 @@ dump_header_only:
lkcd_print(" dh_num_pages: ");
lkcd_print(BITS32() ? "%ld\n" : "%d\n", dh->dh_num_pages);
lkcd_print(" dh_panic_string: %s%s", dh->dh_panic_string,
- dh && dh->dh_panic_string &&
- strstr(dh->dh_panic_string, "\n") ? "" : "\n");
+ dh && strstr(dh->dh_panic_string, "\n") ? "" : "\n");
lkcd_print(" dh_time: %s\n",
strip_linefeeds(ctime(&(dh->dh_time.tv_sec))));
diff --git a/lkcd_v8.c b/lkcd_v8.c
index 3b355e056123..4167fa5e4292 100644
--- a/lkcd_v8.c
+++ b/lkcd_v8.c
@@ -543,8 +543,7 @@ dump_header_only:
lkcd_print(" dh_num_pages: ");
lkcd_print(BITS32() ? "%ld\n" : "%d\n", dh->dh_num_pages);
lkcd_print(" dh_panic_string: %s%s", dh->dh_panic_string,
- dh && dh->dh_panic_string &&
- strstr(dh->dh_panic_string, "\n") ? "" : "\n");
+ dh && strstr(dh->dh_panic_string, "\n") ? "" : "\n");
tv.tv_sec = dh->dh_time.tv_sec;
lkcd_print(" dh_time: %s\n",
strip_linefeeds(ctime(&(tv.tv_sec))));
--
2.37.1

86
0025-Fix-gcc-11-compiler-warnings-on-gdb-10.2-gdb-symtab..patch
View File

@ -1,86 +0,0 @@
From bdbf5887d6259ea3108d4fa674f3794adad54d52 Mon Sep 17 00:00:00 2001
From: Kazuhito Hagio <k-hagio-ab@nec.com>
Date: Thu, 1 Sep 2022 13:42:28 +0900
Subject: [PATCH 25/29] Fix gcc-11 compiler warnings on gdb-10.2/gdb/symtab.c
Without the patch, the following gcc-11 compiler warnings are emitted
for gdb-10.2/gdb/symtab.c:
symtab.c: In function 'void gdb_get_datatype(gnu_request*)':
symtab.c:7131:31: warning: ISO C++17 does not allow 'register' storage class specifier [-Wregister]
7131 | register struct type *type;
| ^~~~
symtab.c:7132:31: warning: ISO C++17 does not allow 'register' storage class specifier [-Wregister]
7132 | register struct type *typedef_type;
| ^~~~~~~~~~~~
...
Usually we don't fix compiler warnings for gdb, but these are emitted
even by "make clean ; make warn", which doesn't recompile the whole
gdb, so it would be better to fix.
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
gdb-10.2.patch | 41 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 41 insertions(+)
diff --git a/gdb-10.2.patch b/gdb-10.2.patch
index 642cd6a78a1d..91edfb338445 100644
--- a/gdb-10.2.patch
+++ b/gdb-10.2.patch
@@ -9,6 +9,7 @@
# to all subsequent patch applications.
tar xvzmf gdb-10.2.tar.gz \
+ gdb-10.2/gdb/symtab.c \
gdb-10.2/gdb/printcmd.c \
gdb-10.2/gdb/symfile.c \
gdb-10.2/gdb/Makefile.in
@@ -1696,3 +1697,43 @@ exit 0
/* If the nearest symbol is too far away, don't print anything symbolic. */
+--- gdb-10.2/gdb/symtab.c.orig
++++ gdb-10.2/gdb/symtab.c
+@@ -7128,8 +7128,8 @@ gdb_get_line_number(struct gnu_request *
+ static void
+ gdb_get_datatype(struct gnu_request *req)
+ {
+- register struct type *type;
+- register struct type *typedef_type;
++ struct type *type;
++ struct type *typedef_type;
+ expression_up expr;
+ struct symbol *sym;
+ struct value *val;
+@@ -7235,7 +7235,7 @@ gdb_get_datatype(struct gnu_request *req
+ static void
+ dump_enum(struct type *type, struct gnu_request *req)
+ {
+- register int i;
++ int i;
+ int len;
+ long long lastval;
+
+@@ -7271,7 +7271,7 @@ dump_enum(struct type *type, struct gnu_
+ static void
+ eval_enum(struct type *type, struct gnu_request *req)
+ {
+- register int i;
++ int i;
+ int len;
+ long long lastval;
+
+@@ -7298,7 +7298,7 @@ eval_enum(struct type *type, struct gnu_
+ static void
+ get_member_data(struct gnu_request *req, struct type *type, long offset, int is_first)
+ {
+- register short i;
++ short i;
+ struct field *nextfield;
+ short nfields;
+ struct type *typedef_type, *target_type;
--
2.37.1

42
0026-Fix-for-the-invalid-linux_banner-pointer-issue.patch
View File

@ -1,42 +0,0 @@
From 4ea3a806d11f000f2eb1ddc72c2b7a543e319f64 Mon Sep 17 00:00:00 2001
From: Lianbo Jiang <lijiang@redhat.com>
Date: Fri, 16 Sep 2022 14:00:01 +0800
Subject: [PATCH 26/29] Fix for the invalid linux_banner pointer issue
Currently, crash may fail with the following error:
# ./crash -s vmlinux vmcore
WARNING: invalid linux_banner pointer: 65762078756e694c
crash: vmlinux and vmcore do not match!
The reason is that the type of the symbol in the data segment may be
defined as 'D' or 'd'. The crash only handled the type 'D', but it
didn't deal with the type 'd'. For example:
# nm vmlinux | grep linux_banner
ffffffff827cfa80 d linux_banner
It has been observed that a vmlinux compiled by clang has this type.
Let's add the type 'd' recognition to solve such issue.
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
kernel.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel.c b/kernel.c
index a521ef30cdb0..aa030e8097ea 100644
--- a/kernel.c
+++ b/kernel.c
@@ -1060,7 +1060,7 @@ verify_version(void)
if (!(sp = symbol_search("linux_banner")))
error(FATAL, "linux_banner symbol does not exist?\n");
else if ((sp->type == 'R') || (sp->type == 'r') ||
- (THIS_KERNEL_VERSION >= LINUX(2,6,11) && sp->type == 'D') ||
+ (THIS_KERNEL_VERSION >= LINUX(2,6,11) && (sp->type == 'D' || sp->type == 'd')) ||
(machine_type("ARM") && sp->type == 'T') ||
(machine_type("ARM64")))
linux_banner = symbol_value("linux_banner");
--
2.37.1

57
0027-Fix-kmem-failing-to-print-task-context-when-address-.patch
View File

@ -1,57 +0,0 @@
From ad1397a73594d65aaad9d0b9a94a1dd75d8c61dd Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Mon, 19 Sep 2022 17:49:21 +0800
Subject: [PATCH 27/29] Fix "kmem" failing to print task context when address
is vmalloced stack
When kernel enabled CONFIG_VMAP_STACK, stack can be allocated to
vmalloced area. Currently crash didn't handle the case, as a result,
"kmem" will not print the task context as expected. This patch fix the
bug by checking if the address is a vmalloced stack first.
Before:
crash> kmem ffffb7efce9bbe28
VMAP_AREA VM_STRUCT ADDRESS RANGE SIZE
ffff94eb9102c640 ffff94eb9102b140 ffffb7efce9b8000 - ffffb7efce9bd000 20480
PAGE PHYSICAL MAPPING INDEX CNT FLAGS
ffffdd28220dc000 1883700000 0 0 1 50000000000000
After:
crash> kmem ffffb7efce9bbe28
PID: 847
COMMAND: "khungtaskd"
TASK: ffff94f8038f4000 [THREAD_INFO: ffff94f8038f4000]
CPU: 72
STATE: TASK_RUNNING (PANIC)
VMAP_AREA VM_STRUCT ADDRESS RANGE SIZE
ffff94eb9102c640 ffff94eb9102b140 ffffb7efce9b8000 - ffffb7efce9bd000 20480
PAGE PHYSICAL MAPPING INDEX CNT FLAGS
ffffdd28220dc000 1883700000 0 0 1 50000000000000
Signed-off-by: Tao Liu <ltao@redhat.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
memory.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/memory.c b/memory.c
index 7339f0cd0224..9ab578134fa1 100644
--- a/memory.c
+++ b/memory.c
@@ -13477,6 +13477,10 @@ kmem_search(struct meminfo *mi)
* Check for a valid mapped address.
*/
if ((mi->memtype == KVADDR) && IS_VMALLOC_ADDR(mi->spec_addr)) {
+ if ((task = stkptr_to_task(vaddr)) && (tc = task_to_context(task))) {
+ show_context(tc);
+ fprintf(fp, "\n");
+ }
if (kvtop(NULL, mi->spec_addr, &paddr, 0)) {
mi->flags = orig_flags | VMLIST_VERIFY;
dump_vmlist(mi);
--
2.37.1

49
0028-Fix-page-offset-issue-when-converting-physical-to-vi.patch
View File

@ -1,49 +0,0 @@
From 60cb8650a0126abda661c44d198ebde514eca3e2 Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Mon, 19 Sep 2022 17:49:22 +0800
Subject: [PATCH 28/29] Fix page offset issue when converting physical to
virtual address
When trying to convert a physical address to its virtual
address in dump_vmap_area() and dump_vmlist(), the vi->retval
is added by 2 values: the page aligned address "pcheck"
and page offset address "PAGEOFFSET(paddr)".
However "paddr" is given by "pcheck", is also page aligned,
so "PAGEOFFSET(paddr)" is always 0.
In this patch, we will use PAGEOFFSET(vi->spec_addr) to give the
page offset, vi->spec_addr is the physical address we'd like
to convert, which contains the correct page offset.
Signed-off-by: Tao Liu <ltao@redhat.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
memory.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/memory.c b/memory.c
index 9ab578134fa1..1b6f9ba17e57 100644
--- a/memory.c
+++ b/memory.c
@@ -8861,7 +8861,7 @@ dump_vmlist(struct meminfo *vi)
(vi->spec_addr < (paddr+PAGESIZE()))) {
if (vi->flags & GET_PHYS_TO_VMALLOC) {
vi->retval = pcheck +
- PAGEOFFSET(paddr);
+ PAGEOFFSET(vi->spec_addr);
return;
} else
fprintf(fp,
@@ -9010,7 +9010,7 @@ dump_vmap_area(struct meminfo *vi)
(vi->spec_addr < (paddr+PAGESIZE()))) {
if (vi->flags & GET_PHYS_TO_VMALLOC) {
vi->retval = pcheck +
- PAGEOFFSET(paddr);
+ PAGEOFFSET(vi->spec_addr);
FREEBUF(ld->list_ptr);
return;
} else
--
2.37.1

76
0029-Let-kmem-print-task-context-with-physical-address.patch
View File

@ -1,76 +0,0 @@
From 3b5e3e1583a1f596360c04e8a322e30cf88f27ab Mon Sep 17 00:00:00 2001
From: Tao Liu <ltao@redhat.com>
Date: Mon, 19 Sep 2022 17:49:23 +0800
Subject: [PATCH 29/29] Let "kmem" print task context with physical address
Patch [1] enables "kmem" to print task context if the given virtual
address is a vmalloced stack.
This patch lets "kmem" print task context also when the given address
is a physical address.
Before:
crash> kmem 1883700e28
VMAP_AREA VM_STRUCT ADDRESS RANGE SIZE
ffff94eb9102c640 ffff94eb9102b140 ffffb7efce9b8000 - ffffb7efce9bd000 20480
PAGE PHYSICAL MAPPING INDEX CNT FLAGS
ffffdd28220dc000 1883700000 0 0 1 50000000000000
After:
crash> kmem 1883700e28
PID: 847
COMMAND: "khungtaskd"
TASK: ffff94f8038f4000 [THREAD_INFO: ffff94f8038f4000]
CPU: 72
STATE: TASK_RUNNING (PANIC)
VMAP_AREA VM_STRUCT ADDRESS RANGE SIZE
ffff94eb9102c640 ffff94eb9102b140 ffffb7efce9b8000 - ffffb7efce9bd000 20480
PAGE PHYSICAL MAPPING INDEX CNT FLAGS
ffffdd28220dc000 1883700000 0 0 1 50000000000000
[1]: https://listman.redhat.com/archives/crash-utility/2022-September/010115.html
[ kh: squashed the 4/4 patch into 3/4 ]
Signed-off-by: Tao Liu <ltao@redhat.com>
Signed-off-by: Kazuhito Hagio <k-hagio-ab@nec.com>
Signed-off-by: Lianbo Jiang <lijiang@redhat.com>
---
memory.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/memory.c b/memory.c
index 1b6f9ba17e57..c80ef61bdcf7 100644
--- a/memory.c
+++ b/memory.c
@@ -13506,6 +13506,10 @@ kmem_search(struct meminfo *mi)
mi->flags &= ~GET_PHYS_TO_VMALLOC;
if (mi->retval) {
+ if ((task = stkptr_to_task(mi->retval)) && (tc = task_to_context(task))) {
+ show_context(tc);
+ fprintf(fp, "\n");
+ }
if ((sp = value_search(mi->retval, &offset))) {
show_symbol(sp, offset,
SHOW_LINENUM | SHOW_RADIX());
@@ -13562,11 +13566,11 @@ kmem_search(struct meminfo *mi)
/*
* Check whether it's a current task or stack address.
*/
- if ((mi->memtype == KVADDR) && (task = vaddr_in_task_struct(vaddr)) &&
+ if ((mi->memtype & (KVADDR|PHYSADDR)) && (task = vaddr_in_task_struct(vaddr)) &&
(tc = task_to_context(task))) {
show_context(tc);
fprintf(fp, "\n");
- } else if ((mi->memtype == KVADDR) && (task = stkptr_to_task(vaddr)) &&
+ } else if ((mi->memtype & (KVADDR|PHYSADDR)) && (task = stkptr_to_task(vaddr)) &&
(tc = task_to_context(task))) {
show_context(tc);
fprintf(fp, "\n");
--
2.37.1

8
crash-8.0.1_build.patch → crash-8.0.2_build.patch
View File

@ -1,5 +1,5 @@
--- crash-8.0.1/Makefile.orig --- crash-8.0.2/Makefile.orig
+++ crash-8.0.1/Makefile +++ crash-8.0.2/Makefile
@@ -204,7 +204,7 @@ GDB_FLAGS= @@ -204,7 +204,7 @@ GDB_FLAGS=
# TARGET_CFLAGS will be configured automatically by configure # TARGET_CFLAGS will be configured automatically by configure
TARGET_CFLAGS= TARGET_CFLAGS=
@ -18,8 +18,8 @@
@echo "../../${PROGRAM} ../../${PROGRAM}lib.a" > ${GDB}/gdb/mergeobj @echo "../../${PROGRAM} ../../${PROGRAM}lib.a" > ${GDB}/gdb/mergeobj
@rm -f ${PROGRAM} @rm -f ${PROGRAM}
@if [ ! -f ${GDB}/config.status ]; then \ @if [ ! -f ${GDB}/config.status ]; then \
--- crash-8.0.1/configure.c.orig --- crash-8.0.2/configure.c.orig
+++ crash-8.0.1/configure.c +++ crash-8.0.2/configure.c
@@ -810,7 +810,8 @@ build_configure(struct supported_gdb_version *sp) @@ -810,7 +810,8 @@ build_configure(struct supported_gdb_version *sp)
fprintf(fp2, "%s\n", sp->GDB); fprintf(fp2, "%s\n", sp->GDB);
sprintf(target_data.gdb_version, "%s", &sp->GDB[4]); sprintf(target_data.gdb_version, "%s", &sp->GDB[4]);

105
crash.spec
View File

@ -3,8 +3,8 @@
# #
Summary: Kernel analysis utility for live systems, netdump, diskdump, kdump, LKCD or mcore dumpfiles Summary: Kernel analysis utility for live systems, netdump, diskdump, kdump, LKCD or mcore dumpfiles
Name: crash Name: crash
Version: 8.0.1 Version: 8.0.2
Release: 4%{?dist} Release: 1%{?dist}
License: GPLv3 License: GPLv3
Source0: https://github.com/crash-utility/crash/archive/crash-%{version}.tar.gz Source0: https://github.com/crash-utility/crash/archive/crash-%{version}.tar.gz
Source1: http://ftp.gnu.org/gnu/gdb/gdb-10.2.tar.gz Source1: http://ftp.gnu.org/gnu/gdb/gdb-10.2.tar.gz
@ -18,53 +18,8 @@ Requires: binutils
Provides: bundled(libiberty) Provides: bundled(libiberty)
Provides: bundled(gdb) = 10.2 Provides: bundled(gdb) = 10.2
Patch0: lzo_snappy_zstd.patch Patch0: lzo_snappy_zstd.patch
Patch1: crash-8.0.1_build.patch Patch1: crash-8.0.2_build.patch
Patch2: 0001-ppc64-update-the-NR_CPUS-to-8192.patch Patch2: crash-8.0.0-5-gdb-cdefs.patch
Patch3: 0002-sbitmapq-remove-struct-and-member-validation-in-sbit.patch
Patch4: 0003-sbitmapq-fix-invalid-offset-for-sbitmap_queue_alloc_.patch
Patch5: 0004-sbitmapq-fix-invalid-offset-for-sbitmap_queue_round_.patch
Patch6: 0005-sbitmapq-fix-invalid-offset-for-sbitmap_word_depth-o.patch
Patch7: 0006-Makefile-add-missing-crash_target.o-to-be-cleaned.patch
Patch8: 0007-bt-x86_64-filter-out-idle-task-stack.patch
Patch9: 0008-bt-arm64-add-support-for-bt-n-idle.patch
Patch10: 0009-gdb-print-details-of-unnamed-struct-and-union.patch
Patch11: 0010-Enhance-dev-d-D-options-to-support-blk-mq-sbitmap.patch
Patch12: 0011-Fix-for-dev-d-D-options-to-support-blk-mq-change-on-.patch
Patch13: 0012-Doc-update-man-page-for-the-bpf-and-sbitmapq-command.patch
Patch14: 0013-sbitmapq-Fix-for-sbitmap_queue-without-ws_active-mem.patch
Patch15: 0014-sbitmapq-Fix-for-sbitmap_word-without-cleared-member.patch
Patch16: 0015-sbitmapq-Fix-for-sbitmap_queue-without-min_shallow_d.patch
Patch17: 0016-Make-dev-d-D-options-parse-sbitmap-on-Linux-4.18-and.patch
Patch18: 0001-sbitmapq-Fix-for-kernels-without-struct-wait_queue_h.patch
Patch19: 0002-sbitmapq-Limit-kernels-without-sbitmap-again.patch
Patch20: 0003-Fix-for-dev-command-on-Linux-5.11-and-later.patch
Patch21: 0004-Extend-field-length-of-task-attributes.patch
Patch22: 0005-ppc64-fix-bt-for-S-case.patch
Patch23: 0006-ppc64-dynamically-allocate-h-w-interrupt-stack.patch
Patch24: 0007-ppc64-rename-ppc64_paca_init-to-ppc64_paca_percpu_of.patch
Patch25: 0008-ppc64-handle-backtrace-when-CPU-is-in-an-emergency-s.patch
Patch26: 0009-ppc64-print-emergency-stacks-info-with-mach-command.patch
Patch27: 0010-ppc64-use-a-variable-for-machdep-machspec.patch
Patch28: 0011-arm64-Fix-for-st-_stext_vmlinux-not-initialized-when.patch
Patch29: 0012-Fix-gcc-11-compiler-warnings-on-filesys.c.patch
Patch30: 0013-Fix-gcc-11-compiler-warning-on-symbols.c.patch
Patch31: 0014-Fix-gcc-11-compiler-warning-on-makedumpfile.c.patch
Patch32: 0015-Fix-gcc-11-compiler-warning-on-kvmdump.c.patch
Patch33: 0016-x86_64-Fix-for-AMD-SME-issue.patch
Patch34: 0017-Makefile-Fix-unnecessary-re-patching-with-coreutils-.patch
Patch35: 0018-gdb-fix-for-assigning-NULL-to-std-string.patch
Patch36: 0019-arm64-use-TCR_EL1_T1SZ-to-get-the-correct-info-if-va.patch
Patch37: 0020-Fix-task-R-by-adding-end-identifier-for-union-in-tas.patch
Patch38: 0021-Let-gdb-get-kernel-module-symbols-info-from-crash.patch
Patch39: 0022-x86_64-Correct-the-identifier-when-locating-the-call.patch
Patch40: 0023-Add-debian-ubuntu-vmlinux-location-to-default-search.patch
Patch41: 0024-Fix-gcc-12-compiler-warnings-on-lkcd_-.c.patch
Patch42: 0025-Fix-gcc-11-compiler-warnings-on-gdb-10.2-gdb-symtab..patch
Patch43: 0026-Fix-for-the-invalid-linux_banner-pointer-issue.patch
Patch44: 0027-Fix-kmem-failing-to-print-task-context-when-address-.patch
Patch45: 0028-Fix-page-offset-issue-when-converting-physical-to-vi.patch
Patch46: 0029-Let-kmem-print-task-context-with-physical-address.patch
Patch47: crash-8.0.0-5-gdb-cdefs.patch
%description %description
The core analysis suite is a self-contained tool that can be used to The core analysis suite is a self-contained tool that can be used to
@ -85,54 +40,9 @@ offered by Mission Critical Linux, or the LKCD kernel patch.
%prep %prep
%setup -n %{name}-%{version} -q %setup -n %{name}-%{version} -q
%patch0 -p1 -b lzo_snappy_zstd.patch %patch0 -p1 -b lzo_snappy_zstd.patch
%patch1 -p1 -b crash-8.0.1_build.patch %patch1 -p1 -b crash-8.0.2_build.patch
%patch2 -p1
%patch3 -p1
%patch4 -p1
%patch5 -p1
%patch6 -p1
%patch7 -p1
%patch8 -p1
%patch9 -p1
%patch10 -p1
%patch11 -p1
%patch12 -p1
%patch13 -p1
%patch14 -p1
%patch15 -p1
%patch16 -p1
%patch17 -p1
%patch18 -p1
%patch19 -p1
%patch20 -p1
%patch21 -p1
%patch22 -p1
%patch23 -p1
%patch24 -p1
%patch25 -p1
%patch26 -p1
%patch27 -p1
%patch28 -p1
%patch29 -p1
%patch30 -p1
%patch31 -p1
%patch32 -p1
%patch33 -p1
%patch34 -p1
%patch35 -p1
%patch36 -p1
%patch37 -p1
%patch38 -p1
%patch39 -p1
%patch40 -p1
%patch41 -p1
%patch42 -p1
%patch43 -p1
%patch44 -p1
%patch45 -p1
%patch46 -p1
%ifarch ppc64le %ifarch ppc64le
%patch47 -p1 -b crash-8.0.0-5-gdb-cdefs.patch %patch2 -p1 -b crash-8.0.0-5-gdb-cdefs.patch
%endif %endif
@ -160,6 +70,9 @@ cp -p defs.h %{buildroot}%{_includedir}/crash
%{_includedir}/* %{_includedir}/*
%changelog %changelog
* Thu Nov 17 2022 Lianbo Jiang <lijiang@redhat.com> - 8.0.2-1
- Rebase to upstream crash 8.0.2
* Thu Sep 22 2022 Lianbo Jiang <lijiang@redhat.com> - 8.0.1-4 * Thu Sep 22 2022 Lianbo Jiang <lijiang@redhat.com> - 8.0.1-4
- Update to the latest upstream commit <3b5e3e1583a1> - Update to the latest upstream commit <3b5e3e1583a1>

8
lzo_snappy_zstd.patch
View File

@ -1,5 +1,5 @@
--- crash-8.0.1/Makefile.orig --- crash-8.0.2/Makefile.orig
+++ crash-8.0.1/Makefile +++ crash-8.0.2/Makefile
@@ -256,7 +256,7 @@ all: make_configure @@ -256,7 +256,7 @@ all: make_configure
gdb_merge: force gdb_merge: force
@if [ ! -f ${GDB}/README ]; then \ @if [ ! -f ${GDB}/README ]; then \
@ -9,8 +9,8 @@
@echo "../../${PROGRAM} ../../${PROGRAM}lib.a" > ${GDB}/gdb/mergeobj @echo "../../${PROGRAM} ../../${PROGRAM}lib.a" > ${GDB}/gdb/mergeobj
@rm -f ${PROGRAM} @rm -f ${PROGRAM}
@if [ ! -f ${GDB}/config.status ]; then \ @if [ ! -f ${GDB}/config.status ]; then \
--- crash-8.0.1/diskdump.c.orig --- crash-8.0.2/diskdump.c.orig
+++ crash-8.0.1/diskdump.c +++ crash-8.0.2/diskdump.c
@@ -23,6 +23,9 @@ @@ -23,6 +23,9 @@
* GNU General Public License for more details. * GNU General Public License for more details.
*/ */

2
sources
View File

@ -1,2 +1,2 @@
SHA512 (crash-8.0.1.tar.gz) = ef9fe84dd5efa1b0570f71a8dd7af398a2ff35e9dc9fbcbeafd0f5ff503c7c4da93a33ffddfbac672fe12788dd3712808d6726ba9161ce282a7c76c3e2dc0793 SHA512 (crash-8.0.2.tar.gz) = 9ff24d1206e9376e83690f76c817a48a68ff6adce677fad70335a73550a59c9af6e4753c1199f22eafa60c137156313244bbf98ed01bc2b066f41d324738ef6b
SHA512 (gdb-10.2.tar.gz) = aa89caf47c1c84366020377d47e7c51ddbc48e5b7686f244e38797c8eb88411cf57fcdc37eb669961efb41ceeac4181747f429625fd1acce7712cb9a1fea9c41 SHA512 (gdb-10.2.tar.gz) = aa89caf47c1c84366020377d47e7c51ddbc48e5b7686f244e38797c8eb88411cf57fcdc37eb669961efb41ceeac4181747f429625fd1acce7712cb9a1fea9c41