[Qemu-devel] [PULL 0/2] Block patches
[PULL for-6.1 0/3] Block patches
1
The following changes since commit 00928a421d47f49691cace1207481b7aad31b1f1:
1
The following changes since commit 3521ade3510eb5cefb2e27a101667f25dad89935:
2
2
3
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20180626' into staging (2018-06-26 18:23:49 +0100)
3
Merge remote-tracking branch 'remotes/thuth-gitlab/tags/pull-request-2021-07-29' into staging (2021-07-29 13:17:20 +0100)
4
4
5
are available in the Git repository at:
5
are available in the Git repository at:
6
6
7
git://github.com/stefanha/qemu.git tags/block-pull-request
7
https://gitlab.com/stefanha/qemu.git tags/block-pull-request
8
8
9
for you to fetch changes up to ed6e2161715c527330f936d44af4c547f25f687e:
9
for you to fetch changes up to cc8eecd7f105a1dff5876adeb238a14696061a4a:
10
10
11
linux-aio: properly bubble up errors from initialization (2018-06-27 13:06:34 +0100)
11
MAINTAINERS: Added myself as a reviewer for the NVMe Block Driver (2021-07-29 17:17:34 +0100)
12
12
13
----------------------------------------------------------------
13
----------------------------------------------------------------
14
Pull request
14
Pull request
15
15
16
* Gracefully handle Linux AIO init failure
16
The main fix here is for io_uring. Spurious -EAGAIN errors can happen and the
17
request needs to be resubmitted.
18
19
The MAINTAINERS changes carry no risk and we might as well include them in QEMU
20
6.1.
17
21
18
----------------------------------------------------------------
22
----------------------------------------------------------------
19
23
20
Nishanth Aravamudan (1):
24
Fabian Ebner (1):
21
linux-aio: properly bubble up errors from initialization
25
block/io_uring: resubmit when result is -EAGAIN
22
26
23
Stefan Hajnoczi (1):
27
Philippe Mathieu-Daudé (1):
24
compiler: add a sizeof_field() macro
28
MAINTAINERS: Added myself as a reviewer for the NVMe Block Driver
25
29
26
include/block/aio.h | 3 +++
30
Stefano Garzarella (1):
27
include/block/raw-aio.h | 2 +-
31
MAINTAINERS: add Stefano Garzarella as io_uring reviewer
28
include/hw/xen/io/ring.h | 2 +-
32
29
include/qemu/compiler.h | 2 ++
33
MAINTAINERS | 2 ++
30
accel/tcg/translate-all.c | 2 +-
34
block/io_uring.c | 16 +++++++++++++++-
31
block/file-posix.c | 33 ++++++++++++++++++++++++++++-----
35
2 files changed, 17 insertions(+), 1 deletion(-)
32
block/linux-aio.c | 12 +++++++++---
33
hw/display/xenfb.c | 4 ++--
34
hw/net/rocker/rocker_of_dpa.c | 2 +-
35
hw/net/virtio-net.c | 2 +-
36
stubs/linux-aio.c | 2 +-
37
target/i386/kvm.c | 2 +-
38
target/ppc/arch_dump.c | 10 +++++-----
39
target/s390x/arch_dump.c | 20 ++++++++++----------
40
util/async.c | 14 +++++++++++---
41
15 files changed, 77 insertions(+), 35 deletions(-)
42
36
43
--
37
--
44
2.17.1
38
2.31.1
45
39
46
diff view generated by jsdifflib
New patch
[PULL for-6.1 1/3] MAINTAINERS: add Stefano Garzarella as io_uring reviewer
1
From: Stefano Garzarella <sgarzare@redhat.com>
1
2
3
I've been working with io_uring for a while so I'd like to help
4
with reviews.
5
6
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
7
Message-Id: <20210728131515.131045-1-sgarzare@redhat.com>
8
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9
---
10
MAINTAINERS | 1 +
11
1 file changed, 1 insertion(+)
12
13
diff --git a/MAINTAINERS b/MAINTAINERS
14
index XXXXXXX..XXXXXXX 100644
15
--- a/MAINTAINERS
16
+++ b/MAINTAINERS
17
@@ -XXX,XX +XXX,XX @@ Linux io_uring
18
M: Aarushi Mehta <mehta.aaru20@gmail.com>
19
M: Julia Suvorova <jusual@redhat.com>
20
M: Stefan Hajnoczi <stefanha@redhat.com>
21
+R: Stefano Garzarella <sgarzare@redhat.com>
22
L: qemu-block@nongnu.org
23
S: Maintained
24
F: block/io_uring.c
25
--
26
2.31.1
27
diff view generated by jsdifflib
[Qemu-devel] [PULL 2/2] linux-aio: properly bubble up errors from initialization
[PULL for-6.1 2/3] block/io_uring: resubmit when result is -EAGAIN
1
From: Nishanth Aravamudan <naravamudan@digitalocean.com>
1
From: Fabian Ebner <f.ebner@proxmox.com>
2
2
3
laio_init() can fail for a couple of reasons, which will lead to a NULL
3
Linux SCSI can throw spurious -EAGAIN in some corner cases in its
4
pointer dereference in laio_attach_aio_context().
4
completion path, which will end up being the result in the completed
5
io_uring request.
5
6
6
To solve this, add a aio_setup_linux_aio() function which is called
7
Resubmitting such requests should allow block jobs to complete, even
7
early in raw_open_common. If this fails, propagate the error up. The
8
if such spurious errors are encountered.
8
signature of aio_get_linux_aio() was not modified, because it seems
9
preferable to return the actual errno from the possible failing
10
initialization calls.
11
9
12
Additionally, when the AioContext changes, we need to associate a
10
Co-authored-by: Stefan Hajnoczi <stefanha@gmail.com>
13
LinuxAioState with the new AioContext. Use the bdrv_attach_aio_context
11
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
14
callback and call the new aio_setup_linux_aio(), which will allocate a
12
Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
15
new AioContext if needed, and return errors on failures. If it fails for
13
Message-id: 20210729091029.65369-1-f.ebner@proxmox.com
16
any reason, fallback to threaded AIO with an error message, as the
17
device is already in-use by the guest.
18
19
Add an assert that aio_get_linux_aio() cannot return NULL.
20
21
Signed-off-by: Nishanth Aravamudan <naravamudan@digitalocean.com>
22
Message-id: 20180622193700.6523-1-naravamudan@digitalocean.com
23
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
14
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
24
---
15
---
25
include/block/aio.h | 3 +++
16
block/io_uring.c | 16 +++++++++++++++-
26
include/block/raw-aio.h | 2 +-
17
1 file changed, 15 insertions(+), 1 deletion(-)
27
block/file-posix.c | 33 ++++++++++++++++++++++++++++-----
28
block/linux-aio.c | 12 +++++++++---
29
stubs/linux-aio.c | 2 +-
30
util/async.c | 14 +++++++++++---
31
6 files changed, 53 insertions(+), 13 deletions(-)
32
18
33
diff --git a/include/block/aio.h b/include/block/aio.h
19
diff --git a/block/io_uring.c b/block/io_uring.c
34
index XXXXXXX..XXXXXXX 100644
20
index XXXXXXX..XXXXXXX 100644
35
--- a/include/block/aio.h
21
--- a/block/io_uring.c
36
+++ b/include/block/aio.h
22
+++ b/block/io_uring.c
37
@@ -XXX,XX +XXX,XX @@ GSource *aio_get_g_source(AioContext *ctx);
23
@@ -XXX,XX +XXX,XX @@ static void luring_process_completions(LuringState *s)
38
/* Return the ThreadPool bound to this AioContext */
24
total_bytes = ret + luringcb->total_read;
39
struct ThreadPool *aio_get_thread_pool(AioContext *ctx);
25
40
26
if (ret < 0) {
41
+/* Setup the LinuxAioState bound to this AioContext */
27
- if (ret == -EINTR) {
42
+struct LinuxAioState *aio_setup_linux_aio(AioContext *ctx, Error **errp);
28
+ /*
43
+
29
+ * Only writev/readv/fsync requests on regular files or host block
44
/* Return the LinuxAioState bound to this AioContext */
30
+ * devices are submitted. Therefore -EAGAIN is not expected but it's
45
struct LinuxAioState *aio_get_linux_aio(AioContext *ctx);
31
+ * known to happen sometimes with Linux SCSI. Submit again and hope
46
32
+ * the request completes successfully.
47
diff --git a/include/block/raw-aio.h b/include/block/raw-aio.h
33
+ *
48
index XXXXXXX..XXXXXXX 100644
34
+ * For more information, see:
49
--- a/include/block/raw-aio.h
35
+ * https://lore.kernel.org/io-uring/20210727165811.284510-3-axboe@kernel.dk/T/#u
50
+++ b/include/block/raw-aio.h
36
+ *
51
@@ -XXX,XX +XXX,XX @@
37
+ * If the code is changed to submit other types of requests in the
52
/* linux-aio.c - Linux native implementation */
38
+ * future, then this workaround may need to be extended to deal with
53
#ifdef CONFIG_LINUX_AIO
39
+ * genuine -EAGAIN results that should not be resubmitted
54
typedef struct LinuxAioState LinuxAioState;
40
+ * immediately.
55
-LinuxAioState *laio_init(void);
41
+ */
56
+LinuxAioState *laio_init(Error **errp);
42
+ if (ret == -EINTR || ret == -EAGAIN) {
57
void laio_cleanup(LinuxAioState *s);
43
luring_resubmit(s, luringcb);
58
int coroutine_fn laio_co_submit(BlockDriverState *bs, LinuxAioState *s, int fd,
44
continue;
59
uint64_t offset, QEMUIOVector *qiov, int type);
45
}
60
diff --git a/block/file-posix.c b/block/file-posix.c
61
index XXXXXXX..XXXXXXX 100644
62
--- a/block/file-posix.c
63
+++ b/block/file-posix.c
64
@@ -XXX,XX +XXX,XX @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
65
66
#ifdef CONFIG_LINUX_AIO
67
/* Currently Linux does AIO only for files opened with O_DIRECT */
68
- if (s->use_linux_aio && !(s->open_flags & O_DIRECT)) {
69
- error_setg(errp, "aio=native was specified, but it requires "
70
- "cache.direct=on, which was not specified.");
71
- ret = -EINVAL;
72
- goto fail;
73
+ if (s->use_linux_aio) {
74
+ if (!(s->open_flags & O_DIRECT)) {
75
+ error_setg(errp, "aio=native was specified, but it requires "
76
+ "cache.direct=on, which was not specified.");
77
+ ret = -EINVAL;
78
+ goto fail;
79
+ }
80
+ if (!aio_setup_linux_aio(bdrv_get_aio_context(bs), errp)) {
81
+ error_prepend(errp, "Unable to use native AIO: ");
82
+ goto fail;
83
+ }
84
}
85
#else
86
if (s->use_linux_aio) {
87
@@ -XXX,XX +XXX,XX @@ static BlockAIOCB *raw_aio_flush(BlockDriverState *bs,
88
return paio_submit(bs, s->fd, 0, NULL, 0, cb, opaque, QEMU_AIO_FLUSH);
89
}
90
91
+static void raw_aio_attach_aio_context(BlockDriverState *bs,
92
+ AioContext *new_context)
93
+{
94
+#ifdef CONFIG_LINUX_AIO
95
+ BDRVRawState *s = bs->opaque;
96
+ if (s->use_linux_aio) {
97
+ Error *local_err;
98
+ if (!aio_setup_linux_aio(new_context, &local_err)) {
99
+ error_reportf_err(local_err, "Unable to use native AIO, "
100
+ "falling back to thread pool: ");
101
+ s->use_linux_aio = false;
102
+ }
103
+ }
104
+#endif
105
+}
106
+
107
static void raw_close(BlockDriverState *bs)
108
{
109
BDRVRawState *s = bs->opaque;
110
@@ -XXX,XX +XXX,XX @@ BlockDriver bdrv_file = {
111
.bdrv_refresh_limits = raw_refresh_limits,
112
.bdrv_io_plug = raw_aio_plug,
113
.bdrv_io_unplug = raw_aio_unplug,
114
+ .bdrv_attach_aio_context = raw_aio_attach_aio_context,
115
116
.bdrv_truncate = raw_truncate,
117
.bdrv_getlength = raw_getlength,
118
diff --git a/block/linux-aio.c b/block/linux-aio.c
119
index XXXXXXX..XXXXXXX 100644
120
--- a/block/linux-aio.c
121
+++ b/block/linux-aio.c
122
@@ -XXX,XX +XXX,XX @@
123
#include "block/raw-aio.h"
124
#include "qemu/event_notifier.h"
125
#include "qemu/coroutine.h"
126
+#include "qapi/error.h"
127
128
#include <libaio.h>
129
130
@@ -XXX,XX +XXX,XX @@ void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context)
131
qemu_laio_poll_cb);
132
}
133
134
-LinuxAioState *laio_init(void)
135
+LinuxAioState *laio_init(Error **errp)
136
{
137
+ int rc;
138
LinuxAioState *s;
139
140
s = g_malloc0(sizeof(*s));
141
- if (event_notifier_init(&s->e, false) < 0) {
142
+ rc = event_notifier_init(&s->e, false);
143
+ if (rc < 0) {
144
+ error_setg_errno(errp, -rc, "failed to to initialize event notifier");
145
goto out_free_state;
146
}
147
148
- if (io_setup(MAX_EVENTS, &s->ctx) != 0) {
149
+ rc = io_setup(MAX_EVENTS, &s->ctx);
150
+ if (rc < 0) {
151
+ error_setg_errno(errp, -rc, "failed to create linux AIO context");
152
goto out_close_efd;
153
}
154
155
diff --git a/stubs/linux-aio.c b/stubs/linux-aio.c
156
index XXXXXXX..XXXXXXX 100644
157
--- a/stubs/linux-aio.c
158
+++ b/stubs/linux-aio.c
159
@@ -XXX,XX +XXX,XX @@ void laio_attach_aio_context(LinuxAioState *s, AioContext *new_context)
160
abort();
161
}
162
163
-LinuxAioState *laio_init(void)
164
+LinuxAioState *laio_init(Error **errp)
165
{
166
abort();
167
}
168
diff --git a/util/async.c b/util/async.c
169
index XXXXXXX..XXXXXXX 100644
170
--- a/util/async.c
171
+++ b/util/async.c
172
@@ -XXX,XX +XXX,XX @@ ThreadPool *aio_get_thread_pool(AioContext *ctx)
173
}
174
175
#ifdef CONFIG_LINUX_AIO
176
-LinuxAioState *aio_get_linux_aio(AioContext *ctx)
177
+LinuxAioState *aio_setup_linux_aio(AioContext *ctx, Error **errp)
178
{
179
if (!ctx->linux_aio) {
180
- ctx->linux_aio = laio_init();
181
- laio_attach_aio_context(ctx->linux_aio, ctx);
182
+ ctx->linux_aio = laio_init(errp);
183
+ if (ctx->linux_aio) {
184
+ laio_attach_aio_context(ctx->linux_aio, ctx);
185
+ }
186
}
187
return ctx->linux_aio;
188
}
189
+
190
+LinuxAioState *aio_get_linux_aio(AioContext *ctx)
191
+{
192
+ assert(ctx->linux_aio);
193
+ return ctx->linux_aio;
194
+}
195
#endif
196
197
void aio_notify(AioContext *ctx)
198
--
46
--
199
2.17.1
47
2.31.1
200
48
201
diff view generated by jsdifflib
[Qemu-devel] [PULL 1/2] compiler: add a sizeof_field() macro
[PULL for-6.1 3/3] MAINTAINERS: Added myself as a reviewer for the NVMe Block Driver
1
Determining the size of a field is useful when you don't have a struct
1
From: Philippe Mathieu-Daudé <philmd@redhat.com>
2
variable handy. Open-coding this is ugly.
3
2
4
This patch adds the sizeof_field() macro, which is similar to
3
I'm interested in following the activity around the NVMe bdrv.
5
typeof_field(). Existing instances are updated to use the macro.
6
4
7
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
5
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
8
Reviewed-by: John Snow <jsnow@redhat.com>
6
Message-id: 20210728183340.2018313-1-philmd@redhat.com
9
Message-id: 20180614164431.29305-1-stefanha@redhat.com
10
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
7
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
11
---
8
---
12
include/hw/xen/io/ring.h | 2 +-
9
MAINTAINERS | 1 +
13
include/qemu/compiler.h | 2 ++
10
1 file changed, 1 insertion(+)
14
accel/tcg/translate-all.c | 2 +-
15
hw/display/xenfb.c | 4 ++--
16
hw/net/rocker/rocker_of_dpa.c | 2 +-
17
hw/net/virtio-net.c | 2 +-
18
target/i386/kvm.c | 2 +-
19
target/ppc/arch_dump.c | 10 +++++-----
20
target/s390x/arch_dump.c | 20 ++++++++++----------
21
9 files changed, 24 insertions(+), 22 deletions(-)
22
11
23
diff --git a/include/hw/xen/io/ring.h b/include/hw/xen/io/ring.h
12
diff --git a/MAINTAINERS b/MAINTAINERS
24
index XXXXXXX..XXXXXXX 100644
13
index XXXXXXX..XXXXXXX 100644
25
--- a/include/hw/xen/io/ring.h
14
--- a/MAINTAINERS
26
+++ b/include/hw/xen/io/ring.h
15
+++ b/MAINTAINERS
27
@@ -XXX,XX +XXX,XX @@ typedef unsigned int RING_IDX;
16
@@ -XXX,XX +XXX,XX @@ F: block/null.c
28
*/
17
NVMe Block Driver
29
#define __CONST_RING_SIZE(_s, _sz) \
18
M: Stefan Hajnoczi <stefanha@redhat.com>
30
(__RD32(((_sz) - offsetof(struct _s##_sring, ring)) / \
19
R: Fam Zheng <fam@euphon.net>
31
-     sizeof(((struct _s##_sring *)0)->ring[0])))
20
+R: Philippe Mathieu-Daudé <philmd@redhat.com>
32
+ sizeof_field(struct _s##_sring, ring[0])))
21
L: qemu-block@nongnu.org
33
/*
22
S: Supported
34
* The same for passing in an actual pointer instead of a name tag.
23
F: block/nvme*
35
*/
36
diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
37
index XXXXXXX..XXXXXXX 100644
38
--- a/include/qemu/compiler.h
39
+++ b/include/qemu/compiler.h
40
@@ -XXX,XX +XXX,XX @@
41
(type *) ((char *) __mptr - offsetof(type, member));})
42
#endif
43
44
+#define sizeof_field(type, field) sizeof(((type *)0)->field)
45
+
46
/* Convert from a base type to a parent type, with compile time checking. */
47
#ifdef __GNUC__
48
#define DO_UPCAST(type, field, dev) ( __extension__ ( { \
49
diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
50
index XXXXXXX..XXXXXXX 100644
51
--- a/accel/tcg/translate-all.c
52
+++ b/accel/tcg/translate-all.c
53
@@ -XXX,XX +XXX,XX @@ struct page_collection {
54
55
/* Make sure all possible CPU event bits fit in tb->trace_vcpu_dstate */
56
QEMU_BUILD_BUG_ON(CPU_TRACE_DSTATE_MAX_EVENTS >
57
- sizeof(((TranslationBlock *)0)->trace_vcpu_dstate)
58
+ sizeof_field(TranslationBlock, trace_vcpu_dstate)
59
* BITS_PER_BYTE);
60
61
/*
62
diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c
63
index XXXXXXX..XXXXXXX 100644
64
--- a/hw/display/xenfb.c
65
+++ b/hw/display/xenfb.c
66
@@ -XXX,XX +XXX,XX @@ static int xenfb_configure_fb(struct XenFB *xenfb, size_t fb_len_lim,
67
int width, int height, int depth,
68
size_t fb_len, int offset, int row_stride)
69
{
70
- size_t mfn_sz = sizeof(*((struct xenfb_page *)0)->pd);
71
- size_t pd_len = sizeof(((struct xenfb_page *)0)->pd) / mfn_sz;
72
+ size_t mfn_sz = sizeof_field(struct xenfb_page, pd[0]);
73
+ size_t pd_len = sizeof_field(struct xenfb_page, pd) / mfn_sz;
74
size_t fb_pages = pd_len * XC_PAGE_SIZE / mfn_sz;
75
size_t fb_len_max = fb_pages * XC_PAGE_SIZE;
76
int max_width, max_height;
77
diff --git a/hw/net/rocker/rocker_of_dpa.c b/hw/net/rocker/rocker_of_dpa.c
78
index XXXXXXX..XXXXXXX 100644
79
--- a/hw/net/rocker/rocker_of_dpa.c
80
+++ b/hw/net/rocker/rocker_of_dpa.c
81
@@ -XXX,XX +XXX,XX @@ typedef struct of_dpa_flow_key {
82
83
/* Width of key which includes field 'f' in u64s, rounded up */
84
#define FLOW_KEY_WIDTH(f) \
85
- DIV_ROUND_UP(offsetof(OfDpaFlowKey, f) + sizeof(((OfDpaFlowKey *)0)->f), \
86
+ DIV_ROUND_UP(offsetof(OfDpaFlowKey, f) + sizeof_field(OfDpaFlowKey, f), \
87
sizeof(uint64_t))
88
89
typedef struct of_dpa_flow_action {
90
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
91
index XXXXXXX..XXXXXXX 100644
92
--- a/hw/net/virtio-net.c
93
+++ b/hw/net/virtio-net.c
94
@@ -XXX,XX +XXX,XX @@
95
* 'container'.
96
*/
97
#define endof(container, field) \
98
- (offsetof(container, field) + sizeof(((container *)0)->field))
99
+ (offsetof(container, field) + sizeof_field(container, field))
100
101
typedef struct VirtIOFeature {
102
uint64_t flags;
103
diff --git a/target/i386/kvm.c b/target/i386/kvm.c
104
index XXXXXXX..XXXXXXX 100644
105
--- a/target/i386/kvm.c
106
+++ b/target/i386/kvm.c
107
@@ -XXX,XX +XXX,XX @@ static int kvm_put_fpu(X86CPU *cpu)
108
#define XSAVE_PKRU 672
109
110
#define XSAVE_BYTE_OFFSET(word_offset) \
111
- ((word_offset) * sizeof(((struct kvm_xsave *)0)->region[0]))
112
+ ((word_offset) * sizeof_field(struct kvm_xsave, region[0]))
113
114
#define ASSERT_OFFSET(word_offset, field) \
115
QEMU_BUILD_BUG_ON(XSAVE_BYTE_OFFSET(word_offset) != \
116
diff --git a/target/ppc/arch_dump.c b/target/ppc/arch_dump.c
117
index XXXXXXX..XXXXXXX 100644
118
--- a/target/ppc/arch_dump.c
119
+++ b/target/ppc/arch_dump.c
120
@@ -XXX,XX +XXX,XX @@ static const struct NoteFuncDescStruct {
121
int contents_size;
122
void (*note_contents_func)(NoteFuncArg *arg, PowerPCCPU *cpu);
123
} note_func[] = {
124
- {sizeof(((Note *)0)->contents.prstatus), ppc_write_elf_prstatus},
125
- {sizeof(((Note *)0)->contents.fpregset), ppc_write_elf_fpregset},
126
- {sizeof(((Note *)0)->contents.vmxregset), ppc_write_elf_vmxregset},
127
- {sizeof(((Note *)0)->contents.vsxregset), ppc_write_elf_vsxregset},
128
- {sizeof(((Note *)0)->contents.speregset), ppc_write_elf_speregset},
129
+ {sizeof_field(Note, contents.prstatus), ppc_write_elf_prstatus},
130
+ {sizeof_field(Note, contents.fpregset), ppc_write_elf_fpregset},
131
+ {sizeof_field(Note, contents.vmxregset), ppc_write_elf_vmxregset},
132
+ {sizeof_field(Note, contents.vsxregset), ppc_write_elf_vsxregset},
133
+ {sizeof_field(Note, contents.speregset), ppc_write_elf_speregset},
134
{ 0, NULL}
135
};
136
137
diff --git a/target/s390x/arch_dump.c b/target/s390x/arch_dump.c
138
index XXXXXXX..XXXXXXX 100644
139
--- a/target/s390x/arch_dump.c
140
+++ b/target/s390x/arch_dump.c
141
@@ -XXX,XX +XXX,XX @@ typedef struct NoteFuncDescStruct {
142
} NoteFuncDesc;
143
144
static const NoteFuncDesc note_core[] = {
145
- {sizeof(((Note *)0)->contents.prstatus), s390x_write_elf64_prstatus},
146
- {sizeof(((Note *)0)->contents.fpregset), s390x_write_elf64_fpregset},
147
+ {sizeof_field(Note, contents.prstatus), s390x_write_elf64_prstatus},
148
+ {sizeof_field(Note, contents.fpregset), s390x_write_elf64_fpregset},
149
{ 0, NULL}
150
};
151
152
static const NoteFuncDesc note_linux[] = {
153
- {sizeof(((Note *)0)->contents.prefix), s390x_write_elf64_prefix},
154
- {sizeof(((Note *)0)->contents.ctrs), s390x_write_elf64_ctrs},
155
- {sizeof(((Note *)0)->contents.timer), s390x_write_elf64_timer},
156
- {sizeof(((Note *)0)->contents.todcmp), s390x_write_elf64_todcmp},
157
- {sizeof(((Note *)0)->contents.todpreg), s390x_write_elf64_todpreg},
158
- {sizeof(((Note *)0)->contents.vregslo), s390x_write_elf64_vregslo},
159
- {sizeof(((Note *)0)->contents.vregshi), s390x_write_elf64_vregshi},
160
- {sizeof(((Note *)0)->contents.gscb), s390x_write_elf64_gscb},
161
+ {sizeof_field(Note, contents.prefix), s390x_write_elf64_prefix},
162
+ {sizeof_field(Note, contents.ctrs), s390x_write_elf64_ctrs},
163
+ {sizeof_field(Note, contents.timer), s390x_write_elf64_timer},
164
+ {sizeof_field(Note, contents.todcmp), s390x_write_elf64_todcmp},
165
+ {sizeof_field(Note, contents.todpreg), s390x_write_elf64_todpreg},
166
+ {sizeof_field(Note, contents.vregslo), s390x_write_elf64_vregslo},
167
+ {sizeof_field(Note, contents.vregshi), s390x_write_elf64_vregshi},
168
+ {sizeof_field(Note, contents.gscb), s390x_write_elf64_gscb},
169
{ 0, NULL}
170
};
171
172
--
24
--
173
2.17.1
25
2.31.1
174
26
175
diff view generated by jsdifflib

Patchew 2.1-devel-b67e4c2

© 2016 - 2025 Red Hat, Inc.