* a problem from blktap?
@ 2009-07-07 9:19 房海峰
2009-07-07 9:47 ` 房海峰
2009-07-07 16:49 ` jake wires
0 siblings, 2 replies; 3+ messages in thread
From: 房海峰 @ 2009-07-07 9:19 UTC (permalink / raw)
To: xen-devel; +Cc: fanghaifeng
[-- Attachment #1.1: Type: text/plain, Size: 11229 bytes --]
hi all,
currently, i am debuging the code derived from block-sync.c, the detail as follow:
=================================================================
struct socket_message_header *msg_hdr;
struct socketmsg_sync_read * msg_body;
int msglen, size;
char * msgbuf = msg;
int ret = 0;
uint64_t offset;
char * iobuf = NULL;
msg_hdr = (struct socket_message_header *)msgbuf;
msg_body = (struct socketmsg_sync_read *)(msgbuf + sizeof(struct socket_message_header));
msglen = msg_hdr->len;
size = msg_body->nb_sectors * msg_body-sector_size;
offset = msg_body->sector * (uint64_t)msg_body->sector_size;
MY_DPRINTF("READ: nb_sectors = %d, sector_size = %d, start_sector = %llu",
msg_body->nb_sectors,
msg_body->sector_size,
msg_body->sector);
//iobuf = calloc(size, sizeof(char));
iobuf = malloc(size);
if (!iobuf)
goto failure;
MY_DPRINTF("IO Buffer memory calloc is ok!");
ret = lseek(image_fd, offset, SEEK_SET);
if (ret != (off_t)-1) {
MY_DPRINTF("after lseek, the position is at %lu", (unsigned long) ret);
MY_DPRINTF("image_fd = %d, iobuf = %p, size = %d", image_fd, iobuf, size);
ret = readn(image_fd, (void*)iobuf, (size_t)size);
if (ret < 0) {
MY_DPRINTF("readn(%s) is error!", strerror(errno));
goto failure;
}
else
{
MY_DPRINTF("readn() is end, and has read [%d] data!", ret);
}
} else
{
MY_DPRINTF("lseek is error(%s)", strerror(errno));
goto failure;
}
MY_DPRINTF("Seek and Read image is ok!");
===================================================================================
the debuging output, which cut from /var/log/message, are as follow:
Jul 7 16:45:07 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(712)] loop for select(4)............
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(715)] select() is end, and ret = 1
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(720)] enter hook_read_message().
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(632)] read() is end, and msglen=32
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(651)] receive SOCK_MSG_READ message.
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(502)] READ: nb_sectors = 2, sector_size = 512, start_sector = 2
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(508)] IO Buffer memory calloc is ok!
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(513)] after lseek, the position is at 1024
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(514)] image_fd = 6, iobuf = 0x804f388, size = 1024
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(66)] call readn(fd[6], ptr[0x804f388], n[1024])
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(70)] call read(fd[6], ptr[0x804f388], nleft[1024])
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(73)] read is error (Invalid argument)
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(74)] nread = -1, nleft = 1024, n = 1024
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(518)] readn(Invalid argument) is error!
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(653)] hook_do_read() is end, and ret = -1
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(670)] do with message is error!
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(722)] out of hook_read_message().
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(712)] loop for select(5)............
Jul 7 16:45:21 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(715)] select() is end, and ret = 0
among the list above, you can find the error message about the "readn" function. to resovle this problem, i have try some methods, BUT all is failure.
varify the mode of opening the file, that is O_RDWR | O_DIRECT | O_LARGEFILE
So, who can help me to do with this problem?
read is error (Invalid argument) ?
thanks a lot!
2009-07-07
房海峰
[-- Attachment #1.2: Type: text/html, Size: 50868 bytes --]
[-- Attachment #2: Type: text/plain, Size: 138 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: a problem from blktap?
2009-07-07 9:19 a problem from blktap? 房海峰
@ 2009-07-07 9:47 ` 房海峰
2009-07-07 16:49 ` jake wires
1 sibling, 0 replies; 3+ messages in thread
From: 房海峰 @ 2009-07-07 9:47 UTC (permalink / raw)
To: 房海峰; +Cc: xen-devel
[-- Attachment #1.1: Type: text/plain, Size: 11486 bytes --]
when i open the large file by the mode of O_DIRECT, Can the malloc() be used?
2009-07-07
房海峰
发件人： 房海峰
发送时间： 2009-07-07 17:20:01
收件人： xen-devel
抄送： fanghaifeng
主题： [Xen-devel] a problem from blktap?
hi all,
currently, i am debuging the code derived from block-sync.c, the detail as follow:
=================================================================
struct socket_message_header *msg_hdr;
struct socketmsg_sync_read * msg_body;
int msglen, size;
char * msgbuf = msg;
int ret = 0;
uint64_t offset;
char * iobuf = NULL;
msg_hdr = (struct socket_message_header *)msgbuf;
msg_body = (struct socketmsg_sync_read *)(msgbuf + sizeof(struct socket_message_header));
msglen = msg_hdr->len;
size = msg_body->nb_sectors * msg_body-sector_size;
offset = msg_body->sector * (uint64_t)msg_body->sector_size;
MY_DPRINTF("READ: nb_sectors = %d, sector_size = %d, start_sector = %llu",
msg_body->nb_sectors,
msg_body->sector_size,
msg_body->sector);
//iobuf = calloc(size, sizeof(char));
iobuf = malloc(size);
if (!iobuf)
goto failure;
MY_DPRINTF("IO Buffer memory calloc is ok!");
ret = lseek(image_fd, offset, SEEK_SET);
if (ret != (off_t)-1) {
MY_DPRINTF("after lseek, the position is at %lu", (unsigned long) ret);
MY_DPRINTF("image_fd = %d, iobuf = %p, size = %d", image_fd, iobuf, size);
ret = readn(image_fd, (void*)iobuf, (size_t)size);
if (ret < 0) {
MY_DPRINTF("readn(%s) is error!", strerror(errno));
goto failure;
}
else
{
MY_DPRINTF("readn() is end, and has read [%d] data!", ret);
}
} else
{
MY_DPRINTF("lseek is error(%s)", strerror(errno));
goto failure;
}
MY_DPRINTF("Seek and Read image is ok!");
===================================================================================
the debuging output, which cut from /var/log/message, are as follow:
Jul 7 16:45:07 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(712)] loop for select(4)............
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(715)] select() is end, and ret = 1
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(720)] enter hook_read_message().
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(632)] read() is end, and msglen=32
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(651)] receive SOCK_MSG_READ message.
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(502)] READ: nb_sectors = 2, sector_size = 512, start_sector = 2
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(508)] IO Buffer memory calloc is ok!
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(513)] after lseek, the position is at 1024
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(514)] image_fd = 6, iobuf = 0x804f388, size = 1024
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(66)] call readn(fd[6], ptr[0x804f388], n[1024])
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(70)] call read(fd[6], ptr[0x804f388], nleft[1024])
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(73)] read is error (Invalid argument)
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(74)] nread = -1, nleft = 1024, n = 1024
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(518)] readn(Invalid argument) is error!
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(653)] hook_do_read() is end, and ret = -1
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(670)] do with message is error!
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(722)] out of hook_read_message().
Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(712)] loop for select(5)............
Jul 7 16:45:21 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(715)] select() is end, and ret = 0
among the list above, you can find the error message about the "readn" function. to resovle this problem, i have try some methods, BUT all is failure.
varify the mode of opening the file, that is O_RDWR | O_DIRECT | O_LARGEFILE
So, who can help me to do with this problem?
read is error (Invalid argument) ?
thanks a lot!
2009-07-07
房海峰
[-- Attachment #1.2: Type: text/html, Size: 53359 bytes --]
[-- Attachment #2: Type: text/plain, Size: 138 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: a problem from blktap?
2009-07-07 9:19 a problem from blktap? 房海峰
2009-07-07 9:47 ` 房海峰
@ 2009-07-07 16:49 ` jake wires
1 sibling, 0 replies; 3+ messages in thread
From: jake wires @ 2009-07-07 16:49 UTC (permalink / raw)
To: 房海峰; +Cc: xen-devel, fanghaifeng
[-- Attachment #1.1: Type: text/plain, Size: 5257 bytes --]
hi,
when you use O_DIRECT, buffers passed to read/write calls must be
sector-aligned. posix_memalign() will give you what you want.
jake
2009/7/7 房海峰 <fhf25@126.com>
> hi all,
>
> currently, i am debuging the code derived from block-sync.c, the detail as
> follow:
>
> =================================================================
>
> struct socket_message_header *msg_hdr;
>
> struct socketmsg_sync_read * msg_body;
>
>
>
> int msglen, size;
>
> char * msgbuf = msg;
>
> int ret = 0;
>
> uint64_t offset;
>
>
>
> char * iobuf = NULL;
>
>
>
> msg_hdr = (struct socket_message_header *)msgbuf;
>
> msg_body = (struct socketmsg_sync_read *)(msgbuf + sizeof(struct socket_message_header));
>
> msglen = msg_hdr->len;
>
>
> size = msg_body->nb_sectors * msg_body-sector_size;
>
> offset = msg_body->sector * (uint64_t)msg_body->sector_size;
>
>
>
> MY_DPRINTF("READ: nb_sectors = %d, sector_size = %d, start_sector = %llu",
>
> msg_body->nb_sectors,
>
> msg_body->sector_size,
>
> msg_body->sector);
>
>
>
> //iobuf = calloc(size, sizeof(char));
>
> iobuf = malloc(size);
>
> if (!iobuf)
>
> goto failure;
>
> MY_DPRINTF("IO Buffer memory calloc is ok!");
>
>
>
> ret = lseek(image_fd, offset, SEEK_SET);
>
> if (ret != (off_t)-1) {
>
>
>
> MY_DPRINTF("after lseek, the position is at %lu", (unsigned long) ret);
>
> MY_DPRINTF("image_fd = %d, iobuf = %p, size = %d", image_fd, iobuf, size);
>
>
>
> ret = readn(image_fd, (void*)iobuf, (size_t)size);
>
> if (ret < 0) {
>
> MY_DPRINTF("readn(%s) is error!", strerror(errno));
>
> goto failure;
>
> }
>
> else
>
> {
>
> MY_DPRINTF("readn() is end, and has read [%d] data!", ret);
>
> }
>
> } else
>
> {
>
> MY_DPRINTF("lseek is error(%s)", strerror(errno));
>
> goto failure;
>
> }
>
> MY_DPRINTF("Seek and Read image is ok!");
>
>
> ===================================================================================
> the debuging output, which cut from /var/log/message, are as follow:
>
>
> Jul 7 16:45:07 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(712)] loop for select(4)............
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(715)] select() is end, and ret = 1
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(720)] enter hook_read_message().
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(632)] read() is end, and msglen=32
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(651)] receive SOCK_MSG_READ message.
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(502)] READ: nb_sectors = 2, sector_size = 512, start_sector = 2
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(508)] IO Buffer memory calloc is ok!
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(513)] after lseek, the position is at 1024
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(514)] image_fd = 6, iobuf = 0x804f388, size = 1024
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(66)] call readn(fd[6], ptr[0x804f388], n[1024])
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(70)] call read(fd[6], ptr[0x804f388], nleft[1024])
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(73)] read is error (Invalid argument)
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [readn(74)] nread = -1, nleft = 1024, n = 1024
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_do_read(518)] readn(Invalid argument) is error!
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(653)] hook_do_read() is end, and ret = -1
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_read_message(670)] do with message is error!
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(722)] out of hook_read_message().
>
> Jul 7 16:45:11 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(712)] loop for select(5)............
>
> Jul 7 16:45:21 domain0 remote-blktapctrl[3963]: [hook_new_tapdisk(715)] select() is end, and ret = 0
>
> among the list above, you can find the error message about the "readn"
> function. to resovle this problem, i have try some methods, BUT all is
> failure.
>
> varify the mode of opening the file, that is O_RDWR | O_DIRECT |
> O_LARGEFILE
>
> So, who can help me to do with this problem?
>
> read is error (Invalid argument) ?
>
>
> thanks a lot!
>
>
> 2009-07-07
> ------------------------------
> 房海峰
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>
>
[-- Attachment #1.2: Type: text/html, Size: 51445 bytes --]
[-- Attachment #2: Type: text/plain, Size: 138 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
^ permalink raw reply [flat|nested] 3+ messages in threadend of thread, other threads:[~2009-07-07 16:49 UTC | newest] Thread overview: 3+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2009-07-07 9:19 a problem from blktap? 房海峰 2009-07-07 9:47 ` 房海峰 2009-07-07 16:49 ` jake wires
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.