xiaoxiang781216 merged PR #15603:
URL: https://github.com/apache/nuttx/pull/15603
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@nu
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2606195953
> @yamt thank you.
kudo for you to deal with a stubborn person like me patiently.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log o
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2606190664
@yamt thank you.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To u
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924629788
##
fs/vfs/fs_uio.c:
##
@@ -82,8 +82,8 @@ ssize_t uio_calc_resid(FAR const struct uio *uio)
void uio_advance(FAR struct uio *uio, size_t sz)
{
FAR const struct iovec
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924628918
##
fs/vfs/fs_uio.c:
##
@@ -82,8 +82,8 @@ ssize_t uio_calc_resid(FAR const struct uio *uio)
void uio_advance(FAR struct uio *uio, size_t sz)
{
FAR const struct iovec *
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924626091
##
fs/vfs/fs_uio.c:
##
@@ -82,8 +82,8 @@ ssize_t uio_calc_resid(FAR const struct uio *uio)
void uio_advance(FAR struct uio *uio, size_t sz)
{
FAR const struct iovec
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924624287
##
fs/vfs/fs_read.c:
##
@@ -50,19 +50,14 @@
*
/
-static ssize_t file_readv_compat(FAR str
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924621279
##
fs/vfs/fs_uio.c:
##
@@ -82,8 +82,8 @@ ssize_t uio_calc_resid(FAR const struct uio *uio)
void uio_advance(FAR struct uio *uio, size_t sz)
{
FAR const struct iovec *
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924617052
##
fs/vfs/fs_read.c:
##
@@ -50,19 +50,14 @@
*
/
-static ssize_t file_readv_compat(FAR s
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924613830
##
fs/vfs/fs_read.c:
##
@@ -74,36 +69,36 @@ static ssize_t file_readv_compat(FAR struct file *filep,
FAR struct uio *uio)
continue;
}
- buffe
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924600320
##
fs/vfs/fs_read.c:
##
@@ -74,36 +69,36 @@ static ssize_t file_readv_compat(FAR struct file *filep,
FAR struct uio *uio)
continue;
}
- buffer
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924599451
##
fs/vfs/fs_uio.c:
##
@@ -131,7 +131,7 @@ int uio_init(FAR struct uio *uio, FAR const struct iovec
*iov, int iovcnt)
resid = uio_calc_resid(uio);
if (resid < 0)
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924594260
##
fs/vfs/fs_read.c:
##
@@ -74,36 +69,36 @@ static ssize_t file_readv_compat(FAR struct file *filep,
FAR struct uio *uio)
continue;
}
- buffe
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924592962
##
fs/vfs/fs_read.c:
##
@@ -361,5 +352,6 @@ ssize_t read(int fd, FAR void *buf, size_t nbytes)
iov.iov_base = buf;
iov.iov_len = nbytes;
+
Review Comment:
do
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924592896
##
fs/vfs/fs_uio.c:
##
@@ -131,7 +131,7 @@ int uio_init(FAR struct uio *uio, FAR const struct iovec
*iov, int iovcnt)
resid = uio_calc_resid(uio);
if (resid < 0)
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924592489
##
fs/vfs/fs_uio.c:
##
@@ -82,8 +82,8 @@ ssize_t uio_calc_resid(FAR const struct uio *uio)
void uio_advance(FAR struct uio *uio, size_t sz)
{
FAR const struct iovec
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924584562
##
fs/vfs/fs_uio.c:
##
@@ -131,7 +131,7 @@ int uio_init(FAR struct uio *uio, FAR const struct iovec
*iov, int iovcnt)
resid = uio_calc_resid(uio);
if (resid < 0)
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924580628
##
fs/vfs/fs_read.c:
##
@@ -50,19 +50,14 @@
*
/
-static ssize_t file_readv_compat(FAR str
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2606106302
> > > > i guess the main disagreement between us is on how large the
overhead is. anchao seems to think it important. i feel it's negligible. none
of us provided any numbers. maybe i can do
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1924580628
##
fs/vfs/fs_read.c:
##
@@ -50,19 +50,14 @@
*
/
-static ssize_t file_readv_compat(FAR str
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2604645799
@yamt please review again
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comme
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1923537264
##
fs/vfs/fs_read.c:
##
@@ -126,18 +113,21 @@ static ssize_t file_readv_compat(FAR struct file *filep,
FAR struct uio *uio)
*
* Input Parameters:
* filep - Fil
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1923536754
##
fs/vfs/fs_read.c:
##
@@ -50,19 +50,14 @@
*
/
-static ssize_t file_readv_compat(FAR s
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1923535040
##
fs/vfs/fs_uio.c:
##
@@ -126,12 +126,14 @@ int uio_init(FAR struct uio *uio, FAR const struct iovec
*iov, int iovcnt)
ssize_t resid;
memset(uio, 0, sizeof(*ui
yamt commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1923468472
##
fs/vfs/fs_read.c:
##
@@ -126,18 +113,21 @@ static ssize_t file_readv_compat(FAR struct file *filep,
FAR struct uio *uio)
*
* Input Parameters:
* filep - File
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2604384099
> > > i guess the main disagreement between us is on how large the overhead
is. anchao seems to think it important. i feel it's negligible. none of us
provided any numbers. maybe i can do
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2604309102
> > i guess the main disagreement between us is on how large the overhead
is. anchao seems to think it important. i feel it's negligible. none of us
provided any numbers. maybe i can do some
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2604298860
> > > @yamt do you agree to merge this patch after discussing with @anchao ?
> >
> >
> > not really.
> > i guess the main disagreement between us is on how large the overhead
is
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2604258856
> i guess the main disagreement between us is on how large the overhead is.
anchao seems to think it important. i feel it's negligible. none of us provided
any numbers. maybe i can do
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2604177996
> > @yamt do you agree to merge this patch after discussing with @anchao ?
>
> not really.
>
> i guess the main disagreement between us is on how large the overhead is.
anchao
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2604001483
> @yamt do you agree to merge this patch after discussing with @anchao ?
not really.
i guess the main disagreement between us is on how large the overhead is.
anchao seems to
xiaoxiang781216 commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2603950728
@yamt do you agree to merge this patch after discussing with @anchao ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Gi
acassis commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2602386064
@anchao @yamt I think this solution is good, it avoid overhead for low end
MCUs and still working for MCUs that support readv/writev
--
This is an automated message from the Apache Git
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601778746
> i agree it's possible. however, i don't agree it's a good trade-off. IMO,
unified api matters more than a small overhead.
I think I may have convinced you a little? Let's go ba
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601667654
> > OK, my opinion has not changed, “uio could be skipped when reading and
writing a single vector buffer“, could we reach a consensus on this option?
>
> i agree it's possible.
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601639703
> > > >
[yamt@67c9bc6](https://github.com/yamt/incubator-nuttx/commit/67c9bc6f370eebb80a6a53d4466cd3da170dcf61)
(not tested at all)
> > >
> > >
> > >

(not tested at all)
> >
> >
> >

(not tested at all)
>
>

(not tested at all)
![image](https://github.com/user-attachments/asset
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601526890
> > * is it intended to drop the EINVAL check for non-iovec cases? why?
>
> In the uio case, you need to accumulate the length, but non-iovec not.
a single size_t value alone can
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601469267
> * is it intended to drop the EINVAL check for non-iovec cases? why?
In the uio case, you need to accumulate the length, but non-iovec not.
> * if you just want to save a func
yamt commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601455518
* is it intended to drop the EINVAL check for non-iovec cases? why?
* if you just want to save a function call for uio_init, i feel it's simpler
to make it a macro and inline the iovcnt ==
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601397384
@yamt we could review this PR first, just some uio logic adjustment
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
anchao commented on PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#issuecomment-2601242619
Point 1 to answer your question
https://github.com/apache/nuttx/pull/15604#issuecomment-2601238122
--
This is an automated message from the Apache Git Service.
To respond to the messag
anchao commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1921702377
##
fs/vfs/fs_read.c:
##
@@ -50,19 +50,16 @@
*
/
-static ssize_t file_readv_compat(FAR s
xiaoxiang781216 commented on code in PR #15603:
URL: https://github.com/apache/nuttx/pull/15603#discussion_r1921697274
##
fs/vfs/fs_write.c:
##
@@ -95,16 +93,6 @@ static ssize_t file_writev_compat(FAR struct file *filep,
{
return ntotal;
}
-
Review
46 matches
Mail list logo