On Tue, 3 Feb 2015, Mario Schuknecht wrote: > I used gadgetfs with PLX3380 controler. I had the problem that was > reported here: > > http://thread.gmane.org/gmane.linux.usb.general/116872/focus=117488 > > Perphaps this is also your problem. The commit that caused the error > is probably this: > > commit 7f7f25e82d54870df24d415a7007fbd327da027b > Author: Al Viro <viro@xxxxxxxxxxxxxxxxxx> > Date: Tue Feb 11 17:49:24 2014 -0500 > > replace checking for ->read/->aio_read presence with check in ->f_mode > > Since we are about to introduce new methods (read_iter/write_iter), the > tests in a bunch of places would have to grow inconveniently. Check > once (at open() time) and store results in ->f_mode as FMODE_CAN_READ > and FMODE_CAN_WRITE resp. It might end up being a temporary measure - > once everything switches from ->aio_{read,write} to ->{read,write}_iter > it might make sense to return to open-coded checks. We'll see... > > Signed-off-by: Al Viro <viro@xxxxxxxxxxxxxxxxxx> > > > If a driver register a read function, it is remembered in a flag in > the open function. > > > --- a/fs/open.c > +++ b/fs/open.c > @@ -725,6 +725,10 @@ static int do_dentry_open(struct file *f, > } > if ((f->f_mode & (FMODE_READ | FMODE_WRITE)) == FMODE_READ) > i_readcount_inc(inode); > + if ((f->f_mode & FMODE_READ) && likely(f->f_op->read || > f->f_op->aio_read)) > + f->f_mode |= FMODE_CAN_READ; > + if ((f->f_mode & FMODE_WRITE) && likely(f->f_op->write || > f->f_op->aio_write)) > + f->f_mode |= FMODE_CAN_WRITE; > > > And if the file is read this flag is checked. > > > --- a/fs/read_write.c > +++ b/fs/read_write.c > @@ -396,7 +396,7 @@ ssize_t vfs_read(struct file *file, char __user > *buf, size_t count, loff_t *pos) > > if (!(file->f_mode & FMODE_READ)) > return -EBADF; > - if (!file->f_op->read && !file->f_op->aio_read) > + if (!(file->f_mode & FMODE_CAN_READ)) > > > > I have fixed the problem for myself. Setting FMODE_CAN_READ flag after > change of file-pointer. > > > diff --git a/drivers/usb/gadget/legacy/inode.c > b/drivers/usb/gadget/legacy/inode.c > index e96077b..1b56bee 100644 > --- a/drivers/usb/gadget/legacy/inode.c > +++ b/drivers/usb/gadget/legacy/inode.c > @@ -857,6 +857,7 @@ ep_config (struct file *fd, const char __user > *buf, size_t len, loff_t *ptr) > } > if (value == 0) { > fd->f_op = &ep_io_operations; > + fd->f_mode |= FMODE_CAN_READ; > value = length; > } > gone: > @@ -1926,6 +1927,7 @@ dev_config (struct file *fd, const char __user > *buf, size_t len, loff_t *ptr) > * kick in after the ep0 descriptor is closed. > */ > fd->f_op = &ep0_io_operations; > + fd->f_mode |= FMODE_CAN_READ; > value = len; > } > return value; > -- > > It seems that there is no maintainer for the gadgetfs. It is located > under the folder "legacy". > Therefore I use the functionfs now, which has a similar functionality. > > > I hope I could help you. > > Mario You should submit a patch with your changes so that they will be available to everybody. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
