On Mon, Feb 18, 2013 at 4:41 PM, Kees Cook <keescook@xxxxxxxxxxxx> wrote:
> On Mon, Feb 18, 2013 at 7:53 AM, Lucas De Marchi
> <lucas.demarchi@xxxxxxxxxxxxxx> wrote:
>> On Tue, Feb 12, 2013 at 8:32 PM, Kees Cook <keescook@xxxxxxxxxxxx> wrote:
>>> When a module is being loaded directly from disk (no compression,
>>> etc), pass the file descriptor to the new finit_module() syscall. If
>>> finit_module is exported by glibc, use it. Otherwise, manually make
>>> the syscall on architectures where it is known to exist.
>>> ---
>>> libkmod/libkmod-file.c | 16 +++++++++++++++-
>>> libkmod/libkmod-module.c | 18 ++++++++++++++++++
>>> libkmod/libkmod-private.h | 2 ++
>>> 3 files changed, 35 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/libkmod/libkmod-file.c b/libkmod/libkmod-file.c
>>> index ced20a8..76585f5 100644
>>> --- a/libkmod/libkmod-file.c
>>> +++ b/libkmod/libkmod-file.c
>>> @@ -52,6 +52,7 @@ struct kmod_file {
>>> gzFile gzf;
>>> #endif
>>> int fd;
>>> + int direct;
>>
>> it's either true or false. It should be bool, not int.
>
> Fixed.
>
>>> off_t size;
>>> void *memory;
>>> const struct file_ops *ops;
>>> @@ -254,9 +255,11 @@ static int load_reg(struct kmod_file *file)
>>> return -errno;
>>>
>>> file->size = st.st_size;
>>> - file->memory = mmap(0, file->size, PROT_READ, MAP_PRIVATE, file->fd, 0);
>>> + file->memory = mmap(NULL, file->size, PROT_READ, MAP_PRIVATE,
>>> + file->fd, 0);
>>> if (file->memory == MAP_FAILED)
>>> return -errno;
>>
>> this should be in patch 3.
>
> Whoops, yes.
>
>>> + file->direct = 1;
>>
>> s/1/true/
>>
>>> return 0;
>>> }
>>>
>>> @@ -300,6 +303,7 @@ struct kmod_file *kmod_file_open(const struct kmod_ctx *ctx,
>>> magic_size_max = itr->magic_size;
>>> }
>>>
>>> + file->direct = 0;
>>> if (magic_size_max > 0) {
>>> char *buf = alloca(magic_size_max + 1);
>>> ssize_t sz;
>>> @@ -353,6 +357,16 @@ off_t kmod_file_get_size(const struct kmod_file *file)
>>> return file->size;
>>> }
>>>
>>> +int kmod_file_get_direct(const struct kmod_file *file)
>>> +{
>>> + return file->direct;
>>> +}
>>> +
>>> +int kmod_file_get_fd(const struct kmod_file *file)
>>> +{
>>> + return file->fd;
>>> +}
>>
>> a leftover from previous patch? there's no reason for this function to exist.
>
> These functions are needed in libkmod-module.c. Since the other
> private members of the struct kmod_file were exported via functions
> like this, it seemed the right thing to do. What would you recommend?
err, my bad: I didn't see kmod_file_get_fd() was actually being used.
Lucas De Marchi
--
To unsubscribe from this list: send the line "unsubscribe linux-modules" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
