[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20171218142138.f1b84a8c1ca072b15b54af33@linux-foundation.org>
Date: Mon, 18 Dec 2017 14:21:38 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Pravin Shedge <pravin.shedge4linux@...il.com>
Cc: fkostenzer@...e.at, andriy.shevchenko@...ux.intel.com,
geert@...ux-m68k.org, paul.gortmaker@...driver.com,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] lib: add module unload support to sort tests
On Sun, 17 Dec 2017 15:19:27 +0530 Pravin Shedge <pravin.shedge4linux@...il.com> wrote:
> test_sort.c perform array-based and linked list sort test. Code allows to
> compile either as a loadable modules or builtin into the kernel.
>
> Current code is not allow to unload the test_sort.ko module after
> successful completion.
>
> This patch add support to unload the "test_sort.ko" module.
>
> ...
>
> --- a/lib/test_sort.c
> +++ b/lib/test_sort.c
> @@ -13,11 +13,12 @@ static int __init cmpint(const void *a, const void *b)
>
> static int __init test_sort_init(void)
> {
> - int *a, i, r = 1, err = -ENOMEM;
> + int *a, i, r = 1;
> + int err = -EAGAIN; /* Fail will directly unload the module */
>
> a = kmalloc_array(TEST_LEN, sizeof(*a), GFP_KERNEL);
> if (!a)
> - return err;
> + return -ENOMEM;
>
> for (i = 0; i < TEST_LEN; i++) {
> r = (r * 725861) % 6599;
> @@ -26,13 +27,12 @@ static int __init test_sort_init(void)
>
> sort(a, TEST_LEN, sizeof(*a), cmpint, NULL);
>
> - err = -EINVAL;
> for (i = 0; i < TEST_LEN-1; i++)
> if (a[i] > a[i+1]) {
> pr_err("test has failed\n");
> + err = -EINVAL;
> goto exit;
> }
> - err = 0;
> pr_info("test passed\n");
> exit:
> kfree(a);
I'm struggling to understand this.
It seems that the current pattern for lib/test_*.c is:
- if test fails, module_init() handler returns -EFOO
- if test succeeds, module_init() handler returns 0
So the module will be auto-unloaded if it failed and will require an
rmmod if it succeeded.
Correct?
And it appears that lib/test_sort.c currently implements the above.
And that your patch changes it so that the module_init() handler always
returns -EFOO, so the module will be aut-unloaded whether or not the
testing passed.
Correct?
If so, why do you think we shiould alter lib/test_sort.c to behave in
this atypical fashion?
Powered by blists - more mailing lists