linux-kernel - Re: [drivers/iio] Question about `iio_gts_build_avail_time

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <9724f90c-111f-40e7-9787-c62787cef96e@gmail.com>
Date: Wed, 13 Mar 2024 10:08:50 +0200
From: Matti Vaittinen <mazziesaccount@...il.com>
To: Chenyuan Yang <chenyuan0y@...il.com>
Cc: jic23@...nel.org, lars@...afoo.de, linux-iio@...r.kernel.org,
 linux-kernel@...r.kernel.org, Zijie Zhao <zzjas98@...il.com>
Subject: Re: [drivers/iio] Question about `iio_gts_build_avail_time_table`

Hi Chenyuan,

On 3/12/24 18:53, Chenyuan Yang wrote:
> Hi Matti,
> 
> I have a question about the "The idea of the check which has been
> removed was to assign the value in
> the array in first free spot if it is bigger than the last value".

Can you please avoid top-posting when discussing on the Linux lists. You 
can find more information from:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html

part of which may be crucial in order to get your changes applied if you 
haven't already familiarized yourself with the kernel development processes.


> 
> -               if (times[idx] < new) {
> -                       times[idx++] = new;
> -                       continue;
> -               }
> +               times[idx] = new;
> 
> It appears that the comparison should perhaps be made with `idx-1`
> rather than `idx`, given that `idx` represents the current number of
> copied values in times, whereas `idx-1` points to the last value.
> Could I have your thoughts on this?

Yes. I implemented the old code wrong as you pointed out.

You may want to take the GTS Kunit test cases:
https://lore.kernel.org/all/6b839dd533fd93b75c2e6f6a8f2286233d4901fb.1704881096.git.mazziesaccount@gmail.com/
which, I think, are already merged in IIO testing branch.

You can test the sorting when you change the order of the times in the 
test case:

+static const struct iio_itime_sel_mul gts_test_itimes[] = {
+	GAIN_SCALE_ITIME_US(400 * 1000, TEST_TSEL_400, 8),
+	GAIN_SCALE_ITIME_US(200 * 1000, TEST_TSEL_200, 4),
+	GAIN_SCALE_ITIME_US(100 * 1000, TEST_TSEL_100, 2),
+	GAIN_SCALE_ITIME_US(50 * 1000, TEST_TSEL_50, 1),
+#define TIMEGAIN_MAX 8
+};

for example to

+static const struct iio_itime_sel_mul gts_test_itimes[] = {
+	GAIN_SCALE_ITIME_US(400 * 1000, TEST_TSEL_400, 8),
+	GAIN_SCALE_ITIME_US(50 * 1000, TEST_TSEL_50, 1),
+	GAIN_SCALE_ITIME_US(200 * 1000, TEST_TSEL_200, 4),
+	GAIN_SCALE_ITIME_US(100 * 1000, TEST_TSEL_100, 2),
+#define TIMEGAIN_MAX 8
+};

Yours,
	-- Matti

-- 
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~