In preparation to enabling -Wvla, remove VLA and replace it
with a fixed-length array instead.
Fixed as part of the directive to remove all VLAs from
the kernel: https://lkml.org/lkml/2018/3/7/621
Signed-off-by: Gustavo A. R. Silva <[email protected]>
---
Notice that due to this change, the field max_interleave is no longer
used after it has been initialized. Maybe it should be removed?
Thanks
--
Gustavo
---
drivers/edac/sb_edac.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/edac/sb_edac.c b/drivers/edac/sb_edac.c
index 8721002..11cb2c9 100644
--- a/drivers/edac/sb_edac.c
+++ b/drivers/edac/sb_edac.c
@@ -110,6 +110,7 @@ static const u32 knl_interleave_list[] = {
0xdc, 0xe4, 0xec, 0xf4, 0xfc, /* 15-19 */
0x104, 0x10c, 0x114, 0x11c, /* 20-23 */
};
+#define MAX_INTERLEAVE ARRAY_SIZE(knl_interleave_list)
struct interleave_pkg {
unsigned char start;
@@ -1899,7 +1900,7 @@ static int get_memory_error_data(struct mem_ctl_info *mci,
int n_rir, n_sads, n_tads, sad_way, sck_xch;
int sad_interl, idx, base_ch;
int interleave_mode, shiftup = 0;
- unsigned sad_interleave[pvt->info.max_interleave];
+ unsigned int sad_interleave[MAX_INTERLEAVE];
u32 reg, dram_rule;
u8 ch_way, sck_way, pkg, sad_ha = 0;
u32 tad_offset;
--
2.7.4
On Fri, Mar 09, 2018 at 09:02:18PM -0600, Gustavo A. R. Silva wrote:
> In preparation to enabling -Wvla, remove VLA and replace it
> with a fixed-length array instead.
>
> Fixed as part of the directive to remove all VLAs from
> the kernel: https://lkml.org/lkml/2018/3/7/621
>
> Signed-off-by: Gustavo A. R. Silva <[email protected]>
> ---
>
> Notice that due to this change, the field max_interleave is no longer
> used after it has been initialized. Maybe it should be removed?
Yes.
> @@ -110,6 +110,7 @@ static const u32 knl_interleave_list[] = {
> 0xdc, 0xe4, 0xec, 0xf4, 0xfc, /* 15-19 */
> 0x104, 0x10c, 0x114, 0x11c, /* 20-23 */
> };
> +#define MAX_INTERLEAVE ARRAY_SIZE(knl_interleave_list)
define that as the max of all interleave lists array sizes so that
people can update it properly when new interleave lists get added.
--
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.
Hi Borislav,
On 03/10/2018 08:04 AM, Borislav Petkov wrote:
> On Fri, Mar 09, 2018 at 09:02:18PM -0600, Gustavo A. R. Silva wrote:
>> In preparation to enabling -Wvla, remove VLA and replace it
>> with a fixed-length array instead.
>>
>> Fixed as part of the directive to remove all VLAs from
>> the kernel: https://lkml.org/lkml/2018/3/7/621
>>
>> Signed-off-by: Gustavo A. R. Silva <[email protected]>
>> ---
>>
>> Notice that due to this change, the field max_interleave is no longer
>> used after it has been initialized. Maybe it should be removed?
>
> Yes.
>
I'll remove it then.
>> @@ -110,6 +110,7 @@ static const u32 knl_interleave_list[] = {
>> 0xdc, 0xe4, 0xec, 0xf4, 0xfc, /* 15-19 */
>> 0x104, 0x10c, 0x114, 0x11c, /* 20-23 */
>> };
>> +#define MAX_INTERLEAVE ARRAY_SIZE(knl_interleave_list)
>
> define that as the max of all interleave lists array sizes so that
> people can update it properly when new interleave lists get added.
>
Do you mean just adding a code comment?
Thanks for the feedback.
--
Gustavo
On Mon, Mar 12, 2018 at 05:52:13PM -0500, Gustavo A. R. Silva wrote:
> Do you mean just adding a code comment?
No, I mean
#define MAX_INTERLEAVE (max_t(unsigned int, ARRAY_SIZE(knl_interleave_list), ...
which computes the max of all three array sizes.
Thx.
--
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.
On 03/13/2018 05:48 AM, Borislav Petkov wrote:
> On Mon, Mar 12, 2018 at 05:52:13PM -0500, Gustavo A. R. Silva wrote:
>> Do you mean just adding a code comment?
>
> No, I mean
>
> #define MAX_INTERLEAVE (max_t(unsigned int, ARRAY_SIZE(knl_interleave_list), ...
>
> which computes the max of all three array sizes.
>
> Thx.
>
I got it.
Exactly this:
#define MAX_INTERLEAVE (max_t(unsigned int,
ARRAY_SIZE(knl_interleave_list),\
max_t(unsigned int,
ARRAY_SIZE(ibridge_interleave_list),\
ARRAY_SIZE(sbridge_interleave_list))))
I'll send v2 shortly.
Thanks!
--
Gustavo