[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20210117133355.10fe86c628279161bb3435d4@linux-foundation.org>
Date: Sun, 17 Jan 2021 13:33:55 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Florian Fainelli <f.fainelli@...il.com>
Cc: linux-kernel@...r.kernel.org, opendmb@...il.com,
Barret Rhoden <brho@...gle.com>, Arnd Bergmann <arnd@...db.de>
Subject: Re: [PATCH] initramfs: Panic with memory information
On Thu, 14 Jan 2021 15:15:16 -0800 Florian Fainelli <f.fainelli@...il.com> wrote:
> On systems with large amounts of reserved memory we may fail to
> successfully complete unpack_to_rootfs() and be left with:
>
> Kernel panic - not syncing: write error
>
> this is not too helpful to understand what happened, so let's wrap the
> panic() calls with a surrounding show_mem() such that we have a chance
> of understanding the memory conditions leading to these allocation
> failures.
Seems sensible.
> @@ -45,6 +46,11 @@ static void __init error(char *x)
> message = x;
> }
>
> +#define panic_show_mem(...) { \
> + show_mem(0, NULL); \
> + panic(__VA_ARGS__); \
> +}
> +
But can we replace nasty macro with pleasing C code?
--- a/init/initramfs.c~initramfs-panic-with-memory-information-fix
+++ a/init/initramfs.c
@@ -46,9 +46,14 @@ static void __init error(char *x)
message = x;
}
-#define panic_show_mem(...) { \
- show_mem(0, NULL); \
- panic(__VA_ARGS__); \
+static void panic_show_mem(const char *fmt, ...)
+{
+ va_list args;
+
+ show_mem(0, NULL);
+ va_start(args, fmt);
+ panic(fmt, args);
+ va_end(args);
}
/* link hash */
_
Powered by blists - more mailing lists