[PATCH] usercopy: Disable CONFIG_HARDENED_USERCOPY_PAGESPAN

[PATCH] usercopy: Disable CONFIG_HARDENED_USERCOPY_PAGESPAN

[Kees Cook]

CONFIG_HARDENED_USERCOPY_PAGESPAN has been mostly broken for a while,
and it has become hard to ignore with some recent scsi changes[1].
While there is a more complete series to replace it with better checks[2],
it should have more soak time in -next. Instead, disable the config now,
with the expectation that it will be fully replaced in the next kernel
release.

[1] https://lore.kernel.org/lkml/20220324064846.GA12961@lst.de/
[2] https://lore.kernel.org/linux-hardening/20220110231530.665970-1-willy@infradead.org/

Suggested-by: Christoph Hellwig <hch@lst.de>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
---
 security/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/security/Kconfig b/security/Kconfig
index 0b847f435beb..88a0bc8c592b 100644
--- a/security/Kconfig
+++ b/security/Kconfig
@@ -166,7 +166,7 @@ config HARDENED_USERCOPY
 config HARDENED_USERCOPY_PAGESPAN
 	bool "Refuse to copy allocations that span multiple pages"
 	depends on HARDENED_USERCOPY
-	depends on EXPERT
+	depends on BROKEN
 	help
 	  When a multi-page allocation is done without __GFP_COMP,
 	  hardened usercopy will reject attempts to copy it. There are,
-- 
2.32.0

Re: [PATCH] usercopy: Disable CONFIG_HARDENED_USERCOPY_PAGESPAN

[Christoph Hellwig]

On Thu, Mar 24, 2022 at 04:02:55PM -0700, Kees Cook wrote:
> CONFIG_HARDENED_USERCOPY_PAGESPAN has been mostly broken for a while,
> and it has become hard to ignore with some recent scsi changes[1].
> While there is a more complete series to replace it with better checks[2],
> it should have more soak time in -next. Instead, disable the config now,
> with the expectation that it will be fully replaced in the next kernel
> release.
> 
> [1] https://lore.kernel.org/lkml/20220324064846.GA12961@lst.de/
> [2] https://lore.kernel.org/linux-hardening/20220110231530.665970-1-willy@infradead.org/
> 
> Suggested-by: Christoph Hellwig <hch@lst.de>
> Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
> Signed-off-by: Kees Cook <keescook@chromium.org>

I'd prefer to just remove the code, but this is better than nothing:

Acked-by: Christoph Hellwig <hch@lst.de>