[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20250214154720.42c1fab53ad7c3d3f8c968aa@linux-foundation.org>
Date: Fri, 14 Feb 2025 15:47:20 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Sourabh Jain <sourabhjain@...ux.ibm.com>
Cc: Venkat Rao Bagalkote <venkat88@...ux.vnet.ibm.com>, Alexander Graf
<graf@...zon.com>, Baoquan he <bhe@...hat.com>, Hari Bathini
<hbathini@...ux.ibm.com>, Madhavan Srinivasan <maddy@...ux.ibm.com>,
"Mike Rapoport (Microsoft)" <rppt@...nel.org>, kexec@...ts.infradead.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] kexec: Fix kexec_locate_mem_hole() for missing
CONFIG_KEXEC_HANDOVER
On Fri, 14 Feb 2025 18:24:02 +0530 Sourabh Jain <sourabhjain@...ux.ibm.com> wrote:
> While adding KHO support, commit 7d128945e003 ("kexec: add KHO support
> to kexec file loads") returns early from kexec_locate_mem_hole() if
> CONFIG_KEXEC_HANDOVER is not defined.
>
> Due to this, kexec_locate_mem_hole() does not locate a hole for the
> kexec segment, and kbuf.mem holds 0x0. This leads to a kexec_file_load
> syscall failure in sanity_check_segment_list(). This impacts both
> kexec and kdump kernels.
Thanks.
Mike, I'll remove the v4 series from mm-unstable to gie you a clean
shot at v5. Please incorporate Sourabh's fix?
Powered by blists - more mailing lists