* [PATCH V3] kexec: copy only happens before uchunk goes to zero
@ 2024-02-22 9:21 yang.zhang
2024-03-01 6:52 ` Baoquan He
0 siblings, 1 reply; 2+ messages in thread
From: yang.zhang @ 2024-02-22 9:21 UTC (permalink / raw)
To: ebiederm; +Cc: linux-kernel, kexec, bhe, yang.zhang
From: "yang.zhang" <yang.zhang@hexintek.com>
When loading segments, ubytes is <= mbytes. When ubytes is exhausted,
there could be remaining mbytes. Then in the while loop, the buf pointer
advancing with mchunk will causing meaningless reading even though it
doesn't harm.
So let's change to make sure that all of the copying and the rest only
happens before uchunk goes to zero.
Acked-by: Baoquan He <bhe@redhat.com>
Signed-off-by: yang.zhang <yang.zhang@hexintek.com>
---
V2 -> V3:
- Add more detailed description in the commit message
v1 -> v2:
- Only copy before uchunk goes to zero
V1: https://lore.kernel.org/lkml/20240130101802.23850-1-gaoshanliukou@163.com/
---
kernel/kexec_core.c | 44 ++++++++++++++++++++++++--------------------
1 file changed, 24 insertions(+), 20 deletions(-)
diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c
index d08fc7b5db97..2fc3d0e3715a 100644
--- a/kernel/kexec_core.c
+++ b/kernel/kexec_core.c
@@ -800,22 +800,24 @@ static int kimage_load_normal_segment(struct kimage *image,
PAGE_SIZE - (maddr & ~PAGE_MASK));
uchunk = min(ubytes, mchunk);
- /* For file based kexec, source pages are in kernel memory */
- if (image->file_mode)
- memcpy(ptr, kbuf, uchunk);
- else
- result = copy_from_user(ptr, buf, uchunk);
+ if (uchunk) {
+ /* For file based kexec, source pages are in kernel memory */
+ if (image->file_mode)
+ memcpy(ptr, kbuf, uchunk);
+ else
+ result = copy_from_user(ptr, buf, uchunk);
+ ubytes -= uchunk;
+ if (image->file_mode)
+ kbuf += uchunk;
+ else
+ buf += uchunk;
+ }
kunmap_local(ptr);
if (result) {
result = -EFAULT;
goto out;
}
- ubytes -= uchunk;
maddr += mchunk;
- if (image->file_mode)
- kbuf += mchunk;
- else
- buf += mchunk;
mbytes -= mchunk;
cond_resched();
@@ -866,11 +868,18 @@ static int kimage_load_crash_segment(struct kimage *image,
memset(ptr + uchunk, 0, mchunk - uchunk);
}
- /* For file based kexec, source pages are in kernel memory */
- if (image->file_mode)
- memcpy(ptr, kbuf, uchunk);
- else
- result = copy_from_user(ptr, buf, uchunk);
+ if (uchunk) {
+ /* For file based kexec, source pages are in kernel memory */
+ if (image->file_mode)
+ memcpy(ptr, kbuf, uchunk);
+ else
+ result = copy_from_user(ptr, buf, uchunk);
+ ubytes -= uchunk;
+ if (image->file_mode)
+ kbuf += uchunk;
+ else
+ buf += uchunk;
+ }
kexec_flush_icache_page(page);
kunmap_local(ptr);
arch_kexec_pre_free_pages(page_address(page), 1);
@@ -878,12 +887,7 @@ static int kimage_load_crash_segment(struct kimage *image,
result = -EFAULT;
goto out;
}
- ubytes -= uchunk;
maddr += mchunk;
- if (image->file_mode)
- kbuf += mchunk;
- else
- buf += mchunk;
mbytes -= mchunk;
cond_resched();
--
2.25.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH V3] kexec: copy only happens before uchunk goes to zero
2024-02-22 9:21 [PATCH V3] kexec: copy only happens before uchunk goes to zero yang.zhang
@ 2024-03-01 6:52 ` Baoquan He
0 siblings, 0 replies; 2+ messages in thread
From: Baoquan He @ 2024-03-01 6:52 UTC (permalink / raw)
To: yang.zhang, akpm; +Cc: ebiederm, linux-kernel, kexec, yang.zhang
CC Andrew
On 02/22/24 at 05:21pm, yang.zhang wrote:
> From: "yang.zhang" <yang.zhang@hexintek.com>
>
> When loading segments, ubytes is <= mbytes. When ubytes is exhausted,
> there could be remaining mbytes. Then in the while loop, the buf pointer
> advancing with mchunk will causing meaningless reading even though it
> doesn't harm.
>
> So let's change to make sure that all of the copying and the rest only
> happens before uchunk goes to zero.
>
> Acked-by: Baoquan He <bhe@redhat.com>
> Signed-off-by: yang.zhang <yang.zhang@hexintek.com>
>
> ---
> V2 -> V3:
> - Add more detailed description in the commit message
> v1 -> v2:
> - Only copy before uchunk goes to zero
>
> V1: https://lore.kernel.org/lkml/20240130101802.23850-1-gaoshanliukou@163.com/
> ---
> kernel/kexec_core.c | 44 ++++++++++++++++++++++++--------------------
> 1 file changed, 24 insertions(+), 20 deletions(-)
>
> diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c
> index d08fc7b5db97..2fc3d0e3715a 100644
> --- a/kernel/kexec_core.c
> +++ b/kernel/kexec_core.c
> @@ -800,22 +800,24 @@ static int kimage_load_normal_segment(struct kimage *image,
> PAGE_SIZE - (maddr & ~PAGE_MASK));
> uchunk = min(ubytes, mchunk);
>
> - /* For file based kexec, source pages are in kernel memory */
> - if (image->file_mode)
> - memcpy(ptr, kbuf, uchunk);
> - else
> - result = copy_from_user(ptr, buf, uchunk);
> + if (uchunk) {
> + /* For file based kexec, source pages are in kernel memory */
> + if (image->file_mode)
> + memcpy(ptr, kbuf, uchunk);
> + else
> + result = copy_from_user(ptr, buf, uchunk);
> + ubytes -= uchunk;
> + if (image->file_mode)
> + kbuf += uchunk;
> + else
> + buf += uchunk;
> + }
> kunmap_local(ptr);
> if (result) {
> result = -EFAULT;
> goto out;
> }
> - ubytes -= uchunk;
> maddr += mchunk;
> - if (image->file_mode)
> - kbuf += mchunk;
> - else
> - buf += mchunk;
> mbytes -= mchunk;
>
> cond_resched();
> @@ -866,11 +868,18 @@ static int kimage_load_crash_segment(struct kimage *image,
> memset(ptr + uchunk, 0, mchunk - uchunk);
> }
>
> - /* For file based kexec, source pages are in kernel memory */
> - if (image->file_mode)
> - memcpy(ptr, kbuf, uchunk);
> - else
> - result = copy_from_user(ptr, buf, uchunk);
> + if (uchunk) {
> + /* For file based kexec, source pages are in kernel memory */
> + if (image->file_mode)
> + memcpy(ptr, kbuf, uchunk);
> + else
> + result = copy_from_user(ptr, buf, uchunk);
> + ubytes -= uchunk;
> + if (image->file_mode)
> + kbuf += uchunk;
> + else
> + buf += uchunk;
> + }
> kexec_flush_icache_page(page);
> kunmap_local(ptr);
> arch_kexec_pre_free_pages(page_address(page), 1);
> @@ -878,12 +887,7 @@ static int kimage_load_crash_segment(struct kimage *image,
> result = -EFAULT;
> goto out;
> }
> - ubytes -= uchunk;
> maddr += mchunk;
> - if (image->file_mode)
> - kbuf += mchunk;
> - else
> - buf += mchunk;
> mbytes -= mchunk;
>
> cond_resched();
> --
> 2.25.1
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2024-03-01 6:52 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-02-22 9:21 [PATCH V3] kexec: copy only happens before uchunk goes to zero yang.zhang
2024-03-01 6:52 ` Baoquan He
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox