From: Phillip Lougher <phillip@squashfs.org.uk>
To: lists@colorremedies.com
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, Phillip Lougher <phillip@squashfs.org.uk>
Subject: Re: 6.0-rc1 BUG squashfs_decompress, and sleeping function called from invalid context at include/linux/sched/mm.h
Date: Sun, 21 Aug 2022 21:44:16 +0100 [thread overview]
Message-ID: <20220821204416.22346-1-phillip@squashfs.org.uk> (raw)
In-Reply-To: <73ea63fe-0c8e-412b-9fb2-94c08933180a@www.fastmail.com>
On 20/08/2022, Chris Murphy <lists@colorremedies.com> wrote:
>Seeing the following on every boot with kernel 6.0-rc1, when booting a Fedora Rawhide Live ISO with qemu-kvm. Full dmesg at:
>https://drive.google.com/file/d/15u38HZD9NSihIvz4P9M0W3dx6FZWq0MX/view?usp=sharing
My fault, it should be fixed with the following patch (untested).
I'll test and send the official patch out tomorrow.
Phillip
diff --git a/fs/squashfs/page_actor.c b/fs/squashfs/page_actor.c
index b23b780d8f42..48c988e3f5d2 100644
--- a/fs/squashfs/page_actor.c
+++ b/fs/squashfs/page_actor.c
@@ -68,20 +68,9 @@ static void *handle_next_page(struct squashfs_page_actor *actor)
if ((actor->next_page == actor->pages) ||
(actor->next_index != actor->page[actor->next_page]->index)) {
- if (actor->alloc_buffer) {
- void *tmp_buffer = kmalloc(PAGE_SIZE, GFP_KERNEL);
-
- if (tmp_buffer) {
- actor->tmp_buffer = tmp_buffer;
- actor->next_index++;
- actor->returned_pages++;
- return tmp_buffer;
- }
- }
-
actor->next_index++;
actor->returned_pages++;
- return ERR_PTR(-ENOMEM);
+ return actor->alloc_buffer ? actor->tmp_buffer : ERR_PTR(-ENOMEM);
}
actor->next_index++;
@@ -96,11 +85,10 @@ static void *direct_first_page(struct squashfs_page_actor *actor)
static void *direct_next_page(struct squashfs_page_actor *actor)
{
- if (actor->pageaddr)
+ if (actor->pageaddr) {
kunmap_local(actor->pageaddr);
-
- kfree(actor->tmp_buffer);
- actor->pageaddr = actor->tmp_buffer = NULL;
+ actor->pageaddr = NULL;
+ }
return handle_next_page(actor);
}
@@ -121,6 +109,16 @@ struct squashfs_page_actor *squashfs_page_actor_init_special(struct squashfs_sb_
if (actor == NULL)
return NULL;
+ if (msblk->decompressor->alloc_buffer) {
+ actor->tmp_buffer = kmalloc(PAGE_SIZE, GFP_KERNEL);
+
+ if (actor->tmp_buffer == NULL) {
+ kfree(actor);
+ return NULL;
+ }
+ } else
+ actor->tmp_buffer = NULL;
+
actor->length = length ? : pages * PAGE_SIZE;
actor->page = page;
actor->pages = pages;
@@ -128,7 +126,6 @@ struct squashfs_page_actor *squashfs_page_actor_init_special(struct squashfs_sb_
actor->returned_pages = 0;
actor->next_index = page[0]->index & ~((1 << (msblk->block_log - PAGE_SHIFT)) - 1);
actor->pageaddr = NULL;
- actor->tmp_buffer = NULL;
actor->alloc_buffer = msblk->decompressor->alloc_buffer;
actor->squashfs_first_page = direct_first_page;
actor->squashfs_next_page = direct_next_page;
--
2.35.1
prev parent reply other threads:[~2022-08-21 20:45 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-20 19:39 Chris Murphy
2022-08-21 20:44 ` Phillip Lougher [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220821204416.22346-1-phillip@squashfs.org.uk \
--to=phillip@squashfs.org.uk \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lists@colorremedies.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox