Re: [PATCH 5.15.y] firmware: stratix10-svc: fix memory leaks and list corruption bugs
"NG, TZE YEE" <[email protected]> Mon, 3 Aug 2026 12:16:43 +0000
| Newsgroups | org.kernel.vger.stable |
|---|---|
| Message-ID | <[email protected]> |
On 31/7/2026 10:39 am, Sasha Levin wrote: > [You don't often get email from [email protected]. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ] > > On Wed, Jul 29, 2026 at 07:36:04PM -0700, [email protected] wrote: >> - pmem = devm_kzalloc(chan->ctrl->dev, sizeof(*pmem), GFP_KERNEL); >> + pmem = kzalloc_obj(*pmem); > > Did you even build test this? > >> @@ -944,16 +946,17 @@ void stratix10_svc_free_memory(struct stratix10_svc_chan *chan, void *kaddr) >> { >> struct stratix10_svc_data_mem *pmem; >> >> + guard(mutex)(&svc_mem_lock); >> + >> list_for_each_entry(pmem, &svc_data_mem, node) > > This wouldn't work either, right? > > -- > Thanks, > Sasha Hi Sasha, Thanks for the review. You are right — this backport is incorrect as submitted. kzalloc_obj() is not available on these stable kernels, and the guard(mutex)(&svc_mem_lock) line was accidentally introduced during rebase; it is not part of the upstream fix and should not be in this backport. The same issues are present in the 5.10.y backport as well. I will send a v2 patches for both 5.15.y and 5.10.y that: - replaces kzalloc_obj() with kzalloc(sizeof(*pmem), GFP_KERNEL) - drops the stray guard(mutex)(&svc_mem_lock) Sorry for the noise. Thanks, Tze Yee