On Tue, Feb 25, 2025 at 2:25 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> On Tue, 25 Feb 2025 13:32:00 -0800 Nhat Pham <nphamcs@xxxxxxxxx> wrote:
>
> > @@ -984,12 +987,19 @@ static void zswap_decompress(struct zswap_entry *entry, struct folio *folio)
> > sg_init_table(&output, 1);
> > sg_set_folio(&output, folio, PAGE_SIZE, 0);
> > acomp_request_set_params(acomp_ctx->req, &input, &output, entry->length, PAGE_SIZE);
> > - BUG_ON(crypto_wait_req(crypto_acomp_decompress(acomp_ctx->req), &acomp_ctx->wait));
> > - BUG_ON(acomp_ctx->req->dlen != PAGE_SIZE);
> > + if (crypto_wait_req(crypto_acomp_decompress(acomp_ctx->req), &acomp_ctx->wait) ||
> > + acomp_ctx->req->dlen != PAGE_SIZE) {
> > + ret = false;
> > + zswap_reject_decompress_fail++;
> > + pr_alert_ratelimited(
> > + "decompression failed on zswap entry with offset %08lx\n",
> > + entry->swpentry.val);
> > + }
> > mutex_unlock(&acomp_ctx->mutex);
>
> This mutex_unlock() isn't present in current kernels. I'd normally just fix
> the reject but a change in the locking environment needs some careful
> checking and retesting, please.
Hah strange. I could have sworn I pulled the latest mm-unstable, but
maybe the git pull failed and I didn't notice. Thanks for picking this
up Andrew! Lemme re-pull and rebase.
