linux-kernel - Re: [PATCH] crypto: marvell/cesa - Avoid empty transfer descriptor

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <aC2xTI1ZuXoZjgjX@gondor.apana.org.au>
Date: Wed, 21 May 2025 18:56:12 +0800
From: Herbert Xu <herbert@...dor.apana.org.au>
To: Corentin Labbe <clabbe.montjoie@...il.com>
Cc: Klaus Kudielka <klaus.kudielka@...il.com>,
	Eric Biggers <ebiggers@...nel.org>, regressions@...ts.linux.dev,
	linux-kernel@...r.kernel.org,
	Linux Crypto Mailing List <linux-crypto@...r.kernel.org>,
	Boris Brezillon <bbrezillon@...nel.org>,
	EBALARD Arnaud <Arnaud.Ebalard@....gouv.fr>,
	Romain Perier <romain.perier@...il.com>,
	Arnd Bergmann <arnd@...db.de>
Subject: Re: [PATCH] crypto: marvell/cesa - Avoid empty transfer descriptor

On Wed, May 21, 2025 at 12:45:18PM +0200, Corentin Labbe wrote:
>
> It fails to build with:
> ERROR: modpost: "arch_sync_dma_for_device" [drivers/crypto/marvell/cesa/marvell-cesa.ko] undefined!

OK, what about this patch?

Thanks,
-- 
Email: Herbert Xu <herbert@...dor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--
diff --git a/drivers/crypto/marvell/cesa/hash.c b/drivers/crypto/marvell/cesa/hash.c
index 6815eddc9068..5c46cd267789 100644
--- a/drivers/crypto/marvell/cesa/hash.c
+++ b/drivers/crypto/marvell/cesa/hash.c
@@ -49,8 +49,7 @@ mv_cesa_ahash_req_iter_next_op(struct mv_cesa_ahash_dma_iter *iter)
 static inline int
 mv_cesa_ahash_dma_alloc_cache(struct mv_cesa_ahash_dma_req *req, gfp_t flags)
 {
-	req->cache = dma_pool_alloc(cesa_dev->dma->cache_pool, flags,
-				    &req->cache_dma);
+	req->cache = kmalloc(CESA_MAX_HASH_BLOCK_SIZE, flags);
 	if (!req->cache)
 		return -ENOMEM;
 
@@ -63,8 +62,8 @@ mv_cesa_ahash_dma_free_cache(struct mv_cesa_ahash_dma_req *req)
 	if (!req->cache)
 		return;
 
-	dma_pool_free(cesa_dev->dma->cache_pool, req->cache,
-		      req->cache_dma);
+	dma_unmap_single(cesa_dev->dev, req->cache_dma, CESA_MAX_HASH_BLOCK_SIZE, DMA_TO_DEVICE);
+	kfree(req->cache);
 }
 
 static int mv_cesa_ahash_dma_alloc_padding(struct mv_cesa_ahash_dma_req *req,
@@ -533,6 +532,13 @@ mv_cesa_ahash_dma_add_cache(struct mv_cesa_tdma_chain *chain,
 
 	memcpy(ahashdreq->cache, creq->cache, creq->cache_ptr);
 
+	ahashdreq->cache_dma = dma_map_single(cesa_dev->dev, ahashdreq->cache, CESA_MAX_HASH_BLOCK_SIZE, DMA_TO_DEVICE);
+	if (dma_mapping_error(cesa_dev->dev, ahashdreq->cache_dma)) {
+		dev_err(cesa_dev->dev, "dma_map_single failed\n");
+		kfree(ahashdreq->cache);
+		return -ENOMEM;
+	}
+
 	return mv_cesa_dma_add_data_transfer(chain,
 					     CESA_SA_DATA_SRAM_OFFSET,
 					     ahashdreq->cache_dma,