[PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[Herbert Xu]

The header file linux/uio.h includes crypto/hash.h which pulls in
most of the Crypto API.  Since linux/uio.h is used throughout the
kernel this means that every tiny bit of change to the Crypto API
causes the entire kernel to get rebuilt.

This patch fixes this by moving it into lib/iov_iter.c instead
where it is actually used.

This patch also fixes the ifdef to use CRYPTO_HASH instead of just
CRYPTO which does not guarantee the existence of ahash.

Finally the prototype of the function has been changed to avoid
the unnecessary use of a void pointer.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/include/linux/uio.h b/include/linux/uio.h
index 9576fd8158d7..67a8ffd31ad8 100644
--- a/include/linux/uio.h
+++ b/include/linux/uio.h
@@ -7,9 +7,9 @@
 
 #include <linux/kernel.h>
 #include <linux/thread_info.h>
-#include <crypto/hash.h>
 #include <uapi/linux/uio.h>
 
+struct ahash_request;
 struct page;
 struct pipe_inode_info;
 
@@ -264,7 +264,7 @@ static inline void iov_iter_reexpand(struct iov_iter *i, size_t count)
 size_t csum_and_copy_to_iter(const void *addr, size_t bytes, void *csump, struct iov_iter *i);
 size_t csum_and_copy_from_iter(void *addr, size_t bytes, __wsum *csum, struct iov_iter *i);
 bool csum_and_copy_from_iter_full(void *addr, size_t bytes, __wsum *csum, struct iov_iter *i);
-size_t hash_and_copy_to_iter(const void *addr, size_t bytes, void *hashp,
+size_t hash_and_copy_to_iter(const void *addr, size_t bytes, struct ahash_request *hash,
 		struct iov_iter *i);
 
 ssize_t import_iovec(int type, const struct iovec __user * uvector,
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 51595bf3af85..ac537111dcc6 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -1,4 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0-only
+#include <crypto/hash.h>
 #include <linux/export.h>
 #include <linux/bvec.h>
 #include <linux/uio.h>
@@ -1563,11 +1564,10 @@ size_t csum_and_copy_to_iter(const void *addr, size_t bytes, void *csump,
 }
 EXPORT_SYMBOL(csum_and_copy_to_iter);
 
-size_t hash_and_copy_to_iter(const void *addr, size_t bytes, void *hashp,
+size_t hash_and_copy_to_iter(const void *addr, size_t bytes, struct ahash_request *hash,
 		struct iov_iter *i)
 {
-#ifdef CONFIG_CRYPTO
-	struct ahash_request *hash = hashp;
+#ifdef CONFIG_CRYPTO_HASH
 	struct scatterlist sg;
 	size_t copied;
 
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[Herbert Xu]

On Thu, Jun 11, 2020 at 05:43:32PM +1000, Herbert Xu wrote:
>
> Finally the prototype of the function has been changed to avoid
> the unnecessary use of a void pointer.

OK that doesn't quite work.  Let me respin without it and instead
add some missing inclusions of crypto/hash.h in files that were
wrongly relying on uio.h to include it.

Cheers,
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[kernel test robot]

[-- Attachment #1: Type: text/plain, Size: 16242 bytes --]

Hi Herbert,

I love your patch! Yet something to improve:

[auto build test ERROR on linus/master]
[also build test ERROR on v5.7 next-20200611]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]

url:    https://github.com/0day-ci/linux/commits/Herbert-Xu/iov_iter-Move-unnecessary-inclusion-of-crypto-hash-h/20200611-154742
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git b29482fde649c72441d5478a4ea2c52c56d97a5e
config: c6x-allyesconfig (attached as .config)
compiler: c6x-elf-gcc (GCC) 9.3.0
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross ARCH=c6x 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All error/warnings (new ones prefixed by >>, old ones prefixed by <<):

drivers/soc/qcom/pdr_interface.c: In function 'pdr_indack_work':
>> drivers/soc/qcom/pdr_interface.c:292:3: error: implicit declaration of function 'kfree' [-Werror=implicit-function-declaration]
292 |   kfree(ind);
|   ^~~~~
drivers/soc/qcom/pdr_interface.c: In function 'pdr_indication_cb':
>> drivers/soc/qcom/pdr_interface.c:328:8: error: implicit declaration of function 'kzalloc' [-Werror=implicit-function-declaration]
328 |  ind = kzalloc(sizeof(*ind), GFP_KERNEL);
|        ^~~~~~~
>> drivers/soc/qcom/pdr_interface.c:328:6: warning: assignment to 'struct pdr_list_node *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
328 |  ind = kzalloc(sizeof(*ind), GFP_KERNEL);
|      ^
drivers/soc/qcom/pdr_interface.c: In function 'pdr_locate_service':
>> drivers/soc/qcom/pdr_interface.c:401:7: warning: assignment to 'struct servreg_get_domain_list_resp *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
401 |  resp = kzalloc(sizeof(*resp), GFP_KERNEL);
|       ^
drivers/soc/qcom/pdr_interface.c: In function 'pdr_add_lookup':
>> drivers/soc/qcom/pdr_interface.c:526:6: warning: assignment to 'struct pdr_service *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
526 |  pds = kzalloc(sizeof(*pds), GFP_KERNEL);
|      ^
drivers/soc/qcom/pdr_interface.c: In function 'pdr_handle_alloc':
>> drivers/soc/qcom/pdr_interface.c:656:6: warning: assignment to 'struct pdr_handle *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
656 |  pdr = kzalloc(sizeof(*pdr), GFP_KERNEL);
|      ^
cc1: some warnings being treated as errors

vim +/kfree +292 drivers/soc/qcom/pdr_interface.c

fbe639b44a8275 Sibi Sankar       2020-03-12  271  
fbe639b44a8275 Sibi Sankar       2020-03-12  272  static void pdr_indack_work(struct work_struct *work)
fbe639b44a8275 Sibi Sankar       2020-03-12  273  {
fbe639b44a8275 Sibi Sankar       2020-03-12  274  	struct pdr_handle *pdr = container_of(work, struct pdr_handle,
fbe639b44a8275 Sibi Sankar       2020-03-12  275  					      indack_work);
fbe639b44a8275 Sibi Sankar       2020-03-12  276  	struct pdr_list_node *ind, *tmp;
fbe639b44a8275 Sibi Sankar       2020-03-12  277  	struct pdr_service *pds;
fbe639b44a8275 Sibi Sankar       2020-03-12  278  
fbe639b44a8275 Sibi Sankar       2020-03-12  279  	list_for_each_entry_safe(ind, tmp, &pdr->indack_list, node) {
fbe639b44a8275 Sibi Sankar       2020-03-12  280  		pds = ind->pds;
fbe639b44a8275 Sibi Sankar       2020-03-12  281  		pdr_send_indack_msg(pdr, pds, ind->transaction_id);
fbe639b44a8275 Sibi Sankar       2020-03-12  282  
fbe639b44a8275 Sibi Sankar       2020-03-12  283  		mutex_lock(&pdr->status_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  284  		pds->state = ind->curr_state;
fbe639b44a8275 Sibi Sankar       2020-03-12  285  		pdr->status(pds->state, pds->service_path, pdr->priv);
fbe639b44a8275 Sibi Sankar       2020-03-12  286  		mutex_unlock(&pdr->status_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  287  
fbe639b44a8275 Sibi Sankar       2020-03-12  288  		mutex_lock(&pdr->list_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  289  		list_del(&ind->node);
fbe639b44a8275 Sibi Sankar       2020-03-12  290  		mutex_unlock(&pdr->list_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  291  
fbe639b44a8275 Sibi Sankar       2020-03-12 @292  		kfree(ind);
fbe639b44a8275 Sibi Sankar       2020-03-12  293  	}
fbe639b44a8275 Sibi Sankar       2020-03-12  294  }
fbe639b44a8275 Sibi Sankar       2020-03-12  295  
fbe639b44a8275 Sibi Sankar       2020-03-12  296  static void pdr_indication_cb(struct qmi_handle *qmi,
fbe639b44a8275 Sibi Sankar       2020-03-12  297  			      struct sockaddr_qrtr *sq,
fbe639b44a8275 Sibi Sankar       2020-03-12  298  			      struct qmi_txn *txn, const void *data)
fbe639b44a8275 Sibi Sankar       2020-03-12  299  {
fbe639b44a8275 Sibi Sankar       2020-03-12  300  	struct pdr_handle *pdr = container_of(qmi, struct pdr_handle,
fbe639b44a8275 Sibi Sankar       2020-03-12  301  					      notifier_hdl);
fbe639b44a8275 Sibi Sankar       2020-03-12  302  	const struct servreg_state_updated_ind *ind_msg = data;
fbe639b44a8275 Sibi Sankar       2020-03-12  303  	struct pdr_list_node *ind;
fbe639b44a8275 Sibi Sankar       2020-03-12  304  	struct pdr_service *pds;
e69b3bede1b2f7 Nathan Chancellor 2020-03-16  305  	bool found = false;
fbe639b44a8275 Sibi Sankar       2020-03-12  306  
fbe639b44a8275 Sibi Sankar       2020-03-12  307  	if (!ind_msg || !ind_msg->service_path[0] ||
fbe639b44a8275 Sibi Sankar       2020-03-12  308  	    strlen(ind_msg->service_path) > SERVREG_NAME_LENGTH)
fbe639b44a8275 Sibi Sankar       2020-03-12  309  		return;
fbe639b44a8275 Sibi Sankar       2020-03-12  310  
fbe639b44a8275 Sibi Sankar       2020-03-12  311  	mutex_lock(&pdr->list_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  312  	list_for_each_entry(pds, &pdr->lookups, node) {
fbe639b44a8275 Sibi Sankar       2020-03-12  313  		if (strcmp(pds->service_path, ind_msg->service_path))
fbe639b44a8275 Sibi Sankar       2020-03-12  314  			continue;
fbe639b44a8275 Sibi Sankar       2020-03-12  315  
fbe639b44a8275 Sibi Sankar       2020-03-12  316  		found = true;
fbe639b44a8275 Sibi Sankar       2020-03-12  317  		break;
fbe639b44a8275 Sibi Sankar       2020-03-12  318  	}
fbe639b44a8275 Sibi Sankar       2020-03-12  319  	mutex_unlock(&pdr->list_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  320  
fbe639b44a8275 Sibi Sankar       2020-03-12  321  	if (!found)
fbe639b44a8275 Sibi Sankar       2020-03-12  322  		return;
fbe639b44a8275 Sibi Sankar       2020-03-12  323  
fbe639b44a8275 Sibi Sankar       2020-03-12  324  	pr_info("PDR: Indication received from %s, state: 0x%x, trans-id: %d\n",
fbe639b44a8275 Sibi Sankar       2020-03-12  325  		ind_msg->service_path, ind_msg->curr_state,
fbe639b44a8275 Sibi Sankar       2020-03-12  326  		ind_msg->transaction_id);
fbe639b44a8275 Sibi Sankar       2020-03-12  327  
fbe639b44a8275 Sibi Sankar       2020-03-12 @328  	ind = kzalloc(sizeof(*ind), GFP_KERNEL);
fbe639b44a8275 Sibi Sankar       2020-03-12  329  	if (!ind)
fbe639b44a8275 Sibi Sankar       2020-03-12  330  		return;
fbe639b44a8275 Sibi Sankar       2020-03-12  331  
fbe639b44a8275 Sibi Sankar       2020-03-12  332  	ind->transaction_id = ind_msg->transaction_id;
fbe639b44a8275 Sibi Sankar       2020-03-12  333  	ind->curr_state = ind_msg->curr_state;
fbe639b44a8275 Sibi Sankar       2020-03-12  334  	ind->pds = pds;
fbe639b44a8275 Sibi Sankar       2020-03-12  335  
fbe639b44a8275 Sibi Sankar       2020-03-12  336  	mutex_lock(&pdr->list_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  337  	list_add_tail(&ind->node, &pdr->indack_list);
fbe639b44a8275 Sibi Sankar       2020-03-12  338  	mutex_unlock(&pdr->list_lock);
fbe639b44a8275 Sibi Sankar       2020-03-12  339  
fbe639b44a8275 Sibi Sankar       2020-03-12  340  	queue_work(pdr->indack_wq, &pdr->indack_work);
fbe639b44a8275 Sibi Sankar       2020-03-12  341  }
fbe639b44a8275 Sibi Sankar       2020-03-12  342  
fbe639b44a8275 Sibi Sankar       2020-03-12  343  static struct qmi_msg_handler qmi_indication_handler[] = {
fbe639b44a8275 Sibi Sankar       2020-03-12  344  	{
fbe639b44a8275 Sibi Sankar       2020-03-12  345  		.type = QMI_INDICATION,
fbe639b44a8275 Sibi Sankar       2020-03-12  346  		.msg_id = SERVREG_STATE_UPDATED_IND_ID,
fbe639b44a8275 Sibi Sankar       2020-03-12  347  		.ei = servreg_state_updated_ind_ei,
fbe639b44a8275 Sibi Sankar       2020-03-12  348  		.decoded_size = sizeof(struct servreg_state_updated_ind),
fbe639b44a8275 Sibi Sankar       2020-03-12  349  		.fn = pdr_indication_cb,
fbe639b44a8275 Sibi Sankar       2020-03-12  350  	},
fbe639b44a8275 Sibi Sankar       2020-03-12  351  	{}
fbe639b44a8275 Sibi Sankar       2020-03-12  352  };
fbe639b44a8275 Sibi Sankar       2020-03-12  353  
fbe639b44a8275 Sibi Sankar       2020-03-12  354  static int pdr_get_domain_list(struct servreg_get_domain_list_req *req,
fbe639b44a8275 Sibi Sankar       2020-03-12  355  			       struct servreg_get_domain_list_resp *resp,
fbe639b44a8275 Sibi Sankar       2020-03-12  356  			       struct pdr_handle *pdr)
fbe639b44a8275 Sibi Sankar       2020-03-12  357  {
fbe639b44a8275 Sibi Sankar       2020-03-12  358  	struct qmi_txn txn;
fbe639b44a8275 Sibi Sankar       2020-03-12  359  	int ret;
fbe639b44a8275 Sibi Sankar       2020-03-12  360  
fbe639b44a8275 Sibi Sankar       2020-03-12  361  	ret = qmi_txn_init(&pdr->locator_hdl, &txn,
fbe639b44a8275 Sibi Sankar       2020-03-12  362  			   servreg_get_domain_list_resp_ei, resp);
fbe639b44a8275 Sibi Sankar       2020-03-12  363  	if (ret < 0)
fbe639b44a8275 Sibi Sankar       2020-03-12  364  		return ret;
fbe639b44a8275 Sibi Sankar       2020-03-12  365  
fbe639b44a8275 Sibi Sankar       2020-03-12  366  	ret = qmi_send_request(&pdr->locator_hdl,
fbe639b44a8275 Sibi Sankar       2020-03-12  367  			       &pdr->locator_addr,
fbe639b44a8275 Sibi Sankar       2020-03-12  368  			       &txn, SERVREG_GET_DOMAIN_LIST_REQ,
fbe639b44a8275 Sibi Sankar       2020-03-12  369  			       SERVREG_GET_DOMAIN_LIST_REQ_MAX_LEN,
fbe639b44a8275 Sibi Sankar       2020-03-12  370  			       servreg_get_domain_list_req_ei,
fbe639b44a8275 Sibi Sankar       2020-03-12  371  			       req);
fbe639b44a8275 Sibi Sankar       2020-03-12  372  	if (ret < 0) {
fbe639b44a8275 Sibi Sankar       2020-03-12  373  		qmi_txn_cancel(&txn);
fbe639b44a8275 Sibi Sankar       2020-03-12  374  		return ret;
fbe639b44a8275 Sibi Sankar       2020-03-12  375  	}
fbe639b44a8275 Sibi Sankar       2020-03-12  376  
fbe639b44a8275 Sibi Sankar       2020-03-12  377  	ret = qmi_txn_wait(&txn, 5 * HZ);
fbe639b44a8275 Sibi Sankar       2020-03-12  378  	if (ret < 0) {
fbe639b44a8275 Sibi Sankar       2020-03-12  379  		pr_err("PDR: %s get domain list txn wait failed: %d\n",
fbe639b44a8275 Sibi Sankar       2020-03-12  380  		       req->service_name, ret);
fbe639b44a8275 Sibi Sankar       2020-03-12  381  		return ret;
fbe639b44a8275 Sibi Sankar       2020-03-12  382  	}
fbe639b44a8275 Sibi Sankar       2020-03-12  383  
fbe639b44a8275 Sibi Sankar       2020-03-12  384  	if (resp->resp.result != QMI_RESULT_SUCCESS_V01) {
fbe639b44a8275 Sibi Sankar       2020-03-12  385  		pr_err("PDR: %s get domain list failed: 0x%x\n",
fbe639b44a8275 Sibi Sankar       2020-03-12  386  		       req->service_name, resp->resp.error);
fbe639b44a8275 Sibi Sankar       2020-03-12  387  		return -EREMOTEIO;
fbe639b44a8275 Sibi Sankar       2020-03-12  388  	}
fbe639b44a8275 Sibi Sankar       2020-03-12  389  
fbe639b44a8275 Sibi Sankar       2020-03-12  390  	return 0;
fbe639b44a8275 Sibi Sankar       2020-03-12  391  }
fbe639b44a8275 Sibi Sankar       2020-03-12  392  
fbe639b44a8275 Sibi Sankar       2020-03-12  393  static int pdr_locate_service(struct pdr_handle *pdr, struct pdr_service *pds)
fbe639b44a8275 Sibi Sankar       2020-03-12  394  {
fbe639b44a8275 Sibi Sankar       2020-03-12  395  	struct servreg_get_domain_list_resp *resp;
fbe639b44a8275 Sibi Sankar       2020-03-12  396  	struct servreg_get_domain_list_req req;
fbe639b44a8275 Sibi Sankar       2020-03-12  397  	struct servreg_location_entry *entry;
fbe639b44a8275 Sibi Sankar       2020-03-12  398  	int domains_read = 0;
fbe639b44a8275 Sibi Sankar       2020-03-12  399  	int ret, i;
fbe639b44a8275 Sibi Sankar       2020-03-12  400  
fbe639b44a8275 Sibi Sankar       2020-03-12 @401  	resp = kzalloc(sizeof(*resp), GFP_KERNEL);
fbe639b44a8275 Sibi Sankar       2020-03-12  402  	if (!resp)
fbe639b44a8275 Sibi Sankar       2020-03-12  403  		return -ENOMEM;
fbe639b44a8275 Sibi Sankar       2020-03-12  404  
fbe639b44a8275 Sibi Sankar       2020-03-12  405  	/* Prepare req message */
fbe639b44a8275 Sibi Sankar       2020-03-12  406  	strcpy(req.service_name, pds->service_name);
fbe639b44a8275 Sibi Sankar       2020-03-12  407  	req.domain_offset_valid = true;
fbe639b44a8275 Sibi Sankar       2020-03-12  408  	req.domain_offset = 0;
fbe639b44a8275 Sibi Sankar       2020-03-12  409  
fbe639b44a8275 Sibi Sankar       2020-03-12  410  	do {
fbe639b44a8275 Sibi Sankar       2020-03-12  411  		req.domain_offset = domains_read;
fbe639b44a8275 Sibi Sankar       2020-03-12  412  		ret = pdr_get_domain_list(&req, resp, pdr);
fbe639b44a8275 Sibi Sankar       2020-03-12  413  		if (ret < 0)
fbe639b44a8275 Sibi Sankar       2020-03-12  414  			goto out;
fbe639b44a8275 Sibi Sankar       2020-03-12  415  
fbe639b44a8275 Sibi Sankar       2020-03-12  416  		for (i = domains_read; i < resp->domain_list_len; i++) {
fbe639b44a8275 Sibi Sankar       2020-03-12  417  			entry = &resp->domain_list[i];
fbe639b44a8275 Sibi Sankar       2020-03-12  418  
fbe639b44a8275 Sibi Sankar       2020-03-12  419  			if (strnlen(entry->name, sizeof(entry->name)) == sizeof(entry->name))
fbe639b44a8275 Sibi Sankar       2020-03-12  420  				continue;
fbe639b44a8275 Sibi Sankar       2020-03-12  421  
fbe639b44a8275 Sibi Sankar       2020-03-12  422  			if (!strcmp(entry->name, pds->service_path)) {
fbe639b44a8275 Sibi Sankar       2020-03-12  423  				pds->service_data_valid = entry->service_data_valid;
fbe639b44a8275 Sibi Sankar       2020-03-12  424  				pds->service_data = entry->service_data;
fbe639b44a8275 Sibi Sankar       2020-03-12  425  				pds->instance = entry->instance;
fbe639b44a8275 Sibi Sankar       2020-03-12  426  				goto out;
fbe639b44a8275 Sibi Sankar       2020-03-12  427  			}
fbe639b44a8275 Sibi Sankar       2020-03-12  428  		}
fbe639b44a8275 Sibi Sankar       2020-03-12  429  
fbe639b44a8275 Sibi Sankar       2020-03-12  430  		/* Update ret to indicate that the service is not yet found */
fbe639b44a8275 Sibi Sankar       2020-03-12  431  		ret = -ENXIO;
fbe639b44a8275 Sibi Sankar       2020-03-12  432  
fbe639b44a8275 Sibi Sankar       2020-03-12  433  		/* Always read total_domains from the response msg */
fbe639b44a8275 Sibi Sankar       2020-03-12  434  		if (resp->domain_list_len > resp->total_domains)
fbe639b44a8275 Sibi Sankar       2020-03-12  435  			resp->domain_list_len = resp->total_domains;
fbe639b44a8275 Sibi Sankar       2020-03-12  436  
fbe639b44a8275 Sibi Sankar       2020-03-12  437  		domains_read += resp->domain_list_len;
fbe639b44a8275 Sibi Sankar       2020-03-12  438  	} while (domains_read < resp->total_domains);
fbe639b44a8275 Sibi Sankar       2020-03-12  439  out:
fbe639b44a8275 Sibi Sankar       2020-03-12  440  	kfree(resp);
fbe639b44a8275 Sibi Sankar       2020-03-12  441  	return ret;
fbe639b44a8275 Sibi Sankar       2020-03-12  442  }
fbe639b44a8275 Sibi Sankar       2020-03-12  443  

:::::: The code at line 292 was first introduced by commit
:::::: fbe639b44a82755d639df1c5d147c93f02ac5a0f soc: qcom: Introduce Protection Domain Restart helpers

:::::: TO: Sibi Sankar <sibis@codeaurora.org>
:::::: CC: Bjorn Andersson <bjorn.andersson@linaro.org>

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 53390 bytes --]

Re: [PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[kernel test robot]

[-- Attachment #1: Type: text/plain, Size: 3509 bytes --]

Hi Herbert,

I love your patch! Perhaps something to improve:

[auto build test WARNING on linus/master]
[also build test WARNING on v5.7 next-20200611]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]

url:    https://github.com/0day-ci/linux/commits/Herbert-Xu/iov_iter-Move-unnecessary-inclusion-of-crypto-hash-h/20200611-154742
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git b29482fde649c72441d5478a4ea2c52c56d97a5e
config: x86_64-randconfig-s021-20200611 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-13) 9.3.0
reproduce:
        # apt-get install sparse
        # sparse version: v0.6.1-250-g42323db3-dirty
        # save the attached .config to linux build tree
        make W=1 C=1 ARCH=x86_64 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__'

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)

>> net/core/datagram.c:512:25: sparse: sparse: incorrect type in argument 6 (incompatible argument 3 (different base types)) @@     expected unsigned long ( *cb )( ... ) @@     got unsigned long ( * )( ... ) @@
>> net/core/datagram.c:512:25: sparse:     expected unsigned long ( *cb )( ... )
>> net/core/datagram.c:512:25: sparse:     got unsigned long ( * )( ... )
   include/net/sock.h:1603:31: sparse: sparse: context imbalance in '__skb_free_datagram_locked' - unexpected unlock

vim +512 net/core/datagram.c

950fcaecd5cc6c0 Sagi Grimberg 2018-12-03  497  
65d69e2505bb64f Sagi Grimberg 2018-12-03  498  /**
65d69e2505bb64f Sagi Grimberg 2018-12-03  499   *	skb_copy_and_hash_datagram_iter - Copy datagram to an iovec iterator
65d69e2505bb64f Sagi Grimberg 2018-12-03  500   *          and update a hash.
65d69e2505bb64f Sagi Grimberg 2018-12-03  501   *	@skb: buffer to copy
65d69e2505bb64f Sagi Grimberg 2018-12-03  502   *	@offset: offset in the buffer to start copying from
65d69e2505bb64f Sagi Grimberg 2018-12-03  503   *	@to: iovec iterator to copy to
65d69e2505bb64f Sagi Grimberg 2018-12-03  504   *	@len: amount of data to copy from buffer to iovec
65d69e2505bb64f Sagi Grimberg 2018-12-03  505   *      @hash: hash request to update
65d69e2505bb64f Sagi Grimberg 2018-12-03  506   */
65d69e2505bb64f Sagi Grimberg 2018-12-03  507  int skb_copy_and_hash_datagram_iter(const struct sk_buff *skb, int offset,
65d69e2505bb64f Sagi Grimberg 2018-12-03  508  			   struct iov_iter *to, int len,
65d69e2505bb64f Sagi Grimberg 2018-12-03  509  			   struct ahash_request *hash)
65d69e2505bb64f Sagi Grimberg 2018-12-03  510  {
65d69e2505bb64f Sagi Grimberg 2018-12-03  511  	return __skb_datagram_iter(skb, offset, to, len, true,
65d69e2505bb64f Sagi Grimberg 2018-12-03 @512  			hash_and_copy_to_iter, hash);
65d69e2505bb64f Sagi Grimberg 2018-12-03  513  }
65d69e2505bb64f Sagi Grimberg 2018-12-03  514  EXPORT_SYMBOL(skb_copy_and_hash_datagram_iter);
65d69e2505bb64f Sagi Grimberg 2018-12-03  515  

:::::: The code at line 512 was first introduced by commit
:::::: 65d69e2505bb64f6a8d7f417f6e46e2a351174c6 datagram: introduce skb_copy_and_hash_datagram_iter helper

:::::: TO: Sagi Grimberg <sagi@lightbitslabs.com>
:::::: CC: Christoph Hellwig <hch@lst.de>

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 27574 bytes --]

Re: [PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[kernel test robot]

[-- Attachment #1: Type: text/plain, Size: 3486 bytes --]

Hi Herbert,

I love your patch! Perhaps something to improve:

[auto build test WARNING on linus/master]
[also build test WARNING on v5.7 next-20200611]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]

url:    https://github.com/0day-ci/linux/commits/Herbert-Xu/iov_iter-Move-unnecessary-inclusion-of-crypto-hash-h/20200611-154742
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git b29482fde649c72441d5478a4ea2c52c56d97a5e
config: i386-randconfig-s001-20200611 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-13) 9.3.0
reproduce:
        # apt-get install sparse
        # sparse version: v0.6.1-250-g42323db3-dirty
        # save the attached .config to linux build tree
        make W=1 C=1 ARCH=i386 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__'

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)

>> net/core/datagram.c:512:25: sparse: sparse: incorrect type in argument 6 (incompatible argument 3 (different base types)) @@     expected unsigned int ( *cb )( ... ) @@     got unsigned int ( * )( ... ) @@
>> net/core/datagram.c:512:25: sparse:     expected unsigned int ( *cb )( ... )
>> net/core/datagram.c:512:25: sparse:     got unsigned int ( * )( ... )
   include/linux/spinlock.h:398:9: sparse: sparse: context imbalance in '__skb_free_datagram_locked' - unexpected unlock

vim +512 net/core/datagram.c

950fcaecd5cc6c Sagi Grimberg 2018-12-03  497  
65d69e2505bb64 Sagi Grimberg 2018-12-03  498  /**
65d69e2505bb64 Sagi Grimberg 2018-12-03  499   *	skb_copy_and_hash_datagram_iter - Copy datagram to an iovec iterator
65d69e2505bb64 Sagi Grimberg 2018-12-03  500   *          and update a hash.
65d69e2505bb64 Sagi Grimberg 2018-12-03  501   *	@skb: buffer to copy
65d69e2505bb64 Sagi Grimberg 2018-12-03  502   *	@offset: offset in the buffer to start copying from
65d69e2505bb64 Sagi Grimberg 2018-12-03  503   *	@to: iovec iterator to copy to
65d69e2505bb64 Sagi Grimberg 2018-12-03  504   *	@len: amount of data to copy from buffer to iovec
65d69e2505bb64 Sagi Grimberg 2018-12-03  505   *      @hash: hash request to update
65d69e2505bb64 Sagi Grimberg 2018-12-03  506   */
65d69e2505bb64 Sagi Grimberg 2018-12-03  507  int skb_copy_and_hash_datagram_iter(const struct sk_buff *skb, int offset,
65d69e2505bb64 Sagi Grimberg 2018-12-03  508  			   struct iov_iter *to, int len,
65d69e2505bb64 Sagi Grimberg 2018-12-03  509  			   struct ahash_request *hash)
65d69e2505bb64 Sagi Grimberg 2018-12-03  510  {
65d69e2505bb64 Sagi Grimberg 2018-12-03  511  	return __skb_datagram_iter(skb, offset, to, len, true,
65d69e2505bb64 Sagi Grimberg 2018-12-03 @512  			hash_and_copy_to_iter, hash);
65d69e2505bb64 Sagi Grimberg 2018-12-03  513  }
65d69e2505bb64 Sagi Grimberg 2018-12-03  514  EXPORT_SYMBOL(skb_copy_and_hash_datagram_iter);
65d69e2505bb64 Sagi Grimberg 2018-12-03  515  

:::::: The code at line 512 was first introduced by commit
:::::: 65d69e2505bb64f6a8d7f417f6e46e2a351174c6 datagram: introduce skb_copy_and_hash_datagram_iter helper

:::::: TO: Sagi Grimberg <sagi@lightbitslabs.com>
:::::: CC: Christoph Hellwig <hch@lst.de>

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 41966 bytes --]

Re: [PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[kernel test robot]

[-- Attachment #1: Type: text/plain, Size: 11726 bytes --]

Hi Herbert,

I love your patch! Yet something to improve:

[auto build test ERROR on linus/master]
[also build test ERROR on v5.7 next-20200611]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]

url:    https://github.com/0day-ci/linux/commits/Herbert-Xu/iov_iter-Move-unnecessary-inclusion-of-crypto-hash-h/20200611-154742
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git b29482fde649c72441d5478a4ea2c52c56d97a5e
config: x86_64-randconfig-m001-20200611 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-13) 9.3.0

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All error/warnings (new ones prefixed by >>, old ones prefixed by <<):

drivers/mtd/mtdpstore.c: In function 'mtdpstore_notify_add':
>> drivers/mtd/mtdpstore.c:419:15: error: implicit declaration of function 'kcalloc' [-Werror=implicit-function-declaration]
419 |  cxt->rmmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
|               ^~~~~~~
>> drivers/mtd/mtdpstore.c:419:13: warning: assignment to 'long unsigned int *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
419 |  cxt->rmmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
|             ^
drivers/mtd/mtdpstore.c:420:15: warning: assignment to 'long unsigned int *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
420 |  cxt->usedmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
|               ^
drivers/mtd/mtdpstore.c:423:14: warning: assignment to 'long unsigned int *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
423 |  cxt->badmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
|              ^
drivers/mtd/mtdpstore.c: In function 'mtdpstore_flush_removed_do':
>> drivers/mtd/mtdpstore.c:452:8: error: implicit declaration of function 'kmalloc' [-Werror=implicit-function-declaration]
452 |  buf = kmalloc(mtd->erasesize, GFP_KERNEL);
|        ^~~~~~~
>> drivers/mtd/mtdpstore.c:452:6: warning: assignment to 'u_char *' {aka 'unsigned char *'} from 'int' makes pointer from integer without a cast [-Wint-conversion]
452 |  buf = kmalloc(mtd->erasesize, GFP_KERNEL);
|      ^
>> drivers/mtd/mtdpstore.c:485:2: error: implicit declaration of function 'kfree' [-Werror=implicit-function-declaration]
485 |  kfree(buf);
|  ^~~~~
cc1: some warnings being treated as errors
--
drivers/dma/sf-pdma/sf-pdma.c: In function 'sf_pdma_alloc_desc':
>> drivers/dma/sf-pdma/sf-pdma.c:65:9: error: implicit declaration of function 'kzalloc'; did you mean 'vzalloc'? [-Werror=implicit-function-declaration]
65 |  desc = kzalloc(sizeof(*desc), GFP_NOWAIT);
|         ^~~~~~~
|         vzalloc
>> drivers/dma/sf-pdma/sf-pdma.c:65:7: warning: assignment to 'struct sf_pdma_desc *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
65 |  desc = kzalloc(sizeof(*desc), GFP_NOWAIT);
|       ^
drivers/dma/sf-pdma/sf-pdma.c: In function 'sf_pdma_free_chan_resources':
>> drivers/dma/sf-pdma/sf-pdma.c:155:2: error: implicit declaration of function 'kfree'; did you mean 'vfree'? [-Werror=implicit-function-declaration]
155 |  kfree(chan->desc);
|  ^~~~~
|  vfree
cc1: some warnings being treated as errors

vim +/kcalloc +419 drivers/mtd/mtdpstore.c

78c08247b9d3e0 WeiXiong Liao 2020-03-25  379  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  380  static void mtdpstore_notify_add(struct mtd_info *mtd)
78c08247b9d3e0 WeiXiong Liao 2020-03-25  381  {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  382  	int ret;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  383  	struct mtdpstore_context *cxt = &oops_cxt;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  384  	struct pstore_blk_config *info = &cxt->info;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  385  	unsigned long longcnt;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  386  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  387  	if (!strcmp(mtd->name, info->device))
78c08247b9d3e0 WeiXiong Liao 2020-03-25  388  		cxt->index = mtd->index;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  389  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  390  	if (mtd->index != cxt->index || cxt->index < 0)
78c08247b9d3e0 WeiXiong Liao 2020-03-25  391  		return;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  392  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  393  	dev_dbg(&mtd->dev, "found matching MTD device %s\n", mtd->name);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  394  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  395  	if (mtd->size < info->kmsg_size * 2) {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  396  		dev_err(&mtd->dev, "MTD partition %d not big enough\n",
78c08247b9d3e0 WeiXiong Liao 2020-03-25  397  				mtd->index);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  398  		return;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  399  	}
78c08247b9d3e0 WeiXiong Liao 2020-03-25  400  	/*
78c08247b9d3e0 WeiXiong Liao 2020-03-25  401  	 * kmsg_size must be aligned to 4096 Bytes, which is limited by
78c08247b9d3e0 WeiXiong Liao 2020-03-25  402  	 * psblk. The default value of kmsg_size is 64KB. If kmsg_size
78c08247b9d3e0 WeiXiong Liao 2020-03-25  403  	 * is larger than erasesize, some errors will occur since mtdpsotre
78c08247b9d3e0 WeiXiong Liao 2020-03-25  404  	 * is designed on it.
78c08247b9d3e0 WeiXiong Liao 2020-03-25  405  	 */
78c08247b9d3e0 WeiXiong Liao 2020-03-25  406  	if (mtd->erasesize < info->kmsg_size) {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  407  		dev_err(&mtd->dev, "eraseblock size of MTD partition %d too small\n",
78c08247b9d3e0 WeiXiong Liao 2020-03-25  408  				mtd->index);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  409  		return;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  410  	}
78c08247b9d3e0 WeiXiong Liao 2020-03-25  411  	if (unlikely(info->kmsg_size % mtd->writesize)) {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  412  		dev_err(&mtd->dev, "record size %lu KB must align to write size %d KB\n",
78c08247b9d3e0 WeiXiong Liao 2020-03-25  413  				info->kmsg_size / 1024,
78c08247b9d3e0 WeiXiong Liao 2020-03-25  414  				mtd->writesize / 1024);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  415  		return;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  416  	}
78c08247b9d3e0 WeiXiong Liao 2020-03-25  417  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  418  	longcnt = BITS_TO_LONGS(div_u64(mtd->size, info->kmsg_size));
78c08247b9d3e0 WeiXiong Liao 2020-03-25 @419  	cxt->rmmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  420  	cxt->usedmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  421  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  422  	longcnt = BITS_TO_LONGS(div_u64(mtd->size, mtd->erasesize));
78c08247b9d3e0 WeiXiong Liao 2020-03-25  423  	cxt->badmap = kcalloc(longcnt, sizeof(long), GFP_KERNEL);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  424  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  425  	cxt->dev.total_size = mtd->size;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  426  	/* just support dmesg right now */
78c08247b9d3e0 WeiXiong Liao 2020-03-25  427  	cxt->dev.flags = PSTORE_FLAGS_DMESG;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  428  	cxt->dev.read = mtdpstore_read;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  429  	cxt->dev.write = mtdpstore_write;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  430  	cxt->dev.erase = mtdpstore_erase;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  431  	cxt->dev.panic_write = mtdpstore_panic_write;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  432  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  433  	ret = register_pstore_device(&cxt->dev);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  434  	if (ret) {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  435  		dev_err(&mtd->dev, "mtd%d register to psblk failed\n",
78c08247b9d3e0 WeiXiong Liao 2020-03-25  436  				mtd->index);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  437  		return;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  438  	}
78c08247b9d3e0 WeiXiong Liao 2020-03-25  439  	cxt->mtd = mtd;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  440  	dev_info(&mtd->dev, "Attached to MTD device %d\n", mtd->index);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  441  }
78c08247b9d3e0 WeiXiong Liao 2020-03-25  442  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  443  static int mtdpstore_flush_removed_do(struct mtdpstore_context *cxt,
78c08247b9d3e0 WeiXiong Liao 2020-03-25  444  		loff_t off, size_t size)
78c08247b9d3e0 WeiXiong Liao 2020-03-25  445  {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  446  	struct mtd_info *mtd = cxt->mtd;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  447  	u_char *buf;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  448  	int ret;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  449  	size_t retlen;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  450  	struct erase_info erase;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  451  
78c08247b9d3e0 WeiXiong Liao 2020-03-25 @452  	buf = kmalloc(mtd->erasesize, GFP_KERNEL);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  453  	if (!buf)
78c08247b9d3e0 WeiXiong Liao 2020-03-25  454  		return -ENOMEM;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  455  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  456  	/* 1st. read to cache */
78c08247b9d3e0 WeiXiong Liao 2020-03-25  457  	ret = mtd_read(mtd, off, mtd->erasesize, &retlen, buf);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  458  	if (mtdpstore_is_io_error(ret))
78c08247b9d3e0 WeiXiong Liao 2020-03-25  459  		goto free;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  460  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  461  	/* 2nd. erase block */
78c08247b9d3e0 WeiXiong Liao 2020-03-25  462  	erase.len = mtd->erasesize;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  463  	erase.addr = off;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  464  	ret = mtd_erase(mtd, &erase);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  465  	if (ret)
78c08247b9d3e0 WeiXiong Liao 2020-03-25  466  		goto free;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  467  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  468  	/* 3rd. write back */
78c08247b9d3e0 WeiXiong Liao 2020-03-25  469  	while (size) {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  470  		unsigned int zonesize = cxt->info.kmsg_size;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  471  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  472  		/* there is valid data on block, write back */
78c08247b9d3e0 WeiXiong Liao 2020-03-25  473  		if (mtdpstore_is_used(cxt, off)) {
78c08247b9d3e0 WeiXiong Liao 2020-03-25  474  			ret = mtd_write(mtd, off, zonesize, &retlen, buf);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  475  			if (ret)
78c08247b9d3e0 WeiXiong Liao 2020-03-25  476  				dev_err(&mtd->dev, "write failure at %lld (%zu of %u written), err %d\n",
78c08247b9d3e0 WeiXiong Liao 2020-03-25  477  						off, retlen, zonesize, ret);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  478  		}
78c08247b9d3e0 WeiXiong Liao 2020-03-25  479  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  480  		off += zonesize;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  481  		size -= min_t(unsigned int, zonesize, size);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  482  	}
78c08247b9d3e0 WeiXiong Liao 2020-03-25  483  
78c08247b9d3e0 WeiXiong Liao 2020-03-25  484  free:
78c08247b9d3e0 WeiXiong Liao 2020-03-25 @485  	kfree(buf);
78c08247b9d3e0 WeiXiong Liao 2020-03-25  486  	return ret;
78c08247b9d3e0 WeiXiong Liao 2020-03-25  487  }
78c08247b9d3e0 WeiXiong Liao 2020-03-25  488  

:::::: The code at line 419 was first introduced by commit
:::::: 78c08247b9d3e03192f8b359aa079024e805a948 mtd: Support kmsg dumper based on pstore/blk

:::::: TO: WeiXiong Liao <liaoweixiong@allwinnertech.com>
:::::: CC: Kees Cook <keescook@chromium.org>

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 33706 bytes --]

[v2 PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[Herbert Xu]

The header file linux/uio.h includes crypto/hash.h which pulls in
most of the Crypto API.  Since linux/uio.h is used throughout the
kernel this means that every tiny bit of change to the Crypto API
causes the entire kernel to get rebuilt.

This patch fixes this by moving it into lib/iov_iter.c instead
where it is actually used.

This patch also fixes the ifdef to use CRYPTO_HASH instead of just
CRYPTO which does not guarantee the existence of ahash.

Unfortunately a number of drivers were relying on linux/uio.h to
provide access to linux/slab.h.  This patch adds inclusions of
linux/slab.h as detected by build failures.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/dma/sf-pdma/sf-pdma.c b/drivers/dma/sf-pdma/sf-pdma.c
index 6d0bec947636..e237d6038407 100644
--- a/drivers/dma/sf-pdma/sf-pdma.c
+++ b/drivers/dma/sf-pdma/sf-pdma.c
@@ -20,6 +20,7 @@
 #include <linux/mod_devicetable.h>
 #include <linux/dma-mapping.h>
 #include <linux/of.h>
+#include <linux/slab.h>
 
 #include "sf-pdma.h"
 
diff --git a/drivers/misc/uacce/uacce.c b/drivers/misc/uacce/uacce.c
index d39307f060bd..5a984df0e95e 100644
--- a/drivers/misc/uacce/uacce.c
+++ b/drivers/misc/uacce/uacce.c
@@ -4,6 +4,7 @@
 #include <linux/iommu.h>
 #include <linux/module.h>
 #include <linux/poll.h>
+#include <linux/slab.h>
 #include <linux/uacce.h>
 
 static struct class *uacce_class;
diff --git a/drivers/soc/qcom/pdr_interface.c b/drivers/soc/qcom/pdr_interface.c
index 17ad3b8698e1..aa2e3fe19c0f 100644
--- a/drivers/soc/qcom/pdr_interface.c
+++ b/drivers/soc/qcom/pdr_interface.c
@@ -5,6 +5,7 @@
 
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/workqueue.h>
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 320d1062068d..d1e03b8cb6bb 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3,6 +3,7 @@
  * Copyright (C) 2007 Oracle.  All rights reserved.
  */
 
+#include <crypto/hash.h>
 #include <linux/kernel.h>
 #include <linux/bio.h>
 #include <linux/buffer_head.h>
diff --git a/include/linux/uio.h b/include/linux/uio.h
index 9576fd8158d7..3835a8a8e9ea 100644
--- a/include/linux/uio.h
+++ b/include/linux/uio.h
@@ -7,7 +7,6 @@
 
 #include <linux/kernel.h>
 #include <linux/thread_info.h>
-#include <crypto/hash.h>
 #include <uapi/linux/uio.h>
 
 struct page;
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 51595bf3af85..2830daf46c73 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -1,4 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0-only
+#include <crypto/hash.h>
 #include <linux/export.h>
 #include <linux/bvec.h>
 #include <linux/uio.h>
@@ -1566,7 +1567,7 @@ EXPORT_SYMBOL(csum_and_copy_to_iter);
 size_t hash_and_copy_to_iter(const void *addr, size_t bytes, void *hashp,
 		struct iov_iter *i)
 {
-#ifdef CONFIG_CRYPTO
+#ifdef CONFIG_CRYPTO_HASH
 	struct ahash_request *hash = hashp;
 	struct scatterlist sg;
 	size_t copied;
diff --git a/drivers/mtd/mtdpstore.c b/drivers/mtd/mtdpstore.c
index a4fe6060b960..a3ae8778f6a9 100644
--- a/drivers/mtd/mtdpstore.c
+++ b/drivers/mtd/mtdpstore.c
@@ -7,6 +7,7 @@
 #include <linux/pstore_blk.h>
 #include <linux/mtd/mtd.h>
 #include <linux/bitops.h>
+#include <linux/slab.h>
 
 static struct mtdpstore_context {
 	int index;
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

[v3 PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[Herbert Xu]

The header file linux/uio.h includes crypto/hash.h which pulls in
most of the Crypto API.  Since linux/uio.h is used throughout the
kernel this means that every tiny bit of change to the Crypto API
causes the entire kernel to get rebuilt.

This patch fixes this by moving it into lib/iov_iter.c instead
where it is actually used.

This patch also fixes the ifdef to use CRYPTO_HASH instead of just
CRYPTO which does not guarantee the existence of ahash.

Unfortunately a number of drivers were relying on linux/uio.h to
provide access to linux/slab.h.  This patch adds inclusions of
linux/slab.h as detected by build failures.

Also skbuff.h was relying on this to provide a declaration for
ahash_request.  This patch adds a forward declaration instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/drivers/dma/sf-pdma/sf-pdma.c b/drivers/dma/sf-pdma/sf-pdma.c
index 6d0bec947636..e237d6038407 100644
--- a/drivers/dma/sf-pdma/sf-pdma.c
+++ b/drivers/dma/sf-pdma/sf-pdma.c
@@ -20,6 +20,7 @@
 #include <linux/mod_devicetable.h>
 #include <linux/dma-mapping.h>
 #include <linux/of.h>
+#include <linux/slab.h>
 
 #include "sf-pdma.h"
 
diff --git a/drivers/misc/uacce/uacce.c b/drivers/misc/uacce/uacce.c
index d39307f060bd..5a984df0e95e 100644
--- a/drivers/misc/uacce/uacce.c
+++ b/drivers/misc/uacce/uacce.c
@@ -4,6 +4,7 @@
 #include <linux/iommu.h>
 #include <linux/module.h>
 #include <linux/poll.h>
+#include <linux/slab.h>
 #include <linux/uacce.h>
 
 static struct class *uacce_class;
diff --git a/drivers/soc/qcom/pdr_interface.c b/drivers/soc/qcom/pdr_interface.c
index 17ad3b8698e1..aa2e3fe19c0f 100644
--- a/drivers/soc/qcom/pdr_interface.c
+++ b/drivers/soc/qcom/pdr_interface.c
@@ -5,6 +5,7 @@
 
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/workqueue.h>
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 320d1062068d..d1e03b8cb6bb 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3,6 +3,7 @@
  * Copyright (C) 2007 Oracle.  All rights reserved.
  */
 
+#include <crypto/hash.h>
 #include <linux/kernel.h>
 #include <linux/bio.h>
 #include <linux/buffer_head.h>
diff --git a/include/linux/uio.h b/include/linux/uio.h
index 9576fd8158d7..3835a8a8e9ea 100644
--- a/include/linux/uio.h
+++ b/include/linux/uio.h
@@ -7,7 +7,6 @@
 
 #include <linux/kernel.h>
 #include <linux/thread_info.h>
-#include <crypto/hash.h>
 #include <uapi/linux/uio.h>
 
 struct page;
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 51595bf3af85..2830daf46c73 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -1,4 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0-only
+#include <crypto/hash.h>
 #include <linux/export.h>
 #include <linux/bvec.h>
 #include <linux/uio.h>
@@ -1566,7 +1567,7 @@ EXPORT_SYMBOL(csum_and_copy_to_iter);
 size_t hash_and_copy_to_iter(const void *addr, size_t bytes, void *hashp,
 		struct iov_iter *i)
 {
-#ifdef CONFIG_CRYPTO
+#ifdef CONFIG_CRYPTO_HASH
 	struct ahash_request *hash = hashp;
 	struct scatterlist sg;
 	size_t copied;
diff --git a/drivers/mtd/mtdpstore.c b/drivers/mtd/mtdpstore.c
index a4fe6060b960..a3ae8778f6a9 100644
--- a/drivers/mtd/mtdpstore.c
+++ b/drivers/mtd/mtdpstore.c
@@ -7,6 +7,7 @@
 #include <linux/pstore_blk.h>
 #include <linux/mtd/mtd.h>
 #include <linux/bitops.h>
+#include <linux/slab.h>
 
 static struct mtdpstore_context {
 	int index;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 3a2ac7072dbb..36df5998d23c 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -238,6 +238,7 @@
 			 SKB_DATA_ALIGN(sizeof(struct sk_buff)) +	\
 			 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
 
+struct ahash_request;
 struct net_device;
 struct scatterlist;
 struct pipe_inode_info;
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [v3 PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[Christoph Hellwig]

On Fri, Jun 12, 2020 at 04:57:37PM +1000, Herbert Xu wrote:
> The header file linux/uio.h includes crypto/hash.h which pulls in
> most of the Crypto API.  Since linux/uio.h is used throughout the
> kernel this means that every tiny bit of change to the Crypto API
> causes the entire kernel to get rebuilt.
> 
> This patch fixes this by moving it into lib/iov_iter.c instead
> where it is actually used.
> 
> This patch also fixes the ifdef to use CRYPTO_HASH instead of just
> CRYPTO which does not guarantee the existence of ahash.
> 
> Unfortunately a number of drivers were relying on linux/uio.h to
> provide access to linux/slab.h.  This patch adds inclusions of
> linux/slab.h as detected by build failures.
> 
> Also skbuff.h was relying on this to provide a declaration for
> ahash_request.  This patch adds a forward declaration instead.
> 
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [v3 PATCH] iov_iter: Move unnecessary inclusion of crypto/hash.h

[Al Viro]

On Fri, Jun 12, 2020 at 04:57:37PM +1000, Herbert Xu wrote:
> The header file linux/uio.h includes crypto/hash.h which pulls in
> most of the Crypto API.  Since linux/uio.h is used throughout the
> kernel this means that every tiny bit of change to the Crypto API
> causes the entire kernel to get rebuilt.
> 
> This patch fixes this by moving it into lib/iov_iter.c instead
> where it is actually used.
> 
> This patch also fixes the ifdef to use CRYPTO_HASH instead of just
> CRYPTO which does not guarantee the existence of ahash.
> 
> Unfortunately a number of drivers were relying on linux/uio.h to
> provide access to linux/slab.h.  This patch adds inclusions of
> linux/slab.h as detected by build failures.
> 
> Also skbuff.h was relying on this to provide a declaration for
> ahash_request.  This patch adds a forward declaration instead.
> 
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

Applied; let it sit in -next for a while to get better build coverage...