Minor improvements for Host1x driver

index a10643aa89aa..a954bd41aa79 100644
@@ -27,10 +27,13 @@ struct host1x_job *host1x_job_alloc(struct host1x_channel *ch,
 				    u32 num_cmdbufs, u32 num_relocs)
 {
 	struct host1x_job *job = NULL;
-	unsigned int num_unpins = num_cmdbufs + num_relocs;
+	unsigned int num_unpins = num_relocs;
 	u64 total;
 	void *mem;
 
+	if (!IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL))
+		num_unpins += num_cmdbufs;
+
 	/* Check that we're not going to overflow */
 	total = sizeof(struct host1x_job) +
 		(u64)num_relocs * sizeof(struct host1x_reloc) +

@@ -183,6 +186,13 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 		job->num_unpins++;
 	}
 
+	/*
+	 * We will copy gathers BO content later, so there is no need to
+	 * hold and pin them.
+	 */
+	if (IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL))
+		return 0;
+
 	for (i = 0; i < job->num_gathers; i++) {
 		struct host1x_job_gather *g = &job->gathers[i];
 		size_t gather_size = 0;

@@ -216,7 +226,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 			goto unpin;
 		}
 
-		if (!IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL) && host->domain) {
+		if (host->domain) {
 			for_each_sg(sgt->sgl, sg, sgt->nents, j)
 				gather_size += sg->length;
 			gather_size = iova_align(&host->iova, gather_size);

index a954bd41aa79..89b6c14b7392 100644
@@ -105,6 +105,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 {
 	struct host1x_client *client = job->client;
 	struct device *dev = client->dev;
+	struct host1x_job_gather *g;
 	struct iommu_domain *domain;
 	unsigned int i;
 	int err;

@@ -194,7 +195,6 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 		return 0;
 
 	for (i = 0; i < job->num_gathers; i++) {
-		struct host1x_job_gather *g = &job->gathers[i];
 		size_t gather_size = 0;
 		struct scatterlist *sg;
 		struct sg_table *sgt;

@@ -204,6 +204,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 		dma_addr_t *phys;
 		unsigned int j;
 
+		g = &job->gathers[i];
 		g->bo = host1x_bo_get(g->bo);
 		if (!g->bo) {
 			err = -EINVAL;

@@ -223,7 +224,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 		sgt = host1x_bo_pin(host->dev, g->bo, phys);
 		if (IS_ERR(sgt)) {
 			err = PTR_ERR(sgt);
-			goto unpin;
+			goto put;
 		}
 
 		if (host->domain) {

@@ -236,7 +237,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 					   host->iova_end >> shift, true);
 			if (!alloc) {
 				err = -ENOMEM;
-				goto unpin;
+				goto put;
 			}
 
 			err = iommu_map_sg(host->domain,

@@ -245,7 +246,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 			if (err == 0) {
 				__free_iova(&host->iova, alloc);
 				err = -EINVAL;
-				goto unpin;
+				goto put;
 			}
 
 			job->unpins[job->num_unpins].size = gather_size;

@@ -255,7 +256,7 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 					 DMA_TO_DEVICE);
 			if (!err) {
 				err = -ENOMEM;
-				goto unpin;
+				goto put;
 			}
 
 			job->unpins[job->num_unpins].dir = DMA_TO_DEVICE;

@@ -273,6 +274,8 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job)
 
 	return 0;
 
+put:
+	host1x_bo_put(g->bo);
 unpin:
 	host1x_job_unpin(job);
 	return err;

index c0392672a842..1b4997bda1c7 100644
@@ -16,6 +16,8 @@
 #include "debug.h"
 #include "channel.h"
 
+static DEFINE_MUTEX(debug_lock);
+
 unsigned int host1x_debug_trace_cmdbuf;
 
 static pid_t host1x_debug_force_timeout_pid;

@@ -52,12 +54,14 @@ static int show_channel(struct host1x_channel *ch, void *data, bool show_fifo)
 	struct output *o = data;
 
 	mutex_lock(&ch->cdma.lock);
+	mutex_lock(&debug_lock);
 
 	if (show_fifo)
 		host1x_hw_show_channel_fifo(m, ch, o);
 
 	host1x_hw_show_channel_cdma(m, ch, o);
 
+	mutex_unlock(&debug_lock);
 	mutex_unlock(&ch->cdma.lock);
 
 	return 0;

index 02125842071c..c25c3fe0a295 100644
@@ -192,8 +192,14 @@ static void show_gather(struct output *o, phys_addr_t phys_addr,
 
 static void show_channel_gathers(struct output *o, struct host1x_cdma *cdma)
 {
+	struct push_buffer *pb = &cdma->push_buffer;
 	struct host1x_job *job;
 
+	host1x_debug_output(o, "Push Buffer at %08x, %d words\n",
+			    pb->dma, pb->size / 4);
+
+	show_gather(o, pb->dma, pb->size / 4, cdma, pb->dma, pb->mapped);
+
 	list_for_each_entry(job, &cdma->sync_queue, list) {
 		unsigned int i;