[parisc-linux] DIFF 2.6.0-test10 SATA dma_address

[parisc-linux] DIFF 2.6.0-test10 SATA dma_address

[Grant Grundler]

Willy, Alan,

SATA support attempts to directly reference scattergather.dma_address
and .dma_length. That's wrong since those are arch specific fields.
alpha, cris, parisc, ppc64, sparc, and sparc64 won't build SATA.

SATA code needs to use "sg_dma_address()" and sg_dma_len()" macros
to reference those fields.

If this patch looks good, please tell me which forum (linux-scsi?)
this needs to be posted to.

Patch appended and also available from:
	ftp://ftp.parisc-linux.org/patches/diff-2.6.0-t10-sata

The code now compiles but I've not yet built a kernel (forgot to
disable KALLSYMS - toolchain bugs) and don't have the HW to test it. 

cheers,
grant


Index: drivers/scsi/libata-core.c
===================================================================
RCS file: /var/cvs/linux-2.6/drivers/scsi/libata-core.c,v
retrieving revision 1.3
diff -u -p -r1.3 libata-core.c
--- drivers/scsi/libata-core.c	24 Nov 2003 03:16:17 -0000	1.3
+++ drivers/scsi/libata-core.c	24 Nov 2003 06:28:18 -0000
@@ -1627,8 +1627,8 @@ static void ata_sg_clean(struct ata_queu
 	if (cmd->use_sg)
 		pci_unmap_sg(ap->host_set->pdev, sg, qc->n_elem, dir);
 	else
-		pci_unmap_single(ap->host_set->pdev, sg[0].dma_address,
-				 sg[0].length, dir);
+		pci_unmap_single(ap->host_set->pdev, sg_dma_address(&sg[0]),
+				 sg_dma_len(&sg[0]), dir);
 
 	qc->flags &= ~ATA_QCFLAG_SG;
 	qc->sg = NULL;
@@ -1651,8 +1651,8 @@ void ata_fill_sg(struct ata_queued_cmd *
 	assert(qc->n_elem > 0);
 
 	for (i = 0; i < qc->n_elem; i++) {
-		ap->prd[i].addr = cpu_to_le32(sg[i].dma_address);
-		ap->prd[i].flags_len = cpu_to_le32(sg[i].length);
+		ap->prd[i].addr = cpu_to_le32(sg_dma_address(&sg[i]));
+		ap->prd[i].flags_len = cpu_to_le32(sg_dma_len(&sg[i]));
 		VPRINTK("PRD[%u] = (0x%X, 0x%X)\n",
 			i, le32_to_cpu(ap->prd[i].addr), le32_to_cpu(ap->prd[i].flags_len));
 	}
@@ -1683,12 +1683,12 @@ static int ata_sg_setup_one(struct ata_q
 
 	sg->page = virt_to_page(cmd->request_buffer);
 	sg->offset = (unsigned long) cmd->request_buffer & ~PAGE_MASK;
-	sg->length = cmd->request_bufflen;
+	sg_dma_len(sg) = cmd->request_bufflen;
 
 	if (!have_sg)
 		return 0;
 
-	sg->dma_address = pci_map_single(ap->host_set->pdev,
+	sg_dma_address(sg) = pci_map_single(ap->host_set->pdev,
 					 cmd->request_buffer,
 					 cmd->request_bufflen, dir);
 
@@ -1909,7 +1909,7 @@ static void ata_pio_sector(struct ata_po
 	qc->cursg_ofs++;
 
 	if (cmd->use_sg)
-		if ((qc->cursg_ofs * ATA_SECT_SIZE) == sg[qc->cursg].length) {
+		if ((qc->cursg_ofs * ATA_SECT_SIZE) == sg_dma_len(&sg[qc->cursg])) {
 			qc->cursg++;
 			qc->cursg_ofs = 0;
 		}
Index: drivers/scsi/sata_promise.c
===================================================================
RCS file: /var/cvs/linux-2.6/drivers/scsi/sata_promise.c,v
retrieving revision 1.3
diff -u -p -r1.3 sata_promise.c
--- drivers/scsi/sata_promise.c	24 Nov 2003 03:16:17 -0000	1.3
+++ drivers/scsi/sata_promise.c	24 Nov 2003 06:28:18 -0000
@@ -603,8 +603,8 @@ static void pdc20621_fill_sg(struct ata_
 	last = qc->n_elem;
 	idx = 0;
 	for (i = 0; i < last; i++) {
-		buf[idx++] = cpu_to_le32(sg[i].dma_address);
-		buf[idx++] = cpu_to_le32(sg[i].length);
+		buf[idx++] = cpu_to_le32(sg_dma_address(&sg[i]));
+		buf[idx++] = cpu_to_le32(sg_dma_len(&sg[i]));
 		total_len += sg[i].length;
 	}
 	buf[idx - 1] |= cpu_to_le32(ATA_PRD_EOT);

Re: [parisc-linux] DIFF 2.6.0-test10 SATA dma_address

[Joel Soete]

Hi Grant,

Grant Grundler wrote:
> Willy, Alan,
> 
> SATA support attempts to directly reference scattergather.dma_address
> and .dma_length. That's wrong since those are arch specific fields.
> alpha, cris, parisc, ppc64, sparc, and sparc64 won't build SATA.
> 
> SATA code needs to use "sg_dma_address()" and sg_dma_len()" macros
> to reference those fields.
> 
> If this patch looks good, please tell me which forum (linux-scsi?)
> this needs to be posted to.
> 
Hmm I don't see any follow up but i just read in "post-halloween-2.6.txt"
(Reporting bugs. section) that
[snip]
     SCSI      - linux-scsi@vger.kernel.org
[snip]


> Patch appended and also available from:
> 	ftp://ftp.parisc-linux.org/patches/diff-2.6.0-t10-sata
> 
> The code now compiles but I've not yet built a kernel (forgot to
> disable KALLSYMS - toolchain bugs) and don't have the HW to test it. 
> 

hth,
	Joel

PS: Sorry I miss to follow this patch :(

Re: [parisc-linux] DIFF 2.6.0-test10 SATA dma_address

[James Bottomley]

On Wed, 2003-12-24 at 07:56, Joel Soete wrote:
> Hmm I don't see any follow up but i just read in "post-halloween-2.6.txt"
> (Reporting bugs. section) that
> [snip]
>      SCSI      - linux-scsi@vger.kernel.org
> [snip]

The SATA maintainer (Jeff Garzik) has already applied this patch.

James