From: Thomas Huth <thuth@redhat.com>
To: jrossi@linux.ibm.com, qemu-devel@nongnu.org, qemu-s390x@nongnu.org
Cc: frankja@linux.ibm.com
Subject: Re: [PATCH 15/18] hw/s390x: Build an IPLB for each boot device
Date: Mon, 30 Sep 2024 13:59:12 +0200 [thread overview]
Message-ID: <b3674e3b-5a23-41a0-a100-7fb2810a2775@redhat.com> (raw)
In-Reply-To: <20240927005117.1679506-16-jrossi@linux.ibm.com>
On 27/09/2024 02.51, jrossi@linux.ibm.com wrote:
> From: Jared Rossi <jrossi@linux.ibm.com>
>
> Build an IPLB for any device with a bootindex (up to a maximum of 8 devices).
>
> The IPLB chain is placed immediately before the BIOS in memory. Because this
> is not a fixed address, the location of the next IPLB and number of remaining
> boot devices is stored in the QIPL global variable for possible later access by
> the guest during IPL.
>
> Signed-off-by: Jared Rossi <jrossi@linux.ibm.com>
>
Note: There's an empty line below your SoB in most of your patch
descriptions ... it's just cosmetics, but in case you touch the patch
anyway, please remove it.
> ---
...
> @@ -484,8 +499,8 @@ static bool s390_gen_initial_iplb(S390IPLState *ipl)
> lp = S390_CCW_MACHINE(qdev_get_machine())->loadparm;
> }
>
> - s390_ipl_convert_loadparm((char *)lp, ipl->iplb.loadparm);
> - ipl->iplb.flags |= DIAG308_FLAGS_LP_VALID;
> + s390_ipl_convert_loadparm((char *)lp, iplb->loadparm);
> + iplb->flags |= DIAG308_FLAGS_LP_VALID;
>
> return true;
> }
> @@ -493,6 +508,58 @@ static bool s390_gen_initial_iplb(S390IPLState *ipl)
> return false;
> }
>
> +static bool s390_init_all_iplbs(S390IPLState *ipl)
> +{
> + int iplb_num = 0;
> + IplParameterBlock iplb_chain[7];
> + DeviceState *dev_st = get_boot_device(0);
> +
> + /*
> + * Parse the boot devices. Generate an IPLB for the first boot device,
> + * which will later be set with DIAG308. Index any fallback boot devices.
> + */
The comment looks like it rather belongs to the whole function, and not to
the if-statement below? So maybe move it at the top?
> + if (!dev_st) {
> + ipl->qipl.chain_len = 0;
> + return false;
> + }
> +
> + iplb_num = 1;
> + s390_build_iplb(dev_st, &ipl->iplb);
> + ipl->iplb.flags |= DIAG308_FLAGS_LP_VALID;
Isn't DIAG308_FLAGS_LP_VALID set within s390_build_iplb() already?
> + while (get_boot_device(iplb_num)) {
> + iplb_num++;
> + }
I'd maybe move the code block below to this spot here, so you've got to
assign ipl->qipl.chain_len only once:
+ if (iplb_num > MAX_IPLB_CHAIN + 1) {
+ warn_report("Excess boot devices defined! %d boot devices found, "
+ "but only the first %d will be considered.",
+ iplb_num, MAX_IPLB_CHAIN + 1);
+
+ iplb_num = MAX_IPLB_CHAIN + 1;
+ }
> + ipl->qipl.chain_len = iplb_num - 1;
> +
> + /*
> + * Build fallback IPLBs for any boot devices above index 0, up to a
> + * maximum amount as defined in ipl.h
> + */
> + if (iplb_num > 1) {
> + if (iplb_num > MAX_IPLB_CHAIN + 1) {
> + warn_report("Excess boot devices defined! %d boot devices found, "
> + "but only the first %d will be considered.",
> + iplb_num, MAX_IPLB_CHAIN + 1);
> +
> + ipl->qipl.chain_len = MAX_IPLB_CHAIN;
> + iplb_num = MAX_IPLB_CHAIN + 1;
> + }
i.e. move this code block -^
and remove the "ipl->qipl.chain_len = MAX_IPLB_CHAIN;" in there.
> + /* Start at 1 because the IPLB for boot index 0 is not chained */
> + for (int i = 1; i < iplb_num; i++) {
> + dev_st = get_boot_device(i);
> + s390_build_iplb(dev_st, &iplb_chain[i - 1]);
> + iplb_chain[i - 1].flags |= DIAG308_FLAGS_LP_VALID;
Again, setting DIAG308_FLAGS_LP_VALID should not be necessary since it is
already done in s390_build_iplb() ?
> + }
> +
> + ipl->qipl.next_iplb = s390_ipl_map_iplb_chain(iplb_chain);
> + }
> +
> + return iplb_num;
> +}
> +
> static bool is_virtio_ccw_device_of_type(IplParameterBlock *iplb,
> int virtio_id)
> {
> @@ -620,7 +687,7 @@ void s390_ipl_reset_request(CPUState *cs, enum s390_reset reset_type)
> * this is the original boot device's SCSI
> * so restore IPL parameter info from it
> */
> - ipl->iplb_valid = s390_gen_initial_iplb(ipl);
> + ipl->iplb_valid = s390_build_iplb(get_boot_device(0), &ipl->iplb);
> }
> }
> if (reset_type == S390_RESET_MODIFIED_CLEAR ||
> @@ -714,7 +781,9 @@ void s390_ipl_prepare_cpu(S390CPU *cpu)
> if (!ipl->kernel || ipl->iplb_valid) {
> cpu->env.psw.addr = ipl->bios_start_addr;
> if (!ipl->iplb_valid) {
> - ipl->iplb_valid = s390_gen_initial_iplb(ipl);
> + ipl->iplb_valid = s390_init_all_iplbs(ipl);
> + } else {
> + ipl->qipl.chain_len = 0;
> }
> }
> s390_ipl_set_boot_menu(ipl);
Thomas
next prev parent reply other threads:[~2024-09-30 11:59 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-27 0:50 [PATCH V2 0/18] s390x: Add Full Boot Order Support jrossi
2024-09-27 0:51 ` [PATCH 01/18] hw/s390x/ipl: Provide more memory to the s390-ccw.img firmware jrossi
2024-09-27 0:51 ` [PATCH 02/18] pc-bios/s390-ccw: Use the libc from SLOF and remove sclp prints jrossi
2024-09-27 0:51 ` [PATCH 03/18] pc-bios/s390-ccw: Link the netboot code into the main s390-ccw.img binary jrossi
2024-09-27 0:51 ` [PATCH 04/18] hw/s390x: Remove the possibility to load the s390-netboot.img binary jrossi
2024-09-27 0:51 ` [PATCH 05/18] pc-bios/s390-ccw: Merge netboot.mak into the main Makefile jrossi
2024-09-27 0:51 ` [PATCH 06/18] docs/system/s390x/bootdevices: Update the documentation about network booting jrossi
2024-09-27 0:51 ` [PATCH 07/18] pc-bios/s390-ccw: Remove panics from ISO IPL path jrossi
2024-09-27 15:02 ` Thomas Huth
2024-09-27 17:15 ` Jared Rossi
2024-09-30 6:11 ` Thomas Huth
2024-09-30 13:10 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 08/18] pc-bios/s390-ccw: Remove panics from ECKD " jrossi
2024-09-27 15:29 ` Thomas Huth
2024-09-27 17:25 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 09/18] pc-bios/s390-ccw: Remove panics from SCSI " jrossi
2024-09-30 7:48 ` Thomas Huth
2024-09-30 10:13 ` Thomas Huth
2024-09-27 0:51 ` [PATCH 10/18] pc-bios/s390-ccw: Remove panics from DASD " jrossi
2024-09-30 8:14 ` Thomas Huth
2024-09-27 0:51 ` [PATCH 11/18] pc-bios/s390-ccw: Remove panics from Netboot " jrossi
2024-09-30 9:39 ` Thomas Huth
2024-09-30 13:15 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 12/18] pc-bios/s390-ccw: Enable failed IPL to return after error jrossi
2024-09-30 10:11 ` Thomas Huth
2024-09-30 13:29 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 13/18] include/hw/s390x: Add include files for common IPL structs jrossi
2024-09-30 10:42 ` Thomas Huth
2024-09-30 13:31 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 14/18] s390x: Add individual loadparm assignment to CCW device jrossi
2024-09-30 11:25 ` Thomas Huth
2024-09-27 0:51 ` [PATCH 15/18] hw/s390x: Build an IPLB for each boot device jrossi
2024-09-30 11:59 ` Thomas Huth [this message]
2024-09-30 13:39 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 16/18] s390x: Rebuild IPLB for SCSI device directly from DIAG308 jrossi
2024-09-30 12:15 ` Thomas Huth
2024-09-30 13:46 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 17/18] pc-bios/s390x: Enable multi-device boot loop jrossi
2024-09-30 12:32 ` Thomas Huth
2024-09-30 13:48 ` Jared Rossi
2024-09-30 13:08 ` Thomas Huth
2024-09-30 13:52 ` Jared Rossi
2024-09-27 0:51 ` [PATCH 18/18] docs/system: Update documentation for s390x IPL jrossi
2024-09-30 12:34 ` Thomas Huth
2024-09-30 13:14 ` [PATCH V2 0/18] s390x: Add Full Boot Order Support Thomas Huth
2024-09-30 14:20 ` Jared Rossi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=b3674e3b-5a23-41a0-a100-7fb2810a2775@redhat.com \
--to=thuth@redhat.com \
--cc=frankja@linux.ibm.com \
--cc=jrossi@linux.ibm.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-s390x@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).